From 2410013d9382b8129702fa3a3bf19a370ae7afc3 Mon Sep 17 00:00:00 2001 From: Dimitry Andric Date: Tue, 16 May 2017 19:47:09 +0000 Subject: [PATCH 1/9] Vendor import of clang trunk r303197: https://llvm.org/svn/llvm-project/cfe/trunk@303197 --- CMakeLists.txt | 13 +- cmake/caches/Apple-stage2.cmake | 7 +- cmake/caches/DistributionExample.cmake | 13 +- docs/CMakeLists.txt | 2 +- docs/ClangFormatStyleOptions.rst | 52 +- docs/ThreadSafetyAnalysis.rst | 4 +- include/clang-c/Index.h | 19 +- include/clang/AST/CXXInheritance.h | 3 +- include/clang/AST/Decl.h | 11 +- include/clang/AST/DeclCXX.h | 25 +- include/clang/AST/ExternalASTMerger.h | 2 + include/clang/AST/RecursiveASTVisitor.h | 2 +- include/clang/Basic/Attr.td | 24 + include/clang/Basic/Builtins.def | 3 + include/clang/Basic/DiagnosticGroups.td | 2 + include/clang/Basic/DiagnosticSemaKinds.td | 28 + include/clang/Basic/TargetOptions.h | 3 +- include/clang/Driver/Options.td | 3 + include/clang/Driver/SanitizerArgs.h | 1 + include/clang/Format/Format.h | 51 +- include/clang/Frontend/CodeGenOptions.def | 2 + include/clang/Lex/MacroInfo.h | 21 - include/clang/Lex/Preprocessor.h | 15 - include/clang/Parse/Parser.h | 7 +- include/clang/Sema/Sema.h | 41 +- include/clang/Tooling/RefactoringCallbacks.h | 50 ++ lib/AST/ASTImporter.cpp | 14 + lib/AST/ASTStructuralEquivalence.cpp | 5 + lib/AST/CXXInheritance.cpp | 114 ++- lib/AST/DeclCXX.cpp | 5 +- lib/AST/ExternalASTMerger.cpp | 6 + lib/AST/ODRHash.cpp | 2 +- lib/AST/Stmt.cpp | 2 +- lib/AST/Type.cpp | 7 +- lib/Basic/Targets.cpp | 5 + lib/CodeGen/BackendUtil.cpp | 5 +- lib/CodeGen/CGBlocks.cpp | 13 +- lib/CodeGen/CGBuiltin.cpp | 34 +- lib/CodeGen/CGCUDANV.cpp | 2 +- lib/CodeGen/CGCleanup.cpp | 3 +- lib/CodeGen/CGDebugInfo.cpp | 23 +- lib/CodeGen/CGDebugInfo.h | 5 +- lib/CodeGen/CGDecl.cpp | 13 +- lib/CodeGen/CGException.cpp | 13 +- lib/CodeGen/CGExpr.cpp | 4 +- lib/CodeGen/CGExprAgg.cpp | 20 +- lib/CodeGen/CGExprConstant.cpp | 10 +- lib/CodeGen/CGExprScalar.cpp | 8 +- lib/CodeGen/CGObjCGNU.cpp | 134 ++-- lib/CodeGen/CGObjCMac.cpp | 170 ++--- lib/CodeGen/CGOpenMPRuntime.cpp | 2 +- lib/CodeGen/CodeGenAction.cpp | 10 +- lib/CodeGen/CodeGenModule.cpp | 2 +- lib/CodeGen/CodeGenTypes.cpp | 2 +- lib/CodeGen/ItaniumCXXABI.cpp | 2 +- lib/CodeGen/TargetInfo.cpp | 24 +- lib/Driver/SanitizerArgs.cpp | 33 +- lib/Driver/ToolChains/Arch/Mips.cpp | 16 +- lib/Driver/ToolChains/Myriad.cpp | 8 +- lib/Driver/ToolChains/WebAssembly.cpp | 5 +- lib/Format/ContinuationIndenter.cpp | 16 + lib/Format/Format.cpp | 26 +- lib/Format/TokenAnnotator.cpp | 34 +- lib/Format/UnwrappedLineParser.cpp | 34 +- lib/Format/WhitespaceManager.cpp | 11 +- lib/Frontend/CompilerInvocation.cpp | 2 + lib/Headers/avxintrin.h | 184 +++-- lib/Headers/emmintrin.h | 281 ++++--- lib/Headers/intrin.h | 6 - lib/Headers/mmintrin.h | 101 ++- lib/Headers/opencl-c.h | 14 +- lib/Headers/pmmintrin.h | 8 +- lib/Headers/prfchwintrin.h | 6 +- lib/Headers/smmintrin.h | 10 +- lib/Headers/tmmintrin.h | 88 ++- lib/Headers/x86intrin.h | 2 - lib/Headers/xmmintrin.h | 28 +- lib/Index/IndexBody.cpp | 47 ++ lib/Index/IndexDecl.cpp | 30 + lib/Index/IndexTypeSourceInfo.cpp | 30 +- lib/Index/IndexingContext.cpp | 15 + lib/Lex/MacroInfo.cpp | 2 - lib/Lex/ModuleMap.cpp | 20 +- lib/Lex/PPDirectives.cpp | 27 +- lib/Lex/Preprocessor.cpp | 7 +- lib/Parse/ParseDecl.cpp | 11 +- lib/Parse/ParseDeclCXX.cpp | 8 +- lib/Parse/ParseExpr.cpp | 45 +- lib/Parse/ParseExprCXX.cpp | 7 +- lib/Parse/ParseTemplate.cpp | 16 +- lib/Sema/CMakeLists.txt | 1 + lib/Sema/Sema.cpp | 32 +- lib/Sema/SemaCast.cpp | 3 +- lib/Sema/SemaCodeComplete.cpp | 92 ++- lib/Sema/SemaDecl.cpp | 259 +++++-- lib/Sema/SemaDeclAttr.cpp | 13 +- lib/Sema/SemaExpr.cpp | 242 +++++- lib/Sema/SemaExprCXX.cpp | 29 +- lib/Sema/SemaExprObjC.cpp | 3 +- lib/Sema/SemaInit.cpp | 40 +- lib/Sema/SemaLookup.cpp | 52 +- lib/Sema/SemaOverload.cpp | 6 +- lib/Sema/SemaStmt.cpp | 54 +- lib/Sema/SemaTemplate.cpp | 196 +++-- lib/Sema/SemaTemplateInstantiateDecl.cpp | 14 + lib/Sema/SemaType.cpp | 5 +- lib/Serialization/ASTReader.cpp | 3 +- lib/Serialization/ASTWriter.cpp | 1 - .../Checkers/BasicObjCFoundationChecks.cpp | 46 +- .../Checkers/BuiltinFunctionChecker.cpp | 16 + .../Checkers/NoReturnFunctionChecker.cpp | 4 +- .../Checkers/RetainCountChecker.cpp | 47 +- lib/StaticAnalyzer/Checkers/SelectorExtras.h | 40 +- .../Checkers/StdLibraryFunctionsChecker.cpp | 5 +- .../Core/ExprEngineCallAndReturn.cpp | 3 +- lib/Tooling/RefactoringCallbacks.cpp | 156 +++- test/Analysis/builtin-assume.c | 8 + test/CXX/drs/dr20xx.cpp | 30 + test/CXX/drs/dr4xx.cpp | 7 +- test/CodeCompletion/member-access.cpp | 80 ++ test/CodeGen/asan-globals-gc.cpp | 15 +- test/CodeGen/asan-no-globals-no-comdat.cpp | 11 + test/CodeGen/mips-aggregate-arg.c | 38 + test/CodeGen/sanitize-recover.c | 10 +- test/CodeGen/sparcv8-inline-asm.c | 11 + test/CodeGen/thinlto_backend.ll | 6 +- test/CodeGen/x86_64-mno-sse.c | 15 + test/CodeGen/xray-customevent.cpp | 28 + test/CodeGenCXX/array-default-argument.cpp | 36 + .../CodeGenCXX/linetable-virtual-variadic.cpp | 6 +- test/CodeGenCXX/vla.cpp | 59 +- test/CodeGenObjC/arc-blocks.m | 10 + test/CodeGenObjC/arc-foreach.m | 6 +- ...amdgpu-debug-info-pointer-address-space.cl | 20 +- .../amdgpu-debug-info-variable-expression.cl | 30 +- .../constant-addr-space-globals.cl | 8 +- test/Driver/fsanitize.c | 9 +- test/Driver/myriad-toolchain.c | 4 +- test/Driver/wasm-toolchain.c | 6 +- test/FixIt/fixit-availability.c | 2 +- test/FixIt/fixit-availability.mm | 42 +- test/Import/conflicting-struct/Inputs/S1.cpp | 6 + test/Import/conflicting-struct/Inputs/S2.cpp | 7 + test/Import/conflicting-struct/test.cpp | 7 + test/Index/Core/index-dependent-source.cpp | 143 ++++ test/Index/Core/index-instantiated-source.cpp | 39 + test/Index/Core/index-source.cpp | 52 ++ test/Index/complete-available.m | 20 + test/Index/get-cursor.m | 11 + test/Misc/warning-flags.c | 4 +- test/Modules/DebugInfoNamespace.cpp | 19 + test/Modules/DebugInfoSubmoduleImport.c | 5 + test/Modules/Inputs/DebugInfoNamespace/A.h | 3 + test/Modules/Inputs/DebugInfoNamespace/B.h | 3 + .../DebugInfoNamespace/module.modulemap | 8 + .../Frameworks/Sub.framework/Headers/B.h | 1 + .../Frameworks/Sub.framework/Headers/Sub.h | 2 + .../Sub.framework/PrivateHeaders/BPriv.h | 1 + .../Sub.framework/PrivateHeaders/SubPriv.h | 1 + .../Inputs/MainA.framework/Headers/A.h | 1 + .../Inputs/MainA.framework/Headers/Main.h | 2 + .../MainA.framework/Modules/module.modulemap | 12 + .../Modules/module.private.modulemap | 12 + .../MainA.framework/PrivateHeaders/APriv.h | 1 + .../MainA.framework/PrivateHeaders/MainPriv.h | 1 + test/Modules/Inputs/SameHeader/A.h | 3 + test/Modules/Inputs/SameHeader/B.h | 4 + test/Modules/Inputs/SameHeader/C.h | 12 + .../Inputs/SameHeader/module.modulemap | 11 + test/Modules/find-privateheaders.m | 13 +- test/Modules/odr_hash.cpp | 33 + test/Modules/preprocess-module.cpp | 5 + test/Modules/redefinition-same-header.m | 20 + test/OpenMP/report_default_DSA.cpp | 18 + test/PCH/cxx-templates.cpp | 8 + test/PCH/cxx-templates.h | 3 + test/Parser/objc-available.m | 6 + test/Preprocessor/predefined-arch-macros.c | 8 + test/Preprocessor/x86_target_features.c | 8 + test/Sema/overloadable.c | 15 + test/Sema/redefinition-same-header.c | 14 + test/Sema/typo-correction.c | 7 + test/Sema/vector-cast.c | 5 +- test/Sema/vector-gcc-compat.c | 330 +++++++++ test/Sema/vector-gcc-compat.cpp | 328 +++++++++ test/Sema/vector-ops.c | 140 ++-- test/Sema/zvector.c | 24 +- test/SemaCXX/constructor-initializer.cpp | 19 + test/SemaCXX/cxx1y-generic-lambdas.cpp | 7 + .../cxx1y-variable-templates_top_level.cpp | 7 +- test/SemaCXX/enable_if.cpp | 14 + test/SemaCXX/for-range-examples.cpp | 34 + test/SemaCXX/invalid-member-expr.cpp | 4 +- test/SemaCXX/modules-ts.cppm | 3 +- test/SemaCXX/type-traits.cpp | 2 +- test/SemaCXX/typo-correction.cpp | 11 +- test/SemaCXX/vector-no-lax.cpp | 2 +- test/SemaCXX/warn-unused-filescoped.cpp | 18 +- test/SemaObjC/method-bad-param.m | 6 + test/SemaObjC/unguarded-availability.m | 21 +- test/SemaObjCXX/interface-return-type.mm | 7 + test/SemaObjCXX/is-base-of.mm | 25 + test/SemaOpenCL/array-init.cl | 20 + test/SemaOpenCL/storageclass.cl | 11 +- test/SemaTemplate/deduction-crash.cpp | 2 +- test/SemaTemplate/default-arguments.cpp | 16 + test/SemaTemplate/explicit-instantiation.cpp | 2 +- .../explicit-specialization-member.cpp | 11 +- .../ms-lookup-template-base-classes.cpp | 3 +- test/SemaTemplate/typo-template-name.cpp | 43 ++ tools/c-index-test/c-index-test.c | 13 + tools/clang-import-test/clang-import-test.cpp | 46 +- tools/libclang/CIndex.cpp | 29 + tools/libclang/libclang.exports | 1 + unittests/Format/FormatTest.cpp | 104 ++- unittests/Format/FormatTestJS.cpp | 69 +- unittests/Format/FormatTestSelective.cpp | 4 +- unittests/Tooling/RecursiveASTVisitorTest.cpp | 8 + .../Tooling/RefactoringCallbacksTest.cpp | 91 ++- www/cxx_dr_status.html | 690 +++++++++--------- 220 files changed, 5201 insertions(+), 1618 deletions(-) create mode 100644 test/Analysis/builtin-assume.c create mode 100644 test/CXX/drs/dr20xx.cpp create mode 100644 test/CodeGen/asan-no-globals-no-comdat.cpp create mode 100644 test/CodeGen/mips-aggregate-arg.c create mode 100644 test/CodeGen/sparcv8-inline-asm.c create mode 100644 test/CodeGen/x86_64-mno-sse.c create mode 100644 test/CodeGen/xray-customevent.cpp create mode 100644 test/CodeGenCXX/array-default-argument.cpp create mode 100644 test/Import/conflicting-struct/Inputs/S1.cpp create mode 100644 test/Import/conflicting-struct/Inputs/S2.cpp create mode 100644 test/Import/conflicting-struct/test.cpp create mode 100644 test/Index/Core/index-dependent-source.cpp create mode 100644 test/Index/Core/index-instantiated-source.cpp create mode 100644 test/Index/complete-available.m create mode 100644 test/Modules/DebugInfoNamespace.cpp create mode 100644 test/Modules/Inputs/DebugInfoNamespace/A.h create mode 100644 test/Modules/Inputs/DebugInfoNamespace/B.h create mode 100644 test/Modules/Inputs/DebugInfoNamespace/module.modulemap create mode 100644 test/Modules/Inputs/MainA.framework/Frameworks/Sub.framework/Headers/B.h create mode 100644 test/Modules/Inputs/MainA.framework/Frameworks/Sub.framework/Headers/Sub.h create mode 100644 test/Modules/Inputs/MainA.framework/Frameworks/Sub.framework/PrivateHeaders/BPriv.h create mode 100644 test/Modules/Inputs/MainA.framework/Frameworks/Sub.framework/PrivateHeaders/SubPriv.h create mode 100644 test/Modules/Inputs/MainA.framework/Headers/A.h create mode 100644 test/Modules/Inputs/MainA.framework/Headers/Main.h create mode 100644 test/Modules/Inputs/MainA.framework/Modules/module.modulemap create mode 100644 test/Modules/Inputs/MainA.framework/Modules/module.private.modulemap create mode 100644 test/Modules/Inputs/MainA.framework/PrivateHeaders/APriv.h create mode 100644 test/Modules/Inputs/MainA.framework/PrivateHeaders/MainPriv.h create mode 100644 test/Modules/Inputs/SameHeader/A.h create mode 100644 test/Modules/Inputs/SameHeader/B.h create mode 100644 test/Modules/Inputs/SameHeader/C.h create mode 100644 test/Modules/Inputs/SameHeader/module.modulemap create mode 100644 test/Modules/redefinition-same-header.m create mode 100644 test/OpenMP/report_default_DSA.cpp create mode 100644 test/Sema/redefinition-same-header.c create mode 100644 test/Sema/vector-gcc-compat.c create mode 100644 test/Sema/vector-gcc-compat.cpp create mode 100644 test/SemaObjCXX/interface-return-type.mm create mode 100644 test/SemaObjCXX/is-base-of.mm create mode 100644 test/SemaOpenCL/array-init.cl create mode 100644 test/SemaTemplate/typo-template-name.cpp diff --git a/CMakeLists.txt b/CMakeLists.txt index ca696b1ce79f..9e43a103b2bc 100644 --- a/CMakeLists.txt +++ b/CMakeLists.txt @@ -579,10 +579,17 @@ if (CLANG_ENABLE_BOOTSTRAP) add_dependencies(clang-bootstrap-deps compiler-rt) endif() + set(C_COMPILER "clang") + set(CXX_COMPILER "clang++") + if(WIN32) + set(C_COMPILER "clang-cl.exe") + set(CXX_COMPILER "clang-cl.exe") + endif() + set(COMPILER_OPTIONS - -DCMAKE_CXX_COMPILER=${LLVM_RUNTIME_OUTPUT_INTDIR}/clang++ - -DCMAKE_C_COMPILER=${LLVM_RUNTIME_OUTPUT_INTDIR}/clang - -DCMAKE_ASM_COMPILER=${LLVM_RUNTIME_OUTPUT_INTDIR}/clang) + -DCMAKE_CXX_COMPILER=${LLVM_RUNTIME_OUTPUT_INTDIR}/${CXX_COMPILER} + -DCMAKE_C_COMPILER=${LLVM_RUNTIME_OUTPUT_INTDIR}/${C_COMPILER} + -DCMAKE_ASM_COMPILER=${LLVM_RUNTIME_OUTPUT_INTDIR}/${C_COMPILER}) if(BOOTSTRAP_LLVM_BUILD_INSTRUMENTED) add_dependencies(clang-bootstrap-deps llvm-profdata) diff --git a/cmake/caches/Apple-stage2.cmake b/cmake/caches/Apple-stage2.cmake index 11c595c1530f..f07973dc0ab2 100644 --- a/cmake/caches/Apple-stage2.cmake +++ b/cmake/caches/Apple-stage2.cmake @@ -13,6 +13,7 @@ set(CLANG_LINKS_TO_CREATE clang++ cc c++ CACHE STRING "") set(CMAKE_MACOSX_RPATH ON CACHE BOOL "") set(LLVM_ENABLE_ZLIB ON CACHE BOOL "") set(LLVM_ENABLE_BACKTRACES OFF CACHE BOOL "") +set(LLVM_ENABLE_MODULES ON CACHE BOOL "") set(LLVM_EXTERNALIZE_DEBUGINFO ON CACHE BOOL "") set(CLANG_PLUGIN_SUPPORT OFF CACHE BOOL "") set(BUG_REPORT_URL "http://developer.apple.com/bugreporter/" CACHE STRING "") @@ -28,8 +29,10 @@ set(LLVM_BUILD_TESTS ON CACHE BOOL "") set(LLVM_ENABLE_LTO ON CACHE BOOL "") set(CMAKE_C_FLAGS "-fno-stack-protector -fno-common -Wno-profile-instr-unprofiled" CACHE STRING "") set(CMAKE_CXX_FLAGS "-fno-stack-protector -fno-common -Wno-profile-instr-unprofiled" CACHE STRING "") -set(CMAKE_C_FLAGS_RELWITHDEBINFO "-O2 -gline-tables-only -DNDEBUG" CACHE STRING "") -set(CMAKE_CXX_FLAGS_RELWITHDEBINFO "-O2 -gline-tables-only -DNDEBUG" CACHE STRING "") +if(LLVM_ENABLE_LTO AND NOT LLVM_ENABLE_LTO STREQUAL "THIN") + set(CMAKE_C_FLAGS_RELWITHDEBINFO "-O2 -gline-tables-only -DNDEBUG" CACHE STRING "") + set(CMAKE_CXX_FLAGS_RELWITHDEBINFO "-O2 -gline-tables-only -DNDEBUG" CACHE STRING "") +endif() set(CMAKE_BUILD_TYPE RelWithDebInfo CACHE STRING "") set(LIBCXX_INSTALL_LIBRARY OFF CACHE BOOL "") diff --git a/cmake/caches/DistributionExample.cmake b/cmake/caches/DistributionExample.cmake index 862f547c1656..551f4ee07eac 100644 --- a/cmake/caches/DistributionExample.cmake +++ b/cmake/caches/DistributionExample.cmake @@ -29,6 +29,13 @@ set(CLANG_BOOTSTRAP_TARGETS # Setup the bootstrap build. set(CLANG_ENABLE_BOOTSTRAP ON CACHE BOOL "") -set(CLANG_BOOTSTRAP_CMAKE_ARGS - -C ${CMAKE_CURRENT_LIST_DIR}/DistributionExample-stage2.cmake - CACHE STRING "") + +if(STAGE2_CACHE_FILE) + set(CLANG_BOOTSTRAP_CMAKE_ARGS + -C ${STAGE2_CACHE_FILE} + CACHE STRING "") +else() + set(CLANG_BOOTSTRAP_CMAKE_ARGS + -C ${CMAKE_CURRENT_LIST_DIR}/DistributionExample-stage2.cmake + CACHE STRING "") +endif() diff --git a/docs/CMakeLists.txt b/docs/CMakeLists.txt index 13b79fdfa534..d2956c18f80c 100644 --- a/docs/CMakeLists.txt +++ b/docs/CMakeLists.txt @@ -91,8 +91,8 @@ endif() endif() if (LLVM_ENABLE_SPHINX) + include(AddSphinxTarget) if (SPHINX_FOUND) - include(AddSphinxTarget) if (${SPHINX_OUTPUT_HTML}) add_sphinx_target(html clang) add_custom_command(TARGET docs-clang-html POST_BUILD diff --git a/docs/ClangFormatStyleOptions.rst b/docs/ClangFormatStyleOptions.rst index ed628e370087..f54acd9b81dd 100644 --- a/docs/ClangFormatStyleOptions.rst +++ b/docs/ClangFormatStyleOptions.rst @@ -209,23 +209,45 @@ the configuration (without a prefix: ``Auto``). float b = 23; std::string ccc = 23; -**AlignEscapedNewlinesLeft** (``bool``) - If ``true``, aligns escaped newlines as far left as possible. - Otherwise puts them into the right-most column. +**AlignEscapedNewlines** (``EscapedNewlineAlignmentStyle``) + Options for aligning backslashes in escaped newlines. - .. code-block:: c++ + Possible values: + + * ``ENAS_DontAlign`` (in configuration: ``DontAlign``) + Don't align escaped newlines. + + .. code-block:: c++ + + #define A \ + int aaaa; \ + int b; \ + int dddddddddd; + + * ``ENAS_Left`` (in configuration: ``Left``) + Align escaped newlines as far left as possible. + + .. code-block:: c++ + + true: + #define A \ + int aaaa; \ + int b; \ + int dddddddddd; + + false: + + * ``ENAS_Right`` (in configuration: ``Right``) + Align escaped newlines in the right-most column. + + .. code-block:: c++ + + #define A \ + int aaaa; \ + int b; \ + int dddddddddd; - true: - #define A \ - int aaaa; \ - int b; \ - int dddddddddd; - false: - #define A \ - int aaaa; \ - int b; \ - int dddddddddd; **AlignOperands** (``bool``) If ``true``, horizontally align operands of binary and ternary @@ -1525,7 +1547,7 @@ the configuration (without a prefix: ``Auto``). Use C++03-compatible syntax. * ``LS_Cpp11`` (in configuration: ``Cpp11``) - Use features of C++11, C++14 and C++1z (e.g. ``A>`` instead of + Use features of C++11, C++14 and C++1z (e.g. ``A>`` instead of ``A >``). * ``LS_Auto`` (in configuration: ``Auto``) diff --git a/docs/ThreadSafetyAnalysis.rst b/docs/ThreadSafetyAnalysis.rst index 47cf4ef3994e..ea8e98a1884b 100644 --- a/docs/ThreadSafetyAnalysis.rst +++ b/docs/ThreadSafetyAnalysis.rst @@ -884,11 +884,11 @@ implementation. // Deprecated. #define PT_GUARDED_VAR \ - THREAD_ANNOTATION_ATTRIBUTE__(pt_guarded) + THREAD_ANNOTATION_ATTRIBUTE__(pt_guarded_var) // Deprecated. #define GUARDED_VAR \ - THREAD_ANNOTATION_ATTRIBUTE__(guarded) + THREAD_ANNOTATION_ATTRIBUTE__(guarded_var) // Replaced by REQUIRES #define EXCLUSIVE_LOCKS_REQUIRED(...) \ diff --git a/include/clang-c/Index.h b/include/clang-c/Index.h index c50ac1b6d248..462d9269f7ad 100644 --- a/include/clang-c/Index.h +++ b/include/clang-c/Index.h @@ -32,7 +32,7 @@ * compatible, thus CINDEX_VERSION_MAJOR is expected to remain stable. */ #define CINDEX_VERSION_MAJOR 0 -#define CINDEX_VERSION_MINOR 38 +#define CINDEX_VERSION_MINOR 39 #define CINDEX_VERSION_ENCODE(major, minor) ( \ ((major) * 10000) \ @@ -4080,6 +4080,23 @@ CINDEX_LINKAGE unsigned clang_Cursor_isObjCOptional(CXCursor C); */ CINDEX_LINKAGE unsigned clang_Cursor_isVariadic(CXCursor C); +/** + * \brief Returns non-zero if the given cursor points to a symbol marked with + * external_source_symbol attribute. + * + * \param language If non-NULL, and the attribute is present, will be set to + * the 'language' string from the attribute. + * + * \param definedIn If non-NULL, and the attribute is present, will be set to + * the 'definedIn' string from the attribute. + * + * \param isGenerated If non-NULL, and the attribute is present, will be set to + * non-zero if the 'generated_declaration' is set in the attribute. + */ +CINDEX_LINKAGE unsigned clang_Cursor_isExternalSymbol(CXCursor C, + CXString *language, CXString *definedIn, + unsigned *isGenerated); + /** * \brief Given a cursor that represents a declaration, return the associated * comment's source range. The range may include multiple consecutive comments diff --git a/include/clang/AST/CXXInheritance.h b/include/clang/AST/CXXInheritance.h index 3cf058f26bc6..a7961ebe8ce6 100644 --- a/include/clang/AST/CXXInheritance.h +++ b/include/clang/AST/CXXInheritance.h @@ -161,7 +161,8 @@ class CXXBasePaths { void ComputeDeclsFound(); bool lookupInBases(ASTContext &Context, const CXXRecordDecl *Record, - CXXRecordDecl::BaseMatchesCallback BaseMatches); + CXXRecordDecl::BaseMatchesCallback BaseMatches, + bool LookupInDependent = false); public: typedef std::list::iterator paths_iterator; diff --git a/include/clang/AST/Decl.h b/include/clang/AST/Decl.h index 573ea55de1fd..facef8e55f7a 100644 --- a/include/clang/AST/Decl.h +++ b/include/clang/AST/Decl.h @@ -966,9 +966,16 @@ class VarDecl : public DeclaratorDecl, public Redeclarable { /// hasLocalStorage - Returns true if a variable with function scope /// is a non-static local variable. bool hasLocalStorage() const { - if (getStorageClass() == SC_None) + if (getStorageClass() == SC_None) { + // OpenCL v1.2 s6.5.3: The __constant or constant address space name is + // used to describe variables allocated in global memory and which are + // accessed inside a kernel(s) as read-only variables. As such, variables + // in constant address space cannot have local storage. + if (getType().getAddressSpace() == LangAS::opencl_constant) + return false; // Second check is for C++11 [dcl.stc]p4. return !isFileVarDecl() && getTSCSpec() == TSCS_unspecified; + } // Global Named Register (GNU extension) if (getStorageClass() == SC_Register && !isLocalVarDeclOrParm()) @@ -2478,7 +2485,7 @@ class FieldDecl : public DeclaratorDecl, public Mergeable { void setCapturedVLAType(const VariableArrayType *VLAType); /// getParent - Returns the parent of this field declaration, which - /// is the struct in which this method is defined. + /// is the struct in which this field is defined. const RecordDecl *getParent() const { return cast(getDeclContext()); } diff --git a/include/clang/AST/DeclCXX.h b/include/clang/AST/DeclCXX.h index 13921a132cfb..6965e8143ff6 100644 --- a/include/clang/AST/DeclCXX.h +++ b/include/clang/AST/DeclCXX.h @@ -1563,10 +1563,13 @@ class CXXRecordDecl : public RecordDecl { /// \param Paths used to record the paths from this class to its base class /// subobjects that match the search criteria. /// + /// \param LookupInDependent can be set to true to extend the search to + /// dependent base classes. + /// /// \returns true if there exists any path from this class to a base class /// subobject that matches the search criteria. - bool lookupInBases(BaseMatchesCallback BaseMatches, - CXXBasePaths &Paths) const; + bool lookupInBases(BaseMatchesCallback BaseMatches, CXXBasePaths &Paths, + bool LookupInDependent = false) const; /// \brief Base-class lookup callback that determines whether the given /// base class specifier refers to a specific class declaration. @@ -1607,6 +1610,16 @@ class CXXRecordDecl : public RecordDecl { static bool FindOrdinaryMember(const CXXBaseSpecifier *Specifier, CXXBasePath &Path, DeclarationName Name); + /// \brief Base-class lookup callback that determines whether there exists + /// a member with the given name. + /// + /// This callback can be used with \c lookupInBases() to find members + /// of the given name within a C++ class hierarchy, including dependent + /// classes. + static bool + FindOrdinaryMemberInDependentClasses(const CXXBaseSpecifier *Specifier, + CXXBasePath &Path, DeclarationName Name); + /// \brief Base-class lookup callback that determines whether there exists /// an OpenMP declare reduction member with the given name. /// @@ -1633,6 +1646,14 @@ class CXXRecordDecl : public RecordDecl { /// \brief Get the indirect primary bases for this class. void getIndirectPrimaryBases(CXXIndirectPrimaryBaseSet& Bases) const; + /// Performs an imprecise lookup of a dependent name in this class. + /// + /// This function does not follow strict semantic rules and should be used + /// only when lookup rules can be relaxed, e.g. indexing. + std::vector + lookupDependentName(const DeclarationName &Name, + llvm::function_ref Filter); + /// Renders and displays an inheritance diagram /// for this C++ class and all of its base classes (transitively) using /// GraphViz. diff --git a/include/clang/AST/ExternalASTMerger.h b/include/clang/AST/ExternalASTMerger.h index 51d0c30ad23b..55459df1fe6b 100644 --- a/include/clang/AST/ExternalASTMerger.h +++ b/include/clang/AST/ExternalASTMerger.h @@ -44,6 +44,8 @@ class ExternalASTMerger : public ExternalASTSource { FindExternalLexicalDecls(const DeclContext *DC, llvm::function_ref IsKindWeWant, SmallVectorImpl &Result) override; + + void CompleteType(TagDecl *Tag) override; }; } // end namespace clang diff --git a/include/clang/AST/RecursiveASTVisitor.h b/include/clang/AST/RecursiveASTVisitor.h index 1b5850a05b37..cd2a39449825 100644 --- a/include/clang/AST/RecursiveASTVisitor.h +++ b/include/clang/AST/RecursiveASTVisitor.h @@ -2326,7 +2326,7 @@ DEF_TRAVERSE_STMT(LambdaExpr, { } TypeLoc TL = S->getCallOperator()->getTypeSourceInfo()->getTypeLoc(); - FunctionProtoTypeLoc Proto = TL.castAs(); + FunctionProtoTypeLoc Proto = TL.getAsAdjusted(); if (S->hasExplicitParameters() && S->hasExplicitResultType()) { // Visit the whole type. diff --git a/include/clang/Basic/Attr.td b/include/clang/Basic/Attr.td index 3eeeb1bdc971..4eb958e3f4d5 100644 --- a/include/clang/Basic/Attr.td +++ b/include/clang/Basic/Attr.td @@ -652,6 +652,30 @@ def Availability : InheritableAttr { .Case("tvos_app_extension", "tvOS (App Extension)") .Case("watchos_app_extension", "watchOS (App Extension)") .Default(llvm::StringRef()); +} +static llvm::StringRef getPlatformNameSourceSpelling(llvm::StringRef Platform) { + return llvm::StringSwitch(Platform) + .Case("ios", "iOS") + .Case("macos", "macOS") + .Case("tvos", "tvOS") + .Case("watchos", "watchOS") + .Case("ios_app_extension", "iOSApplicationExtension") + .Case("macos_app_extension", "macOSApplicationExtension") + .Case("tvos_app_extension", "tvOSApplicationExtension") + .Case("watchos_app_extension", "watchOSApplicationExtension") + .Default(Platform); +} +static llvm::StringRef canonicalizePlatformName(llvm::StringRef Platform) { + return llvm::StringSwitch(Platform) + .Case("iOS", "ios") + .Case("macOS", "macos") + .Case("tvOS", "tvos") + .Case("watchOS", "watchos") + .Case("iOSApplicationExtension", "ios_app_extension") + .Case("macOSApplicationExtension", "macos_app_extension") + .Case("tvOSApplicationExtension", "tvos_app_extension") + .Case("watchOSApplicationExtension", "watchos_app_extension") + .Default(Platform); } }]; let HasCustomParsing = 1; let DuplicatesAllowedWhileMerging = 1; diff --git a/include/clang/Basic/Builtins.def b/include/clang/Basic/Builtins.def index 816ea156f979..a9ec172422ab 100644 --- a/include/clang/Basic/Builtins.def +++ b/include/clang/Basic/Builtins.def @@ -1409,6 +1409,9 @@ LANGBUILTIN(to_private, "v*v*", "tn", OCLC20_LANG) BUILTIN(__builtin_os_log_format_buffer_size, "zcC*.", "p:0:nut") BUILTIN(__builtin_os_log_format, "v*v*cC*.", "p:0:nt") +// Builtins for XRay +BUILTIN(__xray_customevent, "vcC*z", "") + #undef BUILTIN #undef LIBBUILTIN #undef LANGBUILTIN diff --git a/include/clang/Basic/DiagnosticGroups.td b/include/clang/Basic/DiagnosticGroups.td index 05e03fab40fa..e1a41584023c 100644 --- a/include/clang/Basic/DiagnosticGroups.td +++ b/include/clang/Basic/DiagnosticGroups.td @@ -486,6 +486,7 @@ def UnneededInternalDecl : DiagGroup<"unneeded-internal-declaration">; def UnneededMemberFunction : DiagGroup<"unneeded-member-function">; def UnusedPrivateField : DiagGroup<"unused-private-field">; def UnusedFunction : DiagGroup<"unused-function", [UnneededInternalDecl]>; +def UnusedTemplate : DiagGroup<"unused-template", [UnneededInternalDecl]>; def UnusedMemberFunction : DiagGroup<"unused-member-function", [UnneededMemberFunction]>; def UnusedLabel : DiagGroup<"unused-label">; @@ -627,6 +628,7 @@ def Conversion : DiagGroup<"conversion", def Unused : DiagGroup<"unused", [UnusedArgument, UnusedFunction, UnusedLabel, // UnusedParameter, (matches GCC's behavior) + // UnusedTemplate, (clean-up libc++ before enabling) // UnusedMemberFunction, (clean-up llvm before enabling) UnusedPrivateField, UnusedLambdaCapture, UnusedLocalTypedef, UnusedValue, UnusedVariable, diff --git a/include/clang/Basic/DiagnosticSemaKinds.td b/include/clang/Basic/DiagnosticSemaKinds.td index a0c0e5f86449..1db6704f6d1f 100644 --- a/include/clang/Basic/DiagnosticSemaKinds.td +++ b/include/clang/Basic/DiagnosticSemaKinds.td @@ -303,6 +303,8 @@ def note_empty_parens_zero_initialize : Note< "replace parentheses with an initializer to declare a variable">; def warn_unused_function : Warning<"unused function %0">, InGroup, DefaultIgnore; +def warn_unused_template : Warning<"unused %select{function|variable}0 template %1">, + InGroup, DefaultIgnore; def warn_unused_member_function : Warning<"unused member function %0">, InGroup, DefaultIgnore; def warn_used_but_marked_unused: Warning<"%0 was marked unused but was used">, @@ -2463,6 +2465,9 @@ def err_attribute_invalid_size : Error< "vector size not an integral multiple of component size">; def err_attribute_zero_size : Error<"zero vector size">; def err_attribute_size_too_large : Error<"vector size too large">; +def err_typecheck_vector_not_convertable_implict_truncation : Error< + "cannot convert between %select{scalar|vector}0 type %1 and vector type" + " %2 as implicit conversion would cause truncation">; def err_typecheck_vector_not_convertable : Error< "cannot convert between vector values of different size (%0 and %1)">; def err_typecheck_vector_not_convertable_non_scalar : Error< @@ -4607,6 +4612,8 @@ def err_abi_tag_on_redeclaration : Error< "cannot add 'abi_tag' attribute in a redeclaration">; def err_new_abi_tag_on_redeclaration : Error< "'abi_tag' %0 missing in original declaration">; +def note_use_ifdef_guards : Note< + "unguarded header; consider using #ifdef guards or #pragma once">; def note_deleted_dtor_no_operator_delete : Note< "virtual destructor requires an unambiguous, accessible 'operator delete'">; @@ -5775,6 +5782,9 @@ def err_objc_object_assignment : Error< "cannot assign to class object (%0 invalid)">; def err_typecheck_invalid_operands : Error< "invalid operands to binary expression (%0 and %1)">; +def err_typecheck_logical_vector_expr_gnu_cpp_restrict : Error< + "logical expression with vector %select{type %1 and non-vector type %2|types" + " %1 and %2}0 is only supported in C++">; def err_typecheck_sub_ptr_compatible : Error< "%diff{$ and $ are not pointers to compatible types|" "pointers to incompatible types}0,1">; @@ -8186,9 +8196,20 @@ def err_undeclared_use_suggest : Error< "use of undeclared %0; did you mean %1?">; def err_undeclared_var_use_suggest : Error< "use of undeclared identifier %0; did you mean %1?">; +def err_no_template : Error<"no template named %0">; def err_no_template_suggest : Error<"no template named %0; did you mean %1?">; +def err_no_member_template : Error<"no template named %0 in %1">; def err_no_member_template_suggest : Error< "no template named %0 in %1; did you mean %select{|simply }2%3?">; +def err_non_template_in_template_id : Error< + "%0 does not name a template but is followed by template arguments">; +def err_non_template_in_template_id_suggest : Error< + "%0 does not name a template but is followed by template arguments; " + "did you mean %1?">; +def err_non_template_in_member_template_id_suggest : Error< + "member %0 of %1 is not a template; did you mean %select{|simply }2%3?">; +def note_non_template_in_template_id_found : Note< + "non-template declaration found by name lookup">; def err_mem_init_not_member_or_class_suggest : Error< "initializer %0 does not name a non-static data member or base " "class; did you mean the %select{base class|member}1 %2?">; @@ -8875,6 +8896,13 @@ def ext_equivalent_internal_linkage_decl_in_modules : ExtWarn< InGroup>; def note_equivalent_internal_linkage_decl : Note< "declared here%select{ in module '%1'|}0">; + +def note_redefinition_modules_same_file : Note< + "'%0' included multiple times, additional include site in header from module '%1'">; +def note_redefinition_modules_same_file_modulemap : Note< + "consider adding '%0' as part of '%1' definition">; +def note_redefinition_include_same_file : Note< + "'%0' included multiple times, additional include site here">; } let CategoryName = "Coroutines Issue" in { diff --git a/include/clang/Basic/TargetOptions.h b/include/clang/Basic/TargetOptions.h index 2889cce5963b..6ca1ba39c8fb 100644 --- a/include/clang/Basic/TargetOptions.h +++ b/include/clang/Basic/TargetOptions.h @@ -24,8 +24,7 @@ namespace clang { /// \brief Options for controlling the target. class TargetOptions { public: - /// If given, the name of the target triple to compile for. If not given the - /// target will be selected to match the host. + /// The name of the target triple to compile for. std::string Triple; /// When compiling for the device side, contains the triple used to compile diff --git a/include/clang/Driver/Options.td b/include/clang/Driver/Options.td index 31015228f362..d812bd8ec032 100644 --- a/include/clang/Driver/Options.td +++ b/include/clang/Driver/Options.td @@ -827,6 +827,9 @@ def fno_sanitize_address_use_after_scope : Flag<["-"], "fno-sanitize-address-use Group, Flags<[CoreOption, DriverOption]>, HelpText<"Disable use-after-scope detection in AddressSanitizer">; +def fsanitize_address_globals_dead_stripping : Flag<["-"], "fsanitize-address-globals-dead-stripping">, + Group, + HelpText<"Enable linker dead stripping of globals in AddressSanitizer">; def fsanitize_recover : Flag<["-"], "fsanitize-recover">, Group; def fno_sanitize_recover : Flag<["-"], "fno-sanitize-recover">, Flags<[CoreOption, DriverOption]>, diff --git a/include/clang/Driver/SanitizerArgs.h b/include/clang/Driver/SanitizerArgs.h index c7b3e8006dd5..a9645d463fa1 100644 --- a/include/clang/Driver/SanitizerArgs.h +++ b/include/clang/Driver/SanitizerArgs.h @@ -35,6 +35,7 @@ class SanitizerArgs { int AsanFieldPadding = 0; bool AsanSharedRuntime = false; bool AsanUseAfterScope = true; + bool AsanGlobalsDeadStripping = false; bool LinkCXXRuntimes = false; bool NeedPIE = false; bool Stats = false; diff --git a/include/clang/Format/Format.h b/include/clang/Format/Format.h index 9bed253baca2..a963c6369aa9 100644 --- a/include/clang/Format/Format.h +++ b/include/clang/Format/Format.h @@ -98,22 +98,39 @@ struct FormatStyle { /// \endcode bool AlignConsecutiveDeclarations; - /// \brief If ``true``, aligns escaped newlines as far left as possible. - /// Otherwise puts them into the right-most column. - /// \code - /// true: - /// #define A \ - /// int aaaa; \ - /// int b; \ - /// int dddddddddd; - /// - /// false: - /// #define A \ - /// int aaaa; \ - /// int b; \ - /// int dddddddddd; - /// \endcode - bool AlignEscapedNewlinesLeft; + /// \brief Different styles for aligning escaped newlines. + enum EscapedNewlineAlignmentStyle { + /// \brief Don't align escaped newlines. + /// \code + /// #define A \ + /// int aaaa; \ + /// int b; \ + /// int dddddddddd; + /// \endcode + ENAS_DontAlign, + /// \brief Align escaped newlines as far left as possible. + /// \code + /// true: + /// #define A \ + /// int aaaa; \ + /// int b; \ + /// int dddddddddd; + /// + /// false: + /// \endcode + ENAS_Left, + /// \brief Align escaped newlines in the right-most column. + /// \code + /// #define A \ + /// int aaaa; \ + /// int b; \ + /// int dddddddddd; + /// \endcode + ENAS_Right, + }; + + /// \brief Options for aligning backslashes in escaped newlines. + EscapedNewlineAlignmentStyle AlignEscapedNewlines; /// \brief If ``true``, horizontally align operands of binary and ternary /// expressions. @@ -1347,7 +1364,7 @@ struct FormatStyle { AlignAfterOpenBracket == R.AlignAfterOpenBracket && AlignConsecutiveAssignments == R.AlignConsecutiveAssignments && AlignConsecutiveDeclarations == R.AlignConsecutiveDeclarations && - AlignEscapedNewlinesLeft == R.AlignEscapedNewlinesLeft && + AlignEscapedNewlines == R.AlignEscapedNewlines && AlignOperands == R.AlignOperands && AlignTrailingComments == R.AlignTrailingComments && AllowAllParametersOfDeclarationOnNextLine == diff --git a/include/clang/Frontend/CodeGenOptions.def b/include/clang/Frontend/CodeGenOptions.def index 7495ad808c99..251441d38ff8 100644 --- a/include/clang/Frontend/CodeGenOptions.def +++ b/include/clang/Frontend/CodeGenOptions.def @@ -137,6 +137,8 @@ CODEGENOPT(StructPathTBAA , 1, 0) ///< Whether or not to use struct-path TBAA CODEGENOPT(SaveTempLabels , 1, 0) ///< Save temporary labels. CODEGENOPT(SanitizeAddressUseAfterScope , 1, 0) ///< Enable use-after-scope detection ///< in AddressSanitizer +CODEGENOPT(SanitizeAddressGlobalsDeadStripping, 1, 0) ///< Enable linker dead stripping + ///< of globals in AddressSanitizer CODEGENOPT(SanitizeMemoryTrackOrigins, 2, 0) ///< Enable tracking origins in ///< MemorySanitizer CODEGENOPT(SanitizeMemoryUseAfterDtor, 1, 0) ///< Enable use-after-delete detection diff --git a/include/clang/Lex/MacroInfo.h b/include/clang/Lex/MacroInfo.h index 44b7b2e4a474..7da1e7b41ab8 100644 --- a/include/clang/Lex/MacroInfo.h +++ b/include/clang/Lex/MacroInfo.h @@ -105,9 +105,6 @@ class MacroInfo { /// \brief Must warn if the macro is unused at the end of translation unit. bool IsWarnIfUnused : 1; - /// \brief Whether this macro info was loaded from an AST file. - bool FromASTFile : 1; - /// \brief Whether this macro was used as header guard. bool UsedForHeaderGuard : 1; @@ -264,34 +261,16 @@ class MacroInfo { IsDisabled = true; } - /// \brief Determine whether this macro info came from an AST file (such as - /// a precompiled header or module) rather than having been parsed. - bool isFromASTFile() const { return FromASTFile; } - /// \brief Determine whether this macro was used for a header guard. bool isUsedForHeaderGuard() const { return UsedForHeaderGuard; } void setUsedForHeaderGuard(bool Val) { UsedForHeaderGuard = Val; } - /// \brief Retrieve the global ID of the module that owns this particular - /// macro info. - unsigned getOwningModuleID() const { - if (isFromASTFile()) - return *(const unsigned *)(this + 1); - - return 0; - } - void dump() const; private: unsigned getDefinitionLengthSlow(const SourceManager &SM) const; - void setOwningModuleID(unsigned ID) { - assert(isFromASTFile()); - *(unsigned *)(this + 1) = ID; - } - friend class Preprocessor; }; diff --git a/include/clang/Lex/Preprocessor.h b/include/clang/Lex/Preprocessor.h index 0e3f563785d4..114bf70ad59a 100644 --- a/include/clang/Lex/Preprocessor.h +++ b/include/clang/Lex/Preprocessor.h @@ -644,14 +644,6 @@ class Preprocessor { /// of that list. MacroInfoChain *MIChainHead; - struct DeserializedMacroInfoChain { - MacroInfo MI; - unsigned OwningModuleID; // MUST be immediately after the MacroInfo object - // so it can be accessed by MacroInfo::getOwningModuleID(). - DeserializedMacroInfoChain *Next; - }; - DeserializedMacroInfoChain *DeserialMIChainHead; - void updateOutOfDateIdentifier(IdentifierInfo &II) const; public: @@ -1669,10 +1661,6 @@ class Preprocessor { /// \brief Allocate a new MacroInfo object with the provided SourceLocation. MacroInfo *AllocateMacroInfo(SourceLocation L); - /// \brief Allocate a new MacroInfo object loaded from an AST file. - MacroInfo *AllocateDeserializedMacroInfo(SourceLocation L, - unsigned SubModuleID); - /// \brief Turn the specified lexer token into a fully checked and spelled /// filename, e.g. as an operand of \#include. /// @@ -1764,9 +1752,6 @@ class Preprocessor { /// macro name. void updateModuleMacroInfo(const IdentifierInfo *II, ModuleMacroInfo &Info); - /// \brief Allocate a new MacroInfo object. - MacroInfo *AllocateMacroInfo(); - DefMacroDirective *AllocateDefMacroDirective(MacroInfo *MI, SourceLocation Loc); UndefMacroDirective *AllocateUndefMacroDirective(SourceLocation UndefLoc); diff --git a/include/clang/Parse/Parser.h b/include/clang/Parse/Parser.h index 8d0935dec1b6..f5a7e02941a7 100644 --- a/include/clang/Parse/Parser.h +++ b/include/clang/Parse/Parser.h @@ -1488,6 +1488,8 @@ class Parser : public CodeCompletionHandler { K == tok::plusplus || K == tok::minusminus); } + bool diagnoseUnknownTemplateId(ExprResult TemplateName, SourceLocation Less); + ExprResult ParsePostfixExpressionSuffix(ExprResult LHS); ExprResult ParseUnaryExprOrTypeTraitExpression(); ExprResult ParseBuiltinPrimaryExpression(); @@ -2723,10 +2725,7 @@ class Parser : public CodeCompletionHandler { bool ParseGreaterThanInTemplateList(SourceLocation &RAngleLoc, bool ConsumeLastToken, bool ObjCGenericList); - bool ParseTemplateIdAfterTemplateName(TemplateTy Template, - SourceLocation TemplateNameLoc, - const CXXScopeSpec &SS, - bool ConsumeLastToken, + bool ParseTemplateIdAfterTemplateName(bool ConsumeLastToken, SourceLocation &LAngleLoc, TemplateArgList &TemplateArgs, SourceLocation &RAngleLoc); diff --git a/include/clang/Sema/Sema.h b/include/clang/Sema/Sema.h index e5961079f7c2..e910be14f969 100644 --- a/include/clang/Sema/Sema.h +++ b/include/clang/Sema/Sema.h @@ -1074,6 +1074,10 @@ class Sema { /// correctly named definition after the renamed definition. llvm::SmallPtrSet TypoCorrectedFunctionDefinitions; + /// Stack of types that correspond to the parameter entities that are + /// currently being copy-initialized. Can be empty. + llvm::SmallVector CurrentParameterCopyTypes; + void ReadMethodPool(Selector Sel); void updateOutOfDateSelector(Selector Sel); @@ -1456,6 +1460,11 @@ class Sema { /// The modules we're currently parsing. llvm::SmallVector ModuleScopes; + /// Get the module whose scope we are currently within. + Module *getCurrentModule() const { + return ModuleScopes.empty() ? nullptr : ModuleScopes.back().Module; + } + VisibleModuleSet VisibleModules; Module *CachedFakeTopLevelModule; @@ -1466,7 +1475,7 @@ class Sema { /// \brief Make a merged definition of an existing hidden definition \p ND /// visible at the specified location. - void makeMergedDefinitionVisible(NamedDecl *ND, SourceLocation Loc); + void makeMergedDefinitionVisible(NamedDecl *ND); bool isModuleVisible(Module *M) { return VisibleModules.isVisible(M); } @@ -1593,7 +1602,7 @@ class Sema { Scope *S, CXXScopeSpec *SS, ParsedType &SuggestedType, - bool AllowClassTemplates = false); + bool IsTemplateName = false); /// Attempt to behave like MSVC in situations where lookup of an unqualified /// type name has failed in a dependent context. In these situations, we @@ -1738,6 +1747,23 @@ class Sema { TemplateNameKindForDiagnostics getTemplateNameKindForDiagnostics(TemplateName Name); + /// Determine whether it's plausible that E was intended to be a + /// template-name. + bool mightBeIntendedToBeTemplateName(ExprResult E) { + if (!getLangOpts().CPlusPlus || E.isInvalid()) + return false; + if (auto *DRE = dyn_cast(E.get())) + return !DRE->hasExplicitTemplateArgs(); + if (auto *ME = dyn_cast(E.get())) + return !ME->hasExplicitTemplateArgs(); + // Any additional cases recognized here should also be handled by + // diagnoseExprIntendedAsTemplateName. + return false; + } + void diagnoseExprIntendedAsTemplateName(Scope *S, ExprResult TemplateName, + SourceLocation Less, + SourceLocation Greater); + Decl *ActOnDeclarator(Scope *S, Declarator &D); NamedDecl *HandleDeclarator(Scope *S, Declarator &D, @@ -2336,6 +2362,7 @@ class Sema { void MergeVarDeclTypes(VarDecl *New, VarDecl *Old, bool MergeTypeWithOld); void MergeVarDeclExceptionSpecs(VarDecl *New, VarDecl *Old); bool checkVarDeclRedefinition(VarDecl *OldDefn, VarDecl *NewDefn); + void notePreviousDefinition(SourceLocation Old, SourceLocation New); bool MergeCXXFunctionDecl(FunctionDecl *New, FunctionDecl *Old, Scope *S); // AssignmentAction - This is used by all the assignment diagnostic functions @@ -2726,7 +2753,8 @@ class Sema { resolveAddressOfOnlyViableOverloadCandidate(Expr *E, DeclAccessPair &FoundResult); - bool resolveAndFixAddressOfOnlyViableOverloadCandidate(ExprResult &SrcExpr); + bool resolveAndFixAddressOfOnlyViableOverloadCandidate( + ExprResult &SrcExpr, bool DoFunctionPointerConversion = false); FunctionDecl * ResolveSingleFunctionTemplateSpecialization(OverloadExpr *ovl, @@ -3049,7 +3077,8 @@ class Sema { bool IncludeGlobalScope = true); void LookupVisibleDecls(DeclContext *Ctx, LookupNameKind Kind, VisibleDeclConsumer &Consumer, - bool IncludeGlobalScope = true); + bool IncludeGlobalScope = true, + bool IncludeDependentBases = false); enum CorrectTypoKind { CTK_NonError, // CorrectTypo used in a non error recovery situation. @@ -6084,6 +6113,7 @@ class Sema { TemplateArgumentListInfo *ExplicitTemplateArgs, LookupResult &Previous); bool CheckMemberSpecialization(NamedDecl *Member, LookupResult &Previous); + void CompleteMemberSpecialization(NamedDecl *Member, LookupResult &Previous); DeclResult ActOnExplicitInstantiation(Scope *S, @@ -9259,6 +9289,8 @@ class Sema { /// type checking binary operators (subroutines of CreateBuiltinBinOp). QualType InvalidOperands(SourceLocation Loc, ExprResult &LHS, ExprResult &RHS); + QualType InvalidLogicalVectorOperands(SourceLocation Loc, ExprResult &LHS, + ExprResult &RHS); QualType CheckPointerToMemberOperands( // C++ 5.5 ExprResult &LHS, ExprResult &RHS, ExprValueKind &VK, SourceLocation OpLoc, bool isIndirect); @@ -10008,6 +10040,7 @@ class Sema { MacroInfo *MacroInfo, unsigned Argument); void CodeCompleteNaturalLanguage(); + void CodeCompleteAvailabilityPlatformName(); void GatherGlobalCodeCompletions(CodeCompletionAllocator &Allocator, CodeCompletionTUInfo &CCTUInfo, SmallVectorImpl &Results); diff --git a/include/clang/Tooling/RefactoringCallbacks.h b/include/clang/Tooling/RefactoringCallbacks.h index 6ef9ea11f0ae..9862951149a3 100644 --- a/include/clang/Tooling/RefactoringCallbacks.h +++ b/include/clang/Tooling/RefactoringCallbacks.h @@ -47,6 +47,33 @@ class RefactoringCallback : public ast_matchers::MatchFinder::MatchCallback { Replacements Replace; }; +/// \brief Adaptor between \c ast_matchers::MatchFinder and \c +/// tooling::RefactoringTool. +/// +/// Runs AST matchers and stores the \c tooling::Replacements in a map. +class ASTMatchRefactorer { +public: + explicit ASTMatchRefactorer( + std::map &FileToReplaces); + + template + void addMatcher(const T &Matcher, RefactoringCallback *Callback) { + MatchFinder.addMatcher(Matcher, Callback); + Callbacks.push_back(Callback); + } + + void addDynamicMatcher(const ast_matchers::internal::DynTypedMatcher &Matcher, + RefactoringCallback *Callback); + + std::unique_ptr newASTConsumer(); + +private: + friend class RefactoringASTConsumer; + std::vector Callbacks; + ast_matchers::MatchFinder MatchFinder; + std::map &FileToReplaces; +}; + /// \brief Replace the text of the statement bound to \c FromId with the text in /// \c ToText. class ReplaceStmtWithText : public RefactoringCallback { @@ -59,6 +86,29 @@ class ReplaceStmtWithText : public RefactoringCallback { std::string ToText; }; +/// \brief Replace the text of an AST node bound to \c FromId with the result of +/// evaluating the template in \c ToTemplate. +/// +/// Expressions of the form ${NodeName} in \c ToTemplate will be +/// replaced by the text of the node bound to ${NodeName}. The string +/// "$$" will be replaced by "$". +class ReplaceNodeWithTemplate : public RefactoringCallback { +public: + static llvm::Expected> + create(StringRef FromId, StringRef ToTemplate); + void run(const ast_matchers::MatchFinder::MatchResult &Result) override; + +private: + struct TemplateElement { + enum { Literal, Identifier } Type; + std::string Value; + }; + ReplaceNodeWithTemplate(llvm::StringRef FromId, + std::vector Template); + std::string FromId; + std::vector Template; +}; + /// \brief Replace the text of the statement bound to \c FromId with the text of /// the statement bound to \c ToId. class ReplaceStmtWithStmt : public RefactoringCallback { diff --git a/lib/AST/ASTImporter.cpp b/lib/AST/ASTImporter.cpp index 4fb6051d6f58..847638b7bbeb 100644 --- a/lib/AST/ASTImporter.cpp +++ b/lib/AST/ASTImporter.cpp @@ -1622,10 +1622,18 @@ Decl *ASTNodeImporter::VisitRecordDecl(RecordDecl *D) { // We may already have a record of the same name; try to find and match it. RecordDecl *AdoptDecl = nullptr; + RecordDecl *PrevDecl = nullptr; if (!DC->isFunctionOrMethod()) { SmallVector ConflictingDecls; SmallVector FoundDecls; DC->getRedeclContext()->localUncachedLookup(SearchName, FoundDecls); + + if (!FoundDecls.empty()) { + // We're going to have to compare D against potentially conflicting Decls, so complete it. + if (D->hasExternalLexicalStorage() && !D->isCompleteDefinition()) + D->getASTContext().getExternalSource()->CompleteType(D); + } + for (unsigned I = 0, N = FoundDecls.size(); I != N; ++I) { if (!FoundDecls[I]->isInIdentifierNamespace(IDNS)) continue; @@ -1652,6 +1660,8 @@ Decl *ASTNodeImporter::VisitRecordDecl(RecordDecl *D) { } } + PrevDecl = FoundRecord; + if (RecordDecl *FoundDef = FoundRecord->getDefinition()) { if ((SearchName && !D->isCompleteDefinition()) || (D->isCompleteDefinition() && @@ -1744,6 +1754,10 @@ Decl *ASTNodeImporter::VisitRecordDecl(RecordDecl *D) { LexicalDC->addDeclInternal(D2); if (D->isAnonymousStructOrUnion()) D2->setAnonymousStructOrUnion(true); + if (PrevDecl) { + // FIXME: do this for all Redeclarables, not just RecordDecls. + D2->setPreviousDecl(PrevDecl); + } } Importer.Imported(D, D2); diff --git a/lib/AST/ASTStructuralEquivalence.cpp b/lib/AST/ASTStructuralEquivalence.cpp index 8fe72eac4133..9376ee1d4ee4 100644 --- a/lib/AST/ASTStructuralEquivalence.cpp +++ b/lib/AST/ASTStructuralEquivalence.cpp @@ -855,6 +855,11 @@ static bool IsStructurallyEquivalent(StructuralEquivalenceContext &Context, if (CXXRecordDecl *D1CXX = dyn_cast(D1)) { if (CXXRecordDecl *D2CXX = dyn_cast(D2)) { + if (D1CXX->hasExternalLexicalStorage() && + !D1CXX->isCompleteDefinition()) { + D1CXX->getASTContext().getExternalSource()->CompleteType(D1CXX); + } + if (D1CXX->getNumBases() != D2CXX->getNumBases()) { if (Context.Complain) { Context.Diag2(D2->getLocation(), diag::warn_odr_tag_type_inconsistent) diff --git a/lib/AST/CXXInheritance.cpp b/lib/AST/CXXInheritance.cpp index 56fb0464078f..746602d47be5 100644 --- a/lib/AST/CXXInheritance.cpp +++ b/lib/AST/CXXInheritance.cpp @@ -13,6 +13,7 @@ #include "clang/AST/CXXInheritance.h" #include "clang/AST/ASTContext.h" #include "clang/AST/DeclCXX.h" +#include "clang/AST/DeclTemplate.h" #include "clang/AST/RecordLayout.h" #include "llvm/ADT/SetVector.h" #include @@ -174,9 +175,10 @@ bool CXXRecordDecl::forallBases(ForallBasesCallback BaseMatches, return AllMatches; } -bool CXXBasePaths::lookupInBases( - ASTContext &Context, const CXXRecordDecl *Record, - CXXRecordDecl::BaseMatchesCallback BaseMatches) { +bool CXXBasePaths::lookupInBases(ASTContext &Context, + const CXXRecordDecl *Record, + CXXRecordDecl::BaseMatchesCallback BaseMatches, + bool LookupInDependent) { bool FoundPath = false; // The access of the path down to this record. @@ -194,7 +196,7 @@ bool CXXBasePaths::lookupInBases( // the base class scope is not examined during unqualified name lookup // either at the point of definition of the class template or member or // during an instantiation of the class tem- plate or member. - if (BaseType->isDependentType()) + if (!LookupInDependent && BaseType->isDependentType()) continue; // Determine whether we need to visit this base class at all, @@ -262,10 +264,28 @@ bool CXXBasePaths::lookupInBases( return FoundPath; } } else if (VisitBase) { - CXXRecordDecl *BaseRecord - = cast(BaseSpec.getType()->castAs() - ->getDecl()); - if (lookupInBases(Context, BaseRecord, BaseMatches)) { + CXXRecordDecl *BaseRecord; + if (LookupInDependent) { + BaseRecord = nullptr; + const TemplateSpecializationType *TST = + BaseSpec.getType()->getAs(); + if (!TST) { + if (auto *RT = BaseSpec.getType()->getAs()) + BaseRecord = cast(RT->getDecl()); + } else { + TemplateName TN = TST->getTemplateName(); + if (auto *TD = + dyn_cast_or_null(TN.getAsTemplateDecl())) + BaseRecord = TD->getTemplatedDecl(); + } + if (BaseRecord && !BaseRecord->hasDefinition()) + BaseRecord = nullptr; + } else { + BaseRecord = cast( + BaseSpec.getType()->castAs()->getDecl()); + } + if (BaseRecord && + lookupInBases(Context, BaseRecord, BaseMatches, LookupInDependent)) { // C++ [class.member.lookup]p2: // A member name f in one sub-object B hides a member name f in // a sub-object A if A is a base class sub-object of B. Any @@ -299,9 +319,11 @@ bool CXXBasePaths::lookupInBases( } bool CXXRecordDecl::lookupInBases(BaseMatchesCallback BaseMatches, - CXXBasePaths &Paths) const { + CXXBasePaths &Paths, + bool LookupInDependent) const { // If we didn't find anything, report that. - if (!Paths.lookupInBases(getASTContext(), this, BaseMatches)) + if (!Paths.lookupInBases(getASTContext(), this, BaseMatches, + LookupInDependent)) return false; // If we're not recording paths or we won't ever find ambiguities, @@ -387,23 +409,49 @@ bool CXXRecordDecl::FindTagMember(const CXXBaseSpecifier *Specifier, return false; } -bool CXXRecordDecl::FindOrdinaryMember(const CXXBaseSpecifier *Specifier, - CXXBasePath &Path, - DeclarationName Name) { - RecordDecl *BaseRecord = - Specifier->getType()->castAs()->getDecl(); - - const unsigned IDNS = IDNS_Ordinary | IDNS_Tag | IDNS_Member; +static bool findOrdinaryMember(RecordDecl *BaseRecord, CXXBasePath &Path, + DeclarationName Name) { + const unsigned IDNS = clang::Decl::IDNS_Ordinary | clang::Decl::IDNS_Tag | + clang::Decl::IDNS_Member; for (Path.Decls = BaseRecord->lookup(Name); !Path.Decls.empty(); Path.Decls = Path.Decls.slice(1)) { if (Path.Decls.front()->isInIdentifierNamespace(IDNS)) return true; } - + return false; } +bool CXXRecordDecl::FindOrdinaryMember(const CXXBaseSpecifier *Specifier, + CXXBasePath &Path, + DeclarationName Name) { + RecordDecl *BaseRecord = + Specifier->getType()->castAs()->getDecl(); + return findOrdinaryMember(BaseRecord, Path, Name); +} + +bool CXXRecordDecl::FindOrdinaryMemberInDependentClasses( + const CXXBaseSpecifier *Specifier, CXXBasePath &Path, + DeclarationName Name) { + const TemplateSpecializationType *TST = + Specifier->getType()->getAs(); + if (!TST) { + auto *RT = Specifier->getType()->getAs(); + if (!RT) + return false; + return findOrdinaryMember(RT->getDecl(), Path, Name); + } + TemplateName TN = TST->getTemplateName(); + const auto *TD = dyn_cast_or_null(TN.getAsTemplateDecl()); + if (!TD) + return false; + CXXRecordDecl *RD = TD->getTemplatedDecl(); + if (!RD) + return false; + return findOrdinaryMember(RD, Path, Name); +} + bool CXXRecordDecl::FindOMPReductionMember(const CXXBaseSpecifier *Specifier, CXXBasePath &Path, DeclarationName Name) { @@ -438,6 +486,36 @@ FindNestedNameSpecifierMember(const CXXBaseSpecifier *Specifier, return false; } +std::vector CXXRecordDecl::lookupDependentName( + const DeclarationName &Name, + llvm::function_ref Filter) { + std::vector Results; + // Lookup in the class. + DeclContext::lookup_result DirectResult = lookup(Name); + if (!DirectResult.empty()) { + for (const NamedDecl *ND : DirectResult) { + if (Filter(ND)) + Results.push_back(ND); + } + return Results; + } + // Perform lookup into our base classes. + CXXBasePaths Paths; + Paths.setOrigin(this); + if (!lookupInBases( + [&](const CXXBaseSpecifier *Specifier, CXXBasePath &Path) { + return CXXRecordDecl::FindOrdinaryMemberInDependentClasses( + Specifier, Path, Name); + }, + Paths, /*LookupInDependent=*/true)) + return Results; + for (const NamedDecl *ND : Paths.front().Decls) { + if (Filter(ND)) + Results.push_back(ND); + } + return Results; +} + void OverridingMethods::add(unsigned OverriddenSubobject, UniqueVirtualMethod Overriding) { SmallVectorImpl &SubobjectOverrides diff --git a/lib/AST/DeclCXX.cpp b/lib/AST/DeclCXX.cpp index dd8f768c5711..9f87fe12a9cd 100644 --- a/lib/AST/DeclCXX.cpp +++ b/lib/AST/DeclCXX.cpp @@ -1432,8 +1432,9 @@ bool CXXRecordDecl::isAnyDestructorNoReturn() const { // Check base classes destructor for noreturn. for (const auto &Base : bases()) - if (Base.getType()->getAsCXXRecordDecl()->isAnyDestructorNoReturn()) - return true; + if (const CXXRecordDecl *RD = Base.getType()->getAsCXXRecordDecl()) + if (RD->isAnyDestructorNoReturn()) + return true; // Check fields for noreturn. for (const auto *Field : fields()) diff --git a/lib/AST/ExternalASTMerger.cpp b/lib/AST/ExternalASTMerger.cpp index 8849cfc3c80b..1dc472a5f753 100644 --- a/lib/AST/ExternalASTMerger.cpp +++ b/lib/AST/ExternalASTMerger.cpp @@ -178,3 +178,9 @@ void ExternalASTMerger::FindExternalLexicalDecls( } }); } + +void ExternalASTMerger::CompleteType(TagDecl *Tag) { + SmallVector Result; + FindExternalLexicalDecls(Tag, [](Decl::Kind) { return true; }, Result); + Tag->setHasExternalLexicalStorage(false); +} diff --git a/lib/AST/ODRHash.cpp b/lib/AST/ODRHash.cpp index 83168d0924f6..f4d314a6dd0d 100644 --- a/lib/AST/ODRHash.cpp +++ b/lib/AST/ODRHash.cpp @@ -411,7 +411,7 @@ class ODRTypeVisitor : public TypeVisitor { void VisitTypedefType(const TypedefType *T) { AddDecl(T->getDecl()); - Hash.AddQualType(T->getDecl()->getUnderlyingType()); + AddQualType(T->getDecl()->getUnderlyingType().getCanonicalType()); VisitType(T); } }; diff --git a/lib/AST/Stmt.cpp b/lib/AST/Stmt.cpp index 69e65f558f89..2367cadf645c 100644 --- a/lib/AST/Stmt.cpp +++ b/lib/AST/Stmt.cpp @@ -1112,7 +1112,7 @@ void CapturedStmt::setCapturedRegionKind(CapturedRegionKind Kind) { bool CapturedStmt::capturesVariable(const VarDecl *Var) const { for (const auto &I : captures()) { - if (!I.capturesVariable()) + if (!I.capturesVariable() && !I.capturesVariableByCopy()) continue; // This does not handle variable redeclarations. This should be diff --git a/lib/AST/Type.cpp b/lib/AST/Type.cpp index df26233b4796..22d52bcd3f31 100644 --- a/lib/AST/Type.cpp +++ b/lib/AST/Type.cpp @@ -2114,18 +2114,15 @@ bool QualType::isTriviallyCopyableType(const ASTContext &Context) const { if (hasNonTrivialObjCLifetime()) return false; - // C++11 [basic.types]p9 + // C++11 [basic.types]p9 - See Core 2094 // Scalar types, trivially copyable class types, arrays of such types, and - // non-volatile const-qualified versions of these types are collectively + // cv-qualified versions of these types are collectively // called trivially copyable types. QualType CanonicalType = getCanonicalType(); if (CanonicalType->isDependentType()) return false; - if (CanonicalType.isVolatileQualified()) - return false; - // Return false for incomplete types after skipping any incomplete array types // which are expressly allowed by the standard and thus our API. if (CanonicalType->isIncompleteType()) diff --git a/lib/Basic/Targets.cpp b/lib/Basic/Targets.cpp index 33eb0b05ddcd..92c561aa9413 100644 --- a/lib/Basic/Targets.cpp +++ b/lib/Basic/Targets.cpp @@ -6862,6 +6862,11 @@ class SparcTargetInfo : public TargetInfo { case 'N': // Same as 'K' but zext (required for SIMode) case 'O': // The constant 4096 return true; + + case 'f': + case 'e': + info.setAllowsRegister(); + return true; } return false; } diff --git a/lib/CodeGen/BackendUtil.cpp b/lib/CodeGen/BackendUtil.cpp index 0d96f2efa60a..0f07169ac8b0 100644 --- a/lib/CodeGen/BackendUtil.cpp +++ b/lib/CodeGen/BackendUtil.cpp @@ -194,6 +194,8 @@ static void addSanitizerCoveragePass(const PassManagerBuilder &Builder, // where this is not a factor). Also, on ELF this feature requires an assembler // extension that only works with -integrated-as at the moment. static bool asanUseGlobalsGC(const Triple &T, const CodeGenOptions &CGOpts) { + if (!CGOpts.SanitizeAddressGlobalsDeadStripping) + return false; switch (T.getObjectFormat()) { case Triple::MachO: case Triple::COFF: @@ -1071,7 +1073,8 @@ void clang::EmitBackendOutput(DiagnosticsEngine &Diags, // into memory and pass it into runThinLTOBackend, which will run the // function importer and invoke LTO passes. Expected> IndexOrErr = - llvm::getModuleSummaryIndexForFile(CGOpts.ThinLTOIndexFile); + llvm::getModuleSummaryIndexForFile(CGOpts.ThinLTOIndexFile, + /*IgnoreEmptyThinLTOIndexFile*/true); if (!IndexOrErr) { logAllUnhandledErrors(IndexOrErr.takeError(), errs(), "Error loading index file '" + diff --git a/lib/CodeGen/CGBlocks.cpp b/lib/CodeGen/CGBlocks.cpp index 2b2a92dd6019..f1c20e9df1f3 100644 --- a/lib/CodeGen/CGBlocks.cpp +++ b/lib/CodeGen/CGBlocks.cpp @@ -878,7 +878,8 @@ llvm::Value *CodeGenFunction::EmitBlockLiteral(const CGBlockInfo &blockInfo) { // If type is const-qualified, copy the value into the block field. } else if (type.isConstQualified() && - type.getObjCLifetime() == Qualifiers::OCL_Strong) { + type.getObjCLifetime() == Qualifiers::OCL_Strong && + CGM.getCodeGenOpts().OptimizationLevel != 0) { llvm::Value *value = Builder.CreateLoad(src, "captured"); Builder.CreateStore(value, blockField); @@ -960,9 +961,8 @@ llvm::Type *CodeGenModule::getBlockDescriptorType() { // const char *signature; // the block signature // const char *layout; // reserved // }; - BlockDescriptorType = - llvm::StructType::create("struct.__block_descriptor", - UnsignedLongTy, UnsignedLongTy, nullptr); + BlockDescriptorType = llvm::StructType::create( + "struct.__block_descriptor", UnsignedLongTy, UnsignedLongTy); // Now form a pointer to that. unsigned AddrSpace = 0; @@ -986,9 +986,8 @@ llvm::Type *CodeGenModule::getGenericBlockLiteralType() { // struct __block_descriptor *__descriptor; // }; GenericBlockLiteralType = - llvm::StructType::create("struct.__block_literal_generic", - VoidPtrTy, IntTy, IntTy, VoidPtrTy, - BlockDescPtrTy, nullptr); + llvm::StructType::create("struct.__block_literal_generic", VoidPtrTy, + IntTy, IntTy, VoidPtrTy, BlockDescPtrTy); return GenericBlockLiteralType; } diff --git a/lib/CodeGen/CGBuiltin.cpp b/lib/CodeGen/CGBuiltin.cpp index 2f05c0e910e5..50c9e22801c7 100644 --- a/lib/CodeGen/CGBuiltin.cpp +++ b/lib/CodeGen/CGBuiltin.cpp @@ -2769,6 +2769,32 @@ RValue CodeGenFunction::EmitBuiltinExpr(const FunctionDecl *FD, return RValue::get(ConstantInt::get(ConvertType(E->getType()), Layout.size().getQuantity())); } + + case Builtin::BI__xray_customevent: { + if (!ShouldXRayInstrumentFunction()) + return RValue::getIgnored(); + if (const auto *XRayAttr = CurFuncDecl->getAttr()) { + if (XRayAttr->neverXRayInstrument()) + return RValue::getIgnored(); + } + Function *F = CGM.getIntrinsic(Intrinsic::xray_customevent); + auto FTy = F->getFunctionType(); + auto Arg0 = E->getArg(0); + auto Arg0Val = EmitScalarExpr(Arg0); + auto Arg0Ty = Arg0->getType(); + auto PTy0 = FTy->getParamType(0); + if (PTy0 != Arg0Val->getType()) { + if (Arg0Ty->isArrayType()) + Arg0Val = EmitArrayToPointerDecay(Arg0).getPointer(); + else + Arg0Val = Builder.CreatePointerCast(Arg0Val, PTy0); + } + auto Arg1 = EmitScalarExpr(E->getArg(1)); + auto PTy1 = FTy->getParamType(1); + if (PTy1 != Arg1->getType()) + Arg1 = Builder.CreateTruncOrBitCast(Arg1, PTy1); + return RValue::get(Builder.CreateCall(F, {Arg0Val, Arg1})); + } } // If this is an alias for a lib function (e.g. __builtin_sin), emit @@ -4545,7 +4571,7 @@ Value *CodeGenFunction::EmitARMBuiltinExpr(unsigned BuiltinID, Function *F = CGM.getIntrinsic(BuiltinID == ARM::BI__builtin_arm_stlex ? Intrinsic::arm_stlexd : Intrinsic::arm_strexd); - llvm::Type *STy = llvm::StructType::get(Int32Ty, Int32Ty, nullptr); + llvm::Type *STy = llvm::StructType::get(Int32Ty, Int32Ty); Address Tmp = CreateMemTemp(E->getArg(0)->getType()); Value *Val = EmitScalarExpr(E->getArg(0)); @@ -5375,7 +5401,7 @@ Value *CodeGenFunction::EmitAArch64BuiltinExpr(unsigned BuiltinID, Function *F = CGM.getIntrinsic(BuiltinID == AArch64::BI__builtin_arm_stlex ? Intrinsic::aarch64_stlxp : Intrinsic::aarch64_stxp); - llvm::Type *STy = llvm::StructType::get(Int64Ty, Int64Ty, nullptr); + llvm::Type *STy = llvm::StructType::get(Int64Ty, Int64Ty); Address Tmp = CreateMemTemp(E->getArg(0)->getType()); EmitAnyExprToMem(E->getArg(0), Tmp, Qualifiers(), /*init*/ true); @@ -7347,8 +7373,8 @@ Value *CodeGenFunction::EmitX86BuiltinExpr(unsigned BuiltinID, // unsigned int __cpu_type; // unsigned int __cpu_subtype; // unsigned int __cpu_features[1]; - llvm::Type *STy = llvm::StructType::get( - Int32Ty, Int32Ty, Int32Ty, llvm::ArrayType::get(Int32Ty, 1), nullptr); + llvm::Type *STy = llvm::StructType::get(Int32Ty, Int32Ty, Int32Ty, + llvm::ArrayType::get(Int32Ty, 1)); // Grab the global __cpu_model. llvm::Constant *CpuModel = CGM.CreateRuntimeVariable(STy, "__cpu_model"); diff --git a/lib/CodeGen/CGCUDANV.cpp b/lib/CodeGen/CGCUDANV.cpp index 813cd7400186..d24ef0a8a974 100644 --- a/lib/CodeGen/CGCUDANV.cpp +++ b/lib/CodeGen/CGCUDANV.cpp @@ -265,7 +265,7 @@ llvm::Function *CGNVCUDARuntime::makeModuleCtorFunction() { "__cudaRegisterFatBinary"); // struct { int magic, int version, void * gpu_binary, void * dont_care }; llvm::StructType *FatbinWrapperTy = - llvm::StructType::get(IntTy, IntTy, VoidPtrTy, VoidPtrTy, nullptr); + llvm::StructType::get(IntTy, IntTy, VoidPtrTy, VoidPtrTy); llvm::Function *ModuleCtorFunc = llvm::Function::Create( llvm::FunctionType::get(VoidTy, VoidPtrTy, false), diff --git a/lib/CodeGen/CGCleanup.cpp b/lib/CodeGen/CGCleanup.cpp index 437ab7dd4649..e8bcf0a3ac56 100644 --- a/lib/CodeGen/CGCleanup.cpp +++ b/lib/CodeGen/CGCleanup.cpp @@ -51,8 +51,7 @@ DominatingValue::saved_type::save(CodeGenFunction &CGF, RValue rv) { if (rv.isComplex()) { CodeGenFunction::ComplexPairTy V = rv.getComplexVal(); llvm::Type *ComplexTy = - llvm::StructType::get(V.first->getType(), V.second->getType(), - (void*) nullptr); + llvm::StructType::get(V.first->getType(), V.second->getType()); Address addr = CGF.CreateDefaultAlignTempAlloca(ComplexTy, "saved-complex"); CGF.Builder.CreateStore(V.first, CGF.Builder.CreateStructGEP(addr, 0, CharUnits())); diff --git a/lib/CodeGen/CGDebugInfo.cpp b/lib/CodeGen/CGDebugInfo.cpp index 3e5434660567..9d77c61bd52c 100644 --- a/lib/CodeGen/CGDebugInfo.cpp +++ b/lib/CodeGen/CGDebugInfo.cpp @@ -209,7 +209,7 @@ llvm::DIScope *CGDebugInfo::getContextDescriptor(const Decl *Context, // Check namespace. if (const auto *NSDecl = dyn_cast(Context)) - return getOrCreateNameSpace(NSDecl); + return getOrCreateNamespace(NSDecl); if (const auto *RDecl = dyn_cast(Context)) if (!RDecl->isDependentType()) @@ -2860,8 +2860,8 @@ void CGDebugInfo::collectFunctionDeclProps(GlobalDecl GD, llvm::DIFile *Unit, if (DebugKind >= codegenoptions::LimitedDebugInfo) { if (const NamespaceDecl *NSDecl = - dyn_cast_or_null(FD->getDeclContext())) - FDContext = getOrCreateNameSpace(NSDecl); + dyn_cast_or_null(FD->getLexicalDeclContext())) + FDContext = getOrCreateNamespace(NSDecl); else if (const RecordDecl *RDecl = dyn_cast_or_null(FD->getDeclContext())) { llvm::DIScope *Mod = getParentModuleOrNull(RDecl); @@ -3961,7 +3961,7 @@ void CGDebugInfo::EmitUsingDirective(const UsingDirectiveDecl &UD) { CGM.getCodeGenOpts().DebugExplicitImport) { DBuilder.createImportedModule( getCurrentContextDescriptor(cast(UD.getDeclContext())), - getOrCreateNameSpace(NSDecl), + getOrCreateNamespace(NSDecl), getLineNumber(UD.getLocation())); } } @@ -4021,23 +4021,26 @@ CGDebugInfo::EmitNamespaceAlias(const NamespaceAliasDecl &NA) { else R = DBuilder.createImportedDeclaration( getCurrentContextDescriptor(cast(NA.getDeclContext())), - getOrCreateNameSpace(cast(NA.getAliasedNamespace())), + getOrCreateNamespace(cast(NA.getAliasedNamespace())), getLineNumber(NA.getLocation()), NA.getName()); VH.reset(R); return R; } llvm::DINamespace * -CGDebugInfo::getOrCreateNameSpace(const NamespaceDecl *NSDecl) { - NSDecl = NSDecl->getCanonicalDecl(); - auto I = NameSpaceCache.find(NSDecl); - if (I != NameSpaceCache.end()) +CGDebugInfo::getOrCreateNamespace(const NamespaceDecl *NSDecl) { + // Don't canonicalize the NamespaceDecl here: The DINamespace will be uniqued + // if necessary, and this way multiple declarations of the same namespace in + // different parent modules stay distinct. + auto I = NamespaceCache.find(NSDecl); + if (I != NamespaceCache.end()) return cast(I->second); llvm::DIScope *Context = getDeclContextDescriptor(NSDecl); + // Don't trust the context if it is a DIModule (see comment above). llvm::DINamespace *NS = DBuilder.createNameSpace(Context, NSDecl->getName(), NSDecl->isInline()); - NameSpaceCache[NSDecl].reset(NS); + NamespaceCache[NSDecl].reset(NS); return NS; } diff --git a/lib/CodeGen/CGDebugInfo.h b/lib/CodeGen/CGDebugInfo.h index 5050ca0ad3fa..7de48f278994 100644 --- a/lib/CodeGen/CGDebugInfo.h +++ b/lib/CodeGen/CGDebugInfo.h @@ -125,7 +125,7 @@ class CGDebugInfo { /// Cache declarations relevant to DW_TAG_imported_declarations (C++ /// using declarations) that aren't covered by other more specific caches. llvm::DenseMap DeclCache; - llvm::DenseMap NameSpaceCache; + llvm::DenseMap NamespaceCache; llvm::DenseMap NamespaceAliasCache; llvm::DenseMap> @@ -194,8 +194,9 @@ class CGDebugInfo { getOrCreateFunctionType(const Decl *D, QualType FnType, llvm::DIFile *F); /// \return debug info descriptor for vtable. llvm::DIType *getOrCreateVTablePtrType(llvm::DIFile *F); + /// \return namespace descriptor for the given namespace decl. - llvm::DINamespace *getOrCreateNameSpace(const NamespaceDecl *N); + llvm::DINamespace *getOrCreateNamespace(const NamespaceDecl *N); llvm::DIType *CreatePointerLikeType(llvm::dwarf::Tag Tag, const Type *Ty, QualType PointeeTy, llvm::DIFile *F); llvm::DIType *getOrCreateStructPtrType(StringRef Name, llvm::DIType *&Cache); diff --git a/lib/CodeGen/CGDecl.cpp b/lib/CodeGen/CGDecl.cpp index 10a0b46d9028..0fa8eeb1c3e1 100644 --- a/lib/CodeGen/CGDecl.cpp +++ b/lib/CodeGen/CGDecl.cpp @@ -152,7 +152,14 @@ void CodeGenFunction::EmitDecl(const Decl &D) { /// EmitVarDecl - This method handles emission of any variable declaration /// inside a function, including static vars etc. void CodeGenFunction::EmitVarDecl(const VarDecl &D) { - if (D.isStaticLocal()) { + if (D.hasExternalStorage()) + // Don't emit it now, allow it to be emitted lazily on its first use. + return; + + // Some function-scope variable does not have static storage but still + // needs to be emitted like a static variable, e.g. a function-scope + // variable in constant address space in OpenCL. + if (D.getStorageDuration() != SD_Automatic) { llvm::GlobalValue::LinkageTypes Linkage = CGM.getLLVMLinkageVarDefinition(&D, /*isConstant=*/false); @@ -163,10 +170,6 @@ void CodeGenFunction::EmitVarDecl(const VarDecl &D) { return EmitStaticVarDecl(D, Linkage); } - if (D.hasExternalStorage()) - // Don't emit it now, allow it to be emitted lazily on its first use. - return; - if (D.getType().getAddressSpace() == LangAS::opencl_local) return CGM.getOpenCLRuntime().EmitWorkGroupLocalVarDecl(*this, D); diff --git a/lib/CodeGen/CGException.cpp b/lib/CodeGen/CGException.cpp index ca1535182ec1..e65fa863fe31 100644 --- a/lib/CodeGen/CGException.cpp +++ b/lib/CodeGen/CGException.cpp @@ -765,8 +765,8 @@ llvm::BasicBlock *CodeGenFunction::EmitLandingPad() { llvm::BasicBlock *lpad = createBasicBlock("lpad"); EmitBlock(lpad); - llvm::LandingPadInst *LPadInst = Builder.CreateLandingPad( - llvm::StructType::get(Int8PtrTy, Int32Ty, nullptr), 0); + llvm::LandingPadInst *LPadInst = + Builder.CreateLandingPad(llvm::StructType::get(Int8PtrTy, Int32Ty), 0); llvm::Value *LPadExn = Builder.CreateExtractValue(LPadInst, 0); Builder.CreateStore(LPadExn, getExceptionSlot()); @@ -1310,8 +1310,8 @@ llvm::BasicBlock *CodeGenFunction::getTerminateLandingPad() { if (!CurFn->hasPersonalityFn()) CurFn->setPersonalityFn(getOpaquePersonalityFn(CGM, Personality)); - llvm::LandingPadInst *LPadInst = Builder.CreateLandingPad( - llvm::StructType::get(Int8PtrTy, Int32Ty, nullptr), 0); + llvm::LandingPadInst *LPadInst = + Builder.CreateLandingPad(llvm::StructType::get(Int8PtrTy, Int32Ty), 0); LPadInst->addClause(getCatchAllValue(*this)); llvm::Value *Exn = nullptr; @@ -1387,8 +1387,7 @@ llvm::BasicBlock *CodeGenFunction::getEHResumeBlock(bool isCleanup) { llvm::Value *Exn = getExceptionFromSlot(); llvm::Value *Sel = getSelectorFromSlot(); - llvm::Type *LPadType = llvm::StructType::get(Exn->getType(), - Sel->getType(), nullptr); + llvm::Type *LPadType = llvm::StructType::get(Exn->getType(), Sel->getType()); llvm::Value *LPadVal = llvm::UndefValue::get(LPadType); LPadVal = Builder.CreateInsertValue(LPadVal, Exn, 0, "lpad.val"); LPadVal = Builder.CreateInsertValue(LPadVal, Sel, 1, "lpad.val"); @@ -1747,7 +1746,7 @@ void CodeGenFunction::EmitSEHExceptionCodeSave(CodeGenFunction &ParentCGF, // }; // int exceptioncode = exception_pointers->ExceptionRecord->ExceptionCode; llvm::Type *RecordTy = CGM.Int32Ty->getPointerTo(); - llvm::Type *PtrsTy = llvm::StructType::get(RecordTy, CGM.VoidPtrTy, nullptr); + llvm::Type *PtrsTy = llvm::StructType::get(RecordTy, CGM.VoidPtrTy); llvm::Value *Ptrs = Builder.CreateBitCast(SEHInfo, PtrsTy->getPointerTo()); llvm::Value *Rec = Builder.CreateStructGEP(PtrsTy, Ptrs, 0); Rec = Builder.CreateAlignedLoad(Rec, getPointerAlign()); diff --git a/lib/CodeGen/CGExpr.cpp b/lib/CodeGen/CGExpr.cpp index 863b4380da47..cef6292c0e4d 100644 --- a/lib/CodeGen/CGExpr.cpp +++ b/lib/CodeGen/CGExpr.cpp @@ -2859,9 +2859,9 @@ void CodeGenFunction::EmitCfiCheckFail() { EmitTrapCheck(DataIsNotNullPtr); llvm::StructType *SourceLocationTy = - llvm::StructType::get(VoidPtrTy, Int32Ty, Int32Ty, nullptr); + llvm::StructType::get(VoidPtrTy, Int32Ty, Int32Ty); llvm::StructType *CfiCheckFailDataTy = - llvm::StructType::get(Int8Ty, SourceLocationTy, VoidPtrTy, nullptr); + llvm::StructType::get(Int8Ty, SourceLocationTy, VoidPtrTy); llvm::Value *V = Builder.CreateConstGEP2_32( CfiCheckFailDataTy, diff --git a/lib/CodeGen/CGExprAgg.cpp b/lib/CodeGen/CGExprAgg.cpp index 49bbb4808eaa..a05a088f0919 100644 --- a/lib/CodeGen/CGExprAgg.cpp +++ b/lib/CodeGen/CGExprAgg.cpp @@ -512,12 +512,20 @@ void AggExprEmitter::EmitArrayInit(Address DestPtr, llvm::ArrayType *AType, currentElement->addIncoming(element, entryBB); // Emit the actual filler expression. - LValue elementLV = - CGF.MakeAddrLValue(Address(currentElement, elementAlign), elementType); - if (filler) - EmitInitializationToLValue(filler, elementLV); - else - EmitNullInitializationToLValue(elementLV); + { + // C++1z [class.temporary]p5: + // when a default constructor is called to initialize an element of + // an array with no corresponding initializer [...] the destruction of + // every temporary created in a default argument is sequenced before + // the construction of the next array element, if any + CodeGenFunction::RunCleanupsScope CleanupsScope(CGF); + LValue elementLV = + CGF.MakeAddrLValue(Address(currentElement, elementAlign), elementType); + if (filler) + EmitInitializationToLValue(filler, elementLV); + else + EmitNullInitializationToLValue(elementLV); + } // Move on to the next element. llvm::Value *nextElement = diff --git a/lib/CodeGen/CGExprConstant.cpp b/lib/CodeGen/CGExprConstant.cpp index 53c184130709..6b72774c10a5 100644 --- a/lib/CodeGen/CGExprConstant.cpp +++ b/lib/CodeGen/CGExprConstant.cpp @@ -1361,9 +1361,8 @@ llvm::Constant *CodeGenModule::EmitConstantValue(const APValue &Value, Value.getComplexIntImag()); // FIXME: the target may want to specify that this is packed. - llvm::StructType *STy = llvm::StructType::get(Complex[0]->getType(), - Complex[1]->getType(), - nullptr); + llvm::StructType *STy = + llvm::StructType::get(Complex[0]->getType(), Complex[1]->getType()); return llvm::ConstantStruct::get(STy, Complex); } case APValue::Float: { @@ -1384,9 +1383,8 @@ llvm::Constant *CodeGenModule::EmitConstantValue(const APValue &Value, Value.getComplexFloatImag()); // FIXME: the target may want to specify that this is packed. - llvm::StructType *STy = llvm::StructType::get(Complex[0]->getType(), - Complex[1]->getType(), - nullptr); + llvm::StructType *STy = + llvm::StructType::get(Complex[0]->getType(), Complex[1]->getType()); return llvm::ConstantStruct::get(STy, Complex); } case APValue::Vector: { diff --git a/lib/CodeGen/CGExprScalar.cpp b/lib/CodeGen/CGExprScalar.cpp index 70b741651fd1..b8d830ee9f3f 100644 --- a/lib/CodeGen/CGExprScalar.cpp +++ b/lib/CodeGen/CGExprScalar.cpp @@ -89,14 +89,14 @@ struct BinOpInfo { } /// Check if the binop computes a division or a remainder. - bool isDivisionLikeOperation() const { + bool isDivremOp() const { return Opcode == BO_Div || Opcode == BO_Rem || Opcode == BO_DivAssign || Opcode == BO_RemAssign; } /// Check if the binop can result in an integer division by zero. bool mayHaveIntegerDivisionByZero() const { - if (isDivisionLikeOperation()) + if (isDivremOp()) if (auto *CI = dyn_cast(RHS)) return CI->isZero(); return true; @@ -104,7 +104,7 @@ struct BinOpInfo { /// Check if the binop can result in a float division by zero. bool mayHaveFloatDivisionByZero() const { - if (isDivisionLikeOperation()) + if (isDivremOp()) if (auto *CFP = dyn_cast(RHS)) return CFP->isZero(); return true; @@ -2552,6 +2552,7 @@ Value *ScalarExprEmitter::EmitOverflowCheckedBinOp(const BinOpInfo &Ops) { if (isSigned) OpID |= 1; + CodeGenFunction::SanitizerScope SanScope(&CGF); llvm::Type *opTy = CGF.CGM.getTypes().ConvertType(Ops.Ty); llvm::Function *intrinsic = CGF.CGM.getIntrinsic(IID, opTy); @@ -2567,7 +2568,6 @@ Value *ScalarExprEmitter::EmitOverflowCheckedBinOp(const BinOpInfo &Ops) { // If the signed-integer-overflow sanitizer is enabled, emit a call to its // runtime. Otherwise, this is a -ftrapv check, so just emit a trap. if (!isSigned || CGF.SanOpts.has(SanitizerKind::SignedIntegerOverflow)) { - CodeGenFunction::SanitizerScope SanScope(&CGF); llvm::Value *NotOverflow = Builder.CreateNot(overflow); SanitizerMask Kind = isSigned ? SanitizerKind::SignedIntegerOverflow : SanitizerKind::UnsignedIntegerOverflow; diff --git a/lib/CodeGen/CGObjCGNU.cpp b/lib/CodeGen/CGObjCGNU.cpp index 821629c50d4a..c8b8be7f4552 100644 --- a/lib/CodeGen/CGObjCGNU.cpp +++ b/lib/CodeGen/CGObjCGNU.cpp @@ -34,7 +34,6 @@ #include "llvm/IR/LLVMContext.h" #include "llvm/IR/Module.h" #include "llvm/Support/Compiler.h" -#include using namespace clang; using namespace CodeGen; @@ -58,18 +57,19 @@ class LazyRuntimeFunction { /// Initialises the lazy function with the name, return type, and the types /// of the arguments. - LLVM_END_WITH_NULL - void init(CodeGenModule *Mod, const char *name, llvm::Type *RetTy, ...) { + template + void init(CodeGenModule *Mod, const char *name, llvm::Type *RetTy, + Tys *... Types) { CGM = Mod; FunctionName = name; Function = nullptr; - std::vector ArgTys; - va_list Args; - va_start(Args, RetTy); - while (llvm::Type *ArgTy = va_arg(Args, llvm::Type *)) - ArgTys.push_back(ArgTy); - va_end(Args); - FTy = llvm::FunctionType::get(RetTy, ArgTys, false); + if(sizeof...(Tys)) { + SmallVector ArgTys({Types...}); + FTy = llvm::FunctionType::get(RetTy, ArgTys, false); + } + else { + FTy = llvm::FunctionType::get(RetTy, None, false); + } } llvm::FunctionType *getType() { return FTy; } @@ -603,11 +603,10 @@ class CGObjCGCC : public CGObjCGNU { public: CGObjCGCC(CodeGenModule &Mod) : CGObjCGNU(Mod, 8, 2) { // IMP objc_msg_lookup(id, SEL); - MsgLookupFn.init(&CGM, "objc_msg_lookup", IMPTy, IdTy, SelectorTy, - nullptr); + MsgLookupFn.init(&CGM, "objc_msg_lookup", IMPTy, IdTy, SelectorTy); // IMP objc_msg_lookup_super(struct objc_super*, SEL); MsgLookupSuperFn.init(&CGM, "objc_msg_lookup_super", IMPTy, - PtrToObjCSuperTy, SelectorTy, nullptr); + PtrToObjCSuperTy, SelectorTy); } }; @@ -702,52 +701,51 @@ class CGObjCGNUstep : public CGObjCGNU { CGObjCGNUstep(CodeGenModule &Mod) : CGObjCGNU(Mod, 9, 3) { const ObjCRuntime &R = CGM.getLangOpts().ObjCRuntime; - llvm::StructType *SlotStructTy = llvm::StructType::get(PtrTy, - PtrTy, PtrTy, IntTy, IMPTy, nullptr); + llvm::StructType *SlotStructTy = + llvm::StructType::get(PtrTy, PtrTy, PtrTy, IntTy, IMPTy); SlotTy = llvm::PointerType::getUnqual(SlotStructTy); // Slot_t objc_msg_lookup_sender(id *receiver, SEL selector, id sender); SlotLookupFn.init(&CGM, "objc_msg_lookup_sender", SlotTy, PtrToIdTy, - SelectorTy, IdTy, nullptr); + SelectorTy, IdTy); // Slot_t objc_msg_lookup_super(struct objc_super*, SEL); SlotLookupSuperFn.init(&CGM, "objc_slot_lookup_super", SlotTy, - PtrToObjCSuperTy, SelectorTy, nullptr); + PtrToObjCSuperTy, SelectorTy); // If we're in ObjC++ mode, then we want to make if (CGM.getLangOpts().CPlusPlus) { llvm::Type *VoidTy = llvm::Type::getVoidTy(VMContext); // void *__cxa_begin_catch(void *e) - EnterCatchFn.init(&CGM, "__cxa_begin_catch", PtrTy, PtrTy, nullptr); + EnterCatchFn.init(&CGM, "__cxa_begin_catch", PtrTy, PtrTy); // void __cxa_end_catch(void) - ExitCatchFn.init(&CGM, "__cxa_end_catch", VoidTy, nullptr); + ExitCatchFn.init(&CGM, "__cxa_end_catch", VoidTy); // void _Unwind_Resume_or_Rethrow(void*) ExceptionReThrowFn.init(&CGM, "_Unwind_Resume_or_Rethrow", VoidTy, - PtrTy, nullptr); + PtrTy); } else if (R.getVersion() >= VersionTuple(1, 7)) { llvm::Type *VoidTy = llvm::Type::getVoidTy(VMContext); // id objc_begin_catch(void *e) - EnterCatchFn.init(&CGM, "objc_begin_catch", IdTy, PtrTy, nullptr); + EnterCatchFn.init(&CGM, "objc_begin_catch", IdTy, PtrTy); // void objc_end_catch(void) - ExitCatchFn.init(&CGM, "objc_end_catch", VoidTy, nullptr); + ExitCatchFn.init(&CGM, "objc_end_catch", VoidTy); // void _Unwind_Resume_or_Rethrow(void*) - ExceptionReThrowFn.init(&CGM, "objc_exception_rethrow", VoidTy, - PtrTy, nullptr); + ExceptionReThrowFn.init(&CGM, "objc_exception_rethrow", VoidTy, PtrTy); } llvm::Type *VoidTy = llvm::Type::getVoidTy(VMContext); SetPropertyAtomic.init(&CGM, "objc_setProperty_atomic", VoidTy, IdTy, - SelectorTy, IdTy, PtrDiffTy, nullptr); + SelectorTy, IdTy, PtrDiffTy); SetPropertyAtomicCopy.init(&CGM, "objc_setProperty_atomic_copy", VoidTy, - IdTy, SelectorTy, IdTy, PtrDiffTy, nullptr); + IdTy, SelectorTy, IdTy, PtrDiffTy); SetPropertyNonAtomic.init(&CGM, "objc_setProperty_nonatomic", VoidTy, - IdTy, SelectorTy, IdTy, PtrDiffTy, nullptr); + IdTy, SelectorTy, IdTy, PtrDiffTy); SetPropertyNonAtomicCopy.init(&CGM, "objc_setProperty_nonatomic_copy", - VoidTy, IdTy, SelectorTy, IdTy, PtrDiffTy, nullptr); + VoidTy, IdTy, SelectorTy, IdTy, PtrDiffTy); // void objc_setCppObjectAtomic(void *dest, const void *src, void // *helper); CxxAtomicObjectSetFn.init(&CGM, "objc_setCppObjectAtomic", VoidTy, PtrTy, - PtrTy, PtrTy, nullptr); + PtrTy, PtrTy); // void objc_getCppObjectAtomic(void *dest, const void *src, void // *helper); CxxAtomicObjectGetFn.init(&CGM, "objc_getCppObjectAtomic", VoidTy, PtrTy, - PtrTy, PtrTy, nullptr); + PtrTy, PtrTy); } llvm::Constant *GetCppAtomicObjectGetFunction() override { @@ -849,14 +847,14 @@ class CGObjCObjFW: public CGObjCGNU { public: CGObjCObjFW(CodeGenModule &Mod): CGObjCGNU(Mod, 9, 3) { // IMP objc_msg_lookup(id, SEL); - MsgLookupFn.init(&CGM, "objc_msg_lookup", IMPTy, IdTy, SelectorTy, nullptr); + MsgLookupFn.init(&CGM, "objc_msg_lookup", IMPTy, IdTy, SelectorTy); MsgLookupFnSRet.init(&CGM, "objc_msg_lookup_stret", IMPTy, IdTy, - SelectorTy, nullptr); + SelectorTy); // IMP objc_msg_lookup_super(struct objc_super*, SEL); MsgLookupSuperFn.init(&CGM, "objc_msg_lookup_super", IMPTy, - PtrToObjCSuperTy, SelectorTy, nullptr); + PtrToObjCSuperTy, SelectorTy); MsgLookupSuperFnSRet.init(&CGM, "objc_msg_lookup_super_stret", IMPTy, - PtrToObjCSuperTy, SelectorTy, nullptr); + PtrToObjCSuperTy, SelectorTy); } }; } // end anonymous namespace @@ -945,35 +943,34 @@ CGObjCGNU::CGObjCGNU(CodeGenModule &cgm, unsigned runtimeABIVersion, } PtrToIdTy = llvm::PointerType::getUnqual(IdTy); - ObjCSuperTy = llvm::StructType::get(IdTy, IdTy, nullptr); + ObjCSuperTy = llvm::StructType::get(IdTy, IdTy); PtrToObjCSuperTy = llvm::PointerType::getUnqual(ObjCSuperTy); llvm::Type *VoidTy = llvm::Type::getVoidTy(VMContext); // void objc_exception_throw(id); - ExceptionThrowFn.init(&CGM, "objc_exception_throw", VoidTy, IdTy, nullptr); - ExceptionReThrowFn.init(&CGM, "objc_exception_throw", VoidTy, IdTy, nullptr); + ExceptionThrowFn.init(&CGM, "objc_exception_throw", VoidTy, IdTy); + ExceptionReThrowFn.init(&CGM, "objc_exception_throw", VoidTy, IdTy); // int objc_sync_enter(id); - SyncEnterFn.init(&CGM, "objc_sync_enter", IntTy, IdTy, nullptr); + SyncEnterFn.init(&CGM, "objc_sync_enter", IntTy, IdTy); // int objc_sync_exit(id); - SyncExitFn.init(&CGM, "objc_sync_exit", IntTy, IdTy, nullptr); + SyncExitFn.init(&CGM, "objc_sync_exit", IntTy, IdTy); // void objc_enumerationMutation (id) - EnumerationMutationFn.init(&CGM, "objc_enumerationMutation", VoidTy, - IdTy, nullptr); + EnumerationMutationFn.init(&CGM, "objc_enumerationMutation", VoidTy, IdTy); // id objc_getProperty(id, SEL, ptrdiff_t, BOOL) GetPropertyFn.init(&CGM, "objc_getProperty", IdTy, IdTy, SelectorTy, - PtrDiffTy, BoolTy, nullptr); + PtrDiffTy, BoolTy); // void objc_setProperty(id, SEL, ptrdiff_t, id, BOOL, BOOL) SetPropertyFn.init(&CGM, "objc_setProperty", VoidTy, IdTy, SelectorTy, - PtrDiffTy, IdTy, BoolTy, BoolTy, nullptr); + PtrDiffTy, IdTy, BoolTy, BoolTy); // void objc_setPropertyStruct(void*, void*, ptrdiff_t, BOOL, BOOL) - GetStructPropertyFn.init(&CGM, "objc_getPropertyStruct", VoidTy, PtrTy, PtrTy, - PtrDiffTy, BoolTy, BoolTy, nullptr); + GetStructPropertyFn.init(&CGM, "objc_getPropertyStruct", VoidTy, PtrTy, PtrTy, + PtrDiffTy, BoolTy, BoolTy); // void objc_setPropertyStruct(void*, void*, ptrdiff_t, BOOL, BOOL) - SetStructPropertyFn.init(&CGM, "objc_setPropertyStruct", VoidTy, PtrTy, PtrTy, - PtrDiffTy, BoolTy, BoolTy, nullptr); + SetStructPropertyFn.init(&CGM, "objc_setPropertyStruct", VoidTy, PtrTy, PtrTy, + PtrDiffTy, BoolTy, BoolTy); // IMP type llvm::Type *IMPArgs[] = { IdTy, SelectorTy }; @@ -997,21 +994,19 @@ CGObjCGNU::CGObjCGNU(CodeGenModule &cgm, unsigned runtimeABIVersion, // Get functions needed in GC mode // id objc_assign_ivar(id, id, ptrdiff_t); - IvarAssignFn.init(&CGM, "objc_assign_ivar", IdTy, IdTy, IdTy, PtrDiffTy, - nullptr); + IvarAssignFn.init(&CGM, "objc_assign_ivar", IdTy, IdTy, IdTy, PtrDiffTy); // id objc_assign_strongCast (id, id*) StrongCastAssignFn.init(&CGM, "objc_assign_strongCast", IdTy, IdTy, - PtrToIdTy, nullptr); + PtrToIdTy); // id objc_assign_global(id, id*); - GlobalAssignFn.init(&CGM, "objc_assign_global", IdTy, IdTy, PtrToIdTy, - nullptr); + GlobalAssignFn.init(&CGM, "objc_assign_global", IdTy, IdTy, PtrToIdTy); // id objc_assign_weak(id, id*); - WeakAssignFn.init(&CGM, "objc_assign_weak", IdTy, IdTy, PtrToIdTy, nullptr); + WeakAssignFn.init(&CGM, "objc_assign_weak", IdTy, IdTy, PtrToIdTy); // id objc_read_weak(id*); - WeakReadFn.init(&CGM, "objc_read_weak", IdTy, PtrToIdTy, nullptr); + WeakReadFn.init(&CGM, "objc_read_weak", IdTy, PtrToIdTy); // void *objc_memmove_collectable(void*, void *, size_t); MemMoveFn.init(&CGM, "objc_memmove_collectable", PtrTy, PtrTy, PtrTy, - SizeTy, nullptr); + SizeTy); } } @@ -1317,7 +1312,7 @@ CGObjCGNU::GenerateMessageSendSuper(CodeGenFunction &CGF, } } // Cast the pointer to a simplified version of the class structure - llvm::Type *CastTy = llvm::StructType::get(IdTy, IdTy, nullptr); + llvm::Type *CastTy = llvm::StructType::get(IdTy, IdTy); ReceiverClass = Builder.CreateBitCast(ReceiverClass, llvm::PointerType::getUnqual(CastTy)); // Get the superclass pointer @@ -1326,8 +1321,8 @@ CGObjCGNU::GenerateMessageSendSuper(CodeGenFunction &CGF, ReceiverClass = Builder.CreateAlignedLoad(ReceiverClass, CGF.getPointerAlign()); // Construct the structure used to look up the IMP - llvm::StructType *ObjCSuperTy = llvm::StructType::get( - Receiver->getType(), IdTy, nullptr); + llvm::StructType *ObjCSuperTy = + llvm::StructType::get(Receiver->getType(), IdTy); // FIXME: Is this really supposed to be a dynamic alloca? Address ObjCSuper = Address(Builder.CreateAlloca(ObjCSuperTy), @@ -1565,11 +1560,8 @@ GenerateIvarList(ArrayRef IvarNames, IvarList.addInt(IntTy, (int)IvarNames.size()); // Get the ivar structure type. - llvm::StructType *ObjCIvarTy = llvm::StructType::get( - PtrToInt8Ty, - PtrToInt8Ty, - IntTy, - nullptr); + llvm::StructType *ObjCIvarTy = + llvm::StructType::get(PtrToInt8Ty, PtrToInt8Ty, IntTy); // Array of ivar structures. auto Ivars = IvarList.beginArray(ObjCIvarTy); @@ -1611,7 +1603,7 @@ llvm::Constant *CGObjCGNU::GenerateClassStructure( // anyway; the classes will still work with the GNU runtime, they will just // be ignored. llvm::StructType *ClassTy = llvm::StructType::get( - PtrToInt8Ty, // isa + PtrToInt8Ty, // isa PtrToInt8Ty, // super_class PtrToInt8Ty, // name LongTy, // version @@ -1620,18 +1612,18 @@ llvm::Constant *CGObjCGNU::GenerateClassStructure( IVars->getType(), // ivars Methods->getType(), // methods // These are all filled in by the runtime, so we pretend - PtrTy, // dtable - PtrTy, // subclass_list - PtrTy, // sibling_class - PtrTy, // protocols - PtrTy, // gc_object_type + PtrTy, // dtable + PtrTy, // subclass_list + PtrTy, // sibling_class + PtrTy, // protocols + PtrTy, // gc_object_type // New ABI: LongTy, // abi_version IvarOffsets->getType(), // ivar_offsets Properties->getType(), // properties IntPtrTy, // strong_pointers - IntPtrTy, // weak_pointers - nullptr); + IntPtrTy // weak_pointers + ); ConstantInitBuilder Builder(CGM); auto Elements = Builder.beginStruct(ClassTy); diff --git a/lib/CodeGen/CGObjCMac.cpp b/lib/CodeGen/CGObjCMac.cpp index 70d24b791334..9c048423285b 100644 --- a/lib/CodeGen/CGObjCMac.cpp +++ b/lib/CodeGen/CGObjCMac.cpp @@ -105,8 +105,8 @@ class ObjCCommonTypesHelper { llvm::Constant *getMessageSendFp2retFn() const { llvm::Type *params[] = { ObjectPtrTy, SelectorPtrTy }; llvm::Type *longDoubleType = llvm::Type::getX86_FP80Ty(VMContext); - llvm::Type *resultType = - llvm::StructType::get(longDoubleType, longDoubleType, nullptr); + llvm::Type *resultType = + llvm::StructType::get(longDoubleType, longDoubleType); return CGM.CreateRuntimeFunction(llvm::FunctionType::get(resultType, params, true), @@ -5506,17 +5506,15 @@ ObjCCommonTypesHelper::ObjCCommonTypesHelper(CodeGen::CodeGenModule &cgm) // char *name; // char *attributes; // } - PropertyTy = llvm::StructType::create("struct._prop_t", - Int8PtrTy, Int8PtrTy, nullptr); + PropertyTy = llvm::StructType::create("struct._prop_t", Int8PtrTy, Int8PtrTy); // struct _prop_list_t { // uint32_t entsize; // sizeof(struct _prop_t) // uint32_t count_of_properties; // struct _prop_t prop_list[count_of_properties]; // } - PropertyListTy = - llvm::StructType::create("struct._prop_list_t", IntTy, IntTy, - llvm::ArrayType::get(PropertyTy, 0), nullptr); + PropertyListTy = llvm::StructType::create( + "struct._prop_list_t", IntTy, IntTy, llvm::ArrayType::get(PropertyTy, 0)); // struct _prop_list_t * PropertyListPtrTy = llvm::PointerType::getUnqual(PropertyListTy); @@ -5525,9 +5523,8 @@ ObjCCommonTypesHelper::ObjCCommonTypesHelper(CodeGen::CodeGenModule &cgm) // char *method_type; // char *_imp; // } - MethodTy = llvm::StructType::create("struct._objc_method", - SelectorPtrTy, Int8PtrTy, Int8PtrTy, - nullptr); + MethodTy = llvm::StructType::create("struct._objc_method", SelectorPtrTy, + Int8PtrTy, Int8PtrTy); // struct _objc_cache * CacheTy = llvm::StructType::create(VMContext, "struct._objc_cache"); @@ -5540,17 +5537,16 @@ ObjCTypesHelper::ObjCTypesHelper(CodeGen::CodeGenModule &cgm) // SEL name; // char *types; // } - MethodDescriptionTy = - llvm::StructType::create("struct._objc_method_description", - SelectorPtrTy, Int8PtrTy, nullptr); + MethodDescriptionTy = llvm::StructType::create( + "struct._objc_method_description", SelectorPtrTy, Int8PtrTy); // struct _objc_method_description_list { // int count; // struct _objc_method_description[1]; // } - MethodDescriptionListTy = llvm::StructType::create( - "struct._objc_method_description_list", IntTy, - llvm::ArrayType::get(MethodDescriptionTy, 0), nullptr); + MethodDescriptionListTy = + llvm::StructType::create("struct._objc_method_description_list", IntTy, + llvm::ArrayType::get(MethodDescriptionTy, 0)); // struct _objc_method_description_list * MethodDescriptionListPtrTy = @@ -5566,11 +5562,10 @@ ObjCTypesHelper::ObjCTypesHelper(CodeGen::CodeGenModule &cgm) // const char ** extendedMethodTypes; // struct _objc_property_list *class_properties; // } - ProtocolExtensionTy = - llvm::StructType::create("struct._objc_protocol_extension", - IntTy, MethodDescriptionListPtrTy, - MethodDescriptionListPtrTy, PropertyListPtrTy, - Int8PtrPtrTy, PropertyListPtrTy, nullptr); + ProtocolExtensionTy = llvm::StructType::create( + "struct._objc_protocol_extension", IntTy, MethodDescriptionListPtrTy, + MethodDescriptionListPtrTy, PropertyListPtrTy, Int8PtrPtrTy, + PropertyListPtrTy); // struct _objc_protocol_extension * ProtocolExtensionPtrTy = llvm::PointerType::getUnqual(ProtocolExtensionTy); @@ -5582,10 +5577,8 @@ ObjCTypesHelper::ObjCTypesHelper(CodeGen::CodeGenModule &cgm) ProtocolListTy = llvm::StructType::create(VMContext, "struct._objc_protocol_list"); - ProtocolListTy->setBody(llvm::PointerType::getUnqual(ProtocolListTy), - LongTy, - llvm::ArrayType::get(ProtocolTy, 0), - nullptr); + ProtocolListTy->setBody(llvm::PointerType::getUnqual(ProtocolListTy), LongTy, + llvm::ArrayType::get(ProtocolTy, 0)); // struct _objc_protocol { // struct _objc_protocol_extension *isa; @@ -5596,9 +5589,7 @@ ObjCTypesHelper::ObjCTypesHelper(CodeGen::CodeGenModule &cgm) // } ProtocolTy->setBody(ProtocolExtensionPtrTy, Int8PtrTy, llvm::PointerType::getUnqual(ProtocolListTy), - MethodDescriptionListPtrTy, - MethodDescriptionListPtrTy, - nullptr); + MethodDescriptionListPtrTy, MethodDescriptionListPtrTy); // struct _objc_protocol_list * ProtocolListPtrTy = llvm::PointerType::getUnqual(ProtocolListTy); @@ -5612,8 +5603,8 @@ ObjCTypesHelper::ObjCTypesHelper(CodeGen::CodeGenModule &cgm) // char *ivar_type; // int ivar_offset; // } - IvarTy = llvm::StructType::create("struct._objc_ivar", - Int8PtrTy, Int8PtrTy, IntTy, nullptr); + IvarTy = llvm::StructType::create("struct._objc_ivar", Int8PtrTy, Int8PtrTy, + IntTy); // struct _objc_ivar_list * IvarListTy = @@ -5626,9 +5617,8 @@ ObjCTypesHelper::ObjCTypesHelper(CodeGen::CodeGenModule &cgm) MethodListPtrTy = llvm::PointerType::getUnqual(MethodListTy); // struct _objc_class_extension * - ClassExtensionTy = - llvm::StructType::create("struct._objc_class_extension", - IntTy, Int8PtrTy, PropertyListPtrTy, nullptr); + ClassExtensionTy = llvm::StructType::create( + "struct._objc_class_extension", IntTy, Int8PtrTy, PropertyListPtrTy); ClassExtensionPtrTy = llvm::PointerType::getUnqual(ClassExtensionTy); ClassTy = llvm::StructType::create(VMContext, "struct._objc_class"); @@ -5648,18 +5638,9 @@ ObjCTypesHelper::ObjCTypesHelper(CodeGen::CodeGenModule &cgm) // struct _objc_class_ext *ext; // }; ClassTy->setBody(llvm::PointerType::getUnqual(ClassTy), - llvm::PointerType::getUnqual(ClassTy), - Int8PtrTy, - LongTy, - LongTy, - LongTy, - IvarListPtrTy, - MethodListPtrTy, - CachePtrTy, - ProtocolListPtrTy, - Int8PtrTy, - ClassExtensionPtrTy, - nullptr); + llvm::PointerType::getUnqual(ClassTy), Int8PtrTy, LongTy, + LongTy, LongTy, IvarListPtrTy, MethodListPtrTy, CachePtrTy, + ProtocolListPtrTy, Int8PtrTy, ClassExtensionPtrTy); ClassPtrTy = llvm::PointerType::getUnqual(ClassTy); @@ -5673,12 +5654,10 @@ ObjCTypesHelper::ObjCTypesHelper(CodeGen::CodeGenModule &cgm) // struct _objc_property_list *instance_properties;// category's @property // struct _objc_property_list *class_properties; // } - CategoryTy = - llvm::StructType::create("struct._objc_category", - Int8PtrTy, Int8PtrTy, MethodListPtrTy, - MethodListPtrTy, ProtocolListPtrTy, - IntTy, PropertyListPtrTy, PropertyListPtrTy, - nullptr); + CategoryTy = llvm::StructType::create( + "struct._objc_category", Int8PtrTy, Int8PtrTy, MethodListPtrTy, + MethodListPtrTy, ProtocolListPtrTy, IntTy, PropertyListPtrTy, + PropertyListPtrTy); // Global metadata structures @@ -5689,10 +5668,9 @@ ObjCTypesHelper::ObjCTypesHelper(CodeGen::CodeGenModule &cgm) // short cat_def_cnt; // char *defs[cls_def_cnt + cat_def_cnt]; // } - SymtabTy = - llvm::StructType::create("struct._objc_symtab", - LongTy, SelectorPtrTy, ShortTy, ShortTy, - llvm::ArrayType::get(Int8PtrTy, 0), nullptr); + SymtabTy = llvm::StructType::create("struct._objc_symtab", LongTy, + SelectorPtrTy, ShortTy, ShortTy, + llvm::ArrayType::get(Int8PtrTy, 0)); SymtabPtrTy = llvm::PointerType::getUnqual(SymtabTy); // struct _objc_module { @@ -5701,10 +5679,8 @@ ObjCTypesHelper::ObjCTypesHelper(CodeGen::CodeGenModule &cgm) // char *name; // struct _objc_symtab* symtab; // } - ModuleTy = - llvm::StructType::create("struct._objc_module", - LongTy, LongTy, Int8PtrTy, SymtabPtrTy, nullptr); - + ModuleTy = llvm::StructType::create("struct._objc_module", LongTy, LongTy, + Int8PtrTy, SymtabPtrTy); // FIXME: This is the size of the setjmp buffer and should be target // specific. 18 is what's used on 32-bit X86. @@ -5713,10 +5689,9 @@ ObjCTypesHelper::ObjCTypesHelper(CodeGen::CodeGenModule &cgm) // Exceptions llvm::Type *StackPtrTy = llvm::ArrayType::get(CGM.Int8PtrTy, 4); - ExceptionDataTy = - llvm::StructType::create("struct._objc_exception_data", - llvm::ArrayType::get(CGM.Int32Ty,SetJmpBufferSize), - StackPtrTy, nullptr); + ExceptionDataTy = llvm::StructType::create( + "struct._objc_exception_data", + llvm::ArrayType::get(CGM.Int32Ty, SetJmpBufferSize), StackPtrTy); } ObjCNonFragileABITypesHelper::ObjCNonFragileABITypesHelper(CodeGen::CodeGenModule &cgm) @@ -5727,8 +5702,8 @@ ObjCNonFragileABITypesHelper::ObjCNonFragileABITypesHelper(CodeGen::CodeGenModul // struct _objc_method method_list[method_count]; // } MethodListnfABITy = - llvm::StructType::create("struct.__method_list_t", IntTy, IntTy, - llvm::ArrayType::get(MethodTy, 0), nullptr); + llvm::StructType::create("struct.__method_list_t", IntTy, IntTy, + llvm::ArrayType::get(MethodTy, 0)); // struct method_list_t * MethodListnfABIPtrTy = llvm::PointerType::getUnqual(MethodListnfABITy); @@ -5752,14 +5727,12 @@ ObjCNonFragileABITypesHelper::ObjCNonFragileABITypesHelper(CodeGen::CodeGenModul ProtocolListnfABITy = llvm::StructType::create(VMContext, "struct._objc_protocol_list"); - ProtocolnfABITy = - llvm::StructType::create("struct._protocol_t", ObjectPtrTy, Int8PtrTy, - llvm::PointerType::getUnqual(ProtocolListnfABITy), - MethodListnfABIPtrTy, MethodListnfABIPtrTy, - MethodListnfABIPtrTy, MethodListnfABIPtrTy, - PropertyListPtrTy, IntTy, IntTy, Int8PtrPtrTy, - Int8PtrTy, PropertyListPtrTy, - nullptr); + ProtocolnfABITy = llvm::StructType::create( + "struct._protocol_t", ObjectPtrTy, Int8PtrTy, + llvm::PointerType::getUnqual(ProtocolListnfABITy), MethodListnfABIPtrTy, + MethodListnfABIPtrTy, MethodListnfABIPtrTy, MethodListnfABIPtrTy, + PropertyListPtrTy, IntTy, IntTy, Int8PtrPtrTy, Int8PtrTy, + PropertyListPtrTy); // struct _protocol_t* ProtocolnfABIPtrTy = llvm::PointerType::getUnqual(ProtocolnfABITy); @@ -5769,8 +5742,7 @@ ObjCNonFragileABITypesHelper::ObjCNonFragileABITypesHelper(CodeGen::CodeGenModul // struct _protocol_t *[protocol_count]; // } ProtocolListnfABITy->setBody(LongTy, - llvm::ArrayType::get(ProtocolnfABIPtrTy, 0), - nullptr); + llvm::ArrayType::get(ProtocolnfABIPtrTy, 0)); // struct _objc_protocol_list* ProtocolListnfABIPtrTy = llvm::PointerType::getUnqual(ProtocolListnfABITy); @@ -5784,7 +5756,7 @@ ObjCNonFragileABITypesHelper::ObjCNonFragileABITypesHelper(CodeGen::CodeGenModul // } IvarnfABITy = llvm::StructType::create( "struct._ivar_t", llvm::PointerType::getUnqual(IvarOffsetVarTy), - Int8PtrTy, Int8PtrTy, IntTy, IntTy, nullptr); + Int8PtrTy, Int8PtrTy, IntTy, IntTy); // struct _ivar_list_t { // uint32 entsize; // sizeof(struct _ivar_t) @@ -5792,8 +5764,8 @@ ObjCNonFragileABITypesHelper::ObjCNonFragileABITypesHelper(CodeGen::CodeGenModul // struct _iver_t list[count]; // } IvarListnfABITy = - llvm::StructType::create("struct._ivar_list_t", IntTy, IntTy, - llvm::ArrayType::get(IvarnfABITy, 0), nullptr); + llvm::StructType::create("struct._ivar_list_t", IntTy, IntTy, + llvm::ArrayType::get(IvarnfABITy, 0)); IvarListnfABIPtrTy = llvm::PointerType::getUnqual(IvarListnfABITy); @@ -5812,13 +5784,10 @@ ObjCNonFragileABITypesHelper::ObjCNonFragileABITypesHelper(CodeGen::CodeGenModul // } // FIXME. Add 'reserved' field in 64bit abi mode! - ClassRonfABITy = llvm::StructType::create("struct._class_ro_t", - IntTy, IntTy, IntTy, Int8PtrTy, - Int8PtrTy, MethodListnfABIPtrTy, - ProtocolListnfABIPtrTy, - IvarListnfABIPtrTy, - Int8PtrTy, PropertyListPtrTy, - nullptr); + ClassRonfABITy = llvm::StructType::create( + "struct._class_ro_t", IntTy, IntTy, IntTy, Int8PtrTy, Int8PtrTy, + MethodListnfABIPtrTy, ProtocolListnfABIPtrTy, IvarListnfABIPtrTy, + Int8PtrTy, PropertyListPtrTy); // ImpnfABITy - LLVM for id (*)(id, SEL, ...) llvm::Type *params[] = { ObjectPtrTy, SelectorPtrTy }; @@ -5835,11 +5804,9 @@ ObjCNonFragileABITypesHelper::ObjCNonFragileABITypesHelper(CodeGen::CodeGenModul ClassnfABITy = llvm::StructType::create(VMContext, "struct._class_t"); ClassnfABITy->setBody(llvm::PointerType::getUnqual(ClassnfABITy), - llvm::PointerType::getUnqual(ClassnfABITy), - CachePtrTy, + llvm::PointerType::getUnqual(ClassnfABITy), CachePtrTy, llvm::PointerType::getUnqual(ImpnfABITy), - llvm::PointerType::getUnqual(ClassRonfABITy), - nullptr); + llvm::PointerType::getUnqual(ClassRonfABITy)); // LLVM for struct _class_t * ClassnfABIPtrTy = llvm::PointerType::getUnqual(ClassnfABITy); @@ -5854,15 +5821,10 @@ ObjCNonFragileABITypesHelper::ObjCNonFragileABITypesHelper(CodeGen::CodeGenModul // const struct _prop_list_t * const class_properties; // const uint32_t size; // } - CategorynfABITy = llvm::StructType::create("struct._category_t", - Int8PtrTy, ClassnfABIPtrTy, - MethodListnfABIPtrTy, - MethodListnfABIPtrTy, - ProtocolListnfABIPtrTy, - PropertyListPtrTy, - PropertyListPtrTy, - IntTy, - nullptr); + CategorynfABITy = llvm::StructType::create( + "struct._category_t", Int8PtrTy, ClassnfABIPtrTy, MethodListnfABIPtrTy, + MethodListnfABIPtrTy, ProtocolListnfABIPtrTy, PropertyListPtrTy, + PropertyListPtrTy, IntTy); // New types for nonfragile abi messaging. CodeGen::CodeGenTypes &Types = CGM.getTypes(); @@ -5899,9 +5861,8 @@ ObjCNonFragileABITypesHelper::ObjCNonFragileABITypesHelper(CodeGen::CodeGenModul // SUPER_IMP messenger; // SEL name; // }; - SuperMessageRefTy = - llvm::StructType::create("struct._super_message_ref_t", - ImpnfABITy, SelectorPtrTy, nullptr); + SuperMessageRefTy = llvm::StructType::create("struct._super_message_ref_t", + ImpnfABITy, SelectorPtrTy); // SuperMessageRefPtrTy - LLVM for struct _super_message_ref_t* SuperMessageRefPtrTy = llvm::PointerType::getUnqual(SuperMessageRefTy); @@ -5912,10 +5873,9 @@ ObjCNonFragileABITypesHelper::ObjCNonFragileABITypesHelper(CodeGen::CodeGenModul // const char* name; // c++ typeinfo string // Class cls; // }; - EHTypeTy = - llvm::StructType::create("struct._objc_typeinfo", - llvm::PointerType::getUnqual(Int8PtrTy), - Int8PtrTy, ClassnfABIPtrTy, nullptr); + EHTypeTy = llvm::StructType::create("struct._objc_typeinfo", + llvm::PointerType::getUnqual(Int8PtrTy), + Int8PtrTy, ClassnfABIPtrTy); EHTypePtrTy = llvm::PointerType::getUnqual(EHTypeTy); } diff --git a/lib/CodeGen/CGOpenMPRuntime.cpp b/lib/CodeGen/CGOpenMPRuntime.cpp index d1a706b8821e..b256a88c47ad 100644 --- a/lib/CodeGen/CGOpenMPRuntime.cpp +++ b/lib/CodeGen/CGOpenMPRuntime.cpp @@ -728,7 +728,7 @@ CGOpenMPRuntime::CGOpenMPRuntime(CodeGenModule &CGM) IdentTy = llvm::StructType::create( "ident_t", CGM.Int32Ty /* reserved_1 */, CGM.Int32Ty /* flags */, CGM.Int32Ty /* reserved_2 */, CGM.Int32Ty /* reserved_3 */, - CGM.Int8PtrTy /* psource */, nullptr); + CGM.Int8PtrTy /* psource */); KmpCriticalNameTy = llvm::ArrayType::get(CGM.Int32Ty, /*NumElements*/ 8); loadOffloadInfoMetadata(); diff --git a/lib/CodeGen/CodeGenAction.cpp b/lib/CodeGen/CodeGenAction.cpp index f57cbe86c413..c7e30fad7575 100644 --- a/lib/CodeGen/CodeGenAction.cpp +++ b/lib/CodeGen/CodeGenAction.cpp @@ -548,9 +548,9 @@ void BackendConsumer::UnsupportedDiagHandler( StringRef Filename; unsigned Line, Column; - bool BadDebugInfo; - FullSourceLoc Loc = getBestLocationFromDebugLoc(D, BadDebugInfo, Filename, - Line, Column); + bool BadDebugInfo = false; + FullSourceLoc Loc = + getBestLocationFromDebugLoc(D, BadDebugInfo, Filename, Line, Column); Diags.Report(Loc, diag::err_fe_backend_unsupported) << D.getMessage().str(); @@ -572,8 +572,8 @@ void BackendConsumer::EmitOptimizationMessage( StringRef Filename; unsigned Line, Column; bool BadDebugInfo = false; - FullSourceLoc Loc = getBestLocationFromDebugLoc(D, BadDebugInfo, Filename, - Line, Column); + FullSourceLoc Loc = + getBestLocationFromDebugLoc(D, BadDebugInfo, Filename, Line, Column); std::string Msg; raw_string_ostream MsgStream(Msg); diff --git a/lib/CodeGen/CodeGenModule.cpp b/lib/CodeGen/CodeGenModule.cpp index ff26d80fe2b6..a0254797ea43 100644 --- a/lib/CodeGen/CodeGenModule.cpp +++ b/lib/CodeGen/CodeGenModule.cpp @@ -751,7 +751,7 @@ void CodeGenModule::EmitCtorList(CtorList &Fns, const char *GlobalName) { // Get the type of a ctor entry, { i32, void ()*, i8* }. llvm::StructType *CtorStructTy = llvm::StructType::get( - Int32Ty, llvm::PointerType::getUnqual(CtorFTy), VoidPtrTy, nullptr); + Int32Ty, llvm::PointerType::getUnqual(CtorFTy), VoidPtrTy); // Construct the constructor and destructor arrays. ConstantInitBuilder builder(*this); diff --git a/lib/CodeGen/CodeGenTypes.cpp b/lib/CodeGen/CodeGenTypes.cpp index dc24b2040f04..5ed929135880 100644 --- a/lib/CodeGen/CodeGenTypes.cpp +++ b/lib/CodeGen/CodeGenTypes.cpp @@ -490,7 +490,7 @@ llvm::Type *CodeGenTypes::ConvertType(QualType T) { llvm_unreachable("Unexpected undeduced type!"); case Type::Complex: { llvm::Type *EltTy = ConvertType(cast(Ty)->getElementType()); - ResultType = llvm::StructType::get(EltTy, EltTy, nullptr); + ResultType = llvm::StructType::get(EltTy, EltTy); break; } case Type::LValueReference: diff --git a/lib/CodeGen/ItaniumCXXABI.cpp b/lib/CodeGen/ItaniumCXXABI.cpp index dac2d15fa406..66f51305430a 100644 --- a/lib/CodeGen/ItaniumCXXABI.cpp +++ b/lib/CodeGen/ItaniumCXXABI.cpp @@ -499,7 +499,7 @@ llvm::Type * ItaniumCXXABI::ConvertMemberPointerType(const MemberPointerType *MPT) { if (MPT->isMemberDataPointer()) return CGM.PtrDiffTy; - return llvm::StructType::get(CGM.PtrDiffTy, CGM.PtrDiffTy, nullptr); + return llvm::StructType::get(CGM.PtrDiffTy, CGM.PtrDiffTy); } /// In the Itanium and ARM ABIs, method pointers have the form: diff --git a/lib/CodeGen/TargetInfo.cpp b/lib/CodeGen/TargetInfo.cpp index 4ebbef7dfb5b..18367d1602ba 100644 --- a/lib/CodeGen/TargetInfo.cpp +++ b/lib/CodeGen/TargetInfo.cpp @@ -3159,8 +3159,7 @@ GetX86_64ByValArgumentPair(llvm::Type *Lo, llvm::Type *Hi, } } - llvm::StructType *Result = llvm::StructType::get(Lo, Hi, nullptr); - + llvm::StructType *Result = llvm::StructType::get(Lo, Hi); // Verify that the second element is at an 8-byte offset. assert(TD.getStructLayout(Result)->getElementOffset(1) == 8 && @@ -3235,8 +3234,7 @@ classifyReturnType(QualType RetTy) const { case ComplexX87: assert(Hi == ComplexX87 && "Unexpected ComplexX87 classification."); ResType = llvm::StructType::get(llvm::Type::getX86_FP80Ty(getVMContext()), - llvm::Type::getX86_FP80Ty(getVMContext()), - nullptr); + llvm::Type::getX86_FP80Ty(getVMContext())); break; } @@ -3732,7 +3730,7 @@ Address X86_64ABIInfo::EmitVAArg(CodeGenFunction &CGF, Address VAListAddr, CGF.Builder.CreateConstInBoundsByteGEP(RegAddrLo, CharUnits::fromQuantity(16)); llvm::Type *DoubleTy = CGF.DoubleTy; - llvm::StructType *ST = llvm::StructType::get(DoubleTy, DoubleTy, nullptr); + llvm::StructType *ST = llvm::StructType::get(DoubleTy, DoubleTy); llvm::Value *V; Address Tmp = CGF.CreateMemTemp(Ty); Tmp = CGF.Builder.CreateElementBitCast(Tmp, ST); @@ -4637,7 +4635,7 @@ PPC64_SVR4_ABIInfo::classifyReturnType(QualType RetTy) const { llvm::Type *CoerceTy; if (Bits > GPRBits) { CoerceTy = llvm::IntegerType::get(getVMContext(), GPRBits); - CoerceTy = llvm::StructType::get(CoerceTy, CoerceTy, nullptr); + CoerceTy = llvm::StructType::get(CoerceTy, CoerceTy); } else CoerceTy = llvm::IntegerType::get(getVMContext(), llvm::alignTo(Bits, 8)); @@ -6695,6 +6693,14 @@ MipsABIInfo::classifyArgumentType(QualType Ty, uint64_t &Offset) const { return getNaturalAlignIndirect(Ty, RAA == CGCXXABI::RAA_DirectInMemory); } + // Use indirect if the aggregate cannot fit into registers for + // passing arguments according to the ABI + unsigned Threshold = IsO32 ? 16 : 64; + + if(getContext().getTypeSizeInChars(Ty) > CharUnits::fromQuantity(Threshold)) + return ABIArgInfo::getIndirect(CharUnits::fromQuantity(Align), true, + getContext().getTypeAlign(Ty) / 8 > Align); + // If we have reached here, aggregates are passed directly by coercing to // another structure type. Padding is inserted if the offset of the // aggregate is unaligned. @@ -7037,13 +7043,13 @@ ABIArgInfo HexagonABIInfo::classifyArgumentType(QualType Ty) const { ABIArgInfo::getExtend() : ABIArgInfo::getDirect()); } + if (CGCXXABI::RecordArgABI RAA = getRecordArgABI(Ty, getCXXABI())) + return getNaturalAlignIndirect(Ty, RAA == CGCXXABI::RAA_DirectInMemory); + // Ignore empty records. if (isEmptyRecord(getContext(), Ty, true)) return ABIArgInfo::getIgnore(); - if (CGCXXABI::RecordArgABI RAA = getRecordArgABI(Ty, getCXXABI())) - return getNaturalAlignIndirect(Ty, RAA == CGCXXABI::RAA_DirectInMemory); - uint64_t Size = getContext().getTypeSize(Ty); if (Size > 64) return getNaturalAlignIndirect(Ty, /*ByVal=*/true); diff --git a/lib/Driver/SanitizerArgs.cpp b/lib/Driver/SanitizerArgs.cpp index c298302c477c..9ab2e176845c 100644 --- a/lib/Driver/SanitizerArgs.cpp +++ b/lib/Driver/SanitizerArgs.cpp @@ -563,12 +563,18 @@ SanitizerArgs::SanitizerArgs(const ToolChain &TC, } } - if (Arg *A = Args.getLastArg( - options::OPT_fsanitize_address_use_after_scope, - options::OPT_fno_sanitize_address_use_after_scope)) { - AsanUseAfterScope = A->getOption().getID() == - options::OPT_fsanitize_address_use_after_scope; - } + AsanUseAfterScope = Args.hasFlag( + options::OPT_fsanitize_address_use_after_scope, + options::OPT_fno_sanitize_address_use_after_scope, AsanUseAfterScope); + + // As a workaround for a bug in gold 2.26 and earlier, dead stripping of + // globals in ASan is disabled by default on ELF targets. + // See https://sourceware.org/bugzilla/show_bug.cgi?id=19002 + AsanGlobalsDeadStripping = + !TC.getTriple().isOSBinFormatELF() || + Args.hasArg(options::OPT_fsanitize_address_globals_dead_stripping); + } else { + AsanUseAfterScope = false; } // Parse -link-cxx-sanitizer flag. @@ -634,7 +640,7 @@ void SanitizerArgs::addArgs(const ToolChain &TC, const llvm::opt::ArgList &Args, std::make_pair(CoverageNoPrune, "-fsanitize-coverage-no-prune")}; for (auto F : CoverageFlags) { if (CoverageFeatures & F.first) - CmdArgs.push_back(Args.MakeArgString(F.second)); + CmdArgs.push_back(F.second); } if (TC.getTriple().isOSWindows() && needsUbsanRt()) { @@ -687,7 +693,7 @@ void SanitizerArgs::addArgs(const ToolChain &TC, const llvm::opt::ArgList &Args, llvm::utostr(MsanTrackOrigins))); if (MsanUseAfterDtor) - CmdArgs.push_back(Args.MakeArgString("-fsanitize-memory-use-after-dtor")); + CmdArgs.push_back("-fsanitize-memory-use-after-dtor"); // FIXME: Pass these parameters as function attributes, not as -llvm flags. if (!TsanMemoryAccess) { @@ -706,17 +712,20 @@ void SanitizerArgs::addArgs(const ToolChain &TC, const llvm::opt::ArgList &Args, } if (CfiCrossDso) - CmdArgs.push_back(Args.MakeArgString("-fsanitize-cfi-cross-dso")); + CmdArgs.push_back("-fsanitize-cfi-cross-dso"); if (Stats) - CmdArgs.push_back(Args.MakeArgString("-fsanitize-stats")); + CmdArgs.push_back("-fsanitize-stats"); if (AsanFieldPadding) CmdArgs.push_back(Args.MakeArgString("-fsanitize-address-field-padding=" + llvm::utostr(AsanFieldPadding))); if (AsanUseAfterScope) - CmdArgs.push_back(Args.MakeArgString("-fsanitize-address-use-after-scope")); + CmdArgs.push_back("-fsanitize-address-use-after-scope"); + + if (AsanGlobalsDeadStripping) + CmdArgs.push_back("-fsanitize-address-globals-dead-stripping"); // MSan: Workaround for PR16386. // ASan: This is mainly to help LSan with cases such as @@ -724,7 +733,7 @@ void SanitizerArgs::addArgs(const ToolChain &TC, const llvm::opt::ArgList &Args, // We can't make this conditional on -fsanitize=leak, as that flag shouldn't // affect compilation. if (Sanitizers.has(Memory) || Sanitizers.has(Address)) - CmdArgs.push_back(Args.MakeArgString("-fno-assume-sane-operator-new")); + CmdArgs.push_back("-fno-assume-sane-operator-new"); // Require -fvisibility= flag on non-Windows when compiling if vptr CFI is // enabled. diff --git a/lib/Driver/ToolChains/Arch/Mips.cpp b/lib/Driver/ToolChains/Arch/Mips.cpp index cd791af83220..f33542477fb5 100644 --- a/lib/Driver/ToolChains/Arch/Mips.cpp +++ b/lib/Driver/ToolChains/Arch/Mips.cpp @@ -282,18 +282,18 @@ void mips::getMIPSTargetFeatures(const Driver &D, const llvm::Triple &Triple, if (Arg *A = Args.getLastArg(options::OPT_mfp32, options::OPT_mfpxx, options::OPT_mfp64)) { if (A->getOption().matches(options::OPT_mfp32)) - Features.push_back(Args.MakeArgString("-fp64")); + Features.push_back("-fp64"); else if (A->getOption().matches(options::OPT_mfpxx)) { - Features.push_back(Args.MakeArgString("+fpxx")); - Features.push_back(Args.MakeArgString("+nooddspreg")); + Features.push_back("+fpxx"); + Features.push_back("+nooddspreg"); } else - Features.push_back(Args.MakeArgString("+fp64")); + Features.push_back("+fp64"); } else if (mips::shouldUseFPXX(Args, Triple, CPUName, ABIName, FloatABI)) { - Features.push_back(Args.MakeArgString("+fpxx")); - Features.push_back(Args.MakeArgString("+nooddspreg")); + Features.push_back("+fpxx"); + Features.push_back("+nooddspreg"); } else if (mips::isFP64ADefault(Triple, CPUName)) { - Features.push_back(Args.MakeArgString("+fp64")); - Features.push_back(Args.MakeArgString("+nooddspreg")); + Features.push_back("+fp64"); + Features.push_back("+nooddspreg"); } AddTargetFeature(Args, Features, options::OPT_mno_odd_spreg, diff --git a/lib/Driver/ToolChains/Myriad.cpp b/lib/Driver/ToolChains/Myriad.cpp index 2935755c12be..f70ce93c45ce 100644 --- a/lib/Driver/ToolChains/Myriad.cpp +++ b/lib/Driver/ToolChains/Myriad.cpp @@ -43,15 +43,17 @@ void tools::SHAVE::Compiler::ConstructJob(Compilation &C, const JobAction &JA, } CmdArgs.push_back("-DMYRIAD2"); - // Append all -I, -iquote, -isystem paths, defines/undefines, - // 'f' flags, optimize flags, and warning options. + // Append all -I, -iquote, -isystem paths, defines/undefines, 'f' + // flags, 'g' flags, 'M' flags, optimize flags, warning options, + // mcpu flags, mllvm flags, and Xclang flags. // These are spelled the same way in clang and moviCompile. Args.AddAllArgsExcept( CmdArgs, {options::OPT_I_Group, options::OPT_clang_i_Group, options::OPT_std_EQ, options::OPT_D, options::OPT_U, options::OPT_f_Group, options::OPT_f_clang_Group, options::OPT_g_Group, options::OPT_M_Group, - options::OPT_O_Group, options::OPT_W_Group, options::OPT_mcpu_EQ}, + options::OPT_O_Group, options::OPT_W_Group, options::OPT_mcpu_EQ, + options::OPT_mllvm, options::OPT_Xclang}, {options::OPT_fno_split_dwarf_inlining}); Args.hasArg(options::OPT_fno_split_dwarf_inlining); // Claim it if present. diff --git a/lib/Driver/ToolChains/WebAssembly.cpp b/lib/Driver/ToolChains/WebAssembly.cpp index 123a1516f1e7..3471569b6884 100644 --- a/lib/Driver/ToolChains/WebAssembly.cpp +++ b/lib/Driver/ToolChains/WebAssembly.cpp @@ -42,7 +42,7 @@ void wasm::Linker::ConstructJob(Compilation &C, const JobAction &JA, const char *Linker = Args.MakeArgString(ToolChain.GetLinkerPath()); ArgStringList CmdArgs; CmdArgs.push_back("-flavor"); - CmdArgs.push_back("ld"); + CmdArgs.push_back("wasm"); // Enable garbage collection of unused input sections by default, since code // size is of particular importance. This is significantly facilitated by @@ -101,6 +101,9 @@ WebAssembly::WebAssembly(const Driver &D, const llvm::Triple &Triple, : ToolChain(D, Triple, Args) { assert(Triple.isArch32Bit() != Triple.isArch64Bit()); + + getProgramPaths().push_back(getDriver().getInstalledDir()); + getFilePaths().push_back( getDriver().SysRoot + "/lib" + (Triple.isArch32Bit() ? "32" : "64")); } diff --git a/lib/Format/ContinuationIndenter.cpp b/lib/Format/ContinuationIndenter.cpp index 3adb72c11da8..488f9dd582f9 100644 --- a/lib/Format/ContinuationIndenter.cpp +++ b/lib/Format/ContinuationIndenter.cpp @@ -674,6 +674,8 @@ unsigned ContinuationIndenter::getNewLineColumn(const LineState &State) { return State.Stack[State.Stack.size() - 2].LastSpace; return State.FirstIndent; } + if (Current.is(tok::r_paren) && State.Stack.size() > 1) + return State.Stack[State.Stack.size() - 2].LastSpace; if (NextNonComment->is(TT_TemplateString) && NextNonComment->closesScope()) return State.Stack[State.Stack.size() - 2].LastSpace; if (Current.is(tok::identifier) && Current.Next && @@ -920,6 +922,10 @@ void ContinuationIndenter::moveStatePastFakeLParens(LineState &State, NewParenState.NoLineBreak = NewParenState.NoLineBreak || State.Stack.back().NoLineBreakInOperand; + // Don't propagate AvoidBinPacking into subexpressions of arg/param lists. + if (*I > prec::Comma) + NewParenState.AvoidBinPacking = false; + // Indent from 'LastSpace' unless these are fake parentheses encapsulating // a builder type call after 'return' or, if the alignment after opening // brackets is disabled. @@ -1034,13 +1040,20 @@ void ContinuationIndenter::moveStatePastScopeOpener(LineState &State, NestedBlockIndent = Column; } + bool EndsInComma = + Current.MatchingParen && + Current.MatchingParen->getPreviousNonComment() && + Current.MatchingParen->getPreviousNonComment()->is(tok::comma); + AvoidBinPacking = + (Style.Language == FormatStyle::LK_JavaScript && EndsInComma) || (State.Line->MustBeDeclaration && !Style.BinPackParameters) || (!State.Line->MustBeDeclaration && !Style.BinPackArguments) || (Style.ExperimentalAutoDetectBinPacking && (Current.PackingKind == PPK_OnePerLine || (!BinPackInconclusiveFunctions && Current.PackingKind == PPK_Inconclusive))); + if (Current.is(TT_ObjCMethodExpr) && Current.MatchingParen) { if (Style.ColumnLimit) { // If this '[' opens an ObjC call, determine whether all parameters fit @@ -1061,6 +1074,9 @@ void ContinuationIndenter::moveStatePastScopeOpener(LineState &State, } } } + + if (Style.Language == FormatStyle::LK_JavaScript && EndsInComma) + BreakBeforeParameter = true; } // Generally inherit NoLineBreak from the current scope to nested scope. // However, don't do this for non-empty nested blocks, dict literals and diff --git a/lib/Format/Format.cpp b/lib/Format/Format.cpp index f55a623a8d1f..c8677e805179 100644 --- a/lib/Format/Format.cpp +++ b/lib/Format/Format.cpp @@ -171,6 +171,18 @@ template <> struct ScalarEnumerationTraits { } }; +template <> struct ScalarEnumerationTraits { + static void enumeration(IO &IO, FormatStyle::EscapedNewlineAlignmentStyle &Value) { + IO.enumCase(Value, "DontAlign", FormatStyle::ENAS_DontAlign); + IO.enumCase(Value, "Left", FormatStyle::ENAS_Left); + IO.enumCase(Value, "Right", FormatStyle::ENAS_Right); + + // For backward compatibility. + IO.enumCase(Value, "true", FormatStyle::ENAS_Left); + IO.enumCase(Value, "false", FormatStyle::ENAS_Right); + } +}; + template <> struct ScalarEnumerationTraits { static void enumeration(IO &IO, FormatStyle::PointerAlignmentStyle &Value) { IO.enumCase(Value, "Middle", FormatStyle::PAS_Middle); @@ -233,6 +245,7 @@ template <> struct MappingTraits { // For backward compatibility. if (!IO.outputting()) { + IO.mapOptional("AlignEscapedNewlinesLeft", Style.AlignEscapedNewlines); IO.mapOptional("DerivePointerBinding", Style.DerivePointerAlignment); IO.mapOptional("IndentFunctionDeclarationAfterType", Style.IndentWrappedFunctionNames); @@ -247,7 +260,7 @@ template <> struct MappingTraits { Style.AlignConsecutiveAssignments); IO.mapOptional("AlignConsecutiveDeclarations", Style.AlignConsecutiveDeclarations); - IO.mapOptional("AlignEscapedNewlinesLeft", Style.AlignEscapedNewlinesLeft); + IO.mapOptional("AlignEscapedNewlines", Style.AlignEscapedNewlines); IO.mapOptional("AlignOperands", Style.AlignOperands); IO.mapOptional("AlignTrailingComments", Style.AlignTrailingComments); IO.mapOptional("AllowAllParametersOfDeclarationOnNextLine", @@ -498,7 +511,7 @@ FormatStyle getLLVMStyle() { FormatStyle LLVMStyle; LLVMStyle.Language = FormatStyle::LK_Cpp; LLVMStyle.AccessModifierOffset = -2; - LLVMStyle.AlignEscapedNewlinesLeft = false; + LLVMStyle.AlignEscapedNewlines = FormatStyle::ENAS_Right; LLVMStyle.AlignAfterOpenBracket = FormatStyle::BAS_Align; LLVMStyle.AlignOperands = true; LLVMStyle.AlignTrailingComments = true; @@ -587,7 +600,7 @@ FormatStyle getGoogleStyle(FormatStyle::LanguageKind Language) { GoogleStyle.Language = Language; GoogleStyle.AccessModifierOffset = -1; - GoogleStyle.AlignEscapedNewlinesLeft = true; + GoogleStyle.AlignEscapedNewlines = FormatStyle::ENAS_Left; GoogleStyle.AllowShortIfStatementsOnASingleLine = true; GoogleStyle.AllowShortLoopsOnASingleLine = true; GoogleStyle.AlwaysBreakBeforeMultilineStrings = true; @@ -624,9 +637,10 @@ FormatStyle getGoogleStyle(FormatStyle::LanguageKind Language) { GoogleStyle.AllowShortFunctionsOnASingleLine = FormatStyle::SFS_Empty; GoogleStyle.AlwaysBreakBeforeMultilineStrings = false; GoogleStyle.BreakBeforeTernaryOperators = false; - // taze:, @tag followed by { for a lot of JSDoc tags, and @see, which is - // commonly followed by overlong URLs. - GoogleStyle.CommentPragmas = "(taze:|(@[A-Za-z_0-9-]+[ \\t]*{)|@see)"; + // taze:, triple slash directives (`/// <...`), @tag followed by { for a lot + // of JSDoc tags, and @see, which is commonly followed by overlong URLs. + GoogleStyle.CommentPragmas = + "(taze:|^/[ \t]*<|(@[A-Za-z_0-9-]+[ \\t]*{)|@see)"; GoogleStyle.MaxEmptyLinesToKeep = 3; GoogleStyle.NamespaceIndentation = FormatStyle::NI_All; GoogleStyle.SpacesInContainerLiterals = false; diff --git a/lib/Format/TokenAnnotator.cpp b/lib/Format/TokenAnnotator.cpp index c274d7bf07f8..387768a6ee56 100644 --- a/lib/Format/TokenAnnotator.cpp +++ b/lib/Format/TokenAnnotator.cpp @@ -576,9 +576,12 @@ class AnnotatingParser { } break; case tok::kw_for: - if (Style.Language == FormatStyle::LK_JavaScript && Tok->Previous && - Tok->Previous->is(tok::period)) - break; + if (Style.Language == FormatStyle::LK_JavaScript) + if (Tok->Previous && Tok->Previous->is(tok::period)) + break; + // JS' for async ( ... + if (CurrentToken->is(Keywords.kw_async)) + next(); Contexts.back().ColonIsForRangeExpr = true; next(); if (!parseParens()) @@ -1034,8 +1037,9 @@ class AnnotatingParser { if (Style.Language == FormatStyle::LK_JavaScript) { if (Current.is(tok::exclaim)) { if (Current.Previous && - (Current.Previous->isOneOf(tok::identifier, tok::r_paren, - tok::r_square, tok::r_brace) || + (Current.Previous->isOneOf(tok::identifier, tok::kw_namespace, + tok::r_paren, tok::r_square, + tok::r_brace) || Current.Previous->Tok.isLiteral())) { Current.Type = TT_JsNonNullAssertion; return; @@ -2248,6 +2252,10 @@ bool TokenAnnotator::spaceRequiredBefore(const AnnotatedLine &Line, } else if (Style.Language == FormatStyle::LK_JavaScript) { if (Left.is(TT_JsFatArrow)) return true; + // for async ( ... + if (Right.is(tok::l_paren) && Left.is(Keywords.kw_async) && + Left.Previous && Left.Previous->is(tok::kw_for)) + return true; if (Left.is(Keywords.kw_async) && Right.is(tok::l_paren) && Right.MatchingParen) { const FormatToken *Next = Right.MatchingParen->getNextNonComment(); @@ -2462,16 +2470,20 @@ bool TokenAnnotator::mustBreakBefore(const AnnotatedLine &Line, return true; } - // If the last token before a '}' is a comma or a trailing comment, the - // intention is to insert a line break after it in order to make shuffling - // around entries easier. + // If the last token before a '}', ']', or ')' is a comma or a trailing + // comment, the intention is to insert a line break after it in order to make + // shuffling around entries easier. const FormatToken *BeforeClosingBrace = nullptr; - if (Left.isOneOf(tok::l_brace, TT_ArrayInitializerLSquare) && + if ((Left.isOneOf(tok::l_brace, TT_ArrayInitializerLSquare) || + (Style.Language == FormatStyle::LK_JavaScript && + Left.is(tok::l_paren))) && Left.BlockKind != BK_Block && Left.MatchingParen) BeforeClosingBrace = Left.MatchingParen->Previous; else if (Right.MatchingParen && - Right.MatchingParen->isOneOf(tok::l_brace, - TT_ArrayInitializerLSquare)) + (Right.MatchingParen->isOneOf(tok::l_brace, + TT_ArrayInitializerLSquare) || + (Style.Language == FormatStyle::LK_JavaScript && + Right.MatchingParen->is(tok::l_paren)))) BeforeClosingBrace = &Left; if (BeforeClosingBrace && (BeforeClosingBrace->is(tok::comma) || BeforeClosingBrace->isTrailingComment())) diff --git a/lib/Format/UnwrappedLineParser.cpp b/lib/Format/UnwrappedLineParser.cpp index 2d788b52dfda..31c66ffb00a1 100644 --- a/lib/Format/UnwrappedLineParser.cpp +++ b/lib/Format/UnwrappedLineParser.cpp @@ -368,9 +368,10 @@ void UnwrappedLineParser::calculateBraceTypes(bool ExpectClassBody) { (Style.Language == FormatStyle::LK_JavaScript && NextTok->isOneOf(Keywords.kw_of, Keywords.kw_in, Keywords.kw_as)) || + (Style.isCpp() && NextTok->is(tok::l_paren)) || NextTok->isOneOf(tok::comma, tok::period, tok::colon, tok::r_paren, tok::r_square, tok::l_brace, - tok::l_square, tok::l_paren, tok::ellipsis) || + tok::l_square, tok::ellipsis) || (NextTok->is(tok::identifier) && !PrevTok->isOneOf(tok::semi, tok::r_brace, tok::l_brace)) || (NextTok->is(tok::semi) && @@ -476,6 +477,24 @@ static bool isGoogScope(const UnwrappedLine &Line) { return I->Tok->is(tok::l_paren); } +static bool isIIFE(const UnwrappedLine &Line, + const AdditionalKeywords &Keywords) { + // Look for the start of an immediately invoked anonymous function. + // https://en.wikipedia.org/wiki/Immediately-invoked_function_expression + // This is commonly done in JavaScript to create a new, anonymous scope. + // Example: (function() { ... })() + if (Line.Tokens.size() < 3) + return false; + auto I = Line.Tokens.begin(); + if (I->Tok->isNot(tok::l_paren)) + return false; + ++I; + if (I->Tok->isNot(Keywords.kw_function)) + return false; + ++I; + return I->Tok->is(tok::l_paren); +} + static bool ShouldBreakBeforeBrace(const FormatStyle &Style, const FormatToken &InitialToken) { if (InitialToken.is(tok::kw_namespace)) @@ -493,15 +512,16 @@ void UnwrappedLineParser::parseChildBlock() { FormatTok->BlockKind = BK_Block; nextToken(); { - bool GoogScope = - Style.Language == FormatStyle::LK_JavaScript && isGoogScope(*Line); + bool SkipIndent = + (Style.Language == FormatStyle::LK_JavaScript && + (isGoogScope(*Line) || isIIFE(*Line, Keywords))); ScopedLineState LineState(*this); ScopedDeclarationState DeclarationState(*Line, DeclarationScopeStack, /*MustBeDeclaration=*/false); - Line->Level += GoogScope ? 0 : 1; + Line->Level += SkipIndent ? 0 : 1; parseLevel(/*HasOpeningBrace=*/true); flushComments(isOnNewLine(*FormatTok)); - Line->Level -= GoogScope ? 0 : 1; + Line->Level -= SkipIndent ? 0 : 1; } nextToken(); } @@ -1615,6 +1635,10 @@ void UnwrappedLineParser::parseForOrWhileLoop() { assert(FormatTok->isOneOf(tok::kw_for, tok::kw_while, TT_ForEachMacro) && "'for', 'while' or foreach macro expected"); nextToken(); + // JS' for async ( ... + if (Style.Language == FormatStyle::LK_JavaScript && + FormatTok->is(Keywords.kw_async)) + nextToken(); if (FormatTok->Tok.is(tok::l_paren)) parseParens(); if (FormatTok->Tok.is(tok::l_brace)) { diff --git a/lib/Format/WhitespaceManager.cpp b/lib/Format/WhitespaceManager.cpp index 2c1f59324971..3b6311d15487 100644 --- a/lib/Format/WhitespaceManager.cpp +++ b/lib/Format/WhitespaceManager.cpp @@ -517,8 +517,11 @@ void WhitespaceManager::alignTrailingComments(unsigned Start, unsigned End, } void WhitespaceManager::alignEscapedNewlines() { - unsigned MaxEndOfLine = - Style.AlignEscapedNewlinesLeft ? 0 : Style.ColumnLimit; + if (Style.AlignEscapedNewlines == FormatStyle::ENAS_DontAlign) + return; + + bool AlignLeft = Style.AlignEscapedNewlines == FormatStyle::ENAS_Left; + unsigned MaxEndOfLine = AlignLeft ? 0 : Style.ColumnLimit; unsigned StartOfMacro = 0; for (unsigned i = 1, e = Changes.size(); i < e; ++i) { Change &C = Changes[i]; @@ -527,7 +530,7 @@ void WhitespaceManager::alignEscapedNewlines() { MaxEndOfLine = std::max(C.PreviousEndOfTokenColumn + 2, MaxEndOfLine); } else { alignEscapedNewlines(StartOfMacro + 1, i, MaxEndOfLine); - MaxEndOfLine = Style.AlignEscapedNewlinesLeft ? 0 : Style.ColumnLimit; + MaxEndOfLine = AlignLeft ? 0 : Style.ColumnLimit; StartOfMacro = i; } } @@ -602,7 +605,7 @@ void WhitespaceManager::appendNewlineText(std::string &Text, unsigned Newlines, unsigned EscapedNewlineColumn) { if (Newlines > 0) { unsigned Offset = - std::min(EscapedNewlineColumn - 1, PreviousEndOfTokenColumn); + std::min(EscapedNewlineColumn - 2, PreviousEndOfTokenColumn); for (unsigned i = 0; i < Newlines; ++i) { Text.append(EscapedNewlineColumn - Offset - 1, ' '); Text.append(UseCRLF ? "\\\r\n" : "\\\n"); diff --git a/lib/Frontend/CompilerInvocation.cpp b/lib/Frontend/CompilerInvocation.cpp index 96854b8fbc1a..51147b6f9499 100644 --- a/lib/Frontend/CompilerInvocation.cpp +++ b/lib/Frontend/CompilerInvocation.cpp @@ -778,6 +778,8 @@ static bool ParseCodeGenArgs(CodeGenOptions &Opts, ArgList &Args, InputKind IK, Opts.SanitizeAddressUseAfterScope = A->getOption().getID() == OPT_fsanitize_address_use_after_scope; } + Opts.SanitizeAddressGlobalsDeadStripping = + Args.hasArg(OPT_fsanitize_address_globals_dead_stripping); Opts.SSPBufferSize = getLastArgIntValue(Args, OPT_stack_protector_buffer_size, 8, Diags); Opts.StackRealignment = Args.hasArg(OPT_mstackrealign); diff --git a/lib/Headers/avxintrin.h b/lib/Headers/avxintrin.h index cdb7aa4fb626..15d28259dcc0 100644 --- a/lib/Headers/avxintrin.h +++ b/lib/Headers/avxintrin.h @@ -1458,12 +1458,13 @@ _mm256_blendv_ps(__m256 __a, __m256 __b, __m256 __c) /// \brief Computes two dot products in parallel, using the lower and upper /// halves of two [8 x float] vectors as input to the two computations, and /// returning the two dot products in the lower and upper halves of the -/// [8 x float] result. The immediate integer operand controls which input -/// elements will contribute to the dot product, and where the final results -/// are returned. In general, for each dot product, the four corresponding -/// elements of the input vectors are multiplied; the first two and second -/// two products are summed, then the two sums are added to form the final -/// result. +/// [8 x float] result. +/// +/// The immediate integer operand controls which input elements will +/// contribute to the dot product, and where the final results are returned. +/// In general, for each dot product, the four corresponding elements of the +/// input vectors are multiplied; the first two and second two products are +/// summed, then the two sums are added to form the final result. /// /// \headerfile /// @@ -1497,15 +1498,16 @@ _mm256_blendv_ps(__m256 __a, __m256 __b, __m256 __c) /* Vector shuffle */ /// \brief Selects 8 float values from the 256-bit operands of [8 x float], as -/// specified by the immediate value operand. The four selected elements in -/// each operand are copied to the destination according to the bits -/// specified in the immediate operand. The selected elements from the first -/// 256-bit operand are copied to bits [63:0] and bits [191:128] of the -/// destination, and the selected elements from the second 256-bit operand -/// are copied to bits [127:64] and bits [255:192] of the destination. For -/// example, if bits [7:0] of the immediate operand contain a value of 0xFF, -/// the 256-bit destination vector would contain the following values: b[7], -/// b[7], a[7], a[7], b[3], b[3], a[3], a[3]. +/// specified by the immediate value operand. +/// +/// The four selected elements in each operand are copied to the destination +/// according to the bits specified in the immediate operand. The selected +/// elements from the first 256-bit operand are copied to bits [63:0] and +/// bits [191:128] of the destination, and the selected elements from the +/// second 256-bit operand are copied to bits [127:64] and bits [255:192] of +/// the destination. For example, if bits [7:0] of the immediate operand +/// contain a value of 0xFF, the 256-bit destination vector would contain the +/// following values: b[7], b[7], a[7], a[7], b[3], b[3], a[3], a[3]. /// /// \headerfile /// @@ -1557,13 +1559,14 @@ _mm256_blendv_ps(__m256 __a, __m256 __b, __m256 __c) 12 + (((mask) >> 6) & 0x3)); }) /// \brief Selects four double-precision values from the 256-bit operands of -/// [4 x double], as specified by the immediate value operand. The selected -/// elements from the first 256-bit operand are copied to bits [63:0] and -/// bits [191:128] in the destination, and the selected elements from the -/// second 256-bit operand are copied to bits [127:64] and bits [255:192] in -/// the destination. For example, if bits [3:0] of the immediate operand -/// contain a value of 0xF, the 256-bit destination vector would contain the -/// following values: b[3], a[3], b[1], a[1]. +/// [4 x double], as specified by the immediate value operand. +/// +/// The selected elements from the first 256-bit operand are copied to bits +/// [63:0] and bits [191:128] in the destination, and the selected elements +/// from the second 256-bit operand are copied to bits [127:64] and bits +/// [255:192] in the destination. For example, if bits [3:0] of the immediate +/// operand contain a value of 0xF, the 256-bit destination vector would +/// contain the following values: b[3], a[3], b[1], a[1]. /// /// \headerfile /// @@ -1641,9 +1644,11 @@ _mm256_blendv_ps(__m256 __a, __m256 __b, __m256 __c) /// \brief Compares each of the corresponding double-precision values of two /// 128-bit vectors of [2 x double], using the operation specified by the -/// immediate integer operand. Returns a [2 x double] vector consisting of -/// two doubles corresponding to the two comparison results: zero if the -/// comparison is false, and all 1's if the comparison is true. +/// immediate integer operand. +/// +/// Returns a [2 x double] vector consisting of two doubles corresponding to +/// the two comparison results: zero if the comparison is false, and all 1's +/// if the comparison is true. /// /// \headerfile /// @@ -1699,9 +1704,11 @@ _mm256_blendv_ps(__m256 __a, __m256 __b, __m256 __c) /// \brief Compares each of the corresponding values of two 128-bit vectors of /// [4 x float], using the operation specified by the immediate integer -/// operand. Returns a [4 x float] vector consisting of four floats -/// corresponding to the four comparison results: zero if the comparison is -/// false, and all 1's if the comparison is true. +/// operand. +/// +/// Returns a [4 x float] vector consisting of four floats corresponding to +/// the four comparison results: zero if the comparison is false, and all 1's +/// if the comparison is true. /// /// \headerfile /// @@ -1757,9 +1764,11 @@ _mm256_blendv_ps(__m256 __a, __m256 __b, __m256 __c) /// \brief Compares each of the corresponding double-precision values of two /// 256-bit vectors of [4 x double], using the operation specified by the -/// immediate integer operand. Returns a [4 x double] vector consisting of -/// four doubles corresponding to the four comparison results: zero if the -/// comparison is false, and all 1's if the comparison is true. +/// immediate integer operand. +/// +/// Returns a [4 x double] vector consisting of four doubles corresponding to +/// the four comparison results: zero if the comparison is false, and all 1's +/// if the comparison is true. /// /// \headerfile /// @@ -1815,9 +1824,11 @@ _mm256_blendv_ps(__m256 __a, __m256 __b, __m256 __c) /// \brief Compares each of the corresponding values of two 256-bit vectors of /// [8 x float], using the operation specified by the immediate integer -/// operand. Returns a [8 x float] vector consisting of eight floats -/// corresponding to the eight comparison results: zero if the comparison is -/// false, and all 1's if the comparison is true. +/// operand. +/// +/// Returns a [8 x float] vector consisting of eight floats corresponding to +/// the eight comparison results: zero if the comparison is false, and all +/// 1's if the comparison is true. /// /// \headerfile /// @@ -1873,8 +1884,10 @@ _mm256_blendv_ps(__m256 __a, __m256 __b, __m256 __c) /// \brief Compares each of the corresponding scalar double-precision values of /// two 128-bit vectors of [2 x double], using the operation specified by the -/// immediate integer operand. If the result is true, all 64 bits of the -/// destination vector are set; otherwise they are cleared. +/// immediate integer operand. +/// +/// If the result is true, all 64 bits of the destination vector are set; +/// otherwise they are cleared. /// /// \headerfile /// @@ -1930,8 +1943,10 @@ _mm256_blendv_ps(__m256 __a, __m256 __b, __m256 __c) /// \brief Compares each of the corresponding scalar values of two 128-bit /// vectors of [4 x float], using the operation specified by the immediate -/// integer operand. If the result is true, all 32 bits of the destination -/// vector are set; otherwise they are cleared. +/// integer operand. +/// +/// If the result is true, all 32 bits of the destination vector are set; +/// otherwise they are cleared. /// /// \headerfile /// @@ -2536,7 +2551,9 @@ _mm256_unpacklo_ps(__m256 __a, __m256 __b) /// \brief Given two 128-bit floating-point vectors of [2 x double], perform an /// element-by-element comparison of the double-precision element in the /// first source vector and the corresponding element in the second source -/// vector. The EFLAGS register is updated as follows: \n +/// vector. +/// +/// The EFLAGS register is updated as follows: \n /// If there is at least one pair of double-precision elements where the /// sign-bits of both elements are 1, the ZF flag is set to 0. Otherwise the /// ZF flag is set to 1. \n @@ -2563,7 +2580,9 @@ _mm_testz_pd(__m128d __a, __m128d __b) /// \brief Given two 128-bit floating-point vectors of [2 x double], perform an /// element-by-element comparison of the double-precision element in the /// first source vector and the corresponding element in the second source -/// vector. The EFLAGS register is updated as follows: \n +/// vector. +/// +/// The EFLAGS register is updated as follows: \n /// If there is at least one pair of double-precision elements where the /// sign-bits of both elements are 1, the ZF flag is set to 0. Otherwise the /// ZF flag is set to 1. \n @@ -2590,7 +2609,9 @@ _mm_testc_pd(__m128d __a, __m128d __b) /// \brief Given two 128-bit floating-point vectors of [2 x double], perform an /// element-by-element comparison of the double-precision element in the /// first source vector and the corresponding element in the second source -/// vector. The EFLAGS register is updated as follows: \n +/// vector. +/// +/// The EFLAGS register is updated as follows: \n /// If there is at least one pair of double-precision elements where the /// sign-bits of both elements are 1, the ZF flag is set to 0. Otherwise the /// ZF flag is set to 1. \n @@ -2618,7 +2639,9 @@ _mm_testnzc_pd(__m128d __a, __m128d __b) /// \brief Given two 128-bit floating-point vectors of [4 x float], perform an /// element-by-element comparison of the single-precision element in the /// first source vector and the corresponding element in the second source -/// vector. The EFLAGS register is updated as follows: \n +/// vector. +/// +/// The EFLAGS register is updated as follows: \n /// If there is at least one pair of single-precision elements where the /// sign-bits of both elements are 1, the ZF flag is set to 0. Otherwise the /// ZF flag is set to 1. \n @@ -2645,7 +2668,9 @@ _mm_testz_ps(__m128 __a, __m128 __b) /// \brief Given two 128-bit floating-point vectors of [4 x float], perform an /// element-by-element comparison of the single-precision element in the /// first source vector and the corresponding element in the second source -/// vector. The EFLAGS register is updated as follows: \n +/// vector. +/// +/// The EFLAGS register is updated as follows: \n /// If there is at least one pair of single-precision elements where the /// sign-bits of both elements are 1, the ZF flag is set to 0. Otherwise the /// ZF flag is set to 1. \n @@ -2672,7 +2697,9 @@ _mm_testc_ps(__m128 __a, __m128 __b) /// \brief Given two 128-bit floating-point vectors of [4 x float], perform an /// element-by-element comparison of the single-precision element in the /// first source vector and the corresponding element in the second source -/// vector. The EFLAGS register is updated as follows: \n +/// vector. +/// +/// The EFLAGS register is updated as follows: \n /// If there is at least one pair of single-precision elements where the /// sign-bits of both elements are 1, the ZF flag is set to 0. Otherwise the /// ZF flag is set to 1. \n @@ -2700,7 +2727,9 @@ _mm_testnzc_ps(__m128 __a, __m128 __b) /// \brief Given two 256-bit floating-point vectors of [4 x double], perform an /// element-by-element comparison of the double-precision elements in the /// first source vector and the corresponding elements in the second source -/// vector. The EFLAGS register is updated as follows: \n +/// vector. +/// +/// The EFLAGS register is updated as follows: \n /// If there is at least one pair of double-precision elements where the /// sign-bits of both elements are 1, the ZF flag is set to 0. Otherwise the /// ZF flag is set to 1. \n @@ -2727,7 +2756,9 @@ _mm256_testz_pd(__m256d __a, __m256d __b) /// \brief Given two 256-bit floating-point vectors of [4 x double], perform an /// element-by-element comparison of the double-precision elements in the /// first source vector and the corresponding elements in the second source -/// vector. The EFLAGS register is updated as follows: \n +/// vector. +/// +/// The EFLAGS register is updated as follows: \n /// If there is at least one pair of double-precision elements where the /// sign-bits of both elements are 1, the ZF flag is set to 0. Otherwise the /// ZF flag is set to 1. \n @@ -2754,7 +2785,9 @@ _mm256_testc_pd(__m256d __a, __m256d __b) /// \brief Given two 256-bit floating-point vectors of [4 x double], perform an /// element-by-element comparison of the double-precision elements in the /// first source vector and the corresponding elements in the second source -/// vector. The EFLAGS register is updated as follows: \n +/// vector. +/// +/// The EFLAGS register is updated as follows: \n /// If there is at least one pair of double-precision elements where the /// sign-bits of both elements are 1, the ZF flag is set to 0. Otherwise the /// ZF flag is set to 1. \n @@ -2782,7 +2815,9 @@ _mm256_testnzc_pd(__m256d __a, __m256d __b) /// \brief Given two 256-bit floating-point vectors of [8 x float], perform an /// element-by-element comparison of the single-precision element in the /// first source vector and the corresponding element in the second source -/// vector. The EFLAGS register is updated as follows: \n +/// vector. +/// +/// The EFLAGS register is updated as follows: \n /// If there is at least one pair of single-precision elements where the /// sign-bits of both elements are 1, the ZF flag is set to 0. Otherwise the /// ZF flag is set to 1. \n @@ -2809,7 +2844,9 @@ _mm256_testz_ps(__m256 __a, __m256 __b) /// \brief Given two 256-bit floating-point vectors of [8 x float], perform an /// element-by-element comparison of the single-precision element in the /// first source vector and the corresponding element in the second source -/// vector. The EFLAGS register is updated as follows: \n +/// vector. +/// +/// The EFLAGS register is updated as follows: \n /// If there is at least one pair of single-precision elements where the /// sign-bits of both elements are 1, the ZF flag is set to 0. Otherwise the /// ZF flag is set to 1. \n @@ -2836,7 +2873,9 @@ _mm256_testc_ps(__m256 __a, __m256 __b) /// \brief Given two 256-bit floating-point vectors of [8 x float], perform an /// element-by-element comparison of the single-precision elements in the /// first source vector and the corresponding elements in the second source -/// vector. The EFLAGS register is updated as follows: \n +/// vector. +/// +/// The EFLAGS register is updated as follows: \n /// If there is at least one pair of single-precision elements where the /// sign-bits of both elements are 1, the ZF flag is set to 0. Otherwise the /// ZF flag is set to 1. \n @@ -2862,7 +2901,9 @@ _mm256_testnzc_ps(__m256 __a, __m256 __b) } /// \brief Given two 256-bit integer vectors, perform a bit-by-bit comparison -/// of the two source vectors and update the EFLAGS register as follows: \n +/// of the two source vectors. +/// +/// The EFLAGS register is updated as follows: \n /// If there is at least one pair of bits where both bits are 1, the ZF flag /// is set to 0. Otherwise the ZF flag is set to 1. \n /// If there is at least one pair of bits where the bit from the first source @@ -2886,7 +2927,9 @@ _mm256_testz_si256(__m256i __a, __m256i __b) } /// \brief Given two 256-bit integer vectors, perform a bit-by-bit comparison -/// of the two source vectors and update the EFLAGS register as follows: \n +/// of the two source vectors. +/// +/// The EFLAGS register is updated as follows: \n /// If there is at least one pair of bits where both bits are 1, the ZF flag /// is set to 0. Otherwise the ZF flag is set to 1. \n /// If there is at least one pair of bits where the bit from the first source @@ -2910,7 +2953,9 @@ _mm256_testc_si256(__m256i __a, __m256i __b) } /// \brief Given two 256-bit integer vectors, perform a bit-by-bit comparison -/// of the two source vectors and update the EFLAGS register as follows: \n +/// of the two source vectors. +/// +/// The EFLAGS register is updated as follows: \n /// If there is at least one pair of bits where both bits are 1, the ZF flag /// is set to 0. Otherwise the ZF flag is set to 1. \n /// If there is at least one pair of bits where the bit from the first source @@ -4466,9 +4511,10 @@ _mm256_castsi256_si128(__m256i __a) } /// \brief Constructs a 256-bit floating-point vector of [4 x double] from a -/// 128-bit floating-point vector of [2 x double]. The lower 128 bits -/// contain the value of the source vector. The contents of the upper 128 -/// bits are undefined. +/// 128-bit floating-point vector of [2 x double]. +/// +/// The lower 128 bits contain the value of the source vector. The contents +/// of the upper 128 bits are undefined. /// /// \headerfile /// @@ -4486,9 +4532,10 @@ _mm256_castpd128_pd256(__m128d __a) } /// \brief Constructs a 256-bit floating-point vector of [8 x float] from a -/// 128-bit floating-point vector of [4 x float]. The lower 128 bits contain -/// the value of the source vector. The contents of the upper 128 bits are -/// undefined. +/// 128-bit floating-point vector of [4 x float]. +/// +/// The lower 128 bits contain the value of the source vector. The contents +/// of the upper 128 bits are undefined. /// /// \headerfile /// @@ -4506,6 +4553,7 @@ _mm256_castps128_ps256(__m128 __a) } /// \brief Constructs a 256-bit integer vector from a 128-bit integer vector. +/// /// The lower 128 bits contain the value of the source vector. The contents /// of the upper 128 bits are undefined. /// @@ -4586,8 +4634,10 @@ _mm256_zextsi128_si256(__m128i __a) /// \brief Constructs a new 256-bit vector of [8 x float] by first duplicating /// a 256-bit vector of [8 x float] given in the first parameter, and then /// replacing either the upper or the lower 128 bits with the contents of a -/// 128-bit vector of [4 x float] in the second parameter. The immediate -/// integer parameter determines between the upper or the lower 128 bits. +/// 128-bit vector of [4 x float] in the second parameter. +/// +/// The immediate integer parameter determines between the upper or the lower +/// 128 bits. /// /// \headerfile /// @@ -4631,8 +4681,10 @@ _mm256_zextsi128_si256(__m128i __a) /// \brief Constructs a new 256-bit vector of [4 x double] by first duplicating /// a 256-bit vector of [4 x double] given in the first parameter, and then /// replacing either the upper or the lower 128 bits with the contents of a -/// 128-bit vector of [2 x double] in the second parameter. The immediate -/// integer parameter determines between the upper or the lower 128 bits. +/// 128-bit vector of [2 x double] in the second parameter. +/// +/// The immediate integer parameter determines between the upper or the lower +/// 128 bits. /// /// \headerfile /// @@ -4672,8 +4724,10 @@ _mm256_zextsi128_si256(__m128i __a) /// \brief Constructs a new 256-bit integer vector by first duplicating a /// 256-bit integer vector given in the first parameter, and then replacing /// either the upper or the lower 128 bits with the contents of a 128-bit -/// integer vector in the second parameter. The immediate integer parameter -/// determines between the upper or the lower 128 bits. +/// integer vector in the second parameter. +/// +/// The immediate integer parameter determines between the upper or the lower +/// 128 bits. /// /// \headerfile /// diff --git a/lib/Headers/emmintrin.h b/lib/Headers/emmintrin.h index 13b0db22ec44..fa5cefadc52c 100644 --- a/lib/Headers/emmintrin.h +++ b/lib/Headers/emmintrin.h @@ -462,8 +462,9 @@ _mm_cmplt_pd(__m128d __a, __m128d __b) /// \brief Compares each of the corresponding double-precision values of the /// 128-bit vectors of [2 x double] to determine if the values in the first -/// operand are less than or equal to those in the second operand. Each -/// comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true. +/// operand are less than or equal to those in the second operand. +/// +/// Each comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true. /// /// \headerfile /// @@ -482,8 +483,9 @@ _mm_cmple_pd(__m128d __a, __m128d __b) /// \brief Compares each of the corresponding double-precision values of the /// 128-bit vectors of [2 x double] to determine if the values in the first -/// operand are greater than those in the second operand. Each comparison -/// yields 0h for false, FFFFFFFFFFFFFFFFh for true. +/// operand are greater than those in the second operand. +/// +/// Each comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true. /// /// \headerfile /// @@ -502,8 +504,9 @@ _mm_cmpgt_pd(__m128d __a, __m128d __b) /// \brief Compares each of the corresponding double-precision values of the /// 128-bit vectors of [2 x double] to determine if the values in the first -/// operand are greater than or equal to those in the second operand. Each -/// comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true. +/// operand are greater than or equal to those in the second operand. +/// +/// Each comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true. /// /// \headerfile /// @@ -522,9 +525,10 @@ _mm_cmpge_pd(__m128d __a, __m128d __b) /// \brief Compares each of the corresponding double-precision values of the /// 128-bit vectors of [2 x double] to determine if the values in the first -/// operand are ordered with respect to those in the second operand. A pair -/// of double-precision values are "ordered" with respect to each other if -/// neither value is a NaN. Each comparison yields 0h for false, +/// operand are ordered with respect to those in the second operand. +/// +/// A pair of double-precision values are "ordered" with respect to each +/// other if neither value is a NaN. Each comparison yields 0h for false, /// FFFFFFFFFFFFFFFFh for true. /// /// \headerfile @@ -544,9 +548,10 @@ _mm_cmpord_pd(__m128d __a, __m128d __b) /// \brief Compares each of the corresponding double-precision values of the /// 128-bit vectors of [2 x double] to determine if the values in the first -/// operand are unordered with respect to those in the second operand. A pair -/// of double-precision values are "unordered" with respect to each other if -/// one or both values are NaN. Each comparison yields 0h for false, +/// operand are unordered with respect to those in the second operand. +/// +/// A pair of double-precision values are "unordered" with respect to each +/// other if one or both values are NaN. Each comparison yields 0h for false, /// FFFFFFFFFFFFFFFFh for true. /// /// \headerfile @@ -567,8 +572,9 @@ _mm_cmpunord_pd(__m128d __a, __m128d __b) /// \brief Compares each of the corresponding double-precision values of the /// 128-bit vectors of [2 x double] to determine if the values in the first -/// operand are unequal to those in the second operand. Each comparison -/// yields 0h for false, FFFFFFFFFFFFFFFFh for true. +/// operand are unequal to those in the second operand. +/// +/// Each comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true. /// /// \headerfile /// @@ -587,8 +593,9 @@ _mm_cmpneq_pd(__m128d __a, __m128d __b) /// \brief Compares each of the corresponding double-precision values of the /// 128-bit vectors of [2 x double] to determine if the values in the first -/// operand are not less than those in the second operand. Each comparison -/// yields 0h for false, FFFFFFFFFFFFFFFFh for true. +/// operand are not less than those in the second operand. +/// +/// Each comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true. /// /// \headerfile /// @@ -607,8 +614,9 @@ _mm_cmpnlt_pd(__m128d __a, __m128d __b) /// \brief Compares each of the corresponding double-precision values of the /// 128-bit vectors of [2 x double] to determine if the values in the first -/// operand are not less than or equal to those in the second operand. Each -/// comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true. +/// operand are not less than or equal to those in the second operand. +/// +/// Each comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true. /// /// \headerfile /// @@ -627,8 +635,9 @@ _mm_cmpnle_pd(__m128d __a, __m128d __b) /// \brief Compares each of the corresponding double-precision values of the /// 128-bit vectors of [2 x double] to determine if the values in the first -/// operand are not greater than those in the second operand. Each -/// comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true. +/// operand are not greater than those in the second operand. +/// +/// Each comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true. /// /// \headerfile /// @@ -648,6 +657,7 @@ _mm_cmpngt_pd(__m128d __a, __m128d __b) /// \brief Compares each of the corresponding double-precision values of the /// 128-bit vectors of [2 x double] to determine if the values in the first /// operand are not greater than or equal to those in the second operand. +/// /// Each comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true. /// /// \headerfile @@ -666,8 +676,9 @@ _mm_cmpnge_pd(__m128d __a, __m128d __b) } /// \brief Compares the lower double-precision floating-point values in each of -/// the two 128-bit floating-point vectors of [2 x double] for equality. The -/// comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true. +/// the two 128-bit floating-point vectors of [2 x double] for equality. +/// +/// The comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true. /// /// \headerfile /// @@ -690,8 +701,9 @@ _mm_cmpeq_sd(__m128d __a, __m128d __b) /// \brief Compares the lower double-precision floating-point values in each of /// the two 128-bit floating-point vectors of [2 x double] to determine if /// the value in the first parameter is less than the corresponding value in -/// the second parameter. The comparison yields 0h for false, -/// FFFFFFFFFFFFFFFFh for true. +/// the second parameter. +/// +/// The comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true. /// /// \headerfile /// @@ -714,8 +726,9 @@ _mm_cmplt_sd(__m128d __a, __m128d __b) /// \brief Compares the lower double-precision floating-point values in each of /// the two 128-bit floating-point vectors of [2 x double] to determine if /// the value in the first parameter is less than or equal to the -/// corresponding value in the second parameter. The comparison yields 0h for -/// false, FFFFFFFFFFFFFFFFh for true. +/// corresponding value in the second parameter. +/// +/// The comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true. /// /// \headerfile /// @@ -738,8 +751,9 @@ _mm_cmple_sd(__m128d __a, __m128d __b) /// \brief Compares the lower double-precision floating-point values in each of /// the two 128-bit floating-point vectors of [2 x double] to determine if /// the value in the first parameter is greater than the corresponding value -/// in the second parameter. The comparison yields 0h for false, -/// FFFFFFFFFFFFFFFFh for true. +/// in the second parameter. +/// +/// The comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true. /// /// \headerfile /// @@ -763,8 +777,9 @@ _mm_cmpgt_sd(__m128d __a, __m128d __b) /// \brief Compares the lower double-precision floating-point values in each of /// the two 128-bit floating-point vectors of [2 x double] to determine if /// the value in the first parameter is greater than or equal to the -/// corresponding value in the second parameter. The comparison yields 0h for -/// false, FFFFFFFFFFFFFFFFh for true. +/// corresponding value in the second parameter. +/// +/// The comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true. /// /// \headerfile /// @@ -788,9 +803,11 @@ _mm_cmpge_sd(__m128d __a, __m128d __b) /// \brief Compares the lower double-precision floating-point values in each of /// the two 128-bit floating-point vectors of [2 x double] to determine if /// the value in the first parameter is "ordered" with respect to the -/// corresponding value in the second parameter. The comparison yields 0h for -/// false, FFFFFFFFFFFFFFFFh for true. A pair of double-precision values are -/// "ordered" with respect to each other if neither value is a NaN. +/// corresponding value in the second parameter. +/// +/// The comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true. A pair of +/// double-precision values are "ordered" with respect to each other if +/// neither value is a NaN. /// /// \headerfile /// @@ -813,9 +830,11 @@ _mm_cmpord_sd(__m128d __a, __m128d __b) /// \brief Compares the lower double-precision floating-point values in each of /// the two 128-bit floating-point vectors of [2 x double] to determine if /// the value in the first parameter is "unordered" with respect to the -/// corresponding value in the second parameter. The comparison yields 0h -/// for false, FFFFFFFFFFFFFFFFh for true. A pair of double-precision values -/// are "unordered" with respect to each other if one or both values are NaN. +/// corresponding value in the second parameter. +/// +/// The comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true. A pair of +/// double-precision values are "unordered" with respect to each other if one +/// or both values are NaN. /// /// \headerfile /// @@ -839,8 +858,9 @@ _mm_cmpunord_sd(__m128d __a, __m128d __b) /// \brief Compares the lower double-precision floating-point values in each of /// the two 128-bit floating-point vectors of [2 x double] to determine if /// the value in the first parameter is unequal to the corresponding value in -/// the second parameter. The comparison yields 0h for false, -/// FFFFFFFFFFFFFFFFh for true. +/// the second parameter. +/// +/// The comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true. /// /// \headerfile /// @@ -863,8 +883,9 @@ _mm_cmpneq_sd(__m128d __a, __m128d __b) /// \brief Compares the lower double-precision floating-point values in each of /// the two 128-bit floating-point vectors of [2 x double] to determine if /// the value in the first parameter is not less than the corresponding -/// value in the second parameter. The comparison yields 0h for false, -/// FFFFFFFFFFFFFFFFh for true. +/// value in the second parameter. +/// +/// The comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true. /// /// \headerfile /// @@ -887,8 +908,9 @@ _mm_cmpnlt_sd(__m128d __a, __m128d __b) /// \brief Compares the lower double-precision floating-point values in each of /// the two 128-bit floating-point vectors of [2 x double] to determine if /// the value in the first parameter is not less than or equal to the -/// corresponding value in the second parameter. The comparison yields 0h -/// for false, FFFFFFFFFFFFFFFFh for true. +/// corresponding value in the second parameter. +/// +/// The comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true. /// /// \headerfile /// @@ -911,8 +933,9 @@ _mm_cmpnle_sd(__m128d __a, __m128d __b) /// \brief Compares the lower double-precision floating-point values in each of /// the two 128-bit floating-point vectors of [2 x double] to determine if /// the value in the first parameter is not greater than the corresponding -/// value in the second parameter. The comparison yields 0h for false, -/// FFFFFFFFFFFFFFFFh for true. +/// value in the second parameter. +/// +/// The comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true. /// /// \headerfile /// @@ -936,8 +959,9 @@ _mm_cmpngt_sd(__m128d __a, __m128d __b) /// \brief Compares the lower double-precision floating-point values in each of /// the two 128-bit floating-point vectors of [2 x double] to determine if /// the value in the first parameter is not greater than or equal to the -/// corresponding value in the second parameter. The comparison yields 0h -/// for false, FFFFFFFFFFFFFFFFh for true. +/// corresponding value in the second parameter. +/// +/// The comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true. /// /// \headerfile /// @@ -982,7 +1006,9 @@ _mm_comieq_sd(__m128d __a, __m128d __b) /// \brief Compares the lower double-precision floating-point values in each of /// the two 128-bit floating-point vectors of [2 x double] to determine if /// the value in the first parameter is less than the corresponding value in -/// the second parameter. The comparison yields 0 for false, 1 for true. +/// the second parameter. +/// +/// The comparison yields 0 for false, 1 for true. /// /// \headerfile /// @@ -1004,8 +1030,9 @@ _mm_comilt_sd(__m128d __a, __m128d __b) /// \brief Compares the lower double-precision floating-point values in each of /// the two 128-bit floating-point vectors of [2 x double] to determine if /// the value in the first parameter is less than or equal to the -/// corresponding value in the second parameter. The comparison yields 0 for -/// false, 1 for true. +/// corresponding value in the second parameter. +/// +/// The comparison yields 0 for false, 1 for true. /// /// \headerfile /// @@ -1027,7 +1054,9 @@ _mm_comile_sd(__m128d __a, __m128d __b) /// \brief Compares the lower double-precision floating-point values in each of /// the two 128-bit floating-point vectors of [2 x double] to determine if /// the value in the first parameter is greater than the corresponding value -/// in the second parameter. The comparison yields 0 for false, 1 for true. +/// in the second parameter. +/// +/// The comparison yields 0 for false, 1 for true. /// /// \headerfile /// @@ -1049,8 +1078,9 @@ _mm_comigt_sd(__m128d __a, __m128d __b) /// \brief Compares the lower double-precision floating-point values in each of /// the two 128-bit floating-point vectors of [2 x double] to determine if /// the value in the first parameter is greater than or equal to the -/// corresponding value in the second parameter. The comparison yields 0 for -/// false, 1 for true. +/// corresponding value in the second parameter. +/// +/// The comparison yields 0 for false, 1 for true. /// /// \headerfile /// @@ -1072,7 +1102,9 @@ _mm_comige_sd(__m128d __a, __m128d __b) /// \brief Compares the lower double-precision floating-point values in each of /// the two 128-bit floating-point vectors of [2 x double] to determine if /// the value in the first parameter is unequal to the corresponding value in -/// the second parameter. The comparison yields 0 for false, 1 for true. +/// the second parameter. +/// +/// The comparison yields 0 for false, 1 for true. /// /// \headerfile /// @@ -1093,8 +1125,9 @@ _mm_comineq_sd(__m128d __a, __m128d __b) /// \brief Compares the lower double-precision floating-point values in each of /// the two 128-bit floating-point vectors of [2 x double] for equality. The -/// comparison yields 0 for false, 1 for true. If either of the two lower -/// double-precision values is NaN, 1 is returned. +/// comparison yields 0 for false, 1 for true. +/// +/// If either of the two lower double-precision values is NaN, 1 is returned. /// /// \headerfile /// @@ -1117,8 +1150,10 @@ _mm_ucomieq_sd(__m128d __a, __m128d __b) /// \brief Compares the lower double-precision floating-point values in each of /// the two 128-bit floating-point vectors of [2 x double] to determine if /// the value in the first parameter is less than the corresponding value in -/// the second parameter. The comparison yields 0 for false, 1 for true. If -/// either of the two lower double-precision values is NaN, 1 is returned. +/// the second parameter. +/// +/// The comparison yields 0 for false, 1 for true. If either of the two lower +/// double-precision values is NaN, 1 is returned. /// /// \headerfile /// @@ -1141,9 +1176,10 @@ _mm_ucomilt_sd(__m128d __a, __m128d __b) /// \brief Compares the lower double-precision floating-point values in each of /// the two 128-bit floating-point vectors of [2 x double] to determine if /// the value in the first parameter is less than or equal to the -/// corresponding value in the second parameter. The comparison yields 0 for -/// false, 1 for true. If either of the two lower double-precision values is -/// NaN, 1 is returned. +/// corresponding value in the second parameter. +/// +/// The comparison yields 0 for false, 1 for true. If either of the two lower +/// double-precision values is NaN, 1 is returned. /// /// \headerfile /// @@ -1166,8 +1202,10 @@ _mm_ucomile_sd(__m128d __a, __m128d __b) /// \brief Compares the lower double-precision floating-point values in each of /// the two 128-bit floating-point vectors of [2 x double] to determine if /// the value in the first parameter is greater than the corresponding value -/// in the second parameter. The comparison yields 0 for false, 1 for true. -/// If either of the two lower double-precision values is NaN, 0 is returned. +/// in the second parameter. +/// +/// The comparison yields 0 for false, 1 for true. If either of the two lower +/// double-precision values is NaN, 0 is returned. /// /// \headerfile /// @@ -1190,9 +1228,10 @@ _mm_ucomigt_sd(__m128d __a, __m128d __b) /// \brief Compares the lower double-precision floating-point values in each of /// the two 128-bit floating-point vectors of [2 x double] to determine if /// the value in the first parameter is greater than or equal to the -/// corresponding value in the second parameter. The comparison yields 0 for -/// false, 1 for true. If either of the two lower double-precision values -/// is NaN, 0 is returned. +/// corresponding value in the second parameter. +/// +/// The comparison yields 0 for false, 1 for true. If either of the two +/// lower double-precision values is NaN, 0 is returned. /// /// \headerfile /// @@ -1215,8 +1254,10 @@ _mm_ucomige_sd(__m128d __a, __m128d __b) /// \brief Compares the lower double-precision floating-point values in each of /// the two 128-bit floating-point vectors of [2 x double] to determine if /// the value in the first parameter is unequal to the corresponding value in -/// the second parameter. The comparison yields 0 for false, 1 for true. If -/// either of the two lower double-precision values is NaN, 0 is returned. +/// the second parameter. +/// +/// The comparison yields 0 for false, 1 for true. If either of the two lower +/// double-precision values is NaN, 0 is returned. /// /// \headerfile /// @@ -1278,8 +1319,9 @@ _mm_cvtps_pd(__m128 __a) /// \brief Converts the lower two integer elements of a 128-bit vector of /// [4 x i32] into two double-precision floating-point values, returned in a -/// 128-bit vector of [2 x double]. The upper two elements of the input -/// vector are unused. +/// 128-bit vector of [2 x double]. +/// +/// The upper two elements of the input vector are unused. /// /// \headerfile /// @@ -1287,7 +1329,9 @@ _mm_cvtps_pd(__m128 __a) /// /// \param __a /// A 128-bit integer vector of [4 x i32]. The lower two integer elements are -/// converted to double-precision values. The upper two elements are unused. +/// converted to double-precision values. +/// +/// The upper two elements are unused. /// \returns A 128-bit vector of [2 x double] containing the converted values. static __inline__ __m128d __DEFAULT_FN_ATTRS _mm_cvtepi32_pd(__m128i __a) @@ -1409,10 +1453,11 @@ _mm_cvtss_sd(__m128d __a, __m128 __b) /// \brief Converts the two double-precision floating-point elements of a /// 128-bit vector of [2 x double] into two signed 32-bit integer values, -/// returned in the lower 64 bits of a 128-bit vector of [4 x i32]. If the -/// result of either conversion is inexact, the result is truncated (rounded -/// towards zero) regardless of the current MXCSR setting. The upper 64 bits -/// of the result vector are set to zero. +/// returned in the lower 64 bits of a 128-bit vector of [4 x i32]. +/// +/// If the result of either conversion is inexact, the result is truncated +/// (rounded towards zero) regardless of the current MXCSR setting. The upper +/// 64 bits of the result vector are set to zero. /// /// \headerfile /// @@ -1466,9 +1511,10 @@ _mm_cvtpd_pi32(__m128d __a) /// \brief Converts the two double-precision floating-point elements of a /// 128-bit vector of [2 x double] into two signed 32-bit integer values, -/// returned in a 64-bit vector of [2 x i32]. If the result of either -/// conversion is inexact, the result is truncated (rounded towards zero) -/// regardless of the current MXCSR setting. +/// returned in a 64-bit vector of [2 x i32]. +/// +/// If the result of either conversion is inexact, the result is truncated +/// (rounded towards zero) regardless of the current MXCSR setting. /// /// \headerfile /// @@ -1980,8 +2026,9 @@ _mm_storel_pd(double *__dp, __m128d __a) /// \brief Adds the corresponding elements of two 128-bit vectors of [16 x i8], /// saving the lower 8 bits of each sum in the corresponding element of a -/// 128-bit result vector of [16 x i8]. The integer elements of both -/// parameters can be either signed or unsigned. +/// 128-bit result vector of [16 x i8]. +/// +/// The integer elements of both parameters can be either signed or unsigned. /// /// \headerfile /// @@ -2001,8 +2048,9 @@ _mm_add_epi8(__m128i __a, __m128i __b) /// \brief Adds the corresponding elements of two 128-bit vectors of [8 x i16], /// saving the lower 16 bits of each sum in the corresponding element of a -/// 128-bit result vector of [8 x i16]. The integer elements of both -/// parameters can be either signed or unsigned. +/// 128-bit result vector of [8 x i16]. +/// +/// The integer elements of both parameters can be either signed or unsigned. /// /// \headerfile /// @@ -2022,8 +2070,9 @@ _mm_add_epi16(__m128i __a, __m128i __b) /// \brief Adds the corresponding elements of two 128-bit vectors of [4 x i32], /// saving the lower 32 bits of each sum in the corresponding element of a -/// 128-bit result vector of [4 x i32]. The integer elements of both -/// parameters can be either signed or unsigned. +/// 128-bit result vector of [4 x i32]. +/// +/// The integer elements of both parameters can be either signed or unsigned. /// /// \headerfile /// @@ -2061,8 +2110,9 @@ _mm_add_si64(__m64 __a, __m64 __b) /// \brief Adds the corresponding elements of two 128-bit vectors of [2 x i64], /// saving the lower 64 bits of each sum in the corresponding element of a -/// 128-bit result vector of [2 x i64]. The integer elements of both -/// parameters can be either signed or unsigned. +/// 128-bit result vector of [2 x i64]. +/// +/// The integer elements of both parameters can be either signed or unsigned. /// /// \headerfile /// @@ -2208,10 +2258,12 @@ _mm_avg_epu16(__m128i __a, __m128i __b) /// \brief Multiplies the corresponding elements of two 128-bit signed [8 x i16] /// vectors, producing eight intermediate 32-bit signed integer products, and /// adds the consecutive pairs of 32-bit products to form a 128-bit signed -/// [4 x i32] vector. For example, bits [15:0] of both parameters are -/// multiplied producing a 32-bit product, bits [31:16] of both parameters -/// are multiplied producing a 32-bit product, and the sum of those two -/// products becomes bits [31:0] of the result. +/// [4 x i32] vector. +/// +/// For example, bits [15:0] of both parameters are multiplied producing a +/// 32-bit product, bits [31:16] of both parameters are multiplied producing +/// a 32-bit product, and the sum of those two products becomes bits [31:0] +/// of the result. /// /// \headerfile /// @@ -3146,8 +3198,9 @@ _mm_cmpgt_epi8(__m128i __a, __m128i __b) /// \brief Compares each of the corresponding signed 16-bit values of the /// 128-bit integer vectors to determine if the values in the first operand -/// are greater than those in the second operand. Each comparison yields 0h -/// for false, FFFFh for true. +/// are greater than those in the second operand. +/// +/// Each comparison yields 0h for false, FFFFh for true. /// /// \headerfile /// @@ -3166,8 +3219,9 @@ _mm_cmpgt_epi16(__m128i __a, __m128i __b) /// \brief Compares each of the corresponding signed 32-bit values of the /// 128-bit integer vectors to determine if the values in the first operand -/// are greater than those in the second operand. Each comparison yields 0h -/// for false, FFFFFFFFh for true. +/// are greater than those in the second operand. +/// +/// Each comparison yields 0h for false, FFFFFFFFh for true. /// /// \headerfile /// @@ -3186,8 +3240,9 @@ _mm_cmpgt_epi32(__m128i __a, __m128i __b) /// \brief Compares each of the corresponding signed 8-bit values of the 128-bit /// integer vectors to determine if the values in the first operand are less -/// than those in the second operand. Each comparison yields 0h for false, -/// FFh for true. +/// than those in the second operand. +/// +/// Each comparison yields 0h for false, FFh for true. /// /// \headerfile /// @@ -3206,8 +3261,9 @@ _mm_cmplt_epi8(__m128i __a, __m128i __b) /// \brief Compares each of the corresponding signed 16-bit values of the /// 128-bit integer vectors to determine if the values in the first operand -/// are less than those in the second operand. Each comparison yields 0h for -/// false, FFFFh for true. +/// are less than those in the second operand. +/// +/// Each comparison yields 0h for false, FFFFh for true. /// /// \headerfile /// @@ -3226,8 +3282,9 @@ _mm_cmplt_epi16(__m128i __a, __m128i __b) /// \brief Compares each of the corresponding signed 32-bit values of the /// 128-bit integer vectors to determine if the values in the first operand -/// are less than those in the second operand. Each comparison yields 0h for -/// false, FFFFFFFFh for true. +/// are less than those in the second operand. +/// +/// Each comparison yields 0h for false, FFFFFFFFh for true. /// /// \headerfile /// @@ -3925,10 +3982,11 @@ _mm_storeu_si128(__m128i *__p, __m128i __b) /// \brief Moves bytes selected by the mask from the first operand to the /// specified unaligned memory location. When a mask bit is 1, the -/// corresponding byte is written, otherwise it is not written. To minimize -/// caching, the date is flagged as non-temporal (unlikely to be used again -/// soon). Exception and trap behavior for elements not selected for storage -/// to memory are implementation dependent. +/// corresponding byte is written, otherwise it is not written. +/// +/// To minimize caching, the date is flagged as non-temporal (unlikely to be +/// used again soon). Exception and trap behavior for elements not selected +/// for storage to memory are implementation dependent. /// /// \headerfile /// @@ -3972,8 +4030,10 @@ _mm_storel_epi64(__m128i *__p, __m128i __a) } /// \brief Stores a 128-bit floating point vector of [2 x double] to a 128-bit -/// aligned memory location. To minimize caching, the data is flagged as -/// non-temporal (unlikely to be used again soon). +/// aligned memory location. +/// +/// To minimize caching, the data is flagged as non-temporal (unlikely to be +/// used again soon). /// /// \headerfile /// @@ -3990,6 +4050,7 @@ _mm_stream_pd(double *__p, __m128d __a) } /// \brief Stores a 128-bit integer vector to a 128-bit aligned memory location. +/// /// To minimize caching, the data is flagged as non-temporal (unlikely to be /// used again soon). /// @@ -4007,8 +4068,9 @@ _mm_stream_si128(__m128i *__p, __m128i __a) __builtin_nontemporal_store((__v2di)__a, (__v2di*)__p); } -/// \brief Stores a 32-bit integer value in the specified memory location. To -/// minimize caching, the data is flagged as non-temporal (unlikely to be +/// \brief Stores a 32-bit integer value in the specified memory location. +/// +/// To minimize caching, the data is flagged as non-temporal (unlikely to be /// used again soon). /// /// \headerfile @@ -4026,8 +4088,9 @@ _mm_stream_si32(int *__p, int __a) } #ifdef __x86_64__ -/// \brief Stores a 64-bit integer value in the specified memory location. To -/// minimize caching, the data is flagged as non-temporal (unlikely to be +/// \brief Stores a 64-bit integer value in the specified memory location. +/// +/// To minimize caching, the data is flagged as non-temporal (unlikely to be /// used again soon). /// /// \headerfile diff --git a/lib/Headers/intrin.h b/lib/Headers/intrin.h index 38d9407abed9..881d05c0d164 100644 --- a/lib/Headers/intrin.h +++ b/lib/Headers/intrin.h @@ -85,9 +85,6 @@ void __inwordstring(unsigned short, unsigned short *, unsigned long); void __lidt(void *); unsigned __int64 __ll_lshift(unsigned __int64, int); __int64 __ll_rshift(__int64, int); -void __llwpcb(void *); -unsigned char __lwpins32(unsigned int, unsigned int, unsigned int); -void __lwpval32(unsigned int, unsigned int, unsigned int); unsigned int __lzcnt(unsigned int); unsigned short __lzcnt16(unsigned short); static __inline__ @@ -126,7 +123,6 @@ unsigned __int64 __readmsr(unsigned long); unsigned __int64 __readpmc(unsigned long); unsigned long __segmentlimit(unsigned long); void __sidt(void *); -void *__slwpcb(void); static __inline__ void __stosb(unsigned char *, unsigned char, size_t); static __inline__ @@ -227,8 +223,6 @@ void __incgsbyte(unsigned long); void __incgsdword(unsigned long); void __incgsqword(unsigned long); void __incgsword(unsigned long); -unsigned char __lwpins64(unsigned __int64, unsigned int, unsigned int); -void __lwpval64(unsigned __int64, unsigned int, unsigned int); unsigned __int64 __lzcnt64(unsigned __int64); static __inline__ void __movsq(unsigned long long *, unsigned long long const *, size_t); diff --git a/lib/Headers/mmintrin.h b/lib/Headers/mmintrin.h index 2b3618398cbf..5a7968bec842 100644 --- a/lib/Headers/mmintrin.h +++ b/lib/Headers/mmintrin.h @@ -608,10 +608,11 @@ _mm_subs_pi16(__m64 __m1, __m64 __m2) /// \brief Subtracts each 8-bit unsigned integer element of the second 64-bit /// integer vector of [8 x i8] from the corresponding 8-bit unsigned integer -/// element of the first 64-bit integer vector of [8 x i8]. If an element of -/// the first vector is less than the corresponding element of the second -/// vector, the result is saturated to 0. The results are packed into a -/// 64-bit integer vector of [8 x i8]. +/// element of the first 64-bit integer vector of [8 x i8]. +/// +/// If an element of the first vector is less than the corresponding element +/// of the second vector, the result is saturated to 0. The results are +/// packed into a 64-bit integer vector of [8 x i8]. /// /// \headerfile /// @@ -631,10 +632,11 @@ _mm_subs_pu8(__m64 __m1, __m64 __m2) /// \brief Subtracts each 16-bit unsigned integer element of the second 64-bit /// integer vector of [4 x i16] from the corresponding 16-bit unsigned -/// integer element of the first 64-bit integer vector of [4 x i16]. If an -/// element of the first vector is less than the corresponding element of the -/// second vector, the result is saturated to 0. The results are packed into -/// a 64-bit integer vector of [4 x i16]. +/// integer element of the first 64-bit integer vector of [4 x i16]. +/// +/// If an element of the first vector is less than the corresponding element +/// of the second vector, the result is saturated to 0. The results are +/// packed into a 64-bit integer vector of [4 x i16]. /// /// \headerfile /// @@ -657,9 +659,11 @@ _mm_subs_pu16(__m64 __m1, __m64 __m2) /// element of the second 64-bit integer vector of [4 x i16] and get four /// 32-bit products. Adds adjacent pairs of products to get two 32-bit sums. /// The lower 32 bits of these two sums are packed into a 64-bit integer -/// vector of [2 x i32]. For example, bits [15:0] of both parameters are -/// multiplied, bits [31:16] of both parameters are multiplied, and the sum -/// of both results is written to bits [31:0] of the result. +/// vector of [2 x i32]. +/// +/// For example, bits [15:0] of both parameters are multiplied, bits [31:16] +/// of both parameters are multiplied, and the sum of both results is written +/// to bits [31:0] of the result. /// /// \headerfile /// @@ -851,10 +855,11 @@ _mm_slli_si64(__m64 __m, int __count) /// \brief Right-shifts each 16-bit integer element of the first parameter, /// which is a 64-bit integer vector of [4 x i16], by the number of bits -/// specified by the second parameter, which is a 64-bit integer. High-order -/// bits are filled with the sign bit of the initial value of each 16-bit -/// element. The 16-bit results are packed into a 64-bit integer vector of -/// [4 x i16]. +/// specified by the second parameter, which is a 64-bit integer. +/// +/// High-order bits are filled with the sign bit of the initial value of each +/// 16-bit element. The 16-bit results are packed into a 64-bit integer +/// vector of [4 x i16]. /// /// \headerfile /// @@ -874,6 +879,7 @@ _mm_sra_pi16(__m64 __m, __m64 __count) /// \brief Right-shifts each 16-bit integer element of a 64-bit integer vector /// of [4 x i16] by the number of bits specified by a 32-bit integer. +/// /// High-order bits are filled with the sign bit of the initial value of each /// 16-bit element. The 16-bit results are packed into a 64-bit integer /// vector of [4 x i16]. @@ -896,10 +902,11 @@ _mm_srai_pi16(__m64 __m, int __count) /// \brief Right-shifts each 32-bit integer element of the first parameter, /// which is a 64-bit integer vector of [2 x i32], by the number of bits -/// specified by the second parameter, which is a 64-bit integer. High-order -/// bits are filled with the sign bit of the initial value of each 32-bit -/// element. The 32-bit results are packed into a 64-bit integer vector of -/// [2 x i32]. +/// specified by the second parameter, which is a 64-bit integer. +/// +/// High-order bits are filled with the sign bit of the initial value of each +/// 32-bit element. The 32-bit results are packed into a 64-bit integer +/// vector of [2 x i32]. /// /// \headerfile /// @@ -919,6 +926,7 @@ _mm_sra_pi32(__m64 __m, __m64 __count) /// \brief Right-shifts each 32-bit integer element of a 64-bit integer vector /// of [2 x i32] by the number of bits specified by a 32-bit integer. +/// /// High-order bits are filled with the sign bit of the initial value of each /// 32-bit element. The 32-bit results are packed into a 64-bit integer /// vector of [2 x i32]. @@ -941,9 +949,10 @@ _mm_srai_pi32(__m64 __m, int __count) /// \brief Right-shifts each 16-bit integer element of the first parameter, /// which is a 64-bit integer vector of [4 x i16], by the number of bits -/// specified by the second parameter, which is a 64-bit integer. High-order -/// bits are cleared. The 16-bit results are packed into a 64-bit integer -/// vector of [4 x i16]. +/// specified by the second parameter, which is a 64-bit integer. +/// +/// High-order bits are cleared. The 16-bit results are packed into a 64-bit +/// integer vector of [4 x i16]. /// /// \headerfile /// @@ -963,6 +972,7 @@ _mm_srl_pi16(__m64 __m, __m64 __count) /// \brief Right-shifts each 16-bit integer element of a 64-bit integer vector /// of [4 x i16] by the number of bits specified by a 32-bit integer. +/// /// High-order bits are cleared. The 16-bit results are packed into a 64-bit /// integer vector of [4 x i16]. /// @@ -984,9 +994,10 @@ _mm_srli_pi16(__m64 __m, int __count) /// \brief Right-shifts each 32-bit integer element of the first parameter, /// which is a 64-bit integer vector of [2 x i32], by the number of bits -/// specified by the second parameter, which is a 64-bit integer. High-order -/// bits are cleared. The 32-bit results are packed into a 64-bit integer -/// vector of [2 x i32]. +/// specified by the second parameter, which is a 64-bit integer. +/// +/// High-order bits are cleared. The 32-bit results are packed into a 64-bit +/// integer vector of [2 x i32]. /// /// \headerfile /// @@ -1006,6 +1017,7 @@ _mm_srl_pi32(__m64 __m, __m64 __count) /// \brief Right-shifts each 32-bit integer element of a 64-bit integer vector /// of [2 x i32] by the number of bits specified by a 32-bit integer. +/// /// High-order bits are cleared. The 32-bit results are packed into a 64-bit /// integer vector of [2 x i32]. /// @@ -1026,8 +1038,9 @@ _mm_srli_pi32(__m64 __m, int __count) } /// \brief Right-shifts the first 64-bit integer parameter by the number of bits -/// specified by the second 64-bit integer parameter. High-order bits are -/// cleared. +/// specified by the second 64-bit integer parameter. +/// +/// High-order bits are cleared. /// /// \headerfile /// @@ -1046,7 +1059,9 @@ _mm_srl_si64(__m64 __m, __m64 __count) /// \brief Right-shifts the first parameter, which is a 64-bit integer, by the /// number of bits specified by the second parameter, which is a 32-bit -/// integer. High-order bits are cleared. +/// integer. +/// +/// High-order bits are cleared. /// /// \headerfile /// @@ -1140,8 +1155,9 @@ _mm_xor_si64(__m64 __m1, __m64 __m2) /// \brief Compares the 8-bit integer elements of two 64-bit integer vectors of /// [8 x i8] to determine if the element of the first vector is equal to the -/// corresponding element of the second vector. The comparison yields 0 for -/// false, 0xFF for true. +/// corresponding element of the second vector. +/// +/// The comparison yields 0 for false, 0xFF for true. /// /// \headerfile /// @@ -1161,8 +1177,9 @@ _mm_cmpeq_pi8(__m64 __m1, __m64 __m2) /// \brief Compares the 16-bit integer elements of two 64-bit integer vectors of /// [4 x i16] to determine if the element of the first vector is equal to the -/// corresponding element of the second vector. The comparison yields 0 for -/// false, 0xFFFF for true. +/// corresponding element of the second vector. +/// +/// The comparison yields 0 for false, 0xFFFF for true. /// /// \headerfile /// @@ -1182,8 +1199,9 @@ _mm_cmpeq_pi16(__m64 __m1, __m64 __m2) /// \brief Compares the 32-bit integer elements of two 64-bit integer vectors of /// [2 x i32] to determine if the element of the first vector is equal to the -/// corresponding element of the second vector. The comparison yields 0 for -/// false, 0xFFFFFFFF for true. +/// corresponding element of the second vector. +/// +/// The comparison yields 0 for false, 0xFFFFFFFF for true. /// /// \headerfile /// @@ -1203,8 +1221,9 @@ _mm_cmpeq_pi32(__m64 __m1, __m64 __m2) /// \brief Compares the 8-bit integer elements of two 64-bit integer vectors of /// [8 x i8] to determine if the element of the first vector is greater than -/// the corresponding element of the second vector. The comparison yields 0 -/// for false, 0xFF for true. +/// the corresponding element of the second vector. +/// +/// The comparison yields 0 for false, 0xFF for true. /// /// \headerfile /// @@ -1224,8 +1243,9 @@ _mm_cmpgt_pi8(__m64 __m1, __m64 __m2) /// \brief Compares the 16-bit integer elements of two 64-bit integer vectors of /// [4 x i16] to determine if the element of the first vector is greater than -/// the corresponding element of the second vector. The comparison yields 0 -/// for false, 0xFFFF for true. +/// the corresponding element of the second vector. +/// +/// The comparison yields 0 for false, 0xFFFF for true. /// /// \headerfile /// @@ -1245,8 +1265,9 @@ _mm_cmpgt_pi16(__m64 __m1, __m64 __m2) /// \brief Compares the 32-bit integer elements of two 64-bit integer vectors of /// [2 x i32] to determine if the element of the first vector is greater than -/// the corresponding element of the second vector. The comparison yields 0 -/// for false, 0xFFFFFFFF for true. +/// the corresponding element of the second vector. +/// +/// The comparison yields 0 for false, 0xFFFFFFFF for true. /// /// \headerfile /// diff --git a/lib/Headers/opencl-c.h b/lib/Headers/opencl-c.h index 6452d5c987f0..58c8daf3a536 100644 --- a/lib/Headers/opencl-c.h +++ b/lib/Headers/opencl-c.h @@ -14962,6 +14962,7 @@ float __purefn __ovld read_imagef(read_only image2d_array_msaa_depth_t image, in #endif //cl_khr_gl_msaa_sharing // OpenCL Extension v2.0 s9.18 - Mipmaps +#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0 #ifdef cl_khr_mipmap_image float4 __purefn __ovld read_imagef(read_only image1d_t image, sampler_t sampler, float coord, float lod); @@ -15037,6 +15038,7 @@ int4 __purefn __ovld read_imagei(read_only image3d_t image, sampler_t sampler, f uint4 __purefn __ovld read_imageui(read_only image3d_t image, sampler_t sampler, float4 coord, float lod); #endif //cl_khr_mipmap_image +#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0 /** * Sampler-less Image Access @@ -15135,6 +15137,7 @@ float __purefn __ovld read_imagef(read_write image2d_msaa_depth_t image, int2 co float __purefn __ovld read_imagef(read_write image2d_array_msaa_depth_t image, int4 coord, int sample); #endif //cl_khr_gl_msaa_sharing +#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0 #ifdef cl_khr_mipmap_image float4 __purefn __ovld read_imagef(read_write image1d_t image, sampler_t sampler, float coord, float lod); int4 __purefn __ovld read_imagei(read_write image1d_t image, sampler_t sampler, float coord, float lod); @@ -15208,6 +15211,7 @@ float4 __purefn __ovld read_imagef(read_write image3d_t image, sampler_t sampler int4 __purefn __ovld read_imagei(read_write image3d_t image, sampler_t sampler, float4 coord, float lod); uint4 __purefn __ovld read_imageui(read_write image3d_t image, sampler_t sampler, float4 coord, float lod); #endif //cl_khr_mipmap_image +#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0 // Image read functions returning half4 type #ifdef cl_khr_fp16 @@ -15319,6 +15323,7 @@ void __ovld write_imagef(write_only image2d_array_depth_t image, int4 coord, flo #endif //cl_khr_depth_images // OpenCL Extension v2.0 s9.18 - Mipmaps +#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0 #ifdef cl_khr_mipmap_image void __ovld write_imagef(write_only image1d_t image, int coord, int lod, float4 color); void __ovld write_imagei(write_only image1d_t image, int coord, int lod, int4 color); @@ -15345,6 +15350,7 @@ void __ovld write_imagei(write_only image3d_t image, int4 coord, int lod, int4 c void __ovld write_imageui(write_only image3d_t image, int4 coord, int lod, uint4 color); #endif #endif //cl_khr_mipmap_image +#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0 // Image write functions for half4 type #ifdef cl_khr_fp16 @@ -15391,6 +15397,7 @@ void __ovld write_imagef(read_write image2d_depth_t image, int2 coord, float col void __ovld write_imagef(read_write image2d_array_depth_t image, int4 coord, float color); #endif //cl_khr_depth_images +#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0 #ifdef cl_khr_mipmap_image void __ovld write_imagef(read_write image1d_t image, int coord, int lod, float4 color); void __ovld write_imagei(read_write image1d_t image, int coord, int lod, int4 color); @@ -15417,6 +15424,7 @@ void __ovld write_imagei(read_write image3d_t image, int4 coord, int lod, int4 c void __ovld write_imageui(read_write image3d_t image, int4 coord, int lod, uint4 color); #endif #endif //cl_khr_mipmap_image +#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0 // Image write functions for half4 type #ifdef cl_khr_fp16 @@ -15559,6 +15567,7 @@ int __ovld __cnfn get_image_depth(read_write image3d_t image); #endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0 // OpenCL Extension v2.0 s9.18 - Mipmaps +#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0 #ifdef cl_khr_mipmap_image /** * Return the image miplevels. @@ -15574,11 +15583,9 @@ int __ovld get_image_num_mip_levels(write_only image2d_t image); int __ovld get_image_num_mip_levels(write_only image3d_t image); #endif -#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0 int __ovld get_image_num_mip_levels(read_write image1d_t image); int __ovld get_image_num_mip_levels(read_write image2d_t image); int __ovld get_image_num_mip_levels(read_write image3d_t image); -#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0 int __ovld get_image_num_mip_levels(read_only image1d_array_t image); int __ovld get_image_num_mip_levels(read_only image2d_array_t image); @@ -15590,14 +15597,13 @@ int __ovld get_image_num_mip_levels(write_only image2d_array_t image); int __ovld get_image_num_mip_levels(write_only image2d_array_depth_t image); int __ovld get_image_num_mip_levels(write_only image2d_depth_t image); -#if __OPENCL_C_VERSION__ >= CL_VERSION_2_0 int __ovld get_image_num_mip_levels(read_write image1d_array_t image); int __ovld get_image_num_mip_levels(read_write image2d_array_t image); int __ovld get_image_num_mip_levels(read_write image2d_array_depth_t image); int __ovld get_image_num_mip_levels(read_write image2d_depth_t image); -#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0 #endif //cl_khr_mipmap_image +#endif //__OPENCL_C_VERSION__ >= CL_VERSION_2_0 /** * Return the channel data type. Valid values are: diff --git a/lib/Headers/pmmintrin.h b/lib/Headers/pmmintrin.h index a479d9ed2911..559ece2e3974 100644 --- a/lib/Headers/pmmintrin.h +++ b/lib/Headers/pmmintrin.h @@ -31,9 +31,11 @@ __attribute__((__always_inline__, __nodebug__, __target__("sse3"))) /// \brief Loads data from an unaligned memory location to elements in a 128-bit -/// vector. If the address of the data is not 16-byte aligned, the -/// instruction may read two adjacent aligned blocks of memory to retrieve -/// the requested data. +/// vector. +/// +/// If the address of the data is not 16-byte aligned, the instruction may +/// read two adjacent aligned blocks of memory to retrieve the requested +/// data. /// /// \headerfile /// diff --git a/lib/Headers/prfchwintrin.h b/lib/Headers/prfchwintrin.h index a3789126ef07..b52f31da2706 100644 --- a/lib/Headers/prfchwintrin.h +++ b/lib/Headers/prfchwintrin.h @@ -50,8 +50,10 @@ _m_prefetch(void *__P) /// the L1 data cache and sets the cache-coherency to modified. This /// provides a hint to the processor that the cache line will be modified. /// It is intended for use when the cache line will be written to shortly -/// after the prefetch is performed. Note that the effect of this intrinsic -/// is dependent on the processor implementation. +/// after the prefetch is performed. +/// +/// Note that the effect of this intrinsic is dependent on the processor +/// implementation. /// /// \headerfile /// diff --git a/lib/Headers/smmintrin.h b/lib/Headers/smmintrin.h index 1c94aca69381..c2fa5a452bce 100644 --- a/lib/Headers/smmintrin.h +++ b/lib/Headers/smmintrin.h @@ -586,7 +586,9 @@ _mm_mul_epi32 (__m128i __V1, __m128i __V2) /* SSE4 Floating Point Dot Product Instructions. */ /// \brief Computes the dot product of the two 128-bit vectors of [4 x float] /// and returns it in the elements of the 128-bit result vector of -/// [4 x float]. The immediate integer operand controls which input elements +/// [4 x float]. +/// +/// The immediate integer operand controls which input elements /// will contribute to the dot product, and where the final results are /// returned. /// @@ -620,7 +622,9 @@ _mm_mul_epi32 (__m128i __V1, __m128i __V2) /// \brief Computes the dot product of the two 128-bit vectors of [2 x double] /// and returns it in the elements of the 128-bit result vector of -/// [2 x double]. The immediate integer operand controls which input +/// [2 x double]. +/// +/// The immediate integer operand controls which input /// elements will contribute to the dot product, and where the final results /// are returned. /// @@ -875,7 +879,7 @@ _mm_max_epu32 (__m128i __V1, __m128i __V2) /// int _mm_extract_ps(__m128 X, const int N); /// \endcode /// -/// This intrinsic corresponds to the VEXTRACTPS / EXTRACTPS +/// This intrinsic corresponds to the VEXTRACTPS / EXTRACTPS /// instruction. /// /// \param X diff --git a/lib/Headers/tmmintrin.h b/lib/Headers/tmmintrin.h index 80664043a06f..042bfc7e3b0d 100644 --- a/lib/Headers/tmmintrin.h +++ b/lib/Headers/tmmintrin.h @@ -469,10 +469,11 @@ _mm_hsubs_pi16(__m64 __a, __m64 __b) /// values contained in the first source operand and packed 8-bit signed /// integer values contained in the second source operand, adds pairs of /// contiguous products with signed saturation, and writes the 16-bit sums to -/// the corresponding bits in the destination. For example, bits [7:0] of -/// both operands are multiplied, bits [15:8] of both operands are -/// multiplied, and the sum of both results is written to bits [15:0] of the -/// destination. +/// the corresponding bits in the destination. +/// +/// For example, bits [7:0] of both operands are multiplied, bits [15:8] of +/// both operands are multiplied, and the sum of both results is written to +/// bits [15:0] of the destination. /// /// \headerfile /// @@ -502,10 +503,11 @@ _mm_maddubs_epi16(__m128i __a, __m128i __b) /// values contained in the first source operand and packed 8-bit signed /// integer values contained in the second source operand, adds pairs of /// contiguous products with signed saturation, and writes the 16-bit sums to -/// the corresponding bits in the destination. For example, bits [7:0] of -/// both operands are multiplied, bits [15:8] of both operands are -/// multiplied, and the sum of both results is written to bits [15:0] of the -/// destination. +/// the corresponding bits in the destination. +/// +/// For example, bits [7:0] of both operands are multiplied, bits [15:8] of +/// both operands are multiplied, and the sum of both results is written to +/// bits [15:0] of the destination. /// /// \headerfile /// @@ -619,13 +621,14 @@ _mm_shuffle_pi8(__m64 __a, __m64 __b) } /// \brief For each 8-bit integer in the first source operand, perform one of -/// the following actions as specified by the second source operand: If the -/// byte in the second source is negative, calculate the two's complement of -/// the corresponding byte in the first source, and write that value to the -/// destination. If the byte in the second source is positive, copy the -/// corresponding byte from the first source to the destination. If the byte -/// in the second source is zero, clear the corresponding byte in the -/// destination. +/// the following actions as specified by the second source operand. +/// +/// If the byte in the second source is negative, calculate the two's +/// complement of the corresponding byte in the first source, and write that +/// value to the destination. If the byte in the second source is positive, +/// copy the corresponding byte from the first source to the destination. If +/// the byte in the second source is zero, clear the corresponding byte in +/// the destination. /// /// \headerfile /// @@ -644,13 +647,14 @@ _mm_sign_epi8(__m128i __a, __m128i __b) } /// \brief For each 16-bit integer in the first source operand, perform one of -/// the following actions as specified by the second source operand: If the -/// word in the second source is negative, calculate the two's complement of -/// the corresponding word in the first source, and write that value to the -/// destination. If the word in the second source is positive, copy the -/// corresponding word from the first source to the destination. If the word -/// in the second source is zero, clear the corresponding word in the -/// destination. +/// the following actions as specified by the second source operand. +/// +/// If the word in the second source is negative, calculate the two's +/// complement of the corresponding word in the first source, and write that +/// value to the destination. If the word in the second source is positive, +/// copy the corresponding word from the first source to the destination. If +/// the word in the second source is zero, clear the corresponding word in +/// the destination. /// /// \headerfile /// @@ -669,8 +673,9 @@ _mm_sign_epi16(__m128i __a, __m128i __b) } /// \brief For each 32-bit integer in the first source operand, perform one of -/// the following actions as specified by the second source operand: If the -/// doubleword in the second source is negative, calculate the two's +/// the following actions as specified by the second source operand. +/// +/// If the doubleword in the second source is negative, calculate the two's /// complement of the corresponding word in the first source, and write that /// value to the destination. If the doubleword in the second source is /// positive, copy the corresponding word from the first source to the @@ -694,13 +699,14 @@ _mm_sign_epi32(__m128i __a, __m128i __b) } /// \brief For each 8-bit integer in the first source operand, perform one of -/// the following actions as specified by the second source operand: If the -/// byte in the second source is negative, calculate the two's complement of -/// the corresponding byte in the first source, and write that value to the -/// destination. If the byte in the second source is positive, copy the -/// corresponding byte from the first source to the destination. If the byte -/// in the second source is zero, clear the corresponding byte in the -/// destination. +/// the following actions as specified by the second source operand. +/// +/// If the byte in the second source is negative, calculate the two's +/// complement of the corresponding byte in the first source, and write that +/// value to the destination. If the byte in the second source is positive, +/// copy the corresponding byte from the first source to the destination. If +/// the byte in the second source is zero, clear the corresponding byte in +/// the destination. /// /// \headerfile /// @@ -719,13 +725,14 @@ _mm_sign_pi8(__m64 __a, __m64 __b) } /// \brief For each 16-bit integer in the first source operand, perform one of -/// the following actions as specified by the second source operand: If the -/// word in the second source is negative, calculate the two's complement of -/// the corresponding word in the first source, and write that value to the -/// destination. If the word in the second source is positive, copy the -/// corresponding word from the first source to the destination. If the word -/// in the second source is zero, clear the corresponding word in the -/// destination. +/// the following actions as specified by the second source operand. +/// +/// If the word in the second source is negative, calculate the two's +/// complement of the corresponding word in the first source, and write that +/// value to the destination. If the word in the second source is positive, +/// copy the corresponding word from the first source to the destination. If +/// the word in the second source is zero, clear the corresponding word in +/// the destination. /// /// \headerfile /// @@ -744,8 +751,9 @@ _mm_sign_pi16(__m64 __a, __m64 __b) } /// \brief For each 32-bit integer in the first source operand, perform one of -/// the following actions as specified by the second source operand: If the -/// doubleword in the second source is negative, calculate the two's +/// the following actions as specified by the second source operand. +/// +/// If the doubleword in the second source is negative, calculate the two's /// complement of the corresponding doubleword in the first source, and /// write that value to the destination. If the doubleword in the second /// source is positive, copy the corresponding doubleword from the first diff --git a/lib/Headers/x86intrin.h b/lib/Headers/x86intrin.h index ef1d02948c8b..31ee7b82dd53 100644 --- a/lib/Headers/x86intrin.h +++ b/lib/Headers/x86intrin.h @@ -88,6 +88,4 @@ #include #endif -/* FIXME: LWP */ - #endif /* __X86INTRIN_H */ diff --git a/lib/Headers/xmmintrin.h b/lib/Headers/xmmintrin.h index 5c312c08efb6..9773acb840a5 100644 --- a/lib/Headers/xmmintrin.h +++ b/lib/Headers/xmmintrin.h @@ -2331,8 +2331,10 @@ _mm_mulhi_pu16(__m64 __a, __m64 __b) /// \brief Conditionally copies the values from each 8-bit element in the first /// 64-bit integer vector operand to the specified memory location, as /// specified by the most significant bit in the corresponding element in the -/// second 64-bit integer vector operand. To minimize caching, the data is -/// flagged as non-temporal (unlikely to be used again soon). +/// second 64-bit integer vector operand. +/// +/// To minimize caching, the data is flagged as non-temporal +/// (unlikely to be used again soon). /// /// \headerfile /// @@ -2815,11 +2817,12 @@ _mm_cvtpi32x2_ps(__m64 __a, __m64 __b) /// \brief Converts each single-precision floating-point element of a 128-bit /// floating-point vector of [4 x float] into a 16-bit signed integer, and -/// packs the results into a 64-bit integer vector of [4 x i16]. If the -/// floating-point element is NaN or infinity, or if the floating-point -/// element is greater than 0x7FFFFFFF or less than -0x8000, it is converted -/// to 0x8000. Otherwise if the floating-point element is greater than -/// 0x7FFF, it is converted to 0x7FFF. +/// packs the results into a 64-bit integer vector of [4 x i16]. +/// +/// If the floating-point element is NaN or infinity, or if the +/// floating-point element is greater than 0x7FFFFFFF or less than -0x8000, +/// it is converted to 0x8000. Otherwise if the floating-point element is +/// greater than 0x7FFF, it is converted to 0x7FFF. /// /// \headerfile /// @@ -2845,11 +2848,12 @@ _mm_cvtps_pi16(__m128 __a) /// \brief Converts each single-precision floating-point element of a 128-bit /// floating-point vector of [4 x float] into an 8-bit signed integer, and /// packs the results into the lower 32 bits of a 64-bit integer vector of -/// [8 x i8]. The upper 32 bits of the vector are set to 0. If the -/// floating-point element is NaN or infinity, or if the floating-point -/// element is greater than 0x7FFFFFFF or less than -0x80, it is converted -/// to 0x80. Otherwise if the floating-point element is greater than 0x7F, -/// it is converted to 0x7F. +/// [8 x i8]. The upper 32 bits of the vector are set to 0. +/// +/// If the floating-point element is NaN or infinity, or if the +/// floating-point element is greater than 0x7FFFFFFF or less than -0x80, it +/// is converted to 0x80. Otherwise if the floating-point element is greater +/// than 0x7F, it is converted to 0x7F. /// /// \headerfile /// diff --git a/lib/Index/IndexBody.cpp b/lib/Index/IndexBody.cpp index 7f09290de40f..08d233fb83e0 100644 --- a/lib/Index/IndexBody.cpp +++ b/lib/Index/IndexBody.cpp @@ -150,6 +150,53 @@ class BodyIndexer : public RecursiveASTVisitor { Parent, ParentDC, Roles, Relations, E); } + bool indexDependentReference( + const Expr *E, const Type *T, const DeclarationNameInfo &NameInfo, + llvm::function_ref Filter) { + if (!T) + return true; + const TemplateSpecializationType *TST = + T->getAs(); + if (!TST) + return true; + TemplateName TN = TST->getTemplateName(); + const ClassTemplateDecl *TD = + dyn_cast_or_null(TN.getAsTemplateDecl()); + if (!TD) + return true; + CXXRecordDecl *RD = TD->getTemplatedDecl(); + if (!RD->hasDefinition()) + return true; + RD = RD->getDefinition(); + std::vector Symbols = + RD->lookupDependentName(NameInfo.getName(), Filter); + // FIXME: Improve overload handling. + if (Symbols.size() != 1) + return true; + SourceLocation Loc = NameInfo.getLoc(); + if (Loc.isInvalid()) + Loc = E->getLocStart(); + SmallVector Relations; + SymbolRoleSet Roles = getRolesForRef(E, Relations); + return IndexCtx.handleReference(Symbols[0], Loc, Parent, ParentDC, Roles, + Relations, E); + } + + bool VisitCXXDependentScopeMemberExpr(CXXDependentScopeMemberExpr *E) { + const DeclarationNameInfo &Info = E->getMemberNameInfo(); + return indexDependentReference( + E, E->getBaseType().getTypePtrOrNull(), Info, + [](const NamedDecl *D) { return D->isCXXInstanceMember(); }); + } + + bool VisitDependentScopeDeclRefExpr(DependentScopeDeclRefExpr *E) { + const DeclarationNameInfo &Info = E->getNameInfo(); + const NestedNameSpecifier *NNS = E->getQualifier(); + return indexDependentReference( + E, NNS->getAsType(), Info, + [](const NamedDecl *D) { return !D->isCXXInstanceMember(); }); + } + bool VisitDesignatedInitExpr(DesignatedInitExpr *E) { for (DesignatedInitExpr::Designator &D : llvm::reverse(E->designators())) { if (D.isFieldDesignator() && D.getField()) diff --git a/lib/Index/IndexDecl.cpp b/lib/Index/IndexDecl.cpp index 7de70a10b692..203862c46e98 100644 --- a/lib/Index/IndexDecl.cpp +++ b/lib/Index/IndexDecl.cpp @@ -52,6 +52,22 @@ class IndexingDeclVisitor : public ConstDeclVisitor { return MD && !MD->isImplicit() && MD->isThisDeclarationADefinition(); } + void handleTemplateArgumentLoc(const TemplateArgumentLoc &TALoc, + const NamedDecl *Parent, + const DeclContext *DC) { + const TemplateArgumentLocInfo &LocInfo = TALoc.getLocInfo(); + switch (TALoc.getArgument().getKind()) { + case TemplateArgument::Expression: + IndexCtx.indexBody(LocInfo.getAsExpr(), Parent, DC); + break; + case TemplateArgument::Type: + IndexCtx.indexTypeSourceInfo(LocInfo.getAsTypeSourceInfo(), Parent, DC); + break; + default: + break; + } + } + void handleDeclarator(const DeclaratorDecl *D, const NamedDecl *Parent = nullptr, bool isIBType = false) { @@ -233,6 +249,12 @@ class IndexingDeclVisitor : public ConstDeclVisitor { Dtor->getParent(), Dtor->getDeclContext()); } } + // Template specialization arguments. + if (const ASTTemplateArgumentListInfo *TemplateArgInfo = + D->getTemplateSpecializationArgsAsWritten()) { + for (const auto &Arg : TemplateArgInfo->arguments()) + handleTemplateArgumentLoc(Arg, D, D->getLexicalDeclContext()); + } if (D->isThisDeclarationADefinition()) { const Stmt *Body = D->getBody(); @@ -522,6 +544,14 @@ class IndexingDeclVisitor : public ConstDeclVisitor { return true; } + bool VisitNamespaceAliasDecl(const NamespaceAliasDecl *D) { + TRY_DECL(D, IndexCtx.handleDecl(D)); + IndexCtx.indexNestedNameSpecifierLoc(D->getQualifierLoc(), D); + IndexCtx.handleReference(D->getAliasedNamespace(), D->getTargetNameLoc(), D, + D->getLexicalDeclContext()); + return true; + } + bool VisitUsingDecl(const UsingDecl *D) { const DeclContext *DC = D->getDeclContext()->getRedeclContext(); const NamedDecl *Parent = dyn_cast(DC); diff --git a/lib/Index/IndexTypeSourceInfo.cpp b/lib/Index/IndexTypeSourceInfo.cpp index 44d1241fb930..ae27ebe6ea4c 100644 --- a/lib/Index/IndexTypeSourceInfo.cpp +++ b/lib/Index/IndexTypeSourceInfo.cpp @@ -141,6 +141,34 @@ class TypeIndexer : public RecursiveASTVisitor { return true; } + bool VisitDependentNameTypeLoc(DependentNameTypeLoc TL) { + const DependentNameType *DNT = TL.getTypePtr(); + const NestedNameSpecifier *NNS = DNT->getQualifier(); + const Type *T = NNS->getAsType(); + if (!T) + return true; + const TemplateSpecializationType *TST = + T->getAs(); + if (!TST) + return true; + TemplateName TN = TST->getTemplateName(); + const ClassTemplateDecl *TD = + dyn_cast_or_null(TN.getAsTemplateDecl()); + if (!TD) + return true; + CXXRecordDecl *RD = TD->getTemplatedDecl(); + if (!RD->hasDefinition()) + return true; + RD = RD->getDefinition(); + DeclarationName Name(DNT->getIdentifier()); + std::vector Symbols = RD->lookupDependentName( + Name, [](const NamedDecl *ND) { return isa(ND); }); + if (Symbols.size() != 1) + return true; + return IndexCtx.handleReference(Symbols[0], TL.getNameLoc(), Parent, + ParentDC, SymbolRoleSet(), Relations); + } + bool TraverseStmt(Stmt *S) { IndexCtx.indexBody(S, Parent, ParentDC); return true; @@ -184,7 +212,7 @@ void IndexingContext::indexNestedNameSpecifierLoc(NestedNameSpecifierLoc NNS, if (!DC) DC = Parent->getLexicalDeclContext(); - SourceLocation Loc = NNS.getSourceRange().getBegin(); + SourceLocation Loc = NNS.getLocalBeginLoc(); switch (NNS.getNestedNameSpecifier()->getKind()) { case NestedNameSpecifier::Identifier: diff --git a/lib/Index/IndexingContext.cpp b/lib/Index/IndexingContext.cpp index 709a23657b07..5cebb198460f 100644 --- a/lib/Index/IndexingContext.cpp +++ b/lib/Index/IndexingContext.cpp @@ -124,6 +124,10 @@ bool IndexingContext::isTemplateImplicitInstantiation(const Decl *D) { TKind = FD->getTemplateSpecializationKind(); } else if (auto *VD = dyn_cast(D)) { TKind = VD->getTemplateSpecializationKind(); + } else if (isa(D)) { + if (const auto *Parent = + dyn_cast(D->getDeclContext())) + TKind = Parent->getSpecializationKind(); } switch (TKind) { case TSK_Undeclared: @@ -159,6 +163,17 @@ static const Decl *adjustTemplateImplicitInstantiation(const Decl *D) { return FD->getTemplateInstantiationPattern(); } else if (auto *VD = dyn_cast(D)) { return VD->getTemplateInstantiationPattern(); + } else if (const auto *FD = dyn_cast(D)) { + if (const auto *Parent = + dyn_cast(D->getDeclContext())) { + const CXXRecordDecl *Pattern = Parent->getTemplateInstantiationPattern(); + for (const NamedDecl *ND : Pattern->lookup(FD->getDeclName())) { + if (ND->isImplicit()) + continue; + if (isa(ND)) + return ND; + } + } } return nullptr; } diff --git a/lib/Lex/MacroInfo.cpp b/lib/Lex/MacroInfo.cpp index bec434085e3a..1e5deeb1919b 100644 --- a/lib/Lex/MacroInfo.cpp +++ b/lib/Lex/MacroInfo.cpp @@ -29,7 +29,6 @@ MacroInfo::MacroInfo(SourceLocation DefLoc) IsUsed(false), IsAllowRedefinitionsWithoutWarning(false), IsWarnIfUnused(false), - FromASTFile(false), UsedForHeaderGuard(false) { } @@ -137,7 +136,6 @@ LLVM_DUMP_METHOD void MacroInfo::dump() const { if (IsAllowRedefinitionsWithoutWarning) Out << " allow_redefinitions_without_warning"; if (IsWarnIfUnused) Out << " warn_if_unused"; - if (FromASTFile) Out << " imported"; if (UsedForHeaderGuard) Out << " header_guard"; Out << "\n #define "; diff --git a/lib/Lex/ModuleMap.cpp b/lib/Lex/ModuleMap.cpp index 70d37d3d7082..6f44dc757e85 100644 --- a/lib/Lex/ModuleMap.cpp +++ b/lib/Lex/ModuleMap.cpp @@ -1485,7 +1485,19 @@ void ModuleMapParser::parseModuleDecl() { // Determine whether this (sub)module has already been defined. if (Module *Existing = Map.lookupModuleQualified(ModuleName, ActiveModule)) { - if (Existing->DefinitionLoc.isInvalid() && !ActiveModule) { + // We might see a (re)definition of a module that we already have a + // definition for in two cases: + // - If we loaded one definition from an AST file and we've just found a + // corresponding definition in a module map file, or + bool LoadedFromASTFile = Existing->DefinitionLoc.isInvalid(); + // - If we're building a (preprocessed) module and we've just loaded the + // module map file from which it was created. + bool ParsedAsMainInput = + Map.LangOpts.getCompilingModule() == LangOptions::CMK_ModuleMap && + Map.LangOpts.CurrentModule == ModuleName && + SourceMgr.getDecomposedLoc(ModuleNameLoc).first != + SourceMgr.getDecomposedLoc(Existing->DefinitionLoc).first; + if (!ActiveModule && (LoadedFromASTFile || ParsedAsMainInput)) { // Skip the module definition. skipUntil(MMToken::RBrace); if (Tok.is(MMToken::RBrace)) @@ -1901,8 +1913,10 @@ void ModuleMapParser::parseHeaderDecl(MMToken::TokenKind LeadingToken, // 'framework module FrameworkName.Private', since a 'Private.Framework' // does not usually exist. However, since both are currently widely used // for private modules, make sure we find the right path in both cases. - RelativePathName.resize(ActiveModule->IsFramework ? 0 - : RelativePathLength); + if (ActiveModule->IsFramework && ActiveModule->Name == "Private") + RelativePathName.clear(); + else + RelativePathName.resize(RelativePathLength); FullPathName.resize(FullPathLength); llvm::sys::path::append(RelativePathName, "PrivateHeaders", Header.FileName); diff --git a/lib/Lex/PPDirectives.cpp b/lib/Lex/PPDirectives.cpp index 06fee8e5b0a8..faf8809e4eb4 100644 --- a/lib/Lex/PPDirectives.cpp +++ b/lib/Lex/PPDirectives.cpp @@ -54,35 +54,12 @@ using namespace clang; // Utility Methods for Preprocessor Directive Handling. //===----------------------------------------------------------------------===// -MacroInfo *Preprocessor::AllocateMacroInfo() { - MacroInfoChain *MIChain = BP.Allocate(); - MIChain->Next = MIChainHead; +MacroInfo *Preprocessor::AllocateMacroInfo(SourceLocation L) { + auto *MIChain = new (BP) MacroInfoChain{L, MIChainHead}; MIChainHead = MIChain; return &MIChain->MI; } -MacroInfo *Preprocessor::AllocateMacroInfo(SourceLocation L) { - MacroInfo *MI = AllocateMacroInfo(); - new (MI) MacroInfo(L); - return MI; -} - -MacroInfo *Preprocessor::AllocateDeserializedMacroInfo(SourceLocation L, - unsigned SubModuleID) { - static_assert(alignof(MacroInfo) >= sizeof(SubModuleID), - "alignment for MacroInfo is less than the ID"); - DeserializedMacroInfoChain *MIChain = - BP.Allocate(); - MIChain->Next = DeserialMIChainHead; - DeserialMIChainHead = MIChain; - - MacroInfo *MI = &MIChain->MI; - new (MI) MacroInfo(L); - MI->FromASTFile = true; - MI->setOwningModuleID(SubModuleID); - return MI; -} - DefMacroDirective *Preprocessor::AllocateDefMacroDirective(MacroInfo *MI, SourceLocation Loc) { return new (BP) DefMacroDirective(MI, Loc); diff --git a/lib/Lex/Preprocessor.cpp b/lib/Lex/Preprocessor.cpp index e409ab036535..dce8c1efda23 100644 --- a/lib/Lex/Preprocessor.cpp +++ b/lib/Lex/Preprocessor.cpp @@ -88,7 +88,7 @@ Preprocessor::Preprocessor(std::shared_ptr PPOpts, CurDirLookup(nullptr), CurLexerKind(CLK_Lexer), CurLexerSubmodule(nullptr), Callbacks(nullptr), CurSubmoduleState(&NullSubmoduleState), MacroArgCache(nullptr), - Record(nullptr), MIChainHead(nullptr), DeserialMIChainHead(nullptr) { + Record(nullptr), MIChainHead(nullptr) { OwnsHeaderSearch = OwnsHeaders; CounterValue = 0; // __COUNTER__ starts at 0. @@ -169,11 +169,6 @@ Preprocessor::~Preprocessor() { std::fill(TokenLexerCache, TokenLexerCache + NumCachedTokenLexers, nullptr); CurTokenLexer.reset(); - while (DeserializedMacroInfoChain *I = DeserialMIChainHead) { - DeserialMIChainHead = I->Next; - I->~DeserializedMacroInfoChain(); - } - // Free any cached MacroArgs. for (MacroArgs *ArgList = MacroArgCache; ArgList;) ArgList = ArgList->deallocate(); diff --git a/lib/Parse/ParseDecl.cpp b/lib/Parse/ParseDecl.cpp index 1465d21ac5ee..4ccee74eaa90 100644 --- a/lib/Parse/ParseDecl.cpp +++ b/lib/Parse/ParseDecl.cpp @@ -2577,9 +2577,9 @@ bool Parser::ParseImplicitInt(DeclSpec &DS, CXXScopeSpec *SS, // and attempt to recover. ParsedType T; IdentifierInfo *II = Tok.getIdentifierInfo(); + bool IsTemplateName = getLangOpts().CPlusPlus && NextToken().is(tok::less); Actions.DiagnoseUnknownTypeName(II, Loc, getCurScope(), SS, T, - getLangOpts().CPlusPlus && - NextToken().is(tok::less)); + IsTemplateName); if (T) { // The action has suggested that the type T could be used. Set that as // the type in the declaration specifiers, consume the would-be type @@ -2604,6 +2604,13 @@ bool Parser::ParseImplicitInt(DeclSpec &DS, CXXScopeSpec *SS, DS.SetRangeEnd(Tok.getLocation()); ConsumeToken(); + // Eat any following template arguments. + if (IsTemplateName) { + SourceLocation LAngle, RAngle; + TemplateArgList Args; + ParseTemplateIdAfterTemplateName(true, LAngle, Args, RAngle); + } + // TODO: Could inject an invalid typedef decl in an enclosing scope to // avoid rippling error messages on subsequent uses of the same type, // could be useful if #include was forgotten. diff --git a/lib/Parse/ParseDeclCXX.cpp b/lib/Parse/ParseDeclCXX.cpp index ad7b319676e9..e6cf65e36cfa 100644 --- a/lib/Parse/ParseDeclCXX.cpp +++ b/lib/Parse/ParseDeclCXX.cpp @@ -1137,8 +1137,8 @@ TypeResult Parser::ParseBaseTypeSpecifier(SourceLocation &BaseLoc, if (!Template) { TemplateArgList TemplateArgs; SourceLocation LAngleLoc, RAngleLoc; - ParseTemplateIdAfterTemplateName(nullptr, IdLoc, SS, true, LAngleLoc, - TemplateArgs, RAngleLoc); + ParseTemplateIdAfterTemplateName(true, LAngleLoc, TemplateArgs, + RAngleLoc); return true; } @@ -1530,8 +1530,8 @@ void Parser::ParseClassSpecifier(tok::TokenKind TagTokKind, // a class (or template thereof). TemplateArgList TemplateArgs; SourceLocation LAngleLoc, RAngleLoc; - if (ParseTemplateIdAfterTemplateName( - nullptr, NameLoc, SS, true, LAngleLoc, TemplateArgs, RAngleLoc)) { + if (ParseTemplateIdAfterTemplateName(true, LAngleLoc, TemplateArgs, + RAngleLoc)) { // We couldn't parse the template argument list at all, so don't // try to give any location information for the list. LAngleLoc = RAngleLoc = SourceLocation(); diff --git a/lib/Parse/ParseExpr.cpp b/lib/Parse/ParseExpr.cpp index 3e02e46ddc7d..727fd3500991 100644 --- a/lib/Parse/ParseExpr.cpp +++ b/lib/Parse/ParseExpr.cpp @@ -235,6 +235,30 @@ bool Parser::isNotExpressionStart() { return isKnownToBeDeclarationSpecifier(); } +/// We've parsed something that could plausibly be intended to be a template +/// name (\p LHS) followed by a '<' token, and the following code can't possibly +/// be an expression. Determine if this is likely to be a template-id and if so, +/// diagnose it. +bool Parser::diagnoseUnknownTemplateId(ExprResult LHS, SourceLocation Less) { + TentativeParsingAction TPA(*this); + // FIXME: We could look at the token sequence in a lot more detail here. + if (SkipUntil(tok::greater, tok::greatergreater, tok::greatergreatergreater, + StopAtSemi | StopBeforeMatch)) { + TPA.Commit(); + + SourceLocation Greater; + ParseGreaterThanInTemplateList(Greater, true, false); + Actions.diagnoseExprIntendedAsTemplateName(getCurScope(), LHS, + Less, Greater); + return true; + } + + // There's no matching '>' token, this probably isn't supposed to be + // interpreted as a template-id. Parse it as an (ill-formed) comparison. + TPA.Revert(); + return false; +} + static bool isFoldOperator(prec::Level Level) { return Level > prec::Unknown && Level != prec::Conditional; } @@ -276,6 +300,16 @@ Parser::ParseRHSOfBinaryExpression(ExprResult LHS, prec::Level MinPrec) { return LHS; } + // If a '<' token is followed by a type that can be a template argument and + // cannot be an expression, then this is ill-formed, but might be intended + // to be a template-id. + if (OpToken.is(tok::less) && Actions.mightBeIntendedToBeTemplateName(LHS) && + (isKnownToBeDeclarationSpecifier() || + Tok.isOneOf(tok::greater, tok::greatergreater, + tok::greatergreatergreater)) && + diagnoseUnknownTemplateId(LHS, OpToken.getLocation())) + return ExprError(); + // If the next token is an ellipsis, then this is a fold-expression. Leave // it alone so we can handle it in the paren expression. if (isFoldOperator(NextTokPrec) && Tok.is(tok::ellipsis)) { @@ -2989,6 +3023,11 @@ Optional Parser::ParseAvailabilitySpec() { return AvailabilitySpec(ConsumeToken()); } else { // Parse the platform name. + if (Tok.is(tok::code_completion)) { + Actions.CodeCompleteAvailabilityPlatformName(); + cutOffParsing(); + return None; + } if (Tok.isNot(tok::identifier)) { Diag(Tok, diag::err_avail_query_expected_platform_name); return None; @@ -3001,12 +3040,14 @@ Optional Parser::ParseAvailabilitySpec() { if (Version.empty()) return None; - StringRef Platform = PlatformIdentifier->Ident->getName(); + StringRef GivenPlatform = PlatformIdentifier->Ident->getName(); + StringRef Platform = + AvailabilityAttr::canonicalizePlatformName(GivenPlatform); if (AvailabilityAttr::getPrettyPlatformName(Platform).empty()) { Diag(PlatformIdentifier->Loc, diag::err_avail_query_unrecognized_platform_name) - << Platform; + << GivenPlatform; return None; } diff --git a/lib/Parse/ParseExprCXX.cpp b/lib/Parse/ParseExprCXX.cpp index 671a815911f3..56093f685617 100644 --- a/lib/Parse/ParseExprCXX.cpp +++ b/lib/Parse/ParseExprCXX.cpp @@ -2114,11 +2114,8 @@ bool Parser::ParseUnqualifiedIdTemplateId(CXXScopeSpec &SS, // Parse the enclosed template argument list. SourceLocation LAngleLoc, RAngleLoc; TemplateArgList TemplateArgs; - if (Tok.is(tok::less) && - ParseTemplateIdAfterTemplateName(Template, Id.StartLocation, - SS, true, LAngleLoc, - TemplateArgs, - RAngleLoc)) + if (Tok.is(tok::less) && ParseTemplateIdAfterTemplateName( + true, LAngleLoc, TemplateArgs, RAngleLoc)) return true; if (Id.getKind() == UnqualifiedId::IK_Identifier || diff --git a/lib/Parse/ParseTemplate.cpp b/lib/Parse/ParseTemplate.cpp index d2b18e7c0a81..6a81e14ed496 100644 --- a/lib/Parse/ParseTemplate.cpp +++ b/lib/Parse/ParseTemplate.cpp @@ -886,22 +886,12 @@ bool Parser::ParseGreaterThanInTemplateList(SourceLocation &RAngleLoc, /// list ('<' template-parameter-list [opt] '>') and placing the /// results into a form that can be transferred to semantic analysis. /// -/// \param Template the template declaration produced by isTemplateName -/// -/// \param TemplateNameLoc the source location of the template name -/// -/// \param SS if non-NULL, the nested-name-specifier preceding the -/// template name. -/// /// \param ConsumeLastToken if true, then we will consume the last /// token that forms the template-id. Otherwise, we will leave the /// last token in the stream (e.g., so that it can be replaced with an /// annotation token). bool -Parser::ParseTemplateIdAfterTemplateName(TemplateTy Template, - SourceLocation TemplateNameLoc, - const CXXScopeSpec &SS, - bool ConsumeLastToken, +Parser::ParseTemplateIdAfterTemplateName(bool ConsumeLastToken, SourceLocation &LAngleLoc, TemplateArgList &TemplateArgs, SourceLocation &RAngleLoc) { @@ -983,9 +973,7 @@ bool Parser::AnnotateTemplateIdToken(TemplateTy Template, TemplateNameKind TNK, // Parse the enclosed template argument list. SourceLocation LAngleLoc, RAngleLoc; TemplateArgList TemplateArgs; - bool Invalid = ParseTemplateIdAfterTemplateName(Template, - TemplateNameLoc, - SS, false, LAngleLoc, + bool Invalid = ParseTemplateIdAfterTemplateName(false, LAngleLoc, TemplateArgs, RAngleLoc); diff --git a/lib/Sema/CMakeLists.txt b/lib/Sema/CMakeLists.txt index 7a5973299f91..7d9ae621c93d 100644 --- a/lib/Sema/CMakeLists.txt +++ b/lib/Sema/CMakeLists.txt @@ -3,6 +3,7 @@ set(LLVM_LINK_COMPONENTS ) if (MSVC) + set_source_files_properties(SemaDeclAttr.cpp PROPERTIES COMPILE_FLAGS /bigobj) set_source_files_properties(SemaExpr.cpp PROPERTIES COMPILE_FLAGS /bigobj) endif() diff --git a/lib/Sema/Sema.cpp b/lib/Sema/Sema.cpp index 2f493fa5fbef..ca1d27e9505f 100644 --- a/lib/Sema/Sema.cpp +++ b/lib/Sema/Sema.cpp @@ -477,6 +477,13 @@ static bool ShouldRemoveFromUnused(Sema *SemaRef, const DeclaratorDecl *D) { return true; if (const FunctionDecl *FD = dyn_cast(D)) { + // If this is a function template and none of its specializations is used, + // we should warn. + if (FunctionTemplateDecl *Template = FD->getDescribedFunctionTemplate()) + for (const auto *Spec : Template->specializations()) + if (ShouldRemoveFromUnused(SemaRef, Spec)) + return true; + // UnusedFileScopedDecls stores the first declaration. // The declaration may have become definition so check again. const FunctionDecl *DeclToCheck; @@ -500,6 +507,13 @@ static bool ShouldRemoveFromUnused(Sema *SemaRef, const DeclaratorDecl *D) { VD->isUsableInConstantExpressions(SemaRef->Context)) return true; + if (VarTemplateDecl *Template = VD->getDescribedVarTemplate()) + // If this is a variable template and none of its specializations is used, + // we should warn. + for (const auto *Spec : Template->specializations()) + if (ShouldRemoveFromUnused(SemaRef, Spec)) + return true; + // UnusedFileScopedDecls stores the first declaration. // The declaration may have become definition so check again. const VarDecl *DeclToCheck = VD->getDefinition(); @@ -905,10 +919,14 @@ void Sema::ActOnEndOfTranslationUnit() { << /*function*/0 << DiagD->getDeclName(); } } else { - Diag(DiagD->getLocation(), - isa(DiagD) ? diag::warn_unused_member_function - : diag::warn_unused_function) - << DiagD->getDeclName(); + if (FD->getDescribedFunctionTemplate()) + Diag(DiagD->getLocation(), diag::warn_unused_template) + << /*function*/0 << DiagD->getDeclName(); + else + Diag(DiagD->getLocation(), + isa(DiagD) ? diag::warn_unused_member_function + : diag::warn_unused_function) + << DiagD->getDeclName(); } } else { const VarDecl *DiagD = cast(*I)->getDefinition(); @@ -924,7 +942,11 @@ void Sema::ActOnEndOfTranslationUnit() { Diag(DiagD->getLocation(), diag::warn_unused_const_variable) << DiagD->getDeclName(); } else { - Diag(DiagD->getLocation(), diag::warn_unused_variable) + if (DiagD->getDescribedVarTemplate()) + Diag(DiagD->getLocation(), diag::warn_unused_template) + << /*variable*/1 << DiagD->getDeclName(); + else + Diag(DiagD->getLocation(), diag::warn_unused_variable) << DiagD->getDeclName(); } } diff --git a/lib/Sema/SemaCast.cpp b/lib/Sema/SemaCast.cpp index 7e91709e67da..7d534263f468 100644 --- a/lib/Sema/SemaCast.cpp +++ b/lib/Sema/SemaCast.cpp @@ -1871,7 +1871,8 @@ static bool fixOverloadedReinterpretCastExpr(Sema &Self, QualType DestType, // No guarantees that ResolveAndFixSingleFunctionTemplateSpecialization // preserves Result. Result = E; - if (!Self.resolveAndFixAddressOfOnlyViableOverloadCandidate(Result)) + if (!Self.resolveAndFixAddressOfOnlyViableOverloadCandidate( + Result, /*DoFunctionPointerConversion=*/true)) return false; return Result.isUsable(); } diff --git a/lib/Sema/SemaCodeComplete.cpp b/lib/Sema/SemaCodeComplete.cpp index cfac3f1dc1de..8fb2f4139236 100644 --- a/lib/Sema/SemaCodeComplete.cpp +++ b/lib/Sema/SemaCodeComplete.cpp @@ -3869,6 +3869,41 @@ static void AddObjCProperties( } } +static void AddRecordMembersCompletionResults(Sema &SemaRef, + ResultBuilder &Results, Scope *S, + QualType BaseType, + RecordDecl *RD) { + // Indicate that we are performing a member access, and the cv-qualifiers + // for the base object type. + Results.setObjectTypeQualifiers(BaseType.getQualifiers()); + + // Access to a C/C++ class, struct, or union. + Results.allowNestedNameSpecifiers(); + CodeCompletionDeclConsumer Consumer(Results, SemaRef.CurContext); + SemaRef.LookupVisibleDecls(RD, Sema::LookupMemberName, Consumer, + SemaRef.CodeCompleter->includeGlobals(), + /*IncludeDependentBases=*/true); + + if (SemaRef.getLangOpts().CPlusPlus) { + if (!Results.empty()) { + // The "template" keyword can follow "->" or "." in the grammar. + // However, we only want to suggest the template keyword if something + // is dependent. + bool IsDependent = BaseType->isDependentType(); + if (!IsDependent) { + for (Scope *DepScope = S; DepScope; DepScope = DepScope->getParent()) + if (DeclContext *Ctx = DepScope->getEntity()) { + IsDependent = Ctx->isDependentContext(); + break; + } + } + + if (IsDependent) + Results.AddResult(CodeCompletionResult("template")); + } + } +} + void Sema::CodeCompleteMemberReferenceExpr(Scope *S, Expr *Base, SourceLocation OpLoc, bool IsArrow, bool IsBaseExprStatement) { @@ -3879,8 +3914,6 @@ void Sema::CodeCompleteMemberReferenceExpr(Scope *S, Expr *Base, if (ConvertedBase.isInvalid()) return; Base = ConvertedBase.get(); - - typedef CodeCompletionResult Result; QualType BaseType = Base->getType(); @@ -3915,34 +3948,18 @@ void Sema::CodeCompleteMemberReferenceExpr(Scope *S, Expr *Base, &ResultBuilder::IsMember); Results.EnterNewScope(); if (const RecordType *Record = BaseType->getAs()) { - // Indicate that we are performing a member access, and the cv-qualifiers - // for the base object type. - Results.setObjectTypeQualifiers(BaseType.getQualifiers()); - - // Access to a C/C++ class, struct, or union. - Results.allowNestedNameSpecifiers(); - CodeCompletionDeclConsumer Consumer(Results, CurContext); - LookupVisibleDecls(Record->getDecl(), LookupMemberName, Consumer, - CodeCompleter->includeGlobals()); - - if (getLangOpts().CPlusPlus) { - if (!Results.empty()) { - // The "template" keyword can follow "->" or "." in the grammar. - // However, we only want to suggest the template keyword if something - // is dependent. - bool IsDependent = BaseType->isDependentType(); - if (!IsDependent) { - for (Scope *DepScope = S; DepScope; DepScope = DepScope->getParent()) - if (DeclContext *Ctx = DepScope->getEntity()) { - IsDependent = Ctx->isDependentContext(); - break; - } - } - - if (IsDependent) - Results.AddResult(Result("template")); - } + AddRecordMembersCompletionResults(*this, Results, S, BaseType, + Record->getDecl()); + } else if (const auto *TST = BaseType->getAs()) { + TemplateName TN = TST->getTemplateName(); + if (const auto *TD = + dyn_cast_or_null(TN.getAsTemplateDecl())) { + CXXRecordDecl *RD = TD->getTemplatedDecl(); + AddRecordMembersCompletionResults(*this, Results, S, BaseType, RD); } + } else if (const auto *ICNT = BaseType->getAs()) { + if (auto *RD = ICNT->getDecl()) + AddRecordMembersCompletionResults(*this, Results, S, BaseType, RD); } else if (!IsArrow && BaseType->isObjCObjectPointerType()) { // Objective-C property reference. AddedPropertiesSet AddedProperties; @@ -7811,6 +7828,23 @@ void Sema::CodeCompleteNaturalLanguage() { nullptr, 0); } +void Sema::CodeCompleteAvailabilityPlatformName() { + ResultBuilder Results(*this, CodeCompleter->getAllocator(), + CodeCompleter->getCodeCompletionTUInfo(), + CodeCompletionContext::CCC_Other); + Results.EnterNewScope(); + static const char *Platforms[] = {"macOS", "iOS", "watchOS", "tvOS"}; + for (const char *Platform : llvm::makeArrayRef(Platforms)) { + Results.AddResult(CodeCompletionResult(Platform)); + Results.AddResult(CodeCompletionResult(Results.getAllocator().CopyString( + Twine(Platform) + "ApplicationExtension"))); + } + Results.ExitScope(); + HandleCodeCompleteResults(this, CodeCompleter, + CodeCompletionContext::CCC_Other, Results.data(), + Results.size()); +} + void Sema::GatherGlobalCodeCompletions(CodeCompletionAllocator &Allocator, CodeCompletionTUInfo &CCTUInfo, SmallVectorImpl &Results) { diff --git a/lib/Sema/SemaDecl.cpp b/lib/Sema/SemaDecl.cpp index 2612023f59db..2e069a9defaa 100644 --- a/lib/Sema/SemaDecl.cpp +++ b/lib/Sema/SemaDecl.cpp @@ -64,22 +64,45 @@ namespace { class TypeNameValidatorCCC : public CorrectionCandidateCallback { public: - TypeNameValidatorCCC(bool AllowInvalid, bool WantClass=false, - bool AllowTemplates=false) - : AllowInvalidDecl(AllowInvalid), WantClassName(WantClass), - AllowTemplates(AllowTemplates) { - WantExpressionKeywords = false; - WantCXXNamedCasts = false; - WantRemainingKeywords = false; + TypeNameValidatorCCC(bool AllowInvalid, bool WantClass = false, + bool AllowTemplates = false, + bool AllowNonTemplates = true) + : AllowInvalidDecl(AllowInvalid), WantClassName(WantClass), + AllowTemplates(AllowTemplates), AllowNonTemplates(AllowNonTemplates) { + WantExpressionKeywords = false; + WantCXXNamedCasts = false; + WantRemainingKeywords = false; } bool ValidateCandidate(const TypoCorrection &candidate) override { if (NamedDecl *ND = candidate.getCorrectionDecl()) { + if (!AllowInvalidDecl && ND->isInvalidDecl()) + return false; + + if (getAsTypeTemplateDecl(ND)) + return AllowTemplates; + bool IsType = isa(ND) || isa(ND); - bool AllowedTemplate = AllowTemplates && getAsTypeTemplateDecl(ND); - return (IsType || AllowedTemplate) && - (AllowInvalidDecl || !ND->isInvalidDecl()); + if (!IsType) + return false; + + if (AllowNonTemplates) + return true; + + // An injected-class-name of a class template (specialization) is valid + // as a template or as a non-template. + if (AllowTemplates) { + auto *RD = dyn_cast(ND); + if (!RD || !RD->isInjectedClassName()) + return false; + RD = cast(RD->getDeclContext()); + return RD->getDescribedClassTemplate() || + isa(RD); + } + + return false; } + return !WantClassName && candidate.isKeyword(); } @@ -87,6 +110,7 @@ class TypeNameValidatorCCC : public CorrectionCandidateCallback { bool AllowInvalidDecl; bool WantClassName; bool AllowTemplates; + bool AllowNonTemplates; }; } // end anonymous namespace @@ -627,7 +651,7 @@ void Sema::DiagnoseUnknownTypeName(IdentifierInfo *&II, Scope *S, CXXScopeSpec *SS, ParsedType &SuggestedType, - bool AllowClassTemplates) { + bool IsTemplateName) { // Don't report typename errors for editor placeholders. if (II->isEditorPlaceholder()) return; @@ -639,28 +663,41 @@ void Sema::DiagnoseUnknownTypeName(IdentifierInfo *&II, if (TypoCorrection Corrected = CorrectTypo(DeclarationNameInfo(II, IILoc), LookupOrdinaryName, S, SS, llvm::make_unique( - false, false, AllowClassTemplates), + false, false, IsTemplateName, !IsTemplateName), CTK_ErrorRecovery)) { + // FIXME: Support error recovery for the template-name case. + bool CanRecover = !IsTemplateName; if (Corrected.isKeyword()) { // We corrected to a keyword. - diagnoseTypo(Corrected, PDiag(diag::err_unknown_typename_suggest) << II); + diagnoseTypo(Corrected, + PDiag(IsTemplateName ? diag::err_no_template_suggest + : diag::err_unknown_typename_suggest) + << II); II = Corrected.getCorrectionAsIdentifierInfo(); } else { // We found a similarly-named type or interface; suggest that. if (!SS || !SS->isSet()) { diagnoseTypo(Corrected, - PDiag(diag::err_unknown_typename_suggest) << II); + PDiag(IsTemplateName ? diag::err_no_template_suggest + : diag::err_unknown_typename_suggest) + << II, CanRecover); } else if (DeclContext *DC = computeDeclContext(*SS, false)) { std::string CorrectedStr(Corrected.getAsString(getLangOpts())); bool DroppedSpecifier = Corrected.WillReplaceSpecifier() && II->getName().equals(CorrectedStr); diagnoseTypo(Corrected, - PDiag(diag::err_unknown_nested_typename_suggest) - << II << DC << DroppedSpecifier << SS->getRange()); + PDiag(IsTemplateName + ? diag::err_no_member_template_suggest + : diag::err_unknown_nested_typename_suggest) + << II << DC << DroppedSpecifier << SS->getRange(), + CanRecover); } else { llvm_unreachable("could not have corrected a typo here"); } + if (!CanRecover) + return; + CXXScopeSpec tmpSS; if (Corrected.getCorrectionSpecifier()) tmpSS.MakeTrivial(Context, Corrected.getCorrectionSpecifier(), @@ -675,7 +712,7 @@ void Sema::DiagnoseUnknownTypeName(IdentifierInfo *&II, return; } - if (getLangOpts().CPlusPlus) { + if (getLangOpts().CPlusPlus && !IsTemplateName) { // See if II is a class template that the user forgot to pass arguments to. UnqualifiedId Name; Name.setIdentifier(II, IILoc); @@ -700,10 +737,13 @@ void Sema::DiagnoseUnknownTypeName(IdentifierInfo *&II, // (struct, union, enum) from Parser::ParseImplicitInt here, instead? if (!SS || (!SS->isSet() && !SS->isInvalid())) - Diag(IILoc, diag::err_unknown_typename) << II; + Diag(IILoc, IsTemplateName ? diag::err_no_template + : diag::err_unknown_typename) + << II; else if (DeclContext *DC = computeDeclContext(*SS, false)) - Diag(IILoc, diag::err_typename_nested_not_found) - << II << DC << SS->getRange(); + Diag(IILoc, IsTemplateName ? diag::err_no_member_template + : diag::err_typename_nested_not_found) + << II << DC << SS->getRange(); else if (isDependentScopeSpecifier(*SS)) { unsigned DiagID = diag::err_typename_missing; if (getLangOpts().MSVCCompat && isMicrosoftMissingTypename(SS, S)) @@ -1488,6 +1528,11 @@ bool Sema::ShouldWarnIfUnusedFileScopedDecl(const DeclaratorDecl *D) const { if (const FunctionDecl *FD = dyn_cast(D)) { if (FD->getTemplateSpecializationKind() == TSK_ImplicitInstantiation) return false; + // A non-out-of-line declaration of a member specialization was implicitly + // instantiated; it's the out-of-line declaration that we're interested in. + if (FD->getTemplateSpecializationKind() == TSK_ExplicitSpecialization && + FD->getMemberSpecializationInfo() && !FD->isOutOfLine()) + return false; if (const CXXMethodDecl *MD = dyn_cast(FD)) { if (MD->isVirtual() || IsDisallowedCopyOrAssign(MD)) @@ -1514,6 +1559,10 @@ bool Sema::ShouldWarnIfUnusedFileScopedDecl(const DeclaratorDecl *D) const { if (VD->isStaticDataMember() && VD->getTemplateSpecializationKind() == TSK_ImplicitInstantiation) return false; + if (VD->isStaticDataMember() && + VD->getTemplateSpecializationKind() == TSK_ExplicitSpecialization && + VD->getMemberSpecializationInfo() && !VD->isOutOfLine()) + return false; if (VD->isInline() && !isMainFileLoc(*this, VD->getLocation())) return false; @@ -1972,7 +2021,7 @@ bool Sema::isIncompatibleTypedef(TypeDecl *Old, TypedefNameDecl *New) { Diag(New->getLocation(), diag::err_redefinition_variably_modified_typedef) << Kind << NewType; if (Old->getLocation().isValid()) - Diag(Old->getLocation(), diag::note_previous_definition); + notePreviousDefinition(Old->getLocation(), New->getLocation()); New->setInvalidDecl(); return true; } @@ -1985,7 +2034,7 @@ bool Sema::isIncompatibleTypedef(TypeDecl *Old, TypedefNameDecl *New) { Diag(New->getLocation(), diag::err_redefinition_different_typedef) << Kind << NewType << OldType; if (Old->getLocation().isValid()) - Diag(Old->getLocation(), diag::note_previous_definition); + notePreviousDefinition(Old->getLocation(), New->getLocation()); New->setInvalidDecl(); return true; } @@ -2052,7 +2101,7 @@ void Sema::MergeTypedefNameDecl(Scope *S, TypedefNameDecl *New, NamedDecl *OldD = OldDecls.getRepresentativeDecl(); if (OldD->getLocation().isValid()) - Diag(OldD->getLocation(), diag::note_previous_definition); + notePreviousDefinition(OldD->getLocation(), New->getLocation()); return New->setInvalidDecl(); } @@ -2078,7 +2127,7 @@ void Sema::MergeTypedefNameDecl(Scope *S, TypedefNameDecl *New, New->setTypeSourceInfo(OldTD->getTypeSourceInfo()); // Make the old tag definition visible. - makeMergedDefinitionVisible(Hidden, NewTag->getLocation()); + makeMergedDefinitionVisible(Hidden); // If this was an unscoped enumeration, yank all of its enumerators // out of the scope. @@ -2144,7 +2193,7 @@ void Sema::MergeTypedefNameDecl(Scope *S, TypedefNameDecl *New, Diag(New->getLocation(), diag::err_redefinition) << New->getDeclName(); - Diag(Old->getLocation(), diag::note_previous_definition); + notePreviousDefinition(Old->getLocation(), New->getLocation()); return New->setInvalidDecl(); } @@ -2165,7 +2214,7 @@ void Sema::MergeTypedefNameDecl(Scope *S, TypedefNameDecl *New, Diag(New->getLocation(), diag::ext_redefinition_of_typedef) << New->getDeclName(); - Diag(Old->getLocation(), diag::note_previous_definition); + notePreviousDefinition(Old->getLocation(), New->getLocation()); } /// DeclhasAttr - returns true if decl Declaration already has the target @@ -2452,7 +2501,10 @@ static void checkNewAttributesAfterDef(Sema &S, Decl *New, const Decl *Old) { ? diag::err_alias_after_tentative : diag::err_redefinition; S.Diag(VD->getLocation(), Diag) << VD->getDeclName(); - S.Diag(Def->getLocation(), diag::note_previous_definition); + if (Diag == diag::err_redefinition) + S.notePreviousDefinition(Def->getLocation(), VD->getLocation()); + else + S.Diag(Def->getLocation(), diag::note_previous_definition); VD->setInvalidDecl(); } ++I; @@ -2839,7 +2891,7 @@ bool Sema::MergeFunctionDecl(FunctionDecl *New, NamedDecl *&OldD, } else { Diag(New->getLocation(), diag::err_redefinition_different_kind) << New->getDeclName(); - Diag(OldD->getLocation(), diag::note_previous_definition); + notePreviousDefinition(OldD->getLocation(), New->getLocation()); return true; } } @@ -2876,7 +2928,7 @@ bool Sema::MergeFunctionDecl(FunctionDecl *New, NamedDecl *&OldD, !Old->hasAttr()) { Diag(New->getLocation(), diag::err_internal_linkage_redeclaration) << New->getDeclName(); - Diag(Old->getLocation(), diag::note_previous_definition); + notePreviousDefinition(Old->getLocation(), New->getLocation()); New->dropAttr(); } @@ -3604,9 +3656,9 @@ void Sema::MergeVarDecl(VarDecl *New, LookupResult &Previous) { } if (!Old) { Diag(New->getLocation(), diag::err_redefinition_different_kind) - << New->getDeclName(); - Diag(Previous.getRepresentativeDecl()->getLocation(), - diag::note_previous_definition); + << New->getDeclName(); + notePreviousDefinition(Previous.getRepresentativeDecl()->getLocation(), + New->getLocation()); return New->setInvalidDecl(); } @@ -3635,7 +3687,7 @@ void Sema::MergeVarDecl(VarDecl *New, LookupResult &Previous) { Old->getStorageClass() == SC_None && !Old->hasAttr()) { Diag(New->getLocation(), diag::warn_weak_import) << New->getDeclName(); - Diag(Old->getLocation(), diag::note_previous_definition); + notePreviousDefinition(Old->getLocation(), New->getLocation()); // Remove weak_import attribute on new declaration. New->dropAttr(); } @@ -3644,7 +3696,7 @@ void Sema::MergeVarDecl(VarDecl *New, LookupResult &Previous) { !Old->hasAttr()) { Diag(New->getLocation(), diag::err_internal_linkage_redeclaration) << New->getDeclName(); - Diag(Old->getLocation(), diag::note_previous_definition); + notePreviousDefinition(Old->getLocation(), New->getLocation()); New->dropAttr(); } @@ -3801,6 +3853,67 @@ void Sema::MergeVarDecl(VarDecl *New, LookupResult &Previous) { New->setImplicitlyInline(); } +void Sema::notePreviousDefinition(SourceLocation Old, SourceLocation New) { + SourceManager &SrcMgr = getSourceManager(); + auto FNewDecLoc = SrcMgr.getDecomposedLoc(New); + auto FOldDecLoc = SrcMgr.getDecomposedLoc(Old); + auto *FNew = SrcMgr.getFileEntryForID(FNewDecLoc.first); + auto *FOld = SrcMgr.getFileEntryForID(FOldDecLoc.first); + auto &HSI = PP.getHeaderSearchInfo(); + StringRef HdrFilename = SrcMgr.getFilename(SrcMgr.getSpellingLoc(Old)); + + auto noteFromModuleOrInclude = [&](SourceLocation &Loc, + SourceLocation &IncLoc) -> bool { + Module *Mod = nullptr; + // Redefinition errors with modules are common with non modular mapped + // headers, example: a non-modular header H in module A that also gets + // included directly in a TU. Pointing twice to the same header/definition + // is confusing, try to get better diagnostics when modules is on. + if (getLangOpts().Modules) { + auto ModLoc = SrcMgr.getModuleImportLoc(Old); + if (!ModLoc.first.isInvalid()) + Mod = HSI.getModuleMap().inferModuleFromLocation( + FullSourceLoc(Loc, SrcMgr)); + } + + if (IncLoc.isValid()) { + if (Mod) { + Diag(IncLoc, diag::note_redefinition_modules_same_file) + << HdrFilename.str() << Mod->getFullModuleName(); + if (!Mod->DefinitionLoc.isInvalid()) + Diag(Mod->DefinitionLoc, diag::note_defined_here) + << Mod->getFullModuleName(); + } else { + Diag(IncLoc, diag::note_redefinition_include_same_file) + << HdrFilename.str(); + } + return true; + } + + return false; + }; + + // Is it the same file and same offset? Provide more information on why + // this leads to a redefinition error. + bool EmittedDiag = false; + if (FNew == FOld && FNewDecLoc.second == FOldDecLoc.second) { + SourceLocation OldIncLoc = SrcMgr.getIncludeLoc(FOldDecLoc.first); + SourceLocation NewIncLoc = SrcMgr.getIncludeLoc(FNewDecLoc.first); + EmittedDiag = noteFromModuleOrInclude(Old, OldIncLoc); + EmittedDiag |= noteFromModuleOrInclude(New, NewIncLoc); + + // If the header has no guards, emit a note suggesting one. + if (FOld && !HSI.isFileMultipleIncludeGuarded(FOld)) + Diag(Old, diag::note_use_ifdef_guards); + + if (EmittedDiag) + return; + } + + // Redefinition coming from different files or couldn't do better above. + Diag(Old, diag::note_previous_definition); +} + /// We've just determined that \p Old and \p New both appear to be definitions /// of the same variable. Either diagnose or fix the problem. bool Sema::checkVarDeclRedefinition(VarDecl *Old, VarDecl *New) { @@ -3816,12 +3929,12 @@ bool Sema::checkVarDeclRedefinition(VarDecl *Old, VarDecl *New) { // Make the canonical definition visible. if (auto *OldTD = Old->getDescribedVarTemplate()) - makeMergedDefinitionVisible(OldTD, New->getLocation()); - makeMergedDefinitionVisible(Old, New->getLocation()); + makeMergedDefinitionVisible(OldTD); + makeMergedDefinitionVisible(Old); return false; } else { Diag(New->getLocation(), diag::err_redefinition) << New; - Diag(Old->getLocation(), diag::note_previous_definition); + notePreviousDefinition(Old->getLocation(), New->getLocation()); New->setInvalidDecl(); return true; } @@ -6706,6 +6819,9 @@ NamedDecl *Sema::ActOnVariableDeclarator( return NewTemplate; } + if (IsMemberSpecialization && !NewVD->isInvalidDecl()) + CompleteMemberSpecialization(NewVD, Previous); + return NewVD; } @@ -8919,12 +9035,17 @@ Sema::ActOnFunctionDeclarator(Scope *S, Declarator &D, DeclContext *DC, } } + MarkUnusedFileScopedDecl(NewFD); + if (getLangOpts().CPlusPlus) { if (FunctionTemplate) { if (NewFD->isInvalidDecl()) FunctionTemplate->setInvalidDecl(); return FunctionTemplate; } + + if (isMemberSpecialization && !NewFD->isInvalidDecl()) + CompleteMemberSpecialization(NewFD, Previous); } if (NewFD->hasAttr()) { @@ -8964,8 +9085,6 @@ Sema::ActOnFunctionDeclarator(Scope *S, Declarator &D, DeclContext *DC, } } - MarkUnusedFileScopedDecl(NewFD); - // Here we have an function template explicit specialization at class scope. // The actually specialization will be postponed to template instatiation // time via the ClassScopeFunctionSpecializationDecl node. @@ -9182,7 +9301,9 @@ bool Sema::CheckFunctionDeclaration(Scope *S, FunctionDecl *NewFD, if (OldTemplateDecl->getTemplatedDecl()->isDeleted()) { FunctionDecl *const OldTemplatedDecl = OldTemplateDecl->getTemplatedDecl(); + // FIXME: This assert will not hold in the presence of modules. assert(OldTemplatedDecl->getCanonicalDecl() == OldTemplatedDecl); + // FIXME: We need an update record for this AST mutation. OldTemplatedDecl->setDeletedAsWritten(false); } } @@ -10273,23 +10394,36 @@ void Sema::AddInitializerToDecl(Decl *RealDecl, Expr *Init, bool DirectInit) { VDecl->setInit(Init); if (VDecl->isLocalVarDecl()) { + // Don't check the initializer if the declaration is malformed. + if (VDecl->isInvalidDecl()) { + // do nothing + + // OpenCL v1.2 s6.5.3: __constant locals must be constant-initialized. + // This is true even in OpenCL C++. + } else if (VDecl->getType().getAddressSpace() == LangAS::opencl_constant) { + CheckForConstantInitializer(Init, DclT); + + // Otherwise, C++ does not restrict the initializer. + } else if (getLangOpts().CPlusPlus) { + // do nothing + // C99 6.7.8p4: All the expressions in an initializer for an object that has // static storage duration shall be constant expressions or string literals. - // C++ does not have this restriction. - if (!getLangOpts().CPlusPlus && !VDecl->isInvalidDecl()) { + } else if (VDecl->getStorageClass() == SC_Static) { + CheckForConstantInitializer(Init, DclT); + + // C89 is stricter than C99 for aggregate initializers. + // C89 6.5.7p3: All the expressions [...] in an initializer list + // for an object that has aggregate or union type shall be + // constant expressions. + } else if (!getLangOpts().C99 && VDecl->getType()->isAggregateType() && + isa(Init)) { const Expr *Culprit; - if (VDecl->getStorageClass() == SC_Static) - CheckForConstantInitializer(Init, DclT); - // C89 is stricter than C99 for non-static aggregate types. - // C89 6.5.7p3: All the expressions [...] in an initializer list - // for an object that has aggregate or union type shall be - // constant expressions. - else if (!getLangOpts().C99 && VDecl->getType()->isAggregateType() && - isa(Init) && - !Init->isConstantInitializer(Context, false, &Culprit)) + if (!Init->isConstantInitializer(Context, false, &Culprit)) { Diag(Culprit->getExprLoc(), diag::ext_aggregate_init_not_constant) << Culprit->getSourceRange(); + } } } else if (VDecl->isStaticDataMember() && !VDecl->isInline() && VDecl->getLexicalDeclContext()->isRecord()) { @@ -11023,8 +11157,7 @@ static bool hasDependentAlignment(VarDecl *VD) { /// FinalizeDeclaration - called by ParseDeclarationAfterDeclarator to perform /// any semantic actions necessary after any initializer has been attached. -void -Sema::FinalizeDeclaration(Decl *ThisDecl) { +void Sema::FinalizeDeclaration(Decl *ThisDecl) { // Note that we are no longer parsing the initializer for this declaration. ParsingInitForAutoVars.erase(ThisDecl); @@ -11189,9 +11322,8 @@ Sema::FinalizeDeclaration(Decl *ThisDecl) { if (DC->getRedeclContext()->isFileContext() && VD->isExternallyVisible()) AddPushedVisibilityAttribute(VD); - // FIXME: Warn on unused templates. - if (VD->isFileVarDecl() && !VD->getDescribedVarTemplate() && - !isa(VD)) + // FIXME: Warn on unused var template partial specializations. + if (VD->isFileVarDecl() && !isa(VD)) MarkUnusedFileScopedDecl(VD); // Now we have parsed the initializer and can update the table of magic @@ -11765,9 +11897,8 @@ Sema::CheckForFunctionRedefinition(FunctionDecl *FD, Definition->getNumTemplateParameterLists())) { SkipBody->ShouldSkip = true; if (auto *TD = Definition->getDescribedFunctionTemplate()) - makeMergedDefinitionVisible(TD, FD->getLocation()); - makeMergedDefinitionVisible(const_cast(Definition), - FD->getLocation()); + makeMergedDefinitionVisible(TD); + makeMergedDefinitionVisible(const_cast(Definition)); return; } @@ -13421,7 +13552,7 @@ Decl *Sema::ActOnTag(Scope *S, unsigned TagSpec, TagUseKind TUK, // we already have. Make the existing definition visible and // use it in place of this one. SkipBody->ShouldSkip = true; - makeMergedDefinitionVisible(Hidden, KWLoc); + makeMergedDefinitionVisible(Hidden); return Def; } else if (!IsExplicitSpecializationAfterInstantiation) { // A redeclaration in function prototype scope in C isn't @@ -13430,7 +13561,8 @@ Decl *Sema::ActOnTag(Scope *S, unsigned TagSpec, TagUseKind TUK, Diag(NameLoc, diag::warn_redefinition_in_param_list) << Name; else Diag(NameLoc, diag::err_redefinition) << Name; - Diag(Def->getLocation(), diag::note_previous_definition); + notePreviousDefinition(Def->getLocation(), + NameLoc.isValid() ? NameLoc : KWLoc); // If this is a redefinition, recover by making this // struct be anonymous, which will make any later // references get the previous definition. @@ -13520,7 +13652,7 @@ Decl *Sema::ActOnTag(Scope *S, unsigned TagSpec, TagUseKind TUK, // The tag name clashes with something else in the target scope, // issue an error and recover by making this tag be anonymous. Diag(NameLoc, diag::err_redefinition_different_kind) << Name; - Diag(PrevDecl->getLocation(), diag::note_previous_definition); + notePreviousDefinition(PrevDecl->getLocation(), NameLoc); Name = nullptr; Invalid = true; } @@ -13753,6 +13885,9 @@ Decl *Sema::ActOnTag(Scope *S, unsigned TagSpec, TagUseKind TUK, // record. AddPushedVisibilityAttribute(New); + if (isMemberSpecialization && !New->isInvalidDecl()) + CompleteMemberSpecialization(New, Previous); + OwnedDecl = true; // In C++, don't return an invalid declaration. We can't recover well from // the cases where we make the type anonymous. @@ -15221,7 +15356,7 @@ Decl *Sema::ActOnEnumConstant(Scope *S, Decl *theEnumDecl, Decl *lastEnumConst, Diag(IdLoc, diag::err_redefinition_of_enumerator) << Id; else Diag(IdLoc, diag::err_redefinition) << Id; - Diag(PrevDecl->getLocation(), diag::note_previous_definition); + notePreviousDefinition(PrevDecl->getLocation(), IdLoc); return nullptr; } } diff --git a/lib/Sema/SemaDeclAttr.cpp b/lib/Sema/SemaDeclAttr.cpp index 97d273f6ddb6..3de792e4e406 100644 --- a/lib/Sema/SemaDeclAttr.cpp +++ b/lib/Sema/SemaDeclAttr.cpp @@ -7230,6 +7230,13 @@ class DiagnoseUnguardedAvailability SemaRef.Context.getTargetInfo().getPlatformMinVersion()); } + bool TraverseDecl(Decl *D) { + // Avoid visiting nested functions to prevent duplicate warnings. + if (!D || isa(D)) + return true; + return Base::TraverseDecl(D); + } + bool TraverseStmt(Stmt *S) { if (!S) return true; @@ -7243,6 +7250,8 @@ class DiagnoseUnguardedAvailability bool TraverseIfStmt(IfStmt *If); + bool TraverseLambdaExpr(LambdaExpr *E) { return true; } + bool VisitObjCMessageExpr(ObjCMessageExpr *Msg) { if (ObjCMethodDecl *D = Msg->getMethodDecl()) DiagnoseDeclAvailability( @@ -7346,7 +7355,9 @@ void DiagnoseUnguardedAvailability::DiagnoseDeclAvailability( llvm::raw_string_ostream FixItOS(FixItString); FixItOS << "if (" << (SemaRef.getLangOpts().ObjC1 ? "@available" : "__builtin_available") - << "(" << SemaRef.getASTContext().getTargetInfo().getPlatformName() + << "(" + << AvailabilityAttr::getPlatformNameSourceSpelling( + SemaRef.getASTContext().getTargetInfo().getPlatformName()) << " " << Introduced.getAsString() << ", *)) {\n" << Indentation << ExtraIndentation; FixitDiag << FixItHint::CreateInsertion(IfInsertionLoc, FixItOS.str()); diff --git a/lib/Sema/SemaExpr.cpp b/lib/Sema/SemaExpr.cpp index 849e978e2d86..14efc9672061 100644 --- a/lib/Sema/SemaExpr.cpp +++ b/lib/Sema/SemaExpr.cpp @@ -5277,6 +5277,9 @@ ExprResult Sema::ActOnCallExpr(Scope *Scope, Expr *Fn, SourceLocation LParenLoc, // We aren't supposed to apply this logic if there's an '&' involved. if (!find.HasFormOfMemberPointer) { + if (Expr::hasAnyTypeDependentArguments(ArgExprs)) + return new (Context) CallExpr( + Context, Fn, ArgExprs, Context.DependentTy, VK_RValue, RParenLoc); OverloadExpr *ovl = find.Expression; if (UnresolvedLookupExpr *ULE = dyn_cast(ovl)) return BuildOverloadedCallExpr( @@ -8028,6 +8031,33 @@ QualType Sema::InvalidOperands(SourceLocation Loc, ExprResult &LHS, return QualType(); } +// Diagnose cases where a scalar was implicitly converted to a vector and +// diagnose the underlying types. Otherwise, diagnose the error +// as invalid vector logical operands for non-C++ cases. +QualType Sema::InvalidLogicalVectorOperands(SourceLocation Loc, ExprResult &LHS, + ExprResult &RHS) { + QualType LHSType = LHS.get()->IgnoreImpCasts()->getType(); + QualType RHSType = RHS.get()->IgnoreImpCasts()->getType(); + + bool LHSNatVec = LHSType->isVectorType(); + bool RHSNatVec = RHSType->isVectorType(); + + if (!(LHSNatVec && RHSNatVec)) { + Expr *Vector = LHSNatVec ? LHS.get() : RHS.get(); + Expr *NonVector = !LHSNatVec ? LHS.get() : RHS.get(); + Diag(Loc, diag::err_typecheck_logical_vector_expr_gnu_cpp_restrict) + << 0 << Vector->getType() << NonVector->IgnoreImpCasts()->getType() + << Vector->getSourceRange(); + return QualType(); + } + + Diag(Loc, diag::err_typecheck_logical_vector_expr_gnu_cpp_restrict) + << 1 << LHSType << RHSType << LHS.get()->getSourceRange() + << RHS.get()->getSourceRange(); + + return QualType(); +} + /// Try to convert a value of non-vector type to a vector type by converting /// the type to the element type of the vector and then performing a splat. /// If the language is OpenCL, we only use conversions that promote scalar @@ -8075,6 +8105,162 @@ static bool tryVectorConvertAndSplat(Sema &S, ExprResult *scalar, return false; } +/// Test if a (constant) integer Int can be casted to another integer type +/// IntTy without losing precision. +static bool canConvertIntToOtherIntTy(Sema &S, ExprResult *Int, + QualType OtherIntTy) { + QualType IntTy = Int->get()->getType().getUnqualifiedType(); + + // Reject cases where the value of the Int is unknown as that would + // possibly cause truncation, but accept cases where the scalar can be + // demoted without loss of precision. + llvm::APSInt Result; + bool CstInt = Int->get()->EvaluateAsInt(Result, S.Context); + int Order = S.Context.getIntegerTypeOrder(OtherIntTy, IntTy); + bool IntSigned = IntTy->hasSignedIntegerRepresentation(); + bool OtherIntSigned = OtherIntTy->hasSignedIntegerRepresentation(); + + if (CstInt) { + // If the scalar is constant and is of a higher order and has more active + // bits that the vector element type, reject it. + unsigned NumBits = IntSigned + ? (Result.isNegative() ? Result.getMinSignedBits() + : Result.getActiveBits()) + : Result.getActiveBits(); + if (Order < 0 && S.Context.getIntWidth(OtherIntTy) < NumBits) + return true; + + // If the signedness of the scalar type and the vector element type + // differs and the number of bits is greater than that of the vector + // element reject it. + return (IntSigned != OtherIntSigned && + NumBits > S.Context.getIntWidth(OtherIntTy)); + } + + // Reject cases where the value of the scalar is not constant and it's + // order is greater than that of the vector element type. + return (Order < 0); +} + +/// Test if a (constant) integer Int can be casted to floating point type +/// FloatTy without losing precision. +static bool canConvertIntTyToFloatTy(Sema &S, ExprResult *Int, + QualType FloatTy) { + QualType IntTy = Int->get()->getType().getUnqualifiedType(); + + // Determine if the integer constant can be expressed as a floating point + // number of the appropiate type. + llvm::APSInt Result; + bool CstInt = Int->get()->EvaluateAsInt(Result, S.Context); + uint64_t Bits = 0; + if (CstInt) { + // Reject constants that would be truncated if they were converted to + // the floating point type. Test by simple to/from conversion. + // FIXME: Ideally the conversion to an APFloat and from an APFloat + // could be avoided if there was a convertFromAPInt method + // which could signal back if implicit truncation occurred. + llvm::APFloat Float(S.Context.getFloatTypeSemantics(FloatTy)); + Float.convertFromAPInt(Result, IntTy->hasSignedIntegerRepresentation(), + llvm::APFloat::rmTowardZero); + llvm::APSInt ConvertBack(S.Context.getIntWidth(IntTy), + !IntTy->hasSignedIntegerRepresentation()); + bool Ignored = false; + Float.convertToInteger(ConvertBack, llvm::APFloat::rmNearestTiesToEven, + &Ignored); + if (Result != ConvertBack) + return true; + } else { + // Reject types that cannot be fully encoded into the mantissa of + // the float. + Bits = S.Context.getTypeSize(IntTy); + unsigned FloatPrec = llvm::APFloat::semanticsPrecision( + S.Context.getFloatTypeSemantics(FloatTy)); + if (Bits > FloatPrec) + return true; + } + + return false; +} + +/// Attempt to convert and splat Scalar into a vector whose types matches +/// Vector following GCC conversion rules. The rule is that implicit +/// conversion can occur when Scalar can be casted to match Vector's element +/// type without causing truncation of Scalar. +static bool tryGCCVectorConvertAndSplat(Sema &S, ExprResult *Scalar, + ExprResult *Vector) { + QualType ScalarTy = Scalar->get()->getType().getUnqualifiedType(); + QualType VectorTy = Vector->get()->getType().getUnqualifiedType(); + const VectorType *VT = VectorTy->getAs(); + + assert(!isa(VT) && + "ExtVectorTypes should not be handled here!"); + + QualType VectorEltTy = VT->getElementType(); + + // Reject cases where the vector element type or the scalar element type are + // not integral or floating point types. + if (!VectorEltTy->isArithmeticType() || !ScalarTy->isArithmeticType()) + return true; + + // The conversion to apply to the scalar before splatting it, + // if necessary. + CastKind ScalarCast = CK_NoOp; + + // Accept cases where the vector elements are integers and the scalar is + // an integer. + // FIXME: Notionally if the scalar was a floating point value with a precise + // integral representation, we could cast it to an appropriate integer + // type and then perform the rest of the checks here. GCC will perform + // this conversion in some cases as determined by the input language. + // We should accept it on a language independent basis. + if (VectorEltTy->isIntegralType(S.Context) && + ScalarTy->isIntegralType(S.Context) && + S.Context.getIntegerTypeOrder(VectorEltTy, ScalarTy)) { + + if (canConvertIntToOtherIntTy(S, Scalar, VectorEltTy)) + return true; + + ScalarCast = CK_IntegralCast; + } else if (VectorEltTy->isRealFloatingType()) { + if (ScalarTy->isRealFloatingType()) { + + // Reject cases where the scalar type is not a constant and has a higher + // Order than the vector element type. + llvm::APFloat Result(0.0); + bool CstScalar = Scalar->get()->EvaluateAsFloat(Result, S.Context); + int Order = S.Context.getFloatingTypeOrder(VectorEltTy, ScalarTy); + if (!CstScalar && Order < 0) + return true; + + // If the scalar cannot be safely casted to the vector element type, + // reject it. + if (CstScalar) { + bool Truncated = false; + Result.convert(S.Context.getFloatTypeSemantics(VectorEltTy), + llvm::APFloat::rmNearestTiesToEven, &Truncated); + if (Truncated) + return true; + } + + ScalarCast = CK_FloatingCast; + } else if (ScalarTy->isIntegralType(S.Context)) { + if (canConvertIntTyToFloatTy(S, Scalar, VectorEltTy)) + return true; + + ScalarCast = CK_IntegralToFloating; + } else + return true; + } + + // Adjust scalar if desired. + if (Scalar) { + if (ScalarCast != CK_NoOp) + *Scalar = S.ImpCastExprToType(Scalar->get(), VectorEltTy, ScalarCast); + *Scalar = S.ImpCastExprToType(Scalar->get(), VectorTy, CK_VectorSplat); + } + return false; +} + QualType Sema::CheckVectorOperands(ExprResult &LHS, ExprResult &RHS, SourceLocation Loc, bool IsCompAssign, bool AllowBothBool, @@ -8143,19 +8329,29 @@ QualType Sema::CheckVectorOperands(ExprResult &LHS, ExprResult &RHS, } } - // If there's an ext-vector type and a scalar, try to convert the scalar to + // If there's a vector type and a scalar, try to convert the scalar to // the vector element type and splat. - // FIXME: this should also work for regular vector types as supported in GCC. - if (!RHSVecType && isa(LHSVecType)) { - if (!tryVectorConvertAndSplat(*this, &RHS, RHSType, - LHSVecType->getElementType(), LHSType)) - return LHSType; + if (!RHSVecType) { + if (isa(LHSVecType)) { + if (!tryVectorConvertAndSplat(*this, &RHS, RHSType, + LHSVecType->getElementType(), LHSType)) + return LHSType; + } else { + if (!tryGCCVectorConvertAndSplat(*this, &RHS, &LHS)) + return LHSType; + } } - if (!LHSVecType && isa(RHSVecType)) { - if (!tryVectorConvertAndSplat(*this, (IsCompAssign ? nullptr : &LHS), - LHSType, RHSVecType->getElementType(), - RHSType)) - return RHSType; + if (!LHSVecType) { + if (isa(RHSVecType)) { + if (!tryVectorConvertAndSplat(*this, (IsCompAssign ? nullptr : &LHS), + LHSType, RHSVecType->getElementType(), + RHSType)) + return RHSType; + } else { + if (LHS.get()->getValueKind() == VK_LValue || + !tryGCCVectorConvertAndSplat(*this, &LHS, &RHS)) + return RHSType; + } } // FIXME: The code below also handles conversion between vectors and @@ -8208,6 +8404,22 @@ QualType Sema::CheckVectorOperands(ExprResult &LHS, ExprResult &RHS, return QualType(); } + + // If there is a vector type that is not a ExtVector and a scalar, we reach + // this point if scalar could not be converted to the vector's element type + // without truncation. + if ((RHSVecType && !isa(RHSVecType)) || + (LHSVecType && !isa(LHSVecType))) { + QualType Scalar = LHSVecType ? RHSType : LHSType; + QualType Vector = LHSVecType ? LHSType : RHSType; + unsigned ScalarOrVector = LHSVecType && RHSVecType ? 1 : 0; + Diag(Loc, + diag::err_typecheck_vector_not_convertable_implict_truncation) + << ScalarOrVector << Scalar << Vector; + + return QualType(); + } + // Otherwise, use the generic diagnostic. Diag(Loc, diag::err_typecheck_vector_not_convertable) << LHSType << RHSType @@ -9827,6 +10039,12 @@ QualType Sema::CheckVectorLogicalOperands(ExprResult &LHS, ExprResult &RHS, if (getLangOpts().OpenCL && getLangOpts().OpenCLVersion < 120 && vType->hasFloatingRepresentation()) return InvalidOperands(Loc, LHS, RHS); + // FIXME: The check for C++ here is for GCC compatibility. GCC rejects the + // usage of the logical operators && and || with vectors in C. This + // check could be notionally dropped. + if (!getLangOpts().CPlusPlus && + !(isa(vType->getAs()))) + return InvalidLogicalVectorOperands(Loc, LHS, RHS); return GetSignedVectorType(LHS.get()->getType()); } @@ -11770,6 +11988,8 @@ ExprResult Sema::CreateBuiltinUnaryOp(SourceLocation OpLoc, resultType = GetSignedVectorType(resultType); break; } else { + // FIXME: GCC's vector extension permits the usage of '!' with a vector + // type in C++. We should allow that here too. return ExprError(Diag(OpLoc, diag::err_typecheck_unary_expr) << resultType << Input.get()->getSourceRange()); } diff --git a/lib/Sema/SemaExprCXX.cpp b/lib/Sema/SemaExprCXX.cpp index 9b88cddbc969..8500b748a3ec 100644 --- a/lib/Sema/SemaExprCXX.cpp +++ b/lib/Sema/SemaExprCXX.cpp @@ -4720,10 +4720,24 @@ static bool EvaluateBinaryTypeTrait(Sema &Self, TypeTrait BTT, QualType LhsT, // regard to cv-qualifiers. const RecordType *lhsRecord = LhsT->getAs(); - if (!lhsRecord) return false; - const RecordType *rhsRecord = RhsT->getAs(); - if (!rhsRecord) return false; + if (!rhsRecord || !lhsRecord) { + const ObjCObjectType *LHSObjTy = LhsT->getAs(); + const ObjCObjectType *RHSObjTy = RhsT->getAs(); + if (!LHSObjTy || !RHSObjTy) + return false; + + ObjCInterfaceDecl *BaseInterface = LHSObjTy->getInterface(); + ObjCInterfaceDecl *DerivedInterface = RHSObjTy->getInterface(); + if (!BaseInterface || !DerivedInterface) + return false; + + if (Self.RequireCompleteType( + KeyLoc, RhsT, diag::err_incomplete_type_used_in_type_trait_expr)) + return false; + + return BaseInterface->isSuperClassOf(DerivedInterface); + } assert(Self.Context.hasSameUnqualifiedType(LhsT, RhsT) == (lhsRecord == rhsRecord)); @@ -5342,6 +5356,15 @@ QualType Sema::CXXCheckConditionalOperands(ExprResult &Cond, ExprResult &LHS, // C++11 [expr.cond]p1 // The first expression is contextually converted to bool. + // + // FIXME; GCC's vector extension permits the use of a?b:c where the type of + // a is that of a integer vector with the same number of elements and + // size as the vectors of b and c. If one of either b or c is a scalar + // it is implicitly converted to match the type of the vector. + // Otherwise the expression is ill-formed. If both b and c are scalars, + // then b and c are checked and converted to the type of a if possible. + // Unlike the OpenCL ?: operator, the expression is evaluated as + // (a[0] != 0 ? b[0] : c[0], .. , a[n] != 0 ? b[n] : c[n]). if (!Cond.get()->isTypeDependent()) { ExprResult CondRes = CheckCXXBooleanCondition(Cond.get()); if (CondRes.isInvalid()) diff --git a/lib/Sema/SemaExprObjC.cpp b/lib/Sema/SemaExprObjC.cpp index a44e9243e3c5..28581bad1a7a 100644 --- a/lib/Sema/SemaExprObjC.cpp +++ b/lib/Sema/SemaExprObjC.cpp @@ -4241,8 +4241,7 @@ void Sema::diagnoseARCUnbridgedCast(Expr *e) { castType = cast->getTypeAsWritten(); CCK = CCK_OtherCast; } else { - castType = cast->getType(); - CCK = CCK_ImplicitConversion; + llvm_unreachable("Unexpected ImplicitCastExpr"); } ARCConversionTypeClass castACTC = diff --git a/lib/Sema/SemaInit.cpp b/lib/Sema/SemaInit.cpp index d0f530010a0d..32024cb335dc 100644 --- a/lib/Sema/SemaInit.cpp +++ b/lib/Sema/SemaInit.cpp @@ -1209,7 +1209,7 @@ void InitListChecker::CheckSubElementType(const InitializedEntity &Entity, } else { assert((ElemType->isRecordType() || ElemType->isVectorType() || - ElemType->isClkEventT()) && "Unexpected type"); + ElemType->isOpenCLSpecificType()) && "Unexpected type"); // C99 6.7.8p13: // @@ -8296,8 +8296,46 @@ Sema::PerformCopyInitialization(const InitializedEntity &Entity, AllowExplicit); InitializationSequence Seq(*this, Entity, Kind, InitE, TopLevelOfInitList); + // Prevent infinite recursion when performing parameter copy-initialization. + const bool ShouldTrackCopy = + Entity.isParameterKind() && Seq.isConstructorInitialization(); + if (ShouldTrackCopy) { + if (llvm::find(CurrentParameterCopyTypes, Entity.getType()) != + CurrentParameterCopyTypes.end()) { + Seq.SetOverloadFailure( + InitializationSequence::FK_ConstructorOverloadFailed, + OR_No_Viable_Function); + + // Try to give a meaningful diagnostic note for the problematic + // constructor. + const auto LastStep = Seq.step_end() - 1; + assert(LastStep->Kind == + InitializationSequence::SK_ConstructorInitialization); + const FunctionDecl *Function = LastStep->Function.Function; + auto Candidate = + llvm::find_if(Seq.getFailedCandidateSet(), + [Function](const OverloadCandidate &Candidate) -> bool { + return Candidate.Viable && + Candidate.Function == Function && + Candidate.Conversions.size() > 0; + }); + if (Candidate != Seq.getFailedCandidateSet().end() && + Function->getNumParams() > 0) { + Candidate->Viable = false; + Candidate->FailureKind = ovl_fail_bad_conversion; + Candidate->Conversions[0].setBad(BadConversionSequence::no_conversion, + InitE, + Function->getParamDecl(0)->getType()); + } + } + CurrentParameterCopyTypes.push_back(Entity.getType()); + } + ExprResult Result = Seq.Perform(*this, Entity, Kind, InitE); + if (ShouldTrackCopy) + CurrentParameterCopyTypes.pop_back(); + return Result; } diff --git a/lib/Sema/SemaLookup.cpp b/lib/Sema/SemaLookup.cpp index ce76e14982db..c5b579a4b2e9 100644 --- a/lib/Sema/SemaLookup.cpp +++ b/lib/Sema/SemaLookup.cpp @@ -1382,8 +1382,8 @@ Module *Sema::getOwningModule(Decl *Entity) { return M; } -void Sema::makeMergedDefinitionVisible(NamedDecl *ND, SourceLocation Loc) { - if (auto *M = PP.getModuleContainingLocation(Loc)) +void Sema::makeMergedDefinitionVisible(NamedDecl *ND) { + if (auto *M = getCurrentModule()) Context.mergeDefinitionIntoModule(ND, M); else // We're not building a module; just make the definition visible. @@ -1393,7 +1393,7 @@ void Sema::makeMergedDefinitionVisible(NamedDecl *ND, SourceLocation Loc) { // visible too. They're not (necessarily) within a mergeable DeclContext. if (auto *TD = dyn_cast(ND)) for (auto *Param : *TD->getTemplateParameters()) - makeMergedDefinitionVisible(Param, Loc); + makeMergedDefinitionVisible(Param); } /// \brief Find the module in which the given declaration was defined. @@ -3445,7 +3445,8 @@ static void LookupVisibleDecls(DeclContext *Ctx, LookupResult &Result, bool QualifiedNameLookup, bool InBaseClass, VisibleDeclConsumer &Consumer, - VisibleDeclsRecord &Visited) { + VisibleDeclsRecord &Visited, + bool IncludeDependentBases = false) { if (!Ctx) return; @@ -3501,7 +3502,8 @@ static void LookupVisibleDecls(DeclContext *Ctx, LookupResult &Result, ShadowContextRAII Shadow(Visited); for (auto I : Ctx->using_directives()) { LookupVisibleDecls(I->getNominatedNamespace(), Result, - QualifiedNameLookup, InBaseClass, Consumer, Visited); + QualifiedNameLookup, InBaseClass, Consumer, Visited, + IncludeDependentBases); } } @@ -3513,14 +3515,28 @@ static void LookupVisibleDecls(DeclContext *Ctx, LookupResult &Result, for (const auto &B : Record->bases()) { QualType BaseType = B.getType(); - // Don't look into dependent bases, because name lookup can't look - // there anyway. - if (BaseType->isDependentType()) - continue; - - const RecordType *Record = BaseType->getAs(); - if (!Record) - continue; + RecordDecl *RD; + if (BaseType->isDependentType()) { + if (!IncludeDependentBases) { + // Don't look into dependent bases, because name lookup can't look + // there anyway. + continue; + } + const auto *TST = BaseType->getAs(); + if (!TST) + continue; + TemplateName TN = TST->getTemplateName(); + const auto *TD = + dyn_cast_or_null(TN.getAsTemplateDecl()); + if (!TD) + continue; + RD = TD->getTemplatedDecl(); + } else { + const auto *Record = BaseType->getAs(); + if (!Record) + continue; + RD = Record->getDecl(); + } // FIXME: It would be nice to be able to determine whether referencing // a particular member would be ambiguous. For example, given @@ -3543,8 +3559,8 @@ static void LookupVisibleDecls(DeclContext *Ctx, LookupResult &Result, // Find results in this base class (and its bases). ShadowContextRAII Shadow(Visited); - LookupVisibleDecls(Record->getDecl(), Result, QualifiedNameLookup, - true, Consumer, Visited); + LookupVisibleDecls(RD, Result, QualifiedNameLookup, true, Consumer, + Visited, IncludeDependentBases); } } @@ -3713,7 +3729,8 @@ void Sema::LookupVisibleDecls(Scope *S, LookupNameKind Kind, void Sema::LookupVisibleDecls(DeclContext *Ctx, LookupNameKind Kind, VisibleDeclConsumer &Consumer, - bool IncludeGlobalScope) { + bool IncludeGlobalScope, + bool IncludeDependentBases) { LookupResult Result(*this, DeclarationName(), SourceLocation(), Kind); Result.setAllowHidden(Consumer.includeHiddenDecls()); VisibleDeclsRecord Visited; @@ -3721,7 +3738,8 @@ void Sema::LookupVisibleDecls(DeclContext *Ctx, LookupNameKind Kind, Visited.visitedContext(Context.getTranslationUnitDecl()); ShadowContextRAII Shadow(Visited); ::LookupVisibleDecls(Ctx, Result, /*QualifiedNameLookup=*/true, - /*InBaseClass=*/false, Consumer, Visited); + /*InBaseClass=*/false, Consumer, Visited, + IncludeDependentBases); } /// LookupOrCreateLabel - Do a name lookup of a label with the specified name. diff --git a/lib/Sema/SemaOverload.cpp b/lib/Sema/SemaOverload.cpp index 782c377e3202..51794160278c 100644 --- a/lib/Sema/SemaOverload.cpp +++ b/lib/Sema/SemaOverload.cpp @@ -11210,12 +11210,12 @@ Sema::resolveAddressOfOnlyViableOverloadCandidate(Expr *E, /// \brief Given an overloaded function, tries to turn it into a non-overloaded /// function reference using resolveAddressOfOnlyViableOverloadCandidate. This /// will perform access checks, diagnose the use of the resultant decl, and, if -/// necessary, perform a function-to-pointer decay. +/// requested, potentially perform a function-to-pointer decay. /// /// Returns false if resolveAddressOfOnlyViableOverloadCandidate fails. /// Otherwise, returns true. This may emit diagnostics and return true. bool Sema::resolveAndFixAddressOfOnlyViableOverloadCandidate( - ExprResult &SrcExpr) { + ExprResult &SrcExpr, bool DoFunctionPointerConverion) { Expr *E = SrcExpr.get(); assert(E->getType() == Context.OverloadTy && "SrcExpr must be an overload"); @@ -11230,7 +11230,7 @@ bool Sema::resolveAndFixAddressOfOnlyViableOverloadCandidate( DiagnoseUseOfDecl(Found, E->getExprLoc()); CheckAddressOfMemberAccess(E, DAP); Expr *Fixed = FixOverloadedFunctionReference(E, DAP, Found); - if (Fixed->getType()->isFunctionType()) + if (DoFunctionPointerConverion && Fixed->getType()->isFunctionType()) SrcExpr = DefaultFunctionArrayConversion(Fixed, /*Diagnose=*/false); else SrcExpr = Fixed; diff --git a/lib/Sema/SemaStmt.cpp b/lib/Sema/SemaStmt.cpp index 5d7eada28717..33a8f9c4afa3 100644 --- a/lib/Sema/SemaStmt.cpp +++ b/lib/Sema/SemaStmt.cpp @@ -2268,9 +2268,57 @@ Sema::BuildCXXForRangeStmt(SourceLocation ForLoc, SourceLocation CoawaitLoc, BoundExpr = IntegerLiteral::Create( Context, CAT->getSize(), Context.getPointerDiffType(), RangeLoc); else if (const VariableArrayType *VAT = - dyn_cast(UnqAT)) - BoundExpr = VAT->getSizeExpr(); - else { + dyn_cast(UnqAT)) { + // For a variably modified type we can't just use the expression within + // the array bounds, since we don't want that to be re-evaluated here. + // Rather, we need to determine what it was when the array was first + // created - so we resort to using sizeof(vla)/sizeof(element). + // For e.g. + // void f(int b) { + // int vla[b]; + // b = -1; <-- This should not affect the num of iterations below + // for (int &c : vla) { .. } + // } + + // FIXME: This results in codegen generating IR that recalculates the + // run-time number of elements (as opposed to just using the IR Value + // that corresponds to the run-time value of each bound that was + // generated when the array was created.) If this proves too embarassing + // even for unoptimized IR, consider passing a magic-value/cookie to + // codegen that then knows to simply use that initial llvm::Value (that + // corresponds to the bound at time of array creation) within + // getelementptr. But be prepared to pay the price of increasing a + // customized form of coupling between the two components - which could + // be hard to maintain as the codebase evolves. + + ExprResult SizeOfVLAExprR = ActOnUnaryExprOrTypeTraitExpr( + EndVar->getLocation(), UETT_SizeOf, + /*isType=*/true, + CreateParsedType(VAT->desugar(), Context.getTrivialTypeSourceInfo( + VAT->desugar(), RangeLoc)) + .getAsOpaquePtr(), + EndVar->getSourceRange()); + if (SizeOfVLAExprR.isInvalid()) + return StmtError(); + + ExprResult SizeOfEachElementExprR = ActOnUnaryExprOrTypeTraitExpr( + EndVar->getLocation(), UETT_SizeOf, + /*isType=*/true, + CreateParsedType(VAT->desugar(), + Context.getTrivialTypeSourceInfo( + VAT->getElementType(), RangeLoc)) + .getAsOpaquePtr(), + EndVar->getSourceRange()); + if (SizeOfEachElementExprR.isInvalid()) + return StmtError(); + + BoundExpr = + ActOnBinOp(S, EndVar->getLocation(), tok::slash, + SizeOfVLAExprR.get(), SizeOfEachElementExprR.get()); + if (BoundExpr.isInvalid()) + return StmtError(); + + } else { // Can't be a DependentSizedArrayType or an IncompleteArrayType since // UnqAT is not incomplete and Range is not type-dependent. llvm_unreachable("Unexpected array type in for-range"); diff --git a/lib/Sema/SemaTemplate.cpp b/lib/Sema/SemaTemplate.cpp index 61b4df40964c..a479d1027533 100644 --- a/lib/Sema/SemaTemplate.cpp +++ b/lib/Sema/SemaTemplate.cpp @@ -455,6 +455,85 @@ void Sema::LookupTemplateName(LookupResult &Found, } } +void Sema::diagnoseExprIntendedAsTemplateName(Scope *S, ExprResult TemplateName, + SourceLocation Less, + SourceLocation Greater) { + if (TemplateName.isInvalid()) + return; + + DeclarationNameInfo NameInfo; + CXXScopeSpec SS; + LookupNameKind LookupKind; + + DeclContext *LookupCtx = nullptr; + NamedDecl *Found = nullptr; + + // Figure out what name we looked up. + if (auto *ME = dyn_cast(TemplateName.get())) { + NameInfo = ME->getMemberNameInfo(); + SS.Adopt(ME->getQualifierLoc()); + LookupKind = LookupMemberName; + LookupCtx = ME->getBase()->getType()->getAsCXXRecordDecl(); + Found = ME->getMemberDecl(); + } else { + auto *DRE = cast(TemplateName.get()); + NameInfo = DRE->getNameInfo(); + SS.Adopt(DRE->getQualifierLoc()); + LookupKind = LookupOrdinaryName; + Found = DRE->getFoundDecl(); + } + + // Try to correct the name by looking for templates and C++ named casts. + struct TemplateCandidateFilter : CorrectionCandidateCallback { + TemplateCandidateFilter() { + WantTypeSpecifiers = false; + WantExpressionKeywords = false; + WantRemainingKeywords = false; + WantCXXNamedCasts = true; + }; + bool ValidateCandidate(const TypoCorrection &Candidate) override { + if (auto *ND = Candidate.getCorrectionDecl()) + return isAcceptableTemplateName(ND->getASTContext(), ND, true); + return Candidate.isKeyword(); + } + }; + + DeclarationName Name = NameInfo.getName(); + if (TypoCorrection Corrected = + CorrectTypo(NameInfo, LookupKind, S, &SS, + llvm::make_unique(), + CTK_ErrorRecovery, LookupCtx)) { + auto *ND = Corrected.getFoundDecl(); + if (ND) + ND = isAcceptableTemplateName(Context, ND, + /*AllowFunctionTemplates*/ true); + if (ND || Corrected.isKeyword()) { + if (LookupCtx) { + std::string CorrectedStr(Corrected.getAsString(getLangOpts())); + bool DroppedSpecifier = Corrected.WillReplaceSpecifier() && + Name.getAsString() == CorrectedStr; + diagnoseTypo(Corrected, + PDiag(diag::err_non_template_in_member_template_id_suggest) + << Name << LookupCtx << DroppedSpecifier + << SS.getRange(), false); + } else { + diagnoseTypo(Corrected, + PDiag(diag::err_non_template_in_template_id_suggest) + << Name, false); + } + if (Found) + Diag(Found->getLocation(), + diag::note_non_template_in_template_id_found); + return; + } + } + + Diag(NameInfo.getLoc(), diag::err_non_template_in_template_id) + << Name << SourceRange(Less, Greater); + if (Found) + Diag(Found->getLocation(), diag::note_non_template_in_template_id_found); +} + /// ActOnDependentIdExpression - Handle a dependent id-expression that /// was just parsed. This is only possible with an explicit scope /// specifier naming a dependent type. @@ -1251,8 +1330,8 @@ Sema::CheckClassTemplate(Scope *S, unsigned TagSpec, TagUseKind TUK, auto *Tmpl = cast(Hidden)->getDescribedClassTemplate(); assert(Tmpl && "original definition of a class template is not a " "class template?"); - makeMergedDefinitionVisible(Hidden, KWLoc); - makeMergedDefinitionVisible(Tmpl, KWLoc); + makeMergedDefinitionVisible(Hidden); + makeMergedDefinitionVisible(Tmpl); return Def; } @@ -7352,7 +7431,7 @@ Sema::ActOnClassTemplateSpecialization(Scope *S, unsigned TagSpec, NamedDecl *Hidden = nullptr; if (Def && SkipBody && !hasVisibleDefinition(Def, &Hidden)) { SkipBody->ShouldSkip = true; - makeMergedDefinitionVisible(Hidden, KWLoc); + makeMergedDefinitionVisible(Hidden); // From here on out, treat this as just a redeclaration. TUK = TUK_Declaration; } else if (Def) { @@ -7825,6 +7904,9 @@ bool Sema::CheckFunctionTemplateSpecialization( // C++11 [dcl.constexpr]p1: An explicit specialization of a constexpr // function can differ from the template declaration with respect to // the constexpr specifier. + // FIXME: We need an update record for this AST mutation. + // FIXME: What if there are multiple such prior declarations (for instance, + // from different modules)? Specialization->setConstexpr(FD->isConstexpr()); } @@ -7872,9 +7954,11 @@ bool Sema::CheckFunctionTemplateSpecialization( // flag to not-deleted, so that we can inherit that information from 'FD'. if (Specialization->isDeleted() && !SpecInfo->isExplicitSpecialization() && !Specialization->getCanonicalDecl()->isReferenced()) { + // FIXME: This assert will not hold in the presence of modules. assert( Specialization->getCanonicalDecl() == Specialization && "This must be the only existing declaration of this specialization"); + // FIXME: We need an update record for this AST mutation. Specialization->setDeletedAsWritten(false); } SpecInfo->setTemplateSpecializationKind(TSK_ExplicitSpecialization); @@ -7987,8 +8071,11 @@ Sema::CheckMemberSpecialization(NamedDecl *Member, LookupResult &Previous) { return false; } - // If this is a friend, just bail out here before we start turning - // things into explicit specializations. + // A member specialization in a friend declaration isn't really declaring + // an explicit specialization, just identifying a specific (possibly implicit) + // specialization. Don't change the template specialization kind. + // + // FIXME: Is this really valid? Other compilers reject. if (Member->getFriendObjectKind() != Decl::FOK_None) { // Preserve instantiation information. if (InstantiatedFrom && isa(Member)) { @@ -8038,66 +8125,36 @@ Sema::CheckMemberSpecialization(NamedDecl *Member, LookupResult &Previous) { false)) return true; - // Note that this is an explicit instantiation of a member. - // the original declaration to note that it is an explicit specialization - // (if it was previously an implicit instantiation). This latter step - // makes bookkeeping easier. - if (isa(Member)) { + // Note that this member specialization is an "instantiation of" the + // corresponding member of the original template. + if (auto *MemberFunction = dyn_cast(Member)) { FunctionDecl *InstantiationFunction = cast(Instantiation); if (InstantiationFunction->getTemplateSpecializationKind() == TSK_ImplicitInstantiation) { - InstantiationFunction->setTemplateSpecializationKind( - TSK_ExplicitSpecialization); - InstantiationFunction->setLocation(Member->getLocation()); // Explicit specializations of member functions of class templates do not // inherit '=delete' from the member function they are specializing. if (InstantiationFunction->isDeleted()) { + // FIXME: This assert will not hold in the presence of modules. assert(InstantiationFunction->getCanonicalDecl() == InstantiationFunction); + // FIXME: We need an update record for this AST mutation. InstantiationFunction->setDeletedAsWritten(false); } } - cast(Member)->setInstantiationOfMemberFunction( - cast(InstantiatedFrom), - TSK_ExplicitSpecialization); - MarkUnusedFileScopedDecl(InstantiationFunction); - } else if (isa(Member)) { - VarDecl *InstantiationVar = cast(Instantiation); - if (InstantiationVar->getTemplateSpecializationKind() == - TSK_ImplicitInstantiation) { - InstantiationVar->setTemplateSpecializationKind( - TSK_ExplicitSpecialization); - InstantiationVar->setLocation(Member->getLocation()); - } - - cast(Member)->setInstantiationOfStaticDataMember( + MemberFunction->setInstantiationOfMemberFunction( + cast(InstantiatedFrom), TSK_ExplicitSpecialization); + } else if (auto *MemberVar = dyn_cast(Member)) { + MemberVar->setInstantiationOfStaticDataMember( cast(InstantiatedFrom), TSK_ExplicitSpecialization); - MarkUnusedFileScopedDecl(InstantiationVar); - } else if (isa(Member)) { - CXXRecordDecl *InstantiationClass = cast(Instantiation); - if (InstantiationClass->getTemplateSpecializationKind() == - TSK_ImplicitInstantiation) { - InstantiationClass->setTemplateSpecializationKind( - TSK_ExplicitSpecialization); - InstantiationClass->setLocation(Member->getLocation()); - } - - cast(Member)->setInstantiationOfMemberClass( - cast(InstantiatedFrom), - TSK_ExplicitSpecialization); - } else { - assert(isa(Member) && "Only member enums remain"); - EnumDecl *InstantiationEnum = cast(Instantiation); - if (InstantiationEnum->getTemplateSpecializationKind() == - TSK_ImplicitInstantiation) { - InstantiationEnum->setTemplateSpecializationKind( - TSK_ExplicitSpecialization); - InstantiationEnum->setLocation(Member->getLocation()); - } - - cast(Member)->setInstantiationOfMemberEnum( + } else if (auto *MemberClass = dyn_cast(Member)) { + MemberClass->setInstantiationOfMemberClass( + cast(InstantiatedFrom), TSK_ExplicitSpecialization); + } else if (auto *MemberEnum = dyn_cast(Member)) { + MemberEnum->setInstantiationOfMemberEnum( cast(InstantiatedFrom), TSK_ExplicitSpecialization); + } else { + llvm_unreachable("unknown member specialization kind"); } // Save the caller the trouble of having to figure out which declaration @@ -8107,6 +8164,43 @@ Sema::CheckMemberSpecialization(NamedDecl *Member, LookupResult &Previous) { return false; } +/// Complete the explicit specialization of a member of a class template by +/// updating the instantiated member to be marked as an explicit specialization. +/// +/// \param OrigD The member declaration instantiated from the template. +/// \param Loc The location of the explicit specialization of the member. +template +static void completeMemberSpecializationImpl(Sema &S, DeclT *OrigD, + SourceLocation Loc) { + if (OrigD->getTemplateSpecializationKind() != TSK_ImplicitInstantiation) + return; + + // FIXME: Inform AST mutation listeners of this AST mutation. + // FIXME: If there are multiple in-class declarations of the member (from + // multiple modules, or a declaration and later definition of a member type), + // should we update all of them? + OrigD->setTemplateSpecializationKind(TSK_ExplicitSpecialization); + OrigD->setLocation(Loc); +} + +void Sema::CompleteMemberSpecialization(NamedDecl *Member, + LookupResult &Previous) { + NamedDecl *Instantiation = cast(Member->getCanonicalDecl()); + if (Instantiation == Member) + return; + + if (auto *Function = dyn_cast(Instantiation)) + completeMemberSpecializationImpl(*this, Function, Member->getLocation()); + else if (auto *Var = dyn_cast(Instantiation)) + completeMemberSpecializationImpl(*this, Var, Member->getLocation()); + else if (auto *Record = dyn_cast(Instantiation)) + completeMemberSpecializationImpl(*this, Record, Member->getLocation()); + else if (auto *Enum = dyn_cast(Instantiation)) + completeMemberSpecializationImpl(*this, Enum, Member->getLocation()); + else + llvm_unreachable("unknown member specialization kind"); +} + /// \brief Check the scope of an explicit instantiation. /// /// \returns true if a serious error occurs, false otherwise. diff --git a/lib/Sema/SemaTemplateInstantiateDecl.cpp b/lib/Sema/SemaTemplateInstantiateDecl.cpp index 9a71a17561c7..03df6fde6c80 100644 --- a/lib/Sema/SemaTemplateInstantiateDecl.cpp +++ b/lib/Sema/SemaTemplateInstantiateDecl.cpp @@ -1849,6 +1849,19 @@ Decl *TemplateDeclInstantiator::VisitFunctionDecl(FunctionDecl *D, } } } + + // Check the template parameter list against the previous declaration. The + // goal here is to pick up default arguments added since the friend was + // declared; we know the template parameter lists match, since otherwise + // we would not have picked this template as the previous declaration. + if (TemplateParams && FunctionTemplate->getPreviousDecl()) { + SemaRef.CheckTemplateParameterList( + TemplateParams, + FunctionTemplate->getPreviousDecl()->getTemplateParameters(), + Function->isThisDeclarationADefinition() + ? Sema::TPC_FriendFunctionTemplateDefinition + : Sema::TPC_FriendFunctionTemplate); + } } if (Function->isLocalExternDecl() && !Function->getPreviousDecl()) @@ -3660,6 +3673,7 @@ TemplateDeclInstantiator::InitFunctionInstantiation(FunctionDecl *New, New->setType(SemaRef.Context.getFunctionType( NewProto->getReturnType(), NewProto->getParamTypes(), EPI)); } else { + Sema::ContextRAII SwitchContext(SemaRef, New); SemaRef.SubstExceptionSpec(New, Proto, TemplateArgs); } } diff --git a/lib/Sema/SemaType.cpp b/lib/Sema/SemaType.cpp index bcc66bbd1c0a..3992179fabae 100644 --- a/lib/Sema/SemaType.cpp +++ b/lib/Sema/SemaType.cpp @@ -2285,8 +2285,9 @@ bool Sema::CheckFunctionReturnType(QualType T, SourceLocation Loc) { // Methods cannot return interface types. All ObjC objects are // passed by reference. if (T->isObjCObjectType()) { - Diag(Loc, diag::err_object_cannot_be_passed_returned_by_value) << 0 << T; - return 0; + Diag(Loc, diag::err_object_cannot_be_passed_returned_by_value) + << 0 << T << FixItHint::CreateInsertion(Loc, "*"); + return true; } return false; diff --git a/lib/Serialization/ASTReader.cpp b/lib/Serialization/ASTReader.cpp index 61b5a822c552..ef8481488302 100644 --- a/lib/Serialization/ASTReader.cpp +++ b/lib/Serialization/ASTReader.cpp @@ -1534,9 +1534,8 @@ MacroInfo *ASTReader::ReadMacroRecord(ModuleFile &F, uint64_t Offset) { return Macro; unsigned NextIndex = 1; // Skip identifier ID. - SubmoduleID SubModID = getGlobalSubmoduleID(F, Record[NextIndex++]); SourceLocation Loc = ReadSourceLocation(F, Record, NextIndex); - MacroInfo *MI = PP.AllocateDeserializedMacroInfo(Loc, SubModID); + MacroInfo *MI = PP.AllocateMacroInfo(Loc); MI->setDefinitionEndLoc(ReadSourceLocation(F, Record, NextIndex)); MI->setIsUsed(Record[NextIndex++]); MI->setUsedForHeaderGuard(Record[NextIndex++]); diff --git a/lib/Serialization/ASTWriter.cpp b/lib/Serialization/ASTWriter.cpp index 8e4b217a44cd..b6c0cb2815fb 100644 --- a/lib/Serialization/ASTWriter.cpp +++ b/lib/Serialization/ASTWriter.cpp @@ -2413,7 +2413,6 @@ void ASTWriter::WritePreprocessor(const Preprocessor &PP, bool IsModule) { } AddIdentifierRef(Name, Record); - Record.push_back(inferSubmoduleIDFromLocation(MI->getDefinitionLoc())); AddSourceLocation(MI->getDefinitionLoc(), Record); AddSourceLocation(MI->getDefinitionEndLoc(), Record); Record.push_back(MI->isUsed()); diff --git a/lib/StaticAnalyzer/Checkers/BasicObjCFoundationChecks.cpp b/lib/StaticAnalyzer/Checkers/BasicObjCFoundationChecks.cpp index 1ea85d60c9e9..371187747f03 100644 --- a/lib/StaticAnalyzer/Checkers/BasicObjCFoundationChecks.cpp +++ b/lib/StaticAnalyzer/Checkers/BasicObjCFoundationChecks.cpp @@ -233,19 +233,16 @@ void NilArgChecker::checkPreObjCMessage(const ObjCMethodCall &msg, if (StringSelectors.empty()) { ASTContext &Ctx = C.getASTContext(); Selector Sels[] = { - getKeywordSelector(Ctx, "caseInsensitiveCompare", nullptr), - getKeywordSelector(Ctx, "compare", nullptr), - getKeywordSelector(Ctx, "compare", "options", nullptr), - getKeywordSelector(Ctx, "compare", "options", "range", nullptr), - getKeywordSelector(Ctx, "compare", "options", "range", "locale", - nullptr), - getKeywordSelector(Ctx, "componentsSeparatedByCharactersInSet", - nullptr), - getKeywordSelector(Ctx, "initWithFormat", - nullptr), - getKeywordSelector(Ctx, "localizedCaseInsensitiveCompare", nullptr), - getKeywordSelector(Ctx, "localizedCompare", nullptr), - getKeywordSelector(Ctx, "localizedStandardCompare", nullptr), + getKeywordSelector(Ctx, "caseInsensitiveCompare"), + getKeywordSelector(Ctx, "compare"), + getKeywordSelector(Ctx, "compare", "options"), + getKeywordSelector(Ctx, "compare", "options", "range"), + getKeywordSelector(Ctx, "compare", "options", "range", "locale"), + getKeywordSelector(Ctx, "componentsSeparatedByCharactersInSet"), + getKeywordSelector(Ctx, "initWithFormat"), + getKeywordSelector(Ctx, "localizedCaseInsensitiveCompare"), + getKeywordSelector(Ctx, "localizedCompare"), + getKeywordSelector(Ctx, "localizedStandardCompare"), }; for (Selector KnownSel : Sels) StringSelectors[KnownSel] = 0; @@ -262,16 +259,15 @@ void NilArgChecker::checkPreObjCMessage(const ObjCMethodCall &msg, if (ArrayWithObjectSel.isNull()) { ASTContext &Ctx = C.getASTContext(); - ArrayWithObjectSel = getKeywordSelector(Ctx, "arrayWithObject", nullptr); - AddObjectSel = getKeywordSelector(Ctx, "addObject", nullptr); + ArrayWithObjectSel = getKeywordSelector(Ctx, "arrayWithObject"); + AddObjectSel = getKeywordSelector(Ctx, "addObject"); InsertObjectAtIndexSel = - getKeywordSelector(Ctx, "insertObject", "atIndex", nullptr); + getKeywordSelector(Ctx, "insertObject", "atIndex"); ReplaceObjectAtIndexWithObjectSel = - getKeywordSelector(Ctx, "replaceObjectAtIndex", "withObject", nullptr); + getKeywordSelector(Ctx, "replaceObjectAtIndex", "withObject"); SetObjectAtIndexedSubscriptSel = - getKeywordSelector(Ctx, "setObject", "atIndexedSubscript", nullptr); - ArrayByAddingObjectSel = - getKeywordSelector(Ctx, "arrayByAddingObject", nullptr); + getKeywordSelector(Ctx, "setObject", "atIndexedSubscript"); + ArrayByAddingObjectSel = getKeywordSelector(Ctx, "arrayByAddingObject"); } if (S == ArrayWithObjectSel || S == AddObjectSel || @@ -292,13 +288,11 @@ void NilArgChecker::checkPreObjCMessage(const ObjCMethodCall &msg, if (DictionaryWithObjectForKeySel.isNull()) { ASTContext &Ctx = C.getASTContext(); DictionaryWithObjectForKeySel = - getKeywordSelector(Ctx, "dictionaryWithObject", "forKey", nullptr); - SetObjectForKeySel = - getKeywordSelector(Ctx, "setObject", "forKey", nullptr); + getKeywordSelector(Ctx, "dictionaryWithObject", "forKey"); + SetObjectForKeySel = getKeywordSelector(Ctx, "setObject", "forKey"); SetObjectForKeyedSubscriptSel = - getKeywordSelector(Ctx, "setObject", "forKeyedSubscript", nullptr); - RemoveObjectForKeySel = - getKeywordSelector(Ctx, "removeObjectForKey", nullptr); + getKeywordSelector(Ctx, "setObject", "forKeyedSubscript"); + RemoveObjectForKeySel = getKeywordSelector(Ctx, "removeObjectForKey"); } if (S == DictionaryWithObjectForKeySel || S == SetObjectForKeySel) { diff --git a/lib/StaticAnalyzer/Checkers/BuiltinFunctionChecker.cpp b/lib/StaticAnalyzer/Checkers/BuiltinFunctionChecker.cpp index 8c2aef21b3ca..48d6cd8a527c 100644 --- a/lib/StaticAnalyzer/Checkers/BuiltinFunctionChecker.cpp +++ b/lib/StaticAnalyzer/Checkers/BuiltinFunctionChecker.cpp @@ -41,6 +41,22 @@ bool BuiltinFunctionChecker::evalCall(const CallExpr *CE, default: return false; + case Builtin::BI__builtin_assume: { + assert (CE->arg_begin() != CE->arg_end()); + SVal ArgSVal = state->getSVal(CE->getArg(0), LCtx); + if (ArgSVal.isUndef()) + return true; // Return true to model purity. + + state = state->assume(ArgSVal.castAs(), true); + // FIXME: do we want to warn here? Not right now. The most reports might + // come from infeasible paths, thus being false positives. + if (!state) + return true; + + C.addTransition(state); + return true; + } + case Builtin::BI__builtin_unpredictable: case Builtin::BI__builtin_expect: case Builtin::BI__builtin_assume_aligned: diff --git a/lib/StaticAnalyzer/Checkers/NoReturnFunctionChecker.cpp b/lib/StaticAnalyzer/Checkers/NoReturnFunctionChecker.cpp index c1deadef4202..8a5c769b6b50 100644 --- a/lib/StaticAnalyzer/Checkers/NoReturnFunctionChecker.cpp +++ b/lib/StaticAnalyzer/Checkers/NoReturnFunctionChecker.cpp @@ -123,14 +123,14 @@ void NoReturnFunctionChecker::checkPostObjCMessage(const ObjCMethodCall &Msg, case 4: lazyInitKeywordSelector(HandleFailureInFunctionSel, C.getASTContext(), "handleFailureInFunction", "file", "lineNumber", - "description", nullptr); + "description"); if (Sel != HandleFailureInFunctionSel) return; break; case 5: lazyInitKeywordSelector(HandleFailureInMethodSel, C.getASTContext(), "handleFailureInMethod", "object", "file", - "lineNumber", "description", nullptr); + "lineNumber", "description"); if (Sel != HandleFailureInMethodSel) return; break; diff --git a/lib/StaticAnalyzer/Checkers/RetainCountChecker.cpp b/lib/StaticAnalyzer/Checkers/RetainCountChecker.cpp index 3f6ae6222ce0..89b1291c4f46 100644 --- a/lib/StaticAnalyzer/Checkers/RetainCountChecker.cpp +++ b/lib/StaticAnalyzer/Checkers/RetainCountChecker.cpp @@ -703,31 +703,30 @@ class RetainSummaryManager { ObjCMethodSummaries[ObjCSummaryKey(ClsII, S)] = Summ; } + template void addMethodSummary(IdentifierInfo *ClsII, ObjCMethodSummariesTy &Summaries, - const RetainSummary *Summ, va_list argp) { - Selector S = getKeywordSelector(Ctx, argp); + const RetainSummary *Summ, Keywords *... Kws) { + Selector S = getKeywordSelector(Ctx, Kws...); Summaries[ObjCSummaryKey(ClsII, S)] = Summ; } - void addInstMethSummary(const char* Cls, const RetainSummary * Summ, ...) { - va_list argp; - va_start(argp, Summ); - addMethodSummary(&Ctx.Idents.get(Cls), ObjCMethodSummaries, Summ, argp); - va_end(argp); + template + void addInstMethSummary(const char *Cls, const RetainSummary *Summ, + Keywords *... Kws) { + addMethodSummary(&Ctx.Idents.get(Cls), ObjCMethodSummaries, Summ, Kws...); } - void addClsMethSummary(const char* Cls, const RetainSummary * Summ, ...) { - va_list argp; - va_start(argp, Summ); - addMethodSummary(&Ctx.Idents.get(Cls),ObjCClassMethodSummaries, Summ, argp); - va_end(argp); + template + void addClsMethSummary(const char *Cls, const RetainSummary *Summ, + Keywords *... Kws) { + addMethodSummary(&Ctx.Idents.get(Cls), ObjCClassMethodSummaries, Summ, + Kws...); } - void addClsMethSummary(IdentifierInfo *II, const RetainSummary * Summ, ...) { - va_list argp; - va_start(argp, Summ); - addMethodSummary(II, ObjCClassMethodSummaries, Summ, argp); - va_end(argp); + template + void addClsMethSummary(IdentifierInfo *II, const RetainSummary *Summ, + Keywords *... Kws) { + addMethodSummary(II, ObjCClassMethodSummaries, Summ, Kws...); } public: @@ -1640,20 +1639,16 @@ void RetainSummaryManager::InitializeMethodSummaries() { addClassMethSummary("NSAutoreleasePool", "new", NoTrackYet); // Create summaries QCRenderer/QCView -createSnapShotImageOfType: - addInstMethSummary("QCRenderer", AllocSumm, - "createSnapshotImageOfType", nullptr); - addInstMethSummary("QCView", AllocSumm, - "createSnapshotImageOfType", nullptr); + addInstMethSummary("QCRenderer", AllocSumm, "createSnapshotImageOfType"); + addInstMethSummary("QCView", AllocSumm, "createSnapshotImageOfType"); // Create summaries for CIContext, 'createCGImage' and // 'createCGLayerWithSize'. These objects are CF objects, and are not // automatically garbage collected. - addInstMethSummary("CIContext", CFAllocSumm, - "createCGImage", "fromRect", nullptr); + addInstMethSummary("CIContext", CFAllocSumm, "createCGImage", "fromRect"); addInstMethSummary("CIContext", CFAllocSumm, "createCGImage", "fromRect", - "format", "colorSpace", nullptr); - addInstMethSummary("CIContext", CFAllocSumm, "createCGLayerWithSize", "info", - nullptr); + "format", "colorSpace"); + addInstMethSummary("CIContext", CFAllocSumm, "createCGLayerWithSize", "info"); } //===----------------------------------------------------------------------===// diff --git a/lib/StaticAnalyzer/Checkers/SelectorExtras.h b/lib/StaticAnalyzer/Checkers/SelectorExtras.h index 41f70d7d5b69..b11d070c629b 100644 --- a/lib/StaticAnalyzer/Checkers/SelectorExtras.h +++ b/lib/StaticAnalyzer/Checkers/SelectorExtras.h @@ -11,48 +11,26 @@ #define LLVM_CLANG_LIB_STATICANALYZER_CHECKERS_SELECTOREXTRAS_H #include "clang/AST/ASTContext.h" -#include namespace clang { namespace ento { -static inline Selector getKeywordSelectorImpl(ASTContext &Ctx, - const char *First, - va_list argp) { - SmallVector II; - II.push_back(&Ctx.Idents.get(First)); - - while (const char *s = va_arg(argp, const char *)) - II.push_back(&Ctx.Idents.get(s)); +template +static inline Selector getKeywordSelector(ASTContext &Ctx, + IdentifierInfos *... IIs) { + static_assert(sizeof...(IdentifierInfos), + "keyword selectors must have at least one argument"); + SmallVector II({&Ctx.Idents.get(IIs)...}); return Ctx.Selectors.getSelector(II.size(), &II[0]); } -static inline Selector getKeywordSelector(ASTContext &Ctx, va_list argp) { - const char *First = va_arg(argp, const char *); - assert(First && "keyword selectors must have at least one argument"); - return getKeywordSelectorImpl(Ctx, First, argp); -} - -LLVM_END_WITH_NULL -static inline Selector getKeywordSelector(ASTContext &Ctx, - const char *First, ...) { - va_list argp; - va_start(argp, First); - Selector result = getKeywordSelectorImpl(Ctx, First, argp); - va_end(argp); - return result; -} - -LLVM_END_WITH_NULL +template static inline void lazyInitKeywordSelector(Selector &Sel, ASTContext &Ctx, - const char *First, ...) { + IdentifierInfos *... IIs) { if (!Sel.isNull()) return; - va_list argp; - va_start(argp, First); - Sel = getKeywordSelectorImpl(Ctx, First, argp); - va_end(argp); + Sel = getKeywordSelector(Ctx, IIs...); } static inline void lazyInitNullarySelector(Selector &Sel, ASTContext &Ctx, diff --git a/lib/StaticAnalyzer/Checkers/StdLibraryFunctionsChecker.cpp b/lib/StaticAnalyzer/Checkers/StdLibraryFunctionsChecker.cpp index 93ad17cffb34..2f9f5d2d9cf8 100644 --- a/lib/StaticAnalyzer/Checkers/StdLibraryFunctionsChecker.cpp +++ b/lib/StaticAnalyzer/Checkers/StdLibraryFunctionsChecker.cpp @@ -440,7 +440,10 @@ StdLibraryFunctionsChecker::findFunctionSummary(const FunctionDecl *FD, BasicValueFactory &BVF = SVB.getBasicValueFactory(); initFunctionSummaries(BVF); - std::string Name = FD->getQualifiedNameAsString(); + IdentifierInfo *II = FD->getIdentifier(); + if (!II) + return None; + StringRef Name = II->getName(); if (Name.empty() || !C.isCLibraryFunction(FD, Name)) return None; diff --git a/lib/StaticAnalyzer/Core/ExprEngineCallAndReturn.cpp b/lib/StaticAnalyzer/Core/ExprEngineCallAndReturn.cpp index 39d88bfda148..caf86b26b66d 100644 --- a/lib/StaticAnalyzer/Core/ExprEngineCallAndReturn.cpp +++ b/lib/StaticAnalyzer/Core/ExprEngineCallAndReturn.cpp @@ -447,6 +447,7 @@ bool ExprEngine::inlineCall(const CallEvent &Call, const Decl *D, Bldr.takeNodes(Pred); NumInlinedCalls++; + Engine.FunctionSummaries->bumpNumTimesInlined(D); // Mark the decl as visited. if (VisitedCallees) @@ -868,8 +869,6 @@ bool ExprEngine::shouldInlineCall(const CallEvent &Call, const Decl *D, || IsRecursive)) return false; - Engine.FunctionSummaries->bumpNumTimesInlined(D); - return true; } diff --git a/lib/Tooling/RefactoringCallbacks.cpp b/lib/Tooling/RefactoringCallbacks.cpp index e900c23e4f64..9fd333ca554e 100644 --- a/lib/Tooling/RefactoringCallbacks.cpp +++ b/lib/Tooling/RefactoringCallbacks.cpp @@ -9,8 +9,13 @@ // // //===----------------------------------------------------------------------===// -#include "clang/Lex/Lexer.h" #include "clang/Tooling/RefactoringCallbacks.h" +#include "clang/ASTMatchers/ASTMatchFinder.h" +#include "clang/Basic/SourceLocation.h" +#include "clang/Lex/Lexer.h" + +using llvm::StringError; +using llvm::make_error; namespace clang { namespace tooling { @@ -20,18 +25,62 @@ tooling::Replacements &RefactoringCallback::getReplacements() { return Replace; } -static Replacement replaceStmtWithText(SourceManager &Sources, - const Stmt &From, - StringRef Text) { - return tooling::Replacement(Sources, CharSourceRange::getTokenRange( - From.getSourceRange()), Text); +ASTMatchRefactorer::ASTMatchRefactorer( + std::map &FileToReplaces) + : FileToReplaces(FileToReplaces) {} + +void ASTMatchRefactorer::addDynamicMatcher( + const ast_matchers::internal::DynTypedMatcher &Matcher, + RefactoringCallback *Callback) { + MatchFinder.addDynamicMatcher(Matcher, Callback); + Callbacks.push_back(Callback); } -static Replacement replaceStmtWithStmt(SourceManager &Sources, - const Stmt &From, + +class RefactoringASTConsumer : public ASTConsumer { +public: + explicit RefactoringASTConsumer(ASTMatchRefactorer &Refactoring) + : Refactoring(Refactoring) {} + + void HandleTranslationUnit(ASTContext &Context) override { + // The ASTMatchRefactorer is re-used between translation units. + // Clear the matchers so that each Replacement is only emitted once. + for (const auto &Callback : Refactoring.Callbacks) { + Callback->getReplacements().clear(); + } + Refactoring.MatchFinder.matchAST(Context); + for (const auto &Callback : Refactoring.Callbacks) { + for (const auto &Replacement : Callback->getReplacements()) { + llvm::Error Err = + Refactoring.FileToReplaces[Replacement.getFilePath()].add( + Replacement); + if (Err) { + llvm::errs() << "Skipping replacement " << Replacement.toString() + << " due to this error:\n" + << toString(std::move(Err)) << "\n"; + } + } + } + } + +private: + ASTMatchRefactorer &Refactoring; +}; + +std::unique_ptr ASTMatchRefactorer::newASTConsumer() { + return llvm::make_unique(*this); +} + +static Replacement replaceStmtWithText(SourceManager &Sources, const Stmt &From, + StringRef Text) { + return tooling::Replacement( + Sources, CharSourceRange::getTokenRange(From.getSourceRange()), Text); +} +static Replacement replaceStmtWithStmt(SourceManager &Sources, const Stmt &From, const Stmt &To) { - return replaceStmtWithText(Sources, From, Lexer::getSourceText( - CharSourceRange::getTokenRange(To.getSourceRange()), - Sources, LangOptions())); + return replaceStmtWithText( + Sources, From, + Lexer::getSourceText(CharSourceRange::getTokenRange(To.getSourceRange()), + Sources, LangOptions())); } ReplaceStmtWithText::ReplaceStmtWithText(StringRef FromId, StringRef ToText) @@ -103,5 +152,90 @@ void ReplaceIfStmtWithItsBody::run( } } +ReplaceNodeWithTemplate::ReplaceNodeWithTemplate( + llvm::StringRef FromId, std::vector Template) + : FromId(FromId), Template(std::move(Template)) {} + +llvm::Expected> +ReplaceNodeWithTemplate::create(StringRef FromId, StringRef ToTemplate) { + std::vector ParsedTemplate; + for (size_t Index = 0; Index < ToTemplate.size();) { + if (ToTemplate[Index] == '$') { + if (ToTemplate.substr(Index, 2) == "$$") { + Index += 2; + ParsedTemplate.push_back( + TemplateElement{TemplateElement::Literal, "$"}); + } else if (ToTemplate.substr(Index, 2) == "${") { + size_t EndOfIdentifier = ToTemplate.find("}", Index); + if (EndOfIdentifier == std::string::npos) { + return make_error( + "Unterminated ${...} in replacement template near " + + ToTemplate.substr(Index), + llvm::inconvertibleErrorCode()); + } + std::string SourceNodeName = + ToTemplate.substr(Index + 2, EndOfIdentifier - Index - 2); + ParsedTemplate.push_back( + TemplateElement{TemplateElement::Identifier, SourceNodeName}); + Index = EndOfIdentifier + 1; + } else { + return make_error( + "Invalid $ in replacement template near " + + ToTemplate.substr(Index), + llvm::inconvertibleErrorCode()); + } + } else { + size_t NextIndex = ToTemplate.find('$', Index + 1); + ParsedTemplate.push_back( + TemplateElement{TemplateElement::Literal, + ToTemplate.substr(Index, NextIndex - Index)}); + Index = NextIndex; + } + } + return std::unique_ptr( + new ReplaceNodeWithTemplate(FromId, std::move(ParsedTemplate))); +} + +void ReplaceNodeWithTemplate::run( + const ast_matchers::MatchFinder::MatchResult &Result) { + const auto &NodeMap = Result.Nodes.getMap(); + + std::string ToText; + for (const auto &Element : Template) { + switch (Element.Type) { + case TemplateElement::Literal: + ToText += Element.Value; + break; + case TemplateElement::Identifier: { + auto NodeIter = NodeMap.find(Element.Value); + if (NodeIter == NodeMap.end()) { + llvm::errs() << "Node " << Element.Value + << " used in replacement template not bound in Matcher \n"; + llvm::report_fatal_error("Unbound node in replacement template."); + } + CharSourceRange Source = + CharSourceRange::getTokenRange(NodeIter->second.getSourceRange()); + ToText += Lexer::getSourceText(Source, *Result.SourceManager, + Result.Context->getLangOpts()); + break; + } + } + } + if (NodeMap.count(FromId) == 0) { + llvm::errs() << "Node to be replaced " << FromId + << " not bound in query.\n"; + llvm::report_fatal_error("FromId node not bound in MatchResult"); + } + auto Replacement = + tooling::Replacement(*Result.SourceManager, &NodeMap.at(FromId), ToText, + Result.Context->getLangOpts()); + llvm::Error Err = Replace.add(Replacement); + if (Err) { + llvm::errs() << "Query and replace failed in " << Replacement.getFilePath() + << "! " << llvm::toString(std::move(Err)) << "\n"; + llvm::report_fatal_error("Replacement failed"); + } +} + } // end namespace tooling } // end namespace clang diff --git a/test/Analysis/builtin-assume.c b/test/Analysis/builtin-assume.c new file mode 100644 index 000000000000..00d651d9e3be --- /dev/null +++ b/test/Analysis/builtin-assume.c @@ -0,0 +1,8 @@ +// RUN: %clang_analyze_cc1 -analyzer-checker=core,debug.ExprInspection -verify %s + +void clang_analyzer_eval(int); + +void f(int i) { + __builtin_assume(i < 10); + clang_analyzer_eval(i < 15); // expected-warning {{TRUE}} +} diff --git a/test/CXX/drs/dr20xx.cpp b/test/CXX/drs/dr20xx.cpp new file mode 100644 index 000000000000..b97a9a46bc85 --- /dev/null +++ b/test/CXX/drs/dr20xx.cpp @@ -0,0 +1,30 @@ +// RUN: %clang_cc1 -std=c++98 -triple x86_64-unknown-unknown %s -verify -fexceptions -fcxx-exceptions -pedantic-errors \ +// RUN: -Wno-variadic-macros -Wno-c11-extensions +// RUN: %clang_cc1 -std=c++11 -triple x86_64-unknown-unknown %s -verify -fexceptions -fcxx-exceptions -pedantic-errors +// RUN: %clang_cc1 -std=c++14 -triple x86_64-unknown-unknown %s -verify -fexceptions -fcxx-exceptions -pedantic-errors +// RUN: %clang_cc1 -std=c++1z -triple x86_64-unknown-unknown %s -verify -fexceptions -fcxx-exceptions -pedantic-errors + +// expected-no-diagnostics + +#if __cplusplus < 201103L +#define static_assert(...) _Static_assert(__VA_ARGS__) +#endif + +namespace dr2094 { // dr2094: 5.0 + struct A { int n; }; + struct B { volatile int n; }; + static_assert(__is_trivially_copyable(volatile int), ""); + static_assert(__is_trivially_copyable(const volatile int), ""); + static_assert(__is_trivially_copyable(const volatile int[]), ""); + static_assert(__is_trivially_copyable(A), ""); + static_assert(__is_trivially_copyable(volatile A), ""); + static_assert(__is_trivially_copyable(const volatile A), ""); + static_assert(__is_trivially_copyable(const volatile A[]), ""); + static_assert(__is_trivially_copyable(B), ""); + + static_assert(__is_trivially_constructible(A, A const&), ""); + static_assert(__is_trivially_constructible(B, B const&), ""); + + static_assert(__is_trivially_assignable(A, const A&), ""); + static_assert(__is_trivially_assignable(B, const B&), ""); +} diff --git a/test/CXX/drs/dr4xx.cpp b/test/CXX/drs/dr4xx.cpp index 3ea226a745f6..a55bb91be558 100644 --- a/test/CXX/drs/dr4xx.cpp +++ b/test/CXX/drs/dr4xx.cpp @@ -1202,16 +1202,15 @@ namespace dr495 { // dr495: 3.5 long n2 = s2; } -namespace dr496 { // dr496: no +namespace dr496 { // dr496: sup dr2094 struct A { int n; }; struct B { volatile int n; }; int check1[ __is_trivially_copyable(const int) ? 1 : -1]; - int check2[!__is_trivially_copyable(volatile int) ? 1 : -1]; + // This checks the dr2094 behavior, not dr496 + int check2[ __is_trivially_copyable(volatile int) ? 1 : -1]; int check3[ __is_trivially_constructible(A, const A&) ? 1 : -1]; - // FIXME: This is wrong. int check4[ __is_trivially_constructible(B, const B&) ? 1 : -1]; int check5[ __is_trivially_assignable(A, const A&) ? 1 : -1]; - // FIXME: This is wrong. int check6[ __is_trivially_assignable(B, const B&) ? 1 : -1]; } diff --git a/test/CodeCompletion/member-access.cpp b/test/CodeCompletion/member-access.cpp index 66872272ee6d..53af121951bb 100644 --- a/test/CodeCompletion/member-access.cpp +++ b/test/CodeCompletion/member-access.cpp @@ -66,3 +66,83 @@ struct Bar { // Make sure this also doesn't crash // RUN: %clang_cc1 -fsyntax-only -code-completion-at=%s:47:14 %s + + +template +class BaseTemplate { +public: + T baseTemplateFunction(); + + T baseTemplateField; +}; + +template +class TemplateClass: public Base1 , public BaseTemplate { +public: + T function() { } + T field; + + void overload1(const T &); + void overload1(const S &); +}; + +template +void completeDependentMembers(TemplateClass &object, + TemplateClass *object2) { + object.field; + object2->field; +// CHECK-CC2: baseTemplateField : [#T#][#BaseTemplate::#]baseTemplateField +// CHECK-CC2: baseTemplateFunction : [#T#][#BaseTemplate::#]baseTemplateFunction() +// CHECK-CC2: field : [#T#]field +// CHECK-CC2: function : [#T#]function() +// CHECK-CC2: member1 : [#int#][#Base1::#]member1 +// CHECK-CC2: member2 : [#float#][#Base1::#]member2 +// CHECK-CC2: overload1 : [#void#]overload1(<#const T &#>) +// CHECK-CC2: overload1 : [#void#]overload1(<#const S &#>) + +// RUN: %clang_cc1 -fsyntax-only -code-completion-at=%s:92:10 %s -o - | FileCheck -check-prefix=CHECK-CC2 %s +// RUN: %clang_cc1 -fsyntax-only -code-completion-at=%s:93:12 %s -o - | FileCheck -check-prefix=CHECK-CC2 %s +} + + +void completeDependentSpecializedMembers(TemplateClass &object, + TemplateClass *object2) { + object.field; + object2->field; +// CHECK-CC3: baseTemplateField : [#int#][#BaseTemplate::#]baseTemplateField +// CHECK-CC3: baseTemplateFunction : [#int#][#BaseTemplate::#]baseTemplateFunction() +// CHECK-CC3: field : [#int#]field +// CHECK-CC3: function : [#int#]function() +// CHECK-CC3: member1 : [#int#][#Base1::#]member1 +// CHECK-CC3: member2 : [#float#][#Base1::#]member2 +// CHECK-CC3: overload1 : [#void#]overload1(<#const int &#>) +// CHECK-CC3: overload1 : [#void#]overload1(<#const double &#>) + +// RUN: %clang_cc1 -fsyntax-only -code-completion-at=%s:110:10 %s -o - | FileCheck -check-prefix=CHECK-CC3 %s +// RUN: %clang_cc1 -fsyntax-only -code-completion-at=%s:111:12 %s -o - | FileCheck -check-prefix=CHECK-CC3 %s +} + +template +class Template { +public: + BaseTemplate o1; + BaseTemplate o2; + + void function() { + o1.baseTemplateField; +// CHECK-CC4: BaseTemplate : BaseTemplate:: +// CHECK-CC4: baseTemplateField : [#int#]baseTemplateField +// CHECK-CC4: baseTemplateFunction : [#int#]baseTemplateFunction() +// RUN: %clang_cc1 -fsyntax-only -code-completion-at=%s:132:8 %s -o - | FileCheck -check-prefix=CHECK-CC4 %s + o2.baseTemplateField; +// CHECK-CC5: BaseTemplate : BaseTemplate:: +// CHECK-CC5: baseTemplateField : [#T#]baseTemplateField +// CHECK-CC5: baseTemplateFunction : [#T#]baseTemplateFunction() +// RUN: %clang_cc1 -fsyntax-only -code-completion-at=%s:137:8 %s -o - | FileCheck -check-prefix=CHECK-CC5 %s + this->o1; +// CHECK-CC6: [#void#]function() +// CHECK-CC6: o1 : [#BaseTemplate#]o1 +// CHECK-CC6: o2 : [#BaseTemplate#]o2 +// RUN: %clang_cc1 -fsyntax-only -code-completion-at=%s:142:11 %s -o - | FileCheck -check-prefix=CHECK-CC6 %s + } +}; diff --git a/test/CodeGen/asan-globals-gc.cpp b/test/CodeGen/asan-globals-gc.cpp index 6d64f41dda8d..58ce5f067563 100644 --- a/test/CodeGen/asan-globals-gc.cpp +++ b/test/CodeGen/asan-globals-gc.cpp @@ -1,5 +1,16 @@ -// RUN: %clang_cc1 -fsanitize=address -emit-llvm -o - -triple x86_64-windows-msvc %s | FileCheck %s --check-prefix=WITH-GC -// RUN: %clang_cc1 -fsanitize=address -emit-llvm -o - -triple x86_64-windows-msvc -fdata-sections %s | FileCheck %s --check-prefix=WITH-GC +// RUN: %clang_cc1 -fsanitize=address -fsanitize-address-globals-dead-stripping -emit-llvm -o - -triple x86_64-linux %s | FileCheck %s --check-prefix=WITHOUT-GC +// RUN: %clang_cc1 -fsanitize=address -fsanitize-address-globals-dead-stripping -fdata-sections -emit-llvm -o - -triple x86_64-linux %s | FileCheck %s --check-prefix=WITH-GC +// RUN: %clang_cc1 -fsanitize=address -fsanitize-address-globals-dead-stripping -fno-integrated-as -fdata-sections -emit-llvm -o - -triple x86_64-linux %s | FileCheck %s --check-prefix=WITHOUT-GC +// RUN: %clang_cc1 -fsanitize=address -fsanitize-address-globals-dead-stripping -fno-integrated-as -emit-llvm -o - -triple x86_64-linux %s | FileCheck %s --check-prefix=WITHOUT-GC +// RUN: %clang_cc1 -fsanitize=address -fdata-sections -emit-llvm -o - -triple x86_64-linux %s | FileCheck %s --check-prefix=WITHOUT-GC + +// RUN: %clang_cc1 -fsanitize=address -fsanitize-address-globals-dead-stripping -fno-data-sections -emit-llvm -o - -triple x86_64-windows-msvc %s | FileCheck %s --check-prefix=WITH-GC +// RUN: %clang_cc1 -fsanitize=address -fsanitize-address-globals-dead-stripping -fdata-sections -emit-llvm -o - -triple x86_64-windows-msvc %s | FileCheck %s --check-prefix=WITH-GC +// RUN: %clang_cc1 -fsanitize=address -fdata-sections -emit-llvm -o - -triple x86_64-windows-msvc %s | FileCheck %s --check-prefix=WITHOUT-GC + +// RUN: %clang_cc1 -fsanitize=address -fsanitize-address-globals-dead-stripping -fno-data-sections -emit-llvm -o - -triple x86_64-apple-macosx11 %s | FileCheck %s --check-prefix=WITH-GC +// RUN: %clang_cc1 -fsanitize=address -fsanitize-address-globals-dead-stripping -fdata-sections -emit-llvm -o - -triple x86_64-apple-macosx11 %s | FileCheck %s --check-prefix=WITH-GC +// RUN: %clang_cc1 -fsanitize=address -fdata-sections -emit-llvm -o - -triple x86_64-apple-macosx11 %s | FileCheck %s --check-prefix=WITHOUT-GC int global; diff --git a/test/CodeGen/asan-no-globals-no-comdat.cpp b/test/CodeGen/asan-no-globals-no-comdat.cpp new file mode 100644 index 000000000000..4637346f9c41 --- /dev/null +++ b/test/CodeGen/asan-no-globals-no-comdat.cpp @@ -0,0 +1,11 @@ +// Test that on Linux asan constructor is placed in a comdat iff globals-gc is on. +// Even if there are no globals in the module. + +// RUN: %clang_cc1 -fsanitize=address -fsanitize-address-globals-dead-stripping -emit-llvm -o - -triple x86_64-linux %s | FileCheck %s --check-prefix=WITHOUT-GC +// RUN: %clang_cc1 -fsanitize=address -fsanitize-address-globals-dead-stripping -fdata-sections -emit-llvm -o - -triple x86_64-linux %s | FileCheck %s --check-prefix=WITH-GC +// RUN: %clang_cc1 -fsanitize=address -fsanitize-address-globals-dead-stripping -fno-integrated-as -fdata-sections -emit-llvm -o - -triple x86_64-linux %s | FileCheck %s --check-prefix=WITHOUT-GC +// RUN: %clang_cc1 -fsanitize=address -fsanitize-address-globals-dead-stripping -fno-integrated-as -emit-llvm -o - -triple x86_64-linux %s | FileCheck %s --check-prefix=WITHOUT-GC +// RUN: %clang_cc1 -fsanitize=address -fdata-sections -emit-llvm -o - -triple x86_64-linux %s | FileCheck %s --check-prefix=WITHOUT-GC + +// WITH-GC: define internal void @asan.module_ctor() comdat { +// WITHOUT-GC: define internal void @asan.module_ctor() { diff --git a/test/CodeGen/mips-aggregate-arg.c b/test/CodeGen/mips-aggregate-arg.c new file mode 100644 index 000000000000..ccf30df7c22a --- /dev/null +++ b/test/CodeGen/mips-aggregate-arg.c @@ -0,0 +1,38 @@ +// RUN: %clang_cc1 -triple mipsel-unknown-linux-gnu -S -emit-llvm -o - %s | FileCheck -check-prefix=O32 %s +// RUN: %clang_cc1 -triple mips64el-unknown-linux-gnu -S -emit-llvm -o - %s -target-abi n32 | FileCheck -check-prefix=N32-N64 %s +// RUN: %clang_cc1 -triple mips64el-unknown-linux-gnu -S -emit-llvm -o - %s -target-abi n64 | FileCheck -check-prefix=N32-N64 %s + +struct t1 { + char t1[10]; +}; + +struct t2 { + char t2[20]; +}; + +struct t3 { + char t3[65]; +}; + +extern struct t1 g1; +extern struct t2 g2; +extern struct t3 g3; +extern void f1(struct t1); +extern void f2(struct t2); +extern void f3(struct t3); + +void f() { + +// O32: call void @f1(i32 inreg %{{[0-9]+}}, i32 inreg %{{[0-9]+}}, i16 inreg %{{[0-9]+}}) +// O32: call void @f2(%struct.t2* byval align 4 %{{.*}}) +// O32: call void @f3(%struct.t3* byval align 4 %{{.*}}) + +// N32-N64: call void @f1(i64 inreg %{{[0-9]+}}, i16 inreg %{{[0-9]+}}) +// N32-N64: call void @f2(i64 inreg %{{[0-9]+}}, i64 inreg %{{[0-9]+}}, i32 inreg %{{[0-9]+}}) +// N32-N64: call void @f3(%struct.t3* byval align 8 %{{.*}}) + + f1(g1); + f2(g2); + f3(g3); +} + diff --git a/test/CodeGen/sanitize-recover.c b/test/CodeGen/sanitize-recover.c index d714d58c7f26..6358d9d04aa1 100644 --- a/test/CodeGen/sanitize-recover.c +++ b/test/CodeGen/sanitize-recover.c @@ -7,12 +7,12 @@ void test() { extern volatile unsigned x, y, z; - // RECOVER: uadd.with.overflow.i32 - // RECOVER: ubsan_handle_add_overflow( + // RECOVER: uadd.with.overflow.i32{{.*}}, !nosanitize + // RECOVER: ubsan_handle_add_overflow({{.*}}, !nosanitize // RECOVER-NOT: unreachable - // ABORT: uadd.with.overflow.i32 - // ABORT: ubsan_handle_add_overflow_abort( - // ABORT: unreachable + // ABORT: uadd.with.overflow.i32{{.*}}, !nosanitize + // ABORT: ubsan_handle_add_overflow_abort({{.*}}, !nosanitize + // ABORT: unreachable{{.*}}, !nosanitize x = y + z; } diff --git a/test/CodeGen/sparcv8-inline-asm.c b/test/CodeGen/sparcv8-inline-asm.c new file mode 100644 index 000000000000..711a2a0afbb0 --- /dev/null +++ b/test/CodeGen/sparcv8-inline-asm.c @@ -0,0 +1,11 @@ +// RUN: %clang_cc1 -triple sparc-unknown-unknown -emit-llvm %s -o - | FileCheck %s + +// CHECK: define float @fabsf(float %a) +// CHECK: %{{.*}} = call float asm sideeffect "fabss $1, $0;", "=e,f"(float %{{.*}}) #1 +float fabsf(float a) { + float res; + __asm __volatile__("fabss %1, %0;" + : /* reg out*/ "=e"(res) + : /* reg in */ "f"(a)); + return res; +} diff --git a/test/CodeGen/thinlto_backend.ll b/test/CodeGen/thinlto_backend.ll index 813bb62c1a29..86f30c0374fc 100644 --- a/test/CodeGen/thinlto_backend.ll +++ b/test/CodeGen/thinlto_backend.ll @@ -12,10 +12,10 @@ ; RUN: %clang -O2 -o %t4.o -x ir %t1.o -c -fthinlto-index=bad.thinlto.bc 2>&1 | FileCheck %s -check-prefix=CHECK-ERROR1 ; CHECK-ERROR1: Error loading index file 'bad.thinlto.bc' -; Ensure we ignore empty index file under -ignore-empty-index-file, and run -; non-ThinLTO compilation which would not import f2 +; Ensure we ignore empty index file, and run non-ThinLTO compilation which +; would not import f2 ; RUN: touch %t4.thinlto.bc -; RUN: %clang -target x86_64-unknown-linux-gnu -O2 -o %t4.o -x ir %t1.o -c -fthinlto-index=%t4.thinlto.bc -mllvm -ignore-empty-index-file +; RUN: %clang -target x86_64-unknown-linux-gnu -O2 -o %t4.o -x ir %t1.o -c -fthinlto-index=%t4.thinlto.bc ; RUN: llvm-nm %t4.o | FileCheck --check-prefix=CHECK-OBJ-IGNORE-EMPTY %s ; CHECK-OBJ-IGNORE-EMPTY: T f1 ; CHECK-OBJ-IGNORE-EMPTY: U f2 diff --git a/test/CodeGen/x86_64-mno-sse.c b/test/CodeGen/x86_64-mno-sse.c new file mode 100644 index 000000000000..43a695ae3cd3 --- /dev/null +++ b/test/CodeGen/x86_64-mno-sse.c @@ -0,0 +1,15 @@ +// RUN: %clang_cc1 -triple x86_64-linux -target-feature -sse -target-feature -sse2 -S -o /dev/null -verify %s +// REQUIRES: x86-registered-target + +double f1(void) { // expected-error {{SSE register return with SSE disabled}} + return 1.4; +} +extern double g; +void f2(void) { // expected-error {{SSE register return with SSE disabled}} + g = f1(); +} +void take_double(double); +void pass_double(void) { + // FIXME: Still asserts. + //take_double(1.5); +} diff --git a/test/CodeGen/xray-customevent.cpp b/test/CodeGen/xray-customevent.cpp new file mode 100644 index 000000000000..359d92df938a --- /dev/null +++ b/test/CodeGen/xray-customevent.cpp @@ -0,0 +1,28 @@ +// RUN: %clang_cc1 -fxray-instrument -x c++ -std=c++11 -triple x86_64-unknown-unknown -emit-llvm -o - %s | FileCheck %s + +// CHECK-LABEL: @_Z16alwaysInstrumentv +[[clang::xray_always_instrument]] void alwaysInstrument() { + static constexpr char kPhase[] = "instrument"; + __xray_customevent(kPhase, 10); + // CHECK: call void @llvm.xray.customevent(i8*{{.*}}, i32 10) +} + +// CHECK-LABEL: @_Z15neverInstrumentv +[[clang::xray_never_instrument]] void neverInstrument() { + static constexpr char kPhase[] = "never"; + __xray_customevent(kPhase, 5); + // CHECK-NOT: call void @llvm.xray.customevent(i8*{{.*}}, i32 5) +} + +// CHECK-LABEL: @_Z21conditionalInstrumenti +[[clang::xray_always_instrument]] void conditionalInstrument(int v) { + static constexpr char kTrue[] = "true"; + static constexpr char kUntrue[] = "untrue"; + if (v % 2) + __xray_customevent(kTrue, 4); + else + __xray_customevent(kUntrue, 6); + + // CHECK: call void @llvm.xray.customevent(i8*{{.*}}, i32 4) + // CHECK: call void @llvm.xray.customevent(i8*{{.*}}, i32 6) +} diff --git a/test/CodeGenCXX/array-default-argument.cpp b/test/CodeGenCXX/array-default-argument.cpp new file mode 100644 index 000000000000..a07e3908392a --- /dev/null +++ b/test/CodeGenCXX/array-default-argument.cpp @@ -0,0 +1,36 @@ +// RUN: %clang_cc1 -emit-llvm -o - %s -triple %itanium_abi_triple | FileCheck %s +// RUN: %clang_cc1 -emit-llvm -o - %s -triple %itanium_abi_triple -std=c++98 -fexceptions -fcxx-exceptions | FileCheck %s --check-prefix=CHECK --check-prefix=CHECK-EH + +struct A { + A(); + ~A(); +}; + +struct B { + B(A = A()); + ~B(); +}; + +void f(); +// CHECK-LABEL: define void @_Z1gv() +void g() { + // CHECK: br label %[[LOOP:.*]] + + // [[LOOP]]: + // CHECK: {{call|invoke}} {{.*}} @_ZN1AC1Ev([[TEMPORARY:.*]]) + // CHECK-EH: unwind label %[[PARTIAL_ARRAY_LPAD:.*]] + // CHECK: {{call|invoke}} {{.*}} @_ZN1BC1E1A({{.*}}, [[TEMPORARY]]) + // CHECK-EH: unwind label %[[A_AND_PARTIAL_ARRAY_LPAD:.*]] + // CHECK: {{call|invoke}} {{.*}} @_ZN1AD1Ev([[TEMPORARY]]) + // CHECK-EH: unwind label %[[PARTIAL_ARRAY_LPAD]] + // CHECK: getelementptr {{.*}}, i{{[0-9]*}} 1 + // CHECK: icmp eq + // CHECK: br i1 {{.*}} label %[[LOOP]] + B b[5]; + + // CHECK: {{call|invoke}} void @_Z1fv() + f(); + + // CHECK-NOT: @_ZN1AD1Ev( + // CHECK: {{call|invoke}} {{.*}} @_ZN1BD1Ev( +} diff --git a/test/CodeGenCXX/linetable-virtual-variadic.cpp b/test/CodeGenCXX/linetable-virtual-variadic.cpp index 6f966416867a..cd746cdfdfe2 100644 --- a/test/CodeGenCXX/linetable-virtual-variadic.cpp +++ b/test/CodeGenCXX/linetable-virtual-variadic.cpp @@ -12,8 +12,10 @@ void Derived::VariadicFunction(...) { } // CHECK: define void @_ZN7Derived16VariadicFunctionEz({{.*}} !dbg ![[SP:[0-9]+]] // CHECK: ret void, !dbg ![[LOC:[0-9]+]] -// CHECK-LABEL: define void @_ZT{{.+}}N7Derived16VariadicFunctionEz( -// CHECK: ret void, !dbg ![[LOC:[0-9]+]] +// CHECK: define void @_ZT{{.+}}N7Derived16VariadicFunctionEz({{.*}} !dbg ![[SP_I:[0-9]+]] +// CHECK: ret void, !dbg ![[LOC_I:[0-9]+]] // // CHECK: ![[SP]] = distinct !DISubprogram(name: "VariadicFunction" // CHECK: ![[LOC]] = !DILocation({{.*}}scope: ![[SP]]) +// CHECK: ![[SP_I]] = distinct !DISubprogram(name: "VariadicFunction" +// CHECK: ![[LOC_I]] = !DILocation({{.*}}scope: ![[SP_I]]) diff --git a/test/CodeGenCXX/vla.cpp b/test/CodeGenCXX/vla.cpp index 4e22bba7d719..957a9f9568b3 100644 --- a/test/CodeGenCXX/vla.cpp +++ b/test/CodeGenCXX/vla.cpp @@ -1,4 +1,4 @@ -// RUN: %clang_cc1 -triple x86_64-apple-darwin %s -emit-llvm -o - | FileCheck %s +// RUN: %clang_cc1 -std=c++11 -triple x86_64-apple-darwin %s -emit-llvm -o - | FileCheck %s template struct S { @@ -54,3 +54,60 @@ void test0(void *array, int n) { // CHECK-NEXT: ret void } + + +void test2(int b) { + // CHECK-LABEL: define void {{.*}}test2{{.*}}(i32 %b) + int varr[b]; + // get the address of %b by checking the first store that stores it + //CHECK: store i32 %b, i32* [[PTR_B:%.*]] + + // get the size of the VLA by getting the first load of the PTR_B + //CHECK: [[VLA_NUM_ELEMENTS_PREZEXT:%.*]] = load i32, i32* [[PTR_B]] + //CHECK-NEXT: [[VLA_NUM_ELEMENTS_PRE:%.*]] = zext i32 [[VLA_NUM_ELEMENTS_PREZEXT]] + + b = 15; + //CHECK: store i32 15, i32* [[PTR_B]] + + // Now get the sizeof, and then divide by the element size + + + //CHECK: [[VLA_SIZEOF:%.*]] = mul nuw i64 4, [[VLA_NUM_ELEMENTS_PRE]] + //CHECK-NEXT: [[VLA_NUM_ELEMENTS_POST:%.*]] = udiv i64 [[VLA_SIZEOF]], 4 + //CHECK-NEXT: [[VLA_END_PTR:%.*]] = getelementptr inbounds i32, i32* {{%.*}}, i64 [[VLA_NUM_ELEMENTS_POST]] + //CHECK-NEXT: store i32* [[VLA_END_PTR]], i32** %__end + for (int d : varr) 0; +} + +void test3(int b, int c) { + // CHECK-LABEL: define void {{.*}}test3{{.*}}(i32 %b, i32 %c) + int varr[b][c]; + // get the address of %b by checking the first store that stores it + //CHECK: store i32 %b, i32* [[PTR_B:%.*]] + //CHECK-NEXT: store i32 %c, i32* [[PTR_C:%.*]] + + // get the size of the VLA by getting the first load of the PTR_B + //CHECK: [[VLA_DIM1_PREZEXT:%.*]] = load i32, i32* [[PTR_B]] + //CHECK-NEXT: [[VLA_DIM1_PRE:%.*]] = zext i32 [[VLA_DIM1_PREZEXT]] + //CHECK: [[VLA_DIM2_PREZEXT:%.*]] = load i32, i32* [[PTR_C]] + //CHECK-NEXT: [[VLA_DIM2_PRE:%.*]] = zext i32 [[VLA_DIM2_PREZEXT]] + + b = 15; + c = 15; + //CHECK: store i32 15, i32* [[PTR_B]] + //CHECK: store i32 15, i32* [[PTR_C]] + // Now get the sizeof, and then divide by the element size + + // multiply the two dimensions, then by the element type and then divide by the sizeof dim2 + //CHECK: [[VLA_DIM1_X_DIM2:%.*]] = mul nuw i64 [[VLA_DIM1_PRE]], [[VLA_DIM2_PRE]] + //CHECK-NEXT: [[VLA_SIZEOF:%.*]] = mul nuw i64 4, [[VLA_DIM1_X_DIM2]] + //CHECK-NEXT: [[VLA_SIZEOF_DIM2:%.*]] = mul nuw i64 4, [[VLA_DIM2_PRE]] + //CHECK-NEXT: [[VLA_NUM_ELEMENTS:%.*]] = udiv i64 [[VLA_SIZEOF]], [[VLA_SIZEOF_DIM2]] + //CHECK-NEXT: [[VLA_END_INDEX:%.*]] = mul nsw i64 [[VLA_NUM_ELEMENTS]], [[VLA_DIM2_PRE]] + //CHECK-NEXT: [[VLA_END_PTR:%.*]] = getelementptr inbounds i32, i32* {{%.*}}, i64 [[VLA_END_INDEX]] + //CHECK-NEXT: store i32* [[VLA_END_PTR]], i32** %__end + + for (auto &d : varr) 0; +} + + diff --git a/test/CodeGenObjC/arc-blocks.m b/test/CodeGenObjC/arc-blocks.m index 69cd7bb297b2..b84d141037e0 100644 --- a/test/CodeGenObjC/arc-blocks.m +++ b/test/CodeGenObjC/arc-blocks.m @@ -752,6 +752,16 @@ void test19(void (^b)(void)) { // CHECK-NEXT: call void @objc_release(i8* [[X]]) // CHECK-NEXT: ret void +// CHECK-UNOPT-LABEL: define void @test20( +// CHECK-UNOPT: [[XADDR:%.*]] = alloca i8* +// CHECK-UNOPT-NEXT: [[BLOCK:%.*]] = alloca <[[BLOCKTY:.*]]> +// CHECK-UNOPT: [[CAPTUREFIELD:%.*]] = getelementptr inbounds <[[BLOCKTY]]>, <[[BLOCKTY]]>* [[BLOCK]], i32 0, i32 5 +// CHECK-UNOPT: [[BLOCKCAPTURED:%.*]] = getelementptr inbounds <[[BLOCKTY]]>, <[[BLOCKTY]]>* [[BLOCK]], i32 0, i32 5 +// CHECK-UNOPT: [[CAPTURED:%.*]] = load i8*, i8** [[XADDR]] +// CHECK-UNOPT: [[RETAINED:%.*]] = call i8* @objc_retain(i8* [[CAPTURED]]) +// CHECK-UNOPT: store i8* [[RETAINED]], i8** [[BLOCKCAPTURED]] +// CHECK-UNOPT: call void @objc_storeStrong(i8** [[CAPTUREFIELD]], i8* null) + // CHECK-LABEL: define internal void @__copy_helper_block // CHECK: [[BLOCKSOURCE:%.*]] = bitcast i8* %{{.*}} to <[[BLOCKTY]]>* // CHECK: [[CAPTUREFIELD:%.*]] = getelementptr inbounds <[[BLOCKTY]]>, <[[BLOCKTY]]>* [[BLOCKSOURCE]], i32 0, i32 5 diff --git a/test/CodeGenObjC/arc-foreach.m b/test/CodeGenObjC/arc-foreach.m index e3b3f1aa5ce4..77cb068187fa 100644 --- a/test/CodeGenObjC/arc-foreach.m +++ b/test/CodeGenObjC/arc-foreach.m @@ -68,7 +68,8 @@ void test0(NSArray *array) { // CHECK-LP64: [[D0:%.*]] = getelementptr inbounds [[BLOCK_T]], [[BLOCK_T]]* [[BLOCK]], i32 0, i32 5 // CHECK-LP64: [[T0:%.*]] = getelementptr inbounds [[BLOCK_T]], [[BLOCK_T]]* [[BLOCK]], i32 0, i32 5 // CHECK-LP64-NEXT: [[T1:%.*]] = load i8*, i8** [[X]] -// CHECK-LP64-NEXT: store i8* [[T1]], i8** [[T0]] +// CHECK-LP64-NEXT: [[T2:%.*]] = call i8* @objc_retain(i8* [[T1]]) +// CHECK-LP64-NEXT: store i8* [[T2]], i8** [[T0]] // CHECK-LP64-NEXT: [[BLOCK1:%.*]] = bitcast [[BLOCK_T]]* [[BLOCK]] // CHECK-LP64-NEXT: call void @use_block(void ()* [[BLOCK1]]) // CHECK-LP64-NEXT: call void @objc_storeStrong(i8** [[D0]], i8* null) @@ -209,7 +210,8 @@ NSArray *array4; // CHECK-LP64: [[T0:%.*]] = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, [[TY]]* }>, <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, [[TY]]* }>* [[BLOCK]], i32 0, i32 5 // CHECK-LP64: [[BC:%.*]] = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, [[TY]]* }>, <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, [[TY]]* }>* [[BLOCK]], i32 0, i32 5 // CHECK-LP64: [[T1:%.*]] = load [[TY]]*, [[TY]]** [[SELF_ADDR]] -// CHECK-LP64: store [[TY]]* [[T1]], [[TY]]** [[BC]], align 8 +// CHECK-LP64: [[T2:%.*]] = bitcast [[TY]]* [[T1]] to i8* +// CHECK-LP64: call i8* @objc_retain(i8* [[T2]]) // CHECK-LP64-OPT-LABEL: define internal void @"\01-[I1 foo2]"( // CHECK-LP64-OPT: [[TY:%.*]]* %self diff --git a/test/CodeGenOpenCL/amdgpu-debug-info-pointer-address-space.cl b/test/CodeGenOpenCL/amdgpu-debug-info-pointer-address-space.cl index c0b30095a679..7baee5eee391 100644 --- a/test/CodeGenOpenCL/amdgpu-debug-info-pointer-address-space.cl +++ b/test/CodeGenOpenCL/amdgpu-debug-info-pointer-address-space.cl @@ -58,16 +58,16 @@ kernel void kernel1( // CHECK-DAG: !DILocalVariable(name: "FuncVar4", scope: !{{[0-9]+}}, file: !{{[0-9]+}}, line: {{[0-9]+}}, type: ![[DWARF_ADDRESS_SPACE_NONE]]) int *FuncVar4 = Tmp1; - // CHECK-DAG: !DILocalVariable(name: "FuncVar5", scope: !{{[0-9]+}}, file: !{{[0-9]+}}, line: {{[0-9]+}}, type: ![[DWARF_ADDRESS_SPACE_NONE]]) - global int *constant FuncVar5 = KernelArg0; - // CHECK-DAG: !DILocalVariable(name: "FuncVar6", scope: !{{[0-9]+}}, file: !{{[0-9]+}}, line: {{[0-9]+}}, type: ![[DWARF_ADDRESS_SPACE_NONE]]) - constant int *constant FuncVar6 = KernelArg1; - // CHECK-DAG: !DILocalVariable(name: "FuncVar7", scope: !{{[0-9]+}}, file: !{{[0-9]+}}, line: {{[0-9]+}}, type: ![[DWARF_ADDRESS_SPACE_LOCAL]]) - local int *constant FuncVar7 = KernelArg2; - // CHECK-DAG: !DILocalVariable(name: "FuncVar8", scope: !{{[0-9]+}}, file: !{{[0-9]+}}, line: {{[0-9]+}}, type: ![[DWARF_ADDRESS_SPACE_PRIVATE]]) - private int *constant FuncVar8 = Tmp0; - // CHECK-DAG: !DILocalVariable(name: "FuncVar9", scope: !{{[0-9]+}}, file: !{{[0-9]+}}, line: {{[0-9]+}}, type: ![[DWARF_ADDRESS_SPACE_NONE]]) - int *constant FuncVar9 = Tmp1; + // CHECK-DAG: distinct !DIGlobalVariable(name: "FuncVar5", scope: !{{[0-9]+}}, file: !{{[0-9]+}}, line: {{[0-9]+}}, type: ![[DWARF_ADDRESS_SPACE_NONE]], isLocal: true, isDefinition: true) + global int *constant FuncVar5 = 0; + // CHECK-DAG: distinct !DIGlobalVariable(name: "FuncVar6", scope: !{{[0-9]+}}, file: !{{[0-9]+}}, line: {{[0-9]+}}, type: ![[DWARF_ADDRESS_SPACE_NONE]], isLocal: true, isDefinition: true) + constant int *constant FuncVar6 = 0; + // CHECK-DAG: distinct !DIGlobalVariable(name: "FuncVar7", scope: !{{[0-9]+}}, file: !{{[0-9]+}}, line: {{[0-9]+}}, type: ![[DWARF_ADDRESS_SPACE_LOCAL]], isLocal: true, isDefinition: true) + local int *constant FuncVar7 = 0; + // CHECK-DAG: distinct !DIGlobalVariable(name: "FuncVar8", scope: !{{[0-9]+}}, file: !{{[0-9]+}}, line: {{[0-9]+}}, type: ![[DWARF_ADDRESS_SPACE_PRIVATE]], isLocal: true, isDefinition: true) + private int *constant FuncVar8 = 0; + // CHECK-DAG: distinct !DIGlobalVariable(name: "FuncVar9", scope: !{{[0-9]+}}, file: !{{[0-9]+}}, line: {{[0-9]+}}, type: ![[DWARF_ADDRESS_SPACE_NONE]], isLocal: true, isDefinition: true) + int *constant FuncVar9 = 0; // CHECK-DAG: distinct !DIGlobalVariable(name: "FuncVar10", scope: !{{[0-9]+}}, file: !{{[0-9]+}}, line: {{[0-9]+}}, type: ![[DWARF_ADDRESS_SPACE_NONE]], isLocal: true, isDefinition: true) global int *local FuncVar10; FuncVar10 = KernelArg0; diff --git a/test/CodeGenOpenCL/amdgpu-debug-info-variable-expression.cl b/test/CodeGenOpenCL/amdgpu-debug-info-variable-expression.cl index a962d3c75aaf..c0a4c21ce760 100644 --- a/test/CodeGenOpenCL/amdgpu-debug-info-variable-expression.cl +++ b/test/CodeGenOpenCL/amdgpu-debug-info-variable-expression.cl @@ -80,21 +80,21 @@ kernel void kernel1( // CHECK-DAG: call void @llvm.dbg.declare(metadata i32 addrspace(4)** {{.*}}, metadata ![[FUNCVAR4]], metadata ![[PRIVATE]]), !dbg !{{[0-9]+}} int *FuncVar4 = Tmp1; - // CHECK-DAG: ![[FUNCVAR5:[0-9]+]] = !DILocalVariable(name: "FuncVar5", scope: !{{[0-9]+}}, file: !{{[0-9]+}}, line: {{[0-9]+}}, type: !{{[0-9]+}}) - // CHECK-DAG: call void @llvm.dbg.declare(metadata i32 addrspace(1)** {{.*}}, metadata ![[FUNCVAR5]], metadata ![[NONE:[0-9]+]]), !dbg !{{[0-9]+}} - global int *constant FuncVar5 = KernelArg0; - // CHECK-DAG: ![[FUNCVAR6:[0-9]+]] = !DILocalVariable(name: "FuncVar6", scope: !{{[0-9]+}}, file: !{{[0-9]+}}, line: {{[0-9]+}}, type: !{{[0-9]+}}) - // CHECK-DAG: call void @llvm.dbg.declare(metadata i32 addrspace(2)** {{.*}}, metadata ![[FUNCVAR6]], metadata ![[NONE]]), !dbg !{{[0-9]+}} - constant int *constant FuncVar6 = KernelArg1; - // CHECK-DAG: ![[FUNCVAR7:[0-9]+]] = !DILocalVariable(name: "FuncVar7", scope: !{{[0-9]+}}, file: !{{[0-9]+}}, line: {{[0-9]+}}, type: !{{[0-9]+}}) - // CHECK-DAG: call void @llvm.dbg.declare(metadata i32 addrspace(3)** {{.*}}, metadata ![[FUNCVAR7]], metadata ![[NONE]]), !dbg !{{[0-9]+}} - local int *constant FuncVar7 = KernelArg2; - // CHECK-DAG: ![[FUNCVAR8:[0-9]+]] = !DILocalVariable(name: "FuncVar8", scope: !{{[0-9]+}}, file: !{{[0-9]+}}, line: {{[0-9]+}}, type: !{{[0-9]+}}) - // CHECK-DAG: call void @llvm.dbg.declare(metadata i32** {{.*}}, metadata ![[FUNCVAR8]], metadata ![[NONE]]), !dbg !{{[0-9]+}} - private int *constant FuncVar8 = Tmp0; - // CHECK-DAG: ![[FUNCVAR9:[0-9]+]] = !DILocalVariable(name: "FuncVar9", scope: !{{[0-9]+}}, file: !{{[0-9]+}}, line: {{[0-9]+}}, type: !{{[0-9]+}}) - // CHECK-DAG: call void @llvm.dbg.declare(metadata i32 addrspace(4)** {{.*}}, metadata ![[FUNCVAR9]], metadata ![[NONE]]), !dbg !{{[0-9]+}} - int *constant FuncVar9 = Tmp1; + // CHECK-DAG: ![[FUNCVAR5:[0-9]+]] = distinct !DIGlobalVariable(name: "FuncVar5", scope: !{{[0-9]+}}, file: !{{[0-9]+}}, line: {{[0-9]+}}, type: !{{[0-9]+}}, isLocal: true, isDefinition: true) + // CHECK-DAG: !DIGlobalVariableExpression(var: ![[FUNCVAR5]]) + global int *constant FuncVar5 = 0; + // CHECK-DAG: ![[FUNCVAR6:[0-9]+]] = distinct !DIGlobalVariable(name: "FuncVar6", scope: !{{[0-9]+}}, file: !{{[0-9]+}}, line: {{[0-9]+}}, type: !{{[0-9]+}}, isLocal: true, isDefinition: true) + // CHECK-DAG: !DIGlobalVariableExpression(var: ![[FUNCVAR6]]) + constant int *constant FuncVar6 = 0; + // CHECK-DAG: ![[FUNCVAR7:[0-9]+]] = distinct !DIGlobalVariable(name: "FuncVar7", scope: !{{[0-9]+}}, file: !{{[0-9]+}}, line: {{[0-9]+}}, type: !{{[0-9]+}}, isLocal: true, isDefinition: true) + // CHECK-DAG: !DIGlobalVariableExpression(var: ![[FUNCVAR7]]) + local int *constant FuncVar7 = 0; + // CHECK-DAG: ![[FUNCVAR8:[0-9]+]] = distinct !DIGlobalVariable(name: "FuncVar8", scope: !{{[0-9]+}}, file: !{{[0-9]+}}, line: {{[0-9]+}}, type: !{{[0-9]+}}, isLocal: true, isDefinition: true) + // CHECK-DAG: !DIGlobalVariableExpression(var: ![[FUNCVAR8]]) + private int *constant FuncVar8 = 0; + // CHECK-DAG: ![[FUNCVAR9:[0-9]+]] = distinct !DIGlobalVariable(name: "FuncVar9", scope: !{{[0-9]+}}, file: !{{[0-9]+}}, line: {{[0-9]+}}, type: !{{[0-9]+}}, isLocal: true, isDefinition: true) + // CHECK-DAG: !DIGlobalVariableExpression(var: ![[FUNCVAR9]]) + int *constant FuncVar9 = 0; // CHECK-DAG: ![[FUNCVAR10:[0-9]+]] = distinct !DIGlobalVariable(name: "FuncVar10", scope: !{{[0-9]+}}, file: !{{[0-9]+}}, line: {{[0-9]+}}, type: !{{[0-9]+}}, isLocal: true, isDefinition: true) // CHECK-DAG: !DIGlobalVariableExpression(var: ![[FUNCVAR10]], expr: ![[LOCAL]]) diff --git a/test/CodeGenOpenCL/constant-addr-space-globals.cl b/test/CodeGenOpenCL/constant-addr-space-globals.cl index 4f0d1ea23e56..7bb970527c26 100644 --- a/test/CodeGenOpenCL/constant-addr-space-globals.cl +++ b/test/CodeGenOpenCL/constant-addr-space-globals.cl @@ -11,10 +11,11 @@ kernel void test(global float *out) { // but create a copy in the original address space (unless a variable itself is // in the constant address space). -void foo(constant const int *p1, const int *p2, const int *p3); +void foo(constant int* p, constant const int *p1, const int *p2, const int *p3); // CHECK: @k.arr1 = internal addrspace(2) constant [3 x i32] [i32 1, i32 2, i32 3] // CHECK: @k.arr2 = private unnamed_addr addrspace(2) constant [3 x i32] [i32 4, i32 5, i32 6] // CHECK: @k.arr3 = private unnamed_addr addrspace(2) constant [3 x i32] [i32 7, i32 8, i32 9] +// CHECK: @k.var1 = internal addrspace(2) constant i32 1 kernel void k(void) { // CHECK-NOT: %arr1 = alloca [3 x i32] constant const int arr1[] = {1, 2, 3}; @@ -23,5 +24,8 @@ kernel void k(void) { // CHECK: %arr3 = alloca [3 x i32] int arr3[] = {7, 8, 9}; - foo(arr1, arr2, arr3); + constant int var1 = 1; + + // CHECK: call spir_func void @foo(i32 addrspace(2)* @k.var1, i32 addrspace(2)* getelementptr inbounds ([3 x i32], [3 x i32] addrspace(2)* @k.arr1, i32 0, i32 0) + foo(&var1, arr1, arr2, arr3); } diff --git a/test/Driver/fsanitize.c b/test/Driver/fsanitize.c index 05e239c74243..41f573aa7316 100644 --- a/test/Driver/fsanitize.c +++ b/test/Driver/fsanitize.c @@ -30,7 +30,7 @@ // RUN: %clang -target x86_64-pc-win32 -fsanitize-coverage=bb %s -### 2>&1 | FileCheck %s --check-prefix=CHECK-COVERAGE-WIN64 // CHECK-COVERAGE-WIN64: "--dependent-lib={{[^"]*}}ubsan_standalone-x86_64.lib" -// RUN: %clang -target x86_64-linux-gnu -fsanitize=integer %s -### 2>&1 | FileCheck %s --check-prefix=CHECK-INTEGER +// RUN: %clang -target x86_64-linux-gnu -fsanitize=integer %s -### 2>&1 | FileCheck %s --check-prefix=CHECK-INTEGER -implicit-check-not="-fsanitize-address-use-after-scope" // CHECK-INTEGER: "-fsanitize={{((signed-integer-overflow|unsigned-integer-overflow|integer-divide-by-zero|shift-base|shift-exponent),?){5}"}} // RUN: %clang -fsanitize=bounds -### -fsyntax-only %s 2>&1 | FileCheck %s --check-prefix=CHECK-BOUNDS @@ -126,6 +126,13 @@ // RUN: %clang -target x86_64-linux-gnu -fsanitize=address %s -### 2>&1 | FileCheck %s --check-prefix=CHECK-ASAN-WITHOUT-USE-AFTER-SCOPE // CHECK-ASAN-WITHOUT-USE-AFTER-SCOPE: -cc1{{.*}}address-use-after-scope +// RUN: %clang -target x86_64-linux-gnu -fsanitize=address -fsanitize-address-globals-dead-stripping %s -### 2>&1 | FileCheck %s --check-prefix=CHECK-ASAN-GLOBALS +// RUN: %clang -target x86_64-linux-gnu -fsanitize=address %s -### 2>&1 | FileCheck %s --check-prefix=CHECK-NO-ASAN-GLOBALS +// RUN: %clang_cl --target=x86_64-windows-msvc -fsanitize=address -fsanitize-address-globals-dead-stripping -### -- %s 2>&1 | FileCheck %s --check-prefix=CHECK-ASAN-GLOBALS +// RUN: %clang_cl --target=x86_64-windows-msvc -fsanitize=address -### -- %s 2>&1 | FileCheck %s --check-prefix=CHECK-ASAN-GLOBALS +// CHECK-ASAN-GLOBALS: -cc1{{.*}}-fsanitize-address-globals-dead-stripping +// CHECK-NO-ASAN-GLOBALS-NOT: -cc1{{.*}}-fsanitize-address-globals-dead-stripping + // RUN: %clang -target x86_64-linux-gnu -fsanitize-memory-track-origins -pie %s -### 2>&1 | FileCheck %s --check-prefix=CHECK-ONLY-TRACK-ORIGINS // CHECK-ONLY-TRACK-ORIGINS: warning: argument unused during compilation: '-fsanitize-memory-track-origins' diff --git a/test/Driver/myriad-toolchain.c b/test/Driver/myriad-toolchain.c index 971e5d3e6961..7bd215bf1a48 100644 --- a/test/Driver/myriad-toolchain.c +++ b/test/Driver/myriad-toolchain.c @@ -54,9 +54,11 @@ // -fno-split-dwarf-inlining is consumed but not passed to moviCompile. // RUN: %clang -target shave-myriad -c -### %s -g -fno-inline-functions \ // RUN: -fno-inline-functions-called-once -Os -Wall -MF dep.d -fno-split-dwarf-inlining \ -// RUN: -ffunction-sections 2>&1 | FileCheck %s -check-prefix=PASSTHRU_OPTIONS +// RUN: -ffunction-sections -Xclang -xclangflag -mllvm -llvm-flag 2>&1 \ +// RUN: | FileCheck %s -check-prefix=PASSTHRU_OPTIONS // PASSTHRU_OPTIONS: "-g" "-fno-inline-functions" "-fno-inline-functions-called-once" // PASSTHRU_OPTIONS: "-Os" "-Wall" "-MF" "dep.d" "-ffunction-sections" +// PASSTHRU_OPTIONS: "-Xclang" "-xclangflag" "-mllvm" "-llvm-flag" // RUN: %clang -target shave-myriad -c %s -o foo.o -### -MD -MF dep.d 2>&1 \ // RUN: | FileCheck %s -check-prefix=MDMF diff --git a/test/Driver/wasm-toolchain.c b/test/Driver/wasm-toolchain.c index d0b029303891..3be60df92672 100644 --- a/test/Driver/wasm-toolchain.c +++ b/test/Driver/wasm-toolchain.c @@ -27,18 +27,18 @@ // RUN: %clang -### -no-canonical-prefixes -target wasm32-unknown-unknown --sysroot=/foo %s 2>&1 | FileCheck -check-prefix=LINK %s // LINK: clang{{.*}}" "-cc1" {{.*}} "-o" "[[temp:[^"]*]]" -// LINK: lld{{.*}}" "-flavor" "ld" "-L/foo/lib32" "crt1.o" "crti.o" "[[temp]]" "-lc" "-lcompiler_rt" "crtn.o" "-o" "a.out" +// LINK: lld{{.*}}" "-flavor" "wasm" "-L/foo/lib32" "crt1.o" "crti.o" "[[temp]]" "-lc" "-lcompiler_rt" "crtn.o" "-o" "a.out" // A basic C link command-line with optimization. WebAssembly is somewhat // special in enabling --gc-sections by default. // RUN: %clang -### -O2 -no-canonical-prefixes -target wasm32-unknown-unknown --sysroot=/foo %s 2>&1 | FileCheck -check-prefix=LINK_OPT %s // LINK_OPT: clang{{.*}}" "-cc1" {{.*}} "-o" "[[temp:[^"]*]]" -// LINK_OPT: lld{{.*}}" "-flavor" "ld" "--gc-sections" "-L/foo/lib32" "crt1.o" "crti.o" "[[temp]]" "-lc" "-lcompiler_rt" "crtn.o" "-o" "a.out" +// LINK_OPT: lld{{.*}}" "-flavor" "wasm" "--gc-sections" "-L/foo/lib32" "crt1.o" "crti.o" "[[temp]]" "-lc" "-lcompiler_rt" "crtn.o" "-o" "a.out" // Ditto, but ensure that a user --no-gc-sections comes after the // default --gc-sections. // RUN: %clang -### -O2 -no-canonical-prefixes -target wasm32-unknown-unknown --sysroot=/foo -Wl,--no-gc-sections %s 2>&1 | FileCheck -check-prefix=NO_GC_SECTIONS %s // NO_GC_SECTIONS: clang{{.*}}" "-cc1" {{.*}} "-o" "[[temp:[^"]*]]" -// NO_GC_SECTIONS: lld{{.*}}" "-flavor" "ld" "--gc-sections" "-L/foo/lib32" "crt1.o" "crti.o" "--no-gc-sections" "[[temp]]" "-lc" "-lcompiler_rt" "crtn.o" "-o" "a.out" +// NO_GC_SECTIONS: lld{{.*}}" "-flavor" "wasm" "--gc-sections" "-L/foo/lib32" "crt1.o" "crti.o" "--no-gc-sections" "[[temp]]" "-lc" "-lcompiler_rt" "crtn.o" "-o" "a.out" diff --git a/test/FixIt/fixit-availability.c b/test/FixIt/fixit-availability.c index fa641b4b98c6..038dee08b13c 100644 --- a/test/FixIt/fixit-availability.c +++ b/test/FixIt/fixit-availability.c @@ -5,6 +5,6 @@ int function(void); void use() { function(); -// CHECK: fix-it:{{.*}}:{[[@LINE-1]]:3-[[@LINE-1]]:3}:"if (__builtin_available(macos 10.12, *)) {\n " +// CHECK: fix-it:{{.*}}:{[[@LINE-1]]:3-[[@LINE-1]]:3}:"if (__builtin_available(macOS 10.12, *)) {\n " // CHECK-NEXT: fix-it:{{.*}}:{[[@LINE-2]]:14-[[@LINE-2]]:14}:"\n } else {\n // Fallback on earlier versions\n }" } diff --git a/test/FixIt/fixit-availability.mm b/test/FixIt/fixit-availability.mm index 6bf8f49ddc0d..d044a73efdb9 100644 --- a/test/FixIt/fixit-availability.mm +++ b/test/FixIt/fixit-availability.mm @@ -7,58 +7,58 @@ void anotherFunction(int function); int use() { function(); -// CHECK: fix-it:{{.*}}:{[[@LINE-1]]:3-[[@LINE-1]]:3}:"if (@available(macos 10.12, *)) {\n " +// CHECK: fix-it:{{.*}}:{[[@LINE-1]]:3-[[@LINE-1]]:3}:"if (@available(macOS 10.12, *)) {\n " // CHECK-NEXT: fix-it:{{.*}}:{[[@LINE-2]]:14-[[@LINE-2]]:14}:"\n } else {\n // Fallback on earlier versions\n }" int y = function(), x = 0; -// CHECK: fix-it:{{.*}}:{[[@LINE-1]]:3-[[@LINE-1]]:3}:"if (@available(macos 10.12, *)) {\n " +// CHECK: fix-it:{{.*}}:{[[@LINE-1]]:3-[[@LINE-1]]:3}:"if (@available(macOS 10.12, *)) {\n " // CHECK-NEXT: fix-it:{{.*}}:{[[@LINE-2]]:29-[[@LINE-2]]:29}:"\n } else {\n // Fallback on earlier versions\n }" x += function(); -// CHECK: fix-it:{{.*}}:{[[@LINE-1]]:3-[[@LINE-1]]:3}:"if (@available(macos 10.12, *)) {\n " +// CHECK: fix-it:{{.*}}:{[[@LINE-1]]:3-[[@LINE-1]]:3}:"if (@available(macOS 10.12, *)) {\n " // CHECK-NEXT: fix-it:{{.*}}:{[[@LINE-2]]:19-[[@LINE-2]]:19}:"\n } else {\n // Fallback on earlier versions\n }" if (1) { x = function(); -// CHECK: fix-it:{{.*}}:{[[@LINE-1]]:5-[[@LINE-1]]:5}:"if (@available(macos 10.12, *)) {\n " +// CHECK: fix-it:{{.*}}:{[[@LINE-1]]:5-[[@LINE-1]]:5}:"if (@available(macOS 10.12, *)) {\n " // CHECK-NEXT: fix-it:{{.*}}:{[[@LINE-2]]:20-[[@LINE-2]]:20}:"\n } else {\n // Fallback on earlier versions\n }" } anotherFunction(function()); -// CHECK: fix-it:{{.*}}:{[[@LINE-1]]:3-[[@LINE-1]]:3}:"if (@available(macos 10.12, *)) {\n " +// CHECK: fix-it:{{.*}}:{[[@LINE-1]]:3-[[@LINE-1]]:3}:"if (@available(macOS 10.12, *)) {\n " // CHECK-NEXT: fix-it:{{.*}}:{[[@LINE-2]]:31-[[@LINE-2]]:31}:"\n } else {\n // Fallback on earlier versions\n }" if (function()) { -// CHECK: fix-it:{{.*}}:{[[@LINE-1]]:3-[[@LINE-1]]:3}:"if (@available(macos 10.12, *)) {\n " +// CHECK: fix-it:{{.*}}:{[[@LINE-1]]:3-[[@LINE-1]]:3}:"if (@available(macOS 10.12, *)) {\n " // CHECK-NEXT: fix-it:{{.*}}:{[[@LINE+1]]:4-[[@LINE+1]]:4}:"\n } else {\n // Fallback on earlier versions\n }" } while (function()) - // CHECK: fix-it:{{.*}}:{[[@LINE-1]]:3-[[@LINE-1]]:3}:"if (@available(macos 10.12, *)) {\n " + // CHECK: fix-it:{{.*}}:{[[@LINE-1]]:3-[[@LINE-1]]:3}:"if (@available(macOS 10.12, *)) {\n " // CHECK-NEXT: fix-it:{{.*}}:{[[@LINE+1]]:6-[[@LINE+1]]:6}:"\n } else {\n // Fallback on earlier versions\n }" ; do function(); -// CHECK: fix-it:{{.*}}:{[[@LINE-1]]:5-[[@LINE-1]]:5}:"if (@available(macos 10.12, *)) {\n " +// CHECK: fix-it:{{.*}}:{[[@LINE-1]]:5-[[@LINE-1]]:5}:"if (@available(macOS 10.12, *)) {\n " // CHECK-NEXT: fix-it:{{.*}}:{[[@LINE-2]]:16-[[@LINE-2]]:16}:"\n } else {\n // Fallback on earlier versions\n }" while (1); for (int i = 0; i < 10; ++i) function(); -// CHECK: fix-it:{{.*}}:{[[@LINE-1]]:5-[[@LINE-1]]:5}:"if (@available(macos 10.12, *)) {\n " +// CHECK: fix-it:{{.*}}:{[[@LINE-1]]:5-[[@LINE-1]]:5}:"if (@available(macOS 10.12, *)) {\n " // CHECK-NEXT: fix-it:{{.*}}:{[[@LINE-2]]:16-[[@LINE-2]]:16}:"\n } else {\n // Fallback on earlier versions\n }" switch (x) { case 0: function(); -// CHECK: fix-it:{{.*}}:{[[@LINE-1]]:5-[[@LINE-1]]:5}:"if (@available(macos 10.12, *)) {\n " +// CHECK: fix-it:{{.*}}:{[[@LINE-1]]:5-[[@LINE-1]]:5}:"if (@available(macOS 10.12, *)) {\n " // CHECK-NEXT: fix-it:{{.*}}:{[[@LINE-2]]:16-[[@LINE-2]]:16}:"\n } else {\n // Fallback on earlier versions\n }" case 2: anotherFunction(1); function(); -// CHECK: fix-it:{{.*}}:{[[@LINE-1]]:5-[[@LINE-1]]:5}:"if (@available(macos 10.12, *)) {\n " +// CHECK: fix-it:{{.*}}:{[[@LINE-1]]:5-[[@LINE-1]]:5}:"if (@available(macOS 10.12, *)) {\n " // CHECK-NEXT: fix-it:{{.*}}:{[[@LINE-2]]:16-[[@LINE-2]]:16}:"\n } else {\n // Fallback on earlier versions\n }" break; default: function(); -// CHECK: fix-it:{{.*}}:{[[@LINE-1]]:5-[[@LINE-1]]:5}:"if (@available(macos 10.12, *)) {\n " +// CHECK: fix-it:{{.*}}:{[[@LINE-1]]:5-[[@LINE-1]]:5}:"if (@available(macOS 10.12, *)) {\n " // CHECK-NEXT: fix-it:{{.*}}:{[[@LINE-2]]:16-[[@LINE-2]]:16}:"\n } else {\n // Fallback on earlier versions\n }" break; } return function(); -// CHECK: fix-it:{{.*}}:{[[@LINE-1]]:3-[[@LINE-1]]:3}:"if (@available(macos 10.12, *)) {\n " +// CHECK: fix-it:{{.*}}:{[[@LINE-1]]:3-[[@LINE-1]]:3}:"if (@available(macOS 10.12, *)) {\n " // CHECK-NEXT: fix-it:{{.*}}:{[[@LINE-2]]:21-[[@LINE-2]]:21}:"\n } else {\n // Fallback on earlier versions\n }" } @@ -72,39 +72,39 @@ int use() { void useInMacros() { MYFUNCTION(); -// CHECK: fix-it:{{.*}}:{[[@LINE-1]]:3-[[@LINE-1]]:3}:"if (@available(macos 10.12, *)) {\n " +// CHECK: fix-it:{{.*}}:{[[@LINE-1]]:3-[[@LINE-1]]:3}:"if (@available(macOS 10.12, *)) {\n " // CHECK-NEXT: fix-it:{{.*}}:{[[@LINE-2]]:16-[[@LINE-2]]:16}:"\n } else {\n // Fallback on earlier versions\n }" MACRO_ARGUMENT_SEMI(function()) -// CHECK: fix-it:{{.*}}:{[[@LINE-1]]:3-[[@LINE-1]]:3}:"if (@available(macos 10.12, *)) {\n " +// CHECK: fix-it:{{.*}}:{[[@LINE-1]]:3-[[@LINE-1]]:3}:"if (@available(macOS 10.12, *)) {\n " // CHECK-NEXT: fix-it:{{.*}}:{[[@LINE-2]]:34-[[@LINE-2]]:34}:"\n } else {\n // Fallback on earlier versions\n }" MACRO_ARGUMENT(function()); -// CHECK: fix-it:{{.*}}:{[[@LINE-1]]:3-[[@LINE-1]]:3}:"if (@available(macos 10.12, *)) {\n " +// CHECK: fix-it:{{.*}}:{[[@LINE-1]]:3-[[@LINE-1]]:3}:"if (@available(macOS 10.12, *)) {\n " // CHECK-NEXT: fix-it:{{.*}}:{[[@LINE-2]]:30-[[@LINE-2]]:30}:"\n } else {\n // Fallback on earlier versions\n }" MACRO_ARGUMENT_2(function()); -// CHECK: fix-it:{{.*}}:{[[@LINE-1]]:3-[[@LINE-1]]:3}:"if (@available(macos 10.12, *)) {\n " +// CHECK: fix-it:{{.*}}:{[[@LINE-1]]:3-[[@LINE-1]]:3}:"if (@available(macOS 10.12, *)) {\n " // CHECK-NEXT: fix-it:{{.*}}:{[[@LINE-2]]:32-[[@LINE-2]]:32}:"\n } else {\n // Fallback on earlier versions\n }" INNER_MACRO -// CHECK: fix-it:{{.*}}:{[[@LINE-1]]:3-[[@LINE-1]]:3}:"if (@available(macos 10.12, *)) {\n " +// CHECK: fix-it:{{.*}}:{[[@LINE-1]]:3-[[@LINE-1]]:3}:"if (@available(macOS 10.12, *)) {\n " // CHECK-NEXT: fix-it:{{.*}}:{[[@LINE-2]]:14-[[@LINE-2]]:14}:"\n } else {\n // Fallback on earlier versions\n }" } void wrapDeclStmtUses() { int x = 0, y = function(); -// CHECK: fix-it:{{.*}}:{[[@LINE-1]]:3-[[@LINE-1]]:3}:"if (@available(macos 10.12, *)) {\n " +// CHECK: fix-it:{{.*}}:{[[@LINE-1]]:3-[[@LINE-1]]:3}:"if (@available(macOS 10.12, *)) {\n " // CHECK-NEXT: fix-it:{{.*}}:{[[@LINE+13]]:22-[[@LINE+13]]:22}:"\n } else {\n // Fallback on earlier versions\n }" { int z = function(); if (z) { } -// CHECK: fix-it:{{.*}}:{[[@LINE-4]]:5-[[@LINE-4]]:5}:"if (@available(macos 10.12, *)) {\n " +// CHECK: fix-it:{{.*}}:{[[@LINE-4]]:5-[[@LINE-4]]:5}:"if (@available(macOS 10.12, *)) {\n " // CHECK-NEXT: fix-it:{{.*}}:{[[@LINE-2]]:6-[[@LINE-2]]:6}:"\n } else {\n // Fallback on earlier versions\n }" } if (y) int z = function(); -// CHECK: fix-it:{{.*}}:{[[@LINE-1]]:5-[[@LINE-1]]:5}:"if (@available(macos 10.12, *)) {\n " +// CHECK: fix-it:{{.*}}:{[[@LINE-1]]:5-[[@LINE-1]]:5}:"if (@available(macOS 10.12, *)) {\n " // CHECK-NEXT: fix-it:{{.*}}:{[[@LINE-2]]:24-[[@LINE-2]]:24}:"\n } else {\n // Fallback on earlier versions\n }" anotherFunction(y); anotherFunction(x); diff --git a/test/Import/conflicting-struct/Inputs/S1.cpp b/test/Import/conflicting-struct/Inputs/S1.cpp new file mode 100644 index 000000000000..a99dba8c7821 --- /dev/null +++ b/test/Import/conflicting-struct/Inputs/S1.cpp @@ -0,0 +1,6 @@ +class T; + +class S { + T *t; + int a; +}; diff --git a/test/Import/conflicting-struct/Inputs/S2.cpp b/test/Import/conflicting-struct/Inputs/S2.cpp new file mode 100644 index 000000000000..de2cb6cd03c7 --- /dev/null +++ b/test/Import/conflicting-struct/Inputs/S2.cpp @@ -0,0 +1,7 @@ +class U { + int b; +}; + +class T { + U u; +}; diff --git a/test/Import/conflicting-struct/test.cpp b/test/Import/conflicting-struct/test.cpp new file mode 100644 index 000000000000..5aad567cd794 --- /dev/null +++ b/test/Import/conflicting-struct/test.cpp @@ -0,0 +1,7 @@ +// RUN: clang-import-test --import %S/Inputs/S1.cpp --import %S/Inputs/S2.cpp -expression %s +void expr() { + S MyS; + T MyT; + MyS.a = 3; + MyT.u.b = 2; +} diff --git a/test/Index/Core/index-dependent-source.cpp b/test/Index/Core/index-dependent-source.cpp new file mode 100644 index 000000000000..8c097b935a37 --- /dev/null +++ b/test/Index/Core/index-dependent-source.cpp @@ -0,0 +1,143 @@ +// RUN: c-index-test core -print-source-symbols -- %s -std=c++14 -target x86_64-apple-macosx10.7 | FileCheck %s + +int invalid; + +class Base { + void baseFunction(); + + int baseField; + + static void staticBaseFunction(); +}; + +template +class BaseTemplate { +public: + T baseTemplateFunction(); + + T baseTemplateField; + + static T baseTemplateVariable; +}; + +template +class TemplateClass: public Base , public BaseTemplate { +public: + ~TemplateClass(); + + T function() { } + + static void staticFunction() { } + + T field; + + static T variable; + + struct Struct { }; + + enum Enum { EnumValue }; + + using TypeAlias = S; + typedef T Typedef; + + void overload1(const T &); + void overload1(const S &); +}; + +template +void indexSimpleDependentDeclarations(const TemplateClass &object) { + // Valid instance members: + object.function(); +// CHECK: [[@LINE-1]]:10 | instance-method/C++ | function | c:@ST>2#T#T@TemplateClass@F@function# | | Ref,Call,RelCall,RelCont | rel: 1 + object.field; +// CHECK: [[@LINE-1]]:10 | field/C++ | field | c:@ST>2#T#T@TemplateClass@FI@field | | Ref,RelCont | rel: 1 + object.baseFunction(); +// CHECK: [[@LINE-1]]:10 | instance-method/C++ | baseFunction | c:@S@Base@F@baseFunction# | __ZN4Base12baseFunctionEv | Ref,Call,RelCall,RelCont | rel: 1 + object.baseField; +// CHECK: [[@LINE-1]]:10 | field/C++ | baseField | c:@S@Base@FI@baseField | | Ref,RelCont | rel: 1 + object.baseTemplateFunction(); +// CHECK: [[@LINE-1]]:10 | instance-method/C++ | baseTemplateFunction | c:@ST>1#T@BaseTemplate@F@baseTemplateFunction# | | Ref,Call,RelCall,RelCont | rel: 1 + object.baseTemplateField; +// CHECK: [[@LINE-1]]:10 | field/C++ | baseTemplateField | c:@ST>1#T@BaseTemplate@FI@baseTemplateField | | Ref,RelCont | rel: 1 + + // Invalid instance members: + object.variable; +// CHECK-NOT: [[@LINE-1]]:10 + object.staticFunction(); +// CHECK-NOT: [[@LINE-1]]:10 + object.Struct; +// CHECK-NOT: [[@LINE-1]]:10 + object.EnumValue; +// CHECK-NOT: [[@LINE-1]]:10 + + // Valid static members: + TemplateClass::staticFunction(); +// CHECK: [[@LINE-1]]:24 | static-method/C++ | staticFunction | c:@ST>2#T#T@TemplateClass@F@staticFunction#S | | Ref,Call,RelCall,RelCont | rel: 1 + TemplateClass::variable; +// CHECK: [[@LINE-1]]:24 | static-property/C++ | variable | c:@ST>2#T#T@TemplateClass@variable | __ZN13TemplateClass8variableE | Ref,RelCont | rel: 1 + TemplateClass::staticBaseFunction(); +// CHECK: [[@LINE-1]]:24 | static-method/C++ | staticBaseFunction | c:@S@Base@F@staticBaseFunction#S | __ZN4Base18staticBaseFunctionEv | Ref,Call,RelCall,RelCont | rel: 1 + TemplateClass::baseTemplateVariable; +// CHECK: [[@LINE-1]]:24 | static-property/C++ | baseTemplateVariable | c:@ST>1#T@BaseTemplate@baseTemplateVariable | __ZN12BaseTemplate20baseTemplateVariableE | Ref,RelCont | rel: 1 + TemplateClass::EnumValue; +// CHECK: [[@LINE-1]]:24 | enumerator/C | EnumValue | c:@ST>2#T#T@TemplateClass@E@Enum@EnumValue | | Ref,RelCont | rel: 1 + TemplateClass::Struct(); +// CHECK: [[@LINE-1]]:24 | struct/C | Struct | c:@ST>2#T#T@TemplateClass@S@Struct | | Ref,Call,RelCall,RelCont | rel: 1 + + // Invalid static members: + TemplateClass::field; +// CHECK-NOT: [[@LINE-1]]:24 + TemplateClass::function(); +// CHECK-NOT: [[@LINE-1]]:24 + + // Valid type names: + typename TemplateClass::Struct Val; +// CHECK: [[@LINE-1]]:33 | struct/C | Struct | c:@ST>2#T#T@TemplateClass@S@Struct | | Ref,RelCont | rel: 1 + typename TemplateClass::Enum EnumVal; +// CHECK: [[@LINE-1]]:33 | enum/C | Enum | c:@ST>2#T#T@TemplateClass@E@Enum | | Ref,RelCont | rel: 1 + typename TemplateClass::TypeAlias Val2; +// CHECK: [[@LINE-1]]:33 | type-alias/C++ | TypeAlias | c:@ST>2#T#T@TemplateClass@TypeAlias | | Ref,RelCont | rel: 1 + typename TemplateClass::Typedef Val3; +// CHECK: [[@LINE-1]]:33 | type-alias/C | Typedef | c:{{.*}}index-dependent-source.cpp@ST>2#T#T@TemplateClass@T@Typedef | | Ref,RelCont | rel: 1 + + // Invalid type names: + typename TemplateClass::field Val4; +// CHECK-NOT: [[@LINE-1]]:33 + typename TemplateClass::staticFunction Val5; +// CHECK-NOT: [[@LINE-1]]:33 + + + object.invalid; +// CHECK-NOT: [[@LINE-1]]:10 + TemplateClass::invalid; +// CHECK-NOT: [[@LINE-1]]:24 +} + +template +void indexDependentOverloads(const TemplateClass &object) { + object.overload1(T()); +// CHECK-NOT: [[@LINE-1]] + object.overload1(S()); +// CHECK-NOT: [[@LINE-1]] + object.overload1(Y()); +// CHECK-NOT: [[@LINE-1]] +} + +template struct UndefinedTemplateClass; + +template +void undefinedTemplateLookup(UndefinedTemplateClass &x) { +// Shouldn't crash! + x.lookup; + typename UndefinedTemplateClass::Type y; +} + +template +struct UserOfUndefinedTemplateClass: UndefinedTemplateClass { }; + +template +void undefinedTemplateLookup2(UserOfUndefinedTemplateClass &x) { +// Shouldn't crash! + x.lookup; + typename UserOfUndefinedTemplateClass::Type y; +} diff --git a/test/Index/Core/index-instantiated-source.cpp b/test/Index/Core/index-instantiated-source.cpp new file mode 100644 index 000000000000..f61795da540c --- /dev/null +++ b/test/Index/Core/index-instantiated-source.cpp @@ -0,0 +1,39 @@ +// RUN: c-index-test core -print-source-symbols -- %s -std=c++14 -target x86_64-apple-macosx10.7 | FileCheck %s +// References to declarations in instantiations should be canonicalized: + +template +class BaseTemplate { +public: + T baseTemplateFunction(); +// CHECK: [[@LINE-1]]:5 | instance-method/C++ | baseTemplateFunction | c:@ST>1#T@BaseTemplate@F@baseTemplateFunction# + + T baseTemplateField; +// CHECK: [[@LINE-1]]:5 | field/C++ | baseTemplateField | c:@ST>1#T@BaseTemplate@FI@baseTemplateField +}; + +template +class TemplateClass: public BaseTemplate { +public: + T function() { return T(); } +// CHECK: [[@LINE-1]]:5 | instance-method/C++ | function | c:@ST>2#T#T@TemplateClass@F@function# + + static void staticFunction() { } +// CHECK: [[@LINE-1]]:15 | static-method/C++ | staticFunction | c:@ST>2#T#T@TemplateClass@F@staticFunction#S + + T field; +// CHECK: [[@LINE-1]]:5 | field/C++ | field | c:@ST>2#T#T@TemplateClass@FI@field +}; + +void canonicalizeInstaniationReferences(TemplateClass &object) { + (void)object.function(); +// CHECK: [[@LINE-1]]:16 | instance-method/C++ | function | c:@ST>2#T#T@TemplateClass@F@function# | + (void)object.field; +// CHECK: [[@LINE-1]]:16 | field/C++ | field | c:@ST>2#T#T@TemplateClass@FI@field | | Ref,RelCont | rel: 1 + (void)object.baseTemplateFunction(); +// CHECK: [[@LINE-1]]:16 | instance-method/C++ | baseTemplateFunction | c:@ST>1#T@BaseTemplate@F@baseTemplateFunction# | + (void)object.baseTemplateField; +// CHECK: [[@LINE-1]]:16 | field/C++ | baseTemplateField | c:@ST>1#T@BaseTemplate@FI@baseTemplateField | | Ref,RelCont | rel: 1 + + TemplateClass::staticFunction(); +// CHECK: [[@LINE-1]]:30 | static-method/C++ | staticFunction | c:@ST>2#T#T@TemplateClass@F@staticFunction#S | : Cls { }; // CHECK-NEXT: [[@LINE-5]]:7 | class(Gen)/C++ | PartialSpecilizationClass | c:@ST>2#T#T@PartialSpecilizationClass | | Ref | rel: 0 // CHECK-NEXT: [[@LINE-6]]:33 | class/C++ | Cls | c:@S@Cls | | Ref | rel: 0 // CHECK-NEXT: [[@LINE-7]]:38 | class/C++ | Cls | c:@S@Cls | | Ref | rel: 0 + +template +void functionSp() { } + +struct Record { + constexpr static int C = 2; +}; + +template<> +void functionSp, Record::C>() { +// CHECK: [[@LINE-1]]:6 | function(Gen,TS)/C++ | functionSp | c:@F@functionSp<#$@S@SpecializationDecl>#$@S@Cls#VI2># | __Z10functionSpI18SpecializationDeclI3ClsELi2EEvv | Def,RelSpecialization | rel: 1 +// CHECK: RelSpecialization | functionSp | c:@FT@>2#T#NIfunctionSp#v# +// CHECK: [[@LINE-3]]:17 | class(Gen)/C++ | SpecializationDecl | c:@ST>1#T@SpecializationDecl | | Ref,RelCont | rel: 1 +// CHECK: [[@LINE-4]]:36 | class/C++ | Cls | c:@S@Cls | | Ref,RelCont | rel: 1 +// CHECK: [[@LINE-5]]:50 | static-property/C++ | C | c:@S@Record@C | __ZN6Record1CE | Ref,RelCont | rel: 1 +// CHECK: [[@LINE-6]]:42 | struct/C++ | Record | c:@S@Record | | Ref,RelCont | rel: 1 +} + +template +class ClassWithCorrectSpecialization { }; + +template<> +class ClassWithCorrectSpecialization, Record::C> { }; +// CHECK: [[@LINE-1]]:38 | class(Gen)/C++ | SpecializationDecl | c:@ST>1#T@SpecializationDecl | | Ref | rel: 0 +// CHECK: [[@LINE-2]]:57 | class/C++ | Cls | c:@S@Cls | | Ref | rel: 0 +// CHECK: [[@LINE-3]]:71 | static-property/C++ | C | c:@S@Record@C | __ZN6Record1CE | Ref,Read | rel: 0 +// CHECK: [[@LINE-4]]:63 | struct/C++ | Record | c:@S@Record | | Ref | rel: 0 + +namespace ns { +// CHECK: [[@LINE-1]]:11 | namespace/C++ | ns | c:@N@ns | | Decl | rel: 0 +namespace inner { +// CHECK: [[@LINE-1]]:11 | namespace/C++ | inner | c:@N@ns@N@inner | | Decl,RelChild | rel: 1 +void func(); + +} +namespace innerAlias = inner; +// CHECK: [[@LINE-1]]:11 | namespace-alias/C++ | innerAlias | c:@N@ns@NA@innerAlias | | Decl,RelChild | rel: 1 +// CHECK: [[@LINE-2]]:24 | namespace/C++ | inner | c:@N@ns@N@inner | | Ref,RelCont | rel: 1 +} + +namespace namespaceAlias = ::ns::innerAlias; +// CHECK: [[@LINE-1]]:11 | namespace-alias/C++ | namespaceAlias | c:@NA@namespaceAlias | | Decl | rel: 0 +// CHECK: [[@LINE-2]]:30 | namespace/C++ | ns | c:@N@ns | | Ref,RelCont | rel: 1 +// CHECK: [[@LINE-3]]:34 | namespace-alias/C++ | innerAlias | c:@N@ns@NA@innerAlias | | Ref,RelCont | rel: 1 + +void ::ns::inner::func() { +// CHECK: [[@LINE-1]]:8 | namespace/C++ | ns | c:@N@ns | | Ref,RelCont | rel: 1 +// CHECK: [[@LINE-2]]:12 | namespace/C++ | inner | c:@N@ns@N@inner | | Ref,RelCont | rel: 1 + ns::innerAlias::func(); +// CHECK: [[@LINE-1]]:3 | namespace/C++ | ns | c:@N@ns | | Ref,RelCont | rel: 1 +// CHECK: [[@LINE-2]]:7 | namespace-alias/C++ | innerAlias | c:@N@ns@NA@innerAlias | | Ref,RelCont | rel: 1 +} diff --git a/test/Index/complete-available.m b/test/Index/complete-available.m new file mode 100644 index 000000000000..8267dbdddc5c --- /dev/null +++ b/test/Index/complete-available.m @@ -0,0 +1,20 @@ +/* The run lines are below, because this test is line- and + column-number sensitive. */ +void atAvailable() { + if (@available(macOS 10.10, *)) { + + } + if (__builtin_available(iOS 8, *)) { + } +} + +// RUN: c-index-test -code-completion-at=%s:4:18 %s | FileCheck -check-prefix=CHECK %s +// RUN: c-index-test -code-completion-at=%s:7:27 %s | FileCheck -check-prefix=CHECK %s +// CHECK: {TypedText iOS} (40) +// CHECK: {TypedText iOSApplicationExtension} (40) +// CHECK: {TypedText macOS} (40) +// CHECK: {TypedText macOSApplicationExtension} (40) +// CHECK: {TypedText tvOS} (40) +// CHECK: {TypedText tvOSApplicationExtension} (40) +// CHECK: {TypedText watchOS} (40) +// CHECK: {TypedText watchOSApplicationExtension} (40) diff --git a/test/Index/get-cursor.m b/test/Index/get-cursor.m index af277d45fdf4..e85d49fb0c3e 100644 --- a/test/Index/get-cursor.m +++ b/test/Index/get-cursor.m @@ -154,6 +154,12 @@ SomeT someVar; typedef MY_TYPE2(SomeT2) { int x; }; SomeT2 someVar2; +#define GEN_DECL(mod_name) __attribute__((external_source_symbol(language="Swift", defined_in=mod_name, generated_declaration))) + +GEN_DECL("some_module") +@interface ExtCls +-(void)method; +@end // RUN: c-index-test -cursor-at=%s:4:28 -cursor-at=%s:5:28 %s | FileCheck -check-prefix=CHECK-PROP %s // CHECK-PROP: ObjCPropertyDecl=foo1:4:26 @@ -226,3 +232,8 @@ SomeT2 someVar2; // CHECK-TRANSPARENT: 147:1 TypeRef=TokenPaste_t:144:9 Extent=[147:1 - 147:13] Spelling=TokenPaste_t ([147:1 - 147:13]) // CHECK-TRANSPARENT: 151:1 TypeRef=SomeT:150:17 (Transparent: struct SomeT) Extent=[151:1 - 151:6] Spelling=SomeT ([151:1 - 151:6]) // CHECK-TRANSPARENT: 155:1 TypeRef=SomeT2:154:18 Extent=[155:1 - 155:7] Spelling=SomeT2 ([155:1 - 155:7]) + +// RUN: c-index-test -cursor-at=%s:160:12 -cursor-at=%s:161:8 %s | FileCheck -check-prefix=CHECK-EXTERNAL %s +// CHECK-EXTERNAL: 160:12 ObjCInterfaceDecl=ExtCls:160:12 (external lang: Swift, defined: some_module, gen: 1) +// CHECK-EXTERNAL: 161:8 ObjCInstanceMethodDecl=method:161:8 (external lang: Swift, defined: some_module, gen: 1) +C \ No newline at end of file diff --git a/test/Misc/warning-flags.c b/test/Misc/warning-flags.c index be02e12e2cbb..5172d3b15a90 100644 --- a/test/Misc/warning-flags.c +++ b/test/Misc/warning-flags.c @@ -6,8 +6,8 @@ This test serves two purposes: (1) It documents all existing warnings that currently have no associated -W flag, and ensures that the list never grows. - If take an existing warning and add a flag, this test will fail. To - fix this test, simply remove that warning from the list below. + If you take an existing warning and add a flag, this test will fail. + To fix this test, simply remove that warning from the list below. (2) It prevents us adding new warnings to Clang that have no -W flag. All new warnings should have -W flags. diff --git a/test/Modules/DebugInfoNamespace.cpp b/test/Modules/DebugInfoNamespace.cpp new file mode 100644 index 000000000000..33add085d8a6 --- /dev/null +++ b/test/Modules/DebugInfoNamespace.cpp @@ -0,0 +1,19 @@ +// RUN: rm -rf %t +// RUN: %clang_cc1 -x objective-c++ -std=c++11 -debug-info-kind=standalone \ +// RUN: -dwarf-ext-refs -fmodules \ +// RUN: -fmodule-format=obj -fimplicit-module-maps \ +// RUN: -triple %itanium_abi_triple -fmodules-cache-path=%t \ +// RUN: %s -I %S/Inputs/DebugInfoNamespace -I %t -emit-llvm -o - \ +// RUN: | FileCheck %s + +#include "A.h" +#include "B.h" +using namespace N; +B b; + +// Verify that the forward decl of B is in module B. +// +// CHECK: !DICompositeType(tag: DW_TAG_structure_type, name: "B", +// CHECK-SAME: scope: ![[N:[0-9]+]] +// CHECK: ![[N]] = !DINamespace(name: "N", scope: ![[B:[0-9]+]]) +// CHECK: ![[B]] = !DIModule(scope: null, name: "B", diff --git a/test/Modules/DebugInfoSubmoduleImport.c b/test/Modules/DebugInfoSubmoduleImport.c index 1b31aada9c6a..b608d300d5fe 100644 --- a/test/Modules/DebugInfoSubmoduleImport.c +++ b/test/Modules/DebugInfoSubmoduleImport.c @@ -2,6 +2,11 @@ // RUN: %clang_cc1 -fmodules -fmodule-format=obj -debug-info-kind=limited -dwarf-ext-refs \ // RUN: -fimplicit-module-maps -x c -fmodules-cache-path=%t -I %S/Inputs \ // RUN: %s -emit-llvm -debugger-tuning=lldb -o - | FileCheck %s +// +// RUN: %clang_cc1 -fmodules -fmodule-format=obj -debug-info-kind=limited -dwarf-ext-refs \ +// RUN: -fimplicit-module-maps -x c -fmodules-cache-path=%t -I %S/Inputs \ +// RUN: -fmodules-local-submodule-visibility \ +// RUN: %s -emit-llvm -debugger-tuning=lldb -o - | FileCheck %s #include "DebugSubmoduleA.h" #include "DebugSubmoduleB.h" diff --git a/test/Modules/Inputs/DebugInfoNamespace/A.h b/test/Modules/Inputs/DebugInfoNamespace/A.h new file mode 100644 index 000000000000..dc5a1cd26ab6 --- /dev/null +++ b/test/Modules/Inputs/DebugInfoNamespace/A.h @@ -0,0 +1,3 @@ +namespace N { + struct A {}; +} diff --git a/test/Modules/Inputs/DebugInfoNamespace/B.h b/test/Modules/Inputs/DebugInfoNamespace/B.h new file mode 100644 index 000000000000..c9033a54d4e5 --- /dev/null +++ b/test/Modules/Inputs/DebugInfoNamespace/B.h @@ -0,0 +1,3 @@ +namespace N { + struct B {}; +} diff --git a/test/Modules/Inputs/DebugInfoNamespace/module.modulemap b/test/Modules/Inputs/DebugInfoNamespace/module.modulemap new file mode 100644 index 000000000000..9300fcf98c64 --- /dev/null +++ b/test/Modules/Inputs/DebugInfoNamespace/module.modulemap @@ -0,0 +1,8 @@ +module A { + header "A.h" + export * +} +module B { + header "B.h" + export * +} diff --git a/test/Modules/Inputs/MainA.framework/Frameworks/Sub.framework/Headers/B.h b/test/Modules/Inputs/MainA.framework/Frameworks/Sub.framework/Headers/B.h new file mode 100644 index 000000000000..761540b09cb3 --- /dev/null +++ b/test/Modules/Inputs/MainA.framework/Frameworks/Sub.framework/Headers/B.h @@ -0,0 +1 @@ +// B.h diff --git a/test/Modules/Inputs/MainA.framework/Frameworks/Sub.framework/Headers/Sub.h b/test/Modules/Inputs/MainA.framework/Frameworks/Sub.framework/Headers/Sub.h new file mode 100644 index 000000000000..fd86e3cf872f --- /dev/null +++ b/test/Modules/Inputs/MainA.framework/Frameworks/Sub.framework/Headers/Sub.h @@ -0,0 +1,2 @@ +// Sub.h +#import "B.h" diff --git a/test/Modules/Inputs/MainA.framework/Frameworks/Sub.framework/PrivateHeaders/BPriv.h b/test/Modules/Inputs/MainA.framework/Frameworks/Sub.framework/PrivateHeaders/BPriv.h new file mode 100644 index 000000000000..4ab49b798c63 --- /dev/null +++ b/test/Modules/Inputs/MainA.framework/Frameworks/Sub.framework/PrivateHeaders/BPriv.h @@ -0,0 +1 @@ +// BPriv.h diff --git a/test/Modules/Inputs/MainA.framework/Frameworks/Sub.framework/PrivateHeaders/SubPriv.h b/test/Modules/Inputs/MainA.framework/Frameworks/Sub.framework/PrivateHeaders/SubPriv.h new file mode 100644 index 000000000000..f6ac6188d65f --- /dev/null +++ b/test/Modules/Inputs/MainA.framework/Frameworks/Sub.framework/PrivateHeaders/SubPriv.h @@ -0,0 +1 @@ +#import "BPriv.h" diff --git a/test/Modules/Inputs/MainA.framework/Headers/A.h b/test/Modules/Inputs/MainA.framework/Headers/A.h new file mode 100644 index 000000000000..975f1f0437bb --- /dev/null +++ b/test/Modules/Inputs/MainA.framework/Headers/A.h @@ -0,0 +1 @@ +// A.h diff --git a/test/Modules/Inputs/MainA.framework/Headers/Main.h b/test/Modules/Inputs/MainA.framework/Headers/Main.h new file mode 100644 index 000000000000..cb8cc00a0c45 --- /dev/null +++ b/test/Modules/Inputs/MainA.framework/Headers/Main.h @@ -0,0 +1,2 @@ +// Main.h +#import "A.h" diff --git a/test/Modules/Inputs/MainA.framework/Modules/module.modulemap b/test/Modules/Inputs/MainA.framework/Modules/module.modulemap new file mode 100644 index 000000000000..4b0b5e955386 --- /dev/null +++ b/test/Modules/Inputs/MainA.framework/Modules/module.modulemap @@ -0,0 +1,12 @@ +framework module MainA { + umbrella header "Main.h" + + module * { export * } + export * + + framework module Sub { + umbrella header "Sub.h" + module * { export * } + export * + } +} diff --git a/test/Modules/Inputs/MainA.framework/Modules/module.private.modulemap b/test/Modules/Inputs/MainA.framework/Modules/module.private.modulemap new file mode 100644 index 000000000000..a8dc5c2be5f0 --- /dev/null +++ b/test/Modules/Inputs/MainA.framework/Modules/module.private.modulemap @@ -0,0 +1,12 @@ +framework module MainA_Private { + umbrella header "MainPriv.h" + + module * { export * } + export * + + explicit framework module Sub { + umbrella header "SubPriv.h" + module * { export * } + export * + } +} diff --git a/test/Modules/Inputs/MainA.framework/PrivateHeaders/APriv.h b/test/Modules/Inputs/MainA.framework/PrivateHeaders/APriv.h new file mode 100644 index 000000000000..6ac683c39c55 --- /dev/null +++ b/test/Modules/Inputs/MainA.framework/PrivateHeaders/APriv.h @@ -0,0 +1 @@ +// APriv.h diff --git a/test/Modules/Inputs/MainA.framework/PrivateHeaders/MainPriv.h b/test/Modules/Inputs/MainA.framework/PrivateHeaders/MainPriv.h new file mode 100644 index 000000000000..68103017ad0b --- /dev/null +++ b/test/Modules/Inputs/MainA.framework/PrivateHeaders/MainPriv.h @@ -0,0 +1 @@ +#import "APriv.h" diff --git a/test/Modules/Inputs/SameHeader/A.h b/test/Modules/Inputs/SameHeader/A.h new file mode 100644 index 000000000000..bebe9c31c28e --- /dev/null +++ b/test/Modules/Inputs/SameHeader/A.h @@ -0,0 +1,3 @@ +#ifndef __A_h__ +#define __A_h__ +#endif diff --git a/test/Modules/Inputs/SameHeader/B.h b/test/Modules/Inputs/SameHeader/B.h new file mode 100644 index 000000000000..c3fe49cd854f --- /dev/null +++ b/test/Modules/Inputs/SameHeader/B.h @@ -0,0 +1,4 @@ +#ifndef __B_h__ +#define __B_h__ +#include "C.h" +#endif diff --git a/test/Modules/Inputs/SameHeader/C.h b/test/Modules/Inputs/SameHeader/C.h new file mode 100644 index 000000000000..33c3316a6517 --- /dev/null +++ b/test/Modules/Inputs/SameHeader/C.h @@ -0,0 +1,12 @@ +#ifndef __C_h__ +#define __C_h__ +int c = 1; + +struct aaa { + int b; +}; + +typedef struct fd_set { + char c; +}; +#endif diff --git a/test/Modules/Inputs/SameHeader/module.modulemap b/test/Modules/Inputs/SameHeader/module.modulemap new file mode 100644 index 000000000000..d0283a71ac97 --- /dev/null +++ b/test/Modules/Inputs/SameHeader/module.modulemap @@ -0,0 +1,11 @@ +module X { + module A { + header "A.h" + export * + } + module B { + header "B.h" + export * + } + export * +} diff --git a/test/Modules/find-privateheaders.m b/test/Modules/find-privateheaders.m index c5e82ac70da2..5720a73f9be8 100644 --- a/test/Modules/find-privateheaders.m +++ b/test/Modules/find-privateheaders.m @@ -1,2 +1,13 @@ -// RUN: %clang_cc1 -fmodules -fsyntax-only -F%S/Inputs %s +// RUN: rm -rf %t.cache +// RUN: %clang_cc1 -fmodules -fsyntax-only -F%S/Inputs -fimplicit-module-maps \ +// RUN: -fmodules-cache-path=%t.cache -Wno-private-module -DBUILD_PUBLIC -verify %s +// RUN: rm -rf %t.cache +// RUN: %clang_cc1 -fmodules -fsyntax-only -F%S/Inputs -fimplicit-module-maps \ +// RUN: -fmodules-cache-path=%t.cache -Wno-private-module -verify %s +//expected-no-diagnostics + +#ifdef BUILD_PUBLIC #import "Main/Main.h" +#else +#import "MainA/MainPriv.h" +#endif diff --git a/test/Modules/odr_hash.cpp b/test/Modules/odr_hash.cpp index 294e925627c6..58814dd6b3fb 100644 --- a/test/Modules/odr_hash.cpp +++ b/test/Modules/odr_hash.cpp @@ -1078,6 +1078,39 @@ S s; #endif } +namespace MultipleTypedefs { +#if defined(FIRST) +typedef int B1; +typedef B1 A1; +struct S1 { + A1 x; +}; +#elif defined(SECOND) +typedef int A1; +struct S1 { + A1 x; +}; +#else +S1 s1; +#endif + +#if defined(FIRST) +struct T2 { int x; }; +typedef T2 B2; +typedef B2 A2; +struct S2 { + T2 x; +}; +#elif defined(SECOND) +struct T2 { int x; }; +typedef T2 A2; +struct S2 { + T2 x; +}; +#else +S2 s2; +#endif +} // Keep macros contained to one file. #ifdef FIRST diff --git a/test/Modules/preprocess-module.cpp b/test/Modules/preprocess-module.cpp index a3b789238388..64af00c471df 100644 --- a/test/Modules/preprocess-module.cpp +++ b/test/Modules/preprocess-module.cpp @@ -19,6 +19,11 @@ // RUN: %clang_cc1 -fmodules -fmodule-name=file -fmodule-file=%t/fwd.pcm -x c++-module-map-cpp-output %t/no-rewrite.ii -emit-module -o %t/no-rewrite.pcm // RUN: %clang_cc1 -fmodules -fmodule-name=file -fmodule-file=%t/fwd.pcm -x c++-module-map-cpp-output %t/rewrite.ii -emit-module -o %t/rewrite.pcm +// Check that we can load the original module map in the same compilation (this +// could happen if we had a redundant -fmodule-map-file= in the original +// build). +// RUN: %clang_cc1 -fmodules -fmodule-name=file -fmodule-file=%t/fwd.pcm -fmodule-map-file=%S/Inputs/preprocess/module.modulemap -x c++-module-map-cpp-output %t/rewrite.ii -emit-module -o /dev/null + // Check the module we built works. // RUN: %clang_cc1 -fmodules -fmodule-file=%t/no-rewrite.pcm %s -verify // RUN: %clang_cc1 -fmodules -fmodule-file=%t/rewrite.pcm %s -verify diff --git a/test/Modules/redefinition-same-header.m b/test/Modules/redefinition-same-header.m new file mode 100644 index 000000000000..f1c6cbbcaa2e --- /dev/null +++ b/test/Modules/redefinition-same-header.m @@ -0,0 +1,20 @@ +// RUN: rm -rf %t.tmp +// RUN: %clang_cc1 -fsyntax-only -I %S/Inputs/SameHeader -fmodules \ +// RUN: -fimplicit-module-maps -fmodules-cache-path=%t.tmp %s -verify + +// expected-error@Inputs/SameHeader/C.h:3 {{redefinition of 'c'}} +// expected-note-re@Inputs/SameHeader/B.h:3 {{'{{.*}}C.h' included multiple times, additional include site in header from module 'X.B'}} +// expected-note@Inputs/SameHeader/module.modulemap:6 {{X.B defined here}} +// expected-note-re@redefinition-same-header.m:20 {{'{{.*}}C.h' included multiple times, additional include site here}} + +// expected-error@Inputs/SameHeader/C.h:5 {{redefinition of 'aaa'}} +// expected-note-re@Inputs/SameHeader/B.h:3 {{'{{.*}}C.h' included multiple times, additional include site in header from module 'X.B'}} +// expected-note@Inputs/SameHeader/module.modulemap:6 {{X.B defined here}} +// expected-note-re@redefinition-same-header.m:20 {{'{{.*}}C.h' included multiple times, additional include site here}} + +// expected-error@Inputs/SameHeader/C.h:9 {{redefinition of 'fd_set'}} +// expected-note-re@Inputs/SameHeader/B.h:3 {{'{{.*}}C.h' included multiple times, additional include site in header from module 'X.B'}} +// expected-note@Inputs/SameHeader/module.modulemap:6 {{X.B defined here}} +// expected-note-re@redefinition-same-header.m:20 {{'{{.*}}C.h' included multiple times, additional include site here}} +#include "A.h" // maps to a modular +#include "C.h" // textual include diff --git a/test/OpenMP/report_default_DSA.cpp b/test/OpenMP/report_default_DSA.cpp new file mode 100644 index 000000000000..d14cd5cbe9a4 --- /dev/null +++ b/test/OpenMP/report_default_DSA.cpp @@ -0,0 +1,18 @@ +// RUN: %clang_cc1 -verify -fopenmp -fopenmp-version=45 %s + +void foo(int x, int n) { + double vec[n]; + for (int iter = 0; iter < x; iter++) { +#pragma omp target teams distribute parallel for map( \ + from \ + : vec [0:n]) default(none) + // expected-error@+1 {{variable 'n' must have explicitly specified data sharing attributes}} + for (int ii = 0; ii < n; ii++) { + // expected-error@+3 {{variable 'iter' must have explicitly specified data sharing attributes}} + // expected-error@+2 {{variable 'vec' must have explicitly specified data sharing attributes}} + // expected-error@+1 {{variable 'x' must have explicitly specified data sharing attributes}} + vec[ii] = iter + ii + x; + } + } +} + diff --git a/test/PCH/cxx-templates.cpp b/test/PCH/cxx-templates.cpp index d50eee0623c5..e241701f50df 100644 --- a/test/PCH/cxx-templates.cpp +++ b/test/PCH/cxx-templates.cpp @@ -108,3 +108,11 @@ namespace cyclic_module_load { template int local_extern::f(); // expected-note {{in instantiation of}} #endif template int local_extern::g(); + +namespace MemberSpecializationLocation { +#ifndef NO_ERRORS + // expected-note@cxx-templates.h:* {{previous}} + template<> float A::n; // expected-error {{redeclaration of 'n' with a different type}} +#endif + int k = A::n; +} diff --git a/test/PCH/cxx-templates.h b/test/PCH/cxx-templates.h index c4a844727691..68b252e7974e 100644 --- a/test/PCH/cxx-templates.h +++ b/test/PCH/cxx-templates.h @@ -358,3 +358,6 @@ namespace rdar15468709c { } } +namespace MemberSpecializationLocation { + template struct A { static int n; }; +} diff --git a/test/Parser/objc-available.m b/test/Parser/objc-available.m index a170721240ce..49bc53930655 100644 --- a/test/Parser/objc-available.m +++ b/test/Parser/objc-available.m @@ -21,6 +21,12 @@ void f() { (void)@available; // expected-error{{expected '('}} } +void prettyPlatformNames() { + (void)@available(iOS 8, tvOS 10, watchOS 3, macOS 10.11, *); + (void)__builtin_available(iOSApplicationExtension 8, tvOSApplicationExtension 10, + watchOSApplicationExtension 3, macOSApplicationExtension 10.11, *); +} + #if __has_builtin(__builtin_available) #error expected // expected-error@-1 {{expected}} diff --git a/test/Preprocessor/predefined-arch-macros.c b/test/Preprocessor/predefined-arch-macros.c index a0eb8cbcca9a..08f4d2573f40 100644 --- a/test/Preprocessor/predefined-arch-macros.c +++ b/test/Preprocessor/predefined-arch-macros.c @@ -1601,6 +1601,7 @@ // CHECK_BDVER1_M32: #define __AES__ 1 // CHECK_BDVER1_M32: #define __AVX__ 1 // CHECK_BDVER1_M32: #define __FMA4__ 1 +// CHECK_BDVER1_M32: #define __LWP__ 1 // CHECK_BDVER1_M32: #define __LZCNT__ 1 // CHECK_BDVER1_M32: #define __MMX__ 1 // CHECK_BDVER1_M32: #define __PCLMUL__ 1 @@ -1630,6 +1631,7 @@ // CHECK_BDVER1_M64: #define __AES__ 1 // CHECK_BDVER1_M64: #define __AVX__ 1 // CHECK_BDVER1_M64: #define __FMA4__ 1 +// CHECK_BDVER1_M64: #define __LWP__ 1 // CHECK_BDVER1_M64: #define __LZCNT__ 1 // CHECK_BDVER1_M64: #define __MMX__ 1 // CHECK_BDVER1_M64: #define __PCLMUL__ 1 @@ -1664,6 +1666,7 @@ // CHECK_BDVER2_M32: #define __F16C__ 1 // CHECK_BDVER2_M32: #define __FMA4__ 1 // CHECK_BDVER2_M32: #define __FMA__ 1 +// CHECK_BDVER2_M32: #define __LWP__ 1 // CHECK_BDVER2_M32: #define __LZCNT__ 1 // CHECK_BDVER2_M32: #define __MMX__ 1 // CHECK_BDVER2_M32: #define __PCLMUL__ 1 @@ -1697,6 +1700,7 @@ // CHECK_BDVER2_M64: #define __F16C__ 1 // CHECK_BDVER2_M64: #define __FMA4__ 1 // CHECK_BDVER2_M64: #define __FMA__ 1 +// CHECK_BDVER2_M64: #define __LWP__ 1 // CHECK_BDVER2_M64: #define __LZCNT__ 1 // CHECK_BDVER2_M64: #define __MMX__ 1 // CHECK_BDVER2_M64: #define __PCLMUL__ 1 @@ -1733,6 +1737,7 @@ // CHECK_BDVER3_M32: #define __FMA4__ 1 // CHECK_BDVER3_M32: #define __FMA__ 1 // CHECK_BDVER3_M32: #define __FSGSBASE__ 1 +// CHECK_BDVER3_M32: #define __LWP__ 1 // CHECK_BDVER3_M32: #define __LZCNT__ 1 // CHECK_BDVER3_M32: #define __MMX__ 1 // CHECK_BDVER3_M32: #define __PCLMUL__ 1 @@ -1768,6 +1773,7 @@ // CHECK_BDVER3_M64: #define __FMA4__ 1 // CHECK_BDVER3_M64: #define __FMA__ 1 // CHECK_BDVER3_M64: #define __FSGSBASE__ 1 +// CHECK_BDVER3_M64: #define __LWP__ 1 // CHECK_BDVER3_M64: #define __LZCNT__ 1 // CHECK_BDVER3_M64: #define __MMX__ 1 // CHECK_BDVER3_M64: #define __PCLMUL__ 1 @@ -1807,6 +1813,7 @@ // CHECK_BDVER4_M32: #define __FMA4__ 1 // CHECK_BDVER4_M32: #define __FMA__ 1 // CHECK_BDVER4_M32: #define __FSGSBASE__ 1 +// CHECK_BDVER4_M32: #define __LWP__ 1 // CHECK_BDVER4_M32: #define __LZCNT__ 1 // CHECK_BDVER4_M32: #define __MMX__ 1 // CHECK_BDVER4_M32: #define __PCLMUL__ 1 @@ -1843,6 +1850,7 @@ // CHECK_BDVER4_M64: #define __FMA4__ 1 // CHECK_BDVER4_M64: #define __FMA__ 1 // CHECK_BDVER4_M64: #define __FSGSBASE__ 1 +// CHECK_BDVER4_M64: #define __LWP__ 1 // CHECK_BDVER4_M64: #define __LZCNT__ 1 // CHECK_BDVER4_M64: #define __MMX__ 1 // CHECK_BDVER4_M64: #define __PCLMUL__ 1 diff --git a/test/Preprocessor/x86_target_features.c b/test/Preprocessor/x86_target_features.c index a201900ba762..ce3835f91f42 100644 --- a/test/Preprocessor/x86_target_features.c +++ b/test/Preprocessor/x86_target_features.c @@ -272,6 +272,14 @@ // AESNOSSE2-NOT: #define __SSE2__ 1 // AESNOSSE2-NOT: #define __SSE3__ 1 +// RUN: %clang -target i386-unknown-unknown -march=pentiumpro -mlwp -x c -E -dM -o - %s | FileCheck -match-full-lines --check-prefix=LWP %s + +// LWP: #define __LWP__ 1 + +// RUN: %clang -target i386-unknown-unknown -march=bdver1 -mno-lwp -x c -E -dM -o - %s | FileCheck -match-full-lines --check-prefix=NOLWP %s + +// NOLWP-NOT: #define __LWP__ 1 + // RUN: %clang -target i386-unknown-unknown -march=pentiumpro -msha -x c -E -dM -o - %s | FileCheck -match-full-lines --check-prefix=SHA %s // SHA: #define __SHA__ 1 diff --git a/test/Sema/overloadable.c b/test/Sema/overloadable.c index f5e17d211910..49d8085651d4 100644 --- a/test/Sema/overloadable.c +++ b/test/Sema/overloadable.c @@ -151,3 +151,18 @@ void dropping_qualifiers_is_incompatible() { foo(ccharbuf); // expected-error{{call to 'foo' is ambiguous}} expected-note@148{{candidate function}} expected-note@149{{candidate function}} foo(vcharbuf); // expected-error{{call to 'foo' is ambiguous}} expected-note@148{{candidate function}} expected-note@149{{candidate function}} } + +// Bug: we used to treat `__typeof__(foo)` as though it was `__typeof__(&foo)` +// if `foo` was overloaded with only one function that could have its address +// taken. +void typeof_function_is_not_a_pointer() { + void not_a_pointer(void *) __attribute__((overloadable)); + void not_a_pointer(char *__attribute__((pass_object_size(1)))) + __attribute__((overloadable)); + + __typeof__(not_a_pointer) *fn; + + void take_fn(void (*)(void *)); + // if take_fn is passed a void (**)(void *), we'll get a warning. + take_fn(fn); +} diff --git a/test/Sema/redefinition-same-header.c b/test/Sema/redefinition-same-header.c new file mode 100644 index 000000000000..be5bd1d71c92 --- /dev/null +++ b/test/Sema/redefinition-same-header.c @@ -0,0 +1,14 @@ +// RUN: rm -rf %t +// RUN: mkdir -p %t +// RUN: echo 'int yyy = 42;' > %t/a.h +// RUN: %clang_cc1 -fsyntax-only %s -I%t -verify + +// expected-error@a.h:1 {{redefinition of 'yyy'}} +// expected-note@a.h:1 {{unguarded header; consider using #ifdef guards or #pragma once}} +// expected-note-re@redefinition-same-header.c:11 {{'{{.*}}a.h' included multiple times, additional include site here}} +// expected-note-re@redefinition-same-header.c:12 {{'{{.*}}a.h' included multiple times, additional include site here}} + +#include "a.h" +#include "a.h" + +int foo() { return yyy; } diff --git a/test/Sema/typo-correction.c b/test/Sema/typo-correction.c index 59f022dfe528..78007015dcae 100644 --- a/test/Sema/typo-correction.c +++ b/test/Sema/typo-correction.c @@ -80,3 +80,10 @@ int h() { g(x, 5 ? z : 0); // expected-error 2 {{use of undeclared identifier}} (x, 5 ? z : 0); // expected-error 2 {{use of undeclared identifier}} } + +__attribute__((overloadable)) void func_overloadable(int); +__attribute__((overloadable)) void func_overloadable(float); + +void overloadable_callexpr(int arg) { + func_overloadable(ar); //expected-error{{use of undeclared identifier}} +} diff --git a/test/Sema/vector-cast.c b/test/Sema/vector-cast.c index ea4acfac6a0b..cf23eb2742c4 100644 --- a/test/Sema/vector-cast.c +++ b/test/Sema/vector-cast.c @@ -53,9 +53,8 @@ void f4() { float2 f2; double d, a, b, c; float64x2_t v = {0.0, 1.0}; - // FIXME: These diagnostics are inaccurate: should complain that 'double' to vector 'float2' involves truncation - f2 += d; // expected-error {{cannot convert between vector values of different size ('float2' (vector of 2 'float' values) and 'double')}} - d += f2; // expected-error {{cannot convert between vector values of different size}} + f2 += d; // expected-error {{cannot convert between scalar type 'double' and vector type 'float2' (vector of 2 'float' values) as implicit conversion would cause truncation}} + d += f2; // expected-error {{assigning to 'double' from incompatible type 'float2' (vector of 2 'float' values)}} a = 3.0 + vget_low_f64(v); b = vget_low_f64(v) + 3.0; c = vget_low_f64(v); diff --git a/test/Sema/vector-gcc-compat.c b/test/Sema/vector-gcc-compat.c new file mode 100644 index 000000000000..9eb0569b25f0 --- /dev/null +++ b/test/Sema/vector-gcc-compat.c @@ -0,0 +1,330 @@ +// RUN: %clang_cc1 %s -verify -fsyntax-only -Weverything -triple x86_64-apple-darwin10 + +// Test the compatibility of clang's vector extensions with gcc's vector +// extensions for C. Notably &&, ||, ?: and ! are not available. +typedef long long v2i64 __attribute__((vector_size(16))); +typedef int v2i32 __attribute__((vector_size(8))); +typedef short v2i16 __attribute__((vector_size(4))); +typedef char v2i8 __attribute__((vector_size(2))); + +typedef unsigned long long v2u64 __attribute__((vector_size(16))); +typedef unsigned int v2u32 __attribute__((vector_size(8))); +typedef unsigned short v2u16 __attribute__((vector_size(4))); +typedef unsigned char v2u8 __attribute__((vector_size(2))); + +typedef float v4f32 __attribute__((vector_size(16))); +typedef double v2f64 __attribute__((vector_size(16))); +typedef double v4f64 __attribute__((vector_size(32))); +typedef int v4i32 __attribute((vector_size(16))); + +void arithmeticTest(void); +void logicTest(void); +void comparisonTest(void); +void floatTestSignedType(char a, short b, int c, long long d); +void floatTestUnsignedType(unsigned char a, unsigned short b, unsigned int c, + unsigned long long d); +void floatTestConstant(void); +void intTestType(char a, short b, int c, long long d); +void intTestTypeUnsigned(unsigned char a, unsigned short b, unsigned int c, + unsigned long long d); +void uintTestType(char a, short b, int c, long long d); +void uintTestTypeUnsigned(unsigned char a, unsigned short b, unsigned int c, + unsigned long long d); +void uintTestConstant(v2u64 v2u64_a, v2u32 v2u32_a, v2u16 v2u16_a, v2u8 v2u8_a); +void intTestConstant(v2i64 v2i64_a, v2i32 v2i32_a, v2i16 v2i16_a, v2i8 v2i8_a); + +void arithmeticTest(void) { + v2i64 v2i64_a = (v2i64){0, 1}; + v2i64 v2i64_r; + + v2i64_r = v2i64_a + 1; + v2i64_r = v2i64_a - 1; + v2i64_r = v2i64_a * 1; + v2i64_r = v2i64_a / 1; + v2i64_r = v2i64_a % 1; + + v2i64_r = 1 + v2i64_a; + v2i64_r = 1 - v2i64_a; + v2i64_r = 1 * v2i64_a; + v2i64_r = 1 / v2i64_a; + v2i64_r = 1 % v2i64_a; + + v2i64_a += 1; + v2i64_a -= 1; + v2i64_a *= 1; + v2i64_a /= 1; + v2i64_a %= 1; +} + +void comparisonTest(void) { + v2i64 v2i64_a = (v2i64){0, 1}; + v2i64 v2i64_r; + + v2i64_r = v2i64_a == 1; + v2i64_r = v2i64_a != 1; + v2i64_r = v2i64_a < 1; + v2i64_r = v2i64_a > 1; + v2i64_r = v2i64_a <= 1; + v2i64_r = v2i64_a >= 1; + + v2i64_r = 1 == v2i64_a; + v2i64_r = 1 != v2i64_a; + v2i64_r = 1 < v2i64_a; + v2i64_r = 1 > v2i64_a; + v2i64_r = 1 <= v2i64_a; + v2i64_r = 1 >= v2i64_a; +} + +void logicTest(void) { + v2i64 v2i64_a = (v2i64){0, 1}; + v2i64 v2i64_b = (v2i64){2, 1}; + v2i64 v2i64_c = (v2i64){3, 1}; + v2i64 v2i64_r; + + v2i64_r = !v2i64_a; // expected-error {{invalid argument type 'v2i64' (vector of 2 'long long' values) to unary expression}} + v2i64_r = ~v2i64_a; + + v2i64_r = v2i64_a ? v2i64_b : v2i64_c; // expected-error {{used type 'v2i64' (vector of 2 'long long' values) where arithmetic or pointer type is required}} + + v2i64_r = v2i64_a & 1; + v2i64_r = v2i64_a | 1; + v2i64_r = v2i64_a ^ 1; + + v2i64_r = 1 & v2i64_a; + v2i64_r = 1 | v2i64_a; + v2i64_r = 1 ^ v2i64_a; + + v2i64_a &= 1; + v2i64_a |= 1; + v2i64_a ^= 1; + + v2i64_r = v2i64_a && 1; // expected-error {{logical expression with vector type 'v2i64' (vector of 2 'long long' values) and non-vector type 'int' is only supported in C++}} + v2i64_r = v2i64_a || 1; // expected-error {{logical expression with vector type 'v2i64' (vector of 2 'long long' values) and non-vector type 'int' is only supported in C++}} + + v2i64_r = v2i64_a && v2i64_a; // expected-error {{logical expression with vector types 'v2i64' (vector of 2 'long long' values) and 'v2i64' is only supported in C++}} + v2i64_r = v2i64_a || v2i64_a; // expected-error {{logical expression with vector types 'v2i64' (vector of 2 'long long' values) and 'v2i64' is only supported in C++}} + + v2i64_r = v2i64_a << 1; + v2i64_r = v2i64_a >> 1; + + v2i64_r = 1 << v2i64_a; + v2i64_r = 1 >> v2i64_a; + + v2i64_a <<= 1; + v2i64_a >>= 1; +} + +// For operations with floating point types, we check that interger constants +// can be respresented, or failing that checking based on the integer types. +void floatTestConstant(void) { + // Test that constants added to floats must be expressible as floating point + // numbers. + v4f32 v4f32_a = {0.4f, 0.4f, 0.4f, 0.4f}; + v4f32_a = v4f32_a + 1; + v4f32_a = v4f32_a + 0xFFFFFF; + v4f32_a = v4f32_a + (-1567563LL); + v4f32_a = v4f32_a + (16777208); + v4f32_a = v4f32_a + (16777219); // expected-error {{cannot convert between scalar type 'int' and vector type 'v4f32' (vector of 4 'float' values) as implicit conversion would cause truncation}} +} + +void floatTestConstantComparison(void); +void doubleTestConstantComparison(void); + +void floatTestConstantComparison(void) { + v4f32 v4f32_a = {0.4f, 0.4f, 0.4f, 0.4f}; + v4i32 v4i32_r; + v4i32_r = v4f32_a > 0.4f; + v4i32_r = v4f32_a >= 0.4f; + v4i32_r = v4f32_a < 0.4f; + v4i32_r = v4f32_a <= 0.4f; + v4i32_r = v4f32_a == 0.4f; // expected-warning {{comparing floating point with == or != is unsafe}} + v4i32_r = v4f32_a != 0.4f; // expected-warning {{comparing floating point with == or != is unsafe}} +} + +void doubleTestConstantComparison(void) { + v2f64 v2f64_a = {0.4, 0.4}; + v2i64 v2i64_r; + v2i64_r = v2f64_a > 0.4; + v2i64_r = v2f64_a >= 0.4; + v2i64_r = v2f64_a < 0.4; + v2i64_r = v2f64_a <= 0.4; + v2i64_r = v2f64_a == 0.4; // expected-warning {{comparing floating point with == or != is unsafe}} + v2i64_r = v2f64_a != 0.4; // expected-warning {{comparing floating point with == or != is unsafe}} +} + +void floatTestUnsignedType(unsigned char a, unsigned short b, unsigned int c, + unsigned long long d) { + v4f32 v4f32_a = {0.4f, 0.4f, 0.4f, 0.4f}; + v4f64 v4f64_b = {0.4, 0.4, 0.4, 0.4}; + + v4f32_a = v4f32_a + a; + v4f32_a = v4f32_a + b; + v4f32_a = v4f32_a + c; // expected-error {{cannot convert between scalar type 'unsigned int' and vector type 'v4f32' (vector of 4 'float' values) as implicit conversion would cause truncation}} + v4f32_a = v4f32_a + d; // expected-error {{cannot convert between scalar type 'unsigned long long' and vector type 'v4f32' (vector of 4 'float' values) as implicit conversion would cause truncation}} + + v4f64_b = v4f64_b + a; + v4f64_b = v4f64_b + b; + v4f64_b = v4f64_b + c; + v4f64_b = v4f64_b + d; // expected-error {{cannot convert between scalar type 'unsigned long long' and vector type 'v4f64' (vector of 4 'double' values) as implicit conversion would cause truncation}} +} + +void floatTestSignedType(char a, short b, int c, long long d) { + v4f32 v4f32_a = {0.4f, 0.4f, 0.4f, 0.4f}; + v4f64 v4f64_b = {0.4, 0.4, 0.4, 0.4}; + + v4f32_a = v4f32_a + a; + v4f32_a = v4f32_a + b; + v4f32_a = v4f32_a + c; // expected-error {{cannot convert between scalar type 'int' and vector type 'v4f32' (vector of 4 'float' values) as implicit conversion would cause truncation}} + v4f32_a = v4f32_a + d; // expected-error {{cannot convert between scalar type 'long long' and vector type 'v4f32' (vector of 4 'float' values) as implicit conversion would cause truncation}} + + v4f64_b = v4f64_b + a; + v4f64_b = v4f64_b + b; + v4f64_b = v4f64_b + c; + v4f64_b = v4f64_b + d; // expected-error {{cannot convert between scalar type 'long long' and vector type 'v4f64' (vector of 4 'double' values) as implicit conversion would cause truncation}} +} + +void intTestType(char a, short b, int c, long long d) { + v2i64 v2i64_a = {1, 2}; + v2i32 v2i32_a = {1, 2}; + v2i16 v2i16_a = {1, 2}; + v2i8 v2i8_a = {1, 2}; + + v2i64_a = v2i64_a + d; + v2i64_a = v2i64_a + c; + v2i64_a = v2i64_a + b; + v2i64_a = v2i64_a + a; + + v2i32_a = v2i32_a + d; // expected-warning {{implicit conversion loses integer precision: 'long long' to 'v2i32' (vector of 2 'int' values)}} + v2i32_a = v2i32_a + c; + v2i32_a = v2i32_a + b; + v2i32_a = v2i32_a + a; + + v2i16_a = v2i16_a + d; // expected-error {{cannot convert between scalar type 'long long' and vector type 'v2i16' (vector of 2 'short' values) as implicit conversion would cause truncation}} + v2i16_a = v2i16_a + c; // expected-warning {{implicit conversion loses integer precision: 'int' to 'v2i16' (vector of 2 'short' values)}} + v2i16_a = v2i16_a + b; + v2i16_a = v2i16_a + a; + + v2i8_a = v2i8_a + d; // expected-error {{cannot convert between scalar type 'long long' and vector type 'v2i8' (vector of 2 'char' values) as implicit conversion would cause truncation}} + v2i8_a = v2i8_a + c; // expected-error {{cannot convert between scalar type 'int' and vector type 'v2i8' (vector of 2 'char' values) as implicit conversion would cause truncation}} + v2i8_a = v2i8_a + b; // expected-warning {{implicit conversion loses integer precision: 'short' to 'v2i8' (vector of 2 'char' values)}} + v2i8_a = v2i8_a + a; +} + +void intTestTypeUnsigned(unsigned char a, unsigned short b, unsigned int c, + unsigned long long d) { + v2i64 v2i64_a = {1, 2}; + v2i32 v2i32_a = {1, 2}; + v2i16 v2i16_a = {1, 2}; + v2i8 v2i8_a = {1, 2}; + + v2i64_a = v2i64_a + d; // expected-error {{cannot convert between scalar type 'unsigned long long' and vector type 'v2i64' (vector of 2 'long long' values) as implicit conversion would cause truncation}} + + v2i64_a = v2i64_a + c; + v2i64_a = v2i64_a + b; + v2i64_a = v2i64_a + a; + + v2i32_a = v2i32_a + d; // expected-warning {{implicit conversion loses integer precision: 'unsigned long long' to 'v2i32' (vector of 2 'int' values)}} + v2i32_a = v2i32_a + c; // expected-error {{cannot convert between scalar type 'unsigned int' and vector type 'v2i32' (vector of 2 'int' values) as implicit conversion would cause truncation}} + v2i32_a = v2i32_a + b; + v2i32_a = v2i32_a + a; + + v2i16_a = v2i16_a + d; // expected-error {{cannot convert between scalar type 'unsigned long long' and vector type 'v2i16' (vector of 2 'short' values) as implicit conversion would cause truncation}} + v2i16_a = v2i16_a + c; // expected-warning {{implicit conversion loses integer precision: 'unsigned int' to 'v2i16' (vector of 2 'short' values)}} + v2i16_a = v2i16_a + b; // expected-error {{cannot convert between scalar type 'unsigned short' and vector type 'v2i16' (vector of 2 'short' values) as implicit conversion would cause truncation}} + v2i16_a = v2i16_a + a; + + v2i8_a = v2i8_a + d; // expected-error {{cannot convert between scalar type 'unsigned long long' and vector type 'v2i8' (vector of 2 'char' values) as implicit conversion would cause truncation}} + v2i8_a = v2i8_a + c; // expected-error {{cannot convert between scalar type 'unsigned int' and vector type 'v2i8' (vector of 2 'char' values) as implicit conversion would cause truncation}} + v2i8_a = v2i8_a + b; // expected-warning {{implicit conversion loses integer precision: 'unsigned short' to 'v2i8' (vector of 2 'char' values)}} + v2i8_a = v2i8_a + a; // expected-error {{cannot convert between scalar type 'unsigned char' and vector type 'v2i8' (vector of 2 'char' values) as implicit conversion would cause truncation}} +} + +void uintTestType(char a, short b, int c, long long d) { + v2u64 v2u64_a = {1, 2}; + v2u32 v2u32_a = {1, 2}; + v2u16 v2u16_a = {1, 2}; + v2u8 v2u8_a = {1, 2}; + + v2u64_a = v2u64_a + d; // expected-warning {{implicit conversion changes signedness: 'long long' to 'v2u64' (vector of 2 'unsigned long long' values)}} + v2u64_a = v2u64_a + c; // expected-warning {{implicit conversion changes signedness: 'int' to 'v2u64' (vector of 2 'unsigned long long' values)}} + v2u64_a = v2u64_a + b; // expected-warning {{implicit conversion changes signedness: 'short' to 'v2u64' (vector of 2 'unsigned long long' values)}} + v2u64_a = v2u64_a + a; // expected-warning {{implicit conversion changes signedness: 'char' to 'v2u64' (vector of 2 'unsigned long long' values)}} + + v2u32_a = v2u32_a + d; // expected-warning {{implicit conversion loses integer precision: 'long long' to 'v2u32' (vector of 2 'unsigned int' values)}} + v2u32_a = v2u32_a + c; // expected-warning {{implicit conversion changes signedness: 'int' to 'v2u32' (vector of 2 'unsigned int' values)}} + v2u32_a = v2u32_a + b; // expected-warning {{implicit conversion changes signedness: 'short' to 'v2u32' (vector of 2 'unsigned int' values)}} + v2u32_a = v2u32_a + a; // expected-warning {{implicit conversion changes signedness: 'char' to 'v2u32' (vector of 2 'unsigned int' values)}} + + v2u16_a = v2u16_a + d; // expected-error {{cannot convert between scalar type 'long long' and vector type 'v2u16' (vector of 2 'unsigned short' values) as implicit conversion would cause truncation}} + v2u16_a = v2u16_a + c; // expected-warning {{implicit conversion loses integer precision: 'int' to 'v2u16' (vector of 2 'unsigned short' values)}} + v2u16_a = v2u16_a + b; // expected-warning {{implicit conversion changes signedness: 'short' to 'v2u16' (vector of 2 'unsigned short' values)}} + v2u16_a = v2u16_a + a; // expected-warning {{implicit conversion changes signedness: 'char' to 'v2u16' (vector of 2 'unsigned short' values)}} + + v2u8_a = v2u8_a + d; // expected-error {{cannot convert between scalar type 'long long' and vector type 'v2u8' (vector of 2 'unsigned char' values) as implicit conversion would cause truncation}} + v2u8_a = v2u8_a + c; // expected-error {{cannot convert between scalar type 'int' and vector type 'v2u8' (vector of 2 'unsigned char' values) as implicit conversion would cause truncation}} + v2u8_a = v2u8_a + b; // expected-warning {{implicit conversion loses integer precision: 'short' to 'v2u8' (vector of 2 'unsigned char' values)}} + v2u8_a = v2u8_a + a; // expected-warning {{implicit conversion changes signedness: 'char' to 'v2u8' (vector of 2 'unsigned char' values)}} +} + +void uintTestTypeUnsigned(unsigned char a, unsigned short b, unsigned int c, + unsigned long long d) { + v2u64 v2u64_a = {1, 2}; + v2u32 v2u32_a = {1, 2}; + v2u16 v2u16_a = {1, 2}; + v2u8 v2u8_a = {1, 2}; + + v2u64_a = v2u64_a + d; + v2u64_a = v2u64_a + c; + v2u64_a = v2u64_a + b; + v2u64_a = v2u64_a + a; + + v2u32_a = v2u32_a + d; // expected-warning {{implicit conversion loses integer precision: 'unsigned long long' to 'v2u32' (vector of 2 'unsigned int' values)}} + v2u32_a = v2u32_a + c; + v2u32_a = v2u32_a + b; + v2u32_a = v2u32_a + a; + + v2u16_a = v2u16_a + d; // expected-error {{cannot convert between scalar type 'unsigned long long' and vector type 'v2u16' (vector of 2 'unsigned short' values) as implicit conversion would cause truncation}} + v2u16_a = v2u16_a + c; // expected-warning {{implicit conversion loses integer precision: 'unsigned int' to 'v2u16' (vector of 2 'unsigned short' values)}} + v2u16_a = v2u16_a + b; + v2u16_a = v2u16_a + a; + + v2u8_a = v2u8_a + d; // expected-error {{cannot convert between scalar type 'unsigned long long' and vector type 'v2u8' (vector of 2 'unsigned char' values) as implicit conversion would cause truncation}} + v2u8_a = v2u8_a + c; // expected-error {{cannot convert between scalar type 'unsigned int' and vector type 'v2u8' (vector of 2 'unsigned char' values) as implicit conversion would cause truncation}} + v2u8_a = v2u8_a + b; // expected-warning {{implicit conversion loses integer precision: 'unsigned short' to 'v2u8' (vector of 2 'unsigned char' values)}} + v2u8_a = v2u8_a + a; +} + +void uintTestConstant(v2u64 v2u64_a, v2u32 v2u32_a, v2u16 v2u16_a, + v2u8 v2u8_a) { + v2u64_a = v2u64_a + 0xFFFFFFFFFFFFFFFF; + v2u32_a = v2u32_a + 0xFFFFFFFF; + v2u16_a = v2u16_a + 0xFFFF; + v2u8_a = v2u8_a + 0xFF; + + v2u32_a = v2u32_a + 0x1FFFFFFFF; // expected-warning {{implicit conversion from 'long' to 'v2u32' (vector of 2 'unsigned int' values) changes value from 8589934591 to 4294967295}} + v2u16_a = v2u16_a + 0x1FFFF; // expected-warning {{implicit conversion from 'int' to 'v2u16' (vector of 2 'unsigned short' values) changes value from 131071 to 65535}} + v2u8_a = v2u8_a + 0x1FF; // expected-error {{cannot convert between scalar type 'int' and vector type 'v2u8' (vector of 2 'unsigned char' values) as implicit conversion would cause truncation}} +} + +void intTestConstant(v2i64 v2i64_a, v2i32 v2i32_a, v2i16 v2i16_a, v2i8 v2i8_a) { + // Legal upper bounds. + v2i64_a = v2i64_a + (long long)0x7FFFFFFFFFFFFFFF; + v2i32_a = v2i32_a + (int)0x7FFFFFFF; + v2i16_a = v2i16_a + (short)0x7FFF; + v2i8_a = v2i8_a + (char)0x7F; + + // Legal lower bounds. + v2i64_a = v2i64_a + (-9223372036854775807); + v2i32_a = v2i32_a + (-2147483648); + v2i16_a = v2i16_a + (-32768); + v2i8_a = v2i8_a + (-128); + + // One increment/decrement more than the type can hold + v2i32_a = v2i32_a + 2147483648; // expected-warning {{implicit conversion from 'long' to 'v2i32' (vector of 2 'int' values) changes value from 2147483648 to -2147483648}} + v2i16_a = v2i16_a + 32768; // expected-warning {{implicit conversion from 'int' to 'v2i16' (vector of 2 'short' values) changes value from 32768 to -32768}} + v2i8_a = v2i8_a + 128; // expected-warning {{implicit conversion from 'int' to 'v2i8' (vector of 2 'char' values) changes value from 128 to -128}} + + v2i32_a = v2i32_a + (-2147483649); // expected-warning {{implicit conversion from 'long' to 'v2i32' (vector of 2 'int' values) changes value from -2147483649 to 2147483647}} + v2i16_a = v2i16_a + (-32769); // expected-warning {{implicit conversion from 'int' to 'v2i16' (vector of 2 'short' values) changes value from -32769 to 32767}} + v2i8_a = v2i8_a + (-129); // expected-error {{cannot convert between scalar type 'int' and vector type 'v2i8' (vector of 2 'char' values) as implicit conversion would cause truncation}} +} diff --git a/test/Sema/vector-gcc-compat.cpp b/test/Sema/vector-gcc-compat.cpp new file mode 100644 index 000000000000..12da314c325f --- /dev/null +++ b/test/Sema/vector-gcc-compat.cpp @@ -0,0 +1,328 @@ +// RUN: %clang_cc1 %s -verify -fsyntax-only -Weverything -std=c++11 -triple x86_64-apple-darwin10 + +// Test the compatibility of clang++'s vector extensions with g++'s vector +// extensions. In comparison to the extensions available in C, the !, ?:, && and +// || operators work on vector types. + +typedef long long v2i64 __attribute__((vector_size(16))); // expected-warning {{'long long' is incompatible with C++98}} +typedef int v2i32 __attribute__((vector_size(8))); +typedef short v2i16 __attribute__((vector_size(4))); +typedef char v2i8 __attribute__((vector_size(2))); + +typedef unsigned long long v2u64 __attribute__((vector_size(16))); // expected-warning {{'long long' is incompatible with C++98}} +typedef unsigned int v2u32 __attribute__((vector_size(8))); +typedef unsigned short v2u16 __attribute__((vector_size(4))); +typedef unsigned char v2u8 __attribute__((vector_size(2))); + +typedef float v4f32 __attribute__((vector_size(16))); +typedef double v2f64 __attribute__((vector_size(16))); +typedef double v4f64 __attribute__((vector_size(32))); +typedef int v4i32 __attribute((vector_size(16))); + +void arithmeticTest(void); +void logicTest(void); +void comparisonTest(void); +void floatTestSignedType(char a, short b, int c, long long d); // expected-warning {{'long long' is incompatible with C++98}} +void floatTestUnsignedType(unsigned char a, unsigned short b, unsigned int c, + unsigned long long d); // expected-warning {{'long long' is incompatible with C++98}} +void floatTestConstant(void); +void intTestType(char a, short b, int c, long long d); // expected-warning {{'long long' is incompatible with C++98}} +void intTestTypeUnsigned(unsigned char a, unsigned short b, unsigned int c, + unsigned long long d); // expected-warning {{'long long' is incompatible with C++98}} +void uintTestType(char a, short b, int c, long long d); // expected-warning {{'long long' is incompatible with C++98}} +void uintTestTypeUnsigned(unsigned char a, unsigned short b, unsigned int c, + unsigned long long d); // expected-warning {{'long long' is incompatible with C++98}} +void uintTestConstant(v2u64 v2u64_a, v2u32 v2u32_a, v2u16 v2u16_a, v2u8 v2u8_a); +void intTestConstant(v2i64 v2i64_a, v2i32 v2i32_a, v2i16 v2i16_a, v2i8 v2i8_a); + +void arithmeticTest(void) { + v2i64 v2i64_a = (v2i64){0, 1}; // expected-warning {{compound literals are a C99-specific feature}} + v2i64 v2i64_r; + + v2i64_r = v2i64_a + 1; + v2i64_r = v2i64_a - 1; + v2i64_r = v2i64_a * 1; + v2i64_r = v2i64_a / 1; + v2i64_r = v2i64_a % 1; + + v2i64_r = 1 + v2i64_a; + v2i64_r = 1 - v2i64_a; + v2i64_r = 1 * v2i64_a; + v2i64_r = 1 / v2i64_a; + v2i64_r = 1 % v2i64_a; + + v2i64_a += 1; + v2i64_a -= 1; + v2i64_a *= 1; + v2i64_a /= 1; + v2i64_a %= 1; +} + +void comparisonTest(void) { + v2i64 v2i64_a = (v2i64){0, 1}; // expected-warning {{compound literals are a C99-specific feature}} + v2i64 v2i64_r; + + v2i64_r = v2i64_a == 1; + v2i64_r = v2i64_a != 1; + v2i64_r = v2i64_a < 1; + v2i64_r = v2i64_a > 1; + v2i64_r = v2i64_a <= 1; + v2i64_r = v2i64_a >= 1; + + v2i64_r = 1 == v2i64_a; + v2i64_r = 1 != v2i64_a; + v2i64_r = 1 < v2i64_a; + v2i64_r = 1 > v2i64_a; + v2i64_r = 1 <= v2i64_a; + v2i64_r = 1 >= v2i64_a; +} + +void logicTest(void) { + v2i64 v2i64_a = (v2i64){0, 1}; // expected-warning {{compound literals are a C99-specific feature}} + v2i64 v2i64_b = (v2i64){2, 1}; // expected-warning {{compound literals are a C99-specific feature}} + v2i64 v2i64_c = (v2i64){3, 1}; // expected-warning {{compound literals are a C99-specific feature}} + v2i64 v2i64_r; + + v2i64_r = !v2i64_a; // expected-error {{invalid argument type 'v2i64' (vector of 2 'long long' values) to unary expression}} + v2i64_r = ~v2i64_a; + + v2i64_r = v2i64_a ? v2i64_b : v2i64_c; // expected-error {{value of type 'v2i64' (vector of 2 'long long' values) is not contextually convertible to 'bool'}} + + v2i64_r = v2i64_a & 1; + v2i64_r = v2i64_a | 1; + v2i64_r = v2i64_a ^ 1; + + v2i64_r = 1 & v2i64_a; + v2i64_r = 1 | v2i64_a; + v2i64_r = 1 ^ v2i64_a; + v2i64_a &= 1; + v2i64_a |= 1; + v2i64_a ^= 1; + + v2i64_r = v2i64_a && 1; + v2i64_r = v2i64_a || 1; + + v2i64_r = v2i64_a << 1; + v2i64_r = v2i64_a >> 1; + + v2i64_r = 1 << v2i64_a; + v2i64_r = 1 >> v2i64_a; + + v2i64_a <<= 1; + v2i64_a >>= 1; +} + +// For operations with floating point types, we check that interger constants +// can be respresented, or failing that checking based on the integer types. +void floatTestConstant(void) { + // Test that constants added to floats must be expressible as floating point + // numbers. + v4f32 v4f32_a = {0.4f, 0.4f, 0.4f, 0.4f}; + v4f32_a = v4f32_a + 1; + v4f32_a = v4f32_a + 0xFFFFFF; + v4f32_a = v4f32_a + (-1567563LL); // expected-warning {{'long long' is incompatible with C++98}} + v4f32_a = v4f32_a + (16777208); + v4f32_a = v4f32_a + (16777219); // expected-error {{cannot convert between scalar type 'int' and vector type 'v4f32' (vector of 4 'float' values) as implicit conversion would cause truncation}} +} + +void floatTestConstantComparison(void); +void doubleTestConstantComparison(void); + +void floatTestConstantComparison(void) { + v4f32 v4f32_a = {0.4f, 0.4f, 0.4f, 0.4f}; + v4i32 v4i32_r; + v4i32_r = v4f32_a > 0.4f; + v4i32_r = v4f32_a >= 0.4f; + v4i32_r = v4f32_a < 0.4f; + v4i32_r = v4f32_a <= 0.4f; + v4i32_r = v4f32_a == 0.4f; // expected-warning {{comparing floating point with == or != is unsafe}} + v4i32_r = v4f32_a != 0.4f; // expected-warning {{comparing floating point with == or != is unsafe}} +} + +void doubleTestConstantComparison(void) { + v2f64 v2f64_a = {0.4, 0.4}; + v2i64 v2i64_r; + v2i64_r = v2f64_a > 0.4; + v2i64_r = v2f64_a >= 0.4; + v2i64_r = v2f64_a < 0.4; + v2i64_r = v2f64_a <= 0.4; + v2i64_r = v2f64_a == 0.4; // expected-warning {{comparing floating point with == or != is unsafe}} + v2i64_r = v2f64_a != 0.4; // expected-warning {{comparing floating point with == or != is unsafe}} +} + +void floatTestUnsignedType(unsigned char a, unsigned short b, unsigned int c, + unsigned long long d) { // expected-warning {{'long long' is incompatible with C++98}} + v4f32 v4f32_a = {0.4f, 0.4f, 0.4f, 0.4f}; + v4f64 v4f64_b = {0.4, 0.4, 0.4, 0.4}; + + v4f32_a = v4f32_a + a; + v4f32_a = v4f32_a + b; + v4f32_a = v4f32_a + c; // expected-error {{cannot convert between scalar type 'unsigned int' and vector type 'v4f32' (vector of 4 'float' values) as implicit conversion would cause truncation}} + v4f32_a = v4f32_a + d; // expected-error {{cannot convert between scalar type 'unsigned long long' and vector type 'v4f32' (vector of 4 'float' values) as implicit conversion would cause truncation}} + + v4f64_b = v4f64_b + a; + v4f64_b = v4f64_b + b; + v4f64_b = v4f64_b + c; + v4f64_b = v4f64_b + d; // expected-error {{cannot convert between scalar type 'unsigned long long' and vector type 'v4f64' (vector of 4 'double' values) as implicit conversion would cause truncation}} +} + +void floatTestSignedType(char a, short b, int c, long long d) { // expected-warning {{'long long' is incompatible with C++98}} + v4f32 v4f32_a = {0.4f, 0.4f, 0.4f, 0.4f}; + v4f64 v4f64_b = {0.4, 0.4, 0.4, 0.4}; + + v4f32_a = v4f32_a + a; + v4f32_a = v4f32_a + b; + v4f32_a = v4f32_a + c; // expected-error {{cannot convert between scalar type 'int' and vector type 'v4f32' (vector of 4 'float' values) as implicit conversion would cause truncation}} + v4f32_a = v4f32_a + d; // expected-error {{cannot convert between scalar type 'long long' and vector type 'v4f32' (vector of 4 'float' values) as implicit conversion would cause truncation}} + + v4f64_b = v4f64_b + a; + v4f64_b = v4f64_b + b; + v4f64_b = v4f64_b + c; + v4f64_b = v4f64_b + d; // expected-error {{cannot convert between scalar type 'long long' and vector type 'v4f64' (vector of 4 'double' values) as implicit conversion would cause truncation}} +} + +void intTestType(char a, short b, int c, long long d) { // expected-warning {{'long long' is incompatible with C++98}} + v2i64 v2i64_a = {1, 2}; + v2i32 v2i32_a = {1, 2}; + v2i16 v2i16_a = {1, 2}; + v2i8 v2i8_a = {1, 2}; + + v2i64_a = v2i64_a + d; + v2i64_a = v2i64_a + c; + v2i64_a = v2i64_a + b; + v2i64_a = v2i64_a + a; + + v2i32_a = v2i32_a + d; // expected-warning {{implicit conversion loses integer precision: 'long long' to 'v2i32' (vector of 2 'int' values)}} + v2i32_a = v2i32_a + c; + v2i32_a = v2i32_a + b; + v2i32_a = v2i32_a + a; + + v2i16_a = v2i16_a + d; // expected-error {{cannot convert between scalar type 'long long' and vector type 'v2i16' (vector of 2 'short' values) as implicit conversion would cause truncation}} + v2i16_a = v2i16_a + c; // expected-warning {{implicit conversion loses integer precision: 'int' to 'v2i16' (vector of 2 'short' values)}} + v2i16_a = v2i16_a + b; + v2i16_a = v2i16_a + a; + + v2i8_a = v2i8_a + d; // expected-error {{cannot convert between scalar type 'long long' and vector type 'v2i8' (vector of 2 'char' values) as implicit conversion would cause truncation}} + v2i8_a = v2i8_a + c; // expected-error {{cannot convert between scalar type 'int' and vector type 'v2i8' (vector of 2 'char' values) as implicit conversion would cause truncation}} + v2i8_a = v2i8_a + b; // expected-warning {{implicit conversion loses integer precision: 'short' to 'v2i8' (vector of 2 'char' values)}} + v2i8_a = v2i8_a + a; +} + +void intTestTypeUnsigned(unsigned char a, unsigned short b, unsigned int c, + unsigned long long d) { // expected-warning {{'long long' is incompatible with C++98}} + v2i64 v2i64_a = {1, 2}; + v2i32 v2i32_a = {1, 2}; + v2i16 v2i16_a = {1, 2}; + v2i8 v2i8_a = {1, 2}; + + v2i64_a = v2i64_a + d; // expected-error {{cannot convert between scalar type 'unsigned long long' and vector type 'v2i64' (vector of 2 'long long' values) as implicit conversion would cause truncation}} + + v2i64_a = v2i64_a + c; + v2i64_a = v2i64_a + b; + v2i64_a = v2i64_a + a; + + v2i32_a = v2i32_a + d; // expected-warning {{implicit conversion loses integer precision: 'unsigned long long' to 'v2i32' (vector of 2 'int' values)}} + v2i32_a = v2i32_a + c; // expected-error {{cannot convert between scalar type 'unsigned int' and vector type 'v2i32' (vector of 2 'int' values) as implicit conversion would cause truncation}} + v2i32_a = v2i32_a + b; + v2i32_a = v2i32_a + a; + + v2i16_a = v2i16_a + d; // expected-error {{cannot convert between scalar type 'unsigned long long' and vector type 'v2i16' (vector of 2 'short' values) as implicit conversion would cause truncation}} + v2i16_a = v2i16_a + c; // expected-warning {{implicit conversion loses integer precision: 'unsigned int' to 'v2i16' (vector of 2 'short' values)}} + v2i16_a = v2i16_a + b; // expected-error {{cannot convert between scalar type 'unsigned short' and vector type 'v2i16' (vector of 2 'short' values) as implicit conversion would cause truncation}} + v2i16_a = v2i16_a + a; + + v2i8_a = v2i8_a + d; // expected-error {{cannot convert between scalar type 'unsigned long long' and vector type 'v2i8' (vector of 2 'char' values) as implicit conversion would cause truncation}} + v2i8_a = v2i8_a + c; // expected-error {{cannot convert between scalar type 'unsigned int' and vector type 'v2i8' (vector of 2 'char' values) as implicit conversion would cause truncation}} + v2i8_a = v2i8_a + b; // expected-warning {{implicit conversion loses integer precision: 'unsigned short' to 'v2i8' (vector of 2 'char' values)}} + v2i8_a = v2i8_a + a; // expected-error {{cannot convert between scalar type 'unsigned char' and vector type 'v2i8' (vector of 2 'char' values) as implicit conversion would cause truncation}} +} + +void uintTestType(char a, short b, int c, long long d) { // expected-warning {{'long long' is incompatible with C++98}} + v2u64 v2u64_a = {1, 2}; + v2u32 v2u32_a = {1, 2}; + v2u16 v2u16_a = {1, 2}; + v2u8 v2u8_a = {1, 2}; + + v2u64_a = v2u64_a + d; // expected-warning {{implicit conversion changes signedness: 'long long' to 'v2u64' (vector of 2 'unsigned long long' values)}} + v2u64_a = v2u64_a + c; // expected-warning {{implicit conversion changes signedness: 'int' to 'v2u64' (vector of 2 'unsigned long long' values)}} + v2u64_a = v2u64_a + b; // expected-warning {{implicit conversion changes signedness: 'short' to 'v2u64' (vector of 2 'unsigned long long' values)}} + v2u64_a = v2u64_a + a; // expected-warning {{implicit conversion changes signedness: 'char' to 'v2u64' (vector of 2 'unsigned long long' values)}} + + v2u32_a = v2u32_a + d; // expected-warning {{implicit conversion loses integer precision: 'long long' to 'v2u32' (vector of 2 'unsigned int' values)}} + v2u32_a = v2u32_a + c; // expected-warning {{implicit conversion changes signedness: 'int' to 'v2u32' (vector of 2 'unsigned int' values)}} + v2u32_a = v2u32_a + b; // expected-warning {{implicit conversion changes signedness: 'short' to 'v2u32' (vector of 2 'unsigned int' values)}} + v2u32_a = v2u32_a + a; // expected-warning {{implicit conversion changes signedness: 'char' to 'v2u32' (vector of 2 'unsigned int' values)}} + + v2u16_a = v2u16_a + d; // expected-error {{cannot convert between scalar type 'long long' and vector type 'v2u16' (vector of 2 'unsigned short' values) as implicit conversion would cause truncation}} + v2u16_a = v2u16_a + c; // expected-warning {{implicit conversion loses integer precision: 'int' to 'v2u16' (vector of 2 'unsigned short' values)}} + v2u16_a = v2u16_a + b; // expected-warning {{implicit conversion changes signedness: 'short' to 'v2u16' (vector of 2 'unsigned short' values)}} + v2u16_a = v2u16_a + a; // expected-warning {{implicit conversion changes signedness: 'char' to 'v2u16' (vector of 2 'unsigned short' values)}} + + v2u8_a = v2u8_a + d; // expected-error {{cannot convert between scalar type 'long long' and vector type 'v2u8' (vector of 2 'unsigned char' values) as implicit conversion would cause truncation}} + v2u8_a = v2u8_a + c; // expected-error {{cannot convert between scalar type 'int' and vector type 'v2u8' (vector of 2 'unsigned char' values) as implicit conversion would cause truncation}} + v2u8_a = v2u8_a + b; // expected-warning {{implicit conversion loses integer precision: 'short' to 'v2u8' (vector of 2 'unsigned char' values)}} + v2u8_a = v2u8_a + a; // expected-warning {{implicit conversion changes signedness: 'char' to 'v2u8' (vector of 2 'unsigned char' values)}} +} + +void uintTestTypeUnsigned(unsigned char a, unsigned short b, unsigned int c, + unsigned long long d) { // expected-warning {{'long long' is incompatible with C++98}} + v2u64 v2u64_a = {1, 2}; + v2u32 v2u32_a = {1, 2}; + v2u16 v2u16_a = {1, 2}; + v2u8 v2u8_a = {1, 2}; + + v2u64_a = v2u64_a + d; + v2u64_a = v2u64_a + c; + v2u64_a = v2u64_a + b; + v2u64_a = v2u64_a + a; + + v2u32_a = v2u32_a + d; // expected-warning {{implicit conversion loses integer precision: 'unsigned long long' to 'v2u32' (vector of 2 'unsigned int' values)}} + v2u32_a = v2u32_a + c; + v2u32_a = v2u32_a + b; + v2u32_a = v2u32_a + a; + + v2u16_a = v2u16_a + d; // expected-error {{cannot convert between scalar type 'unsigned long long' and vector type 'v2u16' (vector of 2 'unsigned short' values) as implicit conversion would cause truncation}} + v2u16_a = v2u16_a + c; // expected-warning {{implicit conversion loses integer precision: 'unsigned int' to 'v2u16' (vector of 2 'unsigned short' values)}} + v2u16_a = v2u16_a + b; + v2u16_a = v2u16_a + a; + + v2u8_a = v2u8_a + d; // expected-error {{cannot convert between scalar type 'unsigned long long' and vector type 'v2u8' (vector of 2 'unsigned char' values) as implicit conversion would cause truncation}} + v2u8_a = v2u8_a + c; // expected-error {{cannot convert between scalar type 'unsigned int' and vector type 'v2u8' (vector of 2 'unsigned char' values) as implicit conversion would cause truncation}} + v2u8_a = v2u8_a + b; // expected-warning {{implicit conversion loses integer precision: 'unsigned short' to 'v2u8' (vector of 2 'unsigned char' values)}} + v2u8_a = v2u8_a + a; +} + +void uintTestConstant(v2u64 v2u64_a, v2u32 v2u32_a, v2u16 v2u16_a, + v2u8 v2u8_a) { + v2u64_a = v2u64_a + 0xFFFFFFFFFFFFFFFF; + v2u32_a = v2u32_a + 0xFFFFFFFF; + v2u16_a = v2u16_a + 0xFFFF; + v2u8_a = v2u8_a + 0xFF; + + v2u32_a = v2u32_a + 0x1FFFFFFFF; // expected-warning {{implicit conversion from 'long' to 'v2u32' (vector of 2 'unsigned int' values) changes value from 8589934591 to 4294967295}} + v2u16_a = v2u16_a + 0x1FFFF; // expected-warning {{implicit conversion from 'int' to 'v2u16' (vector of 2 'unsigned short' values) changes value from 131071 to 65535}} + v2u8_a = v2u8_a + 0x1FF; // expected-error {{cannot convert between scalar type 'int' and vector type 'v2u8' (vector of 2 'unsigned char' values) as implicit conversion would cause truncation}} +} + +void intTestConstant(v2i64 v2i64_a, v2i32 v2i32_a, v2i16 v2i16_a, v2i8 v2i8_a) { + // Legal upper bounds. + v2i64_a = v2i64_a + static_cast(0x7FFFFFFFFFFFFFFF); // expected-warning {{'long long' is incompatible with C++98}} + v2i32_a = v2i32_a + static_cast(0x7FFFFFFF); + v2i16_a = v2i16_a + static_cast(0x7FFF); + v2i8_a = v2i8_a + static_cast(0x7F); + + // Legal lower bounds. + v2i64_a = v2i64_a + (-9223372036854775807); + v2i32_a = v2i32_a + (-2147483648); + v2i16_a = v2i16_a + (-32768); + v2i8_a = v2i8_a + (-128); + + // One increment/decrement more than the type can hold + v2i32_a = v2i32_a + 2147483648; // expected-warning {{implicit conversion from 'long' to 'v2i32' (vector of 2 'int' values) changes value from 2147483648 to -2147483648}} + v2i16_a = v2i16_a + 32768; // expected-warning {{implicit conversion from 'int' to 'v2i16' (vector of 2 'short' values) changes value from 32768 to -32768}} + v2i8_a = v2i8_a + 128; // expected-warning {{implicit conversion from 'int' to 'v2i8' (vector of 2 'char' values) changes value from 128 to -128}} + + v2i32_a = v2i32_a + (-2147483649); // expected-warning {{implicit conversion from 'long' to 'v2i32' (vector of 2 'int' values) changes value from -2147483649 to 2147483647}} + v2i16_a = v2i16_a + (-32769); // expected-warning {{implicit conversion from 'int' to 'v2i16' (vector of 2 'short' values) changes value from -32769 to 32767}} + v2i8_a = v2i8_a + (-129); // expected-error {{cannot convert between scalar type 'int' and vector type 'v2i8' (vector of 2 'char' values) as implicit conversion would cause truncation}} +} diff --git a/test/Sema/vector-ops.c b/test/Sema/vector-ops.c index 9cdd9d2f1748..575f38b972f5 100644 --- a/test/Sema/vector-ops.c +++ b/test/Sema/vector-ops.c @@ -13,11 +13,11 @@ void test1(v2u v2ua, v2s v2sa, v2f v2fa) { (void)(~v2fa); // expected-error{{invalid argument type 'v2f' (vector of 2 'float' values) to unary}} // Comparison operators - v2ua = (v2ua==v2sa); // expected-warning{{incompatible vector types assigning to 'v2u' (vector of 2 'unsigned int' values) from '__attribute__((__vector_size__(2 * sizeof(int)))) int' (vector of 2 'int' values)}} + v2ua = (v2ua==v2sa); // expected-warning{{incompatible vector types assigning to 'v2u' (vector of 2 'unsigned int' values) from '__attribute__((__vector_size__(2 * sizeof(int)))) int' (vector of 2 'int' values}} v2sa = (v2ua==v2sa); // Arrays - int array1[v2ua]; // expected-error{{size of array has non-integer type 'v2u' (vector of 2 'unsigned int' values)}} + int array1[v2ua]; // expected-error{{size of array has non-integer type 'v2u' (vector of 2 'unsigned int' values}} int array2[17]; // FIXME: error message below needs type! (void)(array2[v2ua]); // expected-error{{array subscript is not an integer}} @@ -28,108 +28,108 @@ void test1(v2u v2ua, v2s v2sa, v2f v2fa) { } void testLogicalVecVec(v2u v2ua, v2s v2sa, v2f v2fa) { - // Logical operators - v2ua = v2ua && v2ua; // expected-warning {{incompatible vector types assigning to 'v2u' (vector of 2 'unsigned int' values) from '__attribute__((__vector_size__(2 * sizeof(int)))) int' (vector of 2 'int' values)}} - v2ua = v2ua || v2ua; // expected-warning {{incompatible vector types assigning to 'v2u' (vector of 2 'unsigned int' values) from '__attribute__((__vector_size__(2 * sizeof(int)))) int' (vector of 2 'int' values)}} + v2ua = v2ua && v2ua; // expected-error {{logical expression with vector types 'v2u' (vector of 2 'unsigned int' values) and 'v2u'}} + v2ua = v2ua || v2ua; // expected-error {{logical expression with vector types 'v2u' (vector of 2 'unsigned int' values) and 'v2u'}} - v2ua = v2sa && v2ua; // expected-warning {{incompatible vector types assigning to 'v2u' (vector of 2 'unsigned int' values) from '__attribute__((__vector_size__(2 * sizeof(int)))) int' (vector of 2 'int' values)}} - v2ua = v2sa || v2ua; // expected-warning {{incompatible vector types assigning to 'v2u' (vector of 2 'unsigned int' values) from '__attribute__((__vector_size__(2 * sizeof(int)))) int' (vector of 2 'int' values)}} + v2ua = v2sa && v2ua; // expected-error {{logical expression with vector types 'v2s' (vector of 2 'int' values) and 'v2u' (vector of 2 'unsigned int' values)}} + v2ua = v2sa || v2ua; // expected-error {{logical expression with vector types 'v2s' (vector of 2 'int' values) and 'v2u' (vector of 2 'unsigned int' values)}} - v2ua = v2ua && v2fa; // expected-warning {{incompatible vector types assigning to 'v2u' (vector of 2 'unsigned int' values) from '__attribute__((__vector_size__(2 * sizeof(int)))) int' (vector of 2 'int' values)}} - v2ua = v2ua || v2fa; // expected-warning {{incompatible vector types assigning to 'v2u' (vector of 2 'unsigned int' values) from '__attribute__((__vector_size__(2 * sizeof(int)))) int' (vector of 2 'int' values)}} + v2ua = v2ua && v2fa; // expected-error {{logical expression with vector types 'v2u' (vector of 2 'unsigned int' values) and 'v2f' (vector of 2 'float' values)}} + v2ua = v2ua || v2fa; // expected-error {{logical expression with vector types 'v2u' (vector of 2 'unsigned int' values) and 'v2f' (vector of 2 'float' values)}} - v2ua = v2sa && v2fa; // expected-warning {{incompatible vector types assigning to 'v2u' (vector of 2 'unsigned int' values) from '__attribute__((__vector_size__(2 * sizeof(int)))) int' (vector of 2 'int' values)}} - v2ua = v2sa || v2fa; // expected-warning {{incompatible vector types assigning to 'v2u' (vector of 2 'unsigned int' values) from '__attribute__((__vector_size__(2 * sizeof(int)))) int' (vector of 2 'int' values)}} + v2ua = v2sa && v2fa; // expected-error {{logical expression with vector types 'v2s' (vector of 2 'int' values) and 'v2f' (vector of 2 'float' values)}} + v2ua = v2sa || v2fa; // expected-error {{logical expression with vector types 'v2s' (vector of 2 'int' values) and 'v2f' (vector of 2 'float' values)}} - v2sa = v2sa && v2sa; - v2sa = v2sa || v2sa; + v2sa = v2sa && v2sa; // expected-error {{logical expression with vector types 'v2s' (vector of 2 'int' values) and 'v2s'}} + v2sa = v2sa || v2sa; // expected-error {{logical expression with vector types 'v2s' (vector of 2 'int' values) and 'v2s'}} - v2sa = v2ua && v2ua; - v2sa = v2ua || v2ua; + v2sa = v2ua && v2ua; // expected-error {{logical expression with vector types 'v2u' (vector of 2 'unsigned int' values) and 'v2u'}} + v2sa = v2ua || v2ua; // expected-error {{logical expression with vector types 'v2u' (vector of 2 'unsigned int' values) and 'v2u'}} - v2sa = v2sa && v2ua; - v2sa = v2sa || v2ua; + v2sa = v2sa && v2ua; // expected-error {{logical expression with vector types 'v2s' (vector of 2 'int' values) and 'v2u' (vector of 2 'unsigned int' values)}} + v2sa = v2sa || v2ua; // expected-error {{logical expression with vector types 'v2s' (vector of 2 'int' values) and 'v2u' (vector of 2 'unsigned int' values)}} - v2sa = v2sa && v2fa; - v2sa = v2sa || v2fa; + v2sa = v2sa && v2fa; // expected-error {{logical expression with vector types 'v2s' (vector of 2 'int' values) and 'v2f' (vector of 2 'float' values)}} + v2sa = v2sa || v2fa; // expected-error {{logical expression with vector types 'v2s' (vector of 2 'int' values) and 'v2f' (vector of 2 'float' values)}} - v2sa = v2ua && v2fa; - v2sa = v2ua || v2fa; + v2sa = v2ua && v2fa; // expected-error {{logical expression with vector types 'v2u' (vector of 2 'unsigned int' values) and 'v2f' (vector of 2 'float' values)}} + v2sa = v2ua || v2fa; // expected-error {{logical expression with vector types 'v2u' (vector of 2 'unsigned int' values) and 'v2f' (vector of 2 'float' values)}} - v2fa = v2fa && v2fa; // expected-warning {{incompatible vector types assigning to 'v2f' (vector of 2 'float' values) from '__attribute__((__vector_size__(2 * sizeof(int)))) int' (vector of 2 'int' values)}} - v2fa = v2fa || v2fa; // expected-warning {{incompatible vector types assigning to 'v2f' (vector of 2 'float' values) from '__attribute__((__vector_size__(2 * sizeof(int)))) int' (vector of 2 'int' values)}} + v2fa = v2fa && v2fa; // expected-error {{logical expression with vector types 'v2f' (vector of 2 'float' values) and 'v2f'}} + v2fa = v2fa || v2fa; // expected-error {{logical expression with vector types 'v2f' (vector of 2 'float' values) and 'v2f'}} - v2fa = v2sa && v2fa; // expected-warning {{incompatible vector types assigning to 'v2f' (vector of 2 'float' values) from '__attribute__((__vector_size__(2 * sizeof(int)))) int' (vector of 2 'int' values)}} - v2fa = v2sa || v2fa; // expected-warning {{incompatible vector types assigning to 'v2f' (vector of 2 'float' values) from '__attribute__((__vector_size__(2 * sizeof(int)))) int' (vector of 2 'int' values)}} + v2fa = v2sa && v2fa; // expected-error {{logical expression with vector types 'v2s' (vector of 2 'int' values) and 'v2f' (vector of 2 'float' values)}} + v2fa = v2sa || v2fa; // expected-error {{logical expression with vector types 'v2s' (vector of 2 'int' values) and 'v2f' (vector of 2 'float' values)}} - v2fa = v2ua && v2fa; // expected-warning {{incompatible vector types assigning to 'v2f' (vector of 2 'float' values) from '__attribute__((__vector_size__(2 * sizeof(int)))) int' (vector of 2 'int' values)}} - v2fa = v2ua || v2fa; // expected-warning {{incompatible vector types assigning to 'v2f' (vector of 2 'float' values) from '__attribute__((__vector_size__(2 * sizeof(int)))) int' (vector of 2 'int' values)}} + v2fa = v2ua && v2fa; // expected-error {{logical expression with vector types 'v2u' (vector of 2 'unsigned int' values) and 'v2f' (vector of 2 'float' values)}} + v2fa = v2ua || v2fa; // expected-error {{logical expression with vector types 'v2u' (vector of 2 'unsigned int' values) and 'v2f' (vector of 2 'float' values)}} - v2fa = v2ua && v2ua; // expected-warning {{incompatible vector types assigning to 'v2f' (vector of 2 'float' values) from '__attribute__((__vector_size__(2 * sizeof(int)))) int' (vector of 2 'int' values)}} - v2fa = v2ua || v2ua; // expected-warning {{incompatible vector types assigning to 'v2f' (vector of 2 'float' values) from '__attribute__((__vector_size__(2 * sizeof(int)))) int' (vector of 2 'int' values)}} + v2fa = v2ua && v2ua; // expected-error {{logical expression with vector types 'v2u' (vector of 2 'unsigned int' values) and 'v2u'}} + v2fa = v2ua || v2ua; // expected-error {{logical expression with vector types 'v2u' (vector of 2 'unsigned int' values) and 'v2u'}} - v2fa = v2sa && v2sa; // expected-warning {{incompatible vector types assigning to 'v2f' (vector of 2 'float' values) from '__attribute__((__vector_size__(2 * sizeof(int)))) int' (vector of 2 'int' values)}} - v2fa = v2sa || v2sa; // expected-warning {{incompatible vector types assigning to 'v2f' (vector of 2 'float' values) from '__attribute__((__vector_size__(2 * sizeof(int)))) int' (vector of 2 'int' values)}} + v2fa = v2sa && v2sa; // expected-error {{logical expression with vector types 'v2s' (vector of 2 'int' values) and 'v2s'}} + v2fa = v2sa || v2sa; // expected-error {{logical expression with vector types 'v2s' (vector of 2 'int' values) and 'v2s'}} - v2fa = v2sa && v2ua; // expected-warning {{incompatible vector types assigning to 'v2f' (vector of 2 'float' values) from '__attribute__((__vector_size__(2 * sizeof(int)))) int' (vector of 2 'int' values)}} - v2fa = v2sa || v2ua; // expected-warning {{incompatible vector types assigning to 'v2f' (vector of 2 'float' values) from '__attribute__((__vector_size__(2 * sizeof(int)))) int' (vector of 2 'int' values)}} + v2fa = v2sa && v2ua; // expected-error {{logical expression with vector types 'v2s' (vector of 2 'int' values) and 'v2u' (vector of 2 'unsigned int' values)}} + v2fa = v2sa || v2ua; // expected-error {{logical expression with vector types 'v2s' (vector of 2 'int' values) and 'v2u' (vector of 2 'unsigned int' values)}} } void testLogicalVecScalar(v2u v2ua, v2s v2sa, v2f v2fa) { - unsigned u1; - v2ua = v2ua && u1; // expected-error {{cannot convert between vector values of different size ('v2u' (vector of 2 'unsigned int' values) and 'unsigned int')}} expected-error {{invalid operands to binary expression ('v2u' (vector of 2 'unsigned int' values) and 'unsigned int')}} - v2ua = v2ua || u1; // expected-error {{cannot convert between vector values of different size ('v2u' (vector of 2 'unsigned int' values) and 'unsigned int')}} expected-error {{invalid operands to binary expression ('v2u' (vector of 2 'unsigned int' values) and 'unsigned int')}} + v2ua = v2ua && u1; // expected-error {{logical expression with vector type 'v2u' (vector of 2 'unsigned int' values) and non-vector type 'unsigned int' is only supported in C++}} + v2ua = v2ua || u1; // expected-error {{logical expression with vector type 'v2u' (vector of 2 'unsigned int' values) and non-vector type 'unsigned int' is only supported in C++}} - v2sa = v2sa && u1; // expected-error {{cannot convert between vector values of different size ('v2s' (vector of 2 'int' values) and 'unsigned int')}} expected-error {{invalid operands to binary expression ('v2s' (vector of 2 'int' values) and 'unsigned int')}} - v2sa = v2sa || u1; // expected-error {{cannot convert between vector values of different size ('v2s' (vector of 2 'int' values) and 'unsigned int')}} expected-error {{invalid operands to binary expression ('v2s' (vector of 2 'int' values) and 'unsigned int')}} + v2sa = v2sa && u1; // expected-error {{cannot convert between scalar type 'unsigned int' and vector type 'v2s' (vector of 2 'int' values) as implicit conversion would cause truncation}} expected-error {{invalid operands to binary expression ('v2s' (vector of 2 'int' values) and 'unsigned int')}} + v2sa = v2sa || u1; // expected-error {{cannot convert between scalar type 'unsigned int' and vector type 'v2s' (vector of 2 'int' values) as implicit conversion would cause truncation}} expected-error {{invalid operands to binary expression ('v2s' (vector of 2 'int' values) and 'unsigned int')}} - v2ua = v2sa && u1; // expected-error {{cannot convert between vector values of different size ('v2s' (vector of 2 'int' values) and 'unsigned int')}} expected-error {{invalid operands to binary expression ('v2s' (vector of 2 'int' values) and 'unsigned int')}} - v2ua = v2sa || u1; // expected-error {{cannot convert between vector values of different size ('v2s' (vector of 2 'int' values) and 'unsigned int')}} expected-error {{invalid operands to binary expression ('v2s' (vector of 2 'int' values) and 'unsigned int')}} - v2sa = v2ua && u1; // expected-error {{cannot convert between vector values of different size ('v2u' (vector of 2 'unsigned int' values) and 'unsigned int')}} expected-error {{invalid operands to binary expression ('v2u' (vector of 2 'unsigned int' values) and 'unsigned int')}} - v2sa = v2ua || u1; // expected-error {{cannot convert between vector values of different size ('v2u' (vector of 2 'unsigned int' values) and 'unsigned int')}} expected-error {{invalid operands to binary expression ('v2u' (vector of 2 'unsigned int' values) and 'unsigned int')}} + v2ua = v2sa && u1; // expected-error {{cannot convert between scalar type 'unsigned int' and vector type 'v2s' (vector of 2 'int' values) as implicit conversion would cause truncation}} expected-error {{invalid operands to binary expression ('v2s' (vector of 2 'int' values) and 'unsigned int')}} + v2ua = v2sa || u1; // expected-error {{cannot convert between scalar type 'unsigned int' and vector type 'v2s' (vector of 2 'int' values) as implicit conversion would cause truncation}} expected-error {{invalid operands to binary expression ('v2s' (vector of 2 'int' values) and 'unsigned int')}} + v2sa = v2ua && u1; // expected-error {{logical expression with vector type 'v2u' (vector of 2 'unsigned int' values) and non-vector type 'unsigned int' is only supported in C++}} + v2sa = v2ua || u1; // expected-error {{logical expression with vector type 'v2u' (vector of 2 'unsigned int' values) and non-vector type 'unsigned int' is only supported in C++}} - v2ua = v2fa && u1; // expected-error {{cannot convert between vector values of different size ('v2f' (vector of 2 'float' values) and 'unsigned int')}} expected-error {{invalid operands to binary expression ('v2f' (vector of 2 'float' values) and 'unsigned int')}} - v2ua = v2fa || u1; // expected-error {{cannot convert between vector values of different size ('v2f' (vector of 2 'float' values) and 'unsigned int')}} expected-error {{invalid operands to binary expression ('v2f' (vector of 2 'float' values) and 'unsigned int')}} + v2ua = v2fa && u1; // expected-error {{cannot convert between scalar type 'unsigned int' and vector type 'v2f' (vector of 2 'float' values) as implicit conversion would cause truncation}} expected-error {{invalid operands to binary expression ('v2f' (vector of 2 'float' values) and 'unsigned int')}} + v2ua = v2fa || u1; // expected-error {{cannot convert between scalar type 'unsigned int' and vector type 'v2f' (vector of 2 'float' values) as implicit conversion would cause truncation}} expected-error {{invalid operands to binary expression ('v2f' (vector of 2 'float' values) and 'unsigned int')}} - v2sa = v2fa && u1; // expected-error {{cannot convert between vector values of different size ('v2f' (vector of 2 'float' values) and 'unsigned int')}} expected-error {{invalid operands to binary expression ('v2f' (vector of 2 'float' values) and 'unsigned int')}} - v2sa = v2fa || u1; // expected-error {{cannot convert between vector values of different size ('v2f' (vector of 2 'float' values) and 'unsigned int')}} expected-error {{invalid operands to binary expression ('v2f' (vector of 2 'float' values) and 'unsigned int')}} + v2sa = v2fa && u1; // expected-error {{cannot convert between scalar type 'unsigned int' and vector type 'v2f' (vector of 2 'float' values) as implicit conversion would cause truncation}} expected-error {{invalid operands to binary expression ('v2f' (vector of 2 'float' values) and 'unsigned int')}} + v2sa = v2fa || u1; // expected-error {{cannot convert between scalar type 'unsigned int' and vector type 'v2f' (vector of 2 'float' values) as implicit conversion would cause truncation}} expected-error {{invalid operands to binary expression ('v2f' (vector of 2 'float' values) and 'unsigned int')}} int s1; - v2ua = v2ua && s1; // expected-error {{cannot convert between vector values of different size ('v2u' (vector of 2 'unsigned int' values) and 'int')}} expected-error {{invalid operands to binary expression ('v2u' (vector of 2 'unsigned int' values) and 'int')}} - v2ua = v2ua || s1; // expected-error {{cannot convert between vector values of different size ('v2u' (vector of 2 'unsigned int' values) and 'int')}} expected-error {{invalid operands to binary expression ('v2u' (vector of 2 'unsigned int' values) and 'int')}} + v2ua = v2ua && s1; // expected-error {{logical expression with vector type 'v2u' (vector of 2 'unsigned int' values) and non-vector type 'int' is only supported in C++}} + v2ua = v2ua || s1; // expected-error {{logical expression with vector type 'v2u' (vector of 2 'unsigned int' values) and non-vector type 'int' is only supported in C++}} - v2sa = v2sa && s1; // expected-error {{cannot convert between vector values of different size ('v2s' (vector of 2 'int' values) and 'int')}} expected-error {{invalid operands to binary expression ('v2s' (vector of 2 'int' values) and 'int')}} - v2sa = v2sa || s1; // expected-error {{cannot convert between vector values of different size ('v2s' (vector of 2 'int' values) and 'int')}} expected-error {{invalid operands to binary expression ('v2s' (vector of 2 'int' values) and 'int')}} + v2sa = v2sa && s1; // expected-error {{logical expression with vector type 'v2s' (vector of 2 'int' values) and non-vector type 'int' is only supported in C++}} + v2sa = v2sa || s1; // expected-error {{logical expression with vector type 'v2s' (vector of 2 'int' values) and non-vector type 'int' is only supported in C++}} - v2ua = v2sa && s1; // expected-error {{cannot convert between vector values of different size ('v2s' (vector of 2 'int' values) and 'int')}} expected-error {{invalid operands to binary expression ('v2s' (vector of 2 'int' values) and 'int')}} - v2ua = v2sa || s1; // expected-error {{cannot convert between vector values of different size ('v2s' (vector of 2 'int' values) and 'int')}} expected-error {{invalid operands to binary expression ('v2s' (vector of 2 'int' values) and 'int')}} - v2sa = v2ua && s1; // expected-error {{cannot convert between vector values of different size ('v2u' (vector of 2 'unsigned int' values) and 'int')}} expected-error {{invalid operands to binary expression ('v2u' (vector of 2 'unsigned int' values) and 'int')}} - v2sa = v2ua || s1; // expected-error {{cannot convert between vector values of different size ('v2u' (vector of 2 'unsigned int' values) and 'int')}} expected-error {{invalid operands to binary expression ('v2u' (vector of 2 'unsigned int' values) and 'int')}} + v2ua = v2sa && s1; // expected-error {{logical expression with vector type 'v2s' (vector of 2 'int' values) and non-vector type 'int' is only supported in C++}} + v2ua = v2sa || s1; // expected-error {{logical expression with vector type 'v2s' (vector of 2 'int' values) and non-vector type 'int' is only supported in C++}} - v2ua = v2fa && s1; // expected-error {{cannot convert between vector values of different size ('v2f' (vector of 2 'float' values) and 'int')}} expected-error {{invalid operands to binary expression ('v2f' (vector of 2 'float' values) and 'int')}} - v2ua = v2fa || s1; // expected-error {{cannot convert between vector values of different size ('v2f' (vector of 2 'float' values) and 'int')}} expected-error {{invalid operands to binary expression ('v2f' (vector of 2 'float' values) and 'int')}} + v2sa = v2ua && s1; // expected-error {{logical expression with vector type 'v2u' (vector of 2 'unsigned int' values) and non-vector type 'int' is only supported in C++}} + v2sa = v2ua || s1; // expected-error {{logical expression with vector type 'v2u' (vector of 2 'unsigned int' values) and non-vector type 'int' is only supported in C++}} - v2sa = v2fa && s1; // expected-error {{cannot convert between vector values of different size ('v2f' (vector of 2 'float' values) and 'int')}} expected-error {{invalid operands to binary expression ('v2f' (vector of 2 'float' values) and 'int')}} - v2sa = v2fa || s1; // expected-error {{cannot convert between vector values of different size ('v2f' (vector of 2 'float' values) and 'int')}} expected-error {{invalid operands to binary expression ('v2f' (vector of 2 'float' values) and 'int')}} + v2ua = v2fa && s1; // expected-error {{cannot convert between scalar type 'int' and vector type 'v2f' (vector of 2 'float' values) as implicit conversion would cause truncation}} expected-error {{invalid operands to binary expression ('v2f' (vector of 2 'float' values) and 'int'}} + v2ua = v2fa || s1; // expected-error {{cannot convert between scalar type 'int' and vector type 'v2f' (vector of 2 'float' values) as implicit conversion would cause truncation}} expected-error {{invalid operands to binary expression ('v2f' (vector of 2 'float' values) and 'int'}} + + v2sa = v2fa && s1; // expected-error {{cannot convert between scalar type 'int' and vector type 'v2f' (vector of 2 'float' values) as implicit conversion would cause truncation}} expected-error {{invalid operands to binary expression ('v2f' (vector of 2 'float' values) and 'int'}} + v2sa = v2fa || s1; // expected-error {{cannot convert between scalar type 'int' and vector type 'v2f' (vector of 2 'float' values) as implicit conversion would cause truncation}} expected-error {{invalid operands to binary expression ('v2f' (vector of 2 'float' values) and 'int'}} float f1; - v2ua = v2ua && f1; // expected-error {{cannot convert between vector values of different size ('v2u' (vector of 2 'unsigned int' values) and 'float')}} expected-error {{invalid operands to binary expression ('v2u' (vector of 2 'unsigned int' values) and 'float')}} - v2ua = v2ua || f1; // expected-error {{cannot convert between vector values of different size ('v2u' (vector of 2 'unsigned int' values) and 'float')}} expected-error {{invalid operands to binary expression ('v2u' (vector of 2 'unsigned int' values) and 'float')}} + v2ua = v2ua && f1; // expected-error {{logical expression with vector type 'v2u' (vector of 2 'unsigned int' values) and non-vector type 'float' is only supported in C++}} + v2ua = v2ua || f1; // expected-error {{logical expression with vector type 'v2u' (vector of 2 'unsigned int' values) and non-vector type 'float' is only supported in C++}} - v2sa = v2sa && f1; // expected-error {{cannot convert between vector values of different size ('v2s' (vector of 2 'int' values) and 'float')}} expected-error {{invalid operands to binary expression ('v2s' (vector of 2 'int' values) and 'float')}} - v2sa = v2sa || f1; // expected-error {{cannot convert between vector values of different size ('v2s' (vector of 2 'int' values) and 'float')}} expected-error {{invalid operands to binary expression ('v2s' (vector of 2 'int' values) and 'float')}} + v2sa = v2sa && f1; // expected-error {{logical expression with vector type 'v2s' (vector of 2 'int' values) and non-vector type 'float' is only supported in C++}} + v2sa = v2sa || f1; // expected-error {{logical expression with vector type 'v2s' (vector of 2 'int' values) and non-vector type 'float' is only supported in C++}} - v2ua = v2sa && f1; // expected-error {{cannot convert between vector values of different size ('v2s' (vector of 2 'int' values) and 'float')}} expected-error {{invalid operands to binary expression ('v2s' (vector of 2 'int' values) and 'float')}} - v2ua = v2sa || f1; // expected-error {{cannot convert between vector values of different size ('v2s' (vector of 2 'int' values) and 'float')}} expected-error {{invalid operands to binary expression ('v2s' (vector of 2 'int' values) and 'float')}} - v2sa = v2ua && f1; // expected-error {{cannot convert between vector values of different size ('v2u' (vector of 2 'unsigned int' values) and 'float')}} expected-error {{invalid operands to binary expression ('v2u' (vector of 2 'unsigned int' values) and 'float')}} - v2sa = v2ua || f1; // expected-error {{cannot convert between vector values of different size ('v2u' (vector of 2 'unsigned int' values) and 'float')}} expected-error {{invalid operands to binary expression ('v2u' (vector of 2 'unsigned int' values) and 'float')}} + v2ua = v2sa && f1; // expected-error {{logical expression with vector type 'v2s' (vector of 2 'int' values) and non-vector type 'float' is only supported in C++}} + v2ua = v2sa || f1; // expected-error {{logical expression with vector type 'v2s' (vector of 2 'int' values) and non-vector type 'float' is only supported in C++}} - v2ua = v2fa && f1; // expected-error {{cannot convert between vector values of different size ('v2f' (vector of 2 'float' values) and 'float')}} expected-error {{invalid operands to binary expression ('v2f' (vector of 2 'float' values) and 'float')}} - v2ua = v2fa || f1; // expected-error {{cannot convert between vector values of different size ('v2f' (vector of 2 'float' values) and 'float')}} expected-error {{invalid operands to binary expression ('v2f' (vector of 2 'float' values) and 'float')}} + v2sa = v2ua && f1; // expected-error {{logical expression with vector type 'v2u' (vector of 2 'unsigned int' values) and non-vector type 'float' is only supported in C++}} + v2sa = v2ua || f1; // expected-error {{logical expression with vector type 'v2u' (vector of 2 'unsigned int' values) and non-vector type 'float' is only supported in C++}} - v2sa = v2fa && f1; // expected-error {{cannot convert between vector values of different size ('v2f' (vector of 2 'float' values) and 'float')}} expected-error {{invalid operands to binary expression ('v2f' (vector of 2 'float' values) and 'float')}} - v2sa = v2fa || f1; // expected-error {{cannot convert between vector values of different size ('v2f' (vector of 2 'float' values) and 'float')}} expected-error {{invalid operands to binary expression ('v2f' (vector of 2 'float' values) and 'float')}} + v2ua = v2fa && f1; // expected-error {{logical expression with vector type 'v2f' (vector of 2 'float' values) and non-vector type 'float' is only supported in C++}} + v2ua = v2fa || f1; // expected-error {{logical expression with vector type 'v2f' (vector of 2 'float' values) and non-vector type 'float' is only supported in C++}} + + v2sa = v2fa && f1; // expected-error {{logical expression with vector type 'v2f' (vector of 2 'float' values) and non-vector type 'float' is only supported in C++}} + v2sa = v2fa || f1; // expected-error {{logical expression with vector type 'v2f' (vector of 2 'float' values) and non-vector type 'float' is only supported in C++}} } diff --git a/test/Sema/zvector.c b/test/Sema/zvector.c index d1cf1aa01f4b..740163fcd9d5 100644 --- a/test/Sema/zvector.c +++ b/test/Sema/zvector.c @@ -326,14 +326,14 @@ void foo(void) bc = bc + sc2; // expected-error {{incompatible type}} bc = sc + bc2; // expected-error {{incompatible type}} - sc = sc + sc_scalar; // expected-error {{cannot convert}} - sc = sc + uc_scalar; // expected-error {{cannot convert}} - sc = sc_scalar + sc; // expected-error {{cannot convert}} - sc = uc_scalar + sc; // expected-error {{cannot convert}} - uc = uc + sc_scalar; // expected-error {{cannot convert}} - uc = uc + uc_scalar; // expected-error {{cannot convert}} - uc = sc_scalar + uc; // expected-error {{cannot convert}} - uc = uc_scalar + uc; // expected-error {{cannot convert}} + sc = sc + sc_scalar; + sc = sc + uc_scalar; // expected-error {{cannot convert between scalar type 'unsigned char' and vector type '__vector signed char' (vector of 16 'signed char' values) as implicit conversion would cause truncation}} + sc = sc_scalar + sc; + sc = uc_scalar + sc; // expected-error {{cannot convert between scalar type 'unsigned char' and vector type '__vector signed char' (vector of 16 'signed char' values) as implicit conversion would cause truncation}} + uc = uc + sc_scalar; // expected-error {{implicit conversion changes signedness: 'signed char' to '__vector unsigned char' (vector of 16 'unsigned char' values)}} + uc = uc + uc_scalar; + uc = sc_scalar + uc; // expected-error {{implicit conversion changes signedness: 'signed char' to '__vector unsigned char' (vector of 16 'unsigned char' values)}} + uc = uc_scalar + uc; ss = ss + ss2; us = us + us2; @@ -368,10 +368,10 @@ void foo(void) sc += sl2; // expected-error {{cannot convert}} sc += fd2; // expected-error {{cannot convert}} - sc += sc_scalar; // expected-error {{cannot convert}} - sc += uc_scalar; // expected-error {{cannot convert}} - uc += sc_scalar; // expected-error {{cannot convert}} - uc += uc_scalar; // expected-error {{cannot convert}} + sc += sc_scalar; + sc += uc_scalar; // expected-error {{cannot convert between scalar type 'unsigned char' and vector type '__vector signed char' (vector of 16 'signed char' values) as implicit conversion would cause truncation}} + uc += sc_scalar; // expected-error {{implicit conversion changes signedness: 'signed char' to '__vector unsigned char' (vector of 16 'unsigned char' values)}} + uc += uc_scalar; ss += ss2; us += us2; diff --git a/test/SemaCXX/constructor-initializer.cpp b/test/SemaCXX/constructor-initializer.cpp index c5de33cedb90..102ff1e80d03 100644 --- a/test/SemaCXX/constructor-initializer.cpp +++ b/test/SemaCXX/constructor-initializer.cpp @@ -302,3 +302,22 @@ namespace PR14073 { struct S2 { union { union { int n; }; char c; }; S2() : n(n) {} }; // expected-warning {{field 'n' is uninitialized when used here}} struct S3 { struct { int n; }; S3() : n(n) {} }; // expected-warning {{field 'n' is uninitialized when used here}} } + +namespace PR10758 { +struct A; +struct B { + B (A const &); // expected-note 2 {{candidate constructor not viable: no known conversion from 'const PR10758::B' to 'const PR10758::A &' for 1st argument}} + B (B &); // expected-note 2 {{candidate constructor not viable: 1st argument ('const PR10758::B') would lose const qualifier}} +}; +struct A { + A (B); // expected-note 2 {{passing argument to parameter here}} +}; + +B f(B const &b) { + return b; // expected-error {{no matching constructor for initialization of 'PR10758::B'}} +} + +A f2(const B &b) { + return b; // expected-error {{no matching constructor for initialization of 'PR10758::B'}} +} +} diff --git a/test/SemaCXX/cxx1y-generic-lambdas.cpp b/test/SemaCXX/cxx1y-generic-lambdas.cpp index 1993c6e1853d..9f3c77591a86 100644 --- a/test/SemaCXX/cxx1y-generic-lambdas.cpp +++ b/test/SemaCXX/cxx1y-generic-lambdas.cpp @@ -986,3 +986,10 @@ class Enclosing3 { ); }; } + +namespace PR32638 { + //https://bugs.llvm.org/show_bug.cgi?id=32638 + void test() { + [](auto x) noexcept(noexcept(x)) { } (0); + } +} \ No newline at end of file diff --git a/test/SemaCXX/cxx1y-variable-templates_top_level.cpp b/test/SemaCXX/cxx1y-variable-templates_top_level.cpp index b4963646838c..a78548b6f128 100644 --- a/test/SemaCXX/cxx1y-variable-templates_top_level.cpp +++ b/test/SemaCXX/cxx1y-variable-templates_top_level.cpp @@ -9,7 +9,7 @@ #endif template -T pi = T(3.1415926535897932385); // expected-note {{template is declared here}} +T pi = T(3.1415926535897932385); // expected-note 2{{declared here}} template CONST T cpi = T(3.1415926535897932385); // expected-note {{template is declared here}} @@ -58,10 +58,9 @@ namespace use_in_top_level_funcs { namespace shadow { void foo() { int ipi0 = pi; - int pi; + int pi; // expected-note {{found}} int a = pi; - int ipi = pi; // expected-error {{expected '(' for function-style cast or type construction}} \ - // expected-error {{expected expression}} + int ipi = pi; // expected-error {{'pi' does not name a template but is followed by template arguments; did you mean '::pi'?}} } } diff --git a/test/SemaCXX/enable_if.cpp b/test/SemaCXX/enable_if.cpp index 9a06d3866110..93014f50d508 100644 --- a/test/SemaCXX/enable_if.cpp +++ b/test/SemaCXX/enable_if.cpp @@ -499,3 +499,17 @@ void run() { } } } + +namespace TypeOfFn { + template + struct is_same; + + template struct is_same { + enum { value = 1 }; + }; + + void foo(int a) __attribute__((enable_if(a, ""))); + void foo(float a) __attribute__((enable_if(1, ""))); + + static_assert(is_same<__typeof__(foo)*, decltype(&foo)>::value, ""); +} diff --git a/test/SemaCXX/for-range-examples.cpp b/test/SemaCXX/for-range-examples.cpp index 08a9982c6378..d6b527ff8a50 100644 --- a/test/SemaCXX/for-range-examples.cpp +++ b/test/SemaCXX/for-range-examples.cpp @@ -241,3 +241,37 @@ namespace pr18587 { } } } + +namespace PR32933 { +// https://bugs.llvm.org/show_bug.cgi?id=32933 +void foo () +{ + int b = 1, a[b]; + a[0] = 0; + [&] { for (int c : a) 0; } (); +} + + +int foo(int b) { + int varr[b][(b+=8)]; + b = 15; + [&] { + int i = 0; + for (auto &c : varr) + { + c[0] = ++b; + } + [&] { + int i = 0; + for (auto &c : varr) { + int j = 0; + for(auto &c2 : c) { + ++j; + } + ++i; + } + }(); + }(); + return b; +} +} \ No newline at end of file diff --git a/test/SemaCXX/invalid-member-expr.cpp b/test/SemaCXX/invalid-member-expr.cpp index 172be6b8266d..fd50d328da67 100644 --- a/test/SemaCXX/invalid-member-expr.cpp +++ b/test/SemaCXX/invalid-member-expr.cpp @@ -53,9 +53,7 @@ namespace test3 { namespace rdar11293995 { struct Length { - explicit Length(PassRefPtr); // expected-error {{unknown type name}} \ - expected-error {{expected ')'}} \ - expected-note {{to match this '('}} + explicit Length(PassRefPtr); // expected-error {{no template named 'PassRefPtr}} expected-error {{undeclared identifier 'CalculationValue'}} }; struct LengthSize { diff --git a/test/SemaCXX/modules-ts.cppm b/test/SemaCXX/modules-ts.cppm index 16695f6463a8..d1d7aaa96e6c 100644 --- a/test/SemaCXX/modules-ts.cppm +++ b/test/SemaCXX/modules-ts.cppm @@ -17,7 +17,8 @@ static int m; // ok, internal linkage, so no redefinition error int n; #if TEST >= 2 // expected-error@-2 {{redefinition of '}} -// expected-note@-3 {{previous}} +// expected-note@-3 {{unguarded header; consider using #ifdef guards or #pragma once}} +// expected-note-re@modules-ts.cppm:1 {{'{{.*}}modules-ts.cppm' included multiple times, additional include site here}} #endif #if TEST == 0 diff --git a/test/SemaCXX/type-traits.cpp b/test/SemaCXX/type-traits.cpp index 9da59b93c503..919122576222 100644 --- a/test/SemaCXX/type-traits.cpp +++ b/test/SemaCXX/type-traits.cpp @@ -1256,7 +1256,7 @@ void is_trivially_copyable2() int t33[F(__is_trivially_copyable(ExtDefaulted))]; int t34[T(__is_trivially_copyable(const int))]; - int t35[F(__is_trivially_copyable(volatile int))]; + int t35[T(__is_trivially_copyable(volatile int))]; } struct CStruct { diff --git a/test/SemaCXX/typo-correction.cpp b/test/SemaCXX/typo-correction.cpp index c59ee618f929..2d78f06c5d33 100644 --- a/test/SemaCXX/typo-correction.cpp +++ b/test/SemaCXX/typo-correction.cpp @@ -524,13 +524,16 @@ namespace shadowed_template { template class Fizbin {}; // expected-note {{'::shadowed_template::Fizbin' declared here}} class Baz { int Fizbin(); - // TODO: Teach the parser to recover from the typo correction instead of - // continuing to treat the template name as an implicit-int declaration. - Fizbin qux; // expected-error {{unknown type name 'Fizbin'; did you mean '::shadowed_template::Fizbin'?}} \ - // expected-error {{expected member name or ';' after declaration specifiers}} + Fizbin qux; // expected-error {{no template named 'Fizbin'; did you mean '::shadowed_template::Fizbin'?}} }; } +namespace no_correct_template_id_to_non_template { + struct Frobnatz {}; // expected-note {{declared here}} + Frobnats fn; // expected-error {{unknown type name 'Frobnats'; did you mean 'Frobnatz'?}} + Frobnats fni; // expected-error-re {{no template named 'Frobnats'{{$}}}} +} + namespace PR18852 { void func() { struct foo { diff --git a/test/SemaCXX/vector-no-lax.cpp b/test/SemaCXX/vector-no-lax.cpp index a85f7f9db060..3cedcb1e8ce5 100644 --- a/test/SemaCXX/vector-no-lax.cpp +++ b/test/SemaCXX/vector-no-lax.cpp @@ -4,6 +4,6 @@ typedef int __attribute__((vector_size (16))) vSInt32; vSInt32 foo (vUInt32 a) { vSInt32 b = { 0, 0, 0, 0 }; - b += a; // expected-error{{cannot convert between vector values}} + b += a; // expected-error{{cannot convert between vector type 'vUInt32' (vector of 4 'unsigned int' values) and vector type 'vSInt32' (vector of 4 'int' values) as implicit conversion would cause truncation}} return b; } diff --git a/test/SemaCXX/warn-unused-filescoped.cpp b/test/SemaCXX/warn-unused-filescoped.cpp index 18defee7d04a..93c6bbd7edc9 100644 --- a/test/SemaCXX/warn-unused-filescoped.cpp +++ b/test/SemaCXX/warn-unused-filescoped.cpp @@ -1,5 +1,5 @@ -// RUN: %clang_cc1 -fsyntax-only -verify -Wunused -Wunused-member-function -Wno-unused-local-typedefs -Wno-c++11-extensions -std=c++98 %s -// RUN: %clang_cc1 -fsyntax-only -verify -Wunused -Wunused-member-function -Wno-unused-local-typedefs -std=c++11 %s +// RUN: %clang_cc1 -fsyntax-only -verify -Wunused -Wunused-template -Wunused-member-function -Wno-unused-local-typedefs -Wno-c++11-extensions -std=c++98 %s +// RUN: %clang_cc1 -fsyntax-only -verify -Wunused -Wunused-template -Wunused-member-function -Wno-unused-local-typedefs -std=c++14 %s #ifdef HEADER @@ -65,7 +65,7 @@ namespace { template <> void TS::m() { } // expected-warning{{unused}} template - void tf() { } + void tf() { } // expected-warning{{unused}} template <> void tf() { } // expected-warning{{unused}} struct VS { @@ -200,6 +200,18 @@ void bar() { void func() __attribute__((used)); } static void func() {} } +namespace test9 { +template +static void completeRedeclChainForTemplateSpecialization() { } // expected-warning {{unused}} +} + +namespace test10 { +#if __cplusplus >= 201103L +template +constexpr T pi = T(3.14); // expected-warning {{unused}} +#endif +} + namespace pr19713 { #if __cplusplus >= 201103L // FIXME: We should warn on both of these. diff --git a/test/SemaObjC/method-bad-param.m b/test/SemaObjC/method-bad-param.m index ad67a34edb00..a7f0745ddbad 100644 --- a/test/SemaObjC/method-bad-param.m +++ b/test/SemaObjC/method-bad-param.m @@ -20,6 +20,12 @@ } @end +// Ensure that this function is properly marked as a failure. +void func_with_bad_call(bar* b, foo* f) { + [b cccccc:5]; // expected-warning {{instance method '-cccccc:' not found}} + // expected-note@-17 {{receiver is instance of class declared here}} +} + void somefunc(foo x) {} // expected-error {{interface type 'foo' cannot be passed by value; did you forget * in 'foo'}} foo somefunc2() {} // expected-error {{interface type 'foo' cannot be returned by value; did you forget * in 'foo'}} diff --git a/test/SemaObjC/unguarded-availability.m b/test/SemaObjC/unguarded-availability.m index ae921f4a27b3..071a21ea1b3f 100644 --- a/test/SemaObjC/unguarded-availability.m +++ b/test/SemaObjC/unguarded-availability.m @@ -8,7 +8,7 @@ int func_10_11() AVAILABLE_10_11; // expected-note 4 {{'func_10_11' has been explicitly marked partial here}} #ifdef OBJCPP -// expected-note@+2 2 {{marked partial here}} +// expected-note@+2 6 {{marked partial here}} #endif int func_10_12() AVAILABLE_10_12; // expected-note 6 {{'func_10_12' has been explicitly marked partial here}} @@ -48,7 +48,7 @@ void star_case() { } else func_10_11(); // expected-warning{{'func_10_11' is only available on macOS 10.11 or newer}} expected-note{{enclose 'func_10_11' in an @available check to silence this warning}} - if (@available(macos 10.11, *)) { + if (@available(macOS 10.11, *)) { if (@available(ios 8, *)) { func_10_11(); func_10_12(); // expected-warning{{'func_10_12' is only available on macOS 10.12 or newer}} expected-note{{enclose}} @@ -176,7 +176,7 @@ int instantiate_with_availability_attr() { } int instantiate_availability() { - if (@available(macos 10.12, *)) + if (@available(macOS 10.12, *)) with_availability_attr(); else with_availability_attr(); // expected-warning{{'with_availability_attr' is only available on macOS 10.11 or newer}} expected-warning{{'int_10_12' is only available on macOS 10.12 or newer}} expected-note 2 {{enclose}} @@ -188,4 +188,19 @@ auto topLevelLambda = [] () { func_10_12(); }; +void functionInFunction() { + func_10_12(); // expected-warning{{'func_10_12' is only available on macOS 10.12 or newer}} expected-note{{@available}} + struct DontWarnTwice { + void f() { + func_10_12(); // expected-warning{{'func_10_12' is only available on macOS 10.12 or newer}} expected-note{{@available}} + } + }; + void([] () { + func_10_12(); // expected-warning{{'func_10_12' is only available on macOS 10.12 or newer}} expected-note{{@available}} + }); + (void)(^ { + func_10_12(); // expected-warning{{'func_10_12' is only available on macOS 10.12 or newer}} expected-note{{@available}} + }); +} + #endif diff --git a/test/SemaObjCXX/interface-return-type.mm b/test/SemaObjCXX/interface-return-type.mm new file mode 100644 index 000000000000..9fff8610ae0d --- /dev/null +++ b/test/SemaObjCXX/interface-return-type.mm @@ -0,0 +1,7 @@ +// RUN: %clang_cc1 -fsyntax-only -std=c++11 %s -verify + +@class NSObject; +template struct C { + static T f(); // expected-error {{interface type 'NSObject' cannot be returned by value; did you forget * in 'NSObject'?}} +}; +int g() { NSObject *x = C::f(); }//expected-error {{no member named 'f' in 'C'}} expected-note {{in instantiation of template class 'C' requested here}} diff --git a/test/SemaObjCXX/is-base-of.mm b/test/SemaObjCXX/is-base-of.mm new file mode 100644 index 000000000000..9cf16661b0c5 --- /dev/null +++ b/test/SemaObjCXX/is-base-of.mm @@ -0,0 +1,25 @@ +// RUN: %clang_cc1 -std=c++11 -fsyntax-only -verify %s + +@interface NSObj +@end + +@interface NSChild : NSObj +@end + +static_assert(__is_base_of(NSObj, NSChild), ""); +static_assert(!__is_base_of(NSChild, NSObj), ""); + +static_assert(__is_base_of(NSObj, NSObj), ""); + +static_assert(!__is_base_of(NSObj *, NSChild *), ""); +static_assert(!__is_base_of(NSChild *, NSObj *), ""); + +static_assert(__is_base_of(const volatile NSObj, NSChild), ""); +static_assert(__is_base_of(NSObj, const volatile NSChild), ""); + +@class NSForward; // expected-note{{forward declaration of class}} + +static_assert(!__is_base_of(NSForward, NSObj), ""); +static_assert(!__is_base_of(NSObj, NSForward), ""); // expected-error{{incomplete type 'NSForward'}} + +static_assert(!__is_base_of(id, NSObj), ""); diff --git a/test/SemaOpenCL/array-init.cl b/test/SemaOpenCL/array-init.cl new file mode 100644 index 000000000000..d9691d86dd81 --- /dev/null +++ b/test/SemaOpenCL/array-init.cl @@ -0,0 +1,20 @@ +// RUN: %clang_cc1 %s -verify -pedantic -fsyntax-only -cl-std=CL2.0 +// expected-no-diagnostics + +__kernel void k1(queue_t q1, queue_t q2) { + queue_t q[] = {q1, q2}; +} + +__kernel void k2(read_only pipe int p) { + reserve_id_t i1 = reserve_read_pipe(p, 1); + reserve_id_t i2 = reserve_read_pipe(p, 1); + reserve_id_t i[] = {i1, i2}; +} + +event_t create_event(); +__kernel void k3() { + event_t e1 = create_event(); + event_t e2 = create_event(); + event_t e[] = {e1, e2}; +} + diff --git a/test/SemaOpenCL/storageclass.cl b/test/SemaOpenCL/storageclass.cl index a93f8244dcbd..f457cfd1d3f6 100644 --- a/test/SemaOpenCL/storageclass.cl +++ b/test/SemaOpenCL/storageclass.cl @@ -5,7 +5,7 @@ constant int G2 = 0; int G3 = 0; // expected-error{{program scope variable must reside in constant address space}} global int G4 = 0; // expected-error{{program scope variable must reside in constant address space}} -void kernel foo() { +void kernel foo(int x) { // static is not allowed at local scope before CL2.0 static int S1 = 5; // expected-error{{variables in function scope cannot be declared static}} static constant int S2 = 5; // expected-error{{variables in function scope cannot be declared static}} @@ -15,6 +15,12 @@ void kernel foo() { auto int L3 = 7; // expected-error{{OpenCL version 1.2 does not support the 'auto' storage class specifier}} global int L4; // expected-error{{function scope variable cannot be declared in global address space}} + + constant int L5 = x; // expected-error {{initializer element is not a compile-time constant}} + global int *constant L6 = &G4; + private int *constant L7 = &x; // expected-error {{initializer element is not a compile-time constant}} + constant int *constant L8 = &L1; + local int *constant L9 = &L2; // expected-error {{initializer element is not a compile-time constant}} } static void kernel bar() { // expected-error{{kernel functions cannot be declared static}} @@ -29,4 +35,7 @@ void f() { } global int L3; // expected-error{{function scope variable cannot be declared in global address space}} extern constant float L4; + extern local float L5; // expected-error{{extern variable must reside in constant address space}} + static int L6 = 0; // expected-error{{variables in function scope cannot be declared static}} + static int L7; // expected-error{{variables in function scope cannot be declared static}} } diff --git a/test/SemaTemplate/deduction-crash.cpp b/test/SemaTemplate/deduction-crash.cpp index ff7421a910bd..c94c9db94e06 100644 --- a/test/SemaTemplate/deduction-crash.cpp +++ b/test/SemaTemplate/deduction-crash.cpp @@ -2,7 +2,7 @@ // Note that the error count below doesn't matter. We just want to // make sure that the parser doesn't crash. -// CHECK: 16 errors +// CHECK: 17 errors // PR7511 template diff --git a/test/SemaTemplate/default-arguments.cpp b/test/SemaTemplate/default-arguments.cpp index d3e249db7ee2..b5b042c64a71 100644 --- a/test/SemaTemplate/default-arguments.cpp +++ b/test/SemaTemplate/default-arguments.cpp @@ -207,3 +207,19 @@ Y y2; } // end ns1 } // end ns PR26134 + +namespace friends { + namespace ns { + template struct A { + template friend void f(); + template friend struct X; + }; + template void f(); // expected-warning 0-1{{extension}} + template struct X; + A a; + } + namespace ns { + void g() { f(); } + X *p; + } +} diff --git a/test/SemaTemplate/explicit-instantiation.cpp b/test/SemaTemplate/explicit-instantiation.cpp index 010716dd1426..42d9f4d332a9 100644 --- a/test/SemaTemplate/explicit-instantiation.cpp +++ b/test/SemaTemplate/explicit-instantiation.cpp @@ -95,7 +95,7 @@ namespace PR7622 { struct basic_streambuf; template - struct basic_streambuf{friend bob<>()}; // expected-error{{unknown type name 'bob'}} \ + struct basic_streambuf{friend bob<>()}; // expected-error{{no template named 'bob'}} \ // expected-error{{expected member name or ';' after declaration specifiers}} template struct basic_streambuf; } diff --git a/test/SemaTemplate/explicit-specialization-member.cpp b/test/SemaTemplate/explicit-specialization-member.cpp index f302836c7e4b..c0c36808b492 100644 --- a/test/SemaTemplate/explicit-specialization-member.cpp +++ b/test/SemaTemplate/explicit-specialization-member.cpp @@ -1,4 +1,4 @@ -// RUN: %clang_cc1 -fsyntax-only -verify %s +// RUN: %clang_cc1 -fsyntax-only -verify %s -fcxx-exceptions template struct X0 { typedef T* type; @@ -57,3 +57,12 @@ template struct Helper { template void Helper::func<2>() {} // expected-error {{cannot specialize a member}} \ // expected-error {{no function template matches}} } + +namespace SpecLoc { + template struct A { + static int n; // expected-note {{previous}} + static void f(); // expected-note {{previous}} + }; + template<> float A::n; // expected-error {{different type}} + template<> void A::f() throw(); // expected-error {{does not match}} +} diff --git a/test/SemaTemplate/ms-lookup-template-base-classes.cpp b/test/SemaTemplate/ms-lookup-template-base-classes.cpp index 6afc7091260d..a41248ee1b8e 100644 --- a/test/SemaTemplate/ms-lookup-template-base-classes.cpp +++ b/test/SemaTemplate/ms-lookup-template-base-classes.cpp @@ -347,8 +347,7 @@ template struct B : A { }; template struct C : A { // Incorrect form. - NameFromBase m; // expected-error {{unknown type name 'NameFromBase'}} - //expected-error@-1 {{expected member name or ';' after declaration specifiers}} + NameFromBase m; // expected-error {{no template named 'NameFromBase'}} }; } diff --git a/test/SemaTemplate/typo-template-name.cpp b/test/SemaTemplate/typo-template-name.cpp new file mode 100644 index 000000000000..fe5201a8e26c --- /dev/null +++ b/test/SemaTemplate/typo-template-name.cpp @@ -0,0 +1,43 @@ +// RUN: %clang_cc1 -std=c++1z %s -verify -Wno-unused + +namespace InExpr { + namespace A { + void typo_first_a(); // expected-note {{found}} + template void typo_first_b(); // expected-note 2{{declared here}} + } + void testA() { A::typo_first_a(); } // expected-error {{'typo_first_a' does not name a template but is followed by template arguments; did you mean 'typo_first_b'?}} + + namespace B { + void typo_first_b(); // expected-note {{found}} + } + void testB() { B::typo_first_b(); } // expected-error {{'typo_first_b' does not name a template but is followed by template arguments; did you mean 'A::typo_first_b'?}} + + struct Base { + template static void foo(); // expected-note 4{{declared here}} + int n; + }; + struct Derived : Base { + void foo(); // expected-note {{found}} + }; + // We probably don't want to suggest correcting to .Base::foo + void testMember() { Derived().foo(); } // expected-error-re {{does not name a template but is followed by template arguments{{$}}}} + + struct Derived2 : Base { + void goo(); // expected-note {{found}} + }; + void testMember2() { Derived2().goo(); } // expected-error {{member 'goo' of 'InExpr::Derived2' is not a template; did you mean 'foo'?}} + + void no_correction() { + int foo; // expected-note 3{{found}} + + foo(); // expected-error {{'foo' does not name a template but is followed by template arguments; did you mean 'Base::foo'?}} + foo<>(); // expected-error {{'foo' does not name a template but is followed by template arguments; did you mean 'Base::foo'?}} + foo(); // expected-error {{'foo' does not name a template but is followed by template arguments; did you mean 'Base::foo'?}} + + // These are valid expressions. + foo(0); + foo(false); + foo Imports("import", llvm::cl::ZeroOrMore, llvm::cl::desc("Path to a file containing declarations to import")); +static llvm::cl::opt + Direct("direct", llvm::cl::Optional, + llvm::cl::desc("Use the parsed declarations without indirection")); + static llvm::cl::list ClangArgs("Xcc", llvm::cl::ZeroOrMore, llvm::cl::desc("Argument to pass to the CompilerInvocation"), @@ -172,6 +176,14 @@ BuildCompilerInstance(ArrayRef ClangArgv) { return Ins; } +std::unique_ptr +BuildCompilerInstance(ArrayRef ClangArgs) { + std::vector ClangArgv(ClangArgs.size()); + std::transform(ClangArgs.begin(), ClangArgs.end(), ClangArgv.begin(), + [](const std::string &s) -> const char * { return s.data(); }); + return init_convenience::BuildCompilerInstance(ClangArgv); +} + std::unique_ptr BuildASTContext(CompilerInstance &CI, SelectorTable &ST, Builtin::Context &BC) { auto AST = llvm::make_unique( @@ -205,6 +217,21 @@ void AddExternalSource( CI.getASTContext().getTranslationUnitDecl()->setHasExternalVisibleStorage(); } +std::unique_ptr BuildIndirect(std::unique_ptr &CI) { + std::vector ClangArgv(ClangArgs.size()); + std::transform(ClangArgs.begin(), ClangArgs.end(), ClangArgv.begin(), + [](const std::string &s) -> const char * { return s.data(); }); + std::unique_ptr IndirectCI = + init_convenience::BuildCompilerInstance(ClangArgv); + auto ST = llvm::make_unique(); + auto BC = llvm::make_unique(); + std::unique_ptr AST = + init_convenience::BuildASTContext(*IndirectCI, *ST, *BC); + IndirectCI->setASTContext(AST.release()); + AddExternalSource(*IndirectCI, CI); + return IndirectCI; +} + llvm::Error ParseSource(const std::string &Path, CompilerInstance &CI, CodeGenerator &CG) { SourceManager &SM = CI.getSourceManager(); @@ -231,7 +258,8 @@ Parse(const std::string &Path, std::unique_ptr AST = init_convenience::BuildASTContext(*CI, *ST, *BC); CI->setASTContext(AST.release()); - AddExternalSource(*CI, Imports); + if (Imports.size()) + AddExternalSource(*CI, Imports); auto LLVMCtx = llvm::make_unique(); std::unique_ptr CG = @@ -268,8 +296,21 @@ int main(int argc, const char **argv) { ImportCIs.push_back(std::move(*ImportCI)); } } + std::vector> IndirectCIs; + if (!Direct) { + for (auto &ImportCI : ImportCIs) { + llvm::Expected> IndirectCI = + BuildIndirect(ImportCI); + if (auto E = IndirectCI.takeError()) { + llvm::errs() << llvm::toString(std::move(E)); + exit(-1); + } else { + IndirectCIs.push_back(std::move(*IndirectCI)); + } + } + } llvm::Expected> ExpressionCI = - Parse(Expression, ImportCIs); + Parse(Expression, Direct ? ImportCIs : IndirectCIs); if (auto E = ExpressionCI.takeError()) { llvm::errs() << llvm::toString(std::move(E)); exit(-1); @@ -277,3 +318,4 @@ int main(int argc, const char **argv) { return 0; } } + diff --git a/tools/libclang/CIndex.cpp b/tools/libclang/CIndex.cpp index c251d83e2097..9c795ae9c5b7 100644 --- a/tools/libclang/CIndex.cpp +++ b/tools/libclang/CIndex.cpp @@ -7479,6 +7479,35 @@ unsigned clang_Cursor_isVariadic(CXCursor C) { return 0; } +unsigned clang_Cursor_isExternalSymbol(CXCursor C, + CXString *language, CXString *definedIn, + unsigned *isGenerated) { + if (!clang_isDeclaration(C.kind)) + return 0; + + const Decl *D = getCursorDecl(C); + + auto getExternalSymAttr = [](const Decl *D) -> ExternalSourceSymbolAttr* { + if (auto *attr = D->getAttr()) + return attr; + if (auto *dcd = dyn_cast(D->getDeclContext())) { + if (auto *attr = dcd->getAttr()) + return attr; + } + return nullptr; + }; + if (auto *attr = getExternalSymAttr(D)) { + if (language) + *language = cxstring::createDup(attr->getLanguage()); + if (definedIn) + *definedIn = cxstring::createDup(attr->getDefinedIn()); + if (isGenerated) + *isGenerated = attr->getGeneratedDeclaration(); + return 1; + } + return 0; +} + CXSourceRange clang_Cursor_getCommentRange(CXCursor C) { if (!clang_isDeclaration(C.kind)) return clang_getNullRange(); diff --git a/tools/libclang/libclang.exports b/tools/libclang/libclang.exports index 895dd804b008..d9a406e5741b 100644 --- a/tools/libclang/libclang.exports +++ b/tools/libclang/libclang.exports @@ -35,6 +35,7 @@ clang_Cursor_getReceiverType clang_Cursor_isAnonymous clang_Cursor_isBitField clang_Cursor_isDynamicCall +clang_Cursor_isExternalSymbol clang_Cursor_isNull clang_Cursor_isObjCOptional clang_Cursor_isVariadic diff --git a/unittests/Format/FormatTest.cpp b/unittests/Format/FormatTest.cpp index 9b833a96f16a..076041406c87 100644 --- a/unittests/Format/FormatTest.cpp +++ b/unittests/Format/FormatTest.cpp @@ -342,7 +342,7 @@ TEST_F(FormatTest, FormatIfWithoutCompoundStatement) { verifyFormat("if (a)\n if (b) {\n f();\n }\ng();"); FormatStyle AllowsMergedIf = getLLVMStyle(); - AllowsMergedIf.AlignEscapedNewlinesLeft = true; + AllowsMergedIf.AlignEscapedNewlines = FormatStyle::ENAS_Left; AllowsMergedIf.AllowShortIfStatementsOnASingleLine = true; verifyFormat("if (a)\n" " // comment\n" @@ -2106,7 +2106,7 @@ TEST_F(FormatTest, LayoutStatementsAroundPreprocessorDirectives) { verifyIncompleteFormat("void f(\n" "#if A\n" - " );\n" + ");\n" "#else\n" "#endif"); } @@ -2591,6 +2591,60 @@ TEST_F(FormatTest, BreakingBeforeNonAssigmentOperators) { Style); } +TEST_F(FormatTest, AllowBinPackingInsideArguments) { + FormatStyle Style = getLLVMStyle(); + Style.BreakBeforeBinaryOperators = FormatStyle::BOS_NonAssignment; + Style.BinPackArguments = false; + Style.ColumnLimit = 40; + verifyFormat("void test() {\n" + " someFunction(\n" + " this + argument + is + quite\n" + " + long + so + it + gets + wrapped\n" + " + but + remains + bin - packed);\n" + "}", + Style); + verifyFormat("void test() {\n" + " someFunction(arg1,\n" + " this + argument + is\n" + " + quite + long + so\n" + " + it + gets + wrapped\n" + " + but + remains + bin\n" + " - packed,\n" + " arg3);\n" + "}", + Style); + verifyFormat("void test() {\n" + " someFunction(\n" + " arg1,\n" + " this + argument + has\n" + " + anotherFunc(nested,\n" + " calls + whose\n" + " + arguments\n" + " + are + also\n" + " + wrapped,\n" + " in + addition)\n" + " + to + being + bin - packed,\n" + " arg3);\n" + "}", + Style); + + Style.BreakBeforeBinaryOperators = FormatStyle::BOS_None; + verifyFormat("void test() {\n" + " someFunction(\n" + " arg1,\n" + " this + argument + has +\n" + " anotherFunc(nested,\n" + " calls + whose +\n" + " arguments +\n" + " are + also +\n" + " wrapped,\n" + " in + addition) +\n" + " to + being + bin - packed,\n" + " arg3);\n" + "}", + Style); +} + TEST_F(FormatTest, ConstructorInitializers) { verifyFormat("Constructor() : Initializer(FitsOnTheLine) {}"); verifyFormat("Constructor() : Inttializer(FitsOnTheLine) {}", @@ -4421,7 +4475,7 @@ TEST_F(FormatTest, WrapsTemplateDeclarations) { EXPECT_EQ("static_cast *>(\n" "\n" - " );", + ");", format("static_cast*>(\n" "\n" @@ -6423,7 +6477,7 @@ TEST_F(FormatTest, BreaksStringLiterals) { EXPECT_EQ("\"some text other\";", format("\"some text other\";", Style)); FormatStyle AlignLeft = getLLVMStyleWithColumns(12); - AlignLeft.AlignEscapedNewlinesLeft = true; + AlignLeft.AlignEscapedNewlines = FormatStyle::ENAS_Left; EXPECT_EQ("#define A \\\n" " \"some \" \\\n" " \"text \" \\\n" @@ -6513,7 +6567,7 @@ TEST_F(FormatTest, BreaksStringLiteralsWithin_TMacro) { "#if !TEST\n" " _T(\"XXXXXXXXXXXXXXXXXXXXXXXXXXXXXn\")\n" "#endif\n" - " );", + ");", format("f(\n" "#if !TEST\n" "_T(\"XXXXXXXXXXXXXXXXXXXXXXXXXXXXXn\")\n" @@ -6824,7 +6878,7 @@ TEST_F(FormatTest, ConfigurableUseOfTab) { FormatStyle Tab = getLLVMStyleWithColumns(42); Tab.IndentWidth = 8; Tab.UseTab = FormatStyle::UT_Always; - Tab.AlignEscapedNewlinesLeft = true; + Tab.AlignEscapedNewlines = FormatStyle::ENAS_Left; EXPECT_EQ("if (aaaaaaaa && // q\n" " bb)\t\t// w\n" @@ -7605,14 +7659,21 @@ TEST_F(FormatTest, AlignConsecutiveAssignments) { "int oneTwoThree = 123;\n" "int oneTwo = 12;", Alignment)); - Alignment.AlignEscapedNewlinesLeft = true; + Alignment.AlignEscapedNewlines = FormatStyle::ENAS_DontAlign; + verifyFormat("#define A \\\n" + " int aaaa = 12; \\\n" + " int b = 23; \\\n" + " int ccc = 234; \\\n" + " int dddddddddd = 2345;", + Alignment); + Alignment.AlignEscapedNewlines = FormatStyle::ENAS_Left; verifyFormat("#define A \\\n" " int aaaa = 12; \\\n" " int b = 23; \\\n" " int ccc = 234; \\\n" " int dddddddddd = 2345;", Alignment); - Alignment.AlignEscapedNewlinesLeft = false; + Alignment.AlignEscapedNewlines = FormatStyle::ENAS_Right; verifyFormat("#define A " " \\\n" " int aaaa = 12; " @@ -7879,14 +7940,21 @@ TEST_F(FormatTest, AlignConsecutiveDeclarations) { "}", Alignment)); Alignment.AlignConsecutiveAssignments = false; - Alignment.AlignEscapedNewlinesLeft = true; + Alignment.AlignEscapedNewlines = FormatStyle::ENAS_DontAlign; + verifyFormat("#define A \\\n" + " int aaaa = 12; \\\n" + " float b = 23; \\\n" + " const int ccc = 234; \\\n" + " unsigned dddddddddd = 2345;", + Alignment); + Alignment.AlignEscapedNewlines = FormatStyle::ENAS_Left; verifyFormat("#define A \\\n" " int aaaa = 12; \\\n" " float b = 23; \\\n" " const int ccc = 234; \\\n" " unsigned dddddddddd = 2345;", Alignment); - Alignment.AlignEscapedNewlinesLeft = false; + Alignment.AlignEscapedNewlines = FormatStyle::ENAS_Right; Alignment.ColumnLimit = 30; verifyFormat("#define A \\\n" " int aaaa = 12; \\\n" @@ -8671,7 +8739,6 @@ TEST_F(FormatTest, GetsCorrectBasedOnStyle) { TEST_F(FormatTest, ParsesConfigurationBools) { FormatStyle Style = {}; Style.Language = FormatStyle::LK_Cpp; - CHECK_PARSE_BOOL(AlignEscapedNewlinesLeft); CHECK_PARSE_BOOL(AlignOperands); CHECK_PARSE_BOOL(AlignTrailingComments); CHECK_PARSE_BOOL(AlignConsecutiveAssignments); @@ -8794,6 +8861,19 @@ TEST_F(FormatTest, ParsesConfiguration) { CHECK_PARSE("AlignAfterOpenBracket: true", AlignAfterOpenBracket, FormatStyle::BAS_Align); + Style.AlignEscapedNewlines = FormatStyle::ENAS_Left; + CHECK_PARSE("AlignEscapedNewlines: DontAlign", AlignEscapedNewlines, + FormatStyle::ENAS_DontAlign); + CHECK_PARSE("AlignEscapedNewlines: Left", AlignEscapedNewlines, + FormatStyle::ENAS_Left); + CHECK_PARSE("AlignEscapedNewlines: Right", AlignEscapedNewlines, + FormatStyle::ENAS_Right); + // For backward compatibility: + CHECK_PARSE("AlignEscapedNewlinesLeft: true", AlignEscapedNewlines, + FormatStyle::ENAS_Left); + CHECK_PARSE("AlignEscapedNewlinesLeft: false", AlignEscapedNewlines, + FormatStyle::ENAS_Right); + Style.UseTab = FormatStyle::UT_ForIndentation; CHECK_PARSE("UseTab: Never", UseTab, FormatStyle::UT_Never); CHECK_PARSE("UseTab: ForIndentation", UseTab, FormatStyle::UT_ForIndentation); @@ -9559,7 +9639,7 @@ TEST_F(FormatTest, FormatsLambdas) { // Other corner cases. verifyFormat("void f() {\n" " bar([]() {} // Did not respect SpacesBeforeTrailingComments\n" - " );\n" + " );\n" "}"); // Lambdas created through weird macros. diff --git a/unittests/Format/FormatTestJS.cpp b/unittests/Format/FormatTestJS.cpp index 9144fe17e9ec..ff28f304be99 100644 --- a/unittests/Format/FormatTestJS.cpp +++ b/unittests/Format/FormatTestJS.cpp @@ -367,6 +367,25 @@ TEST_F(FormatTestJS, GoogScopes) { "});"); } +TEST_F(FormatTestJS, IIFEs) { + // Internal calling parens; no semi. + verifyFormat("(function() {\n" + "var a = 1;\n" + "}())"); + // External calling parens; no semi. + verifyFormat("(function() {\n" + "var b = 2;\n" + "})()"); + // Internal calling parens; with semi. + verifyFormat("(function() {\n" + "var c = 3;\n" + "}());"); + // External calling parens; with semi. + verifyFormat("(function() {\n" + "var d = 4;\n" + "})();"); +} + TEST_F(FormatTestJS, GoogModules) { verifyFormat("goog.module('this.is.really.absurdly.long');", getGoogleJSStyleWithColumns(40)); @@ -451,6 +470,16 @@ TEST_F(FormatTestJS, FormatsFreestandingFunctions) { " inner2(a, b);\n" "}"); verifyFormat("function f() {}"); + verifyFormat("function aFunction() {}\n" + "(function f() {\n" + " var x = 1;\n" + "}());\n"); + // Known issue: this should wrap after {}, but calculateBraceTypes + // misclassifies the first braces as a BK_BracedInit. + verifyFormat("function aFunction(){} {\n" + " let x = 1;\n" + " console.log(x);\n" + "}\n"); } TEST_F(FormatTestJS, GeneratorFunctions) { @@ -519,6 +548,39 @@ TEST_F(FormatTestJS, AsyncFunctions) { " // Comment.\n" " return async.then();\n" "}\n"); + verifyFormat("for async (const x of y) {\n" + " console.log(x);\n" + "}\n"); + verifyFormat("function asyncLoop() {\n" + " for async (const x of y) {\n" + " console.log(x);\n" + " }\n" + "}\n"); + +} + +TEST_F(FormatTestJS, FunctionParametersTrailingComma) { + verifyFormat("function trailingComma(\n" + " p1,\n" + " p2,\n" + " p3,\n" + ") {\n" + " a; //\n" + "}\n", + "function trailingComma(p1, p2, p3,) {\n" + " a; //\n" + "}\n"); + verifyFormat("trailingComma(\n" + " p1,\n" + " p2,\n" + " p3,\n" + ");\n", + "trailingComma(p1, p2, p3,);\n"); + verifyFormat("trailingComma(\n" + " p1 // hello\n" + ");\n", + "trailingComma(p1 // hello\n" + ");\n"); } TEST_F(FormatTestJS, ArrayLiterals) { @@ -662,7 +724,7 @@ TEST_F(FormatTestJS, FunctionLiterals) { "})\n" " .doSomethingElse(\n" " // break\n" - " );"); + " );"); Style.ColumnLimit = 33; verifyFormat("f({a: function() { return 1; }});", Style); @@ -829,7 +891,7 @@ TEST_F(FormatTestJS, ArrowFunctions) { "})\n" " .doSomethingElse(\n" " // break\n" - " );"); + " );"); } TEST_F(FormatTestJS, ReturnStatements) { @@ -1770,6 +1832,8 @@ TEST_F(FormatTestJS, NonNullAssertionOperator) { " .foo()!\n" " .foo()!;\n", getGoogleJSStyleWithColumns(20)); + verifyFormat("let x = namespace!;\n"); + verifyFormat("return !!x;\n"); } TEST_F(FormatTestJS, Conditional) { @@ -1785,6 +1849,7 @@ TEST_F(FormatTestJS, ImportComments) { verifyFormat("import {x} from 'x'; // from some location", getGoogleJSStyleWithColumns(25)); verifyFormat("// taze: x from 'location'", getGoogleJSStyleWithColumns(10)); + verifyFormat("/// ", getGoogleJSStyleWithColumns(10)); } TEST_F(FormatTestJS, Exponentiation) { diff --git a/unittests/Format/FormatTestSelective.cpp b/unittests/Format/FormatTestSelective.cpp index 8046d7fab2dd..9ee87b34c217 100644 --- a/unittests/Format/FormatTestSelective.cpp +++ b/unittests/Format/FormatTestSelective.cpp @@ -325,7 +325,7 @@ TEST_F(FormatTestSelective, WrongIndent) { } TEST_F(FormatTestSelective, AlwaysFormatsEntireMacroDefinitions) { - Style.AlignEscapedNewlinesLeft = true; + Style.AlignEscapedNewlines = FormatStyle::ENAS_Left; EXPECT_EQ("int i;\n" "#define A \\\n" " int i; \\\n" @@ -467,7 +467,7 @@ TEST_F(FormatTestSelective, ReformatRegionAdjustsIndent) { TEST_F(FormatTestSelective, UnderstandsTabs) { Style.IndentWidth = 8; Style.UseTab = FormatStyle::UT_Always; - Style.AlignEscapedNewlinesLeft = true; + Style.AlignEscapedNewlines = FormatStyle::ENAS_Left; EXPECT_EQ("void f() {\n" "\tf();\n" "\tg();\n" diff --git a/unittests/Tooling/RecursiveASTVisitorTest.cpp b/unittests/Tooling/RecursiveASTVisitorTest.cpp index 7e08f9619c1c..269bdbb34ab1 100644 --- a/unittests/Tooling/RecursiveASTVisitorTest.cpp +++ b/unittests/Tooling/RecursiveASTVisitorTest.cpp @@ -52,6 +52,14 @@ TEST(RecursiveASTVisitor, TraverseLambdaBodyCanBeOverridden) { EXPECT_TRUE(Visitor.allBodiesHaveBeenTraversed()); } +TEST(RecursiveASTVisitor, VisitsAttributedLambdaExpr) { + LambdaExprVisitor Visitor; + Visitor.ExpectMatch("", 1, 12); + EXPECT_TRUE(Visitor.runOver( + "void f() { [] () __attribute__ (( fastcall )) { return; }(); }", + LambdaExprVisitor::Lang_CXX14)); +} + // Matches the (optional) capture-default of a lambda-introducer. class LambdaDefaultCaptureVisitor : public ExpectedLocationVisitor { diff --git a/unittests/Tooling/RefactoringCallbacksTest.cpp b/unittests/Tooling/RefactoringCallbacksTest.cpp index ad8aa8f98feb..e226522a70f4 100644 --- a/unittests/Tooling/RefactoringCallbacksTest.cpp +++ b/unittests/Tooling/RefactoringCallbacksTest.cpp @@ -7,10 +7,10 @@ // //===----------------------------------------------------------------------===// -#include "clang/Tooling/RefactoringCallbacks.h" #include "RewriterTestContext.h" #include "clang/ASTMatchers/ASTMatchFinder.h" #include "clang/ASTMatchers/ASTMatchers.h" +#include "clang/Tooling/RefactoringCallbacks.h" #include "gtest/gtest.h" namespace clang { @@ -19,11 +19,10 @@ namespace tooling { using namespace ast_matchers; template -void expectRewritten(const std::string &Code, - const std::string &Expected, - const T &AMatcher, - RefactoringCallback &Callback) { - MatchFinder Finder; +void expectRewritten(const std::string &Code, const std::string &Expected, + const T &AMatcher, RefactoringCallback &Callback) { + std::map FileToReplace; + ASTMatchRefactorer Finder(FileToReplace); Finder.addMatcher(AMatcher, &Callback); std::unique_ptr Factory( tooling::newFrontendActionFactory(&Finder)); @@ -31,7 +30,7 @@ void expectRewritten(const std::string &Code, << "Parsing error in \"" << Code << "\""; RewriterTestContext Context; FileID ID = Context.createInMemoryFile("input.cc", Code); - EXPECT_TRUE(tooling::applyAllReplacements(Callback.getReplacements(), + EXPECT_TRUE(tooling::applyAllReplacements(FileToReplace["input.cc"], Context.Rewrite)); EXPECT_EQ(Expected, Context.getRewrittenText(ID)); } @@ -61,18 +60,18 @@ TEST(RefactoringCallbacksTest, ReplacesInteger) { std::string Code = "void f() { int i = 1; }"; std::string Expected = "void f() { int i = 2; }"; ReplaceStmtWithText Callback("id", "2"); - expectRewritten(Code, Expected, id("id", expr(integerLiteral())), - Callback); + expectRewritten(Code, Expected, id("id", expr(integerLiteral())), Callback); } TEST(RefactoringCallbacksTest, ReplacesStmtWithStmt) { std::string Code = "void f() { int i = false ? 1 : i * 2; }"; std::string Expected = "void f() { int i = i * 2; }"; ReplaceStmtWithStmt Callback("always-false", "should-be"); - expectRewritten(Code, Expected, - id("always-false", conditionalOperator( - hasCondition(cxxBoolLiteral(equals(false))), - hasFalseExpression(id("should-be", expr())))), + expectRewritten( + Code, Expected, + id("always-false", + conditionalOperator(hasCondition(cxxBoolLiteral(equals(false))), + hasFalseExpression(id("should-be", expr())))), Callback); } @@ -80,10 +79,10 @@ TEST(RefactoringCallbacksTest, ReplacesIfStmt) { std::string Code = "bool a; void f() { if (a) f(); else a = true; }"; std::string Expected = "bool a; void f() { f(); }"; ReplaceIfStmtWithItsBody Callback("id", true); - expectRewritten(Code, Expected, - id("id", ifStmt( - hasCondition(implicitCastExpr(hasSourceExpression( - declRefExpr(to(varDecl(hasName("a"))))))))), + expectRewritten( + Code, Expected, + id("id", ifStmt(hasCondition(implicitCastExpr(hasSourceExpression( + declRefExpr(to(varDecl(hasName("a"))))))))), Callback); } @@ -92,9 +91,63 @@ TEST(RefactoringCallbacksTest, RemovesEntireIfOnEmptyElse) { std::string Expected = "void f() { }"; ReplaceIfStmtWithItsBody Callback("id", false); expectRewritten(Code, Expected, - id("id", ifStmt(hasCondition(cxxBoolLiteral(equals(false))))), - Callback); + id("id", ifStmt(hasCondition(cxxBoolLiteral(equals(false))))), + Callback); } +TEST(RefactoringCallbacksTest, TemplateJustText) { + std::string Code = "void f() { int i = 1; }"; + std::string Expected = "void f() { FOO }"; + auto Callback = ReplaceNodeWithTemplate::create("id", "FOO"); + EXPECT_FALSE(Callback.takeError()); + expectRewritten(Code, Expected, id("id", declStmt()), **Callback); +} + +TEST(RefactoringCallbacksTest, TemplateSimpleSubst) { + std::string Code = "void f() { int i = 1; }"; + std::string Expected = "void f() { long x = 1; }"; + auto Callback = ReplaceNodeWithTemplate::create("decl", "long x = ${init}"); + EXPECT_FALSE(Callback.takeError()); + expectRewritten(Code, Expected, + id("decl", varDecl(hasInitializer(id("init", expr())))), + **Callback); +} + +TEST(RefactoringCallbacksTest, TemplateLiteral) { + std::string Code = "void f() { int i = 1; }"; + std::string Expected = "void f() { string x = \"$-1\"; }"; + auto Callback = ReplaceNodeWithTemplate::create("decl", + "string x = \"$$-${init}\""); + EXPECT_FALSE(Callback.takeError()); + expectRewritten(Code, Expected, + id("decl", varDecl(hasInitializer(id("init", expr())))), + **Callback); +} + +static void ExpectStringError(const std::string &Expected, + llvm::Error E) { + std::string Found; + handleAllErrors(std::move(E), [&](const llvm::StringError &SE) { + llvm::raw_string_ostream Stream(Found); + SE.log(Stream); + }); + EXPECT_EQ(Expected, Found); +} + +TEST(RefactoringCallbacksTest, TemplateUnterminated) { + auto Callback = ReplaceNodeWithTemplate::create("decl", + "string x = \"$$-${init\""); + ExpectStringError("Unterminated ${...} in replacement template near ${init\"", + Callback.takeError()); +} + +TEST(RefactoringCallbacksTest, TemplateUnknownDollar) { + auto Callback = ReplaceNodeWithTemplate::create("decl", + "string x = \"$<"); + ExpectStringError("Invalid $ in replacement template near $<", + Callback.takeError()); +} + + } // end namespace ast_matchers } // end namespace clang diff --git a/www/cxx_dr_status.html b/www/cxx_dr_status.html index a01e0dd8dddb..b03b3f0ef9c2 100644 --- a/www/cxx_dr_status.html +++ b/www/cxx_dr_status.html @@ -28,7 +28,7 @@

C++ Defect Report Support in Clang

-

Last updated: $Date: 2017-03-17 22:41:20 +0100 (Fri, 17 Mar 2017) $

+

Last updated: $Date: 2017-05-10 00:21:24 +0200 (Wed, 10 May 2017) $

C++ defect report implementation status

@@ -589,7 +589,7 @@ 92 - WP + CD4 Should exception-specifications be part of the type system? Clang 4 (C++17 onwards) @@ -935,11 +935,11 @@ Accessibility and ambiguity N/A - - 150 - open + + 150 + DR Template template parameters and default arguments - Not resolved + Unknown 151 @@ -1310,7 +1310,7 @@ accessible? 212 - DR + CD4 Implicit instantiation is not described clearly enough Unknown @@ -1466,7 +1466,7 @@ accessible? 238 - DR + CD4 Precision and accuracy constraints on floating point Unknown @@ -1490,7 +1490,7 @@ accessible? 242 - DR + CD4 Interpretation of old-style casts Unknown @@ -2019,7 +2019,7 @@ of class templates 330 - DRWP + CD4 Qualification conversions and pointers to arrays of pointers Unknown @@ -2397,7 +2397,7 @@ of class templates 393 - DRWP + CD4 Pointer to array of unknown bound in template argument list in parameter Unknown @@ -3017,7 +3017,7 @@ of class templates 496 CD3 Is a volatile-qualified type really a POD? - No + Superseded by dr2094 497 @@ -3587,7 +3587,7 @@ and POD class 591 - DRWP + CD4 When a dependent base class is the current instantiation No @@ -3695,7 +3695,7 @@ and POD class 609 - DRWP + CD4 What is a “top-level” cv-qualifier? Unknown @@ -5735,7 +5735,7 @@ and POD class 987 - DRWP + CD4 Which declarations introduce namespace members? Unknown @@ -5939,7 +5939,7 @@ and POD class 1021 - DRWP + CD4 Definitions of namespace members Unknown @@ -6509,7 +6509,7 @@ and POD class 1116 - DRWP + CD4 Aliasing of union members Unknown @@ -7295,7 +7295,7 @@ and POD class 1247 - DRWP + CD4 Restriction on alias name appearing in type-id Unknown @@ -7457,7 +7457,7 @@ and POD class 1274 - DRWP + CD4 Common nonterminal for expression and braced-init-list Unknown @@ -7517,7 +7517,7 @@ and POD class 1284 - DR + CD4 Should the lifetime of an array be independent of that of its elements? Unknown @@ -7565,7 +7565,7 @@ and POD class 1292 - DRWP + CD4 Dependent calls with braced-init-lists containing a pack expansion Unknown @@ -7667,7 +7667,7 @@ and POD class 1309 - DRWP + CD4 Incorrect note regarding lookup of a member of the current instantiation Unknown @@ -7703,7 +7703,7 @@ and POD class 1315 - DR + CD4 Restrictions on non-type template arguments in partial specializations Partial @@ -7841,7 +7841,7 @@ and POD class 1338 - DRWP + CD4 Aliasing and allocation functions Unknown @@ -7870,8 +7870,8 @@ and POD class Not resolved - 1343 - tentatively ready + 1343 + DR Sequencing of non-class initialization Unknown @@ -7919,7 +7919,7 @@ and POD class 1351 - DRWP + CD4 Problems with implicitly-declared exception-specifications Unknown @@ -7949,7 +7949,7 @@ and POD class 1356 - DRWP + CD4 Exception specifications of copy assignment operators with virtual bases Unknown @@ -8159,7 +8159,7 @@ and POD class 1391 - DRWP + CD4 Conversions to parameter types with non-deduced template arguments Partial @@ -8182,8 +8182,8 @@ and POD class Unknown - 1395 - ready + 1395 + DR Partial ordering of variadic templates reconsidered Unknown @@ -8195,7 +8195,7 @@ and POD class 1397 - DRWP + CD4 Class completeness in non-static data member initializers Unknown @@ -8489,7 +8489,7 @@ and POD class 1446 - DRWP + CD4 Member function with no ref-qualifier and non-member function with rvalue reference Unknown @@ -8603,7 +8603,7 @@ and POD class 1465 - DRWP + CD4 noexcept and std::bad_array_new_length Unknown @@ -8615,7 +8615,7 @@ and POD class 1467 - DRWP + CD4 List-initialization of aggregate from same-type object Clang 3.7 (C++11 onwards) @@ -8717,7 +8717,7 @@ and POD class 1484 - DRWP + CD4 Unused local classes of function templates Unknown @@ -8753,7 +8753,7 @@ and POD class 1490 - DRWP + CD4 List-initialization from a string literal Clang 3.7 (C++11 onwards) @@ -8765,7 +8765,7 @@ and POD class 1492 - DRWP + CD4 Exception specifications on template destructors Unknown @@ -8789,7 +8789,7 @@ and POD class 1496 - DR + CD4 Triviality with deleted and missing default constructors Unknown @@ -8921,7 +8921,7 @@ and POD class 1518 - DRWP + CD4 Explicit default constructors and copy-list-initialization Clang 4 @@ -9125,7 +9125,7 @@ and POD class 1552 - DRWP + CD4 exception-specifications and defaulted special member functions Unknown @@ -9161,7 +9161,7 @@ and POD class 1558 - DRWP + CD4 Unused arguments in alias template specializations Unknown @@ -9201,11 +9201,11 @@ and POD class Template argument deduction from an initializer list Unknown - - 1565 - drafting + + 1565 + NAD Copy elision and lifetime of initializer_list underlying array - Not resolved + Unknown 1566 @@ -9239,19 +9239,19 @@ and POD class 1571 - DRWP + CD4 cv-qualification for indirect reference binding via conversion function Unknown 1572 - DRWP + CD4 Incorrect example for rvalue reference binding via conversion function Unknown 1573 - DRWP + CD4 Inherited constructor characteristics Clang 3.9 @@ -9347,7 +9347,7 @@ and POD class 1589 - DRWP + CD4 Ambiguous ranking of list-initialization sequences Clang 3.7 (C++11 onwards) @@ -9359,7 +9359,7 @@ and POD class 1591 - DRWP + CD4 Deducing array bound and element type from initializer list Unknown @@ -9389,7 +9389,7 @@ and POD class 1596 - DRWP + CD4 Non-array objects as array[1] Unknown @@ -9413,7 +9413,7 @@ and POD class 1600 - DRWP + CD4 Erroneous reference initialization in example Unknown @@ -9431,7 +9431,7 @@ and POD class 1603 - DRWP + CD4 Errors resulting from giving unnamed namespaces internal linkage Unknown @@ -9497,13 +9497,13 @@ and POD class 1614 - DRWP + CD4 Address of pure virtual function vs odr-use Unknown 1615 - DRWP + CD4 Alignment of types, variables, and members Unknown @@ -9543,11 +9543,11 @@ and POD class Member initializers in anonymous unions Not resolved - + 1622 - drafting + ready Empty aggregate initializer for union - Not resolved + Unknown 1623 @@ -9593,13 +9593,13 @@ and POD class 1630 - DRWP + CD4 Multiple default constructor templates Unknown 1631 - DRWP + CD4 Incorrect overload resolution for single-element initializer-list Clang 3.7 @@ -9611,7 +9611,7 @@ and POD class 1633 - DRWP + CD4 Copy-initialization in member initialization Unknown @@ -9641,13 +9641,13 @@ and POD class 1638 - DR + CD4 Declaring an explicit specialization of a scoped enumeration Yes 1639 - DRWP + CD4 exception-specifications and pointer/pointer-to-member expressions Unknown @@ -9683,7 +9683,7 @@ and POD class 1645 - DRWP + CD4 Identical inheriting constructors via default arguments Clang 3.9 @@ -9725,13 +9725,13 @@ and POD class 1652 - DRWP + CD4 Object addresses in constexpr expressions Unknown 1653 - WP + CD4 Removing deprecated increment of bool Clang 4 (C++17 onwards) @@ -9755,7 +9755,7 @@ and POD class 1657 - WP + CD4 Attributes for namespaces and enumerators Unknown @@ -9845,7 +9845,7 @@ and POD class 1672 - DRWP + CD4 Layout compatibility with multiple empty bases Unknown @@ -9873,11 +9873,11 @@ and POD class auto return type for allocation and deallocation functions Not resolved - + 1677 - drafting + ready Constant initialization via aggregate initialization - Not resolved + Unknown 1678 @@ -9911,7 +9911,7 @@ and POD class 1683 - DRWP + CD4 Incorrect example after constexpr changes Unknown @@ -9929,7 +9929,7 @@ and POD class 1686 - DRWP + CD4 Which variables are “explicitly declared const?” Unknown @@ -9977,7 +9977,7 @@ and POD class 1694 - DRWP + CD4 Restriction on reference to temporary as a constant expression Unknown @@ -9989,7 +9989,7 @@ and POD class 1696 - DRWP + CD4 Temporary lifetime and non-static data member initializers Unknown @@ -10043,7 +10043,7 @@ and POD class 1705 - DRWP + CD4 Unclear specification of “more specialized” Unknown @@ -10061,7 +10061,7 @@ and POD class 1708 - DRWP + CD4 overly-strict requirements for names with C language linkage Unknown @@ -10085,7 +10085,7 @@ and POD class 1712 - DRWP + CD4 constexpr variable template declarations Unknown @@ -10103,7 +10103,7 @@ and POD class 1715 - DRWP + CD4 Access and inherited constructor templates Clang 3.9 @@ -10127,7 +10127,7 @@ and POD class 1719 - DRWP + CD4 Layout compatibility and cv-qualification revisited Unknown @@ -10145,7 +10145,7 @@ and POD class 1722 - DRWP + CD4 Should lambda to function pointer conversion function be noexcept? Unknown @@ -10217,7 +10217,7 @@ and POD class 1734 - DR + CD4 Nontrivial deleted copy functions Unknown @@ -10229,7 +10229,7 @@ and POD class 1736 - DRWP + CD4 Inheriting constructor templates in a local class Clang 3.9 @@ -10277,7 +10277,7 @@ and POD class 1744 - DRWP + CD4 Unordered initialization for variable template specializations Unknown @@ -10301,7 +10301,7 @@ and POD class 1748 - DRWP + CD4 Placement new with a null pointer Clang 3.7 @@ -10313,25 +10313,25 @@ and POD class 1750 - DRWP + CD4 “Argument” vs “parameter” Unknown 1751 - DRWP + CD4 Non-trivial operations vs non-trivial initialization Unknown 1752 - DRWP + CD4 Right-recursion in mem-initializer-list Unknown 1753 - DRWP + CD4 decltype-specifier in nested-name-specifier of destructor Unknown @@ -10349,19 +10349,19 @@ and POD class 1756 - DRWP + CD4 Direct-list-initialization of a non-class object Clang 3.7 1757 - DRWP + CD4 Const integral subobjects Unknown 1758 - DRWP + CD4 Explicit conversion in copy/move list initialization Clang 3.7 @@ -10409,7 +10409,7 @@ and POD class 1766 - DRWP + CD4 Values outside the range of the values of an enumeration Unknown @@ -10457,7 +10457,7 @@ and POD class 1774 - DRWP + CD4 Discrepancy between subobject destruction and stack unwinding Unknown @@ -10469,13 +10469,13 @@ and POD class 1776 - DRWP + CD4 Replacement of class objects containing reference members Unknown 1777 - DRWP + CD4 Empty pack expansion in dynamic-exception-specification Unknown @@ -10487,13 +10487,13 @@ and POD class 1779 - DRWP + CD4 Type dependency of __func__ Unknown 1780 - DRWP + CD4 Explicit instantiation/specialization of generic lambda operator() Unknown @@ -10505,7 +10505,7 @@ and POD class 1782 - DRWP + CD4 Form of initialization for nullptr_t to bool conversion Unknown @@ -10541,7 +10541,7 @@ and POD class 1788 - DRWP + CD4 Sized deallocation of array of non-class type Unknown @@ -10559,7 +10559,7 @@ and POD class 1791 - DRWP + CD4 Incorrect restrictions on cv-qualifier-seq and ref-qualifier Unknown @@ -10571,7 +10571,7 @@ and POD class 1793 - DRWP + CD4 thread_local in explicit specializations Unknown @@ -10583,19 +10583,19 @@ and POD class 1795 - DRWP + CD4 Disambiguating original-namespace-definition and extension-namespace-definition Unknown 1796 - DRWP + CD4 Is all-bits-zero for null characters a meaningful requirement? Unknown 1797 - DRWP + CD4 Are all bit patterns of unsigned char distinct numbers? Unknown @@ -10607,13 +10607,13 @@ and POD class 1799 - DRWP + CD4 mutable and non-explicit const qualification Unknown 1800 - DRWP + CD4 Pointer to member of nested anonymous union Unknown @@ -10625,7 +10625,7 @@ and POD class 1802 - DRWP + CD4 char16_t string literals and surrogate pairs Unknown @@ -10637,25 +10637,25 @@ and POD class 1804 - DRWP + CD4 Partial specialization and friendship Unknown 1805 - DRWP + CD4 Conversions of array operands in conditional-expressions Unknown 1806 - DRWP + CD4 Virtual bases and move-assignment Unknown 1807 - DRWP + CD4 Order of destruction of array elements after an exception Unknown @@ -10667,19 +10667,19 @@ and POD class 1809 - DRWP + CD4 Narrowing and template argument deduction Unknown 1810 - DRWP + CD4 Invalid ud-suffixes Unknown 1811 - DRWP + CD4 Lookup of deallocation function in a virtual destructor definition Unknown @@ -10691,25 +10691,25 @@ and POD class 1813 - DRWP + CD4 Direct vs indirect bases in standard-layout classes Unknown 1814 - DRWP + CD4 Default arguments in lambda-expressions Unknown 1815 - DRWP + CD4 Lifetime extension in aggregate initialization Unknown 1816 - DRWP + CD4 Unclear specification of bit-field values Unknown @@ -10727,7 +10727,7 @@ and POD class 1819 - DRWP + CD4 Acceptable scopes for definition of partial specialization Unknown @@ -10751,19 +10751,19 @@ and POD class 1823 - DRWP + CD4 String literal uniqueness in inline functions Unknown 1824 - DRWP + CD4 Completeness of return type vs point of instantiation Unknown - 1825 - ready + 1825 + DR Partial ordering between variadic and non-variadic function templates Unknown @@ -10793,7 +10793,7 @@ and POD class 1830 - DRWP + CD4 Repeated specifiers Unknown @@ -10805,7 +10805,7 @@ and POD class 1832 - DRWP + CD4 Casting to incomplete enumeration Unknown @@ -10817,7 +10817,7 @@ and POD class 1834 - DRWP + CD4 Constant initialization binding a reference to an xvalue Unknown @@ -10841,7 +10841,7 @@ and POD class 1838 - DRWP + CD4 Definition via unqualified-id and using-declaration Unknown @@ -10871,7 +10871,7 @@ and POD class 1843 - DRWP + CD4 Bit-field in conditional operator with throw operand Unknown @@ -10889,19 +10889,19 @@ and POD class 1846 - DRWP + CD4 Declaring explicitly-defaulted implicitly-deleted functions Unknown 1847 - DRWP + CD4 Clarifying compatibility during partial ordering Unknown 1848 - DRWP + CD4 Parenthesized constructor and destructor declarators Unknown @@ -10913,19 +10913,19 @@ and POD class 1850 - DRWP + CD4 Differences between definition context and point of instantiation Unknown 1851 - DRWP + CD4 decltype(auto) in new-expressions Unknown 1852 - DRWP + CD4 Wording issues regarding decltype(auto) Unknown @@ -10961,7 +10961,7 @@ and POD class 1858 - DRWP + CD4 Comparing pointers to union members Unknown @@ -10971,15 +10971,15 @@ and POD class UTF-16 in char16_t string literals Not resolved - + 1860 - review + ready What is a “direct member?” - Not resolved + Unknown 1861 - DRWP + CD4 Values of a bit-field Unknown @@ -10991,7 +10991,7 @@ and POD class 1863 - DRWP + CD4 Requirements on thrown object type to support std::current_exception() Unknown @@ -11003,13 +11003,13 @@ and POD class 1865 - DRWP + CD4 Pointer arithmetic and multi-level qualification conversions Unknown 1866 - DRWP + CD4 Initializing variant members with non-trivial destructors Unknown @@ -11033,7 +11033,7 @@ and POD class 1870 - DRWP + CD4 Contradictory wording about definitions vs explicit specialization/instantiation Unknown @@ -11045,25 +11045,25 @@ and POD class 1872 - DR + CD4 Instantiations of constexpr templates that cannot appear in constant expressions Unknown 1873 - DRWP + CD4 Protected member access from derived class friends Unknown 1874 - DRWP + CD4 Type vs non-type template parameters with class keyword Unknown 1875 - DRWP + CD4 Reordering declarations in class scope Unknown @@ -11075,13 +11075,13 @@ and POD class 1877 - DRWP + CD4 Return type deduction from return with no operand Unknown 1878 - DRWP + CD4 operator auto template Unknown @@ -11099,13 +11099,13 @@ and POD class 1881 - DRWP + CD4 Standard-layout classes and unnamed bit-fields Unknown 1882 - DRWP + CD4 Reserved names without library use Unknown @@ -11123,25 +11123,25 @@ and POD class 1885 - DRWP + CD4 Return value of a function is underspecified Unknown 1886 - DRWP + CD4 Language linkage for main() Unknown 1887 - DRWP + CD4 Problems with :: as nested-name-specifier Unknown 1888 - DRWP + CD4 Implicitly-declared default constructors and explicit Unknown @@ -11159,13 +11159,13 @@ and POD class 1891 - DRWP + CD4 Move constructor/assignment for closure class Clang 4 1892 - DRWP + CD4 Use of auto in function type Unknown @@ -11183,7 +11183,7 @@ and POD class 1895 - DR + CD4 Deleted conversions in conditional operator operands Unknown @@ -11207,7 +11207,7 @@ and POD class 1899 - DRWP + CD4 Value-dependent constant expressions Unknown @@ -11225,13 +11225,13 @@ and POD class 1902 - DRWP + CD4 What makes a conversion “otherwise ill-formed”? Clang 3.7 1903 - DRWP + CD4 What declarations are introduced by a non-member using-declaration? Unknown @@ -11267,7 +11267,7 @@ and POD class 1909 - DRWP + CD4 Member class template with the same name as the class Yes @@ -11279,7 +11279,7 @@ and POD class 1911 - DRWP + CD4 constexpr constructor with non-literal base class Unknown @@ -11309,7 +11309,7 @@ and POD class 1916 - DRWP + CD4 “Same cv-unqualified type” Unknown @@ -11333,7 +11333,7 @@ and POD class 1920 - DRWP + CD4 Qualification mismatch in pseudo-destructor-name Unknown @@ -11345,7 +11345,7 @@ and POD class 1922 - DRWP + CD4 Injected class template names and default arguments Unknown @@ -11357,19 +11357,19 @@ and POD class 1924 - drafting + review Definition of “literal” and kinds of literals Not resolved 1925 - DRWP + CD4 Bit-field prvalues Unknown 1926 - DRWP + CD4 Potential results of subscript operator Unknown @@ -11387,13 +11387,13 @@ and POD class 1929 - DRWP + CD4 template keyword following namespace nested-name-specifier Unknown 1930 - DR + CD4 init-declarator-list vs member-declarator-list Unknown @@ -11405,7 +11405,7 @@ and POD class 1932 - DR + CD4 Bit-field results of conditional operators Unknown @@ -11453,19 +11453,19 @@ and POD class 1940 - DRWP + CD4 static_assert in anonymous unions Yes 1941 - DRWP + CD4 SFINAE and inherited constructor default arguments Clang 3.9 1942 - DRWP + CD4 Incorrect reference to trailing-return-type Unknown @@ -11489,7 +11489,7 @@ and POD class 1946 - WP + CD4 exception-specifications vs pointer dereference Unknown @@ -11507,7 +11507,7 @@ and POD class 1949 - DRWP + CD4 “sequenced after” instead of “sequenced before” Unknown @@ -11519,13 +11519,13 @@ and POD class 1951 - DRWP + CD4 Cv-qualification and literal types Unknown 1952 - DRWP + CD4 Constant expressions and library undefined behavior Unknown @@ -11543,13 +11543,13 @@ and POD class 1955 - DR + CD4 #elif with invalid controlling expression Unknown 1956 - DRWP + CD4 Reuse of storage of automatic variables Unknown @@ -11561,13 +11561,13 @@ and POD class 1958 - DRWP + CD4 decltype(auto) with parenthesized initializer Unknown 1959 - DRWP + CD4 Inadvertently inherited copy constructor Clang 3.9 @@ -11578,8 +11578,8 @@ and POD class Unknown - 1961 - ready + 1961 + DR Potentially-concurrent actions within a signal handler Unknown @@ -11591,7 +11591,7 @@ and POD class 1963 - DRWP + CD4 Implementation-defined identifier characters Unknown @@ -11609,13 +11609,13 @@ and POD class 1966 - DRWP + CD4 Colon following enumeration elaborated-type-specifier Unknown 1967 - DRWP + CD4 Temporary lifetime and move-elision Unknown @@ -11639,7 +11639,7 @@ and POD class 1971 - DRWP + CD4 Unclear disambiguation of destructor and operator~ Unknown @@ -11663,7 +11663,7 @@ and POD class 1975 - DRWP + CD4 Permissible declarations for exception-specifications Unknown @@ -11681,7 +11681,7 @@ and POD class 1978 - DRWP + CD4 Redundant description of explicit constructor use Unknown @@ -11699,7 +11699,7 @@ and POD class 1981 - DRWP + CD4 Implicit contextual conversions and explicit Unknown @@ -11741,7 +11741,7 @@ and POD class 1988 - DRWP + CD4 Ambiguity between dependent and non-dependent bases in implicit member access Unknown @@ -11753,19 +11753,19 @@ and POD class 1990 - DRWP + CD4 Ambiguity due to optional decl-specifier-seq Unknown 1991 - DRWP + CD4 Inheriting constructors vs default arguments Clang 3.9 1992 - DR + CD4 new (std::nothrow) int[N] can throw Unknown @@ -11783,7 +11783,7 @@ and POD class 1995 - WP + CD4 exception-specifications and non-type template parameters Unknown @@ -11807,19 +11807,19 @@ and POD class 1999 - DRWP + CD4 Representation of source characters as universal-character-names Unknown 2000 - DRWP + CD4 header-name outside #include directive Unknown 2001 - DR + CD4 non-directive is underspecified Unknown @@ -11837,7 +11837,7 @@ and POD class 2004 - DRWP + CD4 Unions with mutable members in constant expressions Unknown @@ -11849,7 +11849,7 @@ and POD class 2006 - DRWP + CD4 Cv-qualified void types Unknown @@ -11861,7 +11861,7 @@ and POD class 2008 - DR + CD4 Default template-arguments underspecified Unknown @@ -11873,7 +11873,7 @@ and POD class 2010 - WP + CD4 exception-specifications and conversion operators Unknown @@ -11885,7 +11885,7 @@ and POD class 2012 - DR + CD4 Lifetime of references Unknown @@ -11903,19 +11903,19 @@ and POD class 2015 - DRWP + CD4 odr-use of deleted virtual functions Unknown 2016 - DRWP + CD4 Confusing wording in description of conversion function Unknown 2017 - DR + CD4 Flowing off end is not equivalent to no-expression return Unknown @@ -11927,7 +11927,7 @@ and POD class 2019 - DRWP + CD4 Member references omitted from description of storage duration Unknown @@ -11945,7 +11945,7 @@ and POD class 2022 - DRWP + CD4 Copy elision in constant expressions Unknown @@ -11957,7 +11957,7 @@ and POD class 2024 - DRWP + CD4 Dependent types and unexpanded parameter packs Unknown @@ -11969,13 +11969,13 @@ and POD class 2026 - DRWP + CD4 Zero-initialization and constexpr Unknown 2027 - DRWP + CD4 Unclear requirements for multiple alignas specifiers Unknown @@ -11999,19 +11999,19 @@ and POD class 2031 - DRWP + CD4 Missing incompatibility for && Unknown 2032 - DR + CD4 Default template-arguments of variable templates Unknown 2033 - DR + CD4 Redundant restriction on partial specialization argument Unknown @@ -12041,25 +12041,25 @@ and POD class 2038 - DR + CD4 Document C++14 incompatibility of new braced deduction rule Unknown 2039 - DR + CD4 Constant conversions to bool Unknown 2040 - DR + CD4 trailing-return-type no longer ambiguous Unknown 2041 - DR + CD4 Namespace for explicit class template specialization Unknown @@ -12077,7 +12077,7 @@ and POD class 2044 - DR + CD4 decltype(auto) and void Unknown @@ -12095,7 +12095,7 @@ and POD class 2047 - DR + CD4 Coordinating “throws anything” specifications Unknown @@ -12125,7 +12125,7 @@ and POD class 2052 - DRWP + CD4 Template argument deduction vs overloaded operators Unknown @@ -12179,7 +12179,7 @@ and POD class 2061 - DR + CD4 Inline namespace after simplifications Unknown @@ -12191,13 +12191,13 @@ and POD class 2063 - DR + CD4 Type/nontype hiding in class scope Unknown 2064 - DR + CD4 Conflicting specifications for dependent decltype-specifiers Unknown @@ -12209,7 +12209,7 @@ and POD class 2066 - DR + CD4 Does type-dependent imply value-dependent? Unknown @@ -12221,13 +12221,13 @@ and POD class 2068 - DR + CD4 When can/must a defaulted virtual destructor be defined? Unknown 2069 - DR + CD4 Do destructors have names? Unknown @@ -12239,7 +12239,7 @@ and POD class 2071 - DR + CD4 typedef with no declarator Unknown @@ -12263,13 +12263,13 @@ and POD class 2075 - DRWP + CD4 Passing short initializer lists to array reference parameters Unknown 2076 - DRWP + CD4 List-initialization of arguments for constructor parameters Unknown @@ -12287,7 +12287,7 @@ and POD class 2079 - DR + CD4 [[ appearing in a balanced-token-seq Unknown @@ -12305,7 +12305,7 @@ and POD class 2082 - DR + CD4 Referring to parameters in unevaluated operands of default arguments Unknown @@ -12317,13 +12317,13 @@ and POD class 2084 - DR + CD4 NSDMIs and deleted union default constructors Unknown 2085 - DR + CD4 Invalid example of adding special member function via default argument Unknown @@ -12359,7 +12359,7 @@ and POD class 2091 - DRWP + CD4 Deducing reference non-type template arguments Unknown @@ -12371,25 +12371,25 @@ and POD class 2093 - DR + CD4 Qualification conversion for pointer-to-member handler matching Unknown - 2094 - tentatively ready + 2094 + DR Trivial copy/move constructor for class with volatile member - Unknown + Clang 5.0 2095 - DR + CD4 Capturing rvalue references to functions by copy Unknown 2096 - DR + CD4 Constraints on literal unions Unknown @@ -12401,25 +12401,25 @@ and POD class 2098 - DR + CD4 Is uncaught_exceptions() per-thread? Unknown 2099 - DR + CD4 Inferring the bound of an array static data member Unknown - 2100 - tentatively ready + 2100 + DR Value-dependent address of static data member of class template Unknown 2101 - DRWP + CD4 Incorrect description of type- and value-dependence Unknown @@ -12437,7 +12437,7 @@ and POD class 2104 - DR + CD4 Internal-linkage constexpr references and ODR requirements Unknown @@ -12449,13 +12449,13 @@ and POD class 2106 - DR + CD4 Unclear restrictions on use of function-type template arguments Unknown 2107 - DR + CD4 Lifetime of temporaries for default arguments in array copying Unknown @@ -12467,7 +12467,7 @@ and POD class 2109 - DR + CD4 Value dependence underspecified Unknown @@ -12491,7 +12491,7 @@ and POD class 2113 - DR + CD4 Incompete specification of types for declarators Unknown @@ -12533,7 +12533,7 @@ and POD class 2120 - DRWP + CD4 Array as first non-static data member in standard-layout class Unknown @@ -12545,7 +12545,7 @@ and POD class 2122 - DR + CD4 Glvalues of void type Unknown @@ -12557,7 +12557,7 @@ and POD class 2124 - DR + CD4 Signature of constructor template Unknown @@ -12587,13 +12587,13 @@ and POD class 2129 - DR + CD4 Non-object prvalues and constant expressions Unknown 2130 - DR + CD4 Over-aligned types in new-expressions Unknown @@ -12635,7 +12635,7 @@ and POD class 2137 - DRWP + CD4 List-initialization from object of same type Unknown @@ -12653,13 +12653,13 @@ and POD class 2140 - DR + CD4 Lvalue-to-rvalue conversion of std::nullptr_t Unknown 2141 - DR + CD4 Ambiguity in new-expression with elaborated-type-specifier Unknown @@ -12670,8 +12670,8 @@ and POD class Unknown - 2143 - ready + 2143 + DR Value-dependency via injected-class-name Unknown @@ -12683,19 +12683,19 @@ and POD class 2145 - DRWP + CD4 Parenthesized declarator in function definition Unknown 2146 - DR + CD4 Scalar object vs memory location in definition of “unsequenced” Unknown 2147 - DR + CD4 Initializer-list arguments and pack deduction Unknown @@ -12731,31 +12731,31 @@ and POD class 2153 - DR + CD4 pure-specifier in friend declaration Unknown 2154 - DR + CD4 Ambiguity of pure-specifier Unknown - 2155 - ready + 2155 + DR Defining classes and enumerations via using-declarations Unknown 2156 - DR + CD4 Definition of enumeration declared by using-declaration Unknown 2157 - DR + CD4 Further disambiguation of enumeration elaborated-type-specifier Unknown @@ -12791,7 +12791,7 @@ and POD class 2163 - DR + CD4 Labels in constexpr functions Unknown @@ -12815,7 +12815,7 @@ and POD class 2167 - DR + CD4 Non-member references with lifetimes within the current evaluation Unknown @@ -12839,7 +12839,7 @@ and POD class 2171 - DRWP + CD4 Triviality of copy constructor with less-qualified parameter Unknown @@ -12855,21 +12855,21 @@ and POD class Partial specialization with non-deduced contexts Not resolved - + 2174 - drafting + ready Unclear rules for friend definitions in templates - Not resolved + Unknown 2175 - DR + CD4 Ambiguity with attribute in conversion operator declaration Unknown 2176 - DR + CD4 Destroying the returned object when a destructor throws Unknown @@ -12893,7 +12893,7 @@ and POD class 2180 - DR + CD4 Virtual bases in destructors and defaulted assignment operators Yes @@ -12915,11 +12915,11 @@ and POD class Problems in description of potential exceptions Not resolved - - 2184 - review + + 2184 + CD4 Missing C compatibility entry for decrement of bool - Not resolved + Unknown 2185 @@ -12957,11 +12957,11 @@ and POD class Insufficient specification of __has_include Not resolved - + 2191 - open + ready Incorrect result for noexcept(typeid(v)) - Not resolved + Unknown 2192 @@ -12975,11 +12975,11 @@ and POD class numeric_limits<int>::radix and digits Not resolved - + 2194 - open + tentatively ready Impossible case in list initialization - Not resolved + Unknown 2195 @@ -12999,11 +12999,11 @@ and POD class Overload resolution and deleted special member functions Not resolved - + 2198 - open + tentatively ready Linkage of enumerators - Not resolved + Unknown 2199 @@ -13017,11 +13017,11 @@ and POD class Conversions in template argument deduction Not resolved - + 2201 - open + tentatively ready Cv-qualification of array types - Not resolved + Unknown 2202 @@ -13041,17 +13041,17 @@ and POD class Naming delegated constructors Not resolved - + 2205 - open + ready Restrictions on use of alignas - Not resolved + Unknown - + 2206 - open + tentatively ready Composite type of object and function pointers - Not resolved + Unknown 2207 @@ -13095,11 +13095,11 @@ and POD class Forward declaration of partial specializations Not resolved - + 2214 - open + tentatively ready Missing requirement on representation of integer values - Not resolved + Unknown 2215 @@ -13119,11 +13119,11 @@ and POD class constexpr constructors for non-literal types Not resolved - + 2218 - open + ready Ambiguity and namespace aliases - Not resolved + Unknown 2219 @@ -13131,11 +13131,11 @@ and POD class Dynamically-unreachable handlers Not resolved - + 2220 - open + tentatively ready Hiding index variable in range-based for - Not resolved + Unknown 2221 @@ -13155,11 +13155,11 @@ and POD class Multiple alignas specifiers Not resolved - + 2224 - open + tentatively ready Member subobjects and base-class casts - Not resolved + Unknown 2225 @@ -13301,7 +13301,7 @@ and POD class 2248 - open + review Problems with sized delete Not resolved @@ -13319,7 +13319,7 @@ and POD class 2251 - open + review Unreachable enumeration list-initialization Not resolved @@ -13365,11 +13365,11 @@ and POD class Storage deallocation during period of destruction Not resolved - + 2259 - open + tentatively ready Unclear context describing ambiguity - Not resolved + Unknown 2260 @@ -13383,11 +13383,11 @@ and POD class Explicit instantiation of in-class friend definition Not resolved - + 2262 - open + tentatively ready Attributes for asm-definition - Not resolved + Unknown 2263 @@ -13438,11 +13438,47 @@ and POD class Not resolved - 2271 - ready + 2271 + DR Aliasing this Unknown + + 2272 + open + Implicit initialization of aggregate members of reference type + Not resolved + + + 2273 + open + Inheriting constructors vs implicit default constructor + Not resolved + + + 2274 + open + Generic lambda capture vs constexpr if + Not resolved + + + 2275 + open + Type-dependence of function template + Not resolved + + + 2276 + open + Dependent noexcept and function type-dependence + Not resolved + + + 2277 + open + Ambiguity inheriting constructors with default arguments + Not resolved + From 7582e3938bb9fb3e4664efdfb2313df29f27b70b Mon Sep 17 00:00:00 2001 From: Dimitry Andric Date: Tue, 16 May 2017 19:47:31 +0000 Subject: [PATCH 2/9] Vendor import of libc++ trunk r303197: https://llvm.org/svn/llvm-project/libcxx/trunk@303197 --- ...eyor-reqs.cmd => appveyor-reqs-install.cmd | 12 +- appveyor.yml | 22 +- docs/CMakeLists.txt | 4 +- docs/TestingLibcxx.rst | 2 +- include/__bsd_locale_fallbacks.h | 31 +- include/__config | 12 +- include/__locale | 19 + include/__mutex_base | 1 + include/__threading_support | 1 + include/__undef_min_max | 4 +- include/algorithm | 127 +++- include/ctype.h | 9 - include/experimental/numeric | 13 +- include/ext/hash_map | 2 +- include/ext/hash_set | 2 +- include/limits | 4 +- include/locale | 3 - include/memory | 10 +- include/numeric | 11 +- include/stdio.h | 5 +- include/stdlib.h | 4 - include/string_view | 8 - .../{limits_win32.h => limits_msvc_win32.h} | 27 +- include/support/win32/locale_mgmt_win32.h | 33 - include/support/win32/locale_win32.h | 32 +- include/support/win32/support.h | 177 ----- include/variant | 25 +- include/wchar.h | 9 +- src/include/atomic_support.h | 2 +- src/locale.cpp | 46 +- src/string.cpp | 5 +- .../runtime/exception_pointer_msvc.ipp | 1 + src/support/win32/locale_win32.cpp | 26 +- src/thread.cpp | 2 +- .../fopen.fail.cpp | 2 + .../rename.fail.cpp | 2 + ...otify_from_pthread_created_thread.pass.cpp | 0 .../thread.thread.this/sleep_for.pass.cpp | 1 - .../function.objects/refwrap/binary.pass.cpp | 0 .../function.objects/refwrap/unary.pass.cpp | 0 .../template.bitset/includes.pass.cpp | 32 + .../tuple/tuple.tuple/empty_member.pass.cpp | 0 .../meta.type.synop/meta.unary.prop.pass.cpp | 6 + .../lit.local.cfg | 2 - .../delete_align_val_t_replace.pass.cpp | 2 +- .../new.delete.array/new_align_val_t.pass.cpp | 2 +- .../new_align_val_t_nothrow.pass.cpp | 2 +- .../new_align_val_t_nothrow_replace.pass.cpp | 2 +- .../new_align_val_t_replace.pass.cpp | 3 +- .../delete_align_val_t_replace.pass.cpp | 2 +- .../new_align_val_t.pass.cpp | 2 +- .../new_align_val_t_nothrow.pass.cpp | 2 +- .../new_align_val_t_nothrow_replace.pass.cpp | 3 +- .../new_align_val_t_replace.pass.cpp | 3 +- .../support.types/byteops/and.assign.pass.cpp | 9 +- .../support.types/byteops/and.pass.cpp | 9 +- .../byteops/enum_direct_init.pass.cpp | 21 + .../byteops/lshift.assign.pass.cpp | 7 +- .../support.types/byteops/lshift.fail.cpp | 5 +- .../support.types/byteops/lshift.pass.cpp | 7 +- .../support.types/byteops/not.pass.cpp | 9 +- .../support.types/byteops/or.assign.pass.cpp | 9 +- .../support.types/byteops/or.pass.cpp | 9 +- .../byteops/rshift.assign.pass.cpp | 7 +- .../support.types/byteops/rshift.fail.cpp | 5 +- .../support.types/byteops/rshift.pass.cpp | 7 +- .../support.types/byteops/to_integer.fail.cpp | 5 +- .../support.types/byteops/to_integer.pass.cpp | 7 +- .../support.types/byteops/xor.assign.pass.cpp | 9 +- .../support.types/byteops/xor.pass.cpp | 9 +- .../locale.stdcvt/codecvt_utf16_out.pass.cpp | 626 ++++++++-------- .../locale.stdcvt/codecvt_utf8_out.pass.cpp | 708 +++++++----------- .../codecvt_utf8_utf16_in.pass.cpp | 546 ++++++-------- .../codecvt_utf8_utf16_out.pass.cpp | 648 +++++++--------- .../conversions.string/converted.pass.cpp | 39 +- .../conversions.string/from_bytes.pass.cpp | 65 +- .../conversions.string/to_bytes.pass.cpp | 65 +- .../numeric.ops/numeric.ops.gcd/gcd.pass.cpp | 45 +- .../numeric.ops/numeric.ops.lcm/lcm.pass.cpp | 49 +- .../re/re.traits/lookup_classname.pass.cpp | 68 +- .../string.cons/T_size_size.pass.cpp | 18 +- .../string.view.modifiers/clear.pass.cpp | 67 -- .../sleep_for_tested_elsewhere.pass.cpp | 22 + .../allocator.adaptor.cnstr/allocs.pass.cpp | 8 +- .../func.not_fn/not_fn.pass.cpp | 2 +- .../func.wrap.func.con/copy_move.pass.cpp | 3 +- .../func.wrap.func.targ/target.pass.cpp | 4 + .../enable_shared_from_this.pass.cpp | 10 + .../meta/meta.rel/is_convertible.pass.cpp | 4 +- .../is_trivially_copyable.pass.cpp | 12 +- .../template.bitset/includes.pass.cpp | 35 +- .../tuple.tuple/tuple.cnstr/dtor.pass.cpp | 4 + .../utility/pairs/pairs.pair/dtor.pass.cpp | 4 + .../variant/variant.visit/visit.pass.cpp | 10 + test/support/archetypes.hpp | 9 + test/support/archetypes.ipp | 4 + .../support/filesystem_dynamic_test_helper.py | 2 +- test/support/filesystem_test_helper.hpp | 4 +- test/support/msvc_stdlib_force_include.hpp | 30 +- .../c1xx_broken_za_ctor_check.pass.cpp | 41 + test/support/test_macros.h | 8 +- test/support/test_workarounds.h | 7 + utils/libcxx/test/config.py | 25 +- utils/libcxx/test/executor.py | 29 +- utils/libcxx/test/format.py | 2 +- www/cxx1z_status.html | 10 +- 106 files changed, 1946 insertions(+), 2180 deletions(-) rename install-appveyor-reqs.cmd => appveyor-reqs-install.cmd (81%) rename include/support/win32/{limits_win32.h => limits_msvc_win32.h} (80%) delete mode 100644 include/support/win32/locale_mgmt_win32.h delete mode 100644 include/support/win32/support.h rename test/{std => libcxx}/input.output/file.streams/c.files/no.global.filesystem.namespace/fopen.fail.cpp (89%) rename test/{std => libcxx}/input.output/file.streams/c.files/no.global.filesystem.namespace/rename.fail.cpp (89%) rename test/{std => libcxx}/thread/thread.condition/PR30202_notify_from_pthread_created_thread.pass.cpp (100%) rename test/{std => libcxx}/thread/thread.threads/thread.thread.this/sleep_for.pass.cpp (99%) rename test/{std => libcxx}/utilities/function.objects/refwrap/binary.pass.cpp (100%) rename test/{std => libcxx}/utilities/function.objects/refwrap/unary.pass.cpp (100%) create mode 100644 test/libcxx/utilities/template.bitset/includes.pass.cpp rename test/{std => libcxx}/utilities/tuple/tuple.tuple/empty_member.pass.cpp (100%) delete mode 100644 test/std/input.output/file.streams/c.files/no.global.filesystem.namespace/lit.local.cfg create mode 100644 test/std/language.support/support.types/byteops/enum_direct_init.pass.cpp delete mode 100644 test/std/strings/string.view/string.view.modifiers/clear.pass.cpp create mode 100644 test/std/thread/thread.threads/thread.thread.this/sleep_for_tested_elsewhere.pass.cpp create mode 100644 test/support/test.workarounds/c1xx_broken_za_ctor_check.pass.cpp diff --git a/install-appveyor-reqs.cmd b/appveyor-reqs-install.cmd similarity index 81% rename from install-appveyor-reqs.cmd rename to appveyor-reqs-install.cmd index ebd72ca0a2da..3a2087af51fe 100644 --- a/install-appveyor-reqs.cmd +++ b/appveyor-reqs-install.cmd @@ -1,11 +1,12 @@ @echo on + if NOT EXIST C:\projects\deps ( mkdir C:\projects\deps ) cd C:\projects\deps ::########################################################################### -:: Setup the path to Clang-cl +:: Setup Compiler ::########################################################################### if NOT EXIST llvm-installer.exe ( appveyor DownloadFile http://llvm.org/pre-releases/win-snapshots/LLVM-5.0.0-r301646-win32.exe -FileName llvm-installer.exe @@ -13,8 +14,13 @@ if NOT EXIST llvm-installer.exe ( if "%CLANG_VERSION%"=="ToT" ( START /WAIT llvm-installer.exe /S /D=C:\"Program Files\LLVM" ) -@set PATH="C:\Program Files\LLVM\bin";%PATH% -clang-cl -v +if DEFINED CLANG_VERSION @set PATH="C:\Program Files\LLVM\bin";%PATH% +if DEFINED CLANG_VERSION clang-cl -v + +if DEFINED MINGW_PATH rename "C:\Program Files\Git\usr\bin\sh.exe" "sh-ignored.exe" +if DEFINED MINGW_PATH @set "PATH=%PATH:C:\Program Files (x86)\Git\bin=%" +if DEFINED MINGW_PATH @set "PATH=%PATH%;%MINGW_PATH%" +if DEFINED MINGW_PATH g++ -v ::########################################################################### :: Install a recent CMake diff --git a/appveyor.yml b/appveyor.yml index 1d9e88a87b15..be69a555d778 100644 --- a/appveyor.yml +++ b/appveyor.yml @@ -11,24 +11,35 @@ configuration: environment: matrix: - APPVEYOR_BUILD_WORKER_IMAGE: Visual Studio 2017 + CMAKE_OPTIONS: -DCMAKE_C_COMPILER=clang-cl.exe -DCMAKE_CXX_COMPILER=clang-cl.exe CLANG_VERSION: ToT MSVC_SETUP_PATH: C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Auxiliary\Build\vcvarsall.bat MSVC_SETUP_ARG: x86 + GENERATOR: Ninja + MAKE_PROGRAM: ninja APPVEYOR_SAVE_CACHE_ON_ERROR: true - APPVEYOR_BUILD_WORKER_IMAGE: Visual Studio 2015 + CMAKE_OPTIONS: -DCMAKE_C_COMPILER=clang-cl.exe -DCMAKE_CXX_COMPILER=clang-cl.exe CLANG_VERSION: 4 MSVC_SETUP_PATH: C:\Program Files (x86)\Microsoft Visual Studio 14.0\VC\vcvarsall.bat MSVC_SETUP_ARG: x86_amd64 + GENERATOR: Ninja + MAKE_PROGRAM: ninja + APPVEYOR_SAVE_CACHE_ON_ERROR: true + - APPVEYOR_BUILD_WORKER_IMAGE: Visual Studio 2015 + MINGW_PATH: C:\mingw-w64\i686-6.3.0-posix-dwarf-rt_v5-rev1\mingw32\bin + GENERATOR: MinGW Makefiles + MAKE_PROGRAM: mingw32-make APPVEYOR_SAVE_CACHE_ON_ERROR: true install: ############################################################################ # All external dependencies are installed in C:\projects\deps ############################################################################ - - call "%APPVEYOR_BUILD_FOLDER%\\install-appveyor-reqs.cmd" + - call "%APPVEYOR_BUILD_FOLDER%\\appveyor-reqs-install.cmd" before_build: - - call "%MSVC_SETUP_PATH%" %MSVC_SETUP_ARG% + - if DEFINED MSVC_SETUP_PATH call "%MSVC_SETUP_PATH%" %MSVC_SETUP_ARG% - cd %APPVEYOR_BUILD_FOLDER% build_script: @@ -39,8 +50,7 @@ build_script: ############################################################################# # Configuration Step ############################################################################# - - cmake -G Ninja %extra_cmake_flags% - -DCMAKE_C_COMPILER=clang-cl.exe -DCMAKE_CXX_COMPILER=clang-cl.exe + - cmake -G "%GENERATOR%" %CMAKE_OPTIONS% "-DCMAKE_BUILD_TYPE=%configuration%" "-DLLVM_PATH=C:\projects\deps\llvm" -DLIBCXX_ENABLE_EXPERIMENTAL_LIBRARY=OFF -DLLVM_LIT_ARGS="-sv --show-xfail --show-unsupported" @@ -49,10 +59,10 @@ build_script: ############################################################################# # Build Step ############################################################################# - - ninja + - "%MAKE_PROGRAM%" test_script: - - ninja check-cxx + - "%MAKE_PROGRAM% check-cxx" on_failure: - appveyor PushArtifact CMakeFiles/CMakeOutput.log diff --git a/docs/CMakeLists.txt b/docs/CMakeLists.txt index f63ee00a5a02..d679761a5adb 100644 --- a/docs/CMakeLists.txt +++ b/docs/CMakeLists.txt @@ -1,9 +1,9 @@ if (LLVM_ENABLE_SPHINX) + include(AddSphinxTarget) if (SPHINX_FOUND) - include(AddSphinxTarget) if (${SPHINX_OUTPUT_HTML}) add_sphinx_target(html libcxx) endif() endif() -endif() \ No newline at end of file +endif() diff --git a/docs/TestingLibcxx.rst b/docs/TestingLibcxx.rst index e4292764c4d0..5c48ebe61ddf 100644 --- a/docs/TestingLibcxx.rst +++ b/docs/TestingLibcxx.rst @@ -119,7 +119,7 @@ configuration. Passing the option on the command line will override the default. .. option:: libcxx_site_config= Specify the site configuration to use when running the tests. This option - overrides the enviroment variable LIBCXX_SITE_CONFIG. + overrides the environment variable LIBCXX_SITE_CONFIG. .. option:: cxx_headers= diff --git a/include/__bsd_locale_fallbacks.h b/include/__bsd_locale_fallbacks.h index cbc8ad226fd2..3425ce659b7e 100644 --- a/include/__bsd_locale_fallbacks.h +++ b/include/__bsd_locale_fallbacks.h @@ -19,27 +19,24 @@ _LIBCPP_BEGIN_NAMESPACE_STD -typedef _VSTD::remove_pointer::type __use_locale_struct; -typedef _VSTD::unique_ptr<__use_locale_struct, decltype(&uselocale)> __locale_raii; - inline _LIBCPP_ALWAYS_INLINE decltype(MB_CUR_MAX) __libcpp_mb_cur_max_l(locale_t __l) { - __locale_raii __current( uselocale(__l), uselocale ); + __libcpp_locale_guard __current(__l); return MB_CUR_MAX; } inline _LIBCPP_ALWAYS_INLINE wint_t __libcpp_btowc_l(int __c, locale_t __l) { - __locale_raii __current( uselocale(__l), uselocale ); + __libcpp_locale_guard __current(__l); return btowc(__c); } inline _LIBCPP_ALWAYS_INLINE int __libcpp_wctob_l(wint_t __c, locale_t __l) { - __locale_raii __current( uselocale(__l), uselocale ); + __libcpp_locale_guard __current(__l); return wctob(__c); } @@ -47,14 +44,14 @@ inline _LIBCPP_ALWAYS_INLINE size_t __libcpp_wcsnrtombs_l(char *__dest, const wchar_t **__src, size_t __nwc, size_t __len, mbstate_t *__ps, locale_t __l) { - __locale_raii __current( uselocale(__l), uselocale ); + __libcpp_locale_guard __current(__l); return wcsnrtombs(__dest, __src, __nwc, __len, __ps); } inline _LIBCPP_ALWAYS_INLINE size_t __libcpp_wcrtomb_l(char *__s, wchar_t __wc, mbstate_t *__ps, locale_t __l) { - __locale_raii __current( uselocale(__l), uselocale ); + __libcpp_locale_guard __current(__l); return wcrtomb(__s, __wc, __ps); } @@ -62,7 +59,7 @@ inline _LIBCPP_ALWAYS_INLINE size_t __libcpp_mbsnrtowcs_l(wchar_t * __dest, const char **__src, size_t __nms, size_t __len, mbstate_t *__ps, locale_t __l) { - __locale_raii __current( uselocale(__l), uselocale ); + __libcpp_locale_guard __current(__l); return mbsnrtowcs(__dest, __src, __nms, __len, __ps); } @@ -70,28 +67,28 @@ inline _LIBCPP_ALWAYS_INLINE size_t __libcpp_mbrtowc_l(wchar_t *__pwc, const char *__s, size_t __n, mbstate_t *__ps, locale_t __l) { - __locale_raii __current( uselocale(__l), uselocale ); + __libcpp_locale_guard __current(__l); return mbrtowc(__pwc, __s, __n, __ps); } inline _LIBCPP_ALWAYS_INLINE int __libcpp_mbtowc_l(wchar_t *__pwc, const char *__pmb, size_t __max, locale_t __l) { - __locale_raii __current( uselocale(__l), uselocale ); + __libcpp_locale_guard __current(__l); return mbtowc(__pwc, __pmb, __max); } inline _LIBCPP_ALWAYS_INLINE size_t __libcpp_mbrlen_l(const char *__s, size_t __n, mbstate_t *__ps, locale_t __l) { - __locale_raii __current( uselocale(__l), uselocale ); + __libcpp_locale_guard __current(__l); return mbrlen(__s, __n, __ps); } inline _LIBCPP_ALWAYS_INLINE lconv *__libcpp_localeconv_l(locale_t __l) { - __locale_raii __current( uselocale(__l), uselocale ); + __libcpp_locale_guard __current(__l); return localeconv(); } @@ -99,7 +96,7 @@ inline _LIBCPP_ALWAYS_INLINE size_t __libcpp_mbsrtowcs_l(wchar_t *__dest, const char **__src, size_t __len, mbstate_t *__ps, locale_t __l) { - __locale_raii __current( uselocale(__l), uselocale ); + __libcpp_locale_guard __current(__l); return mbsrtowcs(__dest, __src, __len, __ps); } @@ -107,7 +104,7 @@ inline int __libcpp_snprintf_l(char *__s, size_t __n, locale_t __l, const char *__format, ...) { va_list __va; va_start(__va, __format); - __locale_raii __current( uselocale(__l), uselocale ); + __libcpp_locale_guard __current(__l); int __res = vsnprintf(__s, __n, __format, __va); va_end(__va); return __res; @@ -117,7 +114,7 @@ inline int __libcpp_asprintf_l(char **__s, locale_t __l, const char *__format, ...) { va_list __va; va_start(__va, __format); - __locale_raii __current( uselocale(__l), uselocale ); + __libcpp_locale_guard __current(__l); int __res = vasprintf(__s, __format, __va); va_end(__va); return __res; @@ -127,7 +124,7 @@ inline int __libcpp_sscanf_l(const char *__s, locale_t __l, const char *__format, ...) { va_list __va; va_start(__va, __format); - __locale_raii __current( uselocale(__l), uselocale ); + __libcpp_locale_guard __current(__l); int __res = vsscanf(__s, __format, __va); va_end(__va); return __res; diff --git a/include/__config b/include/__config index 2a2907494b2b..4ad700e234c2 100644 --- a/include/__config +++ b/include/__config @@ -129,6 +129,12 @@ #define __has_keyword(__x) !(__is_identifier(__x)) +#ifdef __has_include +#define __libcpp_has_include(__x) __has_include(__x) +#else +#define __libcpp_has_include(__x) 0 +#endif + #if defined(__clang__) #define _LIBCPP_COMPILER_CLANG # ifndef __apple_build_version__ @@ -968,7 +974,7 @@ _LIBCPP_FUNC_VIS extern "C" void __sanitizer_annotate_contiguous_container( # if defined(__GNUC__) && ((__GNUC__ >= 5) || (__GNUC__ == 4 && \ (__GNUC_MINOR__ >= 3 || __GNUC_PATCHLEVEL__ >= 2))) && !defined(__GXX_RTTI) # define _LIBCPP_NO_RTTI -# elif defined(_LIBCPP_MSVC) && !defined(_CPPRTTI) +# elif defined(_LIBCPP_COMPILER_MSVC) && !defined(_CPPRTTI) # define _LIBCPP_NO_RTTI # endif #endif @@ -980,6 +986,7 @@ _LIBCPP_FUNC_VIS extern "C" void __sanitizer_annotate_contiguous_container( // Thread API #if !defined(_LIBCPP_HAS_NO_THREADS) && \ !defined(_LIBCPP_HAS_THREAD_API_PTHREAD) && \ + !defined(_LIBCPP_HAS_THREAD_API_WIN32) && \ !defined(_LIBCPP_HAS_THREAD_API_EXTERNAL) # if defined(__FreeBSD__) || \ defined(__Fuchsia__) || \ @@ -987,7 +994,8 @@ _LIBCPP_FUNC_VIS extern "C" void __sanitizer_annotate_contiguous_container( defined(__linux__) || \ defined(__APPLE__) || \ defined(__CloudABI__) || \ - defined(__sun__) + defined(__sun__) || \ + (defined(__MINGW32__) && __libcpp_has_include()) # define _LIBCPP_HAS_THREAD_API_PTHREAD # elif defined(_LIBCPP_WIN32API) # define _LIBCPP_HAS_THREAD_API_WIN32 diff --git a/include/__locale b/include/__locale index 4184e7e03489..cf3ba23b9092 100644 --- a/include/__locale +++ b/include/__locale @@ -49,6 +49,25 @@ _LIBCPP_BEGIN_NAMESPACE_STD +#if !defined(_LIBCPP_LOCALE__L_EXTENSIONS) || defined(_LIBCPP_MSVCRT) +struct __libcpp_locale_guard { + _LIBCPP_INLINE_VISIBILITY + __libcpp_locale_guard(locale_t& __loc) : __old_loc_(uselocale(__loc)) {} + + _LIBCPP_INLINE_VISIBILITY + ~__libcpp_locale_guard() { + if (__old_loc_) + uselocale(__old_loc_); + } + + locale_t __old_loc_; +private: + __libcpp_locale_guard(__libcpp_locale_guard const&); + __libcpp_locale_guard& operator=(__libcpp_locale_guard const&); +}; +#endif + + class _LIBCPP_TYPE_VIS locale; template diff --git a/include/__mutex_base b/include/__mutex_base index a6d5e79c4c86..7f5e2ea2810e 100644 --- a/include/__mutex_base +++ b/include/__mutex_base @@ -15,6 +15,7 @@ #include #include #include <__threading_support> +#include <__undef_min_max> #if !defined(_LIBCPP_HAS_NO_PRAGMA_SYSTEM_HEADER) #pragma GCC system_header diff --git a/include/__threading_support b/include/__threading_support index 080ebd256b8f..385fff32b350 100644 --- a/include/__threading_support +++ b/include/__threading_support @@ -30,6 +30,7 @@ #include #include #include +#include <__undef_min_max> #endif #if defined(_LIBCPP_HAS_THREAD_LIBRARY_EXTERNAL) || \ diff --git a/include/__undef_min_max b/include/__undef_min_max index f4ca091def45..71db3965e3cd 100644 --- a/include/__undef_min_max +++ b/include/__undef_min_max @@ -10,7 +10,7 @@ #ifdef min #if !defined(_LIBCPP_DISABLE_MACRO_CONFLICT_WARNINGS) -#if defined(_LIBCPP_MSVC) +#if defined(_LIBCPP_WARNING) _LIBCPP_WARNING("macro min is incompatible with C++. Try #define NOMINMAX " "before any Windows header. #undefing min") #else @@ -22,7 +22,7 @@ _LIBCPP_WARNING("macro min is incompatible with C++. Try #define NOMINMAX " #ifdef max #if !defined(_LIBCPP_DISABLE_MACRO_CONFLICT_WARNINGS) -#if defined(_LIBCPP_MSVC) +#if defined(_LIBCPP_WARNING) _LIBCPP_WARNING("macro max is incompatible with C++. Try #define NOMINMAX " "before any Windows header. #undefing max") #else diff --git a/include/algorithm b/include/algorithm index c3517a11bb0d..08ca23ff6168 100644 --- a/include/algorithm +++ b/include/algorithm @@ -644,8 +644,8 @@ template #if defined(__IBMCPP__) #include "support/ibm/support.h" #endif -#if defined(_LIBCPP_MSVCRT) || defined(__MINGW32__) -#include "support/win32/support.h" +#if defined(_LIBCPP_COMPILER_MSVC) +#include #endif #include <__undef_min_max> @@ -783,51 +783,132 @@ struct __debug_less // Precondition: __x != 0 inline _LIBCPP_INLINE_VISIBILITY -unsigned -__ctz(unsigned __x) -{ +unsigned __ctz(unsigned __x) { +#ifndef _LIBCPP_COMPILER_MSVC return static_cast(__builtin_ctz(__x)); +#else + static_assert(sizeof(unsigned) == sizeof(unsigned long), ""); + static_assert(sizeof(unsigned long) == 4, ""); + unsigned long where; + // Search from LSB to MSB for first set bit. + // Returns zero if no set bit is found. + if (_BitScanForward(&where, mask)) + return where; + return 32; +#endif } inline _LIBCPP_INLINE_VISIBILITY -unsigned long -__ctz(unsigned long __x) -{ +unsigned long __ctz(unsigned long __x) { +#ifndef _LIBCPP_COMPILER_MSVC return static_cast(__builtin_ctzl(__x)); +#else + static_assert(sizeof(unsigned long) == sizeof(unsigned), ""); + return __ctz(static_cast(__x)); +#endif } inline _LIBCPP_INLINE_VISIBILITY -unsigned long long -__ctz(unsigned long long __x) -{ +unsigned long long __ctz(unsigned long long __x) { +#ifndef _LIBCPP_COMPILER_MSVC return static_cast(__builtin_ctzll(__x)); +#else + unsigned long where; +// Search from LSB to MSB for first set bit. +// Returns zero if no set bit is found. +#if defined(_LIBCPP_HAS_BITSCAN64) + (defined(_M_AMD64) || defined(__x86_64__)) + if (_BitScanForward64(&where, mask)) + return static_cast(where); +#else + // Win32 doesn't have _BitScanForward64 so emulate it with two 32 bit calls. + // Scan the Low Word. + if (_BitScanForward(&where, static_cast(mask))) + return where; + // Scan the High Word. + if (_BitScanForward(&where, static_cast(mask >> 32))) + return where + 32; // Create a bit offset from the LSB. +#endif + return 64; +#endif // _LIBCPP_COMPILER_MSVC } // Precondition: __x != 0 inline _LIBCPP_INLINE_VISIBILITY -unsigned -__clz(unsigned __x) -{ +unsigned __clz(unsigned __x) { +#ifndef _LIBCPP_COMPILER_MSVC return static_cast(__builtin_clz(__x)); +#else + static_assert(sizeof(unsigned) == sizeof(unsigned long), ""); + static_assert(sizeof(unsigned long) == 4, ""); + unsigned long where; + // Search from LSB to MSB for first set bit. + // Returns zero if no set bit is found. + if (_BitScanReverse(&where, mask)) + return 31 - where; + return 32; // Undefined Behavior. +#endif } inline _LIBCPP_INLINE_VISIBILITY -unsigned long -__clz(unsigned long __x) -{ +unsigned long __clz(unsigned long __x) { +#ifndef _LIBCPP_COMPILER_MSVC return static_cast(__builtin_clzl (__x)); +#else + static_assert(sizeof(unsigned) == sizeof(unsigned long), ""); + return __clz(static_cast(__x)); +#endif } inline _LIBCPP_INLINE_VISIBILITY -unsigned long long -__clz(unsigned long long __x) -{ +unsigned long long __clz(unsigned long long __x) { +#ifndef _LIBCPP_COMPILER_MSVC return static_cast(__builtin_clzll(__x)); +#else + unsigned long where; +// BitScanReverse scans from MSB to LSB for first set bit. +// Returns 0 if no set bit is found. +#if defined(_LIBCPP_HAS_BITSCAN64) + if (_BitScanReverse64(&where, mask)) + return static_cast(63 - where); +#else + // Scan the high 32 bits. + if (_BitScanReverse(&where, static_cast(mask >> 32))) + return 63 - (where + 32); // Create a bit offset from the MSB. + // Scan the low 32 bits. + if (_BitScanReverse(&where, static_cast(mask))) + return 63 - where; +#endif + return 64; // Undefined Behavior. +#endif // _LIBCPP_COMPILER_MSVC } -inline _LIBCPP_INLINE_VISIBILITY int __pop_count(unsigned __x) {return __builtin_popcount (__x);} -inline _LIBCPP_INLINE_VISIBILITY int __pop_count(unsigned long __x) {return __builtin_popcountl (__x);} -inline _LIBCPP_INLINE_VISIBILITY int __pop_count(unsigned long long __x) {return __builtin_popcountll(__x);} +inline _LIBCPP_INLINE_VISIBILITY int __pop_count(unsigned __x) { +#ifndef _LIBCPP_COMPILER_MSVC + return __builtin_popcount (__x); +#else + static_assert(sizeof(unsigned) == 4, ""); + return __popcnt(__x); +#endif +} + +inline _LIBCPP_INLINE_VISIBILITY int __pop_count(unsigned long __x) { +#ifndef _LIBCPP_COMPILER_MSVC + return __builtin_popcountl (__x); +#else + static_assert(sizeof(unsigned long) == 4, ""); + return __popcnt(__x); +#endif +} + +inline _LIBCPP_INLINE_VISIBILITY int __pop_count(unsigned long long __x) { +#ifndef _LIBCPP_COMPILER_MSVC + return __builtin_popcountll(__x); +#else + static_assert(sizeof(unsigned long long) == 8, ""); + return __popcnt64(__x); +#endif +} // all_of diff --git a/include/ctype.h b/include/ctype.h index 22d6c49be9e1..e97ff3c48876 100644 --- a/include/ctype.h +++ b/include/ctype.h @@ -40,15 +40,6 @@ int toupper(int c); #ifdef __cplusplus -#if defined(_LIBCPP_MSVCRT) -// We support including .h headers inside 'extern "C"' contexts, so switch -// back to C++ linkage before including these C++ headers. -extern "C++" { - #include "support/win32/support.h" - #include "support/win32/locale_win32.h" -} -#endif // _LIBCPP_MSVCRT - #undef isalnum #undef isalpha #undef isblank diff --git a/include/experimental/numeric b/include/experimental/numeric index d1209dbec15b..6488a68eca6c 100644 --- a/include/experimental/numeric +++ b/include/experimental/numeric @@ -66,11 +66,11 @@ struct __abs<_Result, _Source, false> { template -_LIBCPP_CONSTEXPR _LIBCPP_INLINE_VISIBILITY -_Tp __gcd(_Tp __m, _Tp __n) +_LIBCPP_CONSTEXPR _LIBCPP_HIDDEN +inline _Tp __gcd(_Tp __m, _Tp __n) { static_assert((!is_signed<_Tp>::value), "" ); - return __n == 0 ? __m : __gcd<_Tp>(__n, __m % __n); + return __n == 0 ? __m : _VSTD_LFTS_V2::__gcd<_Tp>(__n, __m % __n); } @@ -84,8 +84,9 @@ gcd(_Tp __m, _Up __n) static_assert((!is_same::type, bool>::value), "Second argument to gcd cannot be bool" ); using _Rp = common_type_t<_Tp,_Up>; using _Wp = make_unsigned_t<_Rp>; - return static_cast<_Rp>(__gcd(static_cast<_Wp>(__abs<_Rp, _Tp>()(__m)), - static_cast<_Wp>(__abs<_Rp, _Up>()(__n)))); + return static_cast<_Rp>(_VSTD_LFTS_V2::__gcd( + static_cast<_Wp>(__abs<_Rp, _Tp>()(__m)), + static_cast<_Wp>(__abs<_Rp, _Up>()(__n)))); } template @@ -100,7 +101,7 @@ lcm(_Tp __m, _Up __n) return 0; using _Rp = common_type_t<_Tp,_Up>; - _Rp __val1 = __abs<_Rp, _Tp>()(__m) / gcd(__m, __n); + _Rp __val1 = __abs<_Rp, _Tp>()(__m) / _VSTD_LFTS_V2::gcd(__m, __n); _Rp __val2 = __abs<_Rp, _Up>()(__n); _LIBCPP_ASSERT((numeric_limits<_Rp>::max() / __val1 > __val2), "Overflow in lcm"); return __val1 * __val2; diff --git a/include/ext/hash_map b/include/ext/hash_map index 66f2b11c0594..998e8f65994e 100644 --- a/include/ext/hash_map +++ b/include/ext/hash_map @@ -207,7 +207,7 @@ template #include #if __DEPRECATED -#if defined(_LIBCPP_MSVC) +#if defined(_LIBCPP_WARNING) _LIBCPP_WARNING("Use of the header is deprecated. Migrate to ") #else # warning Use of the header is deprecated. Migrate to diff --git a/include/ext/hash_set b/include/ext/hash_set index 916ed6910d7d..38f81ed3b5c8 100644 --- a/include/ext/hash_set +++ b/include/ext/hash_set @@ -199,7 +199,7 @@ template #include #if __DEPRECATED -#if defined(_LIBCPP_MSVC) +#if defined(_LIBCPP_WARNING) _LIBCPP_WARNING("Use of the header is deprecated. Migrate to ") #else # warning Use of the header is deprecated. Migrate to diff --git a/include/limits b/include/limits index 609c4d4ed95a..4755c57cf903 100644 --- a/include/limits +++ b/include/limits @@ -111,8 +111,8 @@ template<> class numeric_limits; #include <__undef_min_max> -#if defined(_LIBCPP_MSVCRT) -#include "support/win32/limits_win32.h" +#if defined(_LIBCPP_COMPILER_MSVC) +#include "support/win32/limits_msvc_win32.h" #endif // _LIBCPP_MSVCRT #if defined(__IBMCPP__) diff --git a/include/locale b/include/locale index ad1c1f0083ec..d29a2dc70a5b 100644 --- a/include/locale +++ b/include/locale @@ -233,9 +233,6 @@ _LIBCPP_BEGIN_NAMESPACE_STD #define __cloc_defined #endif -typedef _VSTD::remove_pointer::type __locale_struct; -typedef _VSTD::unique_ptr<__locale_struct, decltype(&freelocale)> __locale_unique_ptr; - // __scan_keyword // Scans [__b, __e) until a match is found in the basic_strings range // [__kb, __ke) or until it can be shown that there is no match in [__kb, __ke). diff --git a/include/memory b/include/memory index 41ab01b46f7e..4201c92dd723 100644 --- a/include/memory +++ b/include/memory @@ -996,11 +996,11 @@ struct __rebind_pointer { // allocator_traits -namespace __has_pointer_type_imp +struct __has_pointer_type_imp { template static __two __test(...); template static char __test(typename _Up::pointer* = 0); -} +}; template struct __has_pointer_type @@ -3924,7 +3924,10 @@ private: template _LIBCPP_INLINE_VISIBILITY - void + typename enable_if* + >::value, + void>::type __enable_weak_this(const enable_shared_from_this<_Yp>* __e, _OrigPtr* __ptr) _NOEXCEPT { @@ -3943,6 +3946,7 @@ private: template friend class _LIBCPP_TEMPLATE_VIS weak_ptr; }; + template inline _LIBCPP_CONSTEXPR diff --git a/include/numeric b/include/numeric index 8f25146938a4..9c98cdbe262e 100644 --- a/include/numeric +++ b/include/numeric @@ -222,11 +222,11 @@ struct __abs<_Result, _Source, false> { template -_LIBCPP_CONSTEXPR _LIBCPP_INLINE_VISIBILITY +_LIBCPP_CONSTEXPR _LIBCPP_HIDDEN _Tp __gcd(_Tp __m, _Tp __n) { static_assert((!is_signed<_Tp>::value), ""); - return __n == 0 ? __m : __gcd<_Tp>(__n, __m % __n); + return __n == 0 ? __m : _VSTD::__gcd<_Tp>(__n, __m % __n); } @@ -240,8 +240,9 @@ gcd(_Tp __m, _Up __n) static_assert((!is_same::type, bool>::value), "Second argument to gcd cannot be bool" ); using _Rp = common_type_t<_Tp,_Up>; using _Wp = make_unsigned_t<_Rp>; - return static_cast<_Rp>(__gcd(static_cast<_Wp>(__abs<_Rp, _Tp>()(__m)), - static_cast<_Wp>(__abs<_Rp, _Up>()(__n)))); + return static_cast<_Rp>(_VSTD::__gcd( + static_cast<_Wp>(__abs<_Rp, _Tp>()(__m)), + static_cast<_Wp>(__abs<_Rp, _Up>()(__n)))); } template @@ -256,7 +257,7 @@ lcm(_Tp __m, _Up __n) return 0; using _Rp = common_type_t<_Tp,_Up>; - _Rp __val1 = __abs<_Rp, _Tp>()(__m) / gcd(__m, __n); + _Rp __val1 = __abs<_Rp, _Tp>()(__m) / _VSTD::gcd(__m, __n); _Rp __val2 = __abs<_Rp, _Up>()(__n); _LIBCPP_ASSERT((numeric_limits<_Rp>::max() / __val1 > __val2), "Overflow in lcm"); return __val1 * __val2; diff --git a/include/stdio.h b/include/stdio.h index 56fb2d83bb28..dc534970f9d8 100644 --- a/include/stdio.h +++ b/include/stdio.h @@ -111,8 +111,9 @@ void perror(const char* s); // snprintf #if defined(_LIBCPP_MSVCRT) -extern "C++" { -#include "support/win32/support.h" +extern "C" { +int vasprintf(char **sptr, const char *__restrict fmt, va_list ap); +int asprintf(char **sptr, const char *__restrict fmt, ...); } #endif diff --git a/include/stdlib.h b/include/stdlib.h index 12fd676a15f5..f11c5e76226e 100644 --- a/include/stdlib.h +++ b/include/stdlib.h @@ -97,10 +97,6 @@ void *aligned_alloc(size_t alignment, size_t size); // C11 extern "C++" { -#ifdef _LIBCPP_MSVCRT -#include "support/win32/locale_win32.h" -#endif // _LIBCPP_MSVCRT - #undef abs #undef div #undef labs diff --git a/include/string_view b/include/string_view index 5c42b36ca565..e59f099f16e7 100644 --- a/include/string_view +++ b/include/string_view @@ -103,7 +103,6 @@ namespace std { constexpr const_pointer data() const noexcept; // 7.7, basic_string_view modifiers - constexpr void clear() noexcept; constexpr void remove_prefix(size_type n); constexpr void remove_suffix(size_type n); constexpr void swap(basic_string_view& s) noexcept; @@ -292,13 +291,6 @@ public: const_pointer data() const _NOEXCEPT { return __data; } // [string.view.modifiers], modifiers: - _LIBCPP_CONSTEXPR_AFTER_CXX11 _LIBCPP_INLINE_VISIBILITY - void clear() _NOEXCEPT - { - __data = nullptr; - __size = 0; - } - _LIBCPP_CONSTEXPR_AFTER_CXX11 _LIBCPP_INLINE_VISIBILITY void remove_prefix(size_type __n) _NOEXCEPT { diff --git a/include/support/win32/limits_win32.h b/include/support/win32/limits_msvc_win32.h similarity index 80% rename from include/support/win32/limits_win32.h rename to include/support/win32/limits_msvc_win32.h index 406cd3025bf8..1ab2e0b6dc1b 100644 --- a/include/support/win32/limits_win32.h +++ b/include/support/win32/limits_msvc_win32.h @@ -1,5 +1,5 @@ // -*- C++ -*- -//===--------------------- support/win32/limits_win32.h -------------------===// +//===------------------ support/win32/limits_msvc_win32.h -----------------===// // // The LLVM Compiler Infrastructure // @@ -8,17 +8,21 @@ // //===----------------------------------------------------------------------===// -#ifndef _LIBCPP_SUPPORT_WIN32_LIMITS_WIN32_H -#define _LIBCPP_SUPPORT_WIN32_LIMITS_WIN32_H +#ifndef _LIBCPP_SUPPORT_WIN32_LIMITS_MSVC_WIN32_H +#define _LIBCPP_SUPPORT_WIN32_LIMITS_MSVC_WIN32_H #if !defined(_LIBCPP_MSVCRT) #error "This header complements the Microsoft C Runtime library, and should not be included otherwise." -#else +#endif +#if defined(__clang__) +#error "This header should only be included when using Microsofts C1XX frontend" +#endif #include // CHAR_BIT #include // limit constants +#include // HUGE_VAL +#include // internal MSVC header providing the needed functionality -#if ! defined(__clang__) #define __CHAR_BIT__ CHAR_BIT #define __FLT_MANT_DIG__ FLT_MANT_DIG @@ -61,19 +65,8 @@ #define __LDBL_DENORM_MIN__ 3.64519953188247460253e-4951L // __builtin replacements/workarounds -#include // HUGE_VAL -#include // internal MSVC header providing the needed functionality -#define __builtin_huge_val() HUGE_VAL -#define __builtin_huge_valf() _FInf._Float #define __builtin_huge_vall() _LInf._Long_double -#define __builtin_nan(__dummy) _Nan._Double -#define __builtin_nanf(__dummy) _FNan._Float #define __builtin_nanl(__dummmy) _LNan._Long_double -#define __builtin_nans(__dummy) _Snan._Double -#define __builtin_nansf(__dummy) _FSnan._Float #define __builtin_nansl(__dummy) _LSnan._Long_double -#endif // ! defined(__clang__) -#endif // _LIBCPP_MSVCRT - -#endif // _LIBCPP_SUPPORT_WIN32_LIMITS_WIN32_H +#endif // _LIBCPP_SUPPORT_WIN32_LIMITS_MSVC_WIN32_H diff --git a/include/support/win32/locale_mgmt_win32.h b/include/support/win32/locale_mgmt_win32.h deleted file mode 100644 index b3316d62596f..000000000000 --- a/include/support/win32/locale_mgmt_win32.h +++ /dev/null @@ -1,33 +0,0 @@ -// -*- C++ -*- -//===----------------- support/win32/locale_mgmt_win32.h ------------------===// -// -// The LLVM Compiler Infrastructure -// -// This file is dual licensed under the MIT and the University of Illinois Open -// Source Licenses. See LICENSE.TXT for details. -// -//===----------------------------------------------------------------------===// - -#ifndef _LIBCPP_SUPPORT_WIN32_LOCALE_MGMT_WIN32_H -#define _LIBCPP_SUPPORT_WIN32_LOCALE_MGMT_WIN32_H - -#include // _locale_t -#define locale_t _locale_t -#define LC_COLLATE_MASK _M_COLLATE -#define LC_CTYPE_MASK _M_CTYPE -#define LC_MONETARY_MASK _M_MONETARY -#define LC_NUMERIC_MASK _M_NUMERIC -#define LC_TIME_MASK _M_TIME -#define LC_MESSAGES_MASK _M_MESSAGES -#define LC_ALL_MASK ( LC_COLLATE_MASK \ - | LC_CTYPE_MASK \ - | LC_MESSAGES_MASK \ - | LC_MONETARY_MASK \ - | LC_NUMERIC_MASK \ - | LC_TIME_MASK ) -#define freelocale _free_locale -// FIXME: base currently unused. Needs manual work to construct the new locale -locale_t newlocale( int mask, const char * locale, locale_t base ); -locale_t uselocale( locale_t newloc ); - -#endif // _LIBCPP_SUPPORT_WIN32_LOCALE_MGMT_WIN32_H diff --git a/include/support/win32/locale_win32.h b/include/support/win32/locale_win32.h index bc717d97990a..7a6c44ca9003 100644 --- a/include/support/win32/locale_win32.h +++ b/include/support/win32/locale_win32.h @@ -12,9 +12,30 @@ #define _LIBCPP_SUPPORT_WIN32_LOCALE_WIN32_H #include <__config> -#include "support/win32/support.h" -#include "support/win32/locale_mgmt_win32.h" #include +#include // _locale_t + +#define LC_COLLATE_MASK _M_COLLATE +#define LC_CTYPE_MASK _M_CTYPE +#define LC_MONETARY_MASK _M_MONETARY +#define LC_NUMERIC_MASK _M_NUMERIC +#define LC_TIME_MASK _M_TIME +#define LC_MESSAGES_MASK _M_MESSAGES +#define LC_ALL_MASK ( LC_COLLATE_MASK \ + | LC_CTYPE_MASK \ + | LC_MESSAGES_MASK \ + | LC_MONETARY_MASK \ + | LC_NUMERIC_MASK \ + | LC_TIME_MASK ) + +#define locale_t _locale_t + +// Locale management functions +#define freelocale _free_locale +// FIXME: base currently unused. Needs manual work to construct the new locale +locale_t newlocale( int mask, const char * locale, locale_t base ); +locale_t uselocale( locale_t newloc ); + lconv *localeconv_l( locale_t loc ); size_t mbrlen_l( const char *__restrict s, size_t n, @@ -88,7 +109,6 @@ _LIBCPP_FUNC_VIS int snprintf_l(char *ret, size_t n, locale_t loc, const char *f _LIBCPP_FUNC_VIS int asprintf_l( char **ret, locale_t loc, const char *format, ... ); _LIBCPP_FUNC_VIS int vasprintf_l( char **ret, locale_t loc, const char *format, va_list ap ); - // not-so-pressing FIXME: use locale to determine blank characters inline int isblank_l( int c, locale_t /*loc*/ ) { @@ -99,10 +119,4 @@ inline int iswblank_l( wint_t c, locale_t /*loc*/ ) return ( c == L' ' || c == L'\t' ); } -#if defined(_LIBCPP_MSVCRT) -inline int isblank( int c, locale_t /*loc*/ ) -{ return ( c == ' ' || c == '\t' ); } -inline int iswblank( wint_t c, locale_t /*loc*/ ) -{ return ( c == L' ' || c == L'\t' ); } -#endif // _LIBCPP_MSVCRT #endif // _LIBCPP_SUPPORT_WIN32_LOCALE_WIN32_H diff --git a/include/support/win32/support.h b/include/support/win32/support.h deleted file mode 100644 index e48b08ddad03..000000000000 --- a/include/support/win32/support.h +++ /dev/null @@ -1,177 +0,0 @@ -// -*- C++ -*- -//===----------------------- support/win32/support.h ----------------------===// -// -// The LLVM Compiler Infrastructure -// -// This file is dual licensed under the MIT and the University of Illinois Open -// Source Licenses. See LICENSE.TXT for details. -// -//===----------------------------------------------------------------------===// - -#ifndef _LIBCPP_SUPPORT_WIN32_SUPPORT_H -#define _LIBCPP_SUPPORT_WIN32_SUPPORT_H - -// Functions and constants used in libc++ that -// are missing from the Windows C library. - -#include // mbstate_t -#include // va_ macros -// "builtins" not implemented here for Clang or GCC as they provide -// implementations. Assuming required for elsewhere else, certainly MSVC. -#if defined(_LIBCPP_COMPILER_MSVC) -#include -#endif -#define swprintf _snwprintf -#define vswprintf _vsnwprintf - -#ifndef NOMINMAX -#define NOMINMAX -#endif - -// The mingw headers already define these as static. -#ifndef __MINGW32__ -extern "C" { - -int vasprintf(char **sptr, const char *__restrict fmt, va_list ap); -int asprintf(char **sptr, const char *__restrict fmt, ...); -size_t mbsnrtowcs(wchar_t *__restrict dst, const char **__restrict src, - size_t nmc, size_t len, mbstate_t *__restrict ps); -size_t wcsnrtombs(char *__restrict dst, const wchar_t **__restrict src, - size_t nwc, size_t len, mbstate_t *__restrict ps); -} -#endif // __MINGW32__ - -#if defined(_LIBCPP_COMPILER_MSVC) - -// Bit builtin's make these assumptions when calling _BitScanForward/Reverse -// etc. These assumptions are expected to be true for Win32/Win64 which this -// file supports. -static_assert(sizeof(unsigned long long) == 8, ""); -static_assert(sizeof(unsigned long) == 4, ""); -static_assert(sizeof(unsigned int) == 4, ""); - -_LIBCPP_ALWAYS_INLINE int __builtin_popcount(unsigned int x) -{ - // Binary: 0101... - static const unsigned int m1 = 0x55555555; - // Binary: 00110011.. - static const unsigned int m2 = 0x33333333; - // Binary: 4 zeros, 4 ones ... - static const unsigned int m4 = 0x0f0f0f0f; - // The sum of 256 to the power of 0,1,2,3... - static const unsigned int h01 = 0x01010101; - // Put count of each 2 bits into those 2 bits. - x -= (x >> 1) & m1; - // Put count of each 4 bits into those 4 bits. - x = (x & m2) + ((x >> 2) & m2); - // Put count of each 8 bits into those 8 bits. - x = (x + (x >> 4)) & m4; - // Returns left 8 bits of x + (x<<8) + (x<<16) + (x<<24). - return (x * h01) >> 24; -} - -_LIBCPP_ALWAYS_INLINE int __builtin_popcountl(unsigned long x) -{ - return __builtin_popcount(static_cast(x)); -} - -_LIBCPP_ALWAYS_INLINE int __builtin_popcountll(unsigned long long x) -{ - // Binary: 0101... - static const unsigned long long m1 = 0x5555555555555555; - // Binary: 00110011.. - static const unsigned long long m2 = 0x3333333333333333; - // Binary: 4 zeros, 4 ones ... - static const unsigned long long m4 = 0x0f0f0f0f0f0f0f0f; - // The sum of 256 to the power of 0,1,2,3... - static const unsigned long long h01 = 0x0101010101010101; - // Put count of each 2 bits into those 2 bits. - x -= (x >> 1) & m1; - // Put count of each 4 bits into those 4 bits. - x = (x & m2) + ((x >> 2) & m2); - // Put count of each 8 bits into those 8 bits. - x = (x + (x >> 4)) & m4; - // Returns left 8 bits of x + (x<<8) + (x<<16) + (x<<24) + ... - return static_cast((x * h01) >> 56); -} - -// Returns the number of trailing 0-bits in x, starting at the least significant -// bit position. If x is 0, the result is undefined. -_LIBCPP_ALWAYS_INLINE int __builtin_ctzll(unsigned long long mask) -{ - unsigned long where; -// Search from LSB to MSB for first set bit. -// Returns zero if no set bit is found. -#if defined(_LIBCPP_HAS_BITSCAN64) - (defined(_M_AMD64) || defined(__x86_64__)) - if (_BitScanForward64(&where, mask)) - return static_cast(where); -#else - // Win32 doesn't have _BitScanForward64 so emulate it with two 32 bit calls. - // Scan the Low Word. - if (_BitScanForward(&where, static_cast(mask))) - return static_cast(where); - // Scan the High Word. - if (_BitScanForward(&where, static_cast(mask >> 32))) - return static_cast(where + 32); // Create a bit offset from the LSB. -#endif - return 64; -} - -_LIBCPP_ALWAYS_INLINE int __builtin_ctzl(unsigned long mask) -{ - unsigned long where; - // Search from LSB to MSB for first set bit. - // Returns zero if no set bit is found. - if (_BitScanForward(&where, mask)) - return static_cast(where); - return 32; -} - -_LIBCPP_ALWAYS_INLINE int __builtin_ctz(unsigned int mask) -{ - // Win32 and Win64 expectations. - static_assert(sizeof(mask) == 4, ""); - static_assert(sizeof(unsigned long) == 4, ""); - return __builtin_ctzl(static_cast(mask)); -} - -// Returns the number of leading 0-bits in x, starting at the most significant -// bit position. If x is 0, the result is undefined. -_LIBCPP_ALWAYS_INLINE int __builtin_clzll(unsigned long long mask) -{ - unsigned long where; -// BitScanReverse scans from MSB to LSB for first set bit. -// Returns 0 if no set bit is found. -#if defined(_LIBCPP_HAS_BITSCAN64) - if (_BitScanReverse64(&where, mask)) - return static_cast(63 - where); -#else - // Scan the high 32 bits. - if (_BitScanReverse(&where, static_cast(mask >> 32))) - return static_cast(63 - - (where + 32)); // Create a bit offset from the MSB. - // Scan the low 32 bits. - if (_BitScanReverse(&where, static_cast(mask))) - return static_cast(63 - where); -#endif - return 64; // Undefined Behavior. -} - -_LIBCPP_ALWAYS_INLINE int __builtin_clzl(unsigned long mask) -{ - unsigned long where; - // Search from LSB to MSB for first set bit. - // Returns zero if no set bit is found. - if (_BitScanReverse(&where, mask)) - return static_cast(31 - where); - return 32; // Undefined Behavior. -} - -_LIBCPP_ALWAYS_INLINE int __builtin_clz(unsigned int x) -{ - return __builtin_clzl(x); -} -#endif // _LIBCPP_MSVC - -#endif // _LIBCPP_SUPPORT_WIN32_SUPPORT_H diff --git a/include/variant b/include/variant index 88f7b240d029..ba15ed8c4a14 100644 --- a/include/variant +++ b/include/variant @@ -425,30 +425,21 @@ struct __base { constexpr auto __fmatrix = __make_fmatrix<_Visitor&&, decltype(_VSTD::forward<_Vs>(__vs).__as_base())...>(); - const size_t __indices[] = {__vs.index()...}; - return __at(__fmatrix, __indices)(_VSTD::forward<_Visitor>(__visitor), - _VSTD::forward<_Vs>(__vs).__as_base()...); + return __at(__fmatrix, __vs.index()...)( + _VSTD::forward<_Visitor>(__visitor), + _VSTD::forward<_Vs>(__vs).__as_base()...); } private: template inline _LIBCPP_INLINE_VISIBILITY - static constexpr const _Tp& __at_impl(const _Tp& __elem, const size_t*) { - return __elem; - } + static constexpr const _Tp& __at(const _Tp& __elem) { return __elem; } - template - inline _LIBCPP_INLINE_VISIBILITY - static constexpr auto&& __at_impl(const array<_Tp, _Np>& __elems, - const size_t* __index) { - return __at_impl(__elems[*__index], __index + 1); - } - - template + template inline _LIBCPP_INLINE_VISIBILITY static constexpr auto&& __at(const array<_Tp, _Np>& __elems, - const size_t (&__indices)[_Ip]) { - return __at_impl(__elems, begin(__indices)); + size_t __index, _Indices... __indices) { + return __at(__elems[__index], __indices...); } template @@ -1140,7 +1131,7 @@ public: : __impl(in_place_index<_Ip>, _VSTD::forward<_Arg>(__arg)) {} template = 0, + class = enable_if_t<(_Ip < sizeof...(_Types)), int>, class _Tp = variant_alternative_t<_Ip, variant<_Types...>>, enable_if_t, int> = 0> inline _LIBCPP_INLINE_VISIBILITY diff --git a/include/wchar.h b/include/wchar.h index c0c6ef754fbe..25a318faff25 100644 --- a/include/wchar.h +++ b/include/wchar.h @@ -166,9 +166,12 @@ inline _LIBCPP_INLINE_VISIBILITY _LIBCPP_PREFERRED_OVERLOAD } #endif -#if defined(__cplusplus) && (defined(_LIBCPP_MSVCRT) || defined(__MINGW32__)) -extern "C++" { -#include // pull in *swprintf defines +#if defined(__cplusplus) && defined(_LIBCPP_MSVCRT) +extern "C" { +size_t mbsnrtowcs(wchar_t *__restrict dst, const char **__restrict src, + size_t nmc, size_t len, mbstate_t *__restrict ps); +size_t wcsnrtombs(char *__restrict dst, const wchar_t **__restrict src, + size_t nwc, size_t len, mbstate_t *__restrict ps); } // extern "C++" #endif // __cplusplus && _LIBCPP_MSVCRT diff --git a/src/include/atomic_support.h b/src/include/atomic_support.h index 378541b23256..08847e630705 100644 --- a/src/include/atomic_support.h +++ b/src/include/atomic_support.h @@ -29,7 +29,7 @@ #endif #if !defined(_LIBCPP_HAS_ATOMIC_BUILTINS) && !defined(_LIBCPP_HAS_NO_THREADS) -# if defined(_LIBCPP_MSVC) +# if defined(_LIBCPP_WARNING) _LIBCPP_WARNING("Building libc++ without __atomic builtins is unsupported") # else # warning Building libc++ without __atomic builtins is unsupported diff --git a/src/locale.cpp b/src/locale.cpp index 1ed9b41fd4ad..4163c2c0af65 100644 --- a/src/locale.cpp +++ b/src/locale.cpp @@ -45,6 +45,24 @@ _LIBCPP_BEGIN_NAMESPACE_STD +struct __libcpp_unique_locale { + __libcpp_unique_locale(const char* nm) : __loc_(newlocale(LC_ALL_MASK, nm, 0)) {} + + ~__libcpp_unique_locale() { + if (__loc_) + freelocale(__loc_); + } + + explicit operator bool() const { return __loc_; } + + locale_t& get() { return __loc_; } + + locale_t __loc_; +private: + __libcpp_unique_locale(__libcpp_unique_locale const&); + __libcpp_unique_locale& operator=(__libcpp_unique_locale const&); +}; + #ifdef __cloc_defined locale_t __cloc() { // In theory this could create a race condition. In practice @@ -4185,7 +4203,7 @@ __widen_from_utf8<32>::~__widen_from_utf8() static bool checked_string_to_wchar_convert(wchar_t& dest, const char* ptr, - __locale_struct* loc) { + locale_t loc) { if (*ptr == '\0') return false; mbstate_t mb = {}; @@ -4200,7 +4218,7 @@ static bool checked_string_to_wchar_convert(wchar_t& dest, static bool checked_string_to_char_convert(char& dest, const char* ptr, - __locale_struct* __loc) { + locale_t __loc) { if (*ptr == '\0') return false; if (!ptr[1]) { @@ -4295,8 +4313,8 @@ numpunct_byname::__init(const char* nm) { if (strcmp(nm, "C") != 0) { - __locale_unique_ptr loc(newlocale(LC_ALL_MASK, nm, 0), freelocale); - if (loc == nullptr) + __libcpp_unique_locale loc(nm); + if (!loc) __throw_runtime_error("numpunct_byname::numpunct_byname" " failed to construct for " + string(nm)); @@ -4333,8 +4351,8 @@ numpunct_byname::__init(const char* nm) { if (strcmp(nm, "C") != 0) { - __locale_unique_ptr loc(newlocale(LC_ALL_MASK, nm, 0), freelocale); - if (loc == nullptr) + __libcpp_unique_locale loc(nm); + if (!loc) __throw_runtime_error("numpunct_byname::numpunct_byname" " failed to construct for " + string(nm)); @@ -5820,8 +5838,8 @@ void moneypunct_byname::init(const char* nm) { typedef moneypunct base; - __locale_unique_ptr loc(newlocale(LC_ALL_MASK, nm, 0), freelocale); - if (loc == nullptr) + __libcpp_unique_locale loc(nm); + if (!loc) __throw_runtime_error("moneypunct_byname" " failed to construct for " + string(nm)); @@ -5864,8 +5882,8 @@ void moneypunct_byname::init(const char* nm) { typedef moneypunct base; - __locale_unique_ptr loc(newlocale(LC_ALL_MASK, nm, 0), freelocale); - if (loc == nullptr) + __libcpp_unique_locale loc(nm); + if (!loc) __throw_runtime_error("moneypunct_byname" " failed to construct for " + string(nm)); @@ -5924,8 +5942,8 @@ void moneypunct_byname::init(const char* nm) { typedef moneypunct base; - __locale_unique_ptr loc(newlocale(LC_ALL_MASK, nm, 0), freelocale); - if (loc == nullptr) + __libcpp_unique_locale loc(nm); + if (!loc) __throw_runtime_error("moneypunct_byname" " failed to construct for " + string(nm)); lconv* lc = __libcpp_localeconv_l(loc.get()); @@ -5989,8 +6007,8 @@ void moneypunct_byname::init(const char* nm) { typedef moneypunct base; - __locale_unique_ptr loc(newlocale(LC_ALL_MASK, nm, 0), freelocale); - if (loc == nullptr) + __libcpp_unique_locale loc(nm); + if (!loc) __throw_runtime_error("moneypunct_byname" " failed to construct for " + string(nm)); diff --git a/src/string.cpp b/src/string.cpp index cd644330b3a7..d7ebdd3e5c9a 100644 --- a/src/string.cpp +++ b/src/string.cpp @@ -13,9 +13,6 @@ #include "cerrno" #include "limits" #include "stdexcept" -#ifdef _LIBCPP_MSVCRT -#include "support/win32/support.h" -#endif // _LIBCPP_MSVCRT #include _LIBCPP_BEGIN_NAMESPACE_STD @@ -430,7 +427,7 @@ get_swprintf() #ifndef _LIBCPP_MSVCRT return swprintf; #else - return static_cast(swprintf); + return static_cast(_snwprintf); #endif } diff --git a/src/support/runtime/exception_pointer_msvc.ipp b/src/support/runtime/exception_pointer_msvc.ipp index a8cd0e8d304d..eab5d30a9487 100644 --- a/src/support/runtime/exception_pointer_msvc.ipp +++ b/src/support/runtime/exception_pointer_msvc.ipp @@ -10,6 +10,7 @@ #include #include +#include // for _CRTIMP2_PURE _CRTIMP2_PURE void __CLRCALL_PURE_OR_CDECL __ExceptionPtrCreate(_Out_ void*); _CRTIMP2_PURE void __CLRCALL_PURE_OR_CDECL __ExceptionPtrDestroy(_Inout_ void*); diff --git a/src/support/win32/locale_win32.cpp b/src/support/win32/locale_win32.cpp index acbf79ac197c..28cb4491781f 100644 --- a/src/support/win32/locale_win32.cpp +++ b/src/support/win32/locale_win32.cpp @@ -13,14 +13,14 @@ #include #include -typedef _VSTD::remove_pointer::type __locale_struct; -typedef _VSTD::unique_ptr<__locale_struct, decltype(&uselocale)> __locale_raii; +using std::__libcpp_locale_guard; // FIXME: base currently unused. Needs manual work to construct the new locale locale_t newlocale( int mask, const char * locale, locale_t /*base*/ ) { return _create_locale( mask, locale ); } + locale_t uselocale( locale_t newloc ) { locale_t old_locale = _get_current_locale(); @@ -36,59 +36,59 @@ locale_t uselocale( locale_t newloc ) } lconv *localeconv_l( locale_t loc ) { - __locale_raii __current( uselocale(loc), uselocale ); + __libcpp_locale_guard __current(loc); return localeconv(); } size_t mbrlen_l( const char *__restrict s, size_t n, mbstate_t *__restrict ps, locale_t loc ) { - __locale_raii __current( uselocale(loc), uselocale ); + __libcpp_locale_guard __current(loc); return mbrlen( s, n, ps ); } size_t mbsrtowcs_l( wchar_t *__restrict dst, const char **__restrict src, size_t len, mbstate_t *__restrict ps, locale_t loc ) { - __locale_raii __current( uselocale(loc), uselocale ); + __libcpp_locale_guard __current(loc); return mbsrtowcs( dst, src, len, ps ); } size_t wcrtomb_l( char *__restrict s, wchar_t wc, mbstate_t *__restrict ps, locale_t loc ) { - __locale_raii __current( uselocale(loc), uselocale ); + __libcpp_locale_guard __current(loc); return wcrtomb( s, wc, ps ); } size_t mbrtowc_l( wchar_t *__restrict pwc, const char *__restrict s, size_t n, mbstate_t *__restrict ps, locale_t loc ) { - __locale_raii __current( uselocale(loc), uselocale ); + __libcpp_locale_guard __current(loc); return mbrtowc( pwc, s, n, ps ); } size_t mbsnrtowcs_l( wchar_t *__restrict dst, const char **__restrict src, size_t nms, size_t len, mbstate_t *__restrict ps, locale_t loc ) { - __locale_raii __current( uselocale(loc), uselocale ); + __libcpp_locale_guard __current(loc); return mbsnrtowcs( dst, src, nms, len, ps ); } size_t wcsnrtombs_l( char *__restrict dst, const wchar_t **__restrict src, size_t nwc, size_t len, mbstate_t *__restrict ps, locale_t loc ) { - __locale_raii __current( uselocale(loc), uselocale ); + __libcpp_locale_guard __current(loc); return wcsnrtombs( dst, src, nwc, len, ps ); } wint_t btowc_l( int c, locale_t loc ) { - __locale_raii __current( uselocale(loc), uselocale ); + __libcpp_locale_guard __current(loc); return btowc( c ); } int wctob_l( wint_t c, locale_t loc ) { - __locale_raii __current( uselocale(loc), uselocale ); + __libcpp_locale_guard __current(loc); return wctob( c ); } int snprintf_l(char *ret, size_t n, locale_t loc, const char *format, ...) { - __locale_raii __current( uselocale(loc), uselocale ); + __libcpp_locale_guard __current(loc); va_list ap; va_start( ap, format ); int result = vsnprintf( ret, n, format, ap ); @@ -106,6 +106,6 @@ int asprintf_l( char **ret, locale_t loc, const char *format, ... ) } int vasprintf_l( char **ret, locale_t loc, const char *format, va_list ap ) { - __locale_raii __current( uselocale(loc), uselocale ); + __libcpp_locale_guard __current(loc); return vasprintf( ret, format, ap ); } diff --git a/src/thread.cpp b/src/thread.cpp index 3f283c385b76..412a8fa0a620 100644 --- a/src/thread.cpp +++ b/src/thread.cpp @@ -99,7 +99,7 @@ thread::hardware_concurrency() _NOEXCEPT #else // defined(CTL_HW) && defined(HW_NCPU) // TODO: grovel through /proc or check cpuid on x86 and similar // instructions on other architectures. -# if defined(_LIBCPP_MSVC) +# if defined(_LIBCPP_WARNING) _LIBCPP_WARNING("hardware_concurrency not yet implemented") # else # warning hardware_concurrency not yet implemented diff --git a/test/std/input.output/file.streams/c.files/no.global.filesystem.namespace/fopen.fail.cpp b/test/libcxx/input.output/file.streams/c.files/no.global.filesystem.namespace/fopen.fail.cpp similarity index 89% rename from test/std/input.output/file.streams/c.files/no.global.filesystem.namespace/fopen.fail.cpp rename to test/libcxx/input.output/file.streams/c.files/no.global.filesystem.namespace/fopen.fail.cpp index 4d83296f086a..31a37229bf0c 100644 --- a/test/std/input.output/file.streams/c.files/no.global.filesystem.namespace/fopen.fail.cpp +++ b/test/libcxx/input.output/file.streams/c.files/no.global.filesystem.namespace/fopen.fail.cpp @@ -7,6 +7,8 @@ // //===----------------------------------------------------------------------===// +// REQUIRES: libcpp-has-no-global-filesystem-namespace + #include int main() { diff --git a/test/std/input.output/file.streams/c.files/no.global.filesystem.namespace/rename.fail.cpp b/test/libcxx/input.output/file.streams/c.files/no.global.filesystem.namespace/rename.fail.cpp similarity index 89% rename from test/std/input.output/file.streams/c.files/no.global.filesystem.namespace/rename.fail.cpp rename to test/libcxx/input.output/file.streams/c.files/no.global.filesystem.namespace/rename.fail.cpp index deca9bf5b551..248ab4d67210 100644 --- a/test/std/input.output/file.streams/c.files/no.global.filesystem.namespace/rename.fail.cpp +++ b/test/libcxx/input.output/file.streams/c.files/no.global.filesystem.namespace/rename.fail.cpp @@ -7,6 +7,8 @@ // //===----------------------------------------------------------------------===// +// REQUIRES: libcpp-has-no-global-filesystem-namespace + #include int main() { diff --git a/test/std/thread/thread.condition/PR30202_notify_from_pthread_created_thread.pass.cpp b/test/libcxx/thread/thread.condition/PR30202_notify_from_pthread_created_thread.pass.cpp similarity index 100% rename from test/std/thread/thread.condition/PR30202_notify_from_pthread_created_thread.pass.cpp rename to test/libcxx/thread/thread.condition/PR30202_notify_from_pthread_created_thread.pass.cpp diff --git a/test/std/thread/thread.threads/thread.thread.this/sleep_for.pass.cpp b/test/libcxx/thread/thread.threads/thread.thread.this/sleep_for.pass.cpp similarity index 99% rename from test/std/thread/thread.threads/thread.thread.this/sleep_for.pass.cpp rename to test/libcxx/thread/thread.threads/thread.thread.this/sleep_for.pass.cpp index 891b084dd32e..b46c2cdec6cb 100644 --- a/test/std/thread/thread.threads/thread.thread.this/sleep_for.pass.cpp +++ b/test/libcxx/thread/thread.threads/thread.thread.this/sleep_for.pass.cpp @@ -35,7 +35,6 @@ void sig_action(int) {} -#include int main() { int ec; diff --git a/test/std/utilities/function.objects/refwrap/binary.pass.cpp b/test/libcxx/utilities/function.objects/refwrap/binary.pass.cpp similarity index 100% rename from test/std/utilities/function.objects/refwrap/binary.pass.cpp rename to test/libcxx/utilities/function.objects/refwrap/binary.pass.cpp diff --git a/test/std/utilities/function.objects/refwrap/unary.pass.cpp b/test/libcxx/utilities/function.objects/refwrap/unary.pass.cpp similarity index 100% rename from test/std/utilities/function.objects/refwrap/unary.pass.cpp rename to test/libcxx/utilities/function.objects/refwrap/unary.pass.cpp diff --git a/test/libcxx/utilities/template.bitset/includes.pass.cpp b/test/libcxx/utilities/template.bitset/includes.pass.cpp new file mode 100644 index 000000000000..2e3c2812e441 --- /dev/null +++ b/test/libcxx/utilities/template.bitset/includes.pass.cpp @@ -0,0 +1,32 @@ +//===----------------------------------------------------------------------===// +// +// The LLVM Compiler Infrastructure +// +// This file is dual licensed under the MIT and the University of Illinois Open +// Source Licenses. See LICENSE.TXT for details. +// +//===----------------------------------------------------------------------===// + +// test that includes , , and + +#include + +#ifndef _LIBCPP_CSTDDEF +#error has not been included +#endif + +#ifndef _LIBCPP_STRING +#error has not been included +#endif + +#ifndef _LIBCPP_STDEXCEPT +#error has not been included +#endif + +#ifndef _LIBCPP_IOSFWD +#error has not been included +#endif + +int main() +{ +} diff --git a/test/std/utilities/tuple/tuple.tuple/empty_member.pass.cpp b/test/libcxx/utilities/tuple/tuple.tuple/empty_member.pass.cpp similarity index 100% rename from test/std/utilities/tuple/tuple.tuple/empty_member.pass.cpp rename to test/libcxx/utilities/tuple/tuple.tuple/empty_member.pass.cpp diff --git a/test/std/experimental/utilities/meta/meta.type.synop/meta.unary.prop.pass.cpp b/test/std/experimental/utilities/meta/meta.type.synop/meta.unary.prop.pass.cpp index e267c6833ab9..bfd385a1b8e3 100644 --- a/test/std/experimental/utilities/meta/meta.type.synop/meta.unary.prop.pass.cpp +++ b/test/std/experimental/utilities/meta/meta.type.synop/meta.unary.prop.pass.cpp @@ -8,6 +8,12 @@ //===----------------------------------------------------------------------===// // UNSUPPORTED: c++98, c++03, c++11 + +// GCC returns true for __is_trivially_constructible(void, int) +// See gcc.gnu.org/PR80682 +// NOTE: This has been fixed in trunk and will be backported soon. +// XFAIL: gcc-7, gcc-6, gcc-5, gcc-4 + // #include diff --git a/test/std/input.output/file.streams/c.files/no.global.filesystem.namespace/lit.local.cfg b/test/std/input.output/file.streams/c.files/no.global.filesystem.namespace/lit.local.cfg deleted file mode 100644 index 4ea670935591..000000000000 --- a/test/std/input.output/file.streams/c.files/no.global.filesystem.namespace/lit.local.cfg +++ /dev/null @@ -1,2 +0,0 @@ -if 'libcpp-has-no-global-filesystem-namespace' not in config.available_features: - config.unsupported = True diff --git a/test/std/language.support/support.dynamic/new.delete/new.delete.array/delete_align_val_t_replace.pass.cpp b/test/std/language.support/support.dynamic/new.delete/new.delete.array/delete_align_val_t_replace.pass.cpp index 41b14a6e5bf3..36f815a819be 100644 --- a/test/std/language.support/support.dynamic/new.delete/new.delete.array/delete_align_val_t_replace.pass.cpp +++ b/test/std/language.support/support.dynamic/new.delete/new.delete.array/delete_align_val_t_replace.pass.cpp @@ -15,7 +15,7 @@ // XFAIL: clang-3, apple-clang-7, apple-clang-8 // None of the current GCC compilers support this. -// XFAIL: gcc +// XFAIL: gcc-5, gcc-6 // XFAIL: with_system_cxx_lib=macosx10.12 // XFAIL: with_system_cxx_lib=macosx10.11 diff --git a/test/std/language.support/support.dynamic/new.delete/new.delete.array/new_align_val_t.pass.cpp b/test/std/language.support/support.dynamic/new.delete/new.delete.array/new_align_val_t.pass.cpp index 4c9da471294d..69f5ac8c9774 100644 --- a/test/std/language.support/support.dynamic/new.delete/new.delete.array/new_align_val_t.pass.cpp +++ b/test/std/language.support/support.dynamic/new.delete/new.delete.array/new_align_val_t.pass.cpp @@ -13,7 +13,7 @@ // UNSUPPORTED: sanitizer-new-delete // FIXME change this to XFAIL. -// UNSUPPORTED: no-aligned-allocation +// UNSUPPORTED: no-aligned-allocation && !gcc // XFAIL: with_system_cxx_lib=macosx10.12 // XFAIL: with_system_cxx_lib=macosx10.11 diff --git a/test/std/language.support/support.dynamic/new.delete/new.delete.array/new_align_val_t_nothrow.pass.cpp b/test/std/language.support/support.dynamic/new.delete/new.delete.array/new_align_val_t_nothrow.pass.cpp index 8e9eff9e206e..36453283caf6 100644 --- a/test/std/language.support/support.dynamic/new.delete/new.delete.array/new_align_val_t_nothrow.pass.cpp +++ b/test/std/language.support/support.dynamic/new.delete/new.delete.array/new_align_val_t_nothrow.pass.cpp @@ -13,7 +13,7 @@ // UNSUPPORTED: sanitizer-new-delete // FIXME turn this into an XFAIL -// UNSUPPORTED: no-aligned-allocation +// UNSUPPORTED: no-aligned-allocation && !gcc // XFAIL: with_system_cxx_lib=macosx10.12 // XFAIL: with_system_cxx_lib=macosx10.11 diff --git a/test/std/language.support/support.dynamic/new.delete/new.delete.array/new_align_val_t_nothrow_replace.pass.cpp b/test/std/language.support/support.dynamic/new.delete/new.delete.array/new_align_val_t_nothrow_replace.pass.cpp index 46f411d15684..228d176a27e0 100644 --- a/test/std/language.support/support.dynamic/new.delete/new.delete.array/new_align_val_t_nothrow_replace.pass.cpp +++ b/test/std/language.support/support.dynamic/new.delete/new.delete.array/new_align_val_t_nothrow_replace.pass.cpp @@ -17,7 +17,7 @@ // XFAIL: with_system_cxx_lib=macosx10.7 // XFAIL: with_system_cxx_lib=macosx10.8 -// XFAIL: no-aligned-allocation +// XFAIL: no-aligned-allocation && !gcc // On Windows libc++ doesn't provide its own definitions for new/delete // but instead depends on the ones in VCRuntime. However VCRuntime does not diff --git a/test/std/language.support/support.dynamic/new.delete/new.delete.array/new_align_val_t_replace.pass.cpp b/test/std/language.support/support.dynamic/new.delete/new.delete.array/new_align_val_t_replace.pass.cpp index 131deb340d2f..15bebcbcbcb8 100644 --- a/test/std/language.support/support.dynamic/new.delete/new.delete.array/new_align_val_t_replace.pass.cpp +++ b/test/std/language.support/support.dynamic/new.delete/new.delete.array/new_align_val_t_replace.pass.cpp @@ -10,7 +10,8 @@ // UNSUPPORTED: c++98, c++03, c++11, c++14 // UNSUPPORTED: sanitizer-new-delete -// XFAIL: no-aligned-allocation +// NOTE: GCC doesn't provide the -faligned-allocation flag to test for +// XFAIL: no-aligned-allocation && !gcc // test operator new replacement diff --git a/test/std/language.support/support.dynamic/new.delete/new.delete.single/delete_align_val_t_replace.pass.cpp b/test/std/language.support/support.dynamic/new.delete/new.delete.single/delete_align_val_t_replace.pass.cpp index 514e5e1044a0..b0e932c96cb0 100644 --- a/test/std/language.support/support.dynamic/new.delete/new.delete.single/delete_align_val_t_replace.pass.cpp +++ b/test/std/language.support/support.dynamic/new.delete/new.delete.single/delete_align_val_t_replace.pass.cpp @@ -14,7 +14,7 @@ // XFAIL: clang-3, apple-clang-7, apple-clang-8 // None of the current GCC compilers support this. -// XFAIL: gcc +// XFAIL: gcc-5, gcc-6 // XFAIL: with_system_cxx_lib=macosx10.12 // XFAIL: with_system_cxx_lib=macosx10.11 diff --git a/test/std/language.support/support.dynamic/new.delete/new.delete.single/new_align_val_t.pass.cpp b/test/std/language.support/support.dynamic/new.delete/new.delete.single/new_align_val_t.pass.cpp index 2dcb2dc2c578..22ea5e2fff89 100644 --- a/test/std/language.support/support.dynamic/new.delete/new.delete.single/new_align_val_t.pass.cpp +++ b/test/std/language.support/support.dynamic/new.delete/new.delete.single/new_align_val_t.pass.cpp @@ -20,7 +20,7 @@ // UNSUPPORTED: sanitizer-new-delete // FIXME turn this into an XFAIL -// UNSUPPORTED: no-aligned-allocation +// UNSUPPORTED: no-aligned-allocation && !gcc // On Windows libc++ doesn't provide its own definitions for new/delete // but instead depends on the ones in VCRuntime. However VCRuntime does not diff --git a/test/std/language.support/support.dynamic/new.delete/new.delete.single/new_align_val_t_nothrow.pass.cpp b/test/std/language.support/support.dynamic/new.delete/new.delete.single/new_align_val_t_nothrow.pass.cpp index e77b41336174..2c5c0f80b839 100644 --- a/test/std/language.support/support.dynamic/new.delete/new.delete.single/new_align_val_t_nothrow.pass.cpp +++ b/test/std/language.support/support.dynamic/new.delete/new.delete.single/new_align_val_t_nothrow.pass.cpp @@ -20,7 +20,7 @@ // UNSUPPORTED: sanitizer-new-delete // FIXME turn this into an XFAIL -// UNSUPPORTED: no-aligned-allocation +// UNSUPPORTED: no-aligned-allocation && !gcc // On Windows libc++ doesn't provide its own definitions for new/delete // but instead depends on the ones in VCRuntime. However VCRuntime does not diff --git a/test/std/language.support/support.dynamic/new.delete/new.delete.single/new_align_val_t_nothrow_replace.pass.cpp b/test/std/language.support/support.dynamic/new.delete/new.delete.single/new_align_val_t_nothrow_replace.pass.cpp index 862ec1ba66a4..b50f49dc6ca7 100644 --- a/test/std/language.support/support.dynamic/new.delete/new.delete.single/new_align_val_t_nothrow_replace.pass.cpp +++ b/test/std/language.support/support.dynamic/new.delete/new.delete.single/new_align_val_t_nothrow_replace.pass.cpp @@ -17,7 +17,8 @@ // XFAIL: with_system_cxx_lib=macosx10.7 // XFAIL: with_system_cxx_lib=macosx10.8 -// XFAIL: no-aligned-allocation +// NOTE: gcc doesn't provide -faligned-allocation flag to test for +// XFAIL: no-aligned-allocation && !gcc // On Windows libc++ doesn't provide its own definitions for new/delete // but instead depends on the ones in VCRuntime. However VCRuntime does not diff --git a/test/std/language.support/support.dynamic/new.delete/new.delete.single/new_align_val_t_replace.pass.cpp b/test/std/language.support/support.dynamic/new.delete/new.delete.single/new_align_val_t_replace.pass.cpp index df3e4c1560a9..66bb613adfd0 100644 --- a/test/std/language.support/support.dynamic/new.delete/new.delete.single/new_align_val_t_replace.pass.cpp +++ b/test/std/language.support/support.dynamic/new.delete/new.delete.single/new_align_val_t_replace.pass.cpp @@ -10,7 +10,8 @@ // UNSUPPORTED: c++98, c++03, c++11, c++14 // UNSUPPORTED: sanitizer-new-delete -// XFAIL: no-aligned-allocation +// NOTE: GCC doesn't provide a -faligned-allocation flag +// XFAIL: no-aligned-allocation && !gcc // test operator new replacement diff --git a/test/std/language.support/support.types/byteops/and.assign.pass.cpp b/test/std/language.support/support.types/byteops/and.assign.pass.cpp index dec241eb0c41..ec1c3805a6cb 100644 --- a/test/std/language.support/support.types/byteops/and.assign.pass.cpp +++ b/test/std/language.support/support.types/byteops/and.assign.pass.cpp @@ -11,9 +11,6 @@ #include // UNSUPPORTED: c++98, c++03, c++11, c++14 -// The following compilers don't like "std::byte b1{1}" -// UNSUPPORTED: clang-3.5, clang-3.6, clang-3.7, clang-3.8 -// UNSUPPORTED: apple-clang-6, apple-clang-7, apple-clang-8.0 // constexpr byte& operator &=(byte l, byte r) noexcept; @@ -26,9 +23,9 @@ constexpr std::byte test(std::byte b1, std::byte b2) { int main () { std::byte b; // not constexpr, just used in noexcept check - constexpr std::byte b1{1}; - constexpr std::byte b8{8}; - constexpr std::byte b9{9}; + constexpr std::byte b1{static_cast(1)}; + constexpr std::byte b8{static_cast(8)}; + constexpr std::byte b9{static_cast(9)}; static_assert(noexcept(b &= b), "" ); diff --git a/test/std/language.support/support.types/byteops/and.pass.cpp b/test/std/language.support/support.types/byteops/and.pass.cpp index 22da6e3e0d4c..6c5419cfc2a2 100644 --- a/test/std/language.support/support.types/byteops/and.pass.cpp +++ b/test/std/language.support/support.types/byteops/and.pass.cpp @@ -11,16 +11,13 @@ #include // UNSUPPORTED: c++98, c++03, c++11, c++14 -// The following compilers don't like "std::byte b1{1}" -// UNSUPPORTED: clang-3.5, clang-3.6, clang-3.7, clang-3.8 -// UNSUPPORTED: apple-clang-6, apple-clang-7, apple-clang-8.0 // constexpr byte operator&(byte l, byte r) noexcept; int main () { - constexpr std::byte b1{1}; - constexpr std::byte b8{8}; - constexpr std::byte b9{9}; + constexpr std::byte b1{static_cast(1)}; + constexpr std::byte b8{static_cast(8)}; + constexpr std::byte b9{static_cast(9)}; static_assert(noexcept(b1 & b8), "" ); diff --git a/test/std/language.support/support.types/byteops/enum_direct_init.pass.cpp b/test/std/language.support/support.types/byteops/enum_direct_init.pass.cpp new file mode 100644 index 000000000000..157626137330 --- /dev/null +++ b/test/std/language.support/support.types/byteops/enum_direct_init.pass.cpp @@ -0,0 +1,21 @@ +//===----------------------------------------------------------------------===// +// +// The LLVM Compiler Infrastructure +// +// This file is dual licensed under the MIT and the University of Illinois Open +// Source Licenses. See LICENSE.TXT for details. +// +//===----------------------------------------------------------------------===// + +#include +#include + +// UNSUPPORTED: c++98, c++03, c++11, c++14 +// The following compilers don't like "std::byte b1{1}" +// XFAIL: clang-3.5, clang-3.6, clang-3.7, clang-3.8 +// XFAIL: apple-clang-6, apple-clang-7, apple-clang-8.0 + +int main () { + constexpr std::byte b{42}; + static_assert(std::to_integer(b) == 42, ""); +} diff --git a/test/std/language.support/support.types/byteops/lshift.assign.pass.cpp b/test/std/language.support/support.types/byteops/lshift.assign.pass.cpp index dad692e14a6b..abfeebf2b024 100644 --- a/test/std/language.support/support.types/byteops/lshift.assign.pass.cpp +++ b/test/std/language.support/support.types/byteops/lshift.assign.pass.cpp @@ -11,9 +11,6 @@ #include // UNSUPPORTED: c++98, c++03, c++11, c++14 -// The following compilers don't like "std::byte b1{1}" -// UNSUPPORTED: clang-3.5, clang-3.6, clang-3.7, clang-3.8 -// UNSUPPORTED: apple-clang-6, apple-clang-7, apple-clang-8.0 // template // constexpr byte& operator<<=(byte& b, IntegerType shift) noexcept; @@ -28,8 +25,8 @@ constexpr std::byte test(std::byte b) { int main () { std::byte b; // not constexpr, just used in noexcept check - constexpr std::byte b2{2}; - constexpr std::byte b3{3}; + constexpr std::byte b2{static_cast(2)}; + constexpr std::byte b3{static_cast(3)}; static_assert(noexcept(b <<= 2), "" ); diff --git a/test/std/language.support/support.types/byteops/lshift.fail.cpp b/test/std/language.support/support.types/byteops/lshift.fail.cpp index 6b1a68f83d25..707258c6df30 100644 --- a/test/std/language.support/support.types/byteops/lshift.fail.cpp +++ b/test/std/language.support/support.types/byteops/lshift.fail.cpp @@ -11,9 +11,6 @@ #include // UNSUPPORTED: c++98, c++03, c++11, c++14 -// The following compilers don't like "std::byte b1{1}" -// UNSUPPORTED: clang-3.5, clang-3.6, clang-3.7, clang-3.8 -// UNSUPPORTED: apple-clang-6, apple-clang-7, apple-clang-8.0 // template // constexpr byte operator <<(byte b, IntegerType shift) noexcept; @@ -21,6 +18,6 @@ // is_integral_v is true. int main () { - constexpr std::byte b1{1}; + constexpr std::byte b1{static_cast(1)}; constexpr std::byte b2 = b1 << 2.0f; } diff --git a/test/std/language.support/support.types/byteops/lshift.pass.cpp b/test/std/language.support/support.types/byteops/lshift.pass.cpp index 39d659023ad8..15ad796cbe9e 100644 --- a/test/std/language.support/support.types/byteops/lshift.pass.cpp +++ b/test/std/language.support/support.types/byteops/lshift.pass.cpp @@ -11,9 +11,6 @@ #include // UNSUPPORTED: c++98, c++03, c++11, c++14 -// The following compilers don't like "std::byte b1{1}" -// UNSUPPORTED: clang-3.5, clang-3.6, clang-3.7, clang-3.8 -// UNSUPPORTED: apple-clang-6, apple-clang-7, apple-clang-8.0 // template // constexpr byte operator <<(byte b, IntegerType shift) noexcept; @@ -21,8 +18,8 @@ // is_integral_v is true. int main () { - constexpr std::byte b1{1}; - constexpr std::byte b3{3}; + constexpr std::byte b1{static_cast(1)}; + constexpr std::byte b3{static_cast(3)}; static_assert(noexcept(b3 << 2), "" ); diff --git a/test/std/language.support/support.types/byteops/not.pass.cpp b/test/std/language.support/support.types/byteops/not.pass.cpp index 734780f194a0..53f1d91faf23 100644 --- a/test/std/language.support/support.types/byteops/not.pass.cpp +++ b/test/std/language.support/support.types/byteops/not.pass.cpp @@ -11,16 +11,13 @@ #include // UNSUPPORTED: c++98, c++03, c++11, c++14 -// The following compilers don't like "std::byte b1{1}" -// UNSUPPORTED: clang-3.5, clang-3.6, clang-3.7, clang-3.8 -// UNSUPPORTED: apple-clang-6, apple-clang-7, apple-clang-8.0 // constexpr byte operator~(byte b) noexcept; int main () { - constexpr std::byte b1{1}; - constexpr std::byte b2{2}; - constexpr std::byte b8{8}; + constexpr std::byte b1{static_cast(1)}; + constexpr std::byte b2{static_cast(2)}; + constexpr std::byte b8{static_cast(8)}; static_assert(noexcept(~b1), "" ); diff --git a/test/std/language.support/support.types/byteops/or.assign.pass.cpp b/test/std/language.support/support.types/byteops/or.assign.pass.cpp index 75d6ab4d0a9d..bb4a84ec6c76 100644 --- a/test/std/language.support/support.types/byteops/or.assign.pass.cpp +++ b/test/std/language.support/support.types/byteops/or.assign.pass.cpp @@ -11,9 +11,6 @@ #include // UNSUPPORTED: c++98, c++03, c++11, c++14 -// The following compilers don't like "std::byte b1{1}" -// UNSUPPORTED: clang-3.5, clang-3.6, clang-3.7, clang-3.8 -// UNSUPPORTED: apple-clang-6, apple-clang-7, apple-clang-8.0 // constexpr byte& operator |=(byte l, byte r) noexcept; @@ -26,9 +23,9 @@ constexpr std::byte test(std::byte b1, std::byte b2) { int main () { std::byte b; // not constexpr, just used in noexcept check - constexpr std::byte b1{1}; - constexpr std::byte b2{2}; - constexpr std::byte b8{8}; + constexpr std::byte b1{static_cast(1)}; + constexpr std::byte b2{static_cast(2)}; + constexpr std::byte b8{static_cast(8)}; static_assert(noexcept(b |= b), "" ); diff --git a/test/std/language.support/support.types/byteops/or.pass.cpp b/test/std/language.support/support.types/byteops/or.pass.cpp index 02c547f1dbb6..e2b734982c09 100644 --- a/test/std/language.support/support.types/byteops/or.pass.cpp +++ b/test/std/language.support/support.types/byteops/or.pass.cpp @@ -11,16 +11,13 @@ #include // UNSUPPORTED: c++98, c++03, c++11, c++14 -// The following compilers don't like "std::byte b1{1}" -// UNSUPPORTED: clang-3.5, clang-3.6, clang-3.7, clang-3.8 -// UNSUPPORTED: apple-clang-6, apple-clang-7, apple-clang-8.0 // constexpr byte operator|(byte l, byte r) noexcept; int main () { - constexpr std::byte b1{1}; - constexpr std::byte b2{2}; - constexpr std::byte b8{8}; + constexpr std::byte b1{static_cast(1)}; + constexpr std::byte b2{static_cast(2)}; + constexpr std::byte b8{static_cast(8)}; static_assert(noexcept(b1 | b2), "" ); diff --git a/test/std/language.support/support.types/byteops/rshift.assign.pass.cpp b/test/std/language.support/support.types/byteops/rshift.assign.pass.cpp index b7e9a24f20a1..82c611cfdfcb 100644 --- a/test/std/language.support/support.types/byteops/rshift.assign.pass.cpp +++ b/test/std/language.support/support.types/byteops/rshift.assign.pass.cpp @@ -11,9 +11,6 @@ #include // UNSUPPORTED: c++98, c++03, c++11, c++14 -// The following compilers don't like "std::byte b1{1}" -// UNSUPPORTED: clang-3.5, clang-3.6, clang-3.7, clang-3.8 -// UNSUPPORTED: apple-clang-6, apple-clang-7, apple-clang-8.0 // template // constexpr byte& operator>>=(byte& b, IntegerType shift) noexcept; @@ -28,8 +25,8 @@ constexpr std::byte test(std::byte b) { int main () { std::byte b; // not constexpr, just used in noexcept check - constexpr std::byte b16{16}; - constexpr std::byte b192{192}; + constexpr std::byte b16{static_cast(16)}; + constexpr std::byte b192{static_cast(192)}; static_assert(noexcept(b >>= 2), "" ); diff --git a/test/std/language.support/support.types/byteops/rshift.fail.cpp b/test/std/language.support/support.types/byteops/rshift.fail.cpp index a0309539a7a6..8e142dd90a5c 100644 --- a/test/std/language.support/support.types/byteops/rshift.fail.cpp +++ b/test/std/language.support/support.types/byteops/rshift.fail.cpp @@ -11,9 +11,6 @@ #include // UNSUPPORTED: c++98, c++03, c++11, c++14 -// The following compilers don't like "std::byte b1{1}" -// UNSUPPORTED: clang-3.5, clang-3.6, clang-3.7, clang-3.8 -// UNSUPPORTED: apple-clang-6, apple-clang-7, apple-clang-8.0 // template // constexpr byte operator >>(byte b, IntegerType shift) noexcept; @@ -21,6 +18,6 @@ // is_integral_v is true. int main () { - constexpr std::byte b1{1}; + constexpr std::byte b1{static_cast(1)}; constexpr std::byte b2 = b1 >> 2.0f; } diff --git a/test/std/language.support/support.types/byteops/rshift.pass.cpp b/test/std/language.support/support.types/byteops/rshift.pass.cpp index 876732545ac8..22d399130bbc 100644 --- a/test/std/language.support/support.types/byteops/rshift.pass.cpp +++ b/test/std/language.support/support.types/byteops/rshift.pass.cpp @@ -11,9 +11,6 @@ #include // UNSUPPORTED: c++98, c++03, c++11, c++14 -// The following compilers don't like "std::byte b1{1}" -// UNSUPPORTED: clang-3.5, clang-3.6, clang-3.7, clang-3.8 -// UNSUPPORTED: apple-clang-6, apple-clang-7, apple-clang-8.0 // template // constexpr byte operator <<(byte b, IntegerType shift) noexcept; @@ -27,8 +24,8 @@ constexpr std::byte test(std::byte b) { int main () { - constexpr std::byte b100{100}; - constexpr std::byte b115{115}; + constexpr std::byte b100{static_cast(100)}; + constexpr std::byte b115{static_cast(115)}; static_assert(noexcept(b100 << 2), "" ); diff --git a/test/std/language.support/support.types/byteops/to_integer.fail.cpp b/test/std/language.support/support.types/byteops/to_integer.fail.cpp index 426ceb7a67de..fb4928956f55 100644 --- a/test/std/language.support/support.types/byteops/to_integer.fail.cpp +++ b/test/std/language.support/support.types/byteops/to_integer.fail.cpp @@ -11,9 +11,6 @@ #include // UNSUPPORTED: c++98, c++03, c++11, c++14 -// The following compilers don't like "std::byte b1{1}" -// UNSUPPORTED: clang-3.5, clang-3.6, clang-3.7, clang-3.8 -// UNSUPPORTED: apple-clang-6, apple-clang-7, apple-clang-8.0 // template // constexpr IntegerType to_integer(byte b) noexcept; @@ -21,6 +18,6 @@ // is_integral_v is true. int main () { - constexpr std::byte b1{1}; + constexpr std::byte b1{static_cast(1)}; auto f = std::to_integer(b1); } diff --git a/test/std/language.support/support.types/byteops/to_integer.pass.cpp b/test/std/language.support/support.types/byteops/to_integer.pass.cpp index 21dff0196a43..4f34363447f6 100644 --- a/test/std/language.support/support.types/byteops/to_integer.pass.cpp +++ b/test/std/language.support/support.types/byteops/to_integer.pass.cpp @@ -11,9 +11,6 @@ #include // UNSUPPORTED: c++98, c++03, c++11, c++14 -// The following compilers don't like "std::byte b1{1}" -// UNSUPPORTED: clang-3.5, clang-3.6, clang-3.7, clang-3.8 -// UNSUPPORTED: apple-clang-6, apple-clang-7, apple-clang-8.0 // template // constexpr IntegerType to_integer(byte b) noexcept; @@ -21,8 +18,8 @@ // is_integral_v is true. int main () { - constexpr std::byte b1{1}; - constexpr std::byte b3{3}; + constexpr std::byte b1{static_cast(1)}; + constexpr std::byte b3{static_cast(3)}; static_assert(noexcept(std::to_integer(b1)), "" ); static_assert(std::is_same(b1))>::value, "" ); diff --git a/test/std/language.support/support.types/byteops/xor.assign.pass.cpp b/test/std/language.support/support.types/byteops/xor.assign.pass.cpp index c9b40177a17c..6a526da3d714 100644 --- a/test/std/language.support/support.types/byteops/xor.assign.pass.cpp +++ b/test/std/language.support/support.types/byteops/xor.assign.pass.cpp @@ -11,9 +11,6 @@ #include // UNSUPPORTED: c++98, c++03, c++11, c++14 -// The following compilers don't like "std::byte b1{1}" -// UNSUPPORTED: clang-3.5, clang-3.6, clang-3.7, clang-3.8 -// UNSUPPORTED: apple-clang-6, apple-clang-7, apple-clang-8.0 // constexpr byte& operator ^=(byte l, byte r) noexcept; @@ -26,9 +23,9 @@ constexpr std::byte test(std::byte b1, std::byte b2) { int main () { std::byte b; // not constexpr, just used in noexcept check - constexpr std::byte b1{1}; - constexpr std::byte b8{8}; - constexpr std::byte b9{9}; + constexpr std::byte b1{static_cast(1)}; + constexpr std::byte b8{static_cast(8)}; + constexpr std::byte b9{static_cast(9)}; static_assert(noexcept(b ^= b), "" ); diff --git a/test/std/language.support/support.types/byteops/xor.pass.cpp b/test/std/language.support/support.types/byteops/xor.pass.cpp index 3d0402b30a55..e8c3b9875785 100644 --- a/test/std/language.support/support.types/byteops/xor.pass.cpp +++ b/test/std/language.support/support.types/byteops/xor.pass.cpp @@ -11,16 +11,13 @@ #include // UNSUPPORTED: c++98, c++03, c++11, c++14 -// The following compilers don't like "std::byte b1{1}" -// UNSUPPORTED: clang-3.5, clang-3.6, clang-3.7, clang-3.8 -// UNSUPPORTED: apple-clang-6, apple-clang-7, apple-clang-8.0 // constexpr byte operator^(byte l, byte r) noexcept; int main () { - constexpr std::byte b1{1}; - constexpr std::byte b8{8}; - constexpr std::byte b9{9}; + constexpr std::byte b1{static_cast(1)}; + constexpr std::byte b8{static_cast(8)}; + constexpr std::byte b9{static_cast(9)}; static_assert(noexcept(b1 ^ b8), "" ); diff --git a/test/std/localization/locale.stdcvt/codecvt_utf16_out.pass.cpp b/test/std/localization/locale.stdcvt/codecvt_utf16_out.pass.cpp index 2eb6978b0e0a..b90c41e59d0d 100644 --- a/test/std/localization/locale.stdcvt/codecvt_utf16_out.pass.cpp +++ b/test/std/localization/locale.stdcvt/codecvt_utf16_out.pass.cpp @@ -25,307 +25,329 @@ #include #include -int main() -{ - { - typedef std::codecvt_utf16 C; - C c; - wchar_t w = 0x40003; - char n[4] = {0}; - const wchar_t* wp = nullptr; - std::mbstate_t m; - char* np = nullptr; - std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+4); - assert(n[0] == char(0xD8)); - assert(n[1] == char(0xC0)); - assert(n[2] == char(0xDC)); - assert(n[3] == char(0x03)); +template +struct TestHelper; +template +struct TestHelper { + static void test(); +}; +template +struct TestHelper { + static void test(); +}; - w = 0x1005; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(n[0] == char(0x10)); - assert(n[1] == char(0x05)); - assert(n[2] == char(0xDC)); - assert(n[3] == char(0x03)); - - w = 0x453; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(n[0] == char(0x04)); - assert(n[1] == char(0x53)); - assert(n[2] == char(0xDC)); - assert(n[3] == char(0x03)); - - w = 0x56; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(n[0] == char(0x00)); - assert(n[1] == char(0x56)); - assert(n[2] == char(0xDC)); - assert(n[3] == char(0x03)); - } - { - typedef std::codecvt_utf16 C; - C c; - wchar_t w = 0x40003; - char n[4] = {0}; - const wchar_t* wp = nullptr; - std::mbstate_t m; - char* np = nullptr; - std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::error); - assert(wp == &w); - assert(np == n); - assert(n[0] == char(0)); - assert(n[1] == char(0)); - assert(n[2] == char(0)); - assert(n[3] == char(0)); - - w = 0x1005; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::error); - assert(wp == &w); - assert(np == n); - assert(n[0] == char(0)); - assert(n[1] == char(0)); - assert(n[2] == char(0)); - assert(n[3] == char(0)); - - w = 0x453; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(n[0] == char(0x04)); - assert(n[1] == char(0x53)); - assert(n[2] == char(0)); - assert(n[3] == char(0)); - - w = 0x56; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(n[0] == char(0x00)); - assert(n[1] == char(0x56)); - assert(n[2] == char(0)); - assert(n[3] == char(0)); - } - { - typedef std::codecvt_utf16 C; - C c; - wchar_t w = 0x40003; - char n[6] = {0}; - const wchar_t* wp = nullptr; - std::mbstate_t m; - char* np = nullptr; - std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+6, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+6); - assert(n[0] == char(0xFE)); - assert(n[1] == char(0xFF)); - assert(n[2] == char(0xD8)); - assert(n[3] == char(0xC0)); - assert(n[4] == char(0xDC)); - assert(n[5] == char(0x03)); - - w = 0x1005; - r = c.out(m, &w, &w+1, wp, n, n+6, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+4); - assert(n[0] == char(0xFE)); - assert(n[1] == char(0xFF)); - assert(n[2] == char(0x10)); - assert(n[3] == char(0x05)); - assert(n[4] == char(0xDC)); - assert(n[5] == char(0x03)); - - w = 0x453; - r = c.out(m, &w, &w+1, wp, n, n+6, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+4); - assert(n[0] == char(0xFE)); - assert(n[1] == char(0xFF)); - assert(n[2] == char(0x04)); - assert(n[3] == char(0x53)); - assert(n[4] == char(0xDC)); - assert(n[5] == char(0x03)); - - w = 0x56; - r = c.out(m, &w, &w+1, wp, n, n+6, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+4); - assert(n[0] == char(0xFE)); - assert(n[1] == char(0xFF)); - assert(n[2] == char(0x00)); - assert(n[3] == char(0x56)); - assert(n[4] == char(0xDC)); - assert(n[5] == char(0x03)); - } - - { - typedef std::codecvt_utf16 C; - C c; - wchar_t w = 0x40003; - char n[4] = {0}; - const wchar_t* wp = nullptr; - std::mbstate_t m; - char* np = nullptr; - std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+4); - assert(n[1] == char(0xD8)); - assert(n[0] == char(0xC0)); - assert(n[3] == char(0xDC)); - assert(n[2] == char(0x03)); - - w = 0x1005; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(n[1] == char(0x10)); - assert(n[0] == char(0x05)); - assert(n[3] == char(0xDC)); - assert(n[2] == char(0x03)); - - w = 0x453; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(n[1] == char(0x04)); - assert(n[0] == char(0x53)); - assert(n[3] == char(0xDC)); - assert(n[2] == char(0x03)); - - w = 0x56; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(n[1] == char(0x00)); - assert(n[0] == char(0x56)); - assert(n[3] == char(0xDC)); - assert(n[2] == char(0x03)); - } - { - typedef std::codecvt_utf16 C; - C c; - wchar_t w = 0x40003; - char n[4] = {0}; - const wchar_t* wp = nullptr; - std::mbstate_t m; - char* np = nullptr; - std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::error); - assert(wp == &w); - assert(np == n); - assert(n[1] == char(0)); - assert(n[0] == char(0)); - assert(n[3] == char(0)); - assert(n[2] == char(0)); - - w = 0x1005; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::error); - assert(wp == &w); - assert(np == n); - assert(n[1] == char(0)); - assert(n[0] == char(0)); - assert(n[3] == char(0)); - assert(n[2] == char(0)); - - w = 0x453; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(n[1] == char(0x04)); - assert(n[0] == char(0x53)); - assert(n[3] == char(0)); - assert(n[2] == char(0)); - - w = 0x56; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(n[1] == char(0x00)); - assert(n[0] == char(0x56)); - assert(n[3] == char(0)); - assert(n[2] == char(0)); - } - { - typedef std::codecvt_utf16 C; - C c; - wchar_t w = 0x40003; - char n[6] = {0}; - const wchar_t* wp = nullptr; - std::mbstate_t m; - char* np = nullptr; - std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+6, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+6); - assert(n[1] == char(0xFE)); - assert(n[0] == char(0xFF)); - assert(n[3] == char(0xD8)); - assert(n[2] == char(0xC0)); - assert(n[5] == char(0xDC)); - assert(n[4] == char(0x03)); - - w = 0x1005; - r = c.out(m, &w, &w+1, wp, n, n+6, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+4); - assert(n[1] == char(0xFE)); - assert(n[0] == char(0xFF)); - assert(n[3] == char(0x10)); - assert(n[2] == char(0x05)); - assert(n[5] == char(0xDC)); - assert(n[4] == char(0x03)); - - w = 0x453; - r = c.out(m, &w, &w+1, wp, n, n+6, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+4); - assert(n[1] == char(0xFE)); - assert(n[0] == char(0xFF)); - assert(n[3] == char(0x04)); - assert(n[2] == char(0x53)); - assert(n[5] == char(0xDC)); - assert(n[4] == char(0x03)); - - w = 0x56; - r = c.out(m, &w, &w+1, wp, n, n+6, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+4); - assert(n[1] == char(0xFE)); - assert(n[0] == char(0xFF)); - assert(n[3] == char(0x00)); - assert(n[2] == char(0x56)); - assert(n[5] == char(0xDC)); - assert(n[4] == char(0x03)); - } +template +void TestHelper::test() { + // Nothing to do, the conversion in unsupported +} + +template +void TestHelper::test() { + { + typedef std::codecvt_utf16 C; + C c; + CharT w = 0x40003; + char n[4] = {0}; + const CharT* wp = nullptr; + std::mbstate_t m; + char* np = nullptr; + std::codecvt_base::result r = c.out(m, &w, &w + 1, wp, n, n + 4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w + 1); + assert(np == n + 4); + assert(n[0] == char(0xD8)); + assert(n[1] == char(0xC0)); + assert(n[2] == char(0xDC)); + assert(n[3] == char(0x03)); + + w = 0x1005; + r = c.out(m, &w, &w + 1, wp, n, n + 4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w + 1); + assert(np == n + 2); + assert(n[0] == char(0x10)); + assert(n[1] == char(0x05)); + assert(n[2] == char(0xDC)); + assert(n[3] == char(0x03)); + + w = 0x453; + r = c.out(m, &w, &w + 1, wp, n, n + 4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w + 1); + assert(np == n + 2); + assert(n[0] == char(0x04)); + assert(n[1] == char(0x53)); + assert(n[2] == char(0xDC)); + assert(n[3] == char(0x03)); + + w = 0x56; + r = c.out(m, &w, &w + 1, wp, n, n + 4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w + 1); + assert(np == n + 2); + assert(n[0] == char(0x00)); + assert(n[1] == char(0x56)); + assert(n[2] == char(0xDC)); + assert(n[3] == char(0x03)); + } + { + typedef std::codecvt_utf16 C; + C c; + CharT w = 0x40003; + char n[4] = {0}; + const CharT* wp = nullptr; + std::mbstate_t m; + char* np = nullptr; + std::codecvt_base::result r = c.out(m, &w, &w + 1, wp, n, n + 4, np); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(n[0] == char(0)); + assert(n[1] == char(0)); + assert(n[2] == char(0)); + assert(n[3] == char(0)); + + w = 0x1005; + r = c.out(m, &w, &w + 1, wp, n, n + 4, np); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(n[0] == char(0)); + assert(n[1] == char(0)); + assert(n[2] == char(0)); + assert(n[3] == char(0)); + + w = 0x453; + r = c.out(m, &w, &w + 1, wp, n, n + 4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w + 1); + assert(np == n + 2); + assert(n[0] == char(0x04)); + assert(n[1] == char(0x53)); + assert(n[2] == char(0)); + assert(n[3] == char(0)); + + w = 0x56; + r = c.out(m, &w, &w + 1, wp, n, n + 4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w + 1); + assert(np == n + 2); + assert(n[0] == char(0x00)); + assert(n[1] == char(0x56)); + assert(n[2] == char(0)); + assert(n[3] == char(0)); + } + { + typedef std::codecvt_utf16 C; + C c; + CharT w = 0x40003; + char n[6] = {0}; + const CharT* wp = nullptr; + std::mbstate_t m; + char* np = nullptr; + std::codecvt_base::result r = c.out(m, &w, &w + 1, wp, n, n + 6, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w + 1); + assert(np == n + 6); + assert(n[0] == char(0xFE)); + assert(n[1] == char(0xFF)); + assert(n[2] == char(0xD8)); + assert(n[3] == char(0xC0)); + assert(n[4] == char(0xDC)); + assert(n[5] == char(0x03)); + + w = 0x1005; + r = c.out(m, &w, &w + 1, wp, n, n + 6, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w + 1); + assert(np == n + 4); + assert(n[0] == char(0xFE)); + assert(n[1] == char(0xFF)); + assert(n[2] == char(0x10)); + assert(n[3] == char(0x05)); + assert(n[4] == char(0xDC)); + assert(n[5] == char(0x03)); + + w = 0x453; + r = c.out(m, &w, &w + 1, wp, n, n + 6, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w + 1); + assert(np == n + 4); + assert(n[0] == char(0xFE)); + assert(n[1] == char(0xFF)); + assert(n[2] == char(0x04)); + assert(n[3] == char(0x53)); + assert(n[4] == char(0xDC)); + assert(n[5] == char(0x03)); + + w = 0x56; + r = c.out(m, &w, &w + 1, wp, n, n + 6, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w + 1); + assert(np == n + 4); + assert(n[0] == char(0xFE)); + assert(n[1] == char(0xFF)); + assert(n[2] == char(0x00)); + assert(n[3] == char(0x56)); + assert(n[4] == char(0xDC)); + assert(n[5] == char(0x03)); + } + + { + typedef std::codecvt_utf16 C; + C c; + CharT w = 0x40003; + char n[4] = {0}; + const CharT* wp = nullptr; + std::mbstate_t m; + char* np = nullptr; + std::codecvt_base::result r = c.out(m, &w, &w + 1, wp, n, n + 4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w + 1); + assert(np == n + 4); + assert(n[1] == char(0xD8)); + assert(n[0] == char(0xC0)); + assert(n[3] == char(0xDC)); + assert(n[2] == char(0x03)); + + w = 0x1005; + r = c.out(m, &w, &w + 1, wp, n, n + 4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w + 1); + assert(np == n + 2); + assert(n[1] == char(0x10)); + assert(n[0] == char(0x05)); + assert(n[3] == char(0xDC)); + assert(n[2] == char(0x03)); + + w = 0x453; + r = c.out(m, &w, &w + 1, wp, n, n + 4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w + 1); + assert(np == n + 2); + assert(n[1] == char(0x04)); + assert(n[0] == char(0x53)); + assert(n[3] == char(0xDC)); + assert(n[2] == char(0x03)); + + w = 0x56; + r = c.out(m, &w, &w + 1, wp, n, n + 4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w + 1); + assert(np == n + 2); + assert(n[1] == char(0x00)); + assert(n[0] == char(0x56)); + assert(n[3] == char(0xDC)); + assert(n[2] == char(0x03)); + } + { + typedef std::codecvt_utf16 C; + C c; + CharT w = 0x40003; + char n[4] = {0}; + const CharT* wp = nullptr; + std::mbstate_t m; + char* np = nullptr; + std::codecvt_base::result r = c.out(m, &w, &w + 1, wp, n, n + 4, np); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(n[1] == char(0)); + assert(n[0] == char(0)); + assert(n[3] == char(0)); + assert(n[2] == char(0)); + + w = 0x1005; + r = c.out(m, &w, &w + 1, wp, n, n + 4, np); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(n[1] == char(0)); + assert(n[0] == char(0)); + assert(n[3] == char(0)); + assert(n[2] == char(0)); + + w = 0x453; + r = c.out(m, &w, &w + 1, wp, n, n + 4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w + 1); + assert(np == n + 2); + assert(n[1] == char(0x04)); + assert(n[0] == char(0x53)); + assert(n[3] == char(0)); + assert(n[2] == char(0)); + + w = 0x56; + r = c.out(m, &w, &w + 1, wp, n, n + 4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w + 1); + assert(np == n + 2); + assert(n[1] == char(0x00)); + assert(n[0] == char(0x56)); + assert(n[3] == char(0)); + assert(n[2] == char(0)); + } + { + typedef std::codecvt_utf16 + C; + C c; + CharT w = 0x40003; + char n[6] = {0}; + const CharT* wp = nullptr; + std::mbstate_t m; + char* np = nullptr; + std::codecvt_base::result r = c.out(m, &w, &w + 1, wp, n, n + 6, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w + 1); + assert(np == n + 6); + assert(n[1] == char(0xFE)); + assert(n[0] == char(0xFF)); + assert(n[3] == char(0xD8)); + assert(n[2] == char(0xC0)); + assert(n[5] == char(0xDC)); + assert(n[4] == char(0x03)); + + w = 0x1005; + r = c.out(m, &w, &w + 1, wp, n, n + 6, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w + 1); + assert(np == n + 4); + assert(n[1] == char(0xFE)); + assert(n[0] == char(0xFF)); + assert(n[3] == char(0x10)); + assert(n[2] == char(0x05)); + assert(n[5] == char(0xDC)); + assert(n[4] == char(0x03)); + + w = 0x453; + r = c.out(m, &w, &w + 1, wp, n, n + 6, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w + 1); + assert(np == n + 4); + assert(n[1] == char(0xFE)); + assert(n[0] == char(0xFF)); + assert(n[3] == char(0x04)); + assert(n[2] == char(0x53)); + assert(n[5] == char(0xDC)); + assert(n[4] == char(0x03)); + + w = 0x56; + r = c.out(m, &w, &w + 1, wp, n, n + 6, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w + 1); + assert(np == n + 4); + assert(n[1] == char(0xFE)); + assert(n[0] == char(0xFF)); + assert(n[3] == char(0x00)); + assert(n[2] == char(0x56)); + assert(n[5] == char(0xDC)); + assert(n[4] == char(0x03)); + } +} + +int main() { + TestHelper::test(); + TestHelper::test(); } diff --git a/test/std/localization/locale.stdcvt/codecvt_utf8_out.pass.cpp b/test/std/localization/locale.stdcvt/codecvt_utf8_out.pass.cpp index 02cf7cf958e7..886fc4416bbb 100644 --- a/test/std/localization/locale.stdcvt/codecvt_utf8_out.pass.cpp +++ b/test/std/localization/locale.stdcvt/codecvt_utf8_out.pass.cpp @@ -25,432 +25,302 @@ #include #include -int main() -{ - { - typedef std::codecvt_utf8 C; - C c; - wchar_t w = 0x40003; - char n[4] = {0}; - const wchar_t* wp = nullptr; - std::mbstate_t m; - char* np = nullptr; - std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+4); - assert(n[0] == char(0xF1)); - assert(n[1] == char(0x80)); - assert(n[2] == char(0x80)); - assert(n[3] == char(0x83)); +template +struct TestHelper; - w = 0x1005; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+3); - assert(n[0] == char(0xE1)); - assert(n[1] == char(0x80)); - assert(n[2] == char(0x85)); - assert(n[3] == char(0x83)); +template +struct TestHelper { + static void test(); +}; - w = 0x453; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(n[0] == char(0xD1)); - assert(n[1] == char(0x93)); - assert(n[2] == char(0x85)); - assert(n[3] == char(0x83)); +template +struct TestHelper { + static void test(); +}; - w = 0x56; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+1); - assert(n[0] == char(0x56)); - assert(n[1] == char(0x93)); - assert(n[2] == char(0x85)); - assert(n[3] == char(0x83)); - } - { - typedef std::codecvt_utf8 C; - C c; - wchar_t w = 0x40003; - char n[4] = {0}; - const wchar_t* wp = nullptr; - std::mbstate_t m; - char* np = nullptr; - std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::error); - assert(wp == &w); - assert(np == n); - assert(n[0] == char(0)); - assert(n[1] == char(0)); - assert(n[2] == char(0)); - assert(n[3] == char(0)); +template +void TestHelper::test() { + { + typedef std::codecvt_utf8 C; + C c; + CharT w = 0x1005; + char n[4] = {0}; + const CharT* wp = nullptr; + std::mbstate_t m; + char* np = nullptr; + std::codecvt_base::result r = c.out(m, &w, &w + 1, wp, n, n + 4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w + 1); + assert(np == n + 3); + assert(n[0] == char(0xE1)); + assert(n[1] == char(0x80)); + assert(n[2] == char(0x85)); + assert(n[3] == char(0)); - w = 0x1005; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::error); - assert(wp == &w); - assert(np == n); - assert(n[0] == char(0)); - assert(n[1] == char(0)); - assert(n[2] == char(0)); - assert(n[3] == char(0)); + w = 0x453; + r = c.out(m, &w, &w + 1, wp, n, n + 4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w + 1); + assert(np == n + 2); + assert(n[0] == char(0xD1)); + assert(n[1] == char(0x93)); + assert(n[2] == char(0x85)); + assert(n[3] == char(0)); - w = 0x453; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(n[0] == char(0xD1)); - assert(n[1] == char(0x93)); - assert(n[2] == char(0)); - assert(n[3] == char(0)); + w = 0x56; + r = c.out(m, &w, &w + 1, wp, n, n + 4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w + 1); + assert(np == n + 1); + assert(n[0] == char(0x56)); + assert(n[1] == char(0x93)); + assert(n[2] == char(0x85)); + assert(n[3] == char(0)); + } + { + typedef std::codecvt_utf8 C; + C c; + CharT w = 0x1005; + char n[4] = {0}; + const CharT* wp = nullptr; + std::mbstate_t m; + char* np = nullptr; + std::codecvt_base::result r = c.out(m, &w, &w + 1, wp, n, n + 4, np); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(n[0] == char(0)); + assert(n[1] == char(0)); + assert(n[2] == char(0)); + assert(n[3] == char(0)); - w = 0x56; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+1); - assert(n[0] == char(0x56)); - assert(n[1] == char(0x93)); - assert(n[2] == char(0)); - assert(n[3] == char(0)); - } - { - typedef std::codecvt_utf8 C; - C c; - wchar_t w = 0x40003; - char n[7] = {0}; - const wchar_t* wp = nullptr; - std::mbstate_t m; - char* np = nullptr; - std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+7, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+7); - assert(n[0] == char(0xEF)); - assert(n[1] == char(0xBB)); - assert(n[2] == char(0xBF)); - assert(n[3] == char(0xF1)); - assert(n[4] == char(0x80)); - assert(n[5] == char(0x80)); - assert(n[6] == char(0x83)); + w = 0x453; + r = c.out(m, &w, &w + 1, wp, n, n + 4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w + 1); + assert(np == n + 2); + assert(n[0] == char(0xD1)); + assert(n[1] == char(0x93)); + assert(n[2] == char(0)); + assert(n[3] == char(0)); - w = 0x1005; - r = c.out(m, &w, &w+1, wp, n, n+7, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+6); - assert(n[0] == char(0xEF)); - assert(n[1] == char(0xBB)); - assert(n[2] == char(0xBF)); - assert(n[3] == char(0xE1)); - assert(n[4] == char(0x80)); - assert(n[5] == char(0x85)); - assert(n[6] == char(0x83)); + w = 0x56; + r = c.out(m, &w, &w + 1, wp, n, n + 4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w + 1); + assert(np == n + 1); + assert(n[0] == char(0x56)); + assert(n[1] == char(0x93)); + assert(n[2] == char(0)); + assert(n[3] == char(0)); + } + { + typedef std::codecvt_utf8 C; + C c; + CharT w = 0x1005; + char n[7] = {0}; + const CharT* wp = nullptr; + std::mbstate_t m; + char* np = nullptr; + std::codecvt_base::result r = c.out(m, &w, &w + 1, wp, n, n + 7, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w + 1); + assert(np == n + 6); + assert(n[0] == char(0xEF)); + assert(n[1] == char(0xBB)); + assert(n[2] == char(0xBF)); + assert(n[3] == char(0xE1)); + assert(n[4] == char(0x80)); + assert(n[5] == char(0x85)); + assert(n[6] == char(0)); - w = 0x453; - r = c.out(m, &w, &w+1, wp, n, n+7, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+5); - assert(n[0] == char(0xEF)); - assert(n[1] == char(0xBB)); - assert(n[2] == char(0xBF)); - assert(n[3] == char(0xD1)); - assert(n[4] == char(0x93)); - assert(n[5] == char(0x85)); - assert(n[6] == char(0x83)); + w = 0x453; + r = c.out(m, &w, &w + 1, wp, n, n + 7, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w + 1); + assert(np == n + 5); + assert(n[0] == char(0xEF)); + assert(n[1] == char(0xBB)); + assert(n[2] == char(0xBF)); + assert(n[3] == char(0xD1)); + assert(n[4] == char(0x93)); + assert(n[5] == char(0x85)); + assert(n[6] == char(0)); - w = 0x56; - r = c.out(m, &w, &w+1, wp, n, n+7, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+4); - assert(n[0] == char(0xEF)); - assert(n[1] == char(0xBB)); - assert(n[2] == char(0xBF)); - assert(n[3] == char(0x56)); - assert(n[4] == char(0x93)); - assert(n[5] == char(0x85)); - assert(n[6] == char(0x83)); - } - { - typedef std::codecvt_utf8 C; - C c; - char32_t w = 0x40003; - char n[4] = {0}; - const char32_t* wp = nullptr; - std::mbstate_t m; - char* np = nullptr; - std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+4); - assert(n[0] == char(0xF1)); - assert(n[1] == char(0x80)); - assert(n[2] == char(0x80)); - assert(n[3] == char(0x83)); - - w = 0x1005; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+3); - assert(n[0] == char(0xE1)); - assert(n[1] == char(0x80)); - assert(n[2] == char(0x85)); - assert(n[3] == char(0x83)); - - w = 0x453; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(n[0] == char(0xD1)); - assert(n[1] == char(0x93)); - assert(n[2] == char(0x85)); - assert(n[3] == char(0x83)); - - w = 0x56; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+1); - assert(n[0] == char(0x56)); - assert(n[1] == char(0x93)); - assert(n[2] == char(0x85)); - assert(n[3] == char(0x83)); - } - { - typedef std::codecvt_utf8 C; - C c; - char32_t w = 0x40003; - char n[4] = {0}; - const char32_t* wp = nullptr; - std::mbstate_t m; - char* np = nullptr; - std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::error); - assert(wp == &w); - assert(np == n); - assert(n[0] == char(0)); - assert(n[1] == char(0)); - assert(n[2] == char(0)); - assert(n[3] == char(0)); - - w = 0x1005; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::error); - assert(wp == &w); - assert(np == n); - assert(n[0] == char(0)); - assert(n[1] == char(0)); - assert(n[2] == char(0)); - assert(n[3] == char(0)); - - w = 0x453; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(n[0] == char(0xD1)); - assert(n[1] == char(0x93)); - assert(n[2] == char(0)); - assert(n[3] == char(0)); - - w = 0x56; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+1); - assert(n[0] == char(0x56)); - assert(n[1] == char(0x93)); - assert(n[2] == char(0)); - assert(n[3] == char(0)); - } - { - typedef std::codecvt_utf8 C; - C c; - char32_t w = 0x40003; - char n[7] = {0}; - const char32_t* wp = nullptr; - std::mbstate_t m; - char* np = nullptr; - std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+7, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+7); - assert(n[0] == char(0xEF)); - assert(n[1] == char(0xBB)); - assert(n[2] == char(0xBF)); - assert(n[3] == char(0xF1)); - assert(n[4] == char(0x80)); - assert(n[5] == char(0x80)); - assert(n[6] == char(0x83)); - - w = 0x1005; - r = c.out(m, &w, &w+1, wp, n, n+7, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+6); - assert(n[0] == char(0xEF)); - assert(n[1] == char(0xBB)); - assert(n[2] == char(0xBF)); - assert(n[3] == char(0xE1)); - assert(n[4] == char(0x80)); - assert(n[5] == char(0x85)); - assert(n[6] == char(0x83)); - - w = 0x453; - r = c.out(m, &w, &w+1, wp, n, n+7, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+5); - assert(n[0] == char(0xEF)); - assert(n[1] == char(0xBB)); - assert(n[2] == char(0xBF)); - assert(n[3] == char(0xD1)); - assert(n[4] == char(0x93)); - assert(n[5] == char(0x85)); - assert(n[6] == char(0x83)); - - w = 0x56; - r = c.out(m, &w, &w+1, wp, n, n+7, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+4); - assert(n[0] == char(0xEF)); - assert(n[1] == char(0xBB)); - assert(n[2] == char(0xBF)); - assert(n[3] == char(0x56)); - assert(n[4] == char(0x93)); - assert(n[5] == char(0x85)); - assert(n[6] == char(0x83)); - } - { - typedef std::codecvt_utf8 C; - C c; - char16_t w = 0x1005; - char n[4] = {0}; - const char16_t* wp = nullptr; - std::mbstate_t m; - char* np = nullptr; - std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+3); - assert(n[0] == char(0xE1)); - assert(n[1] == char(0x80)); - assert(n[2] == char(0x85)); - assert(n[3] == char(0)); - - w = 0x453; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(n[0] == char(0xD1)); - assert(n[1] == char(0x93)); - assert(n[2] == char(0x85)); - assert(n[3] == char(0)); - - w = 0x56; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+1); - assert(n[0] == char(0x56)); - assert(n[1] == char(0x93)); - assert(n[2] == char(0x85)); - assert(n[3] == char(0)); - } - { - typedef std::codecvt_utf8 C; - C c; - char16_t w = 0x1005; - char n[4] = {0}; - const char16_t* wp = nullptr; - std::mbstate_t m; - char* np = nullptr; - std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::error); - assert(wp == &w); - assert(np == n); - assert(n[0] == char(0)); - assert(n[1] == char(0)); - assert(n[2] == char(0)); - assert(n[3] == char(0)); - - w = 0x453; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(n[0] == char(0xD1)); - assert(n[1] == char(0x93)); - assert(n[2] == char(0)); - assert(n[3] == char(0)); - - w = 0x56; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+1); - assert(n[0] == char(0x56)); - assert(n[1] == char(0x93)); - assert(n[2] == char(0)); - assert(n[3] == char(0)); - } - { - typedef std::codecvt_utf8 C; - C c; - char16_t w = 0x1005; - char n[7] = {0}; - const char16_t* wp = nullptr; - std::mbstate_t m; - char* np = nullptr; - std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+7, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+6); - assert(n[0] == char(0xEF)); - assert(n[1] == char(0xBB)); - assert(n[2] == char(0xBF)); - assert(n[3] == char(0xE1)); - assert(n[4] == char(0x80)); - assert(n[5] == char(0x85)); - assert(n[6] == char(0)); - - w = 0x453; - r = c.out(m, &w, &w+1, wp, n, n+7, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+5); - assert(n[0] == char(0xEF)); - assert(n[1] == char(0xBB)); - assert(n[2] == char(0xBF)); - assert(n[3] == char(0xD1)); - assert(n[4] == char(0x93)); - assert(n[5] == char(0x85)); - assert(n[6] == char(0)); - - w = 0x56; - r = c.out(m, &w, &w+1, wp, n, n+7, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+4); - assert(n[0] == char(0xEF)); - assert(n[1] == char(0xBB)); - assert(n[2] == char(0xBF)); - assert(n[3] == char(0x56)); - assert(n[4] == char(0x93)); - assert(n[5] == char(0x85)); - assert(n[6] == char(0)); - } + w = 0x56; + r = c.out(m, &w, &w + 1, wp, n, n + 7, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w + 1); + assert(np == n + 4); + assert(n[0] == char(0xEF)); + assert(n[1] == char(0xBB)); + assert(n[2] == char(0xBF)); + assert(n[3] == char(0x56)); + assert(n[4] == char(0x93)); + assert(n[5] == char(0x85)); + assert(n[6] == char(0)); + } +} + +template +void TestHelper::test() { + { + typedef std::codecvt_utf8 C; + C c; + CharT w = 0x40003; + char n[4] = {0}; + const CharT* wp = nullptr; + std::mbstate_t m; + char* np = nullptr; + std::codecvt_base::result r = c.out(m, &w, &w + 1, wp, n, n + 4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w + 1); + assert(np == n + 4); + assert(n[0] == char(0xF1)); + assert(n[1] == char(0x80)); + assert(n[2] == char(0x80)); + assert(n[3] == char(0x83)); + + w = 0x1005; + r = c.out(m, &w, &w + 1, wp, n, n + 4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w + 1); + assert(np == n + 3); + assert(n[0] == char(0xE1)); + assert(n[1] == char(0x80)); + assert(n[2] == char(0x85)); + assert(n[3] == char(0x83)); + + w = 0x453; + r = c.out(m, &w, &w + 1, wp, n, n + 4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w + 1); + assert(np == n + 2); + assert(n[0] == char(0xD1)); + assert(n[1] == char(0x93)); + assert(n[2] == char(0x85)); + assert(n[3] == char(0x83)); + + w = 0x56; + r = c.out(m, &w, &w + 1, wp, n, n + 4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w + 1); + assert(np == n + 1); + assert(n[0] == char(0x56)); + assert(n[1] == char(0x93)); + assert(n[2] == char(0x85)); + assert(n[3] == char(0x83)); + } + { + typedef std::codecvt_utf8 C; + C c; + CharT w = 0x40003; + char n[4] = {0}; + const CharT* wp = nullptr; + std::mbstate_t m; + char* np = nullptr; + std::codecvt_base::result r = c.out(m, &w, &w + 1, wp, n, n + 4, np); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(n[0] == char(0)); + assert(n[1] == char(0)); + assert(n[2] == char(0)); + assert(n[3] == char(0)); + + w = 0x1005; + r = c.out(m, &w, &w + 1, wp, n, n + 4, np); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(n[0] == char(0)); + assert(n[1] == char(0)); + assert(n[2] == char(0)); + assert(n[3] == char(0)); + + w = 0x453; + r = c.out(m, &w, &w + 1, wp, n, n + 4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w + 1); + assert(np == n + 2); + assert(n[0] == char(0xD1)); + assert(n[1] == char(0x93)); + assert(n[2] == char(0)); + assert(n[3] == char(0)); + + w = 0x56; + r = c.out(m, &w, &w + 1, wp, n, n + 4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w + 1); + assert(np == n + 1); + assert(n[0] == char(0x56)); + assert(n[1] == char(0x93)); + assert(n[2] == char(0)); + assert(n[3] == char(0)); + } + { + typedef std::codecvt_utf8 C; + C c; + CharT w = 0x40003; + char n[7] = {0}; + const CharT* wp = nullptr; + std::mbstate_t m; + char* np = nullptr; + std::codecvt_base::result r = c.out(m, &w, &w + 1, wp, n, n + 7, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w + 1); + assert(np == n + 7); + assert(n[0] == char(0xEF)); + assert(n[1] == char(0xBB)); + assert(n[2] == char(0xBF)); + assert(n[3] == char(0xF1)); + assert(n[4] == char(0x80)); + assert(n[5] == char(0x80)); + assert(n[6] == char(0x83)); + + w = 0x1005; + r = c.out(m, &w, &w + 1, wp, n, n + 7, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w + 1); + assert(np == n + 6); + assert(n[0] == char(0xEF)); + assert(n[1] == char(0xBB)); + assert(n[2] == char(0xBF)); + assert(n[3] == char(0xE1)); + assert(n[4] == char(0x80)); + assert(n[5] == char(0x85)); + assert(n[6] == char(0x83)); + + w = 0x453; + r = c.out(m, &w, &w + 1, wp, n, n + 7, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w + 1); + assert(np == n + 5); + assert(n[0] == char(0xEF)); + assert(n[1] == char(0xBB)); + assert(n[2] == char(0xBF)); + assert(n[3] == char(0xD1)); + assert(n[4] == char(0x93)); + assert(n[5] == char(0x85)); + assert(n[6] == char(0x83)); + + w = 0x56; + r = c.out(m, &w, &w + 1, wp, n, n + 7, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w + 1); + assert(np == n + 4); + assert(n[0] == char(0xEF)); + assert(n[1] == char(0xBB)); + assert(n[2] == char(0xBF)); + assert(n[3] == char(0x56)); + assert(n[4] == char(0x93)); + assert(n[5] == char(0x85)); + assert(n[6] == char(0x83)); + } +} + +int main() { + TestHelper::test(); + TestHelper::test(); + TestHelper::test(); } diff --git a/test/std/localization/locale.stdcvt/codecvt_utf8_utf16_in.pass.cpp b/test/std/localization/locale.stdcvt/codecvt_utf8_utf16_in.pass.cpp index 0cd941764aa0..392d66f2298c 100644 --- a/test/std/localization/locale.stdcvt/codecvt_utf8_utf16_in.pass.cpp +++ b/test/std/localization/locale.stdcvt/codecvt_utf8_utf16_in.pass.cpp @@ -25,348 +25,220 @@ #include #include -int main() -{ - { - typedef std::codecvt_utf8_utf16 C; - C c; - wchar_t w[2] = {0}; - char n[4] = {char(0xF1), char(0x80), char(0x80), char(0x83)}; - wchar_t* wp = nullptr; - std::mbstate_t m; - const char* np = nullptr; - std::codecvt_base::result r = c.in(m, n, n+4, np, w, w+2, wp); - assert(r == std::codecvt_base::ok); - assert(wp == w+2); - assert(np == n+4); - assert(w[0] == 0xD8C0); - assert(w[1] == 0xDC03); +template +struct TestHelper; +template +struct TestHelper { + static void test(); +}; +template +struct TestHelper { + static void test(); +}; - n[0] = char(0xE1); - n[1] = char(0x80); - n[2] = char(0x85); - r = c.in(m, n, n+3, np, w, w+2, wp); - assert(r == std::codecvt_base::ok); - assert(wp == w+1); - assert(np == n+3); - assert(w[0] == 0x1005); +template +void TestHelper::test() { + { + typedef std::codecvt_utf8_utf16 C; + C c; + CharT w[2] = {0}; + char n[4] = {char(0xF1), char(0x80), char(0x80), char(0x83)}; + CharT* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n + 4, np, w, w + 2, wp); + assert(r == std::codecvt_base::error); + assert(wp == w); + assert(np == n); - n[0] = char(0xD1); - n[1] = char(0x93); - r = c.in(m, n, n+2, np, w, w+2, wp); - assert(r == std::codecvt_base::ok); - assert(wp == w+1); - assert(np == n+2); - assert(w[0] == 0x0453); + n[0] = char(0xE1); + n[1] = char(0x80); + n[2] = char(0x85); + r = c.in(m, n, n + 3, np, w, w + 2, wp); + assert(r == std::codecvt_base::error); + assert(wp == w); + assert(np == n); - n[0] = char(0x56); - r = c.in(m, n, n+1, np, w, w+2, wp); - assert(r == std::codecvt_base::ok); - assert(wp == w+1); - assert(np == n+1); - assert(w[0] == 0x0056); - } - { - typedef std::codecvt_utf8_utf16 C; - C c; - wchar_t w[2] = {0}; - char n[4] = {char(0xF1), char(0x80), char(0x80), char(0x83)}; - wchar_t* wp = nullptr; - std::mbstate_t m; - const char* np = nullptr; - std::codecvt_base::result r = c.in(m, n, n+4, np, w, w+2, wp); - assert(r == std::codecvt_base::error); - assert(wp == w); - assert(np == n); + n[0] = char(0xD1); + n[1] = char(0x93); + r = c.in(m, n, n + 2, np, w, w + 2, wp); + assert(r == std::codecvt_base::ok); + assert(wp == w + 1); + assert(np == n + 2); + assert(w[0] == 0x0453); - n[0] = char(0xE1); - n[1] = char(0x80); - n[2] = char(0x85); - r = c.in(m, n, n+3, np, w, w+2, wp); - assert(r == std::codecvt_base::error); - assert(wp == w); - assert(np == n); + n[0] = char(0x56); + r = c.in(m, n, n + 1, np, w, w + 2, wp); + assert(r == std::codecvt_base::ok); + assert(wp == w + 1); + assert(np == n + 1); + assert(w[0] == 0x0056); + } + { + typedef std::codecvt_utf8_utf16 C; + C c; + CharT w[2] = {0}; + char n[7] = {char(0xEF), char(0xBB), char(0xBF), char(0xF1), + char(0x80), char(0x80), char(0x83)}; + CharT* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n + 7, np, w, w + 2, wp); + assert(r == std::codecvt_base::ok); + assert(wp == w + 2); + assert(np == n + 7); + assert(w[0] == 0xD8C0); + assert(w[1] == 0xDC03); - n[0] = char(0xD1); - n[1] = char(0x93); - r = c.in(m, n, n+2, np, w, w+2, wp); - assert(r == std::codecvt_base::ok); - assert(wp == w+1); - assert(np == n+2); - assert(w[0] == 0x0453); + n[0] = char(0xE1); + n[1] = char(0x80); + n[2] = char(0x85); + r = c.in(m, n, n + 3, np, w, w + 2, wp); + assert(r == std::codecvt_base::ok); + assert(wp == w + 1); + assert(np == n + 3); + assert(w[0] == 0x1005); - n[0] = char(0x56); - r = c.in(m, n, n+1, np, w, w+2, wp); - assert(r == std::codecvt_base::ok); - assert(wp == w+1); - assert(np == n+1); - assert(w[0] == 0x0056); - } - { - typedef std::codecvt_utf8_utf16 C; - C c; - wchar_t w[2] = {0}; - char n[7] = {char(0xEF), char(0xBB), char(0xBF), char(0xF1), char(0x80), char(0x80), char(0x83)}; - wchar_t* wp = nullptr; - std::mbstate_t m; - const char* np = nullptr; - std::codecvt_base::result r = c.in(m, n, n+7, np, w, w+2, wp); - assert(r == std::codecvt_base::ok); - assert(wp == w+2); - assert(np == n+7); - assert(w[0] == 0xD8C0); - assert(w[1] == 0xDC03); + n[0] = char(0xD1); + n[1] = char(0x93); + r = c.in(m, n, n + 2, np, w, w + 2, wp); + assert(r == std::codecvt_base::ok); + assert(wp == w + 1); + assert(np == n + 2); + assert(w[0] == 0x0453); - n[0] = char(0xE1); - n[1] = char(0x80); - n[2] = char(0x85); - r = c.in(m, n, n+3, np, w, w+2, wp); - assert(r == std::codecvt_base::ok); - assert(wp == w+1); - assert(np == n+3); - assert(w[0] == 0x1005); - - n[0] = char(0xD1); - n[1] = char(0x93); - r = c.in(m, n, n+2, np, w, w+2, wp); - assert(r == std::codecvt_base::ok); - assert(wp == w+1); - assert(np == n+2); - assert(w[0] == 0x0453); - - n[0] = char(0x56); - r = c.in(m, n, n+1, np, w, w+2, wp); - assert(r == std::codecvt_base::ok); - assert(wp == w+1); - assert(np == n+1); - assert(w[0] == 0x0056); - } - { - typedef std::codecvt_utf8_utf16 C; - C c; - char32_t w[2] = {0}; - char n[4] = {char(0xF1), char(0x80), char(0x80), char(0x83)}; - char32_t* wp = nullptr; - std::mbstate_t m; - const char* np = nullptr; - std::codecvt_base::result r = c.in(m, n, n+4, np, w, w+2, wp); - assert(r == std::codecvt_base::ok); - assert(wp == w+2); - assert(np == n+4); - assert(w[0] == 0xD8C0); - assert(w[1] == 0xDC03); - - n[0] = char(0xE1); - n[1] = char(0x80); - n[2] = char(0x85); - r = c.in(m, n, n+3, np, w, w+2, wp); - assert(r == std::codecvt_base::ok); - assert(wp == w+1); - assert(np == n+3); - assert(w[0] == 0x1005); - - n[0] = char(0xD1); - n[1] = char(0x93); - r = c.in(m, n, n+2, np, w, w+2, wp); - assert(r == std::codecvt_base::ok); - assert(wp == w+1); - assert(np == n+2); - assert(w[0] == 0x0453); - - n[0] = char(0x56); - r = c.in(m, n, n+1, np, w, w+2, wp); - assert(r == std::codecvt_base::ok); - assert(wp == w+1); - assert(np == n+1); - assert(w[0] == 0x0056); - } - { - typedef std::codecvt_utf8_utf16 C; - C c; - char32_t w[2] = {0}; - char n[4] = {char(0xF1), char(0x80), char(0x80), char(0x83)}; - char32_t* wp = nullptr; - std::mbstate_t m; - const char* np = nullptr; - std::codecvt_base::result r = c.in(m, n, n+4, np, w, w+2, wp); - assert(r == std::codecvt_base::error); - assert(wp == w); - assert(np == n); - - n[0] = char(0xE1); - n[1] = char(0x80); - n[2] = char(0x85); - r = c.in(m, n, n+3, np, w, w+2, wp); - assert(r == std::codecvt_base::error); - assert(wp == w); - assert(np == n); - - n[0] = char(0xD1); - n[1] = char(0x93); - r = c.in(m, n, n+2, np, w, w+2, wp); - assert(r == std::codecvt_base::ok); - assert(wp == w+1); - assert(np == n+2); - assert(w[0] == 0x0453); - - n[0] = char(0x56); - r = c.in(m, n, n+1, np, w, w+2, wp); - assert(r == std::codecvt_base::ok); - assert(wp == w+1); - assert(np == n+1); - assert(w[0] == 0x0056); - } - { - typedef std::codecvt_utf8_utf16 C; - C c; - char32_t w[2] = {0}; - char n[7] = {char(0xEF), char(0xBB), char(0xBF), char(0xF1), char(0x80), char(0x80), char(0x83)}; - char32_t* wp = nullptr; - std::mbstate_t m; - const char* np = nullptr; - std::codecvt_base::result r = c.in(m, n, n+7, np, w, w+2, wp); - assert(r == std::codecvt_base::ok); - assert(wp == w+2); - assert(np == n+7); - assert(w[0] == 0xD8C0); - assert(w[1] == 0xDC03); - - n[0] = char(0xE1); - n[1] = char(0x80); - n[2] = char(0x85); - r = c.in(m, n, n+3, np, w, w+2, wp); - assert(r == std::codecvt_base::ok); - assert(wp == w+1); - assert(np == n+3); - assert(w[0] == 0x1005); - - n[0] = char(0xD1); - n[1] = char(0x93); - r = c.in(m, n, n+2, np, w, w+2, wp); - assert(r == std::codecvt_base::ok); - assert(wp == w+1); - assert(np == n+2); - assert(w[0] == 0x0453); - - n[0] = char(0x56); - r = c.in(m, n, n+1, np, w, w+2, wp); - assert(r == std::codecvt_base::ok); - assert(wp == w+1); - assert(np == n+1); - assert(w[0] == 0x0056); - } - { - typedef std::codecvt_utf8_utf16 C; - C c; - char16_t w[2] = {0}; - char n[4] = {char(0xF1), char(0x80), char(0x80), char(0x83)}; - char16_t* wp = nullptr; - std::mbstate_t m; - const char* np = nullptr; - std::codecvt_base::result r = c.in(m, n, n+4, np, w, w+2, wp); - assert(r == std::codecvt_base::ok); - assert(wp == w+2); - assert(np == n+4); - assert(w[0] == 0xD8C0); - assert(w[1] == 0xDC03); - - n[0] = char(0xE1); - n[1] = char(0x80); - n[2] = char(0x85); - r = c.in(m, n, n+3, np, w, w+2, wp); - assert(r == std::codecvt_base::ok); - assert(wp == w+1); - assert(np == n+3); - assert(w[0] == 0x1005); - - n[0] = char(0xD1); - n[1] = char(0x93); - r = c.in(m, n, n+2, np, w, w+2, wp); - assert(r == std::codecvt_base::ok); - assert(wp == w+1); - assert(np == n+2); - assert(w[0] == 0x0453); - - n[0] = char(0x56); - r = c.in(m, n, n+1, np, w, w+2, wp); - assert(r == std::codecvt_base::ok); - assert(wp == w+1); - assert(np == n+1); - assert(w[0] == 0x0056); - } - { - typedef std::codecvt_utf8_utf16 C; - C c; - char16_t w[2] = {0}; - char n[4] = {char(0xF1), char(0x80), char(0x80), char(0x83)}; - char16_t* wp = nullptr; - std::mbstate_t m; - const char* np = nullptr; - std::codecvt_base::result r = c.in(m, n, n+4, np, w, w+2, wp); - assert(r == std::codecvt_base::error); - assert(wp == w); - assert(np == n); - - n[0] = char(0xE1); - n[1] = char(0x80); - n[2] = char(0x85); - r = c.in(m, n, n+3, np, w, w+2, wp); - assert(r == std::codecvt_base::error); - assert(wp == w); - assert(np == n); - - n[0] = char(0xD1); - n[1] = char(0x93); - r = c.in(m, n, n+2, np, w, w+2, wp); - assert(r == std::codecvt_base::ok); - assert(wp == w+1); - assert(np == n+2); - assert(w[0] == 0x0453); - - n[0] = char(0x56); - r = c.in(m, n, n+1, np, w, w+2, wp); - assert(r == std::codecvt_base::ok); - assert(wp == w+1); - assert(np == n+1); - assert(w[0] == 0x0056); - } - { - typedef std::codecvt_utf8_utf16 C; - C c; - char16_t w[2] = {0}; - char n[7] = {char(0xEF), char(0xBB), char(0xBF), char(0xF1), char(0x80), char(0x80), char(0x83)}; - char16_t* wp = nullptr; - std::mbstate_t m; - const char* np = nullptr; - std::codecvt_base::result r = c.in(m, n, n+7, np, w, w+2, wp); - assert(r == std::codecvt_base::ok); - assert(wp == w+2); - assert(np == n+7); - assert(w[0] == 0xD8C0); - assert(w[1] == 0xDC03); - - n[0] = char(0xE1); - n[1] = char(0x80); - n[2] = char(0x85); - r = c.in(m, n, n+3, np, w, w+2, wp); - assert(r == std::codecvt_base::ok); - assert(wp == w+1); - assert(np == n+3); - assert(w[0] == 0x1005); - - n[0] = char(0xD1); - n[1] = char(0x93); - r = c.in(m, n, n+2, np, w, w+2, wp); - assert(r == std::codecvt_base::ok); - assert(wp == w+1); - assert(np == n+2); - assert(w[0] == 0x0453); - - n[0] = char(0x56); - r = c.in(m, n, n+1, np, w, w+2, wp); - assert(r == std::codecvt_base::ok); - assert(wp == w+1); - assert(np == n+1); - assert(w[0] == 0x0056); - } + n[0] = char(0x56); + r = c.in(m, n, n + 1, np, w, w + 2, wp); + assert(r == std::codecvt_base::ok); + assert(wp == w + 1); + assert(np == n + 1); + assert(w[0] == 0x0056); + } +} + +template +void TestHelper::test() { + { + typedef std::codecvt_utf8_utf16 C; + C c; + CharT w[2] = {0}; + char n[4] = {char(0xF1), char(0x80), char(0x80), char(0x83)}; + CharT* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n + 4, np, w, w + 2, wp); + assert(r == std::codecvt_base::ok); + assert(wp == w + 2); + assert(np == n + 4); + assert(w[0] == 0xD8C0); + assert(w[1] == 0xDC03); + + n[0] = char(0xE1); + n[1] = char(0x80); + n[2] = char(0x85); + r = c.in(m, n, n + 3, np, w, w + 2, wp); + assert(r == std::codecvt_base::ok); + assert(wp == w + 1); + assert(np == n + 3); + assert(w[0] == 0x1005); + + n[0] = char(0xD1); + n[1] = char(0x93); + r = c.in(m, n, n + 2, np, w, w + 2, wp); + assert(r == std::codecvt_base::ok); + assert(wp == w + 1); + assert(np == n + 2); + assert(w[0] == 0x0453); + + n[0] = char(0x56); + r = c.in(m, n, n + 1, np, w, w + 2, wp); + assert(r == std::codecvt_base::ok); + assert(wp == w + 1); + assert(np == n + 1); + assert(w[0] == 0x0056); + } + { + typedef std::codecvt_utf8_utf16 C; + C c; + CharT w[2] = {0}; + char n[4] = {char(0xF1), char(0x80), char(0x80), char(0x83)}; + CharT* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n + 4, np, w, w + 2, wp); + assert(r == std::codecvt_base::error); + assert(wp == w); + assert(np == n); + + n[0] = char(0xE1); + n[1] = char(0x80); + n[2] = char(0x85); + r = c.in(m, n, n + 3, np, w, w + 2, wp); + assert(r == std::codecvt_base::error); + assert(wp == w); + assert(np == n); + + n[0] = char(0xD1); + n[1] = char(0x93); + r = c.in(m, n, n + 2, np, w, w + 2, wp); + assert(r == std::codecvt_base::ok); + assert(wp == w + 1); + assert(np == n + 2); + assert(w[0] == 0x0453); + + n[0] = char(0x56); + r = c.in(m, n, n + 1, np, w, w + 2, wp); + assert(r == std::codecvt_base::ok); + assert(wp == w + 1); + assert(np == n + 1); + assert(w[0] == 0x0056); + } + { + typedef std::codecvt_utf8_utf16 C; + C c; + CharT w[2] = {0}; + char n[7] = {char(0xEF), char(0xBB), char(0xBF), char(0xF1), + char(0x80), char(0x80), char(0x83)}; + CharT* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n + 7, np, w, w + 2, wp); + assert(r == std::codecvt_base::ok); + assert(wp == w + 2); + assert(np == n + 7); + assert(w[0] == 0xD8C0); + assert(w[1] == 0xDC03); + + n[0] = char(0xE1); + n[1] = char(0x80); + n[2] = char(0x85); + r = c.in(m, n, n + 3, np, w, w + 2, wp); + assert(r == std::codecvt_base::ok); + assert(wp == w + 1); + assert(np == n + 3); + assert(w[0] == 0x1005); + + n[0] = char(0xD1); + n[1] = char(0x93); + r = c.in(m, n, n + 2, np, w, w + 2, wp); + assert(r == std::codecvt_base::ok); + assert(wp == w + 1); + assert(np == n + 2); + assert(w[0] == 0x0453); + + n[0] = char(0x56); + r = c.in(m, n, n + 1, np, w, w + 2, wp); + assert(r == std::codecvt_base::ok); + assert(wp == w + 1); + assert(np == n + 1); + assert(w[0] == 0x0056); + } +} + +int main() { +#ifndef _WIN32 + TestHelper::test(); +#endif + TestHelper::test(); + TestHelper::test(); } diff --git a/test/std/localization/locale.stdcvt/codecvt_utf8_utf16_out.pass.cpp b/test/std/localization/locale.stdcvt/codecvt_utf8_utf16_out.pass.cpp index 29c5342e4545..ced2a36a4fec 100644 --- a/test/std/localization/locale.stdcvt/codecvt_utf8_utf16_out.pass.cpp +++ b/test/std/localization/locale.stdcvt/codecvt_utf8_utf16_out.pass.cpp @@ -25,391 +25,285 @@ #include #include -int main() -{ - { - typedef std::codecvt_utf8_utf16 C; - C c; - wchar_t w[2] = {0xD8C0, 0xDC03}; - char n[4] = {0}; - const wchar_t* wp = nullptr; - std::mbstate_t m; - char* np = nullptr; - std::codecvt_base::result r = c.out(m, w, w+2, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == w+2); - assert(np == n+4); - assert(n[0] == char(0xF1)); - assert(n[1] == char(0x80)); - assert(n[2] == char(0x80)); - assert(n[3] == char(0x83)); +template +struct TestHelper; +template +struct TestHelper { + static void test(); +}; +template +struct TestHelper { + static void test(); +}; - w[0] = 0x1005; - r = c.out(m, w, w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == w+1); - assert(np == n+3); - assert(n[0] == char(0xE1)); - assert(n[1] == char(0x80)); - assert(n[2] == char(0x85)); +template +void TestHelper::test() { + { + typedef std::codecvt_utf8_utf16 C; + C c; + CharT w[2] = {0xD8C0, 0xDC03}; + char n[4] = {0}; + const CharT* wp = nullptr; + std::mbstate_t m; + char* np = nullptr; + std::codecvt_base::result r = c.out(m, w, w + 2, wp, n, n + 4, np); + assert(r == std::codecvt_base::ok); + assert(wp == w + 2); + assert(np == n + 4); + assert(n[0] == char(0xF1)); + assert(n[1] == char(0x80)); + assert(n[2] == char(0x80)); + assert(n[3] == char(0x83)); - w[0] = 0x453; - r = c.out(m, w, w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == w+1); - assert(np == n+2); - assert(n[0] == char(0xD1)); - assert(n[1] == char(0x93)); + w[0] = 0x1005; + r = c.out(m, w, w + 1, wp, n, n + 4, np); + assert(r == std::codecvt_base::ok); + assert(wp == w + 1); + assert(np == n + 3); + assert(n[0] == char(0xE1)); + assert(n[1] == char(0x80)); + assert(n[2] == char(0x85)); - w[0] = 0x56; - r = c.out(m, w, w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == w+1); - assert(np == n+1); - assert(n[0] == char(0x56)); - } - { - typedef std::codecvt_utf8_utf16 C; - C c; - wchar_t w[2] = {0xD8C0, 0xDC03}; - char n[4] = {0}; - const wchar_t* wp = nullptr; - std::mbstate_t m; - char* np = nullptr; - std::codecvt_base::result r = c.out(m, w, w+2, wp, n, n+4, np); - assert(r == std::codecvt_base::error); - assert(wp == w); - assert(np == n); + w[0] = 0x453; + r = c.out(m, w, w + 1, wp, n, n + 4, np); + assert(r == std::codecvt_base::ok); + assert(wp == w + 1); + assert(np == n + 2); + assert(n[0] == char(0xD1)); + assert(n[1] == char(0x93)); - w[0] = 0x1005; - r = c.out(m, w, w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::error); - assert(wp == w); - assert(np == n); + w[0] = 0x56; + r = c.out(m, w, w + 1, wp, n, n + 4, np); + assert(r == std::codecvt_base::ok); + assert(wp == w + 1); + assert(np == n + 1); + assert(n[0] == char(0x56)); + } + { + typedef std::codecvt_utf8_utf16 C; + C c; + CharT w[2] = {0xD8C0, 0xDC03}; + char n[4] = {0}; + const CharT* wp = nullptr; + std::mbstate_t m; + char* np = nullptr; + std::codecvt_base::result r = c.out(m, w, w + 2, wp, n, n + 4, np); + assert(r == std::codecvt_base::error); + assert(wp == w); + assert(np == n); - w[0] = 0x453; - r = c.out(m, w, w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == w+1); - assert(np == n+2); - assert(n[0] == char(0xD1)); - assert(n[1] == char(0x93)); + w[0] = 0x1005; + r = c.out(m, w, w + 1, wp, n, n + 4, np); + assert(r == std::codecvt_base::error); + assert(wp == w); + assert(np == n); - w[0] = 0x56; - r = c.out(m, w, w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == w+1); - assert(np == n+1); - assert(n[0] == char(0x56)); - } - { - typedef std::codecvt_utf8_utf16 C; - C c; - wchar_t w[2] = {0xD8C0, 0xDC03}; - char n[7] = {0}; - const wchar_t* wp = nullptr; - std::mbstate_t m; - char* np = nullptr; - std::codecvt_base::result r = c.out(m, w, w+2, wp, n, n+7, np); - assert(r == std::codecvt_base::ok); - assert(wp == w+2); - assert(np == n+7); - assert(n[0] == char(0xEF)); - assert(n[1] == char(0xBB)); - assert(n[2] == char(0xBF)); - assert(n[3] == char(0xF1)); - assert(n[4] == char(0x80)); - assert(n[5] == char(0x80)); - assert(n[6] == char(0x83)); + w[0] = 0x453; + r = c.out(m, w, w + 1, wp, n, n + 4, np); + assert(r == std::codecvt_base::ok); + assert(wp == w + 1); + assert(np == n + 2); + assert(n[0] == char(0xD1)); + assert(n[1] == char(0x93)); - w[0] = 0x1005; - r = c.out(m, w, w+1, wp, n, n+7, np); - assert(r == std::codecvt_base::ok); - assert(wp == w+1); - assert(np == n+6); - assert(n[0] == char(0xEF)); - assert(n[1] == char(0xBB)); - assert(n[2] == char(0xBF)); - assert(n[3] == char(0xE1)); - assert(n[4] == char(0x80)); - assert(n[5] == char(0x85)); + w[0] = 0x56; + r = c.out(m, w, w + 1, wp, n, n + 4, np); + assert(r == std::codecvt_base::ok); + assert(wp == w + 1); + assert(np == n + 1); + assert(n[0] == char(0x56)); + } + { + typedef std::codecvt_utf8_utf16 C; + C c; + CharT w[2] = {0xD8C0, 0xDC03}; + char n[7] = {0}; + const CharT* wp = nullptr; + std::mbstate_t m; + char* np = nullptr; + std::codecvt_base::result r = c.out(m, w, w + 2, wp, n, n + 7, np); + assert(r == std::codecvt_base::ok); + assert(wp == w + 2); + assert(np == n + 7); + assert(n[0] == char(0xEF)); + assert(n[1] == char(0xBB)); + assert(n[2] == char(0xBF)); + assert(n[3] == char(0xF1)); + assert(n[4] == char(0x80)); + assert(n[5] == char(0x80)); + assert(n[6] == char(0x83)); - w[0] = 0x453; - r = c.out(m, w, w+1, wp, n, n+7, np); - assert(r == std::codecvt_base::ok); - assert(wp == w+1); - assert(np == n+5); - assert(n[0] == char(0xEF)); - assert(n[1] == char(0xBB)); - assert(n[2] == char(0xBF)); - assert(n[3] == char(0xD1)); - assert(n[4] == char(0x93)); + w[0] = 0x1005; + r = c.out(m, w, w + 1, wp, n, n + 7, np); + assert(r == std::codecvt_base::ok); + assert(wp == w + 1); + assert(np == n + 6); + assert(n[0] == char(0xEF)); + assert(n[1] == char(0xBB)); + assert(n[2] == char(0xBF)); + assert(n[3] == char(0xE1)); + assert(n[4] == char(0x80)); + assert(n[5] == char(0x85)); - w[0] = 0x56; - r = c.out(m, w, w+1, wp, n, n+7, np); - assert(r == std::codecvt_base::ok); - assert(wp == w+1); - assert(np == n+4); - assert(n[0] == char(0xEF)); - assert(n[1] == char(0xBB)); - assert(n[2] == char(0xBF)); - assert(n[3] == char(0x56)); - } - { - typedef std::codecvt_utf8_utf16 C; - C c; - char32_t w[2] = {0xD8C0, 0xDC03}; - char n[4] = {0}; - const char32_t* wp = nullptr; - std::mbstate_t m; - char* np = nullptr; - std::codecvt_base::result r = c.out(m, w, w+2, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == w+2); - assert(np == n+4); - assert(n[0] == char(0xF1)); - assert(n[1] == char(0x80)); - assert(n[2] == char(0x80)); - assert(n[3] == char(0x83)); + w[0] = 0x453; + r = c.out(m, w, w + 1, wp, n, n + 7, np); + assert(r == std::codecvt_base::ok); + assert(wp == w + 1); + assert(np == n + 5); + assert(n[0] == char(0xEF)); + assert(n[1] == char(0xBB)); + assert(n[2] == char(0xBF)); + assert(n[3] == char(0xD1)); + assert(n[4] == char(0x93)); - w[0] = 0x1005; - r = c.out(m, w, w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == w+1); - assert(np == n+3); - assert(n[0] == char(0xE1)); - assert(n[1] == char(0x80)); - assert(n[2] == char(0x85)); - - w[0] = 0x453; - r = c.out(m, w, w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == w+1); - assert(np == n+2); - assert(n[0] == char(0xD1)); - assert(n[1] == char(0x93)); - - w[0] = 0x56; - r = c.out(m, w, w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == w+1); - assert(np == n+1); - assert(n[0] == char(0x56)); - } - { - typedef std::codecvt_utf8_utf16 C; - C c; - char32_t w[2] = {0xD8C0, 0xDC03}; - char n[4] = {0}; - const char32_t* wp = nullptr; - std::mbstate_t m; - char* np = nullptr; - std::codecvt_base::result r = c.out(m, w, w+2, wp, n, n+4, np); - assert(r == std::codecvt_base::error); - assert(wp == w); - assert(np == n); - - w[0] = 0x1005; - r = c.out(m, w, w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::error); - assert(wp == w); - assert(np == n); - - w[0] = 0x453; - r = c.out(m, w, w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == w+1); - assert(np == n+2); - assert(n[0] == char(0xD1)); - assert(n[1] == char(0x93)); - - w[0] = 0x56; - r = c.out(m, w, w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == w+1); - assert(np == n+1); - assert(n[0] == char(0x56)); - } - { - typedef std::codecvt_utf8_utf16 C; - C c; - char32_t w[2] = {0xD8C0, 0xDC03}; - char n[7] = {0}; - const char32_t* wp = nullptr; - std::mbstate_t m; - char* np = nullptr; - std::codecvt_base::result r = c.out(m, w, w+2, wp, n, n+7, np); - assert(r == std::codecvt_base::ok); - assert(wp == w+2); - assert(np == n+7); - assert(n[0] == char(0xEF)); - assert(n[1] == char(0xBB)); - assert(n[2] == char(0xBF)); - assert(n[3] == char(0xF1)); - assert(n[4] == char(0x80)); - assert(n[5] == char(0x80)); - assert(n[6] == char(0x83)); - - w[0] = 0x1005; - r = c.out(m, w, w+1, wp, n, n+7, np); - assert(r == std::codecvt_base::ok); - assert(wp == w+1); - assert(np == n+6); - assert(n[0] == char(0xEF)); - assert(n[1] == char(0xBB)); - assert(n[2] == char(0xBF)); - assert(n[3] == char(0xE1)); - assert(n[4] == char(0x80)); - assert(n[5] == char(0x85)); - - w[0] = 0x453; - r = c.out(m, w, w+1, wp, n, n+7, np); - assert(r == std::codecvt_base::ok); - assert(wp == w+1); - assert(np == n+5); - assert(n[0] == char(0xEF)); - assert(n[1] == char(0xBB)); - assert(n[2] == char(0xBF)); - assert(n[3] == char(0xD1)); - assert(n[4] == char(0x93)); - - w[0] = 0x56; - r = c.out(m, w, w+1, wp, n, n+7, np); - assert(r == std::codecvt_base::ok); - assert(wp == w+1); - assert(np == n+4); - assert(n[0] == char(0xEF)); - assert(n[1] == char(0xBB)); - assert(n[2] == char(0xBF)); - assert(n[3] == char(0x56)); - } - - { - typedef std::codecvt_utf8_utf16 C; - C c; - char16_t w[2] = {0xD8C0, 0xDC03}; - char n[4] = {0}; - const char16_t* wp = nullptr; - std::mbstate_t m; - char* np = nullptr; - std::codecvt_base::result r = c.out(m, w, w+2, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == w+2); - assert(np == n+4); - assert(n[0] == char(0xF1)); - assert(n[1] == char(0x80)); - assert(n[2] == char(0x80)); - assert(n[3] == char(0x83)); - - w[0] = 0x1005; - r = c.out(m, w, w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == w+1); - assert(np == n+3); - assert(n[0] == char(0xE1)); - assert(n[1] == char(0x80)); - assert(n[2] == char(0x85)); - - w[0] = 0x453; - r = c.out(m, w, w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == w+1); - assert(np == n+2); - assert(n[0] == char(0xD1)); - assert(n[1] == char(0x93)); - - w[0] = 0x56; - r = c.out(m, w, w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == w+1); - assert(np == n+1); - assert(n[0] == char(0x56)); - } - { - typedef std::codecvt_utf8_utf16 C; - C c; - char16_t w[2] = {0xD8C0, 0xDC03}; - char n[4] = {0}; - const char16_t* wp = nullptr; - std::mbstate_t m; - char* np = nullptr; - std::codecvt_base::result r = c.out(m, w, w+2, wp, n, n+4, np); - assert(r == std::codecvt_base::error); - assert(wp == w); - assert(np == n); - - w[0] = 0x1005; - r = c.out(m, w, w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::error); - assert(wp == w); - assert(np == n); - - w[0] = 0x453; - r = c.out(m, w, w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == w+1); - assert(np == n+2); - assert(n[0] == char(0xD1)); - assert(n[1] == char(0x93)); - - w[0] = 0x56; - r = c.out(m, w, w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == w+1); - assert(np == n+1); - assert(n[0] == char(0x56)); - } - { - typedef std::codecvt_utf8_utf16 C; - C c; - char16_t w[2] = {0xD8C0, 0xDC03}; - char n[7] = {0}; - const char16_t* wp = nullptr; - std::mbstate_t m; - char* np = nullptr; - std::codecvt_base::result r = c.out(m, w, w+2, wp, n, n+7, np); - assert(r == std::codecvt_base::ok); - assert(wp == w+2); - assert(np == n+7); - assert(n[0] == char(0xEF)); - assert(n[1] == char(0xBB)); - assert(n[2] == char(0xBF)); - assert(n[3] == char(0xF1)); - assert(n[4] == char(0x80)); - assert(n[5] == char(0x80)); - assert(n[6] == char(0x83)); - - w[0] = 0x1005; - r = c.out(m, w, w+1, wp, n, n+7, np); - assert(r == std::codecvt_base::ok); - assert(wp == w+1); - assert(np == n+6); - assert(n[0] == char(0xEF)); - assert(n[1] == char(0xBB)); - assert(n[2] == char(0xBF)); - assert(n[3] == char(0xE1)); - assert(n[4] == char(0x80)); - assert(n[5] == char(0x85)); - - w[0] = 0x453; - r = c.out(m, w, w+1, wp, n, n+7, np); - assert(r == std::codecvt_base::ok); - assert(wp == w+1); - assert(np == n+5); - assert(n[0] == char(0xEF)); - assert(n[1] == char(0xBB)); - assert(n[2] == char(0xBF)); - assert(n[3] == char(0xD1)); - assert(n[4] == char(0x93)); - - w[0] = 0x56; - r = c.out(m, w, w+1, wp, n, n+7, np); - assert(r == std::codecvt_base::ok); - assert(wp == w+1); - assert(np == n+4); - assert(n[0] == char(0xEF)); - assert(n[1] == char(0xBB)); - assert(n[2] == char(0xBF)); - assert(n[3] == char(0x56)); - } + w[0] = 0x56; + r = c.out(m, w, w + 1, wp, n, n + 7, np); + assert(r == std::codecvt_base::ok); + assert(wp == w + 1); + assert(np == n + 4); + assert(n[0] == char(0xEF)); + assert(n[1] == char(0xBB)); + assert(n[2] == char(0xBF)); + assert(n[3] == char(0x56)); + } +} + +template +void TestHelper::test() { + { + typedef std::codecvt_utf8_utf16 C; + C c; + CharT w[2] = {0xD8C0, 0xDC03}; + char n[4] = {0}; + const CharT* wp = nullptr; + std::mbstate_t m; + char* np = nullptr; + std::codecvt_base::result r = c.out(m, w, w + 2, wp, n, n + 4, np); + assert(r == std::codecvt_base::ok); + assert(wp == w + 2); + assert(np == n + 4); + assert(n[0] == char(0xF1)); + assert(n[1] == char(0x80)); + assert(n[2] == char(0x80)); + assert(n[3] == char(0x83)); + + w[0] = 0x1005; + r = c.out(m, w, w + 1, wp, n, n + 4, np); + assert(r == std::codecvt_base::ok); + assert(wp == w + 1); + assert(np == n + 3); + assert(n[0] == char(0xE1)); + assert(n[1] == char(0x80)); + assert(n[2] == char(0x85)); + + w[0] = 0x453; + r = c.out(m, w, w + 1, wp, n, n + 4, np); + assert(r == std::codecvt_base::ok); + assert(wp == w + 1); + assert(np == n + 2); + assert(n[0] == char(0xD1)); + assert(n[1] == char(0x93)); + + w[0] = 0x56; + r = c.out(m, w, w + 1, wp, n, n + 4, np); + assert(r == std::codecvt_base::ok); + assert(wp == w + 1); + assert(np == n + 1); + assert(n[0] == char(0x56)); + } + { + typedef std::codecvt_utf8_utf16 C; + C c; + CharT w[2] = {0xD8C0, 0xDC03}; + char n[4] = {0}; + const CharT* wp = nullptr; + std::mbstate_t m; + char* np = nullptr; + std::codecvt_base::result r = c.out(m, w, w + 2, wp, n, n + 4, np); + assert(r == std::codecvt_base::error); + assert(wp == w); + assert(np == n); + + w[0] = 0x1005; + r = c.out(m, w, w + 1, wp, n, n + 4, np); + assert(r == std::codecvt_base::error); + assert(wp == w); + assert(np == n); + + w[0] = 0x453; + r = c.out(m, w, w + 1, wp, n, n + 4, np); + assert(r == std::codecvt_base::ok); + assert(wp == w + 1); + assert(np == n + 2); + assert(n[0] == char(0xD1)); + assert(n[1] == char(0x93)); + + w[0] = 0x56; + r = c.out(m, w, w + 1, wp, n, n + 4, np); + assert(r == std::codecvt_base::ok); + assert(wp == w + 1); + assert(np == n + 1); + assert(n[0] == char(0x56)); + } + { + typedef std::codecvt_utf8_utf16 C; + C c; + CharT w[2] = {0xD8C0, 0xDC03}; + char n[7] = {0}; + const CharT* wp = nullptr; + std::mbstate_t m; + char* np = nullptr; + std::codecvt_base::result r = c.out(m, w, w + 2, wp, n, n + 7, np); + assert(r == std::codecvt_base::ok); + assert(wp == w + 2); + assert(np == n + 7); + assert(n[0] == char(0xEF)); + assert(n[1] == char(0xBB)); + assert(n[2] == char(0xBF)); + assert(n[3] == char(0xF1)); + assert(n[4] == char(0x80)); + assert(n[5] == char(0x80)); + assert(n[6] == char(0x83)); + + w[0] = 0x1005; + r = c.out(m, w, w + 1, wp, n, n + 7, np); + assert(r == std::codecvt_base::ok); + assert(wp == w + 1); + assert(np == n + 6); + assert(n[0] == char(0xEF)); + assert(n[1] == char(0xBB)); + assert(n[2] == char(0xBF)); + assert(n[3] == char(0xE1)); + assert(n[4] == char(0x80)); + assert(n[5] == char(0x85)); + + w[0] = 0x453; + r = c.out(m, w, w + 1, wp, n, n + 7, np); + assert(r == std::codecvt_base::ok); + assert(wp == w + 1); + assert(np == n + 5); + assert(n[0] == char(0xEF)); + assert(n[1] == char(0xBB)); + assert(n[2] == char(0xBF)); + assert(n[3] == char(0xD1)); + assert(n[4] == char(0x93)); + + w[0] = 0x56; + r = c.out(m, w, w + 1, wp, n, n + 7, np); + assert(r == std::codecvt_base::ok); + assert(wp == w + 1); + assert(np == n + 4); + assert(n[0] == char(0xEF)); + assert(n[1] == char(0xBB)); + assert(n[2] == char(0xBF)); + assert(n[3] == char(0x56)); + } +} + +int main() { +#ifndef _WIN32 + TestHelper::test(); +#endif + TestHelper::test(); + TestHelper::test(); } diff --git a/test/std/localization/locales/locale.convenience/conversions/conversions.string/converted.pass.cpp b/test/std/localization/locales/locale.convenience/conversions/conversions.string/converted.pass.cpp index 06df185757d2..480628f707e2 100644 --- a/test/std/localization/locales/locale.convenience/conversions/conversions.string/converted.pass.cpp +++ b/test/std/localization/locales/locale.convenience/conversions/conversions.string/converted.pass.cpp @@ -17,9 +17,39 @@ #include #include -int main() -{ - typedef std::codecvt_utf8 Codecvt; +template +struct TestHelper; +template +struct TestHelper { + static void test(); +}; +template +struct TestHelper { + static void test(); +}; + +template +void TestHelper::test() { + static_assert((std::is_same::value), ""); + { + typedef std::codecvt_utf8 Codecvt; + typedef std::wstring_convert Myconv; + Myconv myconv; + assert(myconv.converted() == 0); + std::string bs = myconv.to_bytes(L"\x1005"); + assert(myconv.converted() == 1); + bs = myconv.to_bytes(L"\x1005\x65"); + assert(myconv.converted() == 2); + std::wstring ws = myconv.from_bytes("\xE1\x80\x85"); + assert(myconv.converted() == 3); + } +} + +template +void TestHelper::test() { + static_assert((std::is_same::value), ""); + { + typedef std::codecvt_utf8 Codecvt; typedef std::wstring_convert Myconv; Myconv myconv; assert(myconv.converted() == 0); @@ -29,4 +59,7 @@ int main() assert(myconv.converted() == 2); std::wstring ws = myconv.from_bytes("\xF1\x80\x80\x83"); assert(myconv.converted() == 4); + } } + +int main() { TestHelper::test(); } diff --git a/test/std/localization/locales/locale.convenience/conversions/conversions.string/from_bytes.pass.cpp b/test/std/localization/locales/locale.convenience/conversions/conversions.string/from_bytes.pass.cpp index 8705a5f3bced..2e627b739928 100644 --- a/test/std/localization/locales/locale.convenience/conversions/conversions.string/from_bytes.pass.cpp +++ b/test/std/localization/locales/locale.convenience/conversions/conversions.string/from_bytes.pass.cpp @@ -20,20 +20,53 @@ #include #include -int main() -{ - { - std::wstring_convert > myconv; - std::string bs("\xF1\x80\x80\x83"); - std::wstring ws = myconv.from_bytes('a'); - assert(ws == L"a"); - ws = myconv.from_bytes(bs.c_str()); - assert(ws == L"\x40003"); - ws = myconv.from_bytes(bs); - assert(ws == L"\x40003"); - ws = myconv.from_bytes(bs.data(), bs.data() + bs.size()); - assert(ws == L"\x40003"); - ws = myconv.from_bytes(""); - assert(ws.size() == 0); - } +template +struct TestHelper; +template +struct TestHelper { + static void test(); +}; +template +struct TestHelper { + static void test(); +}; + +template +void TestHelper::test() { + static_assert((std::is_same::value), ""); + { + std::wstring_convert > myconv; + std::string bs("\xE1\x80\x85\x00"); + std::wstring ws = myconv.from_bytes('a'); + assert(ws == L"a"); + ws = myconv.from_bytes(bs.c_str()); + assert(ws == L"\x1005"); + ws = myconv.from_bytes(bs); + assert(ws == L"\x1005"); + ws = myconv.from_bytes(bs.data(), bs.data() + bs.size()); + assert(ws == L"\x1005"); + ws = myconv.from_bytes(""); + assert(ws.size() == 0); + } } + +template +void TestHelper::test() { + static_assert((std::is_same::value), ""); + { + std::wstring_convert > myconv; + std::string bs("\xF1\x80\x80\x83"); + std::wstring ws = myconv.from_bytes('a'); + assert(ws == L"a"); + ws = myconv.from_bytes(bs.c_str()); + assert(ws == L"\x40003"); + ws = myconv.from_bytes(bs); + assert(ws == L"\x40003"); + ws = myconv.from_bytes(bs.data(), bs.data() + bs.size()); + assert(ws == L"\x40003"); + ws = myconv.from_bytes(""); + assert(ws.size() == 0); + } +} + +int main() { TestHelper::test(); } diff --git a/test/std/localization/locales/locale.convenience/conversions/conversions.string/to_bytes.pass.cpp b/test/std/localization/locales/locale.convenience/conversions/conversions.string/to_bytes.pass.cpp index 7253a18a70b1..0a6cab73bf9e 100644 --- a/test/std/localization/locales/locale.convenience/conversions/conversions.string/to_bytes.pass.cpp +++ b/test/std/localization/locales/locale.convenience/conversions/conversions.string/to_bytes.pass.cpp @@ -20,20 +20,53 @@ #include #include -int main() -{ - { - std::wstring_convert > myconv; - std::wstring ws(1, L'\x40003'); - std::string bs = myconv.to_bytes(ws[0]); - assert(bs == "\xF1\x80\x80\x83"); - bs = myconv.to_bytes(ws.c_str()); - assert(bs == "\xF1\x80\x80\x83"); - bs = myconv.to_bytes(ws); - assert(bs == "\xF1\x80\x80\x83"); - bs = myconv.to_bytes(ws.data(), ws.data() + ws.size()); - assert(bs == "\xF1\x80\x80\x83"); - bs = myconv.to_bytes(L""); - assert(bs.size() == 0); - } +template +struct TestHelper; +template +struct TestHelper { + static void test(); +}; +template +struct TestHelper { + static void test(); +}; + +template +void TestHelper::test() { + static_assert((std::is_same::value), ""); + { + std::wstring_convert > myconv; + std::wstring ws(1, CharT(0x1005)); + std::string bs = myconv.to_bytes(ws[0]); + assert(bs == "\xE1\x80\x85\x00"); + bs = myconv.to_bytes(ws.c_str()); + assert(bs == "\xE1\x80\x85\x00"); + bs = myconv.to_bytes(ws); + assert(bs == "\xE1\x80\x85\x00"); + bs = myconv.to_bytes(ws.data(), ws.data() + ws.size()); + assert(bs == "\xE1\x80\x85\x00"); + bs = myconv.to_bytes(L""); + assert(bs.size() == 0); + } } + +template +void TestHelper::test() { + static_assert((std::is_same::value), ""); + { + std::wstring_convert > myconv; + std::wstring ws(1, CharT(0x40003)); + std::string bs = myconv.to_bytes(ws[0]); + assert(bs == "\xF1\x80\x80\x83"); + bs = myconv.to_bytes(ws.c_str()); + assert(bs == "\xF1\x80\x80\x83"); + bs = myconv.to_bytes(ws); + assert(bs == "\xF1\x80\x80\x83"); + bs = myconv.to_bytes(ws.data(), ws.data() + ws.size()); + assert(bs == "\xF1\x80\x80\x83"); + bs = myconv.to_bytes(L""); + assert(bs.size() == 0); + } +} + +int main() { TestHelper::test(); } diff --git a/test/std/numerics/numeric.ops/numeric.ops.gcd/gcd.pass.cpp b/test/std/numerics/numeric.ops/numeric.ops.gcd/gcd.pass.cpp index 961b515ef8d8..517a62a0068f 100644 --- a/test/std/numerics/numeric.ops/numeric.ops.gcd/gcd.pass.cpp +++ b/test/std/numerics/numeric.ops/numeric.ops.gcd/gcd.pass.cpp @@ -16,8 +16,10 @@ #include #include +#include +#include #include // for rand() -#include +#include constexpr struct { int x; @@ -36,21 +38,24 @@ constexpr struct { template -constexpr bool test0(Input1 in1, Input2 in2, Output out) +constexpr bool test0(int in1, int in2, int out) { - static_assert((std::is_same::value), "" ); - static_assert((std::is_same::value), "" ); - return out == std::gcd(in1, in2) ? true : (std::abort(), false); + auto value1 = static_cast(in1); + auto value2 = static_cast(in2); + static_assert(std::is_same_v, ""); + static_assert(std::is_same_v, ""); + assert(static_cast(out) == std::gcd(value1, value2)); + return true; } template constexpr bool do_test(int = 0) { - using S1 = typename std::make_signed::type; - using S2 = typename std::make_signed::type; - using U1 = typename std::make_unsigned::type; - using U2 = typename std::make_unsigned::type; + using S1 = std::make_signed_t; + using S2 = std::make_signed_t; + using U1 = std::make_unsigned_t; + using U2 = std::make_unsigned_t; bool accumulate = true; for (auto TC : Cases) { { // Test with two signed types @@ -103,15 +108,15 @@ int main() assert(do_test(non_cce)); assert(do_test(non_cce)); - static_assert(do_test< int8_t>(), ""); - static_assert(do_test(), ""); - static_assert(do_test(), ""); - static_assert(do_test(), ""); + static_assert(do_test(), ""); + static_assert(do_test(), ""); + static_assert(do_test(), ""); + static_assert(do_test(), ""); - assert(do_test< int8_t>(non_cce)); - assert(do_test(non_cce)); - assert(do_test(non_cce)); - assert(do_test(non_cce)); + assert(do_test(non_cce)); + assert(do_test(non_cce)); + assert(do_test(non_cce)); + assert(do_test(non_cce)); static_assert(do_test(), ""); static_assert(do_test(), ""); @@ -133,8 +138,8 @@ int main() // LWG#2837 { - auto res = std::gcd((int64_t)1234, (int32_t)-2147483648); - static_assert( std::is_same::type>::value, ""); - assert(res == 2); + auto res = std::gcd(static_cast(1234), INT32_MIN); + static_assert(std::is_same_v, ""); + assert(res == 2); } } diff --git a/test/std/numerics/numeric.ops/numeric.ops.lcm/lcm.pass.cpp b/test/std/numerics/numeric.ops/numeric.ops.lcm/lcm.pass.cpp index 90d48398f54a..6bd8a4f1e610 100644 --- a/test/std/numerics/numeric.ops/numeric.ops.lcm/lcm.pass.cpp +++ b/test/std/numerics/numeric.ops/numeric.ops.lcm/lcm.pass.cpp @@ -11,12 +11,14 @@ // // template -// constexpr common_type_t<_M,_N> gcd(_M __m, _N __n) +// constexpr common_type_t<_M,_N> lcm(_M __m, _N __n) #include #include +#include +#include #include -#include +#include constexpr struct { int x; @@ -34,21 +36,24 @@ constexpr struct { }; template -constexpr bool test0(Input1 in1, Input2 in2, Output out) +constexpr bool test0(int in1, int in2, int out) { - static_assert((std::is_same::value), "" ); - static_assert((std::is_same::value), "" ); - return out == std::lcm(in1, in2) ? true : (std::abort(), false); + auto value1 = static_cast(in1); + auto value2 = static_cast(in2); + static_assert(std::is_same_v, ""); + static_assert(std::is_same_v, ""); + assert(static_cast(out) == std::lcm(value1, value2)); + return true; } template constexpr bool do_test(int = 0) { - using S1 = typename std::make_signed::type; - using S2 = typename std::make_signed::type; - using U1 = typename std::make_unsigned::type; - using U2 = typename std::make_unsigned::type; + using S1 = std::make_signed_t; + using S2 = std::make_signed_t; + using U1 = std::make_unsigned_t; + using U2 = std::make_unsigned_t; bool accumulate = true; for (auto TC : Cases) { { // Test with two signed types @@ -101,15 +106,15 @@ int main() assert(do_test(non_cce)); assert(do_test(non_cce)); - static_assert(do_test< int8_t>(), ""); - static_assert(do_test(), ""); - static_assert(do_test(), ""); - static_assert(do_test(), ""); + static_assert(do_test(), ""); + static_assert(do_test(), ""); + static_assert(do_test(), ""); + static_assert(do_test(), ""); - assert(do_test< int8_t>(non_cce)); - assert(do_test(non_cce)); - assert(do_test(non_cce)); - assert(do_test(non_cce)); + assert(do_test(non_cce)); + assert(do_test(non_cce)); + assert(do_test(non_cce)); + assert(do_test(non_cce)); static_assert(do_test(), ""); static_assert(do_test(), ""); @@ -131,9 +136,9 @@ int main() // LWG#2837 { - auto res1 = std::lcm((int64_t)1234, (int32_t)-2147483648); - (void) std::lcm(INT_MIN, 2); // this used to trigger UBSAN - static_assert( std::is_same::type>::value, ""); - assert(res1 == 1324997410816LL); + auto res1 = std::lcm(static_cast(1234), INT32_MIN); + (void)std::lcm(INT_MIN, 2UL); // this used to trigger UBSAN + static_assert(std::is_same_v, ""); + assert(res1 == 1324997410816LL); } } diff --git a/test/std/re/re.traits/lookup_classname.pass.cpp b/test/std/re/re.traits/lookup_classname.pass.cpp index 2215b9043d5b..b61f772b2df8 100644 --- a/test/std/re/re.traits/lookup_classname.pass.cpp +++ b/test/std/re/re.traits/lookup_classname.pass.cpp @@ -27,39 +27,61 @@ test(const char_type* A, typename std::regex_traits::char_class_type expected, bool icase = false) { + typedef typename std::regex_traits::char_class_type char_class_type; std::regex_traits t; typedef forward_iterator F; - assert(t.lookup_classname(F(A), F(A + t.length(A)), icase) == expected); + char_class_type result = t.lookup_classname(F(A), F(A + t.length(A)), icase); + assert(result == expected); +} + +template +void +test_w(const char_type* A, + typename std::regex_traits::char_class_type expected, + bool icase = false) +{ + typedef typename std::regex_traits::char_class_type char_class_type; + std::regex_traits t; + typedef forward_iterator F; + char_class_type result = t.lookup_classname(F(A), F(A + t.length(A)), icase); + assert((result & expected) == expected); + LIBCPP_ASSERT((expected | std::regex_traits::__regex_word) == result); + + const bool matches_underscore = t.isctype('_', result); + if (result != expected) + assert(matches_underscore && "expected to match underscore"); + else + assert(!matches_underscore && "should not match underscore"); } int main() { // if __regex_word is not distinct from all the classes, bad things happen // See https://bugs.llvm.org/show_bug.cgi?id=26476 for an example. - assert((std::ctype_base::space & std::regex_traits::__regex_word) == 0); - assert((std::ctype_base::print & std::regex_traits::__regex_word) == 0); - assert((std::ctype_base::cntrl & std::regex_traits::__regex_word) == 0); - assert((std::ctype_base::upper & std::regex_traits::__regex_word) == 0); - assert((std::ctype_base::lower & std::regex_traits::__regex_word) == 0); - assert((std::ctype_base::alpha & std::regex_traits::__regex_word) == 0); - assert((std::ctype_base::digit & std::regex_traits::__regex_word) == 0); - assert((std::ctype_base::punct & std::regex_traits::__regex_word) == 0); - assert((std::ctype_base::xdigit & std::regex_traits::__regex_word) == 0); - assert((std::ctype_base::blank & std::regex_traits::__regex_word) == 0); + LIBCPP_ASSERT((std::ctype_base::space & std::regex_traits::__regex_word) == 0); + LIBCPP_ASSERT((std::ctype_base::print & std::regex_traits::__regex_word) == 0); + LIBCPP_ASSERT((std::ctype_base::cntrl & std::regex_traits::__regex_word) == 0); + LIBCPP_ASSERT((std::ctype_base::upper & std::regex_traits::__regex_word) == 0); + LIBCPP_ASSERT((std::ctype_base::lower & std::regex_traits::__regex_word) == 0); + LIBCPP_ASSERT((std::ctype_base::alpha & std::regex_traits::__regex_word) == 0); + LIBCPP_ASSERT((std::ctype_base::digit & std::regex_traits::__regex_word) == 0); + LIBCPP_ASSERT((std::ctype_base::punct & std::regex_traits::__regex_word) == 0); + LIBCPP_ASSERT((std::ctype_base::xdigit & std::regex_traits::__regex_word) == 0); + LIBCPP_ASSERT((std::ctype_base::blank & std::regex_traits::__regex_word) == 0); test("d", std::ctype_base::digit); test("D", std::ctype_base::digit); test("d", std::ctype_base::digit, true); test("D", std::ctype_base::digit, true); - test("w", std::regex_traits::__regex_word | std::ctype_base::alnum - | std::ctype_base::upper | std::ctype_base::lower); - test("W", std::regex_traits::__regex_word | std::ctype_base::alnum - | std::ctype_base::upper | std::ctype_base::lower); - test("w", std::regex_traits::__regex_word | std::ctype_base::alnum - | std::ctype_base::upper | std::ctype_base::lower, true); - test("W", std::regex_traits::__regex_word | std::ctype_base::alnum - | std::ctype_base::upper | std::ctype_base::lower, true); + test_w("w", std::ctype_base::alnum + | std::ctype_base::upper | std::ctype_base::lower); + test_w("W", std::ctype_base::alnum + | std::ctype_base::upper | std::ctype_base::lower); + test_w("w", std::ctype_base::alnum + | std::ctype_base::upper | std::ctype_base::lower, true); + test_w("W", std::ctype_base::alnum + | std::ctype_base::upper | std::ctype_base::lower, true); test("s", std::ctype_base::space); test("S", std::ctype_base::space); @@ -140,13 +162,13 @@ int main() test(L"d", std::ctype_base::digit, true); test(L"D", std::ctype_base::digit, true); - test(L"w", std::regex_traits::__regex_word | std::ctype_base::alnum + test_w(L"w", std::ctype_base::alnum | std::ctype_base::upper | std::ctype_base::lower); - test(L"W", std::regex_traits::__regex_word | std::ctype_base::alnum + test_w(L"W", std::ctype_base::alnum | std::ctype_base::upper | std::ctype_base::lower); - test(L"w", std::regex_traits::__regex_word | std::ctype_base::alnum + test_w(L"w", std::ctype_base::alnum | std::ctype_base::upper | std::ctype_base::lower, true); - test(L"W", std::regex_traits::__regex_word | std::ctype_base::alnum + test_w(L"W", std::ctype_base::alnum | std::ctype_base::upper | std::ctype_base::lower, true); test(L"s", std::ctype_base::space); diff --git a/test/std/strings/basic.string/string.cons/T_size_size.pass.cpp b/test/std/strings/basic.string/string.cons/T_size_size.pass.cpp index c8d14a1d0ddf..67ac43494a6e 100644 --- a/test/std/strings/basic.string/string.cons/T_size_size.pass.cpp +++ b/test/std/strings/basic.string/string.cons/T_size_size.pass.cpp @@ -27,16 +27,17 @@ template void -test(SV sv, unsigned pos, unsigned n) +test(SV sv, std::size_t pos, std::size_t n) { typedef typename S::traits_type T; typedef typename S::allocator_type A; + typedef typename S::size_type Size; if (pos <= sv.size()) { - S s2(sv, pos, n); + S s2(sv, static_cast(pos), static_cast(n)); LIBCPP_ASSERT(s2.__invariants()); assert(pos <= sv.size()); - unsigned rlen = std::min(sv.size() - pos, n); + std::size_t rlen = std::min(sv.size() - pos, n); assert(s2.size() == rlen); assert(T::compare(s2.data(), sv.data() + pos, rlen) == 0); assert(s2.get_allocator() == A()); @@ -47,7 +48,7 @@ test(SV sv, unsigned pos, unsigned n) { try { - S s2(sv, pos, n); + S s2(sv, static_cast(pos), static_cast(n)); assert(false); } catch (std::out_of_range&) @@ -60,15 +61,16 @@ test(SV sv, unsigned pos, unsigned n) template void -test(SV sv, unsigned pos, unsigned n, const typename S::allocator_type& a) +test(SV sv, std::size_t pos, std::size_t n, const typename S::allocator_type& a) { typedef typename S::traits_type T; + typedef typename S::size_type Size; if (pos <= sv.size()) { - S s2(sv, pos, n, a); + S s2(sv, static_cast(pos), static_cast(n), a); LIBCPP_ASSERT(s2.__invariants()); assert(pos <= sv.size()); - unsigned rlen = std::min(sv.size() - pos, n); + std::size_t rlen = std::min(sv.size() - pos, n); assert(s2.size() == rlen); assert(T::compare(s2.data(), sv.data() + pos, rlen) == 0); assert(s2.get_allocator() == a); @@ -79,7 +81,7 @@ test(SV sv, unsigned pos, unsigned n, const typename S::allocator_type& a) { try { - S s2(sv, pos, n, a); + S s2(sv, static_cast(pos), static_cast(n), a); assert(false); } catch (std::out_of_range&) diff --git a/test/std/strings/string.view/string.view.modifiers/clear.pass.cpp b/test/std/strings/string.view/string.view.modifiers/clear.pass.cpp deleted file mode 100644 index c1137e80acaf..000000000000 --- a/test/std/strings/string.view/string.view.modifiers/clear.pass.cpp +++ /dev/null @@ -1,67 +0,0 @@ -//===----------------------------------------------------------------------===// -// -// The LLVM Compiler Infrastructure -// -// This file is dual licensed under the MIT and the University of Illinois Open -// Source Licenses. See LICENSE.TXT for details. -// -//===----------------------------------------------------------------------===// - - -// - -// void clear() noexcept - -#include -#include - -#include "test_macros.h" - -template -void test ( const CharT *s, size_t len ) { - typedef std::basic_string_view SV; - { - SV sv1 ( s ); - assert ( sv1.size() == len ); - assert ( sv1.data() == s ); - - sv1.clear (); - assert ( sv1.data() == nullptr ); - assert ( sv1.size() == 0 ); - assert ( sv1 == SV()); - } -} - -#if TEST_STD_VER > 11 -constexpr size_t test_ce ( size_t n ) { - typedef std::basic_string_view SV; - SV sv1{ "ABCDEFGHIJKL", n }; - sv1.clear(); - return sv1.size(); -} -#endif - -int main () { - test ( "ABCDE", 5 ); - test ( "a", 1 ); - test ( "", 0 ); - - test ( L"ABCDE", 5 ); - test ( L"a", 1 ); - test ( L"", 0 ); - -#if TEST_STD_VER >= 11 - test ( u"ABCDE", 5 ); - test ( u"a", 1 ); - test ( u"", 0 ); - - test ( U"ABCDE", 5 ); - test ( U"a", 1 ); - test ( U"", 0 ); -#endif - -#if TEST_STD_VER > 11 - static_assert ( test_ce (5) == 0, "" ); -#endif - -} diff --git a/test/std/thread/thread.threads/thread.thread.this/sleep_for_tested_elsewhere.pass.cpp b/test/std/thread/thread.threads/thread.thread.this/sleep_for_tested_elsewhere.pass.cpp new file mode 100644 index 000000000000..3406fff70610 --- /dev/null +++ b/test/std/thread/thread.threads/thread.thread.this/sleep_for_tested_elsewhere.pass.cpp @@ -0,0 +1,22 @@ +// -*- C++ -*- +//===----------------------------------------------------------------------===// +// +// The LLVM Compiler Infrastructure +// +// This file is dual licensed under the MIT and the University of Illinois Open +// Source Licenses. See LICENSE.TXT for details. +// +//===----------------------------------------------------------------------===// + +// + +// template +// void sleep_for(const chrono::duration& rel_time); + +// The std::this_thread::sleep_for test requires POSIX specific headers and +// is therefore non-standard. For this reason the test lives under the 'libcxx' +// subdirectory. + +int main() +{ +} diff --git a/test/std/utilities/allocator.adaptor/allocator.adaptor.cnstr/allocs.pass.cpp b/test/std/utilities/allocator.adaptor/allocator.adaptor.cnstr/allocs.pass.cpp index 0aef2af85ccf..2aa19c6188a5 100644 --- a/test/std/utilities/allocator.adaptor/allocator.adaptor.cnstr/allocs.pass.cpp +++ b/test/std/utilities/allocator.adaptor/allocator.adaptor.cnstr/allocs.pass.cpp @@ -107,5 +107,11 @@ int main() assert((a.inner_allocator() == std::scoped_allocator_adaptor, A3>(A2(5), A3(6)))); } - +// Test for LWG2782 + { + static_assert(!std::is_convertible, A2>::value, ""); + static_assert(!std::is_convertible< + std::scoped_allocator_adaptor>, + std::scoped_allocator_adaptor>>::value, ""); + } } diff --git a/test/std/utilities/function.objects/func.not_fn/not_fn.pass.cpp b/test/std/utilities/function.objects/func.not_fn/not_fn.pass.cpp index 2b19211be9bf..aa8eb3916db9 100644 --- a/test/std/utilities/function.objects/func.not_fn/not_fn.pass.cpp +++ b/test/std/utilities/function.objects/func.not_fn/not_fn.pass.cpp @@ -416,7 +416,7 @@ void throws_in_constructor_test() ThrowsOnCopy() = default; bool operator()() const { assert(false); -#if defined(_LIBCPP_MSVC) +#if defined(TEST_COMPILER_C1XX) __assume(0); #else __builtin_unreachable(); diff --git a/test/std/utilities/function.objects/func.wrap/func.wrap.func/func.wrap.func.con/copy_move.pass.cpp b/test/std/utilities/function.objects/func.wrap/func.wrap.func/func.wrap.func.con/copy_move.pass.cpp index 9d5681a3db76..7516b2e3af2b 100644 --- a/test/std/utilities/function.objects/func.wrap/func.wrap.func/func.wrap.func.con/copy_move.pass.cpp +++ b/test/std/utilities/function.objects/func.wrap/func.wrap.func/func.wrap.func.con/copy_move.pass.cpp @@ -11,7 +11,8 @@ // class function -// function(const function& f); +// function(const function& f); +// function(const function&& f); #include #include diff --git a/test/std/utilities/function.objects/func.wrap/func.wrap.func/func.wrap.func.targ/target.pass.cpp b/test/std/utilities/function.objects/func.wrap/func.wrap.func/func.wrap.func.targ/target.pass.cpp index 53476a274735..7a4678ad1cae 100644 --- a/test/std/utilities/function.objects/func.wrap/func.wrap.func/func.wrap.func.targ/target.pass.cpp +++ b/test/std/utilities/function.objects/func.wrap/func.wrap.func/func.wrap.func.targ/target.pass.cpp @@ -63,6 +63,7 @@ int main() assert(A::count == 1); assert(f.target()); assert(f.target() == 0); + assert(f.target() == nullptr); } assert(A::count == 0); { @@ -70,6 +71,7 @@ int main() assert(A::count == 0); assert(f.target()); assert(f.target() == 0); + assert(f.target() == nullptr); } assert(A::count == 0); { @@ -77,6 +79,7 @@ int main() assert(A::count == 1); assert(f.target()); assert(f.target() == 0); + assert(f.target() == nullptr); } assert(A::count == 0); { @@ -84,6 +87,7 @@ int main() assert(A::count == 0); assert(f.target()); assert(f.target() == 0); + assert(f.target() == nullptr); } assert(A::count == 0); } diff --git a/test/std/utilities/memory/util.smartptr/util.smartptr.enab/enable_shared_from_this.pass.cpp b/test/std/utilities/memory/util.smartptr/util.smartptr.enab/enable_shared_from_this.pass.cpp index 5a0d9259c11d..eb0d0a955fc5 100644 --- a/test/std/utilities/memory/util.smartptr/util.smartptr.enab/enable_shared_from_this.pass.cpp +++ b/test/std/utilities/memory/util.smartptr/util.smartptr.enab/enable_shared_from_this.pass.cpp @@ -49,6 +49,10 @@ struct Bar : public Foo { }; +struct PrivateBase : private std::enable_shared_from_this { +}; + + int main() { { // https://bugs.llvm.org/show_bug.cgi?id=18843 @@ -74,6 +78,12 @@ int main() assert(p == q); assert(!p.owner_before(q) && !q.owner_before(p)); // p and q share ownership } + { + typedef std::shared_ptr APtr; + typedef std::weak_ptr WeakAPtr; + APtr a1 = std::make_shared(); + assert(a1.use_count() == 1); + } // Test LWG issue 2529. Only reset '__weak_ptr_' when it's already expired. // http://cplusplus.github.io/LWG/lwg-active.html#2529. // Test two different ways: diff --git a/test/std/utilities/meta/meta.rel/is_convertible.pass.cpp b/test/std/utilities/meta/meta.rel/is_convertible.pass.cpp index 552c16075da7..20c9eca8e521 100644 --- a/test/std/utilities/meta/meta.rel/is_convertible.pass.cpp +++ b/test/std/utilities/meta/meta.rel/is_convertible.pass.cpp @@ -120,7 +120,9 @@ int main() static_assert((!std::is_convertible::value), ""); static_assert((!std::is_convertible::value), ""); static_assert((!std::is_convertible::value), ""); - static_assert((!std::is_convertible::value), ""); +#if TEST_STD_VER >= 11 + static_assert((!std::is_convertible::value), ""); +#endif static_assert((!std::is_convertible::value), ""); static_assert((!std::is_convertible::value), ""); static_assert((!std::is_convertible::value), ""); diff --git a/test/std/utilities/meta/meta.unary/meta.unary.prop/is_trivially_copyable.pass.cpp b/test/std/utilities/meta/meta.unary/meta.unary.prop/is_trivially_copyable.pass.cpp index 42ecdb3b896a..0bb373c96620 100644 --- a/test/std/utilities/meta/meta.unary/meta.unary.prop/is_trivially_copyable.pass.cpp +++ b/test/std/utilities/meta/meta.unary/meta.unary.prop/is_trivially_copyable.pass.cpp @@ -11,7 +11,9 @@ // is_trivially_copyable -// XFAIL: gcc-4.9 +// These compilers have not implemented Core 2094 which makes volatile +// qualified types trivially copyable. +// XFAIL: clang-3, clang-4, apple-clang, gcc #include #include @@ -22,13 +24,13 @@ void test_is_trivially_copyable() { static_assert( std::is_trivially_copyable::value, ""); static_assert( std::is_trivially_copyable::value, ""); - static_assert(!std::is_trivially_copyable::value, ""); - static_assert(!std::is_trivially_copyable::value, ""); + static_assert( std::is_trivially_copyable::value, ""); + static_assert( std::is_trivially_copyable::value, ""); #if TEST_STD_VER > 14 static_assert( std::is_trivially_copyable_v, ""); static_assert( std::is_trivially_copyable_v, ""); - static_assert(!std::is_trivially_copyable_v, ""); - static_assert(!std::is_trivially_copyable_v, ""); + static_assert( std::is_trivially_copyable_v, ""); + static_assert( std::is_trivially_copyable_v, ""); #endif } diff --git a/test/std/utilities/template.bitset/includes.pass.cpp b/test/std/utilities/template.bitset/includes.pass.cpp index 2e3c2812e441..e640a1b5b7ed 100644 --- a/test/std/utilities/template.bitset/includes.pass.cpp +++ b/test/std/utilities/template.bitset/includes.pass.cpp @@ -11,22 +11,27 @@ #include -#ifndef _LIBCPP_CSTDDEF -#error has not been included -#endif - -#ifndef _LIBCPP_STRING -#error has not been included -#endif - -#ifndef _LIBCPP_STDEXCEPT -#error has not been included -#endif - -#ifndef _LIBCPP_IOSFWD -#error has not been included -#endif +template void test_typedef() {} int main() { + { // test for + std::ptrdiff_t p; ((void)p); + std::size_t s; ((void)s); + std::nullptr_t np; ((void)np); + } + { // test for + std::string s; ((void)s); + } + { // test for + std::logic_error le("blah"); ((void)le); + std::runtime_error re("blah"); ((void)re); + } + { // test for + test_typedef(); + test_typedef(); + test_typedef(); + test_typedef(); + test_typedef(); + } } diff --git a/test/std/utilities/tuple/tuple.tuple/tuple.cnstr/dtor.pass.cpp b/test/std/utilities/tuple/tuple.tuple/tuple.cnstr/dtor.pass.cpp index fbcda44e4065..d7b184f6383c 100644 --- a/test/std/utilities/tuple/tuple.tuple/tuple.cnstr/dtor.pass.cpp +++ b/test/std/utilities/tuple/tuple.tuple/tuple.cnstr/dtor.pass.cpp @@ -15,6 +15,10 @@ // ~tuple(); +// C++17 added: +// The destructor of tuple shall be a trivial destructor +// if (is_trivially_destructible_v && ...) is true. + #include #include #include diff --git a/test/std/utilities/utility/pairs/pairs.pair/dtor.pass.cpp b/test/std/utilities/utility/pairs/pairs.pair/dtor.pass.cpp index 2d87e7ababab..83c55e75b4d2 100644 --- a/test/std/utilities/utility/pairs/pairs.pair/dtor.pass.cpp +++ b/test/std/utilities/utility/pairs/pairs.pair/dtor.pass.cpp @@ -15,6 +15,10 @@ // ~pair() +// C++17 added: +// The destructor of pair shall be a trivial destructor +// if (is_trivially_destructible_v && is_trivially_destructible_v) is true. + #include #include diff --git a/test/std/utilities/variant/variant.visit/visit.pass.cpp b/test/std/utilities/variant/variant.visit/visit.pass.cpp index 05b58c16f3d8..316f2d22b01d 100644 --- a/test/std/utilities/variant/variant.visit/visit.pass.cpp +++ b/test/std/utilities/variant/variant.visit/visit.pass.cpp @@ -94,6 +94,16 @@ void test_call_operator_forwarding() { using Fn = ForwardingCallObject; Fn obj{}; const Fn &cobj = obj; + { // test call operator forwarding - no variant + std::visit(obj); + assert(Fn::check_call<>(CT_NonConst | CT_LValue)); + std::visit(cobj); + assert(Fn::check_call<>(CT_Const | CT_LValue)); + std::visit(std::move(obj)); + assert(Fn::check_call<>(CT_NonConst | CT_RValue)); + std::visit(std::move(cobj)); + assert(Fn::check_call<>(CT_Const | CT_RValue)); + } { // test call operator forwarding - single variant, single arg using V = std::variant; V v(42); diff --git a/test/support/archetypes.hpp b/test/support/archetypes.hpp index f442b592a26f..533f5869b5ab 100644 --- a/test/support/archetypes.hpp +++ b/test/support/archetypes.hpp @@ -5,6 +5,7 @@ #include #include "test_macros.h" +#include "test_workarounds.h" #if TEST_STD_VER >= 11 @@ -14,7 +15,9 @@ template struct DepType : T {}; struct NullBase { +#ifndef TEST_WORKAROUND_C1XX_BROKEN_ZA_CTOR_CHECK protected: +#endif // !TEST_WORKAROUND_C1XX_BROKEN_ZA_CTOR_CHECK NullBase() = default; NullBase(NullBase const&) = default; NullBase& operator=(NullBase const&) = default; @@ -81,7 +84,9 @@ struct TestBase { ++assigned; ++value_assigned; return *this; } +#ifndef TEST_WORKAROUND_C1XX_BROKEN_ZA_CTOR_CHECK protected: +#endif // !TEST_WORKAROUND_C1XX_BROKEN_ZA_CTOR_CHECK ~TestBase() { assert(value != -999); assert(alive > 0); --alive; ++destroyed; value = -999; @@ -144,7 +149,9 @@ struct ValueBase { } //~ValueBase() { assert(value != -999); value = -999; } int value; +#ifndef TEST_WORKAROUND_C1XX_BROKEN_ZA_CTOR_CHECK protected: +#endif // !TEST_WORKAROUND_C1XX_BROKEN_ZA_CTOR_CHECK constexpr static int check_value(int const& val) { #if TEST_STD_VER < 14 return val == -1 || val == 999 ? (TEST_THROW(42), 0) : val; @@ -197,7 +204,9 @@ struct TrivialValueBase { template ::type = true> constexpr TrivialValueBase(std::initializer_list& il, int = 0) : value(static_cast(il.size())) {} int value; +#ifndef TEST_WORKAROUND_C1XX_BROKEN_ZA_CTOR_CHECK protected: +#endif // !TEST_WORKAROUND_C1XX_BROKEN_ZA_CTOR_CHECK constexpr TrivialValueBase() noexcept : value(0) {} }; diff --git a/test/support/archetypes.ipp b/test/support/archetypes.ipp index d8d1e5a9330b..36045017907f 100644 --- a/test/support/archetypes.ipp +++ b/test/support/archetypes.ipp @@ -6,7 +6,11 @@ #define DEFINE_EXPLICIT #endif #ifndef DEFINE_CONSTEXPR +#ifdef TEST_WORKAROUND_EDG_EXPLICIT_CONSTEXPR +#define DEFINE_CONSTEXPR +#else // TEST_WORKAROUND_EDG_EXPLICIT_CONSTEXPR #define DEFINE_CONSTEXPR constexpr +#endif // TEST_WORKAROUND_EDG_EXPLICIT_CONSTEXPR #endif #ifndef DEFINE_ASSIGN_CONSTEXPR #if TEST_STD_VER >= 14 diff --git a/test/support/filesystem_dynamic_test_helper.py b/test/support/filesystem_dynamic_test_helper.py index d2b2810d0992..081e678b6e84 100644 --- a/test/support/filesystem_dynamic_test_helper.py +++ b/test/support/filesystem_dynamic_test_helper.py @@ -25,7 +25,7 @@ def sanitize(p): """ Some of the tests restrict permissions to induce failures. -Before we delete the test enviroment, we have to walk it and re-raise the +Before we delete the test environment, we have to walk it and re-raise the permissions. """ def clean_recursive(root_p): diff --git a/test/support/filesystem_test_helper.hpp b/test/support/filesystem_test_helper.hpp index 6cc0e370a786..755be90351ff 100644 --- a/test/support/filesystem_test_helper.hpp +++ b/test/support/filesystem_test_helper.hpp @@ -227,7 +227,7 @@ struct scoped_test_env } static inline void fs_helper_run(std::string const& raw_cmd) { - // check that the fs test root in the enviroment matches what we were + // check that the fs test root in the environment matches what we were // compiled with. static bool checked = checkDynamicTestRoot(); ((void)checked); @@ -246,7 +246,7 @@ struct scoped_test_env std::abort(); } if (std::string(fs_root) != LIBCXX_FILESYSTEM_DYNAMIC_TEST_ROOT) { - std::printf("ERROR: LIBCXX_FILESYSTEM_DYNAMIC_TEST_ROOT enviroment variable" + std::printf("ERROR: LIBCXX_FILESYSTEM_DYNAMIC_TEST_ROOT environment variable" " must have the same value as when the test was compiled.\n"); std::printf(" Current Value: '%s'\n", fs_root); std::printf(" Expected Value: '%s'\n", LIBCXX_FILESYSTEM_DYNAMIC_TEST_ROOT); diff --git a/test/support/msvc_stdlib_force_include.hpp b/test/support/msvc_stdlib_force_include.hpp index 2b78e507a327..6bcc97952199 100644 --- a/test/support/msvc_stdlib_force_include.hpp +++ b/test/support/msvc_stdlib_force_include.hpp @@ -13,11 +13,13 @@ // This header is force-included when running the libc++ tests against the // MSVC standard library. -// Silence warnings about CRT machinery. -#define _CRT_SECURE_NO_WARNINGS +#ifndef _LIBCXX_IN_DEVCRT + // Silence warnings about CRT machinery. + #define _CRT_SECURE_NO_WARNINGS -// Avoid assertion dialogs. -#define _CRT_SECURE_INVALID_PARAMETER(EXPR) ::abort() + // Avoid assertion dialogs. + #define _CRT_SECURE_INVALID_PARAMETER(EXPR) ::abort() +#endif // _LIBCXX_IN_DEVCRT #include #include @@ -31,6 +33,7 @@ #define _MSVC_STL_VER 42 #endif +#ifndef _LIBCXX_IN_DEVCRT struct AssertionDialogAvoider { AssertionDialogAvoider() { _CrtSetReportMode(_CRT_ASSERT, _CRTDBG_MODE_FILE); @@ -42,6 +45,7 @@ struct AssertionDialogAvoider { }; const AssertionDialogAvoider assertion_dialog_avoider{}; +#endif // _LIBCXX_IN_DEVCRT // MSVC frontend only configurations #if !defined(__clang__) @@ -66,16 +70,18 @@ const AssertionDialogAvoider assertion_dialog_avoider{}; // MSVC has quick_exit() and at_quick_exit(). #define _LIBCPP_HAS_QUICK_EXIT -// atomic_is_lock_free.pass.cpp needs this VS 2015 Update 2 fix. -#define _ENABLE_ATOMIC_ALIGNMENT_FIX +#ifndef _LIBCXX_IN_DEVCRT + // atomic_is_lock_free.pass.cpp needs this VS 2015 Update 2 fix. + #define _ENABLE_ATOMIC_ALIGNMENT_FIX -// Enable features that /std:c++latest removes by default. -#define _HAS_AUTO_PTR_ETC 1 -#define _HAS_FUNCTION_ASSIGN 1 -#define _HAS_OLD_IOSTREAMS_MEMBERS 1 + // Enable features that /std:c++latest removes by default. + #define _HAS_AUTO_PTR_ETC 1 + #define _HAS_FUNCTION_ASSIGN 1 + #define _HAS_OLD_IOSTREAMS_MEMBERS 1 -// Silence warnings about raw pointers and other unchecked iterators. -#define _SCL_SECURE_NO_WARNINGS + // Silence warnings about raw pointers and other unchecked iterators. + #define _SCL_SECURE_NO_WARNINGS +#endif // _LIBCXX_IN_DEVCRT #include diff --git a/test/support/test.workarounds/c1xx_broken_za_ctor_check.pass.cpp b/test/support/test.workarounds/c1xx_broken_za_ctor_check.pass.cpp new file mode 100644 index 000000000000..856574dbcf23 --- /dev/null +++ b/test/support/test.workarounds/c1xx_broken_za_ctor_check.pass.cpp @@ -0,0 +1,41 @@ +//===----------------------------------------------------------------------===// +// +// The LLVM Compiler Infrastructure +// +// This file is dual licensed under the MIT and the University of Illinois Open +// Source Licenses. See LICENSE.TXT for details. +// +//===----------------------------------------------------------------------===// + +// UNSUPPORTED: c++98, c++03 + +// Verify TEST_WORKAROUND_C1XX_BROKEN_ZA_CTOR_CHECK. + +#include + +#include "test_workarounds.h" + +struct X { + X(int) {} + + X(X&&) = default; + X& operator=(X&&) = default; + +private: + X(const X&) = default; + X& operator=(const X&) = default; +}; + +void PushFront(X&&) {} + +template +auto test(int) -> decltype(PushFront(std::declval()), std::true_type{}); +auto test(long) -> std::false_type; + +int main() { +#if defined(TEST_WORKAROUND_C1XX_BROKEN_ZA_CTOR_CHECK) + static_assert(!decltype(test(0))::value, ""); +#else + static_assert(decltype(test(0))::value, ""); +#endif +} diff --git a/test/support/test_macros.h b/test/support/test_macros.h index 46fcb66fecfc..6b573d40fe68 100644 --- a/test/support/test_macros.h +++ b/test/support/test_macros.h @@ -52,10 +52,12 @@ #define TEST_HAS_BUILTIN_IDENTIFIER(X) 0 #endif -#if defined(__clang__) -#define TEST_COMPILER_CLANG +#if defined(__EDG__) +# define TEST_COMPILER_EDG +#elif defined(__clang__) +# define TEST_COMPILER_CLANG # if defined(__apple_build_version__) -# define TEST_COMPILER_APPLE_CLANG +# define TEST_COMPILER_APPLE_CLANG # endif #elif defined(_MSC_VER) # define TEST_COMPILER_C1XX diff --git a/test/support/test_workarounds.h b/test/support/test_workarounds.h index b24c883902e8..614cec7aab86 100644 --- a/test/support/test_workarounds.h +++ b/test/support/test_workarounds.h @@ -13,9 +13,16 @@ #include "test_macros.h" +#if defined(TEST_COMPILER_EDG) +# define TEST_WORKAROUND_EDG_EXPLICIT_CONSTEXPR +#endif + #if defined(TEST_COMPILER_C1XX) # define TEST_WORKAROUND_C1XX_BROKEN_NULLPTR_CONVERSION_OPERATOR # define TEST_WORKAROUND_C1XX_BROKEN_IS_TRIVIALLY_COPYABLE +# ifndef _MSC_EXTENSIONS +# define TEST_WORKAROUND_C1XX_BROKEN_ZA_CTOR_CHECK +# endif #endif #endif // SUPPORT_TEST_WORKAROUNDS_H diff --git a/utils/libcxx/test/config.py b/utils/libcxx/test/config.py index 7f1ae851b9ce..25553c70da31 100644 --- a/utils/libcxx/test/config.py +++ b/utils/libcxx/test/config.py @@ -67,7 +67,7 @@ def __init__(self, lit_config, config): self.abi_library_root = None self.link_shared = self.get_lit_bool('enable_shared', default=True) self.debug_build = self.get_lit_bool('debug_build', default=False) - self.exec_env = {} + self.exec_env = dict(os.environ) self.use_target = False self.use_system_cxx_lib = False self.use_clang_verify = False @@ -160,7 +160,11 @@ def print_config_info(self): # Print as list to prevent "set([...])" from being printed. self.lit_config.note('Using available_features: %s' % list(self.config.available_features)) - self.lit_config.note('Using environment: %r' % self.exec_env) + show_env_vars = {} + for k,v in self.exec_env.items(): + if k not in os.environ or os.environ[k] != v: + show_env_vars[k] = v + self.lit_config.note('Adding environment variables: %r' % show_env_vars) sys.stderr.flush() # Force flushing to avoid broken output on Windows def get_test_format(self): @@ -546,6 +550,7 @@ def configure_default_compile_flags(self): def configure_compile_flags_header_includes(self): support_path = os.path.join(self.libcxx_src_root, 'test', 'support') + self.configure_config_site_header() if self.cxx_stdlib_under_test != 'libstdc++' and \ not self.is_windows: self.cxx.compile_flags += [ @@ -561,7 +566,6 @@ def configure_compile_flags_header_includes(self): '-include', os.path.join(support_path, 'set_windows_crt_report_mode.h') ] - self.configure_config_site_header() cxx_headers = self.get_lit_conf('cxx_headers') if cxx_headers == '' or (cxx_headers is None and self.cxx_stdlib_under_test != 'libc++'): @@ -868,6 +872,9 @@ def configure_warnings(self): # FIXME: Enable the two warnings below. self.cxx.addWarningFlagIfSupported('-Wno-conversion') self.cxx.addWarningFlagIfSupported('-Wno-unused-local-typedef') + # FIXME: Remove this warning once the min/max handling patch lands + # See https://reviews.llvm.org/D33080 + self.cxx.addWarningFlagIfSupported('-Wno-#warnings') std = self.get_lit_conf('std', None) if std in ['c++98', 'c++03']: # The '#define static_assert' provided by libc++ in C++03 mode @@ -1002,18 +1009,8 @@ def configure_substitutions(self): sub.append(('%link', link_str)) sub.append(('%build', build_str)) # Configure exec prefix substitutions. - exec_env_str = '' - if not self.is_windows and len(self.exec_env) != 0: - exec_env_str = 'env ' - for k, v in self.exec_env.items(): - exec_env_str += ' %s=%s' % (k, v) # Configure run env substitution. - exec_str = exec_env_str - if self.lit_config.useValgrind: - exec_str = ' '.join(self.lit_config.valgrindArgs) + exec_env_str - sub.append(('%exec', exec_str)) - # Configure run shortcut - sub.append(('%run', exec_str + ' %t.exe')) + sub.append(('%run', '%t.exe')) # Configure not program substitutions not_py = os.path.join(self.libcxx_src_root, 'utils', 'not.py') not_str = '%s %s ' % (pipes.quote(sys.executable), pipes.quote(not_py)) diff --git a/utils/libcxx/test/executor.py b/utils/libcxx/test/executor.py index 4a189174d915..0ccf96caa8ba 100644 --- a/utils/libcxx/test/executor.py +++ b/utils/libcxx/test/executor.py @@ -38,36 +38,11 @@ def __init__(self): def run(self, exe_path, cmd=None, work_dir='.', file_deps=None, env=None): cmd = cmd or [exe_path] - env_cmd = [] - if env: - env_cmd += ['env'] - env_cmd += ['%s=%s' % (k, v) for k, v in env.items()] if work_dir == '.': work_dir = os.getcwd() - if not self.is_windows: - out, err, rc = executeCommand(env_cmd + cmd, cwd=work_dir) - else: - out, err, rc = executeCommand(cmd, cwd=work_dir, - env=self._build_windows_env(env)) - return (env_cmd + cmd, out, err, rc) + out, err, rc = executeCommand(cmd, cwd=work_dir, env=env) + return (cmd, out, err, rc) - def _build_windows_env(self, exec_env): - # FIXME: Finding Windows DLL's at runtime requires modifying the - # PATH environment variables. However we don't want to print out - # the entire PATH as part of the diagnostic for every failing test. - # Therefore this hack builds a new executable environment that - # merges the current environment and the supplied environment while - # still only printing the supplied environment in diagnostics. - if not self.is_windows or exec_env is None: - return None - new_env = dict(os.environ) - for key, value in exec_env.items(): - if key == 'PATH': - assert value.strip() != '' and "expected non-empty path" - new_env['PATH'] = "%s;%s" % (value, os.environ['PATH']) - else: - new_env[key] = value - return new_env class PrefixExecutor(Executor): """Prefix an executor with some other command wrapper. diff --git a/utils/libcxx/test/format.py b/utils/libcxx/test/format.py index c3bc97187ad4..3ac5472b4720 100644 --- a/utils/libcxx/test/format.py +++ b/utils/libcxx/test/format.py @@ -139,7 +139,7 @@ def _execute(self, test, lit_config): # We can't run ShTest tests with a executor yet. # For now, bail on trying to run them return lit.Test.UNSUPPORTED, 'ShTest format not yet supported' - test.config.enviroment = dict(self.exec_env) + test.config.environment = dict(self.exec_env) return lit.TestRunner._runShTest(test, lit_config, self.execute_external, script, tmpBase) diff --git a/www/cxx1z_status.html b/www/cxx1z_status.html index 867a6c41f9ad..101ad43212e5 100644 --- a/www/cxx1z_status.html +++ b/www/cxx1z_status.html @@ -387,7 +387,7 @@ 2584 ECMAScript IdentityEscape is ambiguousIssaquah 2588[fund.ts.v2] "Convertible to bool" requirement in conjunction and disjunctionIssaquah 2589match_results can't satisfy the requirements of a containerIssaquahComplete - 2591std::function's member template target() should not lead to undefined behaviourIssaquah + 2591std::function's member template target() should not lead to undefined behaviourIssaquahComplete 2598addressof works on temporariesIssaquahComplete 2664operator/ (and other append) semantics not useful if argument has rootIssaquahComplete 2665remove_filename() post condition is incorrectIssaquahComplete @@ -437,7 +437,7 @@ 2768any_cast and move semanticsKonaComplete 2769Redundant const in the return type of any_cast(const any&)KonaComplete 2781Contradictory requirements for std::function and std::reference_wrapperKonaComplete - 2782scoped_allocator_adaptor constructors must be constrainedKona + 2782scoped_allocator_adaptor constructors must be constrainedKonaComplete 2784Resolution to LWG 2484 is missing "otherwise, no effects" and is hard to parseKonaComplete 2785quoted should work with basic_string_viewKonaComplete 2786Annex C should mention shared_ptr changes for array supportKonaComplete @@ -447,7 +447,7 @@ 2790Missing specification of istreambuf_iterator::operator->Kona 2794Missing requirements for allocator pointersKona 2795§[global.functions] provides incorrect example of ADL useKonaComplete - 2796tuple should be a literal typeKona + 2796tuple should be a literal typeKonaComplete 2801Default-constructibility of unique_ptrKonaComplete 2802shared_ptr constructor requirements for a deleterKona 2804Unconditional constexpr default constructor for istream_iteratorKonaComplete @@ -461,7 +461,7 @@ 2837gcd and lcm should support a wider range of input valuesKonaComplete 2838is_literal_type specification needs a little cleanupKonaComplete 2842in_place_t check for optional::optional(U&&) should decay UKonaComplete - 2850std::function move constructor does unnecessary workKona + 2850std::function move constructor does unnecessary workKonaComplete 2853Possible inconsistency in specification of erase in [vector.modifiers]KonaComplete 2855std::throw_with_nested("string_literal")KonaComplete 2857{variant,optional,any}::emplace should return the constructed valueKonaComplete @@ -489,7 +489,7 @@ -

Last Updated: 17-Apr-2017

+

Last Updated: 11-May-2017

From 022ebf5bbf58ca2dd943d3376cc95a6b206db799 Mon Sep 17 00:00:00 2001 From: Dimitry Andric Date: Tue, 16 May 2017 19:47:41 +0000 Subject: [PATCH 3/9] Vendor import of lld trunk r303197: https://llvm.org/svn/llvm-project/lld/trunk@303197 --- COFF/Driver.cpp | 2 +- COFF/ICF.cpp | 4 +- COFF/MapFile.cpp | 4 +- COFF/Writer.cpp | 16 +- ELF/Config.h | 1 - ELF/Driver.cpp | 8 +- ELF/GdbIndex.h | 2 +- ELF/ICF.cpp | 2 +- ELF/InputFiles.cpp | 6 +- ELF/InputSection.cpp | 56 ++- ELF/LinkerScript.cpp | 84 ++-- ELF/LinkerScript.h | 11 +- ELF/MapFile.cpp | 2 +- ELF/OutputSections.cpp | 13 +- ELF/OutputSections.h | 1 + ELF/Relocations.cpp | 80 ++-- ELF/ScriptParser.cpp | 24 +- ELF/Strings.cpp | 4 +- ELF/Symbols.cpp | 9 +- ELF/Symbols.h | 2 +- ELF/SyntheticSections.cpp | 217 ++++----- ELF/SyntheticSections.h | 51 ++- ELF/Target.cpp | 60 ++- ELF/Threads.h | 21 +- ELF/Thunks.cpp | 18 +- ELF/Writer.cpp | 470 ++++++++++---------- ELF/Writer.h | 2 +- docs/CMakeLists.txt | 2 +- include/lld/Core/Parallel.h | 166 ------- include/lld/Core/TaskGroup.h | 65 --- lib/Core/CMakeLists.txt | 1 - lib/Core/TaskGroup.cpp | 141 ------ lib/ReaderWriter/MachO/LayoutPass.cpp | 10 +- test/COFF/constant-export.test | 85 +++- test/COFF/constant-export.yaml | 83 ---- test/ELF/Inputs/i386-static-tls-model1.s | 10 - test/ELF/Inputs/i386-static-tls-model2.s | 9 - test/ELF/Inputs/i386-static-tls-model3.s | 9 - test/ELF/Inputs/i386-static-tls-model4.s | 9 - test/ELF/gdb-index-empty.s | 116 +++++ test/ELF/gdb-index-gc-sections.s | 157 +++++++ test/ELF/i386-static-tls-model.s | 20 - test/ELF/i386-tls-ie-shared.s | 52 +-- test/ELF/incompatible-section-types2.s | 6 +- test/ELF/linkerscript/early-assign-symbol.s | 14 + test/ELF/linkerscript/ehdr_start.s | 14 +- test/ELF/linkerscript/sections-constraint.s | 2 +- test/ELF/linkerscript/sections.s | 5 +- test/ELF/linkerscript/symbol-memoryexpr.s | 33 ++ test/ELF/many-alloc-sections.s | 106 +++++ test/ELF/many-sections.s | 9 +- test/ELF/tls-dynamic-i686.s | 46 +- test/ELF/tls-opt-iele-i686-nopic.s | 44 +- test/ELF/x86-64-reloc-tpoff32-fpic.s | 14 + unittests/CMakeLists.txt | 1 - unittests/CoreTests/CMakeLists.txt | 7 - unittests/CoreTests/ParallelTest.cpp | 46 -- 57 files changed, 1213 insertions(+), 1239 deletions(-) delete mode 100644 include/lld/Core/Parallel.h delete mode 100644 include/lld/Core/TaskGroup.h delete mode 100644 lib/Core/TaskGroup.cpp delete mode 100644 test/COFF/constant-export.yaml delete mode 100644 test/ELF/Inputs/i386-static-tls-model1.s delete mode 100644 test/ELF/Inputs/i386-static-tls-model2.s delete mode 100644 test/ELF/Inputs/i386-static-tls-model3.s delete mode 100644 test/ELF/Inputs/i386-static-tls-model4.s create mode 100644 test/ELF/gdb-index-empty.s create mode 100644 test/ELF/gdb-index-gc-sections.s delete mode 100644 test/ELF/i386-static-tls-model.s create mode 100644 test/ELF/linkerscript/early-assign-symbol.s create mode 100644 test/ELF/linkerscript/symbol-memoryexpr.s create mode 100644 test/ELF/many-alloc-sections.s create mode 100644 test/ELF/x86-64-reloc-tpoff32-fpic.s delete mode 100644 unittests/CoreTests/CMakeLists.txt delete mode 100644 unittests/CoreTests/ParallelTest.cpp diff --git a/COFF/Driver.cpp b/COFF/Driver.cpp index 5a15b5b11507..4c0ea44b875e 100644 --- a/COFF/Driver.cpp +++ b/COFF/Driver.cpp @@ -18,7 +18,6 @@ #include "lld/Driver/Driver.h" #include "llvm/ADT/Optional.h" #include "llvm/ADT/StringSwitch.h" -#include "llvm/LibDriver/LibDriver.h" #include "llvm/Object/ArchiveWriter.h" #include "llvm/Option/Arg.h" #include "llvm/Option/ArgList.h" @@ -29,6 +28,7 @@ #include "llvm/Support/TarWriter.h" #include "llvm/Support/TargetSelect.h" #include "llvm/Support/raw_ostream.h" +#include "llvm/ToolDrivers/llvm-lib/LibDriver.h" #include #include diff --git a/COFF/ICF.cpp b/COFF/ICF.cpp index 9a43f2bd43f5..3b7cc424f0a2 100644 --- a/COFF/ICF.cpp +++ b/COFF/ICF.cpp @@ -21,9 +21,9 @@ #include "Chunks.h" #include "Error.h" #include "Symbols.h" -#include "lld/Core/Parallel.h" #include "llvm/ADT/Hashing.h" #include "llvm/Support/Debug.h" +#include "llvm/Support/Parallel.h" #include "llvm/Support/raw_ostream.h" #include #include @@ -192,7 +192,7 @@ void ICF::forEachClass(std::function Fn) { // Split sections into 256 shards and call Fn in parallel. size_t NumShards = 256; size_t Step = Chunks.size() / NumShards; - parallel_for(size_t(0), NumShards, [&](size_t I) { + for_each_n(parallel::par, size_t(0), NumShards, [&](size_t I) { forEachClassRange(I * Step, (I + 1) * Step, Fn); }); forEachClassRange(Step * NumShards, Chunks.size(), Fn); diff --git a/COFF/MapFile.cpp b/COFF/MapFile.cpp index 4e596e602fee..b63d4672c7d5 100644 --- a/COFF/MapFile.cpp +++ b/COFF/MapFile.cpp @@ -25,7 +25,7 @@ #include "Symbols.h" #include "Writer.h" -#include "lld/Core/Parallel.h" +#include "llvm/Support/Parallel.h" #include "llvm/Support/raw_ostream.h" using namespace llvm; @@ -76,7 +76,7 @@ static SymbolMapTy getSectionSyms(ArrayRef Syms) { static DenseMap getSymbolStrings(ArrayRef Syms) { std::vector Str(Syms.size()); - parallel_for((size_t)0, Syms.size(), [&](size_t I) { + for_each_n(parallel::par, (size_t)0, Syms.size(), [&](size_t I) { raw_string_ostream OS(Str[I]); writeHeader(OS, Syms[I]->getRVA(), 0, 0); OS << indent(2) << toString(*Syms[I]); diff --git a/COFF/Writer.cpp b/COFF/Writer.cpp index 8762b88c4d6b..5c9c8375dadc 100644 --- a/COFF/Writer.cpp +++ b/COFF/Writer.cpp @@ -17,13 +17,13 @@ #include "PDB.h" #include "SymbolTable.h" #include "Symbols.h" -#include "lld/Core/Parallel.h" #include "llvm/ADT/DenseMap.h" #include "llvm/ADT/STLExtras.h" #include "llvm/ADT/StringSwitch.h" #include "llvm/Support/Debug.h" #include "llvm/Support/Endian.h" #include "llvm/Support/FileOutputBuffer.h" +#include "llvm/Support/Parallel.h" #include "llvm/Support/RandomNumberGenerator.h" #include "llvm/Support/raw_ostream.h" #include @@ -745,8 +745,8 @@ void Writer::writeSections() { // ADD instructions). if (Sec->getPermissions() & IMAGE_SCN_CNT_CODE) memset(SecBuf, 0xCC, Sec->getRawSize()); - parallel_for_each(Sec->getChunks().begin(), Sec->getChunks().end(), - [&](Chunk *C) { C->writeTo(SecBuf); }); + for_each(parallel::par, Sec->getChunks().begin(), Sec->getChunks().end(), + [&](Chunk *C) { C->writeTo(SecBuf); }); } } @@ -760,16 +760,14 @@ void Writer::sortExceptionTable() { uint8_t *End = Begin + Sec->getVirtualSize(); if (Config->Machine == AMD64) { struct Entry { ulittle32_t Begin, End, Unwind; }; - parallel_sort( - (Entry *)Begin, (Entry *)End, - [](const Entry &A, const Entry &B) { return A.Begin < B.Begin; }); + sort(parallel::par, (Entry *)Begin, (Entry *)End, + [](const Entry &A, const Entry &B) { return A.Begin < B.Begin; }); return; } if (Config->Machine == ARMNT) { struct Entry { ulittle32_t Begin, Unwind; }; - parallel_sort( - (Entry *)Begin, (Entry *)End, - [](const Entry &A, const Entry &B) { return A.Begin < B.Begin; }); + sort(parallel::par, (Entry *)Begin, (Entry *)End, + [](const Entry &A, const Entry &B) { return A.Begin < B.Begin; }); return; } errs() << "warning: don't know how to handle .pdata.\n"; diff --git a/ELF/Config.h b/ELF/Config.h index 0321c84e7106..57a0e5a5ec73 100644 --- a/ELF/Config.h +++ b/ELF/Config.h @@ -73,7 +73,6 @@ struct VersionDefinition { // Most fields are initialized by the driver. struct Configuration { InputFile *FirstElf = nullptr; - bool HasStaticTlsModel = false; uint8_t OSABI = 0; llvm::CachePruningPolicy ThinLTOCachePolicy; llvm::StringMap SectionStartMap; diff --git a/ELF/Driver.cpp b/ELF/Driver.cpp index c2cfe3c4129e..737c6a6bf114 100644 --- a/ELF/Driver.cpp +++ b/ELF/Driver.cpp @@ -284,7 +284,7 @@ static int getInteger(opt::InputArgList &Args, unsigned Key, int Default) { int V = Default; if (auto *Arg = Args.getLastArg(Key)) { StringRef S = Arg->getValue(); - if (S.getAsInteger(10, V)) + if (!to_integer(S, V, 10)) error(Arg->getSpelling() + ": number expected, but got " + S); } return V; @@ -311,7 +311,7 @@ static uint64_t getZOptionValue(opt::InputArgList &Args, StringRef Key, if (Pos != StringRef::npos && Key == Value.substr(0, Pos)) { Value = Value.substr(Pos + 1); uint64_t Result; - if (Value.getAsInteger(0, Result)) + if (!to_integer(Value, Result)) error("invalid " + Key + ": " + Value); return Result; } @@ -522,7 +522,7 @@ static uint64_t parseSectionAddress(StringRef S, opt::Arg *Arg) { uint64_t VA = 0; if (S.startswith("0x")) S = S.drop_front(2); - if (S.getAsInteger(16, VA)) + if (!to_integer(S, VA, 16)) error("invalid argument: " + toString(Arg)); return VA; } @@ -886,7 +886,7 @@ static uint64_t getImageBase(opt::InputArgList &Args) { StringRef S = Arg->getValue(); uint64_t V; - if (S.getAsInteger(0, V)) { + if (!to_integer(S, V)) { error("-image-base: number expected, but got " + S); return 0; } diff --git a/ELF/GdbIndex.h b/ELF/GdbIndex.h index a36b92714def..03fec64f9bd5 100644 --- a/ELF/GdbIndex.h +++ b/ELF/GdbIndex.h @@ -21,7 +21,7 @@ class InputSection; // Struct represents single entry of address area of gdb index. struct AddressEntry { - InputSectionBase *Section; + InputSection *Section; uint64_t LowAddress; uint64_t HighAddress; size_t CuIndex; diff --git a/ELF/ICF.cpp b/ELF/ICF.cpp index dcf01ea80011..3722d4e3ed2f 100644 --- a/ELF/ICF.cpp +++ b/ELF/ICF.cpp @@ -325,7 +325,7 @@ void ICF::forEachClass(std::function Fn) { // Split sections into 256 shards and call Fn in parallel. size_t NumShards = 256; size_t Step = Sections.size() / NumShards; - parallelFor(0, NumShards, [&](size_t I) { + parallelForEachN(0, NumShards, [&](size_t I) { forEachClassRange(I * Step, (I + 1) * Step, Fn); }); forEachClassRange(Step * NumShards, Sections.size(), Fn); diff --git a/ELF/InputFiles.cpp b/ELF/InputFiles.cpp index 5f94fc9338a4..fe036a644f41 100644 --- a/ELF/InputFiles.cpp +++ b/ELF/InputFiles.cpp @@ -383,9 +383,9 @@ elf::ObjectFile::createInputSection(const Elf_Shdr &Sec, // we see. The eglibc ARM dynamic loaders require the presence of an // attribute section for dlopen to work. // In a full implementation we would merge all attribute sections. - if (In::ARMAttributes == nullptr) { - In::ARMAttributes = make(this, &Sec, Name); - return In::ARMAttributes; + if (InX::ARMAttributes == nullptr) { + InX::ARMAttributes = make(this, &Sec, Name); + return InX::ARMAttributes; } return &InputSection::Discarded; case SHT_RELA: diff --git a/ELF/InputSection.cpp b/ELF/InputSection.cpp index c082f128a9bc..87896ec96b29 100644 --- a/ELF/InputSection.cpp +++ b/ELF/InputSection.cpp @@ -324,7 +324,7 @@ void InputSection::copyRelocations(uint8_t *Buf, ArrayRef Rels) { // section, but for --emit-relocs it is an virtual address. P->r_offset = RelocatedSection->OutSec->Addr + RelocatedSection->getOffset(Rel.r_offset); - P->setSymbolAndType(In::SymTab->getSymbolIndex(&Body), Type, + P->setSymbolAndType(InX::SymTab->getSymbolIndex(&Body), Type, Config->IsMips64EL); if (Body.Type == STT_SECTION) { @@ -400,40 +400,40 @@ getRelocTargetVA(uint32_t Type, int64_t A, typename ELFT::uint P, return Body.getVA(A); case R_GOT: case R_RELAX_TLS_GD_TO_IE_ABS: - return Body.getGotVA() + A; + return Body.getGotVA() + A; case R_GOTONLY_PC: - return In::Got->getVA() + A - P; + return InX::Got->getVA() + A - P; case R_GOTONLY_PC_FROM_END: - return In::Got->getVA() + A - P + In::Got->getSize(); + return InX::Got->getVA() + A - P + InX::Got->getSize(); case R_GOTREL: - return Body.getVA(A) - In::Got->getVA(); + return Body.getVA(A) - InX::Got->getVA(); case R_GOTREL_FROM_END: - return Body.getVA(A) - In::Got->getVA() - In::Got->getSize(); + return Body.getVA(A) - InX::Got->getVA() - InX::Got->getSize(); case R_GOT_FROM_END: case R_RELAX_TLS_GD_TO_IE_END: - return Body.getGotOffset() + A - In::Got->getSize(); + return Body.getGotOffset() + A - InX::Got->getSize(); case R_GOT_OFF: return Body.getGotOffset() + A; case R_GOT_PAGE_PC: case R_RELAX_TLS_GD_TO_IE_PAGE_PC: - return getAArch64Page(Body.getGotVA() + A) - getAArch64Page(P); + return getAArch64Page(Body.getGotVA() + A) - getAArch64Page(P); case R_GOT_PC: case R_RELAX_TLS_GD_TO_IE: - return Body.getGotVA() + A - P; + return Body.getGotVA() + A - P; case R_HINT: case R_NONE: case R_TLSDESC_CALL: llvm_unreachable("cannot relocate hint relocs"); case R_MIPS_GOTREL: - return Body.getVA(A) - In::MipsGot->getGp(); + return Body.getVA(A) - InX::MipsGot->getGp(); case R_MIPS_GOT_GP: - return In::MipsGot->getGp() + A; + return InX::MipsGot->getGp() + A; case R_MIPS_GOT_GP_PC: { // R_MIPS_LO16 expression has R_MIPS_GOT_GP_PC type iif the target // is _gp_disp symbol. In that case we should use the following // formula for calculation "AHL + GP - P + 4". For details see p. 4-19 at // ftp://www.linux-mips.org/pub/linux/mips/doc/ABI/mipsabi.pdf - uint64_t V = In::MipsGot->getGp() + A - P; + uint64_t V = InX::MipsGot->getGp() + A - P; if (Type == R_MIPS_LO16) V += 4; return V; @@ -442,24 +442,21 @@ getRelocTargetVA(uint32_t Type, int64_t A, typename ELFT::uint P, // If relocation against MIPS local symbol requires GOT entry, this entry // should be initialized by 'page address'. This address is high 16-bits // of sum the symbol's value and the addend. - return In::MipsGot->getVA() + - In::MipsGot->getPageEntryOffset(Body, A) - - In::MipsGot->getGp(); + return InX::MipsGot->getVA() + InX::MipsGot->getPageEntryOffset(Body, A) - + InX::MipsGot->getGp(); case R_MIPS_GOT_OFF: case R_MIPS_GOT_OFF32: // In case of MIPS if a GOT relocation has non-zero addend this addend // should be applied to the GOT entry content not to the GOT entry offset. // That is why we use separate expression type. - return In::MipsGot->getVA() + - In::MipsGot->getBodyEntryOffset(Body, A) - - In::MipsGot->getGp(); + return InX::MipsGot->getVA() + InX::MipsGot->getBodyEntryOffset(Body, A) - + InX::MipsGot->getGp(); case R_MIPS_TLSGD: - return In::MipsGot->getVA() + In::MipsGot->getTlsOffset() + - In::MipsGot->getGlobalDynOffset(Body) - - In::MipsGot->getGp(); + return InX::MipsGot->getVA() + InX::MipsGot->getTlsOffset() + + InX::MipsGot->getGlobalDynOffset(Body) - InX::MipsGot->getGp(); case R_MIPS_TLSLD: - return In::MipsGot->getVA() + In::MipsGot->getTlsOffset() + - In::MipsGot->getTlsIndexOff() - In::MipsGot->getGp(); + return InX::MipsGot->getVA() + InX::MipsGot->getTlsOffset() + + InX::MipsGot->getTlsIndexOff() - InX::MipsGot->getGp(); case R_PAGE_PC: case R_PLT_PAGE_PC: if (Body.isUndefined() && !Body.isLocal() && Body.symbol()->isWeak()) @@ -523,19 +520,18 @@ getRelocTargetVA(uint32_t Type, int64_t A, typename ELFT::uint P, case R_SIZE: return Body.getSize() + A; case R_TLSDESC: - return In::Got->getGlobalDynAddr(Body) + A; + return InX::Got->getGlobalDynAddr(Body) + A; case R_TLSDESC_PAGE: - return getAArch64Page(In::Got->getGlobalDynAddr(Body) + A) - + return getAArch64Page(InX::Got->getGlobalDynAddr(Body) + A) - getAArch64Page(P); case R_TLSGD: - return In::Got->getGlobalDynOffset(Body) + A - - In::Got->getSize(); + return InX::Got->getGlobalDynOffset(Body) + A - InX::Got->getSize(); case R_TLSGD_PC: - return In::Got->getGlobalDynAddr(Body) + A - P; + return InX::Got->getGlobalDynAddr(Body) + A - P; case R_TLSLD: - return In::Got->getTlsIndexOff() + A - In::Got->getSize(); + return InX::Got->getTlsIndexOff() + A - InX::Got->getSize(); case R_TLSLD_PC: - return In::Got->getTlsIndexVA() + A - P; + return InX::Got->getTlsIndexVA() + A - P; } llvm_unreachable("Invalid expression"); } diff --git a/ELF/LinkerScript.cpp b/ELF/LinkerScript.cpp index d7858e173c7b..161909abf00d 100644 --- a/ELF/LinkerScript.cpp +++ b/ELF/LinkerScript.cpp @@ -48,8 +48,12 @@ using namespace lld::elf; LinkerScript *elf::Script; uint64_t ExprValue::getValue() const { - if (Sec) - return Sec->getOffset(Val) + Sec->getOutputSection()->Addr; + if (Sec) { + if (Sec->getOutputSection()) + return Sec->getOffset(Val) + Sec->getOutputSection()->Addr; + error("unable to evaluate expression: input section " + Sec->Name + + " has no output section assigned"); + } return Val; } @@ -411,6 +415,7 @@ void LinkerScript::processCommands(OutputSectionFactory &Factory) { if (OutputSection *Sec = Cmd->Sec) { assert(Sec->SectionIndex == INT_MAX); Sec->SectionIndex = I; + SecToCommand[Sec] = Cmd; } } } @@ -440,6 +445,7 @@ void LinkerScript::fabricateDefaultCommands() { auto *OSCmd = make(Sec->Name); OSCmd->Sec = Sec; + SecToCommand[Sec] = OSCmd; // Prefer user supplied address over additional alignment constraint auto I = Config->SectionStartMap.find(Sec->Name); @@ -484,6 +490,7 @@ void LinkerScript::addOrphanSections(OutputSectionFactory &Factory) { auto *Cmd = cast(*I); Factory.addInputSec(S, Name, Cmd->Sec); if (OutputSection *Sec = Cmd->Sec) { + SecToCommand[Sec] = Cmd; unsigned Index = std::distance(Opt.Commands.begin(), I); assert(Sec->SectionIndex == INT_MAX || Sec->SectionIndex == Index); Sec->SectionIndex = Index; @@ -699,6 +706,7 @@ void LinkerScript::adjustSectionsBeforeSorting() { OutSec->SectionIndex = I; OutputSections->push_back(OutSec); Cmd->Sec = OutSec; + SecToCommand[OutSec] = Cmd; } } @@ -822,16 +830,14 @@ void LinkerScript::placeOrphanSections() { // If there is no command corresponding to this output section, // create one and put a InputSectionDescription in it so that both // representations agree on which input sections to use. - auto Pos = std::find_if(CmdIter, E, [&](BaseCommand *Base) { - auto *Cmd = dyn_cast(Base); - return Cmd && Cmd->Name == Name; - }); - if (Pos == E) { - auto *Cmd = make(Name); + OutputSectionCommand *Cmd = getCmd(Sec); + if (!Cmd) { + Cmd = make(Name); Opt.Commands.insert(CmdIter, Cmd); ++CmdIndex; Cmd->Sec = Sec; + SecToCommand[Sec] = Cmd; auto *ISD = make(""); for (InputSection *IS : Sec->Sections) ISD->Sections.push_back(IS); @@ -841,7 +847,11 @@ void LinkerScript::placeOrphanSections() { } // Continue from where we found it. - CmdIndex = (Pos - Opt.Commands.begin()) + 1; + while (*CmdIter != Cmd) { + ++CmdIter; + ++CmdIndex; + } + ++CmdIndex; } } @@ -1000,7 +1010,7 @@ std::vector LinkerScript::createPhdrs() { break; // Assign headers specified by linker script - for (size_t Id : getPhdrIndices(Sec->Name)) { + for (size_t Id : getPhdrIndices(Sec)) { Ret[Id].add(Sec); if (Opt.PhdrsCommands[Id].Flags == UINT_MAX) Ret[Id].p_flags |= Sec->getPhdrFlags(); @@ -1020,11 +1030,16 @@ bool LinkerScript::ignoreInterpSection() { return true; } -Optional LinkerScript::getFiller(StringRef Name) { - for (BaseCommand *Base : Opt.Commands) - if (auto *Cmd = dyn_cast(Base)) - if (Cmd->Name == Name) - return Cmd->Filler; +OutputSectionCommand *LinkerScript::getCmd(OutputSection *Sec) const { + auto I = SecToCommand.find(Sec); + if (I == SecToCommand.end()) + return nullptr; + return I->second; +} + +Optional LinkerScript::getFiller(OutputSection *Sec) { + if (OutputSectionCommand *Cmd = getCmd(Sec)) + return Cmd->Filler; return None; } @@ -1042,26 +1057,16 @@ static void writeInt(uint8_t *Buf, uint64_t Data, uint64_t Size) { } void LinkerScript::writeDataBytes(OutputSection *Sec, uint8_t *Buf) { - auto I = std::find_if(Opt.Commands.begin(), Opt.Commands.end(), - [=](BaseCommand *Base) { - if (auto *Cmd = dyn_cast(Base)) - if (Cmd->Sec == Sec) - return true; - return false; - }); - if (I == Opt.Commands.end()) - return; - auto *Cmd = cast(*I); - for (BaseCommand *Base : Cmd->Commands) - if (auto *Data = dyn_cast(Base)) - writeInt(Buf + Data->Offset, Data->Expression().getValue(), Data->Size); + if (OutputSectionCommand *Cmd = getCmd(Sec)) + for (BaseCommand *Base : Cmd->Commands) + if (auto *Data = dyn_cast(Base)) + writeInt(Buf + Data->Offset, Data->Expression().getValue(), Data->Size); } -bool LinkerScript::hasLMA(StringRef Name) { - for (BaseCommand *Base : Opt.Commands) - if (auto *Cmd = dyn_cast(Base)) - if (Cmd->LMAExpr && Cmd->Name == Name) - return true; +bool LinkerScript::hasLMA(OutputSection *Sec) { + if (OutputSectionCommand *Cmd = getCmd(Sec)) + if (Cmd->LMAExpr) + return true; return false; } @@ -1080,15 +1085,10 @@ ExprValue LinkerScript::getSymbolValue(const Twine &Loc, StringRef S) { bool LinkerScript::isDefined(StringRef S) { return findSymbol(S) != nullptr; } -// Returns indices of ELF headers containing specific section, identified -// by Name. Each index is a zero based number of ELF header listed within -// PHDRS {} script block. -std::vector LinkerScript::getPhdrIndices(StringRef SectionName) { - for (BaseCommand *Base : Opt.Commands) { - auto *Cmd = dyn_cast(Base); - if (!Cmd || Cmd->Name != SectionName) - continue; - +// Returns indices of ELF headers containing specific section. Each index is a +// zero based number of ELF header listed within PHDRS {} script block. +std::vector LinkerScript::getPhdrIndices(OutputSection *Sec) { + if (OutputSectionCommand *Cmd = getCmd(Sec)) { std::vector Ret; for (StringRef PhdrName : Cmd->Phdrs) Ret.push_back(getPhdrIndex(Cmd->Location, PhdrName)); diff --git a/ELF/LinkerScript.h b/ELF/LinkerScript.h index 7bcd21c87602..d0a4d83d72b0 100644 --- a/ELF/LinkerScript.h +++ b/ELF/LinkerScript.h @@ -211,8 +211,9 @@ struct ScriptConfiguration { std::vector ReferencedSymbols; }; -class LinkerScript { -protected: +class LinkerScript final { + llvm::DenseMap SecToCommand; + OutputSectionCommand *getCmd(OutputSection *Sec) const; void assignSymbol(SymbolAssignment *Cmd, bool InSec); void setDot(Expr E, const Twine &Loc, bool InSec); @@ -222,7 +223,7 @@ class LinkerScript { std::vector createInputSectionList(OutputSectionCommand &Cmd); - std::vector getPhdrIndices(StringRef SectionName); + std::vector getPhdrIndices(OutputSection *Sec); size_t getPhdrIndex(const Twine &Loc, StringRef PhdrName); MemoryRegion *findMemoryRegion(OutputSectionCommand *Cmd); @@ -262,8 +263,8 @@ class LinkerScript { std::vector createPhdrs(); bool ignoreInterpSection(); - llvm::Optional getFiller(StringRef Name); - bool hasLMA(StringRef Name); + llvm::Optional getFiller(OutputSection *Sec); + bool hasLMA(OutputSection *Sec); bool shouldKeep(InputSectionBase *S); void assignOffsets(OutputSectionCommand *Cmd); void placeOrphanSections(); diff --git a/ELF/MapFile.cpp b/ELF/MapFile.cpp index af5bc3c2c813..23c63e845c9a 100644 --- a/ELF/MapFile.cpp +++ b/ELF/MapFile.cpp @@ -84,7 +84,7 @@ template DenseMap getSymbolStrings(ArrayRef Syms) { std::vector Str(Syms.size()); - parallelFor(0, Syms.size(), [&](size_t I) { + parallelForEachN(0, Syms.size(), [&](size_t I) { raw_string_ostream OS(Str[I]); writeHeader(OS, Syms[I]->getVA(), Syms[I]->template getSize(), 0); diff --git a/ELF/OutputSections.cpp b/ELF/OutputSections.cpp index cb9c57657af3..dcefd03766d7 100644 --- a/ELF/OutputSections.cpp +++ b/ELF/OutputSections.cpp @@ -133,7 +133,7 @@ template void OutputSection::finalize() { if (isa(First)) return; - this->Link = In::SymTab->OutSec->SectionIndex; + this->Link = InX::SymTab->OutSec->SectionIndex; // sh_info for SHT_REL[A] sections should contain the section header index of // the section to which the relocation applies. InputSectionBase *S = First->getRelocatedSection(); @@ -273,7 +273,7 @@ uint32_t OutputSection::getFiller() { // linker script. If nothing is specified and this is an executable section, // fall back to trap instructions to prevent bad diassembly and detect invalid // jumps to padding. - if (Optional Filler = Script->getFiller(Name)) + if (Optional Filler = Script->getFiller(this)) return *Filler; if (Flags & SHF_EXECINSTR) return Target->TrapInstr; @@ -297,7 +297,7 @@ template void OutputSection::writeTo(uint8_t *Buf) { if (Filler) fill(Buf, Sections.empty() ? Size : Sections[0]->OutSecOff, Filler); - parallelFor(0, Sections.size(), [=](size_t I) { + parallelForEachN(0, Sections.size(), [=](size_t I) { InputSection *Sec = Sections[I]; Sec->writeTo(Buf); @@ -429,8 +429,11 @@ void OutputSectionFactory::addInputSec(InputSectionBase *IS, if (canMergeToProgbits(Sec->Type) && canMergeToProgbits(IS->Type)) Sec->Type = SHT_PROGBITS; else - error("Section has different type from others with the same name " + - toString(IS)); + error("section type mismatch for " + IS->Name + + "\n>>> " + toString(IS) + ": " + + getELFSectionTypeName(Config->EMachine, IS->Type) + + "\n>>> output section " + Sec->Name + ": " + + getELFSectionTypeName(Config->EMachine, Sec->Type)); } Sec->Flags |= Flags; } else { diff --git a/ELF/OutputSections.h b/ELF/OutputSections.h index 6405fb38c6d6..413871b60cf7 100644 --- a/ELF/OutputSections.h +++ b/ELF/OutputSections.h @@ -50,6 +50,7 @@ class OutputSection final : public SectionBase { template void writeHeaderTo(typename ELFT::Shdr *SHdr); unsigned SectionIndex; + unsigned SortRank; uint32_t getPhdrFlags() const; diff --git a/ELF/Relocations.cpp b/ELF/Relocations.cpp index f5db931e9755..ea7477e03842 100644 --- a/ELF/Relocations.cpp +++ b/ELF/Relocations.cpp @@ -106,21 +106,21 @@ static unsigned handleMipsTlsRelocation(uint32_t Type, SymbolBody &Body, InputSectionBase &C, uint64_t Offset, int64_t Addend, RelExpr Expr) { if (Expr == R_MIPS_TLSLD) { - if (In::MipsGot->addTlsIndex() && Config->Pic) - In::RelaDyn->addReloc({Target->TlsModuleIndexRel, In::MipsGot, - In::MipsGot->getTlsIndexOff(), false, + if (InX::MipsGot->addTlsIndex() && Config->Pic) + In::RelaDyn->addReloc({Target->TlsModuleIndexRel, InX::MipsGot, + InX::MipsGot->getTlsIndexOff(), false, nullptr, 0}); C.Relocations.push_back({Expr, Type, Offset, Addend, &Body}); return 1; } if (Expr == R_MIPS_TLSGD) { - if (In::MipsGot->addDynTlsEntry(Body) && Body.isPreemptible()) { - uint64_t Off = In::MipsGot->getGlobalDynOffset(Body); + if (InX::MipsGot->addDynTlsEntry(Body) && Body.isPreemptible()) { + uint64_t Off = InX::MipsGot->getGlobalDynOffset(Body); In::RelaDyn->addReloc( - {Target->TlsModuleIndexRel, In::MipsGot, Off, false, &Body, 0}); + {Target->TlsModuleIndexRel, InX::MipsGot, Off, false, &Body, 0}); if (Body.isPreemptible()) - In::RelaDyn->addReloc({Target->TlsOffsetRel, In::MipsGot, + In::RelaDyn->addReloc({Target->TlsOffsetRel, InX::MipsGot, Off + Config->Wordsize, false, &Body, 0}); } C.Relocations.push_back({Expr, Type, Offset, Addend, &Body}); @@ -156,17 +156,17 @@ static unsigned handleARMTlsRelocation(uint32_t Type, SymbolBody &Body, auto AddTlsReloc = [&](uint64_t Off, uint32_t Type, SymbolBody *Dest, bool Dyn) { if (Dyn) - In::RelaDyn->addReloc({Type, In::Got, Off, false, Dest, 0}); + In::RelaDyn->addReloc({Type, InX::Got, Off, false, Dest, 0}); else - In::Got->Relocations.push_back({R_ABS, Type, Off, 0, Dest}); + InX::Got->Relocations.push_back({R_ABS, Type, Off, 0, Dest}); }; // Local Dynamic is for access to module local TLS variables, while still // being suitable for being dynamically loaded via dlopen. // GOT[e0] is the module index, with a special value of 0 for the current // module. GOT[e1] is unused. There only needs to be one module index entry. - if (Expr == R_TLSLD_PC && In::Got->addTlsIndex()) { - AddTlsReloc(In::Got->getTlsIndexOff(), Target->TlsModuleIndexRel, + if (Expr == R_TLSLD_PC && InX::Got->addTlsIndex()) { + AddTlsReloc(InX::Got->getTlsIndexOff(), Target->TlsModuleIndexRel, NeedDynId ? nullptr : &Body, NeedDynId); C.Relocations.push_back({Expr, Type, Offset, Addend, &Body}); return 1; @@ -176,8 +176,8 @@ static unsigned handleARMTlsRelocation(uint32_t Type, SymbolBody &Body, // the module index and offset of symbol in TLS block we can fill these in // using static GOT relocations. if (Expr == R_TLSGD_PC) { - if (In::Got->addDynTlsEntry(Body)) { - uint64_t Off = In::Got->getGlobalDynOffset(Body); + if (InX::Got->addDynTlsEntry(Body)) { + uint64_t Off = InX::Got->getGlobalDynOffset(Body); AddTlsReloc(Off, Target->TlsModuleIndexRel, &Body, NeedDynId); AddTlsReloc(Off + Config->Wordsize, Target->TlsOffsetRel, &Body, NeedDynOff); @@ -207,10 +207,10 @@ handleTlsRelocation(uint32_t Type, SymbolBody &Body, InputSectionBase &C, bool IsPreemptible = isPreemptible(Body, Type); if (isRelExprOneOf(Expr) && Config->Shared) { - if (In::Got->addDynTlsEntry(Body)) { - uint64_t Off = In::Got->getGlobalDynOffset(Body); - In::RelaDyn->addReloc({Target->TlsDescRel, In::Got, Off, - !IsPreemptible, &Body, 0}); + if (InX::Got->addDynTlsEntry(Body)) { + uint64_t Off = InX::Got->getGlobalDynOffset(Body); + In::RelaDyn->addReloc( + {Target->TlsDescRel, InX::Got, Off, !IsPreemptible, &Body, 0}); } if (Expr != R_TLSDESC_CALL) C.Relocations.push_back({Expr, Type, Offset, Addend, &Body}); @@ -224,10 +224,10 @@ handleTlsRelocation(uint32_t Type, SymbolBody &Body, InputSectionBase &C, {R_RELAX_TLS_LD_TO_LE, Type, Offset, Addend, &Body}); return 2; } - if (In::Got->addTlsIndex()) - In::RelaDyn->addReloc({Target->TlsModuleIndexRel, In::Got, - In::Got->getTlsIndexOff(), false, - nullptr, 0}); + if (InX::Got->addTlsIndex()) + In::RelaDyn->addReloc({Target->TlsModuleIndexRel, InX::Got, + InX::Got->getTlsIndexOff(), false, nullptr, + 0}); C.Relocations.push_back({Expr, Type, Offset, Addend, &Body}); return 1; } @@ -242,19 +242,19 @@ handleTlsRelocation(uint32_t Type, SymbolBody &Body, InputSectionBase &C, if (isRelExprOneOf(Expr)) { if (Config->Shared) { - if (In::Got->addDynTlsEntry(Body)) { - uint64_t Off = In::Got->getGlobalDynOffset(Body); + if (InX::Got->addDynTlsEntry(Body)) { + uint64_t Off = InX::Got->getGlobalDynOffset(Body); In::RelaDyn->addReloc( - {Target->TlsModuleIndexRel, In::Got, Off, false, &Body, 0}); + {Target->TlsModuleIndexRel, InX::Got, Off, false, &Body, 0}); // If the symbol is preemptible we need the dynamic linker to write // the offset too. uint64_t OffsetOff = Off + Config->Wordsize; if (IsPreemptible) - In::RelaDyn->addReloc({Target->TlsOffsetRel, In::Got, - OffsetOff, false, &Body, 0}); + In::RelaDyn->addReloc( + {Target->TlsOffsetRel, InX::Got, OffsetOff, false, &Body, 0}); else - In::Got->Relocations.push_back( + InX::Got->Relocations.push_back( {R_ABS, Target->TlsOffsetRel, OffsetOff, 0, &Body}); } C.Relocations.push_back({Expr, Type, Offset, Addend, &Body}); @@ -268,8 +268,8 @@ handleTlsRelocation(uint32_t Type, SymbolBody &Body, InputSectionBase &C, {Target->adjustRelaxExpr(Type, nullptr, R_RELAX_TLS_GD_TO_IE), Type, Offset, Addend, &Body}); if (!Body.isInGot()) { - In::Got->addEntry(Body); - In::RelaDyn->addReloc({Target->TlsGotRel, In::Got, + InX::Got->addEntry(Body); + In::RelaDyn->addReloc({Target->TlsGotRel, InX::Got, Body.getGotOffset(), false, &Body, 0}); } } else { @@ -518,7 +518,7 @@ template static void addCopyRelSymbol(SharedSymbol *SS) { // See if this symbol is in a read-only segment. If so, preserve the symbol's // memory protection by reserving space in the .bss.rel.ro section. bool IsReadOnly = isReadOnly(SS); - BssSection *Sec = IsReadOnly ? In::BssRelRo : In::Bss; + BssSection *Sec = IsReadOnly ? InX::BssRelRo : InX::Bss; uint64_t Off = Sec->reserveSpace(SymSize, SS->getAlignment()); // Look through the DSO's dynamic symbol table for aliases and create a @@ -774,7 +774,7 @@ static void addPltEntry(PltSection *Plt, GotPltSection *GotPlt, template static void addGotEntry(SymbolBody &Sym, bool Preemptible) { - In::Got->addEntry(Sym); + InX::Got->addEntry(Sym); uint64_t Off = Sym.getGotOffset(); uint32_t DynType; @@ -792,10 +792,10 @@ static void addGotEntry(SymbolBody &Sym, bool Preemptible) { bool Constant = !Preemptible && !(Config->Pic && !isAbsolute(Sym)); if (!Constant) In::RelaDyn->addReloc( - {DynType, In::Got, Off, !Preemptible, &Sym, 0}); + {DynType, InX::Got, Off, !Preemptible, &Sym, 0}); if (Constant || (!Config->IsRela && !Preemptible)) - In::Got->Relocations.push_back({Expr, DynType, Off, 0, &Sym}); + InX::Got->Relocations.push_back({Expr, DynType, Off, 0, &Sym}); } // The reason we have to do this early scan is as follows @@ -856,7 +856,7 @@ static void scanRelocs(InputSectionBase &Sec, ArrayRef Rels) { // needs it to be created. Here we request for that. if (isRelExprOneOf(Expr)) - In::Got->HasGotOffRel = true; + InX::Got->HasGotOffRel = true; // Read an addend. int64_t Addend = computeAddend(Rel, Sec.Data.data()); @@ -874,11 +874,11 @@ static void scanRelocs(InputSectionBase &Sec, ArrayRef Rels) { // If a relocation needs PLT, we create PLT and GOTPLT slots for the symbol. if (needsPlt(Expr) && !Body.isInPlt()) { if (Body.isGnuIFunc() && !Preemptible) - addPltEntry(InX::Iplt, In::IgotPlt, In::RelaIplt, + addPltEntry(InX::Iplt, InX::IgotPlt, In::RelaIplt, Target->IRelativeRel, Body, true); else - addPltEntry(InX::Plt, In::GotPlt, In::RelaPlt, - Target->PltRel, Body, !Preemptible); + addPltEntry(InX::Plt, InX::GotPlt, In::RelaPlt, Target->PltRel, + Body, !Preemptible); } // Create a GOT slot if a relocation needs GOT. @@ -891,9 +891,9 @@ static void scanRelocs(InputSectionBase &Sec, ArrayRef Rels) { // See "Global Offset Table" in Chapter 5 in the following document // for detailed description: // ftp://www.linux-mips.org/pub/linux/mips/doc/ABI/mipsabi.pdf - In::MipsGot->addEntry(Body, Addend, Expr); + InX::MipsGot->addEntry(Body, Addend, Expr); if (Body.isTls() && Body.isPreemptible()) - In::RelaDyn->addReloc({Target->TlsGotRel, In::MipsGot, + In::RelaDyn->addReloc({Target->TlsGotRel, InX::MipsGot, Body.getGotOffset(), false, &Body, 0}); } else if (!Body.isInGot()) { addGotEntry(Body, Preemptible); @@ -927,7 +927,7 @@ static void scanRelocs(InputSectionBase &Sec, ArrayRef Rels) { // a dynamic relocation. // ftp://www.linux-mips.org/pub/linux/mips/doc/ABI/mipsabi.pdf p.4-19 if (Config->EMachine == EM_MIPS) - In::MipsGot->addEntry(Body, Addend, Expr); + InX::MipsGot->addEntry(Body, Addend, Expr); continue; } diff --git a/ELF/ScriptParser.cpp b/ELF/ScriptParser.cpp index 032ecd50f3e3..f1bc245c9256 100644 --- a/ELF/ScriptParser.cpp +++ b/ELF/ScriptParser.cpp @@ -639,7 +639,7 @@ ScriptParser::readOutputSectionDescription(StringRef OutSec) { // We are compatible with ld.gold because it's easier to implement. uint32_t ScriptParser::parseFill(StringRef Tok) { uint32_t V = 0; - if (Tok.getAsInteger(0, V)) + if (!to_integer(Tok, V)) setError("invalid filler expression: " + Tok); uint32_t Buf; @@ -778,23 +778,23 @@ static Optional parseInt(StringRef Tok) { // Hexadecimal uint64_t Val; - if (Tok.startswith_lower("0x") && !Tok.substr(2).getAsInteger(16, Val)) + if (Tok.startswith_lower("0x") && to_integer(Tok.substr(2), Val, 16)) return Val; - if (Tok.endswith_lower("H") && !Tok.drop_back().getAsInteger(16, Val)) + if (Tok.endswith_lower("H") && to_integer(Tok.drop_back(), Val, 16)) return Val; // Decimal if (Tok.endswith_lower("K")) { - if (Tok.drop_back().getAsInteger(10, Val)) + if (!to_integer(Tok.drop_back(), Val, 10)) return None; return Val * 1024; } if (Tok.endswith_lower("M")) { - if (Tok.drop_back().getAsInteger(10, Val)) + if (!to_integer(Tok.drop_back(), Val, 10)) return None; return Val * 1024 * 1024; } - if (Tok.getAsInteger(10, Val)) + if (!to_integer(Tok, Val, 10)) return None; return Val; } @@ -900,10 +900,22 @@ Expr ScriptParser::readPrimary() { StringRef Name = readParenLiteral(); return [=] { return Script->isDefined(Name) ? 1 : 0; }; } + if (Tok == "LENGTH") { + StringRef Name = readParenLiteral(); + if (Script->Opt.MemoryRegions.count(Name) == 0) + setError("memory region not defined: " + Name); + return [=] { return Script->Opt.MemoryRegions[Name].Length; }; + } if (Tok == "LOADADDR") { StringRef Name = readParenLiteral(); return [=] { return Script->getOutputSection(Location, Name)->getLMA(); }; } + if (Tok == "ORIGIN") { + StringRef Name = readParenLiteral(); + if (Script->Opt.MemoryRegions.count(Name) == 0) + setError("memory region not defined: " + Name); + return [=] { return Script->Opt.MemoryRegions[Name].Origin; }; + } if (Tok == "SEGMENT_START") { expect("("); skip(); diff --git a/ELF/Strings.cpp b/ELF/Strings.cpp index 29760b492ba9..2e88bfba0fc1 100644 --- a/ELF/Strings.cpp +++ b/ELF/Strings.cpp @@ -46,7 +46,7 @@ int elf::getPriority(StringRef S) { if (Pos == StringRef::npos) return 65536; int V; - if (S.substr(Pos + 1).getAsInteger(10, V)) + if (!to_integer(S.substr(Pos + 1), V, 10)) return 65536; return V; } @@ -68,7 +68,7 @@ std::vector elf::parseHex(StringRef S) { StringRef B = S.substr(0, 2); S = S.substr(2); uint8_t H; - if (B.getAsInteger(16, H)) { + if (!to_integer(B, H, 16)) { error("not a hexadecimal value: " + B); return {}; } diff --git a/ELF/Symbols.cpp b/ELF/Symbols.cpp index 2090b33e8cd6..7ce1f5354b1b 100644 --- a/ELF/Symbols.cpp +++ b/ELF/Symbols.cpp @@ -163,8 +163,8 @@ uint64_t SymbolBody::getVA(int64_t Addend) const { return OutVA + Addend; } -template typename ELFT::uint SymbolBody::getGotVA() const { - return In::Got->getVA() + getGotOffset(); +uint64_t SymbolBody::getGotVA() const { + return InX::Got->getVA() + getGotOffset(); } uint64_t SymbolBody::getGotOffset() const { @@ -370,11 +370,6 @@ std::string lld::toString(const SymbolBody &B) { return B.getName(); } -template uint32_t SymbolBody::template getGotVA() const; -template uint32_t SymbolBody::template getGotVA() const; -template uint64_t SymbolBody::template getGotVA() const; -template uint64_t SymbolBody::template getGotVA() const; - template uint32_t SymbolBody::template getSize() const; template uint32_t SymbolBody::template getSize() const; template uint64_t SymbolBody::template getSize() const; diff --git a/ELF/Symbols.h b/ELF/Symbols.h index 39a0c0f7b4df..030527f63744 100644 --- a/ELF/Symbols.h +++ b/ELF/Symbols.h @@ -78,7 +78,7 @@ class SymbolBody { uint64_t getVA(int64_t Addend = 0) const; uint64_t getGotOffset() const; - template typename ELFT::uint getGotVA() const; + uint64_t getGotVA() const; uint64_t getGotPltOffset() const; uint64_t getGotPltVA() const; uint64_t getPltVA() const; diff --git a/ELF/SyntheticSections.cpp b/ELF/SyntheticSections.cpp index 9c585e41e9f0..5a2c2c37efd8 100644 --- a/ELF/SyntheticSections.cpp +++ b/ELF/SyntheticSections.cpp @@ -186,7 +186,7 @@ template void MipsOptionsSection::writeTo(uint8_t *Buf) { Options->size = getSize(); if (!Config->Relocatable) - Reginfo.ri_gp_value = In::MipsGot->getGp(); + Reginfo.ri_gp_value = InX::MipsGot->getGp(); memcpy(Buf + sizeof(Elf_Mips_Options), &Reginfo, sizeof(Reginfo)); } @@ -244,7 +244,7 @@ MipsReginfoSection::MipsReginfoSection(Elf_Mips_RegInfo Reginfo) template void MipsReginfoSection::writeTo(uint8_t *Buf) { if (!Config->Relocatable) - Reginfo.ri_gp_value = In::MipsGot->getGp(); + Reginfo.ri_gp_value = InX::MipsGot->getGp(); memcpy(Buf, &Reginfo, sizeof(Reginfo)); } @@ -293,13 +293,12 @@ InputSection *elf::createInterpSection() { return Sec; } -template SymbolBody *elf::addSyntheticLocal(StringRef Name, uint8_t Type, uint64_t Value, uint64_t Size, InputSectionBase *Section) { auto *S = make(Name, /*IsLocal*/ true, STV_DEFAULT, Type, Value, Size, Section, nullptr); - if (In::SymTab) - In::SymTab->addSymbol(S); + if (InX::SymTab) + InX::SymTab->addSymbol(S); return S; } @@ -356,7 +355,7 @@ void BuildIdSection::computeHash( std::vector Hashes(Chunks.size() * HashSize); // Compute hash values. - parallelFor(0, Chunks.size(), [&](size_t I) { + parallelForEachN(0, Chunks.size(), [&](size_t I) { HashFn(Hashes.data() + I * HashSize, Chunks[I]); }); @@ -618,17 +617,16 @@ template void EhFrameSection::writeTo(uint8_t *Buf) { } } -template -GotSection::GotSection() +GotBaseSection::GotBaseSection() : SyntheticSection(SHF_ALLOC | SHF_WRITE, SHT_PROGBITS, Target->GotEntrySize, ".got") {} -template void GotSection::addEntry(SymbolBody &Sym) { +void GotBaseSection::addEntry(SymbolBody &Sym) { Sym.GotIndex = NumEntries; ++NumEntries; } -template bool GotSection::addDynTlsEntry(SymbolBody &Sym) { +bool GotBaseSection::addDynTlsEntry(SymbolBody &Sym) { if (Sym.GlobalDynIndex != -1U) return false; Sym.GlobalDynIndex = NumEntries; @@ -639,7 +637,7 @@ template bool GotSection::addDynTlsEntry(SymbolBody &Sym) { // Reserves TLS entries for a TLS module ID and a TLS block offset. // In total it takes two GOT slots. -template bool GotSection::addTlsIndex() { +bool GotBaseSection::addTlsIndex() { if (TlsIndexOff != uint32_t(-1)) return false; TlsIndexOff = NumEntries * Config->Wordsize; @@ -647,21 +645,19 @@ template bool GotSection::addTlsIndex() { return true; } -template -uint64_t GotSection::getGlobalDynAddr(const SymbolBody &B) const { +uint64_t GotBaseSection::getGlobalDynAddr(const SymbolBody &B) const { return this->getVA() + B.GlobalDynIndex * Config->Wordsize; } -template -uint64_t GotSection::getGlobalDynOffset(const SymbolBody &B) const { +uint64_t GotBaseSection::getGlobalDynOffset(const SymbolBody &B) const { return B.GlobalDynIndex * Config->Wordsize; } -template void GotSection::finalizeContents() { +void GotBaseSection::finalizeContents() { Size = NumEntries * Config->Wordsize; } -template bool GotSection::empty() const { +bool GotBaseSection::empty() const { // If we have a relocation that is relative to GOT (such as GOTOFFREL), // we need to emit a GOT even if it's empty. return NumEntries == 0 && !HasGotOffRel; @@ -1028,24 +1024,15 @@ template void DynamicSection::addEntries() { // Add strings to .dynstr early so that .dynstr's size will be // fixed early. for (StringRef S : Config->AuxiliaryList) - add({DT_AUXILIARY, In::DynStrTab->addString(S)}); + add({DT_AUXILIARY, InX::DynStrTab->addString(S)}); if (!Config->Rpath.empty()) add({Config->EnableNewDtags ? DT_RUNPATH : DT_RPATH, - In::DynStrTab->addString(Config->Rpath)}); + InX::DynStrTab->addString(Config->Rpath)}); for (SharedFile *F : Symtab::X->getSharedFiles()) if (F->isNeeded()) - add({DT_NEEDED, In::DynStrTab->addString(F->SoName)}); + add({DT_NEEDED, InX::DynStrTab->addString(F->SoName)}); if (!Config->SoName.empty()) - add({DT_SONAME, In::DynStrTab->addString(Config->SoName)}); - - if (!Config->Shared && !Config->Relocatable) - add({DT_DEBUG, (uint64_t)0}); -} - -// Add remaining entries to complete .dynamic contents. -template void DynamicSection::finalizeContents() { - if (this->Size) - return; // Already finalized. + add({DT_SONAME, InX::DynStrTab->addString(Config->SoName)}); // Set DT_FLAGS and DT_FLAGS_1. uint32_t DtFlags = 0; @@ -1064,15 +1051,22 @@ template void DynamicSection::finalizeContents() { DtFlags |= DF_ORIGIN; DtFlags1 |= DF_1_ORIGIN; } - if (Config->HasStaticTlsModel) - DtFlags |= DF_STATIC_TLS; if (DtFlags) add({DT_FLAGS, DtFlags}); if (DtFlags1) add({DT_FLAGS_1, DtFlags1}); - this->Link = In::DynStrTab->OutSec->SectionIndex; + if (!Config->Shared && !Config->Relocatable) + add({DT_DEBUG, (uint64_t)0}); +} + +// Add remaining entries to complete .dynamic contents. +template void DynamicSection::finalizeContents() { + if (this->Size) + return; // Already finalized. + + this->Link = InX::DynStrTab->OutSec->SectionIndex; if (In::RelaDyn->OutSec->Size > 0) { bool IsRela = Config->IsRela; add({IsRela ? DT_RELA : DT_REL, In::RelaDyn}); @@ -1093,18 +1087,18 @@ template void DynamicSection::finalizeContents() { add({DT_JMPREL, In::RelaPlt}); add({DT_PLTRELSZ, In::RelaPlt->OutSec->Size}); add({Config->EMachine == EM_MIPS ? DT_MIPS_PLTGOT : DT_PLTGOT, - In::GotPlt}); + InX::GotPlt}); add({DT_PLTREL, uint64_t(Config->IsRela ? DT_RELA : DT_REL)}); } - add({DT_SYMTAB, In::DynSymTab}); + add({DT_SYMTAB, InX::DynSymTab}); add({DT_SYMENT, sizeof(Elf_Sym)}); - add({DT_STRTAB, In::DynStrTab}); - add({DT_STRSZ, In::DynStrTab->getSize()}); + add({DT_STRTAB, InX::DynStrTab}); + add({DT_STRSZ, InX::DynStrTab->getSize()}); if (!Config->ZText) add({DT_TEXTREL, (uint64_t)0}); - if (In::GnuHashTab) - add({DT_GNU_HASH, In::GnuHashTab}); + if (InX::GnuHashTab) + add({DT_GNU_HASH, InX::GnuHashTab}); if (In::HashTab) add({DT_HASH, In::HashTab}); @@ -1142,15 +1136,15 @@ template void DynamicSection::finalizeContents() { add({DT_MIPS_RLD_VERSION, 1}); add({DT_MIPS_FLAGS, RHF_NOTPOT}); add({DT_MIPS_BASE_ADDRESS, Config->ImageBase}); - add({DT_MIPS_SYMTABNO, In::DynSymTab->getNumSymbols()}); - add({DT_MIPS_LOCAL_GOTNO, In::MipsGot->getLocalEntriesNum()}); - if (const SymbolBody *B = In::MipsGot->getFirstGlobalEntry()) + add({DT_MIPS_SYMTABNO, InX::DynSymTab->getNumSymbols()}); + add({DT_MIPS_LOCAL_GOTNO, InX::MipsGot->getLocalEntriesNum()}); + if (const SymbolBody *B = InX::MipsGot->getFirstGlobalEntry()) add({DT_MIPS_GOTSYM, B->DynsymIndex}); else - add({DT_MIPS_GOTSYM, In::DynSymTab->getNumSymbols()}); - add({DT_PLTGOT, In::MipsGot}); - if (In::MipsRldMap) - add({DT_MIPS_RLD_MAP, In::MipsRldMap}); + add({DT_MIPS_GOTSYM, InX::DynSymTab->getNumSymbols()}); + add({DT_PLTGOT, InX::MipsGot}); + if (InX::MipsRldMap) + add({DT_MIPS_RLD_MAP, InX::MipsRldMap}); } this->OutSec->Link = this->Link; @@ -1235,11 +1229,11 @@ template void RelocationSection::writeTo(uint8_t *Buf) { if (Config->IsRela) P->r_addend = Rel.getAddend(); P->r_offset = Rel.getOffset(); - if (Config->EMachine == EM_MIPS && Rel.getInputSec() == In::MipsGot) + if (Config->EMachine == EM_MIPS && Rel.getInputSec() == InX::MipsGot) // Dynamic relocation against MIPS GOT section make deal TLS entries // allocated in the end of the GOT. We need to adjust the offset to take // in account 'local' and 'global' GOT entries. - P->r_offset += In::MipsGot->getTlsOffset(); + P->r_offset += InX::MipsGot->getTlsOffset(); P->setSymbolAndType(Rel.getSymIndex(), Rel.Type, Config->IsMips64EL); } @@ -1259,22 +1253,19 @@ template unsigned RelocationSection::getRelocOffset() { } template void RelocationSection::finalizeContents() { - this->Link = In::DynSymTab ? In::DynSymTab->OutSec->SectionIndex - : In::SymTab->OutSec->SectionIndex; + this->Link = InX::DynSymTab ? InX::DynSymTab->OutSec->SectionIndex + : InX::SymTab->OutSec->SectionIndex; // Set required output section properties. this->OutSec->Link = this->Link; } -template -SymbolTableSection::SymbolTableSection(StringTableSection &StrTabSec) +SymbolTableBaseSection::SymbolTableBaseSection(StringTableSection &StrTabSec) : SyntheticSection(StrTabSec.isDynamic() ? (uint64_t)SHF_ALLOC : 0, StrTabSec.isDynamic() ? SHT_DYNSYM : SHT_SYMTAB, Config->Wordsize, StrTabSec.isDynamic() ? ".dynsym" : ".symtab"), - StrTabSec(StrTabSec) { - this->Entsize = sizeof(Elf_Sym); -} + StrTabSec(StrTabSec) {} // Orders symbols according to their positions in the GOT, // in compliance with MIPS ABI rules. @@ -1296,7 +1287,7 @@ static bool sortMipsSymbols(const SymbolTableEntry &L, // symbols precede global symbols, so we sort symbol entries in this // function. (For .dynsym, we don't do that because symbols for // dynamic linking are inherently all globals.) -template void SymbolTableSection::finalizeContents() { +void SymbolTableBaseSection::finalizeContents() { this->OutSec->Link = StrTabSec.OutSec->SectionIndex; // If it is a .dynsym, there should be no local symbols, but we need @@ -1306,9 +1297,9 @@ template void SymbolTableSection::finalizeContents() { // Because the first symbol entry is a null entry, 1 is the first. this->OutSec->Info = 1; - if (In::GnuHashTab) { + if (InX::GnuHashTab) { // NB: It also sorts Symbols to meet the GNU hash table requirements. - In::GnuHashTab->addSymbols(Symbols); + InX::GnuHashTab->addSymbols(Symbols); } else if (Config->EMachine == EM_MIPS) { std::stable_sort(Symbols.begin(), Symbols.end(), sortMipsSymbols); } @@ -1320,7 +1311,7 @@ template void SymbolTableSection::finalizeContents() { } } -template void SymbolTableSection::postThunkContents() { +void SymbolTableBaseSection::postThunkContents() { if (this->Type == SHT_DYNSYM) return; // move all local symbols before global symbols. @@ -1333,7 +1324,7 @@ template void SymbolTableSection::postThunkContents() { this->OutSec->Info = NumLocals + 1; } -template void SymbolTableSection::addSymbol(SymbolBody *B) { +void SymbolTableBaseSection::addSymbol(SymbolBody *B) { // Adding a local symbol to a .dynsym is a bug. assert(this->Type != SHT_DYNSYM || !B->isLocal()); @@ -1341,8 +1332,7 @@ template void SymbolTableSection::addSymbol(SymbolBody *B) { Symbols.push_back({B, StrTabSec.addString(B->getName(), HashIt)}); } -template -size_t SymbolTableSection::getSymbolIndex(SymbolBody *Body) { +size_t SymbolTableBaseSection::getSymbolIndex(SymbolBody *Body) { auto I = llvm::find_if(Symbols, [&](const SymbolTableEntry &E) { if (E.Symbol == Body) return true; @@ -1358,6 +1348,12 @@ size_t SymbolTableSection::getSymbolIndex(SymbolBody *Body) { return I - Symbols.begin() + 1; } +template +SymbolTableSection::SymbolTableSection(StringTableSection &StrTabSec) + : SymbolTableBaseSection(StrTabSec) { + this->Entsize = sizeof(Elf_Sym); +} + // Write the internal symbol table contents to the output symbol table. template void SymbolTableSection::writeTo(uint8_t *Buf) { // The first entry is a null entry as per the ELF spec. @@ -1450,13 +1446,12 @@ template void SymbolTableSection::writeTo(uint8_t *Buf) { // DSOs very quickly. If you are sure that your dynamic linker knows // about .gnu.hash, you want to specify -hash-style=gnu. Otherwise, a // safe bet is to specify -hash-style=both for backward compatibilty. -template -GnuHashTableSection::GnuHashTableSection() +GnuHashTableSection::GnuHashTableSection() : SyntheticSection(SHF_ALLOC, SHT_GNU_HASH, Config->Wordsize, ".gnu.hash") { } -template void GnuHashTableSection::finalizeContents() { - this->OutSec->Link = In::DynSymTab->OutSec->SectionIndex; +void GnuHashTableSection::finalizeContents() { + this->OutSec->Link = InX::DynSymTab->OutSec->SectionIndex; // Computes bloom filter size in word size. We want to allocate 8 // bits for each symbol. It must be a power of two. @@ -1471,11 +1466,10 @@ template void GnuHashTableSection::finalizeContents() { Size += Symbols.size() * 4; // Hash values } -template -void GnuHashTableSection::writeTo(uint8_t *Buf) { +void GnuHashTableSection::writeTo(uint8_t *Buf) { // Write a header. write32(Buf, NBuckets, Config->Endianness); - write32(Buf + 4, In::DynSymTab->getNumSymbols() - Symbols.size(), + write32(Buf + 4, InX::DynSymTab->getNumSymbols() - Symbols.size(), Config->Endianness); write32(Buf + 8, MaskWords, Config->Endianness); write32(Buf + 12, getShift2(), Config->Endianness); @@ -1494,8 +1488,7 @@ void GnuHashTableSection::writeTo(uint8_t *Buf) { // // [1] Ulrich Drepper (2011), "How To Write Shared Libraries" (Ver. 4.1.2), // p.9, https://www.akkadia.org/drepper/dsohowto.pdf -template -void GnuHashTableSection::writeBloomFilter(uint8_t *Buf) { +void GnuHashTableSection::writeBloomFilter(uint8_t *Buf) { const unsigned C = Config->Wordsize * 8; for (const Entry &Sym : Symbols) { size_t I = (Sym.Hash / C) & (MaskWords - 1); @@ -1506,8 +1499,7 @@ void GnuHashTableSection::writeBloomFilter(uint8_t *Buf) { } } -template -void GnuHashTableSection::writeHashTable(uint8_t *Buf) { +void GnuHashTableSection::writeHashTable(uint8_t *Buf) { // Group symbols by hash value. std::vector> Syms(NBuckets); for (const Entry &Ent : Symbols) @@ -1560,8 +1552,7 @@ static size_t getBucketSize(size_t NumSymbols) { // Add symbols to this symbol hash table. Note that this function // destructively sort a given vector -- which is needed because // GNU-style hash table places some sorting requirements. -template -void GnuHashTableSection::addSymbols(std::vector &V) { +void GnuHashTableSection::addSymbols(std::vector &V) { // We cannot use 'auto' for Mid because GCC 6.1 cannot deduce // its type correctly. std::vector::iterator Mid = @@ -1594,15 +1585,15 @@ HashTableSection::HashTableSection() } template void HashTableSection::finalizeContents() { - this->OutSec->Link = In::DynSymTab->OutSec->SectionIndex; + this->OutSec->Link = InX::DynSymTab->OutSec->SectionIndex; unsigned NumEntries = 2; // nbucket and nchain. - NumEntries += In::DynSymTab->getNumSymbols(); // The chain entries. + NumEntries += InX::DynSymTab->getNumSymbols(); // The chain entries. // Create as many buckets as there are symbols. // FIXME: This is simplistic. We can try to optimize it, but implementing // support for SHT_GNU_HASH is probably even more profitable. - NumEntries += In::DynSymTab->getNumSymbols(); + NumEntries += InX::DynSymTab->getNumSymbols(); this->Size = NumEntries * 4; } @@ -1610,7 +1601,7 @@ template void HashTableSection::writeTo(uint8_t *Buf) { // A 32-bit integer type in the target endianness. typedef typename ELFT::Word Elf_Word; - unsigned NumSymbols = In::DynSymTab->getNumSymbols(); + unsigned NumSymbols = InX::DynSymTab->getNumSymbols(); auto *P = reinterpret_cast(Buf); *P++ = NumSymbols; // nbucket @@ -1619,7 +1610,7 @@ template void HashTableSection::writeTo(uint8_t *Buf) { Elf_Word *Buckets = P; Elf_Word *Chains = P + NumSymbols; - for (const SymbolTableEntry &S : In::DynSymTab->getSymbols()) { + for (const SymbolTableEntry &S : InX::DynSymTab->getSymbols()) { SymbolBody *Body = S.Symbol; StringRef Name = Body->getName(); unsigned I = Body->DynsymIndex; @@ -1706,13 +1697,14 @@ readCuList(DWARFContext &Dwarf, InputSection *Sec) { return Ret; } -static InputSectionBase *findSection(ArrayRef Arr, - uint64_t Offset) { +static InputSection *findSection(ArrayRef Arr, + uint64_t Offset) { for (InputSectionBase *S : Arr) - if (S && S != &InputSection::Discarded) - if (Offset >= S->getOffsetInFile() && - Offset < S->getOffsetInFile() + S->getSize()) - return S; + if (auto *IS = dyn_cast_or_null(S)) + if (IS != &InputSection::Discarded && IS->Live && + Offset >= IS->getOffsetInFile() && + Offset < IS->getOffsetInFile() + IS->getSize()) + return IS; return nullptr; } @@ -1725,10 +1717,10 @@ readAddressArea(DWARFContext &Dwarf, InputSection *Sec, size_t CurrentCU) { CU->collectAddressRanges(Ranges); ArrayRef Sections = Sec->File->getSections(); - for (std::pair &R : Ranges) - if (InputSectionBase *S = findSection(Sections, R.first)) - Ret.push_back({S, R.first - S->getOffsetInFile(), - R.second - S->getOffsetInFile(), CurrentCU}); + for (DWARFAddressRange &R : Ranges) + if (InputSection *S = findSection(Sections, R.LowPC)) + Ret.push_back({S, R.LowPC - S->getOffsetInFile(), + R.HighPC - S->getOffsetInFile(), CurrentCU}); ++CurrentCU; } return Ret; @@ -1951,11 +1943,11 @@ static StringRef getFileDefName() { } template void VersionDefinitionSection::finalizeContents() { - FileDefNameOff = In::DynStrTab->addString(getFileDefName()); + FileDefNameOff = InX::DynStrTab->addString(getFileDefName()); for (VersionDefinition &V : Config->VersionDefinitions) - V.NameOff = In::DynStrTab->addString(V.Name); + V.NameOff = InX::DynStrTab->addString(V.Name); - this->OutSec->Link = In::DynStrTab->OutSec->SectionIndex; + this->OutSec->Link = InX::DynStrTab->OutSec->SectionIndex; // sh_info should be set to the number of definitions. This fact is missed in // documentation, but confirmed by binutils community: @@ -2008,16 +2000,16 @@ VersionTableSection::VersionTableSection() template void VersionTableSection::finalizeContents() { // At the moment of june 2016 GNU docs does not mention that sh_link field // should be set, but Sun docs do. Also readelf relies on this field. - this->OutSec->Link = In::DynSymTab->OutSec->SectionIndex; + this->OutSec->Link = InX::DynSymTab->OutSec->SectionIndex; } template size_t VersionTableSection::getSize() const { - return sizeof(Elf_Versym) * (In::DynSymTab->getSymbols().size() + 1); + return sizeof(Elf_Versym) * (InX::DynSymTab->getSymbols().size() + 1); } template void VersionTableSection::writeTo(uint8_t *Buf) { auto *OutVersym = reinterpret_cast(Buf) + 1; - for (const SymbolTableEntry &S : In::DynSymTab->getSymbols()) { + for (const SymbolTableEntry &S : InX::DynSymTab->getSymbols()) { OutVersym->vs_index = S.Symbol->symbol()->VersionId; ++OutVersym; } @@ -2051,14 +2043,14 @@ void VersionNeedSection::addSymbol(SharedSymbol *SS) { // to create one by adding it to our needed list and creating a dynstr entry // for the soname. if (File->VerdefMap.empty()) - Needed.push_back({File, In::DynStrTab->addString(File->SoName)}); + Needed.push_back({File, InX::DynStrTab->addString(File->SoName)}); typename SharedFile::NeededVer &NV = File->VerdefMap[Ver]; // If we don't already know that we need an Elf_Vernaux for this Elf_Verdef, // prepare to create one by allocating a version identifier and creating a // dynstr entry for the version name. if (NV.Index == 0) { - NV.StrTab = In::DynStrTab->addString(File->getStringTable().data() + - Ver->getAux()->vda_name); + NV.StrTab = InX::DynStrTab->addString(File->getStringTable().data() + + Ver->getAux()->vda_name); NV.Index = NextIndex++; } SS->symbol()->VersionId = NV.Index; @@ -2100,7 +2092,7 @@ template void VersionNeedSection::writeTo(uint8_t *Buf) { } template void VersionNeedSection::finalizeContents() { - this->OutSec->Link = In::DynStrTab->OutSec->SectionIndex; + this->OutSec->Link = InX::DynStrTab->OutSec->SectionIndex; this->OutSec->Info = Needed.size(); } @@ -2187,7 +2179,7 @@ MipsRldMapSection::MipsRldMapSection() void MipsRldMapSection::writeTo(uint8_t *Buf) { // Apply filler from linker script. - Optional Fill = Script->getFiller(this->Name); + Optional Fill = Script->getFiller(this->OutSec); if (!Fill || *Fill == 0) return; @@ -2245,10 +2237,14 @@ BssSection *InX::Bss; BssSection *InX::BssRelRo; BuildIdSection *InX::BuildId; InputSection *InX::Common; +SyntheticSection *InX::Dynamic; StringTableSection *InX::DynStrTab; +SymbolTableBaseSection *InX::DynSymTab; InputSection *InX::Interp; GdbIndexSection *InX::GdbIndex; +GotBaseSection *InX::Got; GotPltSection *InX::GotPlt; +GnuHashTableSection *InX::GnuHashTab; IgotPltSection *InX::IgotPlt; MipsGotSection *InX::MipsGot; MipsRldMapSection *InX::MipsRldMap; @@ -2256,6 +2252,7 @@ PltSection *InX::Plt; PltSection *InX::Iplt; StringTableSection *InX::ShStrTab; StringTableSection *InX::StrTab; +SymbolTableBaseSection *InX::SymTab; template void PltSection::addEntry(SymbolBody &Sym); template void PltSection::addEntry(SymbolBody &Sym); @@ -2272,19 +2269,6 @@ template MergeInputSection *elf::createCommentSection(); template MergeInputSection *elf::createCommentSection(); template MergeInputSection *elf::createCommentSection(); -template SymbolBody *elf::addSyntheticLocal(StringRef, uint8_t, - uint64_t, uint64_t, - InputSectionBase *); -template SymbolBody *elf::addSyntheticLocal(StringRef, uint8_t, - uint64_t, uint64_t, - InputSectionBase *); -template SymbolBody *elf::addSyntheticLocal(StringRef, uint8_t, - uint64_t, uint64_t, - InputSectionBase *); -template SymbolBody *elf::addSyntheticLocal(StringRef, uint8_t, - uint64_t, uint64_t, - InputSectionBase *); - template class elf::MipsAbiFlagsSection; template class elf::MipsAbiFlagsSection; template class elf::MipsAbiFlagsSection; @@ -2320,11 +2304,6 @@ template class elf::SymbolTableSection; template class elf::SymbolTableSection; template class elf::SymbolTableSection; -template class elf::GnuHashTableSection; -template class elf::GnuHashTableSection; -template class elf::GnuHashTableSection; -template class elf::GnuHashTableSection; - template class elf::HashTableSection; template class elf::HashTableSection; template class elf::HashTableSection; diff --git a/ELF/SyntheticSections.h b/ELF/SyntheticSections.h index 1098c58a3baf..0477c601a7df 100644 --- a/ELF/SyntheticSections.h +++ b/ELF/SyntheticSections.h @@ -104,10 +104,9 @@ template class EhFrameSection final : public SyntheticSection { llvm::DenseMap, SymbolBody *>, CieRecord> CieMap; }; -template class GotSection final : public SyntheticSection { +class GotBaseSection : public SyntheticSection { public: - GotSection(); - void writeTo(uint8_t *Buf) override; + GotBaseSection(); size_t getSize() const override { return Size; } void finalizeContents() override; bool empty() const override; @@ -125,12 +124,17 @@ template class GotSection final : public SyntheticSection { // that relies on its address. bool HasGotOffRel = false; -private: +protected: size_t NumEntries = 0; uint32_t TlsIndexOff = -1; uint64_t Size = 0; }; +template class GotSection final : public GotBaseSection { +public: + void writeTo(uint8_t *Buf) override; +}; + // .note.gnu.build-id section. class BuildIdSection : public SyntheticSection { // First 16 bytes are a header. @@ -401,31 +405,35 @@ struct SymbolTableEntry { size_t StrTabOffset; }; -template class SymbolTableSection final : public SyntheticSection { +class SymbolTableBaseSection : public SyntheticSection { public: - typedef typename ELFT::Sym Elf_Sym; - - SymbolTableSection(StringTableSection &StrTabSec); - + SymbolTableBaseSection(StringTableSection &StrTabSec); void finalizeContents() override; void postThunkContents() override; - void writeTo(uint8_t *Buf) override; - size_t getSize() const override { return getNumSymbols() * sizeof(Elf_Sym); } + size_t getSize() const override { return getNumSymbols() * Entsize; } void addSymbol(SymbolBody *Body); unsigned getNumSymbols() const { return Symbols.size() + 1; } size_t getSymbolIndex(SymbolBody *Body); ArrayRef getSymbols() const { return Symbols; } -private: +protected: // A vector of symbols and their string table offsets. std::vector Symbols; StringTableSection &StrTabSec; }; +template +class SymbolTableSection final : public SymbolTableBaseSection { + typedef typename ELFT::Sym Elf_Sym; + +public: + SymbolTableSection(StringTableSection &StrTabSec); + void writeTo(uint8_t *Buf) override; +}; + // Outputs GNU Hash section. For detailed explanation see: // https://blogs.oracle.com/ali/entry/gnu_hash_elf_sections -template class GnuHashTableSection final : public SyntheticSection { public: GnuHashTableSection(); @@ -739,7 +747,7 @@ class ThunkSection : public SyntheticSection { template InputSection *createCommonSection(); InputSection *createInterpSection(); template MergeInputSection *createCommentSection(); -template + SymbolBody *addSyntheticLocal(StringRef Name, uint8_t Type, uint64_t Value, uint64_t Size, InputSectionBase *Section); @@ -750,9 +758,13 @@ struct InX { static BssSection *BssRelRo; static BuildIdSection *BuildId; static InputSection *Common; + static SyntheticSection *Dynamic; static StringTableSection *DynStrTab; + static SymbolTableBaseSection *DynSymTab; + static GnuHashTableSection *GnuHashTab; static InputSection *Interp; static GdbIndexSection *GdbIndex; + static GotBaseSection *Got; static GotPltSection *GotPlt; static IgotPltSection *IgotPlt; static MipsGotSection *MipsGot; @@ -761,36 +773,27 @@ struct InX { static PltSection *Iplt; static StringTableSection *ShStrTab; static StringTableSection *StrTab; + static SymbolTableBaseSection *SymTab; }; template struct In : public InX { - static DynamicSection *Dynamic; - static SymbolTableSection *DynSymTab; static EhFrameHeader *EhFrameHdr; - static GnuHashTableSection *GnuHashTab; - static GotSection *Got; static EhFrameSection *EhFrame; static HashTableSection *HashTab; static RelocationSection *RelaDyn; static RelocationSection *RelaPlt; static RelocationSection *RelaIplt; - static SymbolTableSection *SymTab; static VersionDefinitionSection *VerDef; static VersionTableSection *VerSym; static VersionNeedSection *VerNeed; }; -template DynamicSection *In::Dynamic; -template SymbolTableSection *In::DynSymTab; template EhFrameHeader *In::EhFrameHdr; -template GnuHashTableSection *In::GnuHashTab; -template GotSection *In::Got; template EhFrameSection *In::EhFrame; template HashTableSection *In::HashTab; template RelocationSection *In::RelaDyn; template RelocationSection *In::RelaPlt; template RelocationSection *In::RelaIplt; -template SymbolTableSection *In::SymTab; template VersionDefinitionSection *In::VerDef; template VersionTableSection *In::VerSym; template VersionNeedSection *In::VerNeed; diff --git a/ELF/Target.cpp b/ELF/Target.cpp index 4643c1a919aa..781d7fe3bc3f 100644 --- a/ELF/Target.cpp +++ b/ELF/Target.cpp @@ -351,15 +351,6 @@ X86TargetInfo::X86TargetInfo() { RelExpr X86TargetInfo::getRelExpr(uint32_t Type, const SymbolBody &S, const uint8_t *Loc) const { - // There are 4 different TLS variable models with varying degrees of - // flexibility and performance. LocalExec and InitialExec models are fast but - // less-flexible models. They cannot be used for dlopen(). If they are in use, - // we set DF_STATIC_TLS in the ELF header so that the runtime can reject such - // DSOs. - if (Type == R_386_TLS_LE || Type == R_386_TLS_LE_32 || Type == R_386_TLS_IE || - Type == R_386_TLS_GOTIE) - Config->HasStaticTlsModel = true; - switch (Type) { case R_386_8: case R_386_16: @@ -429,7 +420,7 @@ RelExpr X86TargetInfo::adjustRelaxExpr(uint32_t Type, const uint8_t *Data, } void X86TargetInfo::writeGotPltHeader(uint8_t *Buf) const { - write32le(Buf, In::Dynamic->getVA()); + write32le(Buf, InX::Dynamic->getVA()); } void X86TargetInfo::writeGotPlt(uint8_t *Buf, const SymbolBody &S) const { @@ -460,8 +451,8 @@ void X86TargetInfo::writePltHeader(uint8_t *Buf) const { }; memcpy(Buf, V, sizeof(V)); - uint32_t Ebx = In::Got->getVA() + In::Got->getSize(); - uint32_t GotPlt = In::GotPlt->getVA() - Ebx; + uint32_t Ebx = InX::Got->getVA() + InX::Got->getSize(); + uint32_t GotPlt = InX::GotPlt->getVA() - Ebx; write32le(Buf + 2, GotPlt + 4); write32le(Buf + 8, GotPlt + 8); return; @@ -473,7 +464,7 @@ void X86TargetInfo::writePltHeader(uint8_t *Buf) const { 0x90, 0x90, 0x90, 0x90 // nop }; memcpy(Buf, PltData, sizeof(PltData)); - uint32_t GotPlt = In::GotPlt->getVA(); + uint32_t GotPlt = InX::GotPlt->getVA(); write32le(Buf + 2, GotPlt + 4); write32le(Buf + 8, GotPlt + 8); } @@ -490,7 +481,7 @@ void X86TargetInfo::writePlt(uint8_t *Buf, uint64_t GotPltEntryAddr, if (Config->Pic) { // jmp *foo@GOT(%ebx) - uint32_t Ebx = In::Got->getVA() + In::Got->getSize(); + uint32_t Ebx = InX::Got->getVA() + InX::Got->getSize(); Buf[1] = 0xa3; write32le(Buf + 2, GotPltEntryAddr - Ebx); } else { @@ -718,7 +709,7 @@ void X86_64TargetInfo::writeGotPltHeader(uint8_t *Buf) const { // required, but it is documented in the psabi and the glibc dynamic linker // seems to use it (note that this is relevant for linking ld.so, not any // other program). - write64le(Buf, In::Dynamic->getVA()); + write64le(Buf, InX::Dynamic->getVA()); } template @@ -736,8 +727,8 @@ void X86_64TargetInfo::writePltHeader(uint8_t *Buf) const { 0x0f, 0x1f, 0x40, 0x00 // nop }; memcpy(Buf, PltData, sizeof(PltData)); - uint64_t GotPlt = In::GotPlt->getVA(); - uint64_t Plt = In::Plt->getVA(); + uint64_t GotPlt = InX::GotPlt->getVA(); + uint64_t Plt = InX::Plt->getVA(); write32le(Buf + 2, GotPlt - Plt + 2); // GOTPLT+8 write32le(Buf + 8, GotPlt - Plt + 4); // GOTPLT+16 } @@ -760,7 +751,8 @@ void X86_64TargetInfo::writePlt(uint8_t *Buf, uint64_t GotPltEntryAddr, template bool X86_64TargetInfo::isPicRel(uint32_t Type) const { - return Type != R_X86_64_PC32 && Type != R_X86_64_32; + return Type != R_X86_64_PC32 && Type != R_X86_64_32 && + Type != R_X86_64_TPOFF32; } template @@ -1140,7 +1132,7 @@ uint64_t getPPC64TocBase() { // TOC starts where the first of these sections starts. We always create a // .got when we see a relocation that uses it, so for us the start is always // the .got. - uint64_t TocVA = In::Got->getVA(); + uint64_t TocVA = InX::Got->getVA(); // Per the ppc64-elf-linux ABI, The TOC base is TOC value plus 0x8000 // thus permitting a full 64 Kbytes segment. Note that the glibc startup @@ -1369,7 +1361,7 @@ bool AArch64TargetInfo::isPicRel(uint32_t Type) const { } void AArch64TargetInfo::writeGotPlt(uint8_t *Buf, const SymbolBody &) const { - write64le(Buf, In::Plt->getVA()); + write64le(Buf, InX::Plt->getVA()); } // Page(Expr) is the page address of the expression Expr, defined @@ -1392,8 +1384,8 @@ void AArch64TargetInfo::writePltHeader(uint8_t *Buf) const { }; memcpy(Buf, PltData, sizeof(PltData)); - uint64_t Got = In::GotPlt->getVA(); - uint64_t Plt = In::Plt->getVA(); + uint64_t Got = InX::GotPlt->getVA(); + uint64_t Plt = InX::Plt->getVA(); relocateOne(Buf + 4, R_AARCH64_ADR_PREL_PG_HI21, getAArch64Page(Got + 16) - getAArch64Page(Plt + 4)); relocateOne(Buf + 8, R_AARCH64_LDST64_ABS_LO12_NC, Got + 16); @@ -1746,7 +1738,7 @@ uint32_t ARMTargetInfo::getDynRel(uint32_t Type) const { } void ARMTargetInfo::writeGotPlt(uint8_t *Buf, const SymbolBody &) const { - write32le(Buf, In::Plt->getVA()); + write32le(Buf, InX::Plt->getVA()); } void ARMTargetInfo::writeIgotPlt(uint8_t *Buf, const SymbolBody &S) const { @@ -1763,15 +1755,15 @@ void ARMTargetInfo::writePltHeader(uint8_t *Buf) const { 0x00, 0x00, 0x00, 0x00, // L2: .word &(.got.plt) - L1 - 8 }; memcpy(Buf, PltData, sizeof(PltData)); - uint64_t GotPlt = In::GotPlt->getVA(); - uint64_t L1 = In::Plt->getVA() + 8; + uint64_t GotPlt = InX::GotPlt->getVA(); + uint64_t L1 = InX::Plt->getVA() + 8; write32le(Buf + 16, GotPlt - L1 - 8); } void ARMTargetInfo::addPltHeaderSymbols(InputSectionBase *ISD) const { auto *IS = cast(ISD); - addSyntheticLocal("$a", STT_NOTYPE, 0, 0, IS); - addSyntheticLocal("$d", STT_NOTYPE, 16, 0, IS); + addSyntheticLocal("$a", STT_NOTYPE, 0, 0, IS); + addSyntheticLocal("$d", STT_NOTYPE, 16, 0, IS); } void ARMTargetInfo::writePlt(uint8_t *Buf, uint64_t GotPltEntryAddr, @@ -1793,8 +1785,8 @@ void ARMTargetInfo::writePlt(uint8_t *Buf, uint64_t GotPltEntryAddr, void ARMTargetInfo::addPltSymbols(InputSectionBase *ISD, uint64_t Off) const { auto *IS = cast(ISD); - addSyntheticLocal("$a", STT_NOTYPE, Off, 0, IS); - addSyntheticLocal("$d", STT_NOTYPE, Off + 12, 0, IS); + addSyntheticLocal("$a", STT_NOTYPE, Off, 0, IS); + addSyntheticLocal("$d", STT_NOTYPE, Off + 12, 0, IS); } bool ARMTargetInfo::needsThunk(RelExpr Expr, uint32_t RelocType, @@ -1874,7 +1866,8 @@ void ARMTargetInfo::relocateOne(uint8_t *Loc, uint32_t Type, // BLX (always unconditional) instruction to an ARM Target, select an // unconditional BL. write32le(Loc, 0xeb000000 | (read32le(Loc) & 0x00ffffff)); - // fall through as BL encoding is shared with B + // fall through as BL encoding is shared with B + LLVM_FALLTHROUGH; case R_ARM_JUMP24: case R_ARM_PC24: case R_ARM_PLT32: @@ -1908,7 +1901,8 @@ void ARMTargetInfo::relocateOne(uint8_t *Loc, uint32_t Type, } // Bit 12 is 0 for BLX, 1 for BL write16le(Loc + 2, (read16le(Loc + 2) & ~0x1000) | (Val & 1) << 12); - // Fall through as rest of encoding is the same as B.W + // Fall through as rest of encoding is the same as B.W + LLVM_FALLTHROUGH; case R_ARM_THM_JUMP24: // Encoding B T4, BL T1, BLX T2: Val = S:I1:I2:imm10:imm11:0 // FIXME: Use of I1 and I2 require v6T2ops @@ -2132,7 +2126,7 @@ uint32_t MipsTargetInfo::getDynRel(uint32_t Type) const { template void MipsTargetInfo::writeGotPlt(uint8_t *Buf, const SymbolBody &) const { - write32(Buf, In::Plt->getVA()); + write32(Buf, InX::Plt->getVA()); } template @@ -2201,7 +2195,7 @@ void MipsTargetInfo::writePltHeader(uint8_t *Buf) const { write32(Buf + 24, 0x0320f809); // jalr $25 write32(Buf + 28, 0x2718fffe); // subu $24, $24, 2 - uint64_t GotPlt = In::GotPlt->getVA(); + uint64_t GotPlt = InX::GotPlt->getVA(); writeMipsHi16(Buf, GotPlt); writeMipsLo16(Buf + 4, GotPlt); writeMipsLo16(Buf + 8, GotPlt); diff --git a/ELF/Threads.h b/ELF/Threads.h index 897432e69f8e..e01afd4d3fc9 100644 --- a/ELF/Threads.h +++ b/ELF/Threads.h @@ -61,8 +61,7 @@ #include "Config.h" -#include "lld/Core/Parallel.h" -#include +#include "llvm/Support/Parallel.h" #include namespace lld { @@ -71,19 +70,17 @@ namespace elf { template void parallelForEach(IterTy Begin, IterTy End, FuncTy Fn) { if (Config->Threads) - parallel_for_each(Begin, End, Fn); + for_each(llvm::parallel::par, Begin, End, Fn); else - std::for_each(Begin, End, Fn); + for_each(llvm::parallel::seq, Begin, End, Fn); } -inline void parallelFor(size_t Begin, size_t End, - std::function Fn) { - if (Config->Threads) { - parallel_for(Begin, End, Fn); - } else { - for (size_t I = Begin; I < End; ++I) - Fn(I); - } +inline void parallelForEachN(size_t Begin, size_t End, + std::function Fn) { + if (Config->Threads) + for_each_n(llvm::parallel::par, Begin, End, Fn); + else + for_each_n(llvm::parallel::seq, Begin, End, Fn); } } } diff --git a/ELF/Thunks.cpp b/ELF/Thunks.cpp index 307ca5df2288..80ea69663c01 100644 --- a/ELF/Thunks.cpp +++ b/ELF/Thunks.cpp @@ -124,10 +124,10 @@ void ARMV7ABSLongThunk::writeTo(uint8_t *Buf, ThunkSection &IS) const { template void ARMV7ABSLongThunk::addSymbols(ThunkSection &IS) { - this->ThunkSym = addSyntheticLocal( + this->ThunkSym = addSyntheticLocal( Saver.save("__ARMv7ABSLongThunk_" + this->Destination.getName()), STT_FUNC, this->Offset, size(), &IS); - addSyntheticLocal("$a", STT_NOTYPE, this->Offset, 0, &IS); + addSyntheticLocal("$a", STT_NOTYPE, this->Offset, 0, &IS); } template @@ -145,10 +145,10 @@ void ThumbV7ABSLongThunk::writeTo(uint8_t *Buf, ThunkSection &IS) const { template void ThumbV7ABSLongThunk::addSymbols(ThunkSection &IS) { - this->ThunkSym = addSyntheticLocal( + this->ThunkSym = addSyntheticLocal( Saver.save("__Thumbv7ABSLongThunk_" + this->Destination.getName()), STT_FUNC, this->Offset, size(), &IS); - addSyntheticLocal("$t", STT_NOTYPE, this->Offset, 0, &IS); + addSyntheticLocal("$t", STT_NOTYPE, this->Offset, 0, &IS); } template @@ -168,10 +168,10 @@ void ARMV7PILongThunk::writeTo(uint8_t *Buf, ThunkSection &IS) const { template void ARMV7PILongThunk::addSymbols(ThunkSection &IS) { - this->ThunkSym = addSyntheticLocal( + this->ThunkSym = addSyntheticLocal( Saver.save("__ARMV7PILongThunk_" + this->Destination.getName()), STT_FUNC, this->Offset, size(), &IS); - addSyntheticLocal("$a", STT_NOTYPE, this->Offset, 0, &IS); + addSyntheticLocal("$a", STT_NOTYPE, this->Offset, 0, &IS); } template @@ -191,10 +191,10 @@ void ThumbV7PILongThunk::writeTo(uint8_t *Buf, ThunkSection &IS) const { template void ThumbV7PILongThunk::addSymbols(ThunkSection &IS) { - this->ThunkSym = addSyntheticLocal( + this->ThunkSym = addSyntheticLocal( Saver.save("__ThumbV7PILongThunk_" + this->Destination.getName()), STT_FUNC, this->Offset, size(), &IS); - addSyntheticLocal("$t", STT_NOTYPE, this->Offset, 0, &IS); + addSyntheticLocal("$t", STT_NOTYPE, this->Offset, 0, &IS); } // Write MIPS LA25 thunk code to call PIC function from the non-PIC one. @@ -212,7 +212,7 @@ void MipsThunk::writeTo(uint8_t *Buf, ThunkSection &) const { } template void MipsThunk::addSymbols(ThunkSection &IS) { - this->ThunkSym = addSyntheticLocal( + this->ThunkSym = addSyntheticLocal( Saver.save("__LA25Thunk_" + this->Destination.getName()), STT_FUNC, this->Offset, size(), &IS); } diff --git a/ELF/Writer.cpp b/ELF/Writer.cpp index 7f00e37ce7b0..4cdfce76202c 100644 --- a/ELF/Writer.cpp +++ b/ELF/Writer.cpp @@ -164,11 +164,10 @@ static void combineMergableSections() { uint64_t Flags = MS->Flags & ~(uint64_t)(SHF_GROUP | SHF_COMPRESSED); uint32_t Alignment = std::max(MS->Alignment, MS->Entsize); - auto I = - llvm::find_if(MergeSections, [=](MergeSyntheticSection *Sec) { - return Sec->Name == OutsecName && Sec->Flags == Flags && - Sec->Alignment == Alignment; - }); + auto I = llvm::find_if(MergeSections, [=](MergeSyntheticSection *Sec) { + return Sec->Name == OutsecName && Sec->Flags == Flags && + Sec->Alignment == Alignment; + }); if (I == MergeSections.end()) { MergeSyntheticSection *Syn = make(OutsecName, MS->Type, Flags, Alignment); @@ -312,11 +311,11 @@ template void Writer::createSyntheticSections() { auto Add = [](InputSectionBase *Sec) { InputSections.push_back(Sec); }; - In::DynStrTab = make(".dynstr", true); - In::Dynamic = make>(); + InX::DynStrTab = make(".dynstr", true); + InX::Dynamic = make>(); In::RelaDyn = make>( Config->IsRela ? ".rela.dyn" : ".rel.dyn", Config->ZCombreloc); - In::ShStrTab = make(".shstrtab", false); + InX::ShStrTab = make(".shstrtab", false); Out::ElfHeader = make("", 0, SHF_ALLOC); Out::ElfHeader->Size = sizeof(Elf_Ehdr); @@ -324,41 +323,41 @@ template void Writer::createSyntheticSections() { Out::ProgramHeaders->updateAlignment(Config->Wordsize); if (needsInterpSection()) { - In::Interp = createInterpSection(); - Add(In::Interp); + InX::Interp = createInterpSection(); + Add(InX::Interp); } else { - In::Interp = nullptr; + InX::Interp = nullptr; } if (!Config->Relocatable) Add(createCommentSection()); if (Config->Strip != StripPolicy::All) { - In::StrTab = make(".strtab", false); - In::SymTab = make>(*In::StrTab); + InX::StrTab = make(".strtab", false); + InX::SymTab = make>(*InX::StrTab); } if (Config->BuildId != BuildIdKind::None) { - In::BuildId = make(); - Add(In::BuildId); + InX::BuildId = make(); + Add(InX::BuildId); } - In::Common = createCommonSection(); - if (In::Common) + InX::Common = createCommonSection(); + if (InX::Common) Add(InX::Common); - In::Bss = make(".bss"); - Add(In::Bss); - In::BssRelRo = make(".bss.rel.ro"); - Add(In::BssRelRo); + InX::Bss = make(".bss"); + Add(InX::Bss); + InX::BssRelRo = make(".bss.rel.ro"); + Add(InX::BssRelRo); // Add MIPS-specific sections. bool HasDynSymTab = !Symtab::X->getSharedFiles().empty() || Config->Pic || Config->ExportDynamic; if (Config->EMachine == EM_MIPS) { if (!Config->Shared && HasDynSymTab) { - In::MipsRldMap = make(); - Add(In::MipsRldMap); + InX::MipsRldMap = make(); + Add(InX::MipsRldMap); } if (auto *Sec = MipsAbiFlagsSection::create()) Add(Sec); @@ -369,8 +368,8 @@ template void Writer::createSyntheticSections() { } if (HasDynSymTab) { - In::DynSymTab = make>(*In::DynStrTab); - Add(In::DynSymTab); + InX::DynSymTab = make>(*InX::DynStrTab); + Add(InX::DynSymTab); In::VerSym = make>(); Add(In::VerSym); @@ -384,8 +383,8 @@ template void Writer::createSyntheticSections() { Add(In::VerNeed); if (Config->GnuHash) { - In::GnuHashTab = make>(); - Add(In::GnuHashTab); + InX::GnuHashTab = make(); + Add(InX::GnuHashTab); } if (Config->SysvHash) { @@ -393,29 +392,29 @@ template void Writer::createSyntheticSections() { Add(In::HashTab); } - Add(In::Dynamic); - Add(In::DynStrTab); + Add(InX::Dynamic); + Add(InX::DynStrTab); Add(In::RelaDyn); } // Add .got. MIPS' .got is so different from the other archs, // it has its own class. if (Config->EMachine == EM_MIPS) { - In::MipsGot = make(); - Add(In::MipsGot); + InX::MipsGot = make(); + Add(InX::MipsGot); } else { - In::Got = make>(); - Add(In::Got); + InX::Got = make>(); + Add(InX::Got); } - In::GotPlt = make(); - Add(In::GotPlt); - In::IgotPlt = make(); - Add(In::IgotPlt); + InX::GotPlt = make(); + Add(InX::GotPlt); + InX::IgotPlt = make(); + Add(InX::IgotPlt); if (Config->GdbIndex) { - In::GdbIndex = make(); - Add(In::GdbIndex); + InX::GdbIndex = make(); + Add(InX::GdbIndex); } // We always need to add rel[a].plt to output if it has entries. @@ -431,10 +430,10 @@ template void Writer::createSyntheticSections() { false /*Sort*/); Add(In::RelaIplt); - In::Plt = make(Target->PltHeaderSize); - Add(In::Plt); - In::Iplt = make(0); - Add(In::Iplt); + InX::Plt = make(Target->PltHeaderSize); + Add(InX::Plt); + InX::Iplt = make(0); + Add(InX::Iplt); if (!Config->Relocatable) { if (Config->EhFrameHdr) { @@ -445,11 +444,11 @@ template void Writer::createSyntheticSections() { Add(In::EhFrame); } - if (In::SymTab) - Add(In::SymTab); - Add(In::ShStrTab); - if (In::StrTab) - Add(In::StrTab); + if (InX::SymTab) + Add(InX::SymTab); + Add(InX::ShStrTab); + if (InX::StrTab) + Add(InX::StrTab); } static bool shouldKeepInSymtab(SectionBase *Sec, StringRef SymName, @@ -504,7 +503,7 @@ static bool includeInSymtab(const SymbolBody &B) { // Local symbols are not in the linker's symbol table. This function scans // each object file's symbol table to copy local symbols to the output. template void Writer::copyLocalSymbols() { - if (!In::SymTab) + if (!InX::SymTab) return; for (elf::ObjectFile *F : Symtab::X->getObjectFiles()) { for (SymbolBody *B : F->getLocalSymbols()) { @@ -522,7 +521,7 @@ template void Writer::copyLocalSymbols() { SectionBase *Sec = DR->Section; if (!shouldKeepInSymtab(Sec, B->getName(), *B)) continue; - In::SymTab->addSymbol(B); + InX::SymTab->addSymbol(B); } } } @@ -542,43 +541,17 @@ template void Writer::addSectionSymbols() { auto *Sym = make("", /*IsLocal=*/true, /*StOther=*/0, STT_SECTION, /*Value=*/0, /*Size=*/0, IS, nullptr); - In::SymTab->addSymbol(Sym); + InX::SymTab->addSymbol(Sym); } } -// PPC64 has a number of special SHT_PROGBITS+SHF_ALLOC+SHF_WRITE sections that -// we would like to make sure appear is a specific order to maximize their -// coverage by a single signed 16-bit offset from the TOC base pointer. -// Conversely, the special .tocbss section should be first among all SHT_NOBITS -// sections. This will put it next to the loaded special PPC64 sections (and, -// thus, within reach of the TOC base pointer). -static int getPPC64SectionRank(StringRef SectionName) { - return StringSwitch(SectionName) - .Case(".tocbss", 0) - .Case(".branch_lt", 2) - .Case(".toc", 3) - .Case(".toc1", 4) - .Case(".opd", 5) - .Default(1); -} - -// All sections with SHF_MIPS_GPREL flag should be grouped together -// because data in these sections is addressable with a gp relative address. -static int getMipsSectionRank(const OutputSection *S) { - if ((S->Flags & SHF_MIPS_GPREL) == 0) - return 0; - if (S->Name == ".got") - return 1; - return 2; -} - // Today's loaders have a feature to make segments read-only after // processing dynamic relocations to enhance security. PT_GNU_RELRO // is defined for that. // // This function returns true if a section needs to be put into a // PT_GNU_RELRO segment. -template bool elf::isRelroSection(const OutputSection *Sec) { +bool elf::isRelroSection(const OutputSection *Sec) { if (!Config->ZRelro) return false; @@ -613,27 +586,27 @@ template bool elf::isRelroSection(const OutputSection *Sec) { // .got contains pointers to external symbols. They are resolved by // the dynamic linker when a module is loaded into memory, and after // that they are not expected to change. So, it can be in RELRO. - if (In::Got && Sec == In::Got->OutSec) + if (InX::Got && Sec == InX::Got->OutSec) return true; // .got.plt contains pointers to external function symbols. They are // by default resolved lazily, so we usually cannot put it into RELRO. // However, if "-z now" is given, the lazy symbol resolution is // disabled, which enables us to put it into RELRO. - if (Sec == In::GotPlt->OutSec) + if (Sec == InX::GotPlt->OutSec) return Config->ZNow; // .dynamic section contains data for the dynamic linker, and // there's no need to write to it at runtime, so it's better to put // it into RELRO. - if (Sec == In::Dynamic->OutSec) + if (Sec == InX::Dynamic->OutSec) return true; // .bss.rel.ro is used for copy relocations for read-only symbols. // Since the dynamic linker needs to process copy relocations, the // section cannot be read-only, but once initialized, they shouldn't // change. - if (Sec == In::BssRelRo->OutSec) + if (Sec == InX::BssRelRo->OutSec) return true; // Sections with some special names are put into RELRO. This is a @@ -645,105 +618,149 @@ template bool elf::isRelroSection(const OutputSection *Sec) { S == ".eh_frame" || S == ".openbsd.randomdata"; } -template -static bool compareSectionsNonScript(const OutputSection *A, - const OutputSection *B) { - // Put .interp first because some loaders want to see that section - // on the first page of the executable file when loaded into memory. - bool AIsInterp = A->Name == ".interp"; - bool BIsInterp = B->Name == ".interp"; - if (AIsInterp != BIsInterp) - return AIsInterp; +// We compute a rank for each section. The rank indicates where the +// section should be placed in the file. Instead of using simple +// numbers (0,1,2...), we use a series of flags. One for each decision +// point when placing the section. +// Using flags has two key properties: +// * It is easy to check if a give branch was taken. +// * It is easy two see how similar two ranks are (see getRankProximity). +enum RankFlags { + RF_NOT_ADDR_SET = 1 << 16, + RF_NOT_INTERP = 1 << 15, + RF_NOT_ALLOC = 1 << 14, + RF_WRITE = 1 << 13, + RF_EXEC = 1 << 12, + RF_NON_TLS_BSS = 1 << 11, + RF_NON_TLS_BSS_RO = 1 << 10, + RF_NOT_TLS = 1 << 9, + RF_BSS = 1 << 8, + RF_PPC_NOT_TOCBSS = 1 << 7, + RF_PPC_OPD = 1 << 6, + RF_PPC_TOCL = 1 << 5, + RF_PPC_TOC = 1 << 4, + RF_PPC_BRANCH_LT = 1 << 3, + RF_MIPS_GPREL = 1 << 2, + RF_MIPS_NOT_GOT = 1 << 1 +}; - // Allocatable sections go first to reduce the total PT_LOAD size and - // so debug info doesn't change addresses in actual code. - bool AIsAlloc = A->Flags & SHF_ALLOC; - bool BIsAlloc = B->Flags & SHF_ALLOC; - if (AIsAlloc != BIsAlloc) - return AIsAlloc; - - // We don't have any special requirements for the relative order of two non - // allocatable sections. - if (!AIsAlloc) - return false; +static unsigned getSectionRank(const OutputSection *Sec) { + unsigned Rank = 0; // We want to put section specified by -T option first, so we // can start assigning VA starting from them later. - auto AAddrSetI = Config->SectionStartMap.find(A->Name); - auto BAddrSetI = Config->SectionStartMap.find(B->Name); - bool AHasAddrSet = AAddrSetI != Config->SectionStartMap.end(); - bool BHasAddrSet = BAddrSetI != Config->SectionStartMap.end(); - if (AHasAddrSet != BHasAddrSet) - return AHasAddrSet; - if (AHasAddrSet) - return AAddrSetI->second < BAddrSetI->second; + if (Config->SectionStartMap.count(Sec->Name)) + return Rank; + Rank |= RF_NOT_ADDR_SET; + + // Put .interp first because some loaders want to see that section + // on the first page of the executable file when loaded into memory. + if (Sec->Name == ".interp") + return Rank; + Rank |= RF_NOT_INTERP; + + // Allocatable sections go first to reduce the total PT_LOAD size and + // so debug info doesn't change addresses in actual code. + if (!(Sec->Flags & SHF_ALLOC)) + return Rank | RF_NOT_ALLOC; // We want the read only sections first so that they go in the PT_LOAD // covering the program headers at the start of the file. - bool AIsWritable = A->Flags & SHF_WRITE; - bool BIsWritable = B->Flags & SHF_WRITE; - if (AIsWritable != BIsWritable) - return BIsWritable; + if (Sec->Flags & SHF_WRITE) + Rank |= RF_WRITE; - if (!Config->SingleRoRx) { + if (Sec->Flags & SHF_EXECINSTR) { // For a corresponding reason, put non exec sections first (the program // header PT_LOAD is not executable). // We only do that if we are not using linker scripts, since with linker // scripts ro and rx sections are in the same PT_LOAD, so their relative // order is not important. The same applies for -no-rosegment. - bool AIsExec = A->Flags & SHF_EXECINSTR; - bool BIsExec = B->Flags & SHF_EXECINSTR; - if (AIsExec != BIsExec) - return BIsExec; + if ((Rank & RF_WRITE) || !Config->SingleRoRx) + Rank |= RF_EXEC; } // If we got here we know that both A and B are in the same PT_LOAD. - bool AIsTls = A->Flags & SHF_TLS; - bool BIsTls = B->Flags & SHF_TLS; - bool AIsNoBits = A->Type == SHT_NOBITS; - bool BIsNoBits = B->Type == SHT_NOBITS; + bool IsTls = Sec->Flags & SHF_TLS; + bool IsNoBits = Sec->Type == SHT_NOBITS; // The first requirement we have is to put (non-TLS) nobits sections last. The // reason is that the only thing the dynamic linker will see about them is a // p_memsz that is larger than p_filesz. Seeing that it zeros the end of the // PT_LOAD, so that has to correspond to the nobits sections. - bool AIsNonTlsNoBits = AIsNoBits && !AIsTls; - bool BIsNonTlsNoBits = BIsNoBits && !BIsTls; - if (AIsNonTlsNoBits != BIsNonTlsNoBits) - return BIsNonTlsNoBits; + bool IsNonTlsNoBits = IsNoBits && !IsTls; + if (IsNonTlsNoBits) + Rank |= RF_NON_TLS_BSS; // We place nobits RelRo sections before plain r/w ones, and non-nobits RelRo // sections after r/w ones, so that the RelRo sections are contiguous. - bool AIsRelRo = isRelroSection(A); - bool BIsRelRo = isRelroSection(B); - if (AIsRelRo != BIsRelRo) - return AIsNonTlsNoBits ? AIsRelRo : BIsRelRo; + bool IsRelRo = isRelroSection(Sec); + if (IsNonTlsNoBits && !IsRelRo) + Rank |= RF_NON_TLS_BSS_RO; + if (!IsNonTlsNoBits && IsRelRo) + Rank |= RF_NON_TLS_BSS_RO; // The TLS initialization block needs to be a single contiguous block in a R/W // PT_LOAD, so stick TLS sections directly before the other RelRo R/W // sections. The TLS NOBITS sections are placed here as they don't take up // virtual address space in the PT_LOAD. - if (AIsTls != BIsTls) - return AIsTls; + if (!IsTls) + Rank |= RF_NOT_TLS; // Within the TLS initialization block, the non-nobits sections need to appear // first. - if (AIsNoBits != BIsNoBits) - return BIsNoBits; + if (IsNoBits) + Rank |= RF_BSS; - // Some architectures have additional ordering restrictions for sections - // within the same PT_LOAD. - if (Config->EMachine == EM_PPC64) - return getPPC64SectionRank(A->Name) < getPPC64SectionRank(B->Name); - if (Config->EMachine == EM_MIPS) - return getMipsSectionRank(A) < getMipsSectionRank(B); + // // Some architectures have additional ordering restrictions for sections + // // within the same PT_LOAD. + if (Config->EMachine == EM_PPC64) { + // PPC64 has a number of special SHT_PROGBITS+SHF_ALLOC+SHF_WRITE sections + // that we would like to make sure appear is a specific order to maximize + // their coverage by a single signed 16-bit offset from the TOC base + // pointer. Conversely, the special .tocbss section should be first among + // all SHT_NOBITS sections. This will put it next to the loaded special + // PPC64 sections (and, thus, within reach of the TOC base pointer). + StringRef Name = Sec->Name; + if (Name != ".tocbss") + Rank |= RF_PPC_NOT_TOCBSS; + if (Name == ".opd") + Rank |= RF_PPC_OPD; + + if (Name == ".toc1") + Rank |= RF_PPC_TOCL; + + if (Name == ".toc") + Rank |= RF_PPC_TOC; + + if (Name == ".branch_lt") + Rank |= RF_PPC_BRANCH_LT; + } + if (Config->EMachine == EM_MIPS) { + // All sections with SHF_MIPS_GPREL flag should be grouped together + // because data in these sections is addressable with a gp relative address. + if (Sec->Flags & SHF_MIPS_GPREL) + Rank |= RF_MIPS_GPREL; + + if (Sec->Name != ".got") + Rank |= RF_MIPS_NOT_GOT; + } + + return Rank; +} + +static bool compareSectionsNonScript(const OutputSection *A, + const OutputSection *B) { + if (A->SortRank != B->SortRank) + return A->SortRank < B->SortRank; + if (!(A->SortRank & RF_NOT_ADDR_SET)) + return Config->SectionStartMap.lookup(A->Name) < + Config->SectionStartMap.lookup(B->Name); return false; } // Output section ordering is determined by this function. -template static bool compareSections(const OutputSection *A, const OutputSection *B) { // For now, put sections mentioned in a linker script // first. Sections not on linker script will have a SectionIndex of @@ -753,7 +770,7 @@ static bool compareSections(const OutputSection *A, const OutputSection *B) { if (AIndex != BIndex) return AIndex < BIndex; - return compareSectionsNonScript(A, B); + return compareSectionsNonScript(A, B); } // Program header entry @@ -802,7 +819,7 @@ addOptionalRegular(StringRef Name, SectionBase *Sec, uint64_t Val, // need these symbols, since IRELATIVE relocs are resolved through GOT // and PLT. For details, see http://www.airs.com/blog/archives/403. template void Writer::addRelIpltSymbols() { - if (In::DynSymTab) + if (InX::DynSymTab) return; StringRef S = Config->IsRela ? "__rela_iplt_start" : "__rel_iplt_start"; addOptionalRegular(S, In::RelaIplt, 0, STV_HIDDEN, STB_WEAK); @@ -855,16 +872,19 @@ template void Writer::addReservedSymbols() { // static linking the linker is required to optimize away any references to // __tls_get_addr, so it's not defined anywhere. Create a hidden definition // to avoid the undefined symbol error. - if (!In::DynSymTab) + if (!InX::DynSymTab) Symtab::X->addIgnored("__tls_get_addr"); + // __ehdr_start is the location of ELF file headers. Note that we define + // this symbol unconditionally even when using a linker script, which + // differs from the behavior implemented by GNU linker which only define + // this symbol if ELF headers are in the memory mapped segment. + addOptionalRegular("__ehdr_start", Out::ElfHeader, 0, STV_HIDDEN); + // If linker script do layout we do not need to create any standart symbols. if (Script->Opt.HasSections) return; - // __ehdr_start is the location of ELF file headers. - addOptionalRegular("__ehdr_start", Out::ElfHeader, 0, STV_HIDDEN); - auto Add = [](StringRef S) { return addOptionalRegular(S, Out::ElfHeader, 0, STV_DEFAULT); }; @@ -960,18 +980,36 @@ template void Writer::createSections() { Sec->assignOffsets(); } -static bool canSharePtLoad(const OutputSection &S1, const OutputSection &S2) { - if (!(S1.Flags & SHF_ALLOC) || !(S2.Flags & SHF_ALLOC)) - return false; +// We want to find how similar two ranks are. +// The more branches in getSectionRank that match, the more similar they are. +// Since each branch corresponds to a bit flag, we can just use +// countLeadingZeros. +static unsigned getRankProximity(OutputSection *A, OutputSection *B) { + return countLeadingZeros(A->SortRank ^ B->SortRank); +} - bool S1IsWrite = S1.Flags & SHF_WRITE; - bool S2IsWrite = S2.Flags & SHF_WRITE; - if (S1IsWrite != S2IsWrite) - return false; +// We want to place orphan sections so that they share as much +// characteristics with their neighbors as possible. For example, if +// both are rw, or both are tls. +template +static std::vector::iterator +findOrphanPos(std::vector::iterator B, + std::vector::iterator E) { + OutputSection *Sec = *E; - if (!S1IsWrite) - return true; // RO and RX share a PT_LOAD with linker scripts. - return (S1.Flags & SHF_EXECINSTR) == (S2.Flags & SHF_EXECINSTR); + // Find the first element that has as close a rank as possible. + auto I = std::max_element(B, E, [=](OutputSection *A, OutputSection *B) { + return getRankProximity(Sec, A) < getRankProximity(Sec, B); + }); + if (I == E) + return E; + + // Consider all existing sections with the same proximity. + unsigned Proximity = getRankProximity(Sec, *I); + while (I != E && getRankProximity(Sec, *I) == Proximity && + Sec->SortRank >= (*I)->SortRank) + ++I; + return I; } template void Writer::sortSections() { @@ -979,12 +1017,18 @@ template void Writer::sortSections() { // relative order for SHF_LINK_ORDER sections. if (Config->Relocatable) return; + + if (Script->Opt.HasSections) + Script->adjustSectionsBeforeSorting(); + + for (OutputSection *Sec : OutputSections) + Sec->SortRank = getSectionRank(Sec); + if (!Script->Opt.HasSections) { std::stable_sort(OutputSections.begin(), OutputSections.end(), - compareSectionsNonScript); + compareSectionsNonScript); return; } - Script->adjustSectionsBeforeSorting(); // The order of the sections in the script is arbitrary and may not agree with // compareSectionsNonScript. This means that we cannot easily define a @@ -1004,14 +1048,13 @@ template void Writer::sortSections() { // .d (ro) # not in script // // The way we define an order then is: - // * First put script sections at the start and sort the script and - // non-script sections independently. + // * First put script sections at the start and sort the script sections. // * Move each non-script section to its preferred position. We try // to put each section in the last position where it it can share // a PT_LOAD. std::stable_sort(OutputSections.begin(), OutputSections.end(), - compareSections); + compareSections); auto I = OutputSections.begin(); auto E = OutputSections.end(); @@ -1019,31 +1062,16 @@ template void Writer::sortSections() { std::find_if(OutputSections.begin(), E, [](OutputSection *S) { return S->SectionIndex == INT_MAX; }); while (NonScriptI != E) { - auto BestPos = std::max_element( - I, NonScriptI, [&](OutputSection *&A, OutputSection *&B) { - bool ACanSharePtLoad = canSharePtLoad(**NonScriptI, *A); - bool BCanSharePtLoad = canSharePtLoad(**NonScriptI, *B); - if (ACanSharePtLoad != BCanSharePtLoad) - return BCanSharePtLoad; + auto Pos = findOrphanPos(I, NonScriptI); - bool ACmp = compareSectionsNonScript(*NonScriptI, A); - bool BCmp = compareSectionsNonScript(*NonScriptI, B); - if (ACmp != BCmp) - return BCmp; // FIXME: missing test - - size_t PosA = &A - &OutputSections[0]; - size_t PosB = &B - &OutputSections[0]; - return ACmp ? PosA > PosB : PosA < PosB; - }); - - // max_element only returns NonScriptI if the range is empty. If the range - // is not empty we should consider moving the the element forward one - // position. - if (BestPos != NonScriptI && - !compareSectionsNonScript(*NonScriptI, *BestPos)) - ++BestPos; - std::rotate(BestPos, NonScriptI, NonScriptI + 1); - ++NonScriptI; + // As an optimization, find all sections with the same sort rank + // and insert them with one rotate. + unsigned Rank = (*NonScriptI)->SortRank; + auto End = std::find_if(NonScriptI + 1, E, [=](OutputSection *Sec) { + return Sec->SortRank != Rank; + }); + std::rotate(Pos, NonScriptI, End); + NonScriptI = End; } Script->adjustSectionsAfterSorting(); @@ -1103,8 +1131,8 @@ template void Writer::finalizeSections() { // It should be okay as no one seems to care about the type. // Even the author of gold doesn't remember why gold behaves that way. // https://sourceware.org/ml/binutils/2002-03/msg00360.html - if (In::DynSymTab) - addRegular("_DYNAMIC", In::Dynamic, 0); + if (InX::DynSymTab) + addRegular("_DYNAMIC", InX::Dynamic, 0); // Define __rel[a]_iplt_{start,end} symbols if needed. addRelIpltSymbols(); @@ -1119,10 +1147,10 @@ template void Writer::finalizeSections() { // we can correctly decide if a dynamic relocation is needed. forEachRelSec(scanRelocations); - if (In::Plt && !In::Plt->empty()) - In::Plt->addSymbols(); - if (In::Iplt && !In::Iplt->empty()) - In::Iplt->addSymbols(); + if (InX::Plt && !InX::Plt->empty()) + InX::Plt->addSymbols(); + if (InX::Iplt && !InX::Iplt->empty()) + InX::Iplt->addSymbols(); // Now that we have defined all possible global symbols including linker- // synthesized ones. Visit all symbols to give the finishing touches. @@ -1131,11 +1159,11 @@ template void Writer::finalizeSections() { if (!includeInSymtab(*Body)) continue; - if (In::SymTab) - In::SymTab->addSymbol(Body); + if (InX::SymTab) + InX::SymTab->addSymbol(Body); - if (In::DynSymTab && S->includeInDynsym()) { - In::DynSymTab->addSymbol(Body); + if (InX::DynSymTab && S->includeInDynsym()) { + InX::DynSymTab->addSymbol(Body); if (auto *SS = dyn_cast(Body)) if (cast>(SS->File)->isNeeded()) In::VerNeed->addSymbol(SS); @@ -1161,7 +1189,7 @@ template void Writer::finalizeSections() { unsigned I = 1; for (OutputSection *Sec : OutputSections) { Sec->SectionIndex = I++; - Sec->ShName = In::ShStrTab->addString(Sec->Name); + Sec->ShName = InX::ShStrTab->addString(Sec->Name); } // Binary and relocatable output does not have PHDRS. @@ -1175,15 +1203,14 @@ template void Writer::finalizeSections() { // Dynamic section must be the last one in this list and dynamic // symbol table section (DynSymTab) must be the first one. - applySynthetic({In::DynSymTab, In::Bss, In::BssRelRo, - In::GnuHashTab, In::HashTab, In::SymTab, - In::ShStrTab, In::StrTab, In::VerDef, - In::DynStrTab, In::GdbIndex, In::Got, - In::MipsGot, In::IgotPlt, In::GotPlt, - In::RelaDyn, In::RelaIplt, In::RelaPlt, - In::Plt, In::Iplt, In::Plt, - In::EhFrameHdr, In::VerSym, In::VerNeed, - In::Dynamic}, + applySynthetic({InX::DynSymTab, InX::Bss, InX::BssRelRo, + InX::GnuHashTab, In::HashTab, InX::SymTab, + InX::ShStrTab, InX::StrTab, In::VerDef, + InX::DynStrTab, InX::GdbIndex, InX::Got, + InX::MipsGot, InX::IgotPlt, InX::GotPlt, + In::RelaDyn, In::RelaIplt, In::RelaPlt, + InX::Plt, InX::Iplt, In::EhFrameHdr, + In::VerSym, In::VerNeed, InX::Dynamic}, [](SyntheticSection *SS) { SS->finalizeContents(); }); // Some architectures use small displacements for jump instructions. @@ -1198,7 +1225,7 @@ template void Writer::finalizeSections() { // when no more Thunks are added ThunkCreator TC; if (TC.createThunks(OutputSections)) - applySynthetic({In::MipsGot}, + applySynthetic({InX::MipsGot}, [](SyntheticSection *SS) { SS->updateAllocSize(); }); } // Fill other section headers. The dynamic table is finalized @@ -1214,7 +1241,7 @@ template void Writer::finalizeSections() { [](OutputSection *S) { S->maybeCompress(); }); // createThunks may have added local symbols to the static symbol table - applySynthetic({In::SymTab, In::ShStrTab, In::StrTab}, + applySynthetic({InX::SymTab, InX::ShStrTab, InX::StrTab}, [](SyntheticSection *SS) { SS->postThunkContents(); }); } @@ -1332,7 +1359,7 @@ template std::vector Writer::createPhdrs() { // different flags or is loaded at a discontiguous address using AT linker // script command. uint64_t NewFlags = computeFlags(Sec->getPhdrFlags()); - if (Script->hasLMA(Sec->Name) || Flags != NewFlags) { + if (Script->hasLMA(Sec) || Flags != NewFlags) { Load = AddHdr(PT_LOAD, NewFlags); Flags = NewFlags; } @@ -1349,15 +1376,15 @@ template std::vector Writer::createPhdrs() { Ret.push_back(std::move(TlsHdr)); // Add an entry for .dynamic. - if (In::DynSymTab) - AddHdr(PT_DYNAMIC, In::Dynamic->OutSec->getPhdrFlags()) - ->add(In::Dynamic->OutSec); + if (InX::DynSymTab) + AddHdr(PT_DYNAMIC, InX::Dynamic->OutSec->getPhdrFlags()) + ->add(InX::Dynamic->OutSec); // PT_GNU_RELRO includes all sections that should be marked as // read-only by dynamic linker after proccessing relocations. PhdrEntry RelRo(PT_GNU_RELRO, PF_R); for (OutputSection *Sec : OutputSections) - if (needsPtLoad(Sec) && isRelroSection(Sec)) + if (needsPtLoad(Sec) && isRelroSection(Sec)) RelRo.add(Sec); if (RelRo.First) Ret.push_back(std::move(RelRo)); @@ -1395,7 +1422,7 @@ template std::vector Writer::createPhdrs() { PhdrEntry *Note = nullptr; for (OutputSection *Sec : OutputSections) { if (Sec->Type == SHT_NOTE) { - if (!Note || Script->hasLMA(Sec->Name)) + if (!Note || Script->hasLMA(Sec)) Note = AddHdr(PT_NOTE, PF_R); Note->add(Sec); } else { @@ -1547,7 +1574,7 @@ template uint64_t Writer::getEntryAddr() { if (SymbolBody *B = Symtab::X->find(Config->Entry)) return B->getVA(); uint64_t Addr; - if (!Config->Entry.getAsInteger(0, Addr)) + if (to_integer(Config->Entry, Addr)) return Addr; // Case 4 @@ -1649,7 +1676,7 @@ template void Writer::writeHeader() { EHdr->e_phnum = Phdrs.size(); EHdr->e_shentsize = sizeof(Elf_Shdr); EHdr->e_shnum = OutputSections.size() + 1; - EHdr->e_shstrndx = In::ShStrTab->OutSec->SectionIndex; + EHdr->e_shstrndx = InX::ShStrTab->OutSec->SectionIndex; if (Config->EMachine == EM_ARM) // We don't currently use any features incompatible with EF_ARM_EABI_VER5, @@ -1743,21 +1770,16 @@ template void Writer::writeSections() { } template void Writer::writeBuildId() { - if (!In::BuildId || !In::BuildId->OutSec) + if (!InX::BuildId || !InX::BuildId->OutSec) return; // Compute a hash of all sections of the output file. uint8_t *Start = Buffer->getBufferStart(); uint8_t *End = Start + FileSize; - In::BuildId->writeBuildId({Start, End}); + InX::BuildId->writeBuildId({Start, End}); } template void elf::writeResult(); template void elf::writeResult(); template void elf::writeResult(); template void elf::writeResult(); - -template bool elf::isRelroSection(const OutputSection *); -template bool elf::isRelroSection(const OutputSection *); -template bool elf::isRelroSection(const OutputSection *); -template bool elf::isRelroSection(const OutputSection *); diff --git a/ELF/Writer.h b/ELF/Writer.h index 8b965f7beddb..17fbda394a20 100644 --- a/ELF/Writer.h +++ b/ELF/Writer.h @@ -24,7 +24,7 @@ template class ObjectFile; template class SymbolTable; template void writeResult(); template void markLive(); -template bool isRelroSection(const OutputSection *Sec); +bool isRelroSection(const OutputSection *Sec); // This describes a program header entry. // Each contains type, access flags and range of output sections that will be diff --git a/docs/CMakeLists.txt b/docs/CMakeLists.txt index d4f3b058efb7..112ce35e8cf4 100644 --- a/docs/CMakeLists.txt +++ b/docs/CMakeLists.txt @@ -1,6 +1,6 @@ if (LLVM_ENABLE_SPHINX) + include(AddSphinxTarget) if (SPHINX_FOUND) - include(AddSphinxTarget) if (${SPHINX_OUTPUT_HTML}) add_sphinx_target(html lld) endif() diff --git a/include/lld/Core/Parallel.h b/include/lld/Core/Parallel.h deleted file mode 100644 index 58fa87e85c51..000000000000 --- a/include/lld/Core/Parallel.h +++ /dev/null @@ -1,166 +0,0 @@ -//===- lld/Core/Parallel.h - Parallel utilities ---------------------------===// -// -// The LLVM Linker -// -// This file is distributed under the University of Illinois Open Source -// License. See LICENSE.TXT for details. -// -//===----------------------------------------------------------------------===// - -#ifndef LLD_CORE_PARALLEL_H -#define LLD_CORE_PARALLEL_H - -#include "lld/Core/LLVM.h" -#include "lld/Core/TaskGroup.h" -#include "llvm/Support/MathExtras.h" -#include "llvm/Config/llvm-config.h" - -#include - -#if defined(_MSC_VER) && LLVM_ENABLE_THREADS -#include -#include -#endif - -namespace lld { - -#if !LLVM_ENABLE_THREADS -template -void parallel_sort( - RandomAccessIterator Start, RandomAccessIterator End, - const Comparator &Comp = std::less< - typename std::iterator_traits::value_type>()) { - std::sort(Start, End, Comp); -} -#elif defined(_MSC_VER) -// Use ppl parallel_sort on Windows. -template -void parallel_sort( - RandomAccessIterator Start, RandomAccessIterator End, - const Comparator &Comp = std::less< - typename std::iterator_traits::value_type>()) { - concurrency::parallel_sort(Start, End, Comp); -} -#else -namespace detail { -const ptrdiff_t MinParallelSize = 1024; - -/// \brief Inclusive median. -template -RandomAccessIterator medianOf3(RandomAccessIterator Start, - RandomAccessIterator End, - const Comparator &Comp) { - RandomAccessIterator Mid = Start + (std::distance(Start, End) / 2); - return Comp(*Start, *(End - 1)) - ? (Comp(*Mid, *(End - 1)) ? (Comp(*Start, *Mid) ? Mid : Start) - : End - 1) - : (Comp(*Mid, *Start) ? (Comp(*(End - 1), *Mid) ? Mid : End - 1) - : Start); -} - -template -void parallel_quick_sort(RandomAccessIterator Start, RandomAccessIterator End, - const Comparator &Comp, TaskGroup &TG, size_t Depth) { - // Do a sequential sort for small inputs. - if (std::distance(Start, End) < detail::MinParallelSize || Depth == 0) { - std::sort(Start, End, Comp); - return; - } - - // Partition. - auto Pivot = medianOf3(Start, End, Comp); - // Move Pivot to End. - std::swap(*(End - 1), *Pivot); - Pivot = std::partition(Start, End - 1, [&Comp, End](decltype(*Start) V) { - return Comp(V, *(End - 1)); - }); - // Move Pivot to middle of partition. - std::swap(*Pivot, *(End - 1)); - - // Recurse. - TG.spawn([=, &Comp, &TG] { - parallel_quick_sort(Start, Pivot, Comp, TG, Depth - 1); - }); - parallel_quick_sort(Pivot + 1, End, Comp, TG, Depth - 1); -} -} - -template -void parallel_sort( - RandomAccessIterator Start, RandomAccessIterator End, - const Comparator &Comp = std::less< - typename std::iterator_traits::value_type>()) { - TaskGroup TG; - detail::parallel_quick_sort(Start, End, Comp, TG, - llvm::Log2_64(std::distance(Start, End)) + 1); -} -#endif - -template void parallel_sort(T *Start, T *End) { - parallel_sort(Start, End, std::less()); -} - -#if !LLVM_ENABLE_THREADS -template -void parallel_for_each(IterTy Begin, IterTy End, FuncTy Fn) { - std::for_each(Begin, End, Fn); -} - -template -void parallel_for(IndexTy Begin, IndexTy End, FuncTy Fn) { - for (IndexTy I = Begin; I != End; ++I) - Fn(I); -} -#elif defined(_MSC_VER) -// Use ppl parallel_for_each on Windows. -template -void parallel_for_each(IterTy Begin, IterTy End, FuncTy Fn) { - concurrency::parallel_for_each(Begin, End, Fn); -} - -template -void parallel_for(IndexTy Begin, IndexTy End, FuncTy Fn) { - concurrency::parallel_for(Begin, End, Fn); -} -#else -template -void parallel_for_each(IterTy Begin, IterTy End, FuncTy Fn) { - // TaskGroup has a relatively high overhead, so we want to reduce - // the number of spawn() calls. We'll create up to 1024 tasks here. - // (Note that 1024 is an arbitrary number. This code probably needs - // improving to take the number of available cores into account.) - ptrdiff_t TaskSize = std::distance(Begin, End) / 1024; - if (TaskSize == 0) - TaskSize = 1; - - TaskGroup TG; - while (TaskSize <= std::distance(Begin, End)) { - TG.spawn([=, &Fn] { std::for_each(Begin, Begin + TaskSize, Fn); }); - Begin += TaskSize; - } - TG.spawn([=, &Fn] { std::for_each(Begin, End, Fn); }); -} - -template -void parallel_for(IndexTy Begin, IndexTy End, FuncTy Fn) { - ptrdiff_t TaskSize = (End - Begin) / 1024; - if (TaskSize == 0) - TaskSize = 1; - - TaskGroup TG; - IndexTy I = Begin; - for (; I + TaskSize < End; I += TaskSize) { - TG.spawn([=, &Fn] { - for (IndexTy J = I, E = I + TaskSize; J != E; ++J) - Fn(J); - }); - } - TG.spawn([=, &Fn] { - for (IndexTy J = I; J < End; ++J) - Fn(J); - }); -} -#endif -} // End namespace lld - -#endif // LLD_CORE_PARALLEL_H diff --git a/include/lld/Core/TaskGroup.h b/include/lld/Core/TaskGroup.h deleted file mode 100644 index 82e9122f4ae2..000000000000 --- a/include/lld/Core/TaskGroup.h +++ /dev/null @@ -1,65 +0,0 @@ -//===- lld/Core/TaskGroup.h - Task Group ----------------------------------===// -// -// The LLVM Linker -// -// This file is distributed under the University of Illinois Open Source -// License. See LICENSE.TXT for details. -// -//===----------------------------------------------------------------------===// - -#ifndef LLD_CORE_TASKGROUP_H -#define LLD_CORE_TASKGROUP_H - -#include "lld/Core/LLVM.h" - -#include -#include -#include - -namespace lld { -/// \brief Allows one or more threads to wait on a potentially unknown number of -/// events. -/// -/// A latch starts at \p count. inc() increments this, and dec() decrements it. -/// All calls to sync() will block while the count is not 0. -/// -/// Calling dec() on a Latch with a count of 0 has undefined behaivor. -class Latch { - uint32_t _count; - mutable std::mutex _condMut; - mutable std::condition_variable _cond; - -public: - explicit Latch(uint32_t count = 0) : _count(count) {} - ~Latch() { sync(); } - - void inc() { - std::unique_lock lock(_condMut); - ++_count; - } - - void dec() { - std::unique_lock lock(_condMut); - if (--_count == 0) - _cond.notify_all(); - } - - void sync() const { - std::unique_lock lock(_condMut); - _cond.wait(lock, [&] { return _count == 0; }); - } -}; - -/// \brief Allows launching a number of tasks and waiting for them to finish -/// either explicitly via sync() or implicitly on destruction. -class TaskGroup { - Latch _latch; - -public: - void spawn(std::function f); - - void sync() const { _latch.sync(); } -}; -} - -#endif diff --git a/lib/Core/CMakeLists.txt b/lib/Core/CMakeLists.txt index cdd4e679ffa2..f2bf90509295 100644 --- a/lib/Core/CMakeLists.txt +++ b/lib/Core/CMakeLists.txt @@ -12,7 +12,6 @@ add_lld_library(lldCore Resolver.cpp SymbolTable.cpp TargetOptionsCommandFlags.cpp - TaskGroup.cpp Writer.cpp ADDITIONAL_HEADER_DIRS diff --git a/lib/Core/TaskGroup.cpp b/lib/Core/TaskGroup.cpp deleted file mode 100644 index d4de48ce3dc4..000000000000 --- a/lib/Core/TaskGroup.cpp +++ /dev/null @@ -1,141 +0,0 @@ -//===- lld/Core/TaskGroup.cpp - Task Group --------------------------------===// -// -// The LLVM Linker -// -// This file is distributed under the University of Illinois Open Source -// License. See LICENSE.TXT for details. -// -//===----------------------------------------------------------------------===// - -#include "lld/Core/TaskGroup.h" -#include "llvm/Config/llvm-config.h" - -#include -#include -#include - -#if defined(_MSC_VER) && LLVM_ENABLE_THREADS -#include -#include -#endif - -using namespace lld; - -namespace { - -/// \brief An abstract class that takes closures and runs them asynchronously. -class Executor { -public: - virtual ~Executor() = default; - virtual void add(std::function func) = 0; - - static Executor *getDefaultExecutor(); -}; - -#if !LLVM_ENABLE_THREADS -class SyncExecutor : public Executor { -public: - virtual void add(std::function F) { F(); } -}; - -Executor *Executor::getDefaultExecutor() { - static SyncExecutor Exec; - return &Exec; -} - -#elif defined(_MSC_VER) -/// \brief An Executor that runs tasks via ConcRT. -class ConcRTExecutor : public Executor { - struct Taskish { - Taskish(std::function Task) : Task(Task) {} - - std::function Task; - - static void run(void *P) { - Taskish *Self = static_cast(P); - Self->Task(); - concurrency::Free(Self); - } - }; - -public: - virtual void add(std::function F) { - Concurrency::CurrentScheduler::ScheduleTask( - Taskish::run, new (concurrency::Alloc(sizeof(Taskish))) Taskish(F)); - } -}; - -Executor *Executor::getDefaultExecutor() { - static ConcRTExecutor exec; - return &exec; -} - -#else -/// \brief An implementation of an Executor that runs closures on a thread pool -/// in filo order. -class ThreadPoolExecutor : public Executor { -public: - explicit ThreadPoolExecutor( - unsigned ThreadCount = std::thread::hardware_concurrency()) - : Done(ThreadCount) { - // Spawn all but one of the threads in another thread as spawning threads - // can take a while. - std::thread([&, ThreadCount] { - for (size_t i = 1; i < ThreadCount; ++i) { - std::thread([=] { work(); }).detach(); - } - work(); - }).detach(); - } - - ~ThreadPoolExecutor() override { - std::unique_lock Lock(Mutex); - Stop = true; - Lock.unlock(); - Cond.notify_all(); - // Wait for ~Latch. - } - - void add(std::function F) override { - std::unique_lock Lock(Mutex); - WorkStack.push(F); - Lock.unlock(); - Cond.notify_one(); - } - -private: - void work() { - while (true) { - std::unique_lock Lock(Mutex); - Cond.wait(Lock, [&] { return Stop || !WorkStack.empty(); }); - if (Stop) - break; - auto Task = WorkStack.top(); - WorkStack.pop(); - Lock.unlock(); - Task(); - } - Done.dec(); - } - - std::atomic Stop{false}; - std::stack> WorkStack; - std::mutex Mutex; - std::condition_variable Cond; - Latch Done; -}; - -Executor *Executor::getDefaultExecutor() { - static ThreadPoolExecutor exec; - return &exec; -} -#endif -} - -void TaskGroup::spawn(std::function f) { - _latch.inc(); - Executor::getDefaultExecutor()->add([&, f] { - f(); - _latch.dec(); - }); -} diff --git a/lib/ReaderWriter/MachO/LayoutPass.cpp b/lib/ReaderWriter/MachO/LayoutPass.cpp index 24dbf79d3e3b..7bca07eb16d6 100644 --- a/lib/ReaderWriter/MachO/LayoutPass.cpp +++ b/lib/ReaderWriter/MachO/LayoutPass.cpp @@ -9,12 +9,12 @@ #include "LayoutPass.h" #include "lld/Core/Instrumentation.h" -#include "lld/Core/Parallel.h" #include "lld/Core/PassManager.h" #include "lld/ReaderWriter/MachOLinkingContext.h" #include "llvm/ADT/STLExtras.h" #include "llvm/ADT/Twine.h" #include "llvm/Support/Debug.h" +#include "llvm/Support/Parallel.h" #include #include #include @@ -461,10 +461,10 @@ llvm::Error LayoutPass::perform(SimpleFile &mergedFile) { }); std::vector vec = decorate(atomRange); - parallel_sort(vec.begin(), vec.end(), - [&](const LayoutPass::SortKey &l, const LayoutPass::SortKey &r) -> bool { - return compareAtoms(l, r, _customSorter); - }); + sort(llvm::parallel::par, vec.begin(), vec.end(), + [&](const LayoutPass::SortKey &l, const LayoutPass::SortKey &r) -> bool { + return compareAtoms(l, r, _customSorter); + }); DEBUG(checkTransitivity(vec, _customSorter)); undecorate(atomRange, vec); diff --git a/test/COFF/constant-export.test b/test/COFF/constant-export.test index 18b1f5e30d29..80597660ce19 100644 --- a/test/COFF/constant-export.test +++ b/test/COFF/constant-export.test @@ -1,5 +1,5 @@ # RUN: mkdir -p %t -# RUN: yaml2obj -o %t/constant-export.obj %S/constant-export.yaml +# RUN: yaml2obj -o %t/constant-export.obj %s # RUN: lld-link /machine:x86 /dll /entry:__CFConstantStringClassReference -out:%t/constant-export.dll %t/constant-export.obj # RUN: llvm-readobj -coff-exports %t/constant-export.lib | FileCheck %s @@ -7,3 +7,86 @@ # CHECK: Name type: noprefix # CHECK: Symbol: __imp____CFConstantStringClassReference +--- !COFF +header: + Machine: IMAGE_FILE_MACHINE_I386 + Characteristics: [ ] +sections: + - Name: .text + Characteristics: [ IMAGE_SCN_CNT_CODE, IMAGE_SCN_MEM_EXECUTE, IMAGE_SCN_MEM_READ ] + Alignment: 4 + SectionData: '' + - Name: .data + Characteristics: [ IMAGE_SCN_CNT_INITIALIZED_DATA, IMAGE_SCN_MEM_READ, IMAGE_SCN_MEM_WRITE ] + Alignment: 4 + SectionData: '' + - Name: .bss + Characteristics: [ IMAGE_SCN_CNT_UNINITIALIZED_DATA, IMAGE_SCN_MEM_READ, IMAGE_SCN_MEM_WRITE ] + Alignment: 4 + SectionData: '' + - Name: .drectve + Characteristics: [ IMAGE_SCN_LNK_INFO, IMAGE_SCN_LNK_REMOVE ] + Alignment: 1 + SectionData: 20202D6578706F72743A5F5F5F4346436F6E7374616E74537472696E67436C6173735265666572656E63652C434F4E5354414E54 +symbols: + - Name: .text + Value: 0 + SectionNumber: 1 + SimpleType: IMAGE_SYM_TYPE_NULL + ComplexType: IMAGE_SYM_DTYPE_NULL + StorageClass: IMAGE_SYM_CLASS_STATIC + SectionDefinition: + Length: 0 + NumberOfRelocations: 0 + NumberOfLinenumbers: 0 + CheckSum: 0 + Number: 1 + - Name: .data + Value: 0 + SectionNumber: 2 + SimpleType: IMAGE_SYM_TYPE_NULL + ComplexType: IMAGE_SYM_DTYPE_NULL + StorageClass: IMAGE_SYM_CLASS_STATIC + SectionDefinition: + Length: 0 + NumberOfRelocations: 0 + NumberOfLinenumbers: 0 + CheckSum: 0 + Number: 2 + - Name: .bss + Value: 0 + SectionNumber: 3 + SimpleType: IMAGE_SYM_TYPE_NULL + ComplexType: IMAGE_SYM_DTYPE_NULL + StorageClass: IMAGE_SYM_CLASS_STATIC + SectionDefinition: + Length: 0 + NumberOfRelocations: 0 + NumberOfLinenumbers: 0 + CheckSum: 0 + Number: 3 + - Name: .drectve + Value: 0 + SectionNumber: 4 + SimpleType: IMAGE_SYM_TYPE_NULL + ComplexType: IMAGE_SYM_DTYPE_NULL + StorageClass: IMAGE_SYM_CLASS_STATIC + SectionDefinition: + Length: 52 + NumberOfRelocations: 0 + NumberOfLinenumbers: 0 + CheckSum: 1983959296 + Number: 4 + - Name: '@feat.00' + Value: 1 + SectionNumber: -1 + SimpleType: IMAGE_SYM_TYPE_NULL + ComplexType: IMAGE_SYM_DTYPE_NULL + StorageClass: IMAGE_SYM_CLASS_STATIC + - Name: ___CFConstantStringClassReference + Value: 128 + SectionNumber: 0 + SimpleType: IMAGE_SYM_TYPE_NULL + ComplexType: IMAGE_SYM_DTYPE_NULL + StorageClass: IMAGE_SYM_CLASS_EXTERNAL +... diff --git a/test/COFF/constant-export.yaml b/test/COFF/constant-export.yaml deleted file mode 100644 index 7e44bb70c9d7..000000000000 --- a/test/COFF/constant-export.yaml +++ /dev/null @@ -1,83 +0,0 @@ ---- !COFF -header: - Machine: IMAGE_FILE_MACHINE_I386 - Characteristics: [ ] -sections: - - Name: .text - Characteristics: [ IMAGE_SCN_CNT_CODE, IMAGE_SCN_MEM_EXECUTE, IMAGE_SCN_MEM_READ ] - Alignment: 4 - SectionData: '' - - Name: .data - Characteristics: [ IMAGE_SCN_CNT_INITIALIZED_DATA, IMAGE_SCN_MEM_READ, IMAGE_SCN_MEM_WRITE ] - Alignment: 4 - SectionData: '' - - Name: .bss - Characteristics: [ IMAGE_SCN_CNT_UNINITIALIZED_DATA, IMAGE_SCN_MEM_READ, IMAGE_SCN_MEM_WRITE ] - Alignment: 4 - SectionData: '' - - Name: .drectve - Characteristics: [ IMAGE_SCN_LNK_INFO, IMAGE_SCN_LNK_REMOVE ] - Alignment: 1 - SectionData: 20202D6578706F72743A5F5F5F4346436F6E7374616E74537472696E67436C6173735265666572656E63652C434F4E5354414E54 -symbols: - - Name: .text - Value: 0 - SectionNumber: 1 - SimpleType: IMAGE_SYM_TYPE_NULL - ComplexType: IMAGE_SYM_DTYPE_NULL - StorageClass: IMAGE_SYM_CLASS_STATIC - SectionDefinition: - Length: 0 - NumberOfRelocations: 0 - NumberOfLinenumbers: 0 - CheckSum: 0 - Number: 1 - - Name: .data - Value: 0 - SectionNumber: 2 - SimpleType: IMAGE_SYM_TYPE_NULL - ComplexType: IMAGE_SYM_DTYPE_NULL - StorageClass: IMAGE_SYM_CLASS_STATIC - SectionDefinition: - Length: 0 - NumberOfRelocations: 0 - NumberOfLinenumbers: 0 - CheckSum: 0 - Number: 2 - - Name: .bss - Value: 0 - SectionNumber: 3 - SimpleType: IMAGE_SYM_TYPE_NULL - ComplexType: IMAGE_SYM_DTYPE_NULL - StorageClass: IMAGE_SYM_CLASS_STATIC - SectionDefinition: - Length: 0 - NumberOfRelocations: 0 - NumberOfLinenumbers: 0 - CheckSum: 0 - Number: 3 - - Name: .drectve - Value: 0 - SectionNumber: 4 - SimpleType: IMAGE_SYM_TYPE_NULL - ComplexType: IMAGE_SYM_DTYPE_NULL - StorageClass: IMAGE_SYM_CLASS_STATIC - SectionDefinition: - Length: 52 - NumberOfRelocations: 0 - NumberOfLinenumbers: 0 - CheckSum: 1983959296 - Number: 4 - - Name: '@feat.00' - Value: 1 - SectionNumber: -1 - SimpleType: IMAGE_SYM_TYPE_NULL - ComplexType: IMAGE_SYM_DTYPE_NULL - StorageClass: IMAGE_SYM_CLASS_STATIC - - Name: ___CFConstantStringClassReference - Value: 128 - SectionNumber: 0 - SimpleType: IMAGE_SYM_TYPE_NULL - ComplexType: IMAGE_SYM_DTYPE_NULL - StorageClass: IMAGE_SYM_CLASS_EXTERNAL -... diff --git a/test/ELF/Inputs/i386-static-tls-model1.s b/test/ELF/Inputs/i386-static-tls-model1.s deleted file mode 100644 index e7e584c1fcf1..000000000000 --- a/test/ELF/Inputs/i386-static-tls-model1.s +++ /dev/null @@ -1,10 +0,0 @@ -.section ".tdata", "awT", @progbits -.globl var -var: - -.section .foo, "aw" -.global _start -_start: - movl $var@tpoff, %edx # R_386_TLS_LE_32 - movl %gs:0, %ecx - subl %edx, %eax diff --git a/test/ELF/Inputs/i386-static-tls-model2.s b/test/ELF/Inputs/i386-static-tls-model2.s deleted file mode 100644 index b28a1458742d..000000000000 --- a/test/ELF/Inputs/i386-static-tls-model2.s +++ /dev/null @@ -1,9 +0,0 @@ -.section ".tdata", "awT", @progbits -.globl var -var: - -.section .foo, "aw" -.global _start -_start: - movl %gs:0, %eax - addl var@gotntpoff(%ebx),%eax # R_386_TLS_GOTIE diff --git a/test/ELF/Inputs/i386-static-tls-model3.s b/test/ELF/Inputs/i386-static-tls-model3.s deleted file mode 100644 index f92267ecbdd0..000000000000 --- a/test/ELF/Inputs/i386-static-tls-model3.s +++ /dev/null @@ -1,9 +0,0 @@ -.section ".tdata", "awT", @progbits -.globl var -var: - -.section .foo, "aw" -.global _start -_start: - movl %gs:0, %eax - addl var@indntpoff, %eax #R_386_TLS_IE diff --git a/test/ELF/Inputs/i386-static-tls-model4.s b/test/ELF/Inputs/i386-static-tls-model4.s deleted file mode 100644 index ffb20def4fab..000000000000 --- a/test/ELF/Inputs/i386-static-tls-model4.s +++ /dev/null @@ -1,9 +0,0 @@ -.section ".tdata", "awT", @progbits -.globl var -var: - -.section .foo, "aw" -.global _start -_start: - movl %gs:0, %eax - leal var@ntpoff(%eax), %eax #R_386_TLS_LE diff --git a/test/ELF/gdb-index-empty.s b/test/ELF/gdb-index-empty.s new file mode 100644 index 000000000000..933afed33e2f --- /dev/null +++ b/test/ELF/gdb-index-empty.s @@ -0,0 +1,116 @@ +# RUN: llvm-mc -filetype=obj -triple=x86_64-pc-linux -o %t %s +# RUN: ld.lld --gdb-index --gc-sections -o %t2 %t +# RUN: llvm-dwarfdump -debug-dump=gdb_index %t2 | FileCheck %s + +# CHECK: Address area offset = 0x28, has 0 entries: + +# Generated with: (clang r302976) +# echo "void _start() { __builtin_unreachable(); }" | \ +# clang -Os -g -S -o gdb-index-empty.s -x c - -Xclang -fdebug-compilation-dir -Xclang . + + .text + .file "-" + .globl _start + .type _start,@function +_start: # @_start +.Lfunc_begin0: + .cfi_startproc +# BB#0: # %entry +.Lfunc_end0: + .size _start, .Lfunc_end0-_start + .cfi_endproc + + .file 1 "" + .section .debug_str,"MS",@progbits,1 +.Linfo_string0: + .asciz "clang version 5.0.0 " # string offset=0 +.Linfo_string1: + .asciz "-" # string offset=21 +.Linfo_string2: + .asciz "." # string offset=23 +.Linfo_string3: + .asciz "_start" # string offset=25 + .section .debug_loc,"",@progbits + .section .debug_abbrev,"",@progbits + .byte 1 # Abbreviation Code + .byte 17 # DW_TAG_compile_unit + .byte 1 # DW_CHILDREN_yes + .byte 37 # DW_AT_producer + .byte 14 # DW_FORM_strp + .byte 19 # DW_AT_language + .byte 5 # DW_FORM_data2 + .byte 3 # DW_AT_name + .byte 14 # DW_FORM_strp + .byte 16 # DW_AT_stmt_list + .byte 23 # DW_FORM_sec_offset + .byte 27 # DW_AT_comp_dir + .byte 14 # DW_FORM_strp + .byte 17 # DW_AT_low_pc + .byte 1 # DW_FORM_addr + .byte 18 # DW_AT_high_pc + .byte 6 # DW_FORM_data4 + .byte 0 # EOM(1) + .byte 0 # EOM(2) + .byte 2 # Abbreviation Code + .byte 46 # DW_TAG_subprogram + .byte 0 # DW_CHILDREN_no + .byte 17 # DW_AT_low_pc + .byte 1 # DW_FORM_addr + .byte 18 # DW_AT_high_pc + .byte 6 # DW_FORM_data4 + .byte 64 # DW_AT_frame_base + .byte 24 # DW_FORM_exprloc + .byte 3 # DW_AT_name + .byte 14 # DW_FORM_strp + .byte 58 # DW_AT_decl_file + .byte 11 # DW_FORM_data1 + .byte 59 # DW_AT_decl_line + .byte 11 # DW_FORM_data1 + .byte 63 # DW_AT_external + .byte 25 # DW_FORM_flag_present + .byte 0 # EOM(1) + .byte 0 # EOM(2) + .byte 0 # EOM(3) + .section .debug_info,"",@progbits +.Lcu_begin0: + .long 60 # Length of Unit + .short 4 # DWARF version number + .long .debug_abbrev # Offset Into Abbrev. Section + .byte 8 # Address Size (in bytes) + .byte 1 # Abbrev [1] 0xb:0x35 DW_TAG_compile_unit + .long .Linfo_string0 # DW_AT_producer + .short 12 # DW_AT_language + .long .Linfo_string1 # DW_AT_name + .long .Lline_table_start0 # DW_AT_stmt_list + .long .Linfo_string2 # DW_AT_comp_dir + .quad .Lfunc_begin0 # DW_AT_low_pc + .long .Lfunc_end0-.Lfunc_begin0 # DW_AT_high_pc + .byte 2 # Abbrev [2] 0x2a:0x15 DW_TAG_subprogram + .quad .Lfunc_begin0 # DW_AT_low_pc + .long .Lfunc_end0-.Lfunc_begin0 # DW_AT_high_pc + .byte 1 # DW_AT_frame_base + .byte 87 + .long .Linfo_string3 # DW_AT_name + .byte 1 # DW_AT_decl_file + .byte 1 # DW_AT_decl_line + # DW_AT_external + .byte 0 # End Of Children Mark + .section .debug_ranges,"",@progbits + .section .debug_macinfo,"",@progbits +.Lcu_macro_begin0: + .byte 0 # End Of Macro List Mark + .section .debug_pubnames,"",@progbits + .long .LpubNames_end0-.LpubNames_begin0 # Length of Public Names Info +.LpubNames_begin0: + .short 2 # DWARF Version + .long .Lcu_begin0 # Offset of Compilation Unit Info + .long 64 # Compilation Unit Length + .long 42 # DIE offset + .asciz "_start" # External Name + .long 0 # End Mark +.LpubNames_end0: + + .ident "clang version 5.0.0 " + .section ".note.GNU-stack","",@progbits + .section .debug_line,"",@progbits +.Lline_table_start0: diff --git a/test/ELF/gdb-index-gc-sections.s b/test/ELF/gdb-index-gc-sections.s new file mode 100644 index 000000000000..70a14754656c --- /dev/null +++ b/test/ELF/gdb-index-gc-sections.s @@ -0,0 +1,157 @@ +# RUN: llvm-mc -filetype=obj -triple=x86_64-pc-linux -o %t %s +# RUN: ld.lld --gdb-index --gc-sections -o %t2 %t +# RUN: llvm-dwarfdump -debug-dump=gdb_index %t2 | FileCheck %s + +# CHECK: Address area offset = 0x28, has 1 entries: +# CHECK-NEXT: Low/High address = [0x201000, 0x201001) (Size: 0x1), CU id = 0 + +# Generated with: (clang r302976) +# echo "void _start() {} void dead() {}" | \ +# clang -Os -g -S -ffunction-sections -o gdb-index-gc-sections.s -x c - -Xclang -fdebug-compilation-dir -Xclang . + + .text + .file "-" + .section .text._start,"ax",@progbits + .globl _start + .type _start,@function +_start: # @_start +.Lfunc_begin0: + .file 1 "" + .loc 1 1 0 # :1:0 + .cfi_startproc +# BB#0: # %entry + .loc 1 1 16 prologue_end # :1:16 + retq +.Ltmp0: +.Lfunc_end0: + .size _start, .Lfunc_end0-_start + .cfi_endproc + + .section .text.dead,"ax",@progbits + .globl dead + .type dead,@function +dead: # @dead +.Lfunc_begin1: + .loc 1 1 0 # :1:0 + .cfi_startproc +# BB#0: # %entry + .loc 1 1 31 prologue_end # :1:31 + retq +.Ltmp1: +.Lfunc_end1: + .size dead, .Lfunc_end1-dead + .cfi_endproc + + .section .debug_str,"MS",@progbits,1 +.Linfo_string0: + .asciz "clang version 5.0.0 " # string offset=0 +.Linfo_string1: + .asciz "-" # string offset=21 +.Linfo_string2: + .asciz "." # string offset=23 +.Linfo_string3: + .asciz "_start" # string offset=25 +.Linfo_string4: + .asciz "dead" # string offset=32 + .section .debug_loc,"",@progbits + .section .debug_abbrev,"",@progbits + .byte 1 # Abbreviation Code + .byte 17 # DW_TAG_compile_unit + .byte 1 # DW_CHILDREN_yes + .byte 37 # DW_AT_producer + .byte 14 # DW_FORM_strp + .byte 19 # DW_AT_language + .byte 5 # DW_FORM_data2 + .byte 3 # DW_AT_name + .byte 14 # DW_FORM_strp + .byte 16 # DW_AT_stmt_list + .byte 23 # DW_FORM_sec_offset + .byte 27 # DW_AT_comp_dir + .byte 14 # DW_FORM_strp + .byte 17 # DW_AT_low_pc + .byte 1 # DW_FORM_addr + .byte 85 # DW_AT_ranges + .byte 23 # DW_FORM_sec_offset + .byte 0 # EOM(1) + .byte 0 # EOM(2) + .byte 2 # Abbreviation Code + .byte 46 # DW_TAG_subprogram + .byte 0 # DW_CHILDREN_no + .byte 17 # DW_AT_low_pc + .byte 1 # DW_FORM_addr + .byte 18 # DW_AT_high_pc + .byte 6 # DW_FORM_data4 + .byte 64 # DW_AT_frame_base + .byte 24 # DW_FORM_exprloc + .byte 3 # DW_AT_name + .byte 14 # DW_FORM_strp + .byte 58 # DW_AT_decl_file + .byte 11 # DW_FORM_data1 + .byte 59 # DW_AT_decl_line + .byte 11 # DW_FORM_data1 + .byte 63 # DW_AT_external + .byte 25 # DW_FORM_flag_present + .byte 0 # EOM(1) + .byte 0 # EOM(2) + .byte 0 # EOM(3) + .section .debug_info,"",@progbits +.Lcu_begin0: + .long 81 # Length of Unit + .short 4 # DWARF version number + .long .debug_abbrev # Offset Into Abbrev. Section + .byte 8 # Address Size (in bytes) + .byte 1 # Abbrev [1] 0xb:0x4a DW_TAG_compile_unit + .long .Linfo_string0 # DW_AT_producer + .short 12 # DW_AT_language + .long .Linfo_string1 # DW_AT_name + .long .Lline_table_start0 # DW_AT_stmt_list + .long .Linfo_string2 # DW_AT_comp_dir + .quad 0 # DW_AT_low_pc + .long .Ldebug_ranges0 # DW_AT_ranges + .byte 2 # Abbrev [2] 0x2a:0x15 DW_TAG_subprogram + .quad .Lfunc_begin0 # DW_AT_low_pc + .long .Lfunc_end0-.Lfunc_begin0 # DW_AT_high_pc + .byte 1 # DW_AT_frame_base + .byte 87 + .long .Linfo_string3 # DW_AT_name + .byte 1 # DW_AT_decl_file + .byte 1 # DW_AT_decl_line + # DW_AT_external + .byte 2 # Abbrev [2] 0x3f:0x15 DW_TAG_subprogram + .quad .Lfunc_begin1 # DW_AT_low_pc + .long .Lfunc_end1-.Lfunc_begin1 # DW_AT_high_pc + .byte 1 # DW_AT_frame_base + .byte 87 + .long .Linfo_string4 # DW_AT_name + .byte 1 # DW_AT_decl_file + .byte 1 # DW_AT_decl_line + # DW_AT_external + .byte 0 # End Of Children Mark + .section .debug_ranges,"",@progbits +.Ldebug_ranges0: + .quad .Lfunc_begin0 + .quad .Lfunc_end0 + .quad .Lfunc_begin1 + .quad .Lfunc_end1 + .quad 0 + .quad 0 + .section .debug_macinfo,"",@progbits +.Lcu_macro_begin0: + .byte 0 # End Of Macro List Mark + .section .debug_pubnames,"",@progbits + .long .LpubNames_end0-.LpubNames_begin0 # Length of Public Names Info +.LpubNames_begin0: + .short 2 # DWARF Version + .long .Lcu_begin0 # Offset of Compilation Unit Info + .long 85 # Compilation Unit Length + .long 42 # DIE offset + .asciz "_start" # External Name + .long 63 # DIE offset + .asciz "dead" # External Name + .long 0 # End Mark +.LpubNames_end0: + + .ident "clang version 5.0.0 " + .section ".note.GNU-stack","",@progbits + .section .debug_line,"",@progbits +.Lline_table_start0: diff --git a/test/ELF/i386-static-tls-model.s b/test/ELF/i386-static-tls-model.s deleted file mode 100644 index b2799c4c722f..000000000000 --- a/test/ELF/i386-static-tls-model.s +++ /dev/null @@ -1,20 +0,0 @@ -# REQUIRES: x86 - -# RUN: llvm-mc -filetype=obj -triple=i686-pc-linux %S/Inputs/i386-static-tls-model1.s -o %t.o -# RUN: ld.lld %t.o -o %t1 -shared -# RUN: llvm-readobj -dynamic-table %t1 | FileCheck %s - -# RUN: llvm-mc -filetype=obj -triple=i686-pc-linux %S/Inputs/i386-static-tls-model2.s -o %t.o -# RUN: ld.lld %t.o -o %t2 -shared -# RUN: llvm-readobj -dynamic-table %t2 | FileCheck %s - -# RUN: llvm-mc -filetype=obj -triple=i686-pc-linux %S/Inputs/i386-static-tls-model3.s -o %t.o -# RUN: ld.lld %t.o -o %t3 -shared -# RUN: llvm-readobj -dynamic-table %t3 | FileCheck %s - -# RUN: llvm-mc -filetype=obj -triple=i686-pc-linux %S/Inputs/i386-static-tls-model4.s -o %t.o -# RUN: ld.lld %t.o -o %t4 -shared -# RUN: llvm-readobj -dynamic-table %t4 | FileCheck %s - -# CHECK: DynamicSection [ -# CHECK: FLAGS STATIC_TLS diff --git a/test/ELF/i386-tls-ie-shared.s b/test/ELF/i386-tls-ie-shared.s index c6dccf84a216..8becc3199f95 100644 --- a/test/ELF/i386-tls-ie-shared.s +++ b/test/ELF/i386-tls-ie-shared.s @@ -13,8 +13,8 @@ // GOTRELSHARED-NEXT: SHF_ALLOC // GOTRELSHARED-NEXT: SHF_WRITE // GOTRELSHARED-NEXT: ] -// GOTRELSHARED-NEXT: Address: 0x1060 -// GOTRELSHARED-NEXT: Offset: 0x1060 +// GOTRELSHARED-NEXT: Address: 0x1058 +// GOTRELSHARED-NEXT: Offset: 0x1058 // GOTRELSHARED-NEXT: Size: 16 // GOTRELSHARED-NEXT: Link: 0 // GOTRELSHARED-NEXT: Info: 0 @@ -31,36 +31,36 @@ // GOTRELSHARED-NEXT: 0x202D R_386_RELATIVE - 0x0 // GOTRELSHARED-NEXT: 0x2036 R_386_RELATIVE - 0x0 // GOTRELSHARED-NEXT: 0x203F R_386_RELATIVE - 0x0 -// GOTRELSHARED-NEXT: 0x1060 R_386_TLS_TPOFF tlslocal0 0x0 -// GOTRELSHARED-NEXT: 0x1064 R_386_TLS_TPOFF tlslocal1 0x0 -// GOTRELSHARED-NEXT: 0x1068 R_386_TLS_TPOFF tlsshared0 0x0 -// GOTRELSHARED-NEXT: 0x106C R_386_TLS_TPOFF tlsshared1 0x0 +// GOTRELSHARED-NEXT: 0x1058 R_386_TLS_TPOFF tlslocal0 0x0 +// GOTRELSHARED-NEXT: 0x105C R_386_TLS_TPOFF tlslocal1 0x0 +// GOTRELSHARED-NEXT: 0x1060 R_386_TLS_TPOFF tlsshared0 0x0 +// GOTRELSHARED-NEXT: 0x1064 R_386_TLS_TPOFF tlsshared1 0x0 // GOTRELSHARED-NEXT: } // GOTRELSHARED-NEXT: ] // GOTRELSHARED: 0x6FFFFFFA RELCOUNT 8 // DISASMSHARED: Disassembly of section test: // DISASMSHARED-NEXT: _start: -// (.got)[0] = 0x1060 = 4192 -// (.got)[1] = 0x1064 = 4196 -// (.got)[2] = 0x1068 = 4200 -// (.got)[3] = 0x106C = 4204 -// DISASMSHARED-NEXT: 2000: {{.*}} movl 4192, %ecx -// DISASMSHARED-NEXT: 2006: {{.*}} movl %gs:(%ecx), %eax -// DISASMSHARED-NEXT: 2009: {{.*}} movl 4192, %eax -// DISASMSHARED-NEXT: 200e: {{.*}} movl %gs:(%eax), %eax -// DISASMSHARED-NEXT: 2011: {{.*}} addl 4192, %ecx -// DISASMSHARED-NEXT: 2017: {{.*}} movl %gs:(%ecx), %eax -// DISASMSHARED-NEXT: 201a: {{.*}} movl 4196, %ecx -// DISASMSHARED-NEXT: 2020: {{.*}} movl %gs:(%ecx), %eax -// DISASMSHARED-NEXT: 2023: {{.*}} movl 4196, %eax -// DISASMSHARED-NEXT: 2028: {{.*}} movl %gs:(%eax), %eax -// DISASMSHARED-NEXT: 202b: {{.*}} addl 4196, %ecx -// DISASMSHARED-NEXT: 2031: {{.*}} movl %gs:(%ecx), %eax -// DISASMSHARED-NEXT: 2034: {{.*}} movl 4200, %ecx -// DISASMSHARED-NEXT: 203a: {{.*}} movl %gs:(%ecx), %eax -// DISASMSHARED-NEXT: 203d: {{.*}} addl 4204, %ecx -// DISASMSHARED-NEXT: 2043: {{.*}} movl %gs:(%ecx), %eax +// (.got)[0] = 0x2050 = 8272 +// (.got)[1] = 0x2054 = 8276 +// (.got)[2] = 0x2058 = 8280 +// (.got)[3] = 0x205C = 8284 +// DISASMSHARED-NEXT: 2000: 8b 0d 58 10 00 00 movl 4184, %ecx +// DISASMSHARED-NEXT: 2006: 65 8b 01 movl %gs:(%ecx), %eax +// DISASMSHARED-NEXT: 2009: a1 58 10 00 00 movl 4184, %eax +// DISASMSHARED-NEXT: 200e: 65 8b 00 movl %gs:(%eax), %eax +// DISASMSHARED-NEXT: 2011: 03 0d 58 10 00 00 addl 4184, %ecx +// DISASMSHARED-NEXT: 2017: 65 8b 01 movl %gs:(%ecx), %eax +// DISASMSHARED-NEXT: 201a: 8b 0d 5c 10 00 00 movl 4188, %ecx +// DISASMSHARED-NEXT: 2020: 65 8b 01 movl %gs:(%ecx), %eax +// DISASMSHARED-NEXT: 2023: a1 5c 10 00 00 movl 4188, %eax +// DISASMSHARED-NEXT: 2028: 65 8b 00 movl %gs:(%eax), %eax +// DISASMSHARED-NEXT: 202b: 03 0d 5c 10 00 00 addl 4188, %ecx +// DISASMSHARED-NEXT: 2031: 65 8b 01 movl %gs:(%ecx), %eax +// DISASMSHARED-NEXT: 2034: 8b 0d 60 10 00 00 movl 4192, %ecx +// DISASMSHARED-NEXT: 203a: 65 8b 01 movl %gs:(%ecx), %eax +// DISASMSHARED-NEXT: 203d: 03 0d 64 10 00 00 addl 4196, %ecx +// DISASMSHARED-NEXT: 2043: 65 8b 01 movl %gs:(%ecx), %eax .type tlslocal0,@object .section .tbss,"awT",@nobits diff --git a/test/ELF/incompatible-section-types2.s b/test/ELF/incompatible-section-types2.s index 2cf9b8548aa1..146e680ab271 100644 --- a/test/ELF/incompatible-section-types2.s +++ b/test/ELF/incompatible-section-types2.s @@ -1,7 +1,9 @@ // RUN: llvm-mc -filetype=obj -triple=x86_64-pc-linux %s -o %t.o // RUN: not ld.lld %t.o -o %t 2>&1 | FileCheck %s -// CHECK: error: Section has different type from others with the same name :(.shstrtab) +// CHECK: error: section type mismatch for .shstrtab +// CHECK-NEXT: >>> :(.shstrtab): SHT_STRTAB +// CHECK-NEXT: >>> output section .shstrtab: Unknown -.section .shstrtab,"" +.section .shstrtab,"",@12345 .short 20 diff --git a/test/ELF/linkerscript/early-assign-symbol.s b/test/ELF/linkerscript/early-assign-symbol.s new file mode 100644 index 000000000000..21940c088393 --- /dev/null +++ b/test/ELF/linkerscript/early-assign-symbol.s @@ -0,0 +1,14 @@ +# REQUIRES: x86 +# RUN: llvm-mc -filetype=obj -triple=x86_64-unknown-linux %s -o %t.o + +# RUN: echo "SECTIONS { aaa = 1 + ABSOLUTE(foo - 1); .text : { *(.text*) } }" > %t1.script +# RUN: not ld.lld -o %t --script %t1.script %t.o 2>&1 | FileCheck %s + +# RUN: echo "SECTIONS { aaa = ABSOLUTE(foo - 1) + 1; .text : { *(.text*) } }" > %t2.script +# RUN: not ld.lld -o %t --script %t2.script %t.o 2>&1 | FileCheck %s + +# CHECK: error: unable to evaluate expression: input section .text has no output section assigned + +.section .text +.globl foo +foo: diff --git a/test/ELF/linkerscript/ehdr_start.s b/test/ELF/linkerscript/ehdr_start.s index 935fa2bf3391..4da158a83956 100644 --- a/test/ELF/linkerscript/ehdr_start.s +++ b/test/ELF/linkerscript/ehdr_start.s @@ -2,9 +2,17 @@ # RUN: llvm-mc -filetype=obj -triple=x86_64-unknown-linux %s -o %t.o # RUN: echo "SECTIONS { }" > %t.script -# RUN: not ld.lld %t.o -script %t.script -o %t 2>&1 | FileCheck %s -# CHECK: error: undefined symbol: __ehdr_start -# CHECK: >>> referenced by {{.*}}:(.text+0x0) +# RUN: ld.lld %t.o -script %t.script -o %t +# RUN: llvm-readobj -symbols %t | FileCheck %s +# CHECK: Name: __ehdr_start (1) +# CHECK-NEXT: Value: 0x0 +# CHECK-NEXT: Size: 0 +# CHECK-NEXT: Binding: Local (0x0) +# CHECK-NEXT: Type: None (0x0) +# CHECK-NEXT: Other [ (0x2) +# CHECK-NEXT: STV_HIDDEN (0x2) +# CHECK-NEXT: ] +# CHECK-NEXT: Section: .text (0x1) .text .global _start, __ehdr_start diff --git a/test/ELF/linkerscript/sections-constraint.s b/test/ELF/linkerscript/sections-constraint.s index 4d95ec18336c..796240627170 100644 --- a/test/ELF/linkerscript/sections-constraint.s +++ b/test/ELF/linkerscript/sections-constraint.s @@ -24,8 +24,8 @@ # NO1-NEXT: 0 00000000 # NO1: .writable 00000004 # NO1: .foo.2 00000004 -# NO1: .readable 00000004 # NO1: .foo.1 00000004 +# NO1: .readable 00000004 .global _start _start: diff --git a/test/ELF/linkerscript/sections.s b/test/ELF/linkerscript/sections.s index 69c6f19d078d..d5645c303754 100644 --- a/test/ELF/linkerscript/sections.s +++ b/test/ELF/linkerscript/sections.s @@ -45,8 +45,9 @@ # SEC-ORDER: 3 .shstrtab 0000003b {{[0-9a-f]*}} # SEC-ORDER: 4 .symtab 00000030 {{[0-9a-f]*}} # SEC-ORDER: 5 .strtab 00000008 {{[0-9a-f]*}} -# SEC-ORDER: 6 .data 00000020 {{[0-9a-f]*}} DATA -# SEC-ORDER: 7 .text 0000000e {{[0-9a-f]*}} TEXT DATA +# SEC-ORDER: 6 .comment 00000008 {{[0-9a-f]*}} +# SEC-ORDER: 7 .data 00000020 {{[0-9a-f]*}} DATA +# SEC-ORDER: 8 .text 0000000e {{[0-9a-f]*}} TEXT DATA # .text and .data have swapped names but proper sizes and types. # RUN: echo "SECTIONS { \ diff --git a/test/ELF/linkerscript/symbol-memoryexpr.s b/test/ELF/linkerscript/symbol-memoryexpr.s new file mode 100644 index 000000000000..9c75274e1644 --- /dev/null +++ b/test/ELF/linkerscript/symbol-memoryexpr.s @@ -0,0 +1,33 @@ +# REQUIRES: x86 +# RUN: llvm-mc -filetype=obj -triple=x86_64-unknown-linux %s -o %t + +# RUN: echo "MEMORY { \ +# RUN: ram (rwx) : ORIGIN = 0x8000, LENGTH = 256K \ +# RUN: } \ +# RUN: SECTIONS { \ +# RUN: origin = ORIGIN(ram); \ +# RUN: length = LENGTH(ram); \ +# RUN: end = ORIGIN(ram) + LENGTH(ram); \ +# RUN: }" > %t.script +# RUN: ld.lld -o %t1 --script %t.script %t +# RUN: llvm-objdump -t %t1 | FileCheck %s + +# CHECK: SYMBOL TABLE: +# CHECK-NEXT: 0000000000000000 *UND* 00000000 +# CHECK-NEXT: 0000000000008000 .text 00000000 _start +# CHECK-NEXT: 0000000000008000 *ABS* 00000000 origin +# CHECK-NEXT: 0000000000040000 *ABS* 00000000 length +# CHECK-NEXT: 0000000000048000 *ABS* 00000000 end + +# RUN: echo "SECTIONS { \ +# RUN: no_exist_origin = ORIGIN(ram); \ +# RUN: no_exist_length = LENGTH(ram); \ +# RUN: }" > %t2.script +# RUN: not ld.lld -o %t2 --script %t2.script %t 2>&1 \ +# RUN: | FileCheck -check-prefix=ERR %s +# ERR: {{.*}}.script:1: memory region not defined: ram + + +.global _start +_start: + nop diff --git a/test/ELF/many-alloc-sections.s b/test/ELF/many-alloc-sections.s new file mode 100644 index 000000000000..441e5ff32d08 --- /dev/null +++ b/test/ELF/many-alloc-sections.s @@ -0,0 +1,106 @@ +// RUN: llvm-mc -filetype=obj -triple x86_64-pc-linux-gnu %s -o %t.o +// RUN: echo "SECTIONS { . = SIZEOF_HEADERS; .text : { *(.text) } }" > %t.script +// FIXME: threads are disable because the test is too slow with them (PR32942). +// RUN: ld.lld -T %t.script %t.o -o %t --no-threads +// RUN: llvm-readobj -t %t | FileCheck %s + +// Test that _start is in the correct section. +// CHECK: Name: _start +// CHECK-NEXT: Value: 0x120 +// CHECK-NEXT: Size: 0 +// CHECK-NEXT: Binding: Global +// CHECK-NEXT: Type: None +// CHECK-NEXT: Other: 0 +// CHECK-NEXT: Section: dm + +.macro gen_sections4 x + .section a\x,"a" + .section b\x,"a" + .section c\x,"a" + .section d\x,"a" +.endm + +.macro gen_sections8 x + gen_sections4 a\x + gen_sections4 b\x +.endm + +.macro gen_sections16 x + gen_sections8 a\x + gen_sections8 b\x +.endm + +.macro gen_sections32 x + gen_sections16 a\x + gen_sections16 b\x +.endm + +.macro gen_sections64 x + gen_sections32 a\x + gen_sections32 b\x +.endm + +.macro gen_sections128 x + gen_sections64 a\x + gen_sections64 b\x +.endm + +.macro gen_sections256 x + gen_sections128 a\x + gen_sections128 b\x +.endm + +.macro gen_sections512 x + gen_sections256 a\x + gen_sections256 b\x +.endm + +.macro gen_sections1024 x + gen_sections512 a\x + gen_sections512 b\x +.endm + +.macro gen_sections2048 x + gen_sections1024 a\x + gen_sections1024 b\x +.endm + +.macro gen_sections4096 x + gen_sections2048 a\x + gen_sections2048 b\x +.endm + +.macro gen_sections8192 x + gen_sections4096 a\x + gen_sections4096 b\x +.endm + +.macro gen_sections16384 x + gen_sections8192 a\x + gen_sections8192 b\x +.endm + +.macro gen_sections32768 x + gen_sections16384 a\x + gen_sections16384 b\x +.endm + + .bss + .section bar + +gen_sections32768 a +gen_sections16384 b +gen_sections8192 c +gen_sections4096 d +gen_sections2048 e +gen_sections1024 f +gen_sections512 g +gen_sections128 h +gen_sections64 i +gen_sections32 j +gen_sections16 k +gen_sections8 l +gen_sections4 m + +.global _start +_start: diff --git a/test/ELF/many-sections.s b/test/ELF/many-sections.s index 77e76c20a60d..ae923889ddc1 100644 --- a/test/ELF/many-sections.s +++ b/test/ELF/many-sections.s @@ -11,7 +11,14 @@ // CHECK-NEXT: Section: dm (0xFF00) -// RUN: ld.lld %t -o %t2 +// FIXME: threads are disable because the test is too slow with them (PR32942). +// RUN: ld.lld %t -o %t2 --no-threads +// RUN: llvm-readobj -t %t2 | FileCheck --check-prefix=LINKED %s + +// Test also with a linker script. +// RUN: echo "SECTIONS { . = SIZEOF_HEADERS; .text : { *(.text) } }" > %t.script +// FIXME: threads are disable because the test is too slow with them (PR32942). +// RUN: ld.lld -T %t.script %t -o %t2 --no-threads // RUN: llvm-readobj -t %t2 | FileCheck --check-prefix=LINKED %s // Test that _start is in the correct section. diff --git a/test/ELF/tls-dynamic-i686.s b/test/ELF/tls-dynamic-i686.s index 04fd13822530..ac88e6eaed31 100644 --- a/test/ELF/tls-dynamic-i686.s +++ b/test/ELF/tls-dynamic-i686.s @@ -56,8 +56,8 @@ addl tls1@gotntpoff(%ebx),%eax // CHECK-NEXT: SHF_ALLOC // CHECK-NEXT: SHF_WRITE // CHECK-NEXT: ] -// CHECK-NEXT: Address: 0x3070 -// CHECK-NEXT: Offset: 0x3070 +// CHECK-NEXT: Address: 0x3068 +// CHECK-NEXT: Offset: 0x3068 // CHECK-NEXT: Size: 32 // CHECK-NEXT: Link: 0 // CHECK-NEXT: Info: 0 @@ -66,13 +66,13 @@ addl tls1@gotntpoff(%ebx),%eax // CHECK: Relocations [ // CHECK: Section ({{.+}}) .rel.dyn { -// CHECK-NEXT: 0x3080 R_386_TLS_DTPMOD32 - 0x0 -// CHECK-NEXT: 0x3070 R_386_TLS_DTPMOD32 tls0 0x0 -// CHECK-NEXT: 0x3074 R_386_TLS_DTPOFF32 tls0 0x0 -// CHECK-NEXT: 0x3088 R_386_TLS_TPOFF tls0 0x0 -// CHECK-NEXT: 0x3078 R_386_TLS_DTPMOD32 tls1 0x0 -// CHECK-NEXT: 0x307C R_386_TLS_DTPOFF32 tls1 0x0 -// CHECK-NEXT: 0x308C R_386_TLS_TPOFF tls1 0x0 +// CHECK-NEXT: 0x3078 R_386_TLS_DTPMOD32 - 0x0 +// CHECK-NEXT: 0x3068 R_386_TLS_DTPMOD32 tls0 0x0 +// CHECK-NEXT: 0x306C R_386_TLS_DTPOFF32 tls0 0x0 +// CHECK-NEXT: 0x3080 R_386_TLS_TPOFF tls0 0x0 +// CHECK-NEXT: 0x3070 R_386_TLS_DTPMOD32 tls1 0x0 +// CHECK-NEXT: 0x3074 R_386_TLS_DTPOFF32 tls1 0x0 +// CHECK-NEXT: 0x3084 R_386_TLS_TPOFF tls1 0x0 // CHECK-NEXT: } // DIS: Disassembly of section .text: @@ -80,20 +80,20 @@ addl tls1@gotntpoff(%ebx),%eax // General dynamic model: // -32 and -24 are first and second GOT entries offsets. // Each one is a pair of records. -// DIS-NEXT: 1000: {{.*}} leal -32(,%ebx), %eax -// DIS-NEXT: 1007: {{.*}} calll 100 -// DIS-NEXT: 100c: {{.*}} leal -24(,%ebx), %eax -// DIS-NEXT: 1013: {{.*}} calll 88 +// DIS-NEXT: 1000: 8d 04 1d e0 ff ff ff leal -32(,%ebx), %eax +// DIS-NEXT: 1007: e8 64 00 00 00 calll 100 +// DIS-NEXT: 100c: 8d 04 1d e8 ff ff ff leal -24(,%ebx), %eax +// DIS-NEXT: 1013: e8 58 00 00 00 calll 88 // Local dynamic model: // -16 is a local module tls index offset. -// DIS-NEXT: 1018: {{.*}} leal -16(%ebx), %eax -// DIS-NEXT: 101e: {{.*}} calll 77 -// DIS-NEXT: 1023: {{.*}} leal 8(%eax), %edx -// DIS-NEXT: 1029: {{.*}} leal -16(%ebx), %eax -// DIS-NEXT: 102f: {{.*}} calll 60 -// DIS-NEXT: 1034: {{.*}} leal 12(%eax), %edx +// DIS-NEXT: 1018: 8d 83 f0 ff ff ff leal -16(%ebx), %eax +// DIS-NEXT: 101e: e8 4d 00 00 00 calll 77 +// DIS-NEXT: 1023: 8d 90 08 00 00 00 leal 8(%eax), %edx +// DIS-NEXT: 1029: 8d 83 f0 ff ff ff leal -16(%ebx), %eax +// DIS-NEXT: 102f: e8 3c 00 00 00 calll 60 +// DIS-NEXT: 1034: 8d 90 0c 00 00 00 leal 12(%eax), %edx // Initial exec model: -// DIS-NEXT: 103a: {{.*}} movl %gs:0, %eax -// DIS-NEXT: 1040: {{.*}} addl -8(%ebx), %eax -// DIS-NEXT: 1046: {{.*}} movl %gs:0, %eax -// DIS-NEXT: 104c: {{.*}} addl -4(%ebx), %eax +// DIS-NEXT: 103a: 65 a1 00 00 00 00 movl %gs:0, %eax +// DIS-NEXT: 1040: 03 83 f8 ff ff ff addl -8(%ebx), %eax +// DIS-NEXT: 1046: 65 a1 00 00 00 00 movl %gs:0, %eax +// DIS-NEXT: 104c: 03 83 fc ff ff ff addl -4(%ebx), %eax diff --git a/test/ELF/tls-opt-iele-i686-nopic.s b/test/ELF/tls-opt-iele-i686-nopic.s index a883bce511a6..b6608c16551c 100644 --- a/test/ELF/tls-opt-iele-i686-nopic.s +++ b/test/ELF/tls-opt-iele-i686-nopic.s @@ -13,8 +13,8 @@ // GOTREL-NEXT: SHF_ALLOC // GOTREL-NEXT: SHF_WRITE // GOTREL-NEXT: ] -// GOTREL-NEXT: Address: 0x12060 -// GOTREL-NEXT: Offset: 0x2060 +// GOTREL-NEXT: Address: 0x12058 +// GOTREL-NEXT: Offset: 0x2058 // GOTREL-NEXT: Size: 8 // GOTREL-NEXT: Link: 0 // GOTREL-NEXT: Info: 0 @@ -23,8 +23,8 @@ // GOTREL-NEXT: } // GOTREL: Relocations [ // GOTREL-NEXT: Section ({{.*}}) .rel.dyn { -// GOTREL-NEXT: 0x12060 R_386_TLS_TPOFF tlsshared0 0x0 -// GOTREL-NEXT: 0x12064 R_386_TLS_TPOFF tlsshared1 0x0 +// GOTREL-NEXT: 0x12058 R_386_TLS_TPOFF tlsshared0 0x0 +// GOTREL-NEXT: 0x1205C R_386_TLS_TPOFF tlsshared1 0x0 // GOTREL-NEXT: } // GOTREL-NEXT: ] @@ -32,24 +32,24 @@ // DISASM-NEXT: _start: // 4294967288 = 0xFFFFFFF8 // 4294967292 = 0xFFFFFFFC -// 73824 = (.got)[0] = 0x12060 -// 73828 = (.got)[1] = 0x12064 -// DISASM-NEXT: 11000: {{.*}} movl $4294967288, %ecx -// DISASM-NEXT: 11006: {{.*}} movl %gs:(%ecx), %eax -// DISASM-NEXT: 11009: {{.*}} movl $4294967288, %eax -// DISASM-NEXT: 1100e: {{.*}} movl %gs:(%eax), %eax -// DISASM-NEXT: 11011: {{.*}} addl $4294967288, %ecx -// DISASM-NEXT: 11017: {{.*}} movl %gs:(%ecx), %eax -// DISASM-NEXT: 1101a: {{.*}} movl $4294967292, %ecx -// DISASM-NEXT: 11020: {{.*}} movl %gs:(%ecx), %eax -// DISASM-NEXT: 11023: {{.*}} movl $4294967292, %eax -// DISASM-NEXT: 11028: {{.*}} movl %gs:(%eax), %eax -// DISASM-NEXT: 1102b: {{.*}} addl $4294967292, %ecx -// DISASM-NEXT: 11031: {{.*}} movl %gs:(%ecx), %eax -// DISASM-NEXT: 11034: {{.*}} movl 73824, %ecx -// DISASM-NEXT: 1103a: {{.*}} movl %gs:(%ecx), %eax -// DISASM-NEXT: 1103d: {{.*}} addl 73828, %ecx -// DISASM-NEXT: 11043: {{.*}} movl %gs:(%ecx), %eax +// 73808 = (.got)[0] = 0x12058 +// 73812 = (.got)[1] = 0x1205C +// DISASM-NEXT: 11000: c7 c1 f8 ff ff ff movl $4294967288, %ecx +// DISASM-NEXT: 11006: 65 8b 01 movl %gs:(%ecx), %eax +// DISASM-NEXT: 11009: b8 f8 ff ff ff movl $4294967288, %eax +// DISASM-NEXT: 1100e: 65 8b 00 movl %gs:(%eax), %eax +// DISASM-NEXT: 11011: 81 c1 f8 ff ff ff addl $4294967288, %ecx +// DISASM-NEXT: 11017: 65 8b 01 movl %gs:(%ecx), %eax +// DISASM-NEXT: 1101a: c7 c1 fc ff ff ff movl $4294967292, %ecx +// DISASM-NEXT: 11020: 65 8b 01 movl %gs:(%ecx), %eax +// DISASM-NEXT: 11023: b8 fc ff ff ff movl $4294967292, %eax +// DISASM-NEXT: 11028: 65 8b 00 movl %gs:(%eax), %eax +// DISASM-NEXT: 1102b: 81 c1 fc ff ff ff addl $4294967292, %ecx +// DISASM-NEXT: 11031: 65 8b 01 movl %gs:(%ecx), %eax +// DISASM-NEXT: 11034: 8b 0d 58 20 01 00 movl 73816, %ecx +// DISASM-NEXT: 1103a: 65 8b 01 movl %gs:(%ecx), %eax +// DISASM-NEXT: 1103d: 03 0d 5c 20 01 00 addl 73820, %ecx +// DISASM-NEXT: 11043: 65 8b 01 movl %gs:(%ecx), %eax .type tlslocal0,@object .section .tbss,"awT",@nobits diff --git a/test/ELF/x86-64-reloc-tpoff32-fpic.s b/test/ELF/x86-64-reloc-tpoff32-fpic.s new file mode 100644 index 000000000000..5be3dc317012 --- /dev/null +++ b/test/ELF/x86-64-reloc-tpoff32-fpic.s @@ -0,0 +1,14 @@ +# REQUIRES: x86 +# RUN: llvm-mc -filetype=obj -triple=x86_64-pc-linux %s -o %t.o +# RUN: not ld.lld %t.o -shared -o %t.so 2>&1 | FileCheck %s + +# CHECK: relocation R_X86_64_TPOFF32 cannot be used against shared object; recompile with -fPIC +# CHECK: >>> defined in {{.*}}.o +# CHECK: >>> referenced by {{.*}}.o:(.tdata+0xC) + +.section ".tdata", "awT", @progbits +.globl var +var: + +movq %fs:0, %rax +leaq var@TPOFF(%rax),%rax diff --git a/unittests/CMakeLists.txt b/unittests/CMakeLists.txt index 9cd085398c37..84d35d43f4e8 100644 --- a/unittests/CMakeLists.txt +++ b/unittests/CMakeLists.txt @@ -12,6 +12,5 @@ function(add_lld_unittest test_dirname) target_link_libraries(${test_dirname} ${LLVM_COMMON_LIBS}) endfunction() -add_subdirectory(CoreTests) add_subdirectory(DriverTests) add_subdirectory(MachOTests) diff --git a/unittests/CoreTests/CMakeLists.txt b/unittests/CoreTests/CMakeLists.txt deleted file mode 100644 index 9f68f56a6c03..000000000000 --- a/unittests/CoreTests/CMakeLists.txt +++ /dev/null @@ -1,7 +0,0 @@ -add_lld_unittest(CoreTests - ParallelTest.cpp - ) - -target_link_libraries(CoreTests - lldCore ${LLVM_PTHREAD_LIB} - ) diff --git a/unittests/CoreTests/ParallelTest.cpp b/unittests/CoreTests/ParallelTest.cpp deleted file mode 100644 index bd8507026a07..000000000000 --- a/unittests/CoreTests/ParallelTest.cpp +++ /dev/null @@ -1,46 +0,0 @@ -//===- lld/unittest/ParallelTest.cpp --------------------------------------===// -// -// The LLVM Compiler Infrastructure -// -// This file is distributed under the University of Illinois Open Source -// License. See LICENSE.TXT for details. -// -//===----------------------------------------------------------------------===// -/// -/// \file -/// \brief Parallel.h unit tests. -/// -//===----------------------------------------------------------------------===// - -#include "gtest/gtest.h" -#include "lld/Core/Parallel.h" -#include -#include - -uint32_t array[1024 * 1024]; - -TEST(Parallel, sort) { - std::mt19937 randEngine; - std::uniform_int_distribution dist; - - for (auto &i : array) - i = dist(randEngine); - - lld::parallel_sort(std::begin(array), std::end(array)); - ASSERT_TRUE(std::is_sorted(std::begin(array), std::end(array))); -} - -TEST(Parallel, parallel_for) { - // We need to test the case with a TaskSize > 1. We are white-box testing - // here. The TaskSize is calculated as (End - Begin) / 1024 at the time of - // writing. - uint32_t range[2050]; - std::fill(range, range + 2050, 1); - lld::parallel_for(0, 2049, [&range](size_t I) { ++range[I]; }); - - uint32_t expected[2049]; - std::fill(expected, expected + 2049, 2); - ASSERT_TRUE(std::equal(range, range + 2049, expected)); - // Check that we don't write past the end of the requested range. - ASSERT_EQ(range[2049], 1u); -} From 7af96fb3afd6725a2824a0a5ca5dad34e5e0b056 Mon Sep 17 00:00:00 2001 From: Dimitry Andric Date: Wed, 17 May 2017 20:22:39 +0000 Subject: [PATCH 4/9] Vendor import of llvm trunk r303291: https://llvm.org/svn/llvm-project/llvm/trunk@303291 --- include/llvm/ADT/APInt.h | 4 +- include/llvm/ADT/BitVector.h | 313 ++-- include/llvm/ADT/PostOrderIterator.h | 33 +- include/llvm/ADT/PriorityWorklist.h | 15 +- include/llvm/ADT/SCCIterator.h | 10 +- include/llvm/ADT/Sequence.h | 21 +- include/llvm/ADT/SetVector.h | 22 +- include/llvm/ADT/SmallBitVector.h | 13 + include/llvm/ADT/SmallPtrSet.h | 30 +- include/llvm/ADT/SmallVector.h | 45 +- include/llvm/ADT/SparseBitVector.h | 20 +- include/llvm/ADT/SparseMultiSet.h | 39 +- include/llvm/ADT/SparseSet.h | 22 +- include/llvm/ADT/StringExtras.h | 22 +- include/llvm/ADT/StringMap.h | 88 +- include/llvm/ADT/StringRef.h | 20 +- include/llvm/ADT/StringSet.h | 17 +- include/llvm/ADT/TinyPtrVector.h | 14 +- include/llvm/ADT/UniqueVector.h | 15 +- include/llvm/Analysis/ProfileSummaryInfo.h | 15 + .../llvm/DebugInfo/CodeView/CVTypeVisitor.h | 32 +- .../CodeView/RandomAccessTypeVisitor.h | 15 - include/llvm/DebugInfo/DWARF/DWARFAttribute.h | 4 +- .../DebugInfo/DWARF/DWARFDebugArangeSet.h | 18 +- .../llvm/DebugInfo/DWARF/DWARFDebugAranges.h | 8 +- .../DebugInfo/DWARF/DWARFDebugRangeList.h | 38 +- include/llvm/DebugInfo/DWARF/DWARFDie.h | 44 +- include/llvm/DebugInfo/DWARF/DWARFFormValue.h | 11 +- include/llvm/DebugInfo/DWARF/DWARFGdbIndex.h | 16 +- include/llvm/DebugInfo/DWARF/DWARFRelocMap.h | 11 +- include/llvm/DebugInfo/DWARF/DWARFUnit.h | 12 +- include/llvm/DebugInfo/PDB/Native/TpiStream.h | 1 + include/llvm/IR/IntrinsicsPowerPC.td | 2 + .../Target/GlobalISel/SelectionDAGCompat.td | 1 + lib/Analysis/DependenceAnalysis.cpp | 33 +- lib/Analysis/InlineCost.cpp | 42 +- lib/Analysis/InstructionSimplify.cpp | 18 + lib/Analysis/ProfileSummaryInfo.cpp | 2 +- lib/Analysis/ScalarEvolution.cpp | 43 +- lib/CodeGen/AggressiveAntiDepBreaker.cpp | 5 +- lib/CodeGen/AsmPrinter/CodeViewDebug.cpp | 2 +- .../AsmPrinter/DbgValueHistoryCalculator.cpp | 3 +- lib/CodeGen/GlobalISel/IRTranslator.cpp | 5 + lib/CodeGen/MachineVerifier.cpp | 2 +- lib/CodeGen/RegAllocGreedy.cpp | 11 +- lib/CodeGen/SelectionDAG/DAGCombiner.cpp | 28 +- lib/CodeGen/SpillPlacement.cpp | 4 +- lib/CodeGen/StackColoring.cpp | 6 +- lib/CodeGen/TargetLoweringBase.cpp | 2 +- lib/CodeGen/TargetPassConfig.cpp | 24 +- lib/CodeGen/TargetRegisterInfo.cpp | 3 +- lib/DebugInfo/CodeView/CVTypeDumper.cpp | 22 +- lib/DebugInfo/CodeView/CVTypeVisitor.cpp | 99 +- .../CodeView/RandomAccessTypeVisitor.cpp | 10 +- lib/DebugInfo/CodeView/TypeDumpVisitor.cpp | 3 +- lib/DebugInfo/CodeView/TypeStreamMerger.cpp | 17 +- lib/DebugInfo/DWARF/DWARFContext.cpp | 63 +- .../PDB/Native/PDBTypeServerHandler.cpp | 3 +- .../RuntimeDyld/RuntimeDyld.cpp | 10 +- lib/Support/CrashRecoveryContext.cpp | 130 +- lib/Support/Unix/Path.inc | 30 +- lib/Target/AArch64/AArch64FrameLowering.cpp | 3 +- lib/Target/AArch64/AArch64ISelLowering.cpp | 341 +--- lib/Target/AArch64/AArch64ISelLowering.h | 1 + .../AArch64/AArch64TargetTransformInfo.cpp | 25 + .../AArch64/AArch64TargetTransformInfo.h | 3 + lib/Target/AMDGPU/AMDGPUISelDAGToDAG.cpp | 30 +- lib/Target/AMDGPU/AMDGPUSubtarget.h | 4 + lib/Target/AMDGPU/SIISelLowering.cpp | 3 +- lib/Target/AMDGPU/SIInstrInfo.cpp | 14 +- lib/Target/AMDGPU/SIRegisterInfo.cpp | 26 +- lib/Target/AMDGPU/VOP3Instructions.td | 20 +- lib/Target/ARM/ARMInstructionSelector.cpp | 15 - lib/Target/ARM/Thumb1FrameLowering.cpp | 3 +- lib/Target/Mips/MipsDelaySlotFiller.cpp | 2 +- lib/Target/PowerPC/PPCFrameLowering.cpp | 23 +- lib/Target/PowerPC/PPCISelLowering.cpp | 45 +- lib/Target/PowerPC/PPCISelLowering.h | 1 + lib/Target/PowerPC/PPCInstr64Bit.td | 4 + lib/Target/PowerPC/PPCInstrInfo.cpp | 13 + lib/Target/PowerPC/PPCInstrInfo.td | 12 +- .../SystemZ/SystemZTargetTransformInfo.cpp | 34 +- .../WebAssembly/WebAssemblyRegColoring.cpp | 3 +- .../WebAssembly/known_gcc_test_failures.txt | 3 - lib/Target/X86/X86.td | 3 - lib/Target/X86/X86FixupLEAs.cpp | 269 +-- lib/Target/X86/X86InstructionSelector.cpp | 66 + lib/Target/X86/X86LegalizerInfo.cpp | 5 + lib/Target/X86/X86Subtarget.h | 6 - lib/Target/X86/X86TargetMachine.cpp | 2 - lib/Target/X86/X86TargetTransformInfo.cpp | 18 + lib/Transforms/Coroutines/CoroFrame.cpp | 28 +- .../InstCombine/InstCombineInternal.h | 21 + .../InstCombine/InstructionCombining.cpp | 47 +- lib/Transforms/Scalar/LICM.cpp | 2 +- lib/Transforms/Scalar/LoopIdiomRecognize.cpp | 16 +- lib/Transforms/Scalar/LoopStrengthReduce.cpp | 3 +- lib/Transforms/Scalar/NewGVN.cpp | 68 +- lib/Transforms/Scalar/Reassociate.cpp | 2 +- lib/Transforms/Scalar/SimpleLoopUnswitch.cpp | 37 +- test/Analysis/CostModel/SystemZ/div-pow2.ll | 154 ++ test/Analysis/CostModel/X86/bitreverse.ll | 69 + test/Analysis/CostModel/X86/ctbits-cost.ll | 587 ------- test/Analysis/CostModel/X86/ctlz.ll | 233 +++ test/Analysis/CostModel/X86/ctpop.ll | 133 ++ test/Analysis/CostModel/X86/cttz.ll | 233 +++ test/CodeGen/AArch64/aarch64-addv.ll | 63 +- test/CodeGen/AArch64/aarch64-minmaxv.ll | 424 +---- test/CodeGen/AArch64/arm64-vabs.ll | 42 +- test/CodeGen/AArch64/ldst-zero.ll | 23 +- test/CodeGen/AArch64/misched-stp.ll | 35 +- test/CodeGen/AMDGPU/fmax3.ll | 101 +- test/CodeGen/AMDGPU/fmin3.ll | 100 +- test/CodeGen/AMDGPU/global-constant.ll | 4 +- test/CodeGen/AMDGPU/immv216.ll | 6 +- test/CodeGen/AMDGPU/max3.ll | 91 +- test/CodeGen/AMDGPU/min3.ll | 131 +- test/CodeGen/AMDGPU/packed-op-sel.ll | 266 +++ .../CodeGen/ARM/2011-02-04-AntidepMultidef.ll | 16 +- .../ARM/2012-10-04-AAPCS-byval-align8.ll | 2 +- test/CodeGen/ARM/dag-combine-ldst.ll | 2 +- test/CodeGen/MSP430/vararg.ll | 1 - test/CodeGen/Mips/msa/bmzi_bmnzi.ll | 8 +- test/CodeGen/PowerPC/atomic-2.ll | 14 +- test/CodeGen/PowerPC/atomics-indexed.ll | 14 +- test/CodeGen/PowerPC/atomics-regression.ll | 64 +- test/CodeGen/PowerPC/atomics.ll | 14 +- test/CodeGen/PowerPC/ppcf128sf.ll | 8 +- test/CodeGen/PowerPC/save-bp.ll | 54 + test/CodeGen/PowerPC/save-cr-ppc32svr4.ll | 46 + test/CodeGen/PowerPC/save-crbp-ppc32svr4.ll | 57 + test/CodeGen/SPARC/32abi.ll | 16 +- test/CodeGen/SPARC/64abi.ll | 18 +- test/CodeGen/SystemZ/swift-return.ll | 6 +- test/CodeGen/Thumb/stack-access.ll | 10 +- test/CodeGen/Thumb2/ldr-str-imm12.ll | 4 +- test/CodeGen/X86/GlobalISel/add-scalar.ll | 94 +- test/CodeGen/X86/GlobalISel/legalize-add.mir | 69 +- .../X86/GlobalISel/regbankselect-X32.mir | 36 + .../CodeGen/X86/GlobalISel/select-add-x32.mir | 63 + test/CodeGen/X86/arg-copy-elide.ll | 7 +- test/CodeGen/X86/leaFixup32.mir | 508 ------ test/CodeGen/X86/leaFixup64.mir | 1041 ------------ test/CodeGen/X86/nontemporal.ll | 72 +- test/CodeGen/X86/psubus.ll | 1443 ++++++++--------- test/CodeGen/X86/store-narrow.ll | 5 +- test/CodeGen/X86/swift-return.ll | 6 +- test/CodeGen/X86/win32-spill-xmm.ll | 2 +- test/CodeGen/X86/win64_sibcall.ll | 4 +- test/CodeGen/X86/win64_vararg.ll | 4 +- test/CodeGen/X86/x86-64-ms_abi-vararg.ll | 4 +- .../RuntimeDyld/X86/ELF_x86-64_debug_frame.s | 20 + test/Feature/optnone-llc.ll | 1 + test/MC/AMDGPU/vop3-gfx9.s | 24 + test/TableGen/GlobalISelEmitter.td | 75 + .../Transforms/Coroutines/coro-catchswitch.ll | 88 + test/Transforms/Inline/inline-hot-callee.ll | 10 +- .../InstCombine/canonicalize_branch.ll | 529 +++++- test/Transforms/InstCombine/debuginfo-skip.ll | 44 + test/Transforms/InstSimplify/AndOrXor.ll | 12 +- .../LoopVectorize/AArch64/pr33053.ll | 56 + .../AArch64/reduction-small-size.ll | 26 +- test/Transforms/NewGVN/pr32934.ll | 1 - .../SLPVectorizer/AArch64/gather-root.ll | 40 +- tools/llvm-pdbdump/Analyze.cpp | 15 +- tools/llvm-pdbdump/LLVMOutputStyle.cpp | 32 +- tools/llvm-pdbdump/PdbYaml.cpp | 14 +- tools/llvm-pdbdump/YamlTypeDumper.cpp | 78 +- unittests/ADT/BitVectorTest.cpp | 184 ++- unittests/Analysis/ProfileSummaryInfoTest.cpp | 8 + .../CodeView/RandomAccessVisitorTest.cpp | 1 + .../DebugInfo/PDB/TypeServerHandlerTest.cpp | 20 +- unittests/Support/BinaryStreamTest.cpp | 29 +- unittests/Support/CMakeLists.txt | 1 + unittests/Support/CrashRecoveryTest.cpp | 83 + utils/TableGen/AsmMatcherEmitter.cpp | 4 +- utils/TableGen/GlobalISelEmitter.cpp | 124 +- utils/lit/lit/main.py | 12 +- utils/lit/lit/run.py | 103 +- 179 files changed, 5516 insertions(+), 5357 deletions(-) create mode 100644 test/Analysis/CostModel/SystemZ/div-pow2.ll delete mode 100644 test/Analysis/CostModel/X86/ctbits-cost.ll create mode 100644 test/Analysis/CostModel/X86/ctlz.ll create mode 100644 test/Analysis/CostModel/X86/ctpop.ll create mode 100644 test/Analysis/CostModel/X86/cttz.ll create mode 100644 test/CodeGen/AMDGPU/packed-op-sel.ll create mode 100644 test/CodeGen/PowerPC/save-bp.ll create mode 100644 test/CodeGen/PowerPC/save-cr-ppc32svr4.ll create mode 100644 test/CodeGen/PowerPC/save-crbp-ppc32svr4.ll create mode 100644 test/CodeGen/X86/GlobalISel/regbankselect-X32.mir create mode 100644 test/CodeGen/X86/GlobalISel/select-add-x32.mir delete mode 100644 test/CodeGen/X86/leaFixup32.mir delete mode 100644 test/CodeGen/X86/leaFixup64.mir create mode 100644 test/ExecutionEngine/RuntimeDyld/X86/ELF_x86-64_debug_frame.s create mode 100644 test/Transforms/Coroutines/coro-catchswitch.ll create mode 100644 test/Transforms/InstCombine/debuginfo-skip.ll create mode 100644 test/Transforms/LoopVectorize/AArch64/pr33053.ll create mode 100644 unittests/Support/CrashRecoveryTest.cpp diff --git a/include/llvm/ADT/APInt.h b/include/llvm/ADT/APInt.h index 94fbd1a29bf9..894e5571f8ad 100644 --- a/include/llvm/ADT/APInt.h +++ b/include/llvm/ADT/APInt.h @@ -1067,9 +1067,7 @@ class LLVM_NODISCARD APInt { /// \returns the bit value at bitPosition bool operator[](unsigned bitPosition) const { assert(bitPosition < getBitWidth() && "Bit position out of bounds!"); - return (maskBit(bitPosition) & - (isSingleWord() ? U.VAL : U.pVal[whichWord(bitPosition)])) != - 0; + return (maskBit(bitPosition) & getWord(bitPosition)) != 0; } /// @} diff --git a/include/llvm/ADT/BitVector.h b/include/llvm/ADT/BitVector.h index 4a2af7cd68a6..e68ef5f53d10 100644 --- a/include/llvm/ADT/BitVector.h +++ b/include/llvm/ADT/BitVector.h @@ -15,6 +15,7 @@ #define LLVM_ADT_BITVECTOR_H #include "llvm/ADT/ArrayRef.h" +#include "llvm/ADT/iterator_range.h" #include "llvm/Support/MathExtras.h" #include #include @@ -26,6 +27,50 @@ namespace llvm { +/// ForwardIterator for the bits that are set. +/// Iterators get invalidated when resize / reserve is called. +template class const_set_bits_iterator_impl { + const BitVectorT &Parent; + int Current = 0; + + void advance() { + assert(Current != -1 && "Trying to advance past end."); + Current = Parent.find_next(Current); + } + +public: + const_set_bits_iterator_impl(const BitVectorT &Parent, int Current) + : Parent(Parent), Current(Current) {} + explicit const_set_bits_iterator_impl(const BitVectorT &Parent) + : const_set_bits_iterator_impl(Parent, Parent.find_first()) {} + const_set_bits_iterator_impl(const const_set_bits_iterator_impl &) = default; + + const_set_bits_iterator_impl operator++(int) { + auto Prev = *this; + advance(); + return Prev; + } + + const_set_bits_iterator_impl &operator++() { + advance(); + return *this; + } + + unsigned operator*() const { return Current; } + + bool operator==(const const_set_bits_iterator_impl &Other) const { + assert(&Parent == &Other.Parent && + "Comparing iterators from different BitVectors"); + return Current == Other.Current; + } + + bool operator!=(const const_set_bits_iterator_impl &Other) const { + assert(&Parent == &Other.Parent && + "Comparing iterators from different BitVectors"); + return Current != Other.Current; + } +}; + class BitVector { typedef unsigned long BitWord; @@ -73,6 +118,18 @@ class BitVector { } }; + typedef const_set_bits_iterator_impl const_set_bits_iterator; + typedef const_set_bits_iterator set_iterator; + + const_set_bits_iterator set_bits_begin() const { + return const_set_bits_iterator(*this); + } + const_set_bits_iterator set_bits_end() const { + return const_set_bits_iterator(*this, -1); + } + iterator_range set_bits() const { + return make_range(set_bits_begin(), set_bits_end()); + } /// BitVector default ctor - Creates an empty bitvector. BitVector() : Size(0) {} @@ -146,138 +203,164 @@ class BitVector { return !any(); } - /// find_first - Returns the index of the first set bit, -1 if none - /// of the bits are set. - int find_first() const { - for (unsigned i = 0; i < NumBitWords(size()); ++i) - if (Bits[i] != 0) - return i * BITWORD_SIZE + countTrailingZeros(Bits[i]); + /// find_first_in - Returns the index of the first set bit in the range + /// [Begin, End). Returns -1 if all bits in the range are unset. + int find_first_in(unsigned Begin, unsigned End) const { + assert(Begin <= End && End <= Size); + if (Begin == End) + return -1; + + unsigned FirstWord = Begin / BITWORD_SIZE; + unsigned LastWord = (End - 1) / BITWORD_SIZE; + + // Check subsequent words. + for (unsigned i = FirstWord; i <= LastWord; ++i) { + BitWord Copy = Bits[i]; + + if (i == FirstWord) { + unsigned FirstBit = Begin % BITWORD_SIZE; + Copy &= maskTrailingZeros(FirstBit); + } + + if (i == LastWord) { + unsigned LastBit = (End - 1) % BITWORD_SIZE; + Copy &= maskTrailingOnes(LastBit + 1); + } + if (Copy != 0) + return i * BITWORD_SIZE + countTrailingZeros(Copy); + } return -1; } + /// find_last_in - Returns the index of the last set bit in the range + /// [Begin, End). Returns -1 if all bits in the range are unset. + int find_last_in(unsigned Begin, unsigned End) const { + assert(Begin <= End && End <= Size); + if (Begin == End) + return -1; + + unsigned LastWord = (End - 1) / BITWORD_SIZE; + unsigned FirstWord = Begin / BITWORD_SIZE; + + for (unsigned i = LastWord + 1; i >= FirstWord + 1; --i) { + unsigned CurrentWord = i - 1; + + BitWord Copy = Bits[CurrentWord]; + if (CurrentWord == LastWord) { + unsigned LastBit = (End - 1) % BITWORD_SIZE; + Copy &= maskTrailingOnes(LastBit + 1); + } + + if (CurrentWord == FirstWord) { + unsigned FirstBit = Begin % BITWORD_SIZE; + Copy &= maskTrailingZeros(FirstBit); + } + + if (Copy != 0) + return (CurrentWord + 1) * BITWORD_SIZE - countLeadingZeros(Copy) - 1; + } + + return -1; + } + + /// find_first_unset_in - Returns the index of the first unset bit in the + /// range [Begin, End). Returns -1 if all bits in the range are set. + int find_first_unset_in(unsigned Begin, unsigned End) const { + assert(Begin <= End && End <= Size); + if (Begin == End) + return -1; + + unsigned FirstWord = Begin / BITWORD_SIZE; + unsigned LastWord = (End - 1) / BITWORD_SIZE; + + // Check subsequent words. + for (unsigned i = FirstWord; i <= LastWord; ++i) { + BitWord Copy = Bits[i]; + + if (i == FirstWord) { + unsigned FirstBit = Begin % BITWORD_SIZE; + Copy |= maskTrailingOnes(FirstBit); + } + + if (i == LastWord) { + unsigned LastBit = (End - 1) % BITWORD_SIZE; + Copy |= maskTrailingZeros(LastBit + 1); + } + if (Copy != ~0UL) { + unsigned Result = i * BITWORD_SIZE + countTrailingOnes(Copy); + return Result < size() ? Result : -1; + } + } + return -1; + } + + /// find_last_unset_in - Returns the index of the last unset bit in the + /// range [Begin, End). Returns -1 if all bits in the range are set. + int find_last_unset_in(unsigned Begin, unsigned End) const { + assert(Begin <= End && End <= Size); + if (Begin == End) + return -1; + + unsigned LastWord = (End - 1) / BITWORD_SIZE; + unsigned FirstWord = Begin / BITWORD_SIZE; + + for (unsigned i = LastWord + 1; i >= FirstWord + 1; --i) { + unsigned CurrentWord = i - 1; + + BitWord Copy = Bits[CurrentWord]; + if (CurrentWord == LastWord) { + unsigned LastBit = (End - 1) % BITWORD_SIZE; + Copy |= maskTrailingZeros(LastBit + 1); + } + + if (CurrentWord == FirstWord) { + unsigned FirstBit = Begin % BITWORD_SIZE; + Copy |= maskTrailingOnes(FirstBit); + } + + if (Copy != ~0UL) { + unsigned Result = + (CurrentWord + 1) * BITWORD_SIZE - countLeadingOnes(Copy) - 1; + return Result < Size ? Result : -1; + } + } + return -1; + } + + /// find_first - Returns the index of the first set bit, -1 if none + /// of the bits are set. + int find_first() const { return find_first_in(0, Size); } + /// find_last - Returns the index of the last set bit, -1 if none of the bits /// are set. - int find_last() const { - if (Size == 0) - return -1; - - unsigned N = NumBitWords(size()); - assert(N > 0); - - unsigned i = N - 1; - while (i > 0 && Bits[i] == BitWord(0)) - --i; - - return int((i + 1) * BITWORD_SIZE - countLeadingZeros(Bits[i])) - 1; - } - - /// find_first_unset - Returns the index of the first unset bit, -1 if all - /// of the bits are set. - int find_first_unset() const { - for (unsigned i = 0; i < NumBitWords(size()); ++i) - if (Bits[i] != ~0UL) { - unsigned Result = i * BITWORD_SIZE + countTrailingOnes(Bits[i]); - return Result < size() ? Result : -1; - } - return -1; - } - - /// find_last_unset - Returns the index of the last unset bit, -1 if all of - /// the bits are set. - int find_last_unset() const { - if (Size == 0) - return -1; - - const unsigned N = NumBitWords(size()); - assert(N > 0); - - unsigned i = N - 1; - BitWord W = Bits[i]; - - // The last word in the BitVector has some unused bits, so we need to set - // them all to 1 first. Set them all to 1 so they don't get treated as - // valid unset bits. - unsigned UnusedCount = BITWORD_SIZE - size() % BITWORD_SIZE; - W |= maskLeadingOnes(UnusedCount); - - while (W == ~BitWord(0) && --i > 0) - W = Bits[i]; - - return int((i + 1) * BITWORD_SIZE - countLeadingOnes(W)) - 1; - } + int find_last() const { return find_last_in(0, Size); } /// find_next - Returns the index of the next set bit following the /// "Prev" bit. Returns -1 if the next set bit is not found. - int find_next(unsigned Prev) const { - ++Prev; - if (Prev >= Size) - return -1; + int find_next(unsigned Prev) const { return find_first_in(Prev + 1, Size); } - unsigned WordPos = Prev / BITWORD_SIZE; - unsigned BitPos = Prev % BITWORD_SIZE; - BitWord Copy = Bits[WordPos]; - // Mask off previous bits. - Copy &= maskTrailingZeros(BitPos); + /// find_prev - Returns the index of the first set bit that precedes the + /// the bit at \p PriorTo. Returns -1 if all previous bits are unset. + int find_prev(unsigned PriorTo) const { return find_last_in(0, PriorTo); } - if (Copy != 0) - return WordPos * BITWORD_SIZE + countTrailingZeros(Copy); - - // Check subsequent words. - for (unsigned i = WordPos+1; i < NumBitWords(size()); ++i) - if (Bits[i] != 0) - return i * BITWORD_SIZE + countTrailingZeros(Bits[i]); - return -1; - } + /// find_first_unset - Returns the index of the first unset bit, -1 if all + /// of the bits are set. + int find_first_unset() const { return find_first_unset_in(0, Size); } /// find_next_unset - Returns the index of the next unset bit following the /// "Prev" bit. Returns -1 if all remaining bits are set. int find_next_unset(unsigned Prev) const { - ++Prev; - if (Prev >= Size) - return -1; - - unsigned WordPos = Prev / BITWORD_SIZE; - unsigned BitPos = Prev % BITWORD_SIZE; - BitWord Copy = Bits[WordPos]; - // Mask in previous bits. - BitWord Mask = (1 << BitPos) - 1; - Copy |= Mask; - - if (Copy != ~0UL) - return next_unset_in_word(WordPos, Copy); - - // Check subsequent words. - for (unsigned i = WordPos + 1; i < NumBitWords(size()); ++i) - if (Bits[i] != ~0UL) - return next_unset_in_word(i, Bits[i]); - return -1; + return find_first_unset_in(Prev + 1, Size); } - /// find_prev - Returns the index of the first set bit that precedes the - /// the bit at \p PriorTo. Returns -1 if all previous bits are unset. - int find_prev(unsigned PriorTo) const { - if (PriorTo == 0) - return -1; + /// find_last_unset - Returns the index of the last unset bit, -1 if all of + /// the bits are set. + int find_last_unset() const { return find_last_unset_in(0, Size); } - --PriorTo; - - unsigned WordPos = PriorTo / BITWORD_SIZE; - unsigned BitPos = PriorTo % BITWORD_SIZE; - BitWord Copy = Bits[WordPos]; - // Mask off next bits. - Copy &= maskTrailingOnes(BitPos + 1); - - if (Copy != 0) - return (WordPos + 1) * BITWORD_SIZE - countLeadingZeros(Copy) - 1; - - // Check previous words. - for (unsigned i = 1; i <= WordPos; ++i) { - unsigned Index = WordPos - i; - if (Bits[Index] == 0) - continue; - return (Index + 1) * BITWORD_SIZE - countLeadingZeros(Bits[Index]) - 1; - } - return -1; + /// find_prev_unset - Returns the index of the first unset bit that precedes + /// the bit at \p PriorTo. Returns -1 if all previous bits are set. + int find_prev_unset(unsigned PriorTo) { + return find_last_unset_in(0, PriorTo); } /// clear - Removes all bits from the bitvector. Does not change capacity. diff --git a/include/llvm/ADT/PostOrderIterator.h b/include/llvm/ADT/PostOrderIterator.h index 8fc08eb252eb..a179d29956b1 100644 --- a/include/llvm/ADT/PostOrderIterator.h +++ b/include/llvm/ADT/PostOrderIterator.h @@ -96,24 +96,14 @@ template , public po_iterator_storage { - typedef std::iterator super; - typedef typename GT::NodeRef NodeRef; - typedef typename GT::ChildIteratorType ChildItTy; + using super = std::iterator; + using NodeRef = typename GT::NodeRef; + using ChildItTy = typename GT::ChildIteratorType; // VisitStack - Used to maintain the ordering. Top = current block // First element is basic block pointer, second is the 'next child' to visit std::vector> VisitStack; - void traverseChild() { - while (VisitStack.back().second != GT::child_end(VisitStack.back().first)) { - NodeRef BB = *VisitStack.back().second++; - if (this->insertEdge(Optional(VisitStack.back().first), BB)) { - // If the block is not visited... - VisitStack.push_back(std::make_pair(BB, GT::child_begin(BB))); - } - } - } - po_iterator(NodeRef BB) { this->insertEdge(Optional(), BB); VisitStack.push_back(std::make_pair(BB, GT::child_begin(BB))); @@ -134,8 +124,18 @@ class po_iterator : po_iterator_storage(S) { } // End is when stack is empty. + void traverseChild() { + while (VisitStack.back().second != GT::child_end(VisitStack.back().first)) { + NodeRef BB = *VisitStack.back().second++; + if (this->insertEdge(Optional(VisitStack.back().first), BB)) { + // If the block is not visited... + VisitStack.push_back(std::make_pair(BB, GT::child_begin(BB))); + } + } + } + public: - typedef typename super::pointer pointer; + using pointer = typename super::pointer; // Provide static "constructors"... static po_iterator begin(GraphT G) { @@ -286,7 +286,8 @@ inverse_post_order_ext(const T &G, SetType &S) { template> class ReversePostOrderTraversal { - typedef typename GT::NodeRef NodeRef; + using NodeRef = typename GT::NodeRef; + std::vector Blocks; // Block list in normal PO order void Initialize(NodeRef BB) { @@ -294,7 +295,7 @@ class ReversePostOrderTraversal { } public: - typedef typename std::vector::reverse_iterator rpo_iterator; + using rpo_iterator = typename std::vector::reverse_iterator; ReversePostOrderTraversal(GraphT G) { Initialize(GT::getEntryNode(G)); } diff --git a/include/llvm/ADT/PriorityWorklist.h b/include/llvm/ADT/PriorityWorklist.h index 3198dd438700..35891e931801 100644 --- a/include/llvm/ADT/PriorityWorklist.h +++ b/include/llvm/ADT/PriorityWorklist.h @@ -17,13 +17,14 @@ #define LLVM_ADT_PRIORITYWORKLIST_H #include "llvm/ADT/DenseMap.h" -#include "llvm/ADT/STLExtras.h" -#include "llvm/ADT/Sequence.h" #include "llvm/ADT/SmallVector.h" +#include "llvm/ADT/STLExtras.h" #include "llvm/Support/Compiler.h" #include #include #include +#include +#include #include namespace llvm { @@ -55,11 +56,11 @@ template , typename MapT = DenseMap> class PriorityWorklist { public: - typedef T value_type; - typedef T key_type; - typedef T& reference; - typedef const T& const_reference; - typedef typename MapT::size_type size_type; + using value_type = T; + using key_type = T; + using reference = T&; + using const_reference = const T&; + using size_type = typename MapT::size_type; /// Construct an empty PriorityWorklist PriorityWorklist() = default; diff --git a/include/llvm/ADT/SCCIterator.h b/include/llvm/ADT/SCCIterator.h index 9a8a7b168fce..734a58f87da2 100644 --- a/include/llvm/ADT/SCCIterator.h +++ b/include/llvm/ADT/SCCIterator.h @@ -1,4 +1,4 @@ -//===---- ADT/SCCIterator.h - Strongly Connected Comp. Iter. ----*- C++ -*-===// +//===- ADT/SCCIterator.h - Strongly Connected Comp. Iter. -------*- C++ -*-===// // // The LLVM Compiler Infrastructure // @@ -43,10 +43,10 @@ template > class scc_iterator : public iterator_facade_base< scc_iterator, std::forward_iterator_tag, const std::vector, ptrdiff_t> { - typedef typename GT::NodeRef NodeRef; - typedef typename GT::ChildIteratorType ChildItTy; - typedef std::vector SccTy; - typedef typename scc_iterator::reference reference; + using NodeRef = typename GT::NodeRef; + using ChildItTy = typename GT::ChildIteratorType; + using SccTy = std::vector; + using reference = typename scc_iterator::reference; /// Element of VisitStack during DFS. struct StackElement { diff --git a/include/llvm/ADT/Sequence.h b/include/llvm/ADT/Sequence.h index 5d36831cc128..3d4a897bf9a9 100644 --- a/include/llvm/ADT/Sequence.h +++ b/include/llvm/ADT/Sequence.h @@ -13,27 +13,31 @@ /// //===----------------------------------------------------------------------===// -#ifndef LLVM_ADT_SEQ_H -#define LLVM_ADT_SEQ_H +#ifndef LLVM_ADT_SEQUENCE_H +#define LLVM_ADT_SEQUENCE_H #include "llvm/ADT/iterator.h" #include "llvm/ADT/iterator_range.h" +#include +#include +#include namespace llvm { namespace detail { + template class value_sequence_iterator : public iterator_facade_base, std::random_access_iterator_tag, const ValueT> { - typedef typename value_sequence_iterator::iterator_facade_base BaseT; + using BaseT = typename value_sequence_iterator::iterator_facade_base; ValueT Value; public: - typedef typename BaseT::difference_type difference_type; - typedef typename BaseT::reference reference; + using difference_type = typename BaseT::difference_type; + using reference = typename BaseT::reference; value_sequence_iterator() = default; value_sequence_iterator(const value_sequence_iterator &) = default; @@ -65,7 +69,8 @@ class value_sequence_iterator reference operator*() const { return Value; } }; -} // End detail namespace. + +} // end namespace detail template iterator_range> seq(ValueT Begin, @@ -74,6 +79,6 @@ iterator_range> seq(ValueT Begin, detail::value_sequence_iterator(End)); } -} +} // end namespace llvm -#endif +#endif // LLVM_ADT_SEQUENCE_H diff --git a/include/llvm/ADT/SetVector.h b/include/llvm/ADT/SetVector.h index 13378aa3a04e..04ed52fc543f 100644 --- a/include/llvm/ADT/SetVector.h +++ b/include/llvm/ADT/SetVector.h @@ -40,17 +40,17 @@ template , typename Set = DenseSet> class SetVector { public: - typedef T value_type; - typedef T key_type; - typedef T& reference; - typedef const T& const_reference; - typedef Set set_type; - typedef Vector vector_type; - typedef typename vector_type::const_iterator iterator; - typedef typename vector_type::const_iterator const_iterator; - typedef typename vector_type::const_reverse_iterator reverse_iterator; - typedef typename vector_type::const_reverse_iterator const_reverse_iterator; - typedef typename vector_type::size_type size_type; + using value_type = T; + using key_type = T; + using reference = T&; + using const_reference = const T&; + using set_type = Set; + using vector_type = Vector; + using iterator = typename vector_type::const_iterator; + using const_iterator = typename vector_type::const_iterator; + using reverse_iterator = typename vector_type::const_reverse_iterator; + using const_reverse_iterator = typename vector_type::const_reverse_iterator; + using size_type = typename vector_type::size_type; /// \brief Construct an empty SetVector SetVector() = default; diff --git a/include/llvm/ADT/SmallBitVector.h b/include/llvm/ADT/SmallBitVector.h index 0eeacc162543..0ff427066959 100644 --- a/include/llvm/ADT/SmallBitVector.h +++ b/include/llvm/ADT/SmallBitVector.h @@ -134,6 +134,19 @@ class SmallBitVector { } public: + typedef const_set_bits_iterator_impl const_set_bits_iterator; + typedef const_set_bits_iterator set_iterator; + + const_set_bits_iterator set_bits_begin() const { + return const_set_bits_iterator(*this); + } + const_set_bits_iterator set_bits_end() const { + return const_set_bits_iterator(*this, -1); + } + iterator_range set_bits() const { + return make_range(set_bits_begin(), set_bits_end()); + } + /// Creates an empty bitvector. SmallBitVector() : X(1) {} diff --git a/include/llvm/ADT/SmallPtrSet.h b/include/llvm/ADT/SmallPtrSet.h index 196ab6338047..b49d216e0b6e 100644 --- a/include/llvm/ADT/SmallPtrSet.h +++ b/include/llvm/ADT/SmallPtrSet.h @@ -27,15 +27,13 @@ #include #include -#if LLVM_ENABLE_ABI_BREAKING_CHECKS namespace llvm { + +#if LLVM_ENABLE_ABI_BREAKING_CHECKS template struct ReverseIterate { static bool value; }; template bool ReverseIterate::value = false; -} #endif -namespace llvm { - /// SmallPtrSetImplBase - This is the common code shared among all the /// SmallPtrSet<>'s, which is almost everything. SmallPtrSet has two modes, one /// for small and one for large sets. @@ -92,7 +90,7 @@ class SmallPtrSetImplBase { } public: - typedef unsigned size_type; + using size_type = unsigned; SmallPtrSetImplBase &operator=(const SmallPtrSetImplBase &) = delete; @@ -273,14 +271,14 @@ class SmallPtrSetIteratorImpl { /// SmallPtrSetIterator - This implements a const_iterator for SmallPtrSet. template class SmallPtrSetIterator : public SmallPtrSetIteratorImpl { - typedef PointerLikeTypeTraits PtrTraits; + using PtrTraits = PointerLikeTypeTraits; public: - typedef PtrTy value_type; - typedef PtrTy reference; - typedef PtrTy pointer; - typedef std::ptrdiff_t difference_type; - typedef std::forward_iterator_tag iterator_category; + using value_type = PtrTy; + using reference = PtrTy; + using pointer = PtrTy; + using difference_type = std::ptrdiff_t; + using iterator_category = std::forward_iterator_tag; explicit SmallPtrSetIterator(const void *const *BP, const void *const *E) : SmallPtrSetIteratorImpl(BP, E) {} @@ -351,8 +349,8 @@ struct RoundUpToPowerOfTwo { template class SmallPtrSetImpl : public SmallPtrSetImplBase { using ConstPtrType = typename add_const_past_pointer::type; - typedef PointerLikeTypeTraits PtrTraits; - typedef PointerLikeTypeTraits ConstPtrTraits; + using PtrTraits = PointerLikeTypeTraits; + using ConstPtrTraits = PointerLikeTypeTraits; protected: // Constructors that forward to the base. @@ -365,8 +363,8 @@ class SmallPtrSetImpl : public SmallPtrSetImplBase { : SmallPtrSetImplBase(SmallStorage, SmallSize) {} public: - typedef SmallPtrSetIterator iterator; - typedef SmallPtrSetIterator const_iterator; + using iterator = SmallPtrSetIterator; + using const_iterator = SmallPtrSetIterator; SmallPtrSetImpl(const SmallPtrSetImpl &) = delete; @@ -431,7 +429,7 @@ class SmallPtrSet : public SmallPtrSetImpl { // DenseSet<> instead if you expect many elements in the set. static_assert(SmallSize <= 32, "SmallSize should be small"); - typedef SmallPtrSetImpl BaseT; + using BaseT = SmallPtrSetImpl; // Make sure that SmallSize is a power of two, round up if not. enum { SmallSizePowTwo = RoundUpToPowerOfTwo::Val }; diff --git a/include/llvm/ADT/SmallVector.h b/include/llvm/ADT/SmallVector.h index b9588214023c..bd24eab93b50 100644 --- a/include/llvm/ADT/SmallVector.h +++ b/include/llvm/ADT/SmallVector.h @@ -71,7 +71,7 @@ class SmallVectorTemplateCommon : public SmallVectorBase { // Allocate raw space for N elements of type T. If T has a ctor or dtor, we // don't want it to be automatically run, so we need to represent the space as // something else. Use an array of char of sufficient alignment. - typedef AlignedCharArrayUnion U; + using U = AlignedCharArrayUnion; U FirstEl; // Space after 'FirstEl' is clobbered, do not add any instance vars after it. @@ -96,19 +96,19 @@ class SmallVectorTemplateCommon : public SmallVectorBase { void setEnd(T *P) { this->EndX = P; } public: - typedef size_t size_type; - typedef ptrdiff_t difference_type; - typedef T value_type; - typedef T *iterator; - typedef const T *const_iterator; + using size_type = size_t; + using difference_type = ptrdiff_t; + using value_type = T; + using iterator = T *; + using const_iterator = const T *; - typedef std::reverse_iterator const_reverse_iterator; - typedef std::reverse_iterator reverse_iterator; + using const_reverse_iterator = std::reverse_iterator; + using reverse_iterator = std::reverse_iterator; - typedef T &reference; - typedef const T &const_reference; - typedef T *pointer; - typedef const T *const_pointer; + using reference = T &; + using const_reference = const T &; + using pointer = T *; + using const_pointer = const T *; // forward iterator creation methods. LLVM_ATTRIBUTE_ALWAYS_INLINE @@ -319,12 +319,12 @@ class SmallVectorTemplateBase : public SmallVectorTemplateCommon { /// reduce code duplication based on the SmallVector 'N' template parameter. template class SmallVectorImpl : public SmallVectorTemplateBase::value> { - typedef SmallVectorTemplateBase::value > SuperClass; + using SuperClass = SmallVectorTemplateBase::value>; public: - typedef typename SuperClass::iterator iterator; - typedef typename SuperClass::const_iterator const_iterator; - typedef typename SuperClass::size_type size_type; + using iterator = typename SuperClass::iterator; + using const_iterator = typename SuperClass::const_iterator; + using size_type = typename SuperClass::size_type; protected: // Default ctor - Initialize to empty. @@ -845,8 +845,7 @@ class SmallVector : public SmallVectorImpl { SmallVectorStorage Storage; public: - SmallVector() : SmallVectorImpl(N) { - } + SmallVector() : SmallVectorImpl(N) {} explicit SmallVector(size_t Size, const T &Value = T()) : SmallVectorImpl(N) { @@ -883,16 +882,16 @@ class SmallVector : public SmallVectorImpl { SmallVectorImpl::operator=(::std::move(RHS)); } - const SmallVector &operator=(SmallVector &&RHS) { - SmallVectorImpl::operator=(::std::move(RHS)); - return *this; - } - SmallVector(SmallVectorImpl &&RHS) : SmallVectorImpl(N) { if (!RHS.empty()) SmallVectorImpl::operator=(::std::move(RHS)); } + const SmallVector &operator=(SmallVector &&RHS) { + SmallVectorImpl::operator=(::std::move(RHS)); + return *this; + } + const SmallVector &operator=(SmallVectorImpl &&RHS) { SmallVectorImpl::operator=(::std::move(RHS)); return *this; diff --git a/include/llvm/ADT/SparseBitVector.h b/include/llvm/ADT/SparseBitVector.h index a82cef6028f9..4cbf40c76805 100644 --- a/include/llvm/ADT/SparseBitVector.h +++ b/include/llvm/ADT/SparseBitVector.h @@ -1,4 +1,4 @@ -//===- llvm/ADT/SparseBitVector.h - Efficient Sparse BitVector -*- C++ -*- ===// +//===- llvm/ADT/SparseBitVector.h - Efficient Sparse BitVector --*- C++ -*-===// // // The LLVM Compiler Infrastructure // @@ -41,8 +41,8 @@ namespace llvm { template struct SparseBitVectorElement { public: - typedef unsigned long BitWord; - typedef unsigned size_type; + using BitWord = unsigned long; + using size_type = unsigned; enum { BITWORD_SIZE = sizeof(BitWord) * CHAR_BIT, BITWORDS_PER_ELEMENT = (ElementSize + BITWORD_SIZE - 1) / BITWORD_SIZE, @@ -100,7 +100,7 @@ template struct SparseBitVectorElement { Bits[Idx / BITWORD_SIZE] |= 1L << (Idx % BITWORD_SIZE); } - bool test_and_set (unsigned Idx) { + bool test_and_set(unsigned Idx) { bool old = test(Idx); if (!old) { set(Idx); @@ -254,9 +254,9 @@ template struct SparseBitVectorElement { template class SparseBitVector { - typedef std::list> ElementList; - typedef typename ElementList::iterator ElementListIter; - typedef typename ElementList::const_iterator ElementListConstIter; + using ElementList = std::list>; + using ElementListIter = typename ElementList::iterator; + using ElementListConstIter = typename ElementList::const_iterator; enum { BITWORD_SIZE = SparseBitVectorElement::BITWORD_SIZE }; @@ -421,14 +421,12 @@ class SparseBitVector { }; public: - typedef SparseBitVectorIterator iterator; + using iterator = SparseBitVectorIterator; SparseBitVector() { CurrElementIter = Elements.begin(); } - ~SparseBitVector() = default; - // SparseBitVector copy ctor. SparseBitVector(const SparseBitVector &RHS) { ElementListConstIter ElementIter = RHS.Elements.begin(); @@ -440,6 +438,8 @@ class SparseBitVector { CurrElementIter = Elements.begin (); } + ~SparseBitVector() = default; + // Clear. void clear() { Elements.clear(); diff --git a/include/llvm/ADT/SparseMultiSet.h b/include/llvm/ADT/SparseMultiSet.h index 08da4b68ebaa..b3a413aa3aa5 100644 --- a/include/llvm/ADT/SparseMultiSet.h +++ b/include/llvm/ADT/SparseMultiSet.h @@ -1,4 +1,4 @@ -//===--- llvm/ADT/SparseMultiSet.h - Sparse multiset ------------*- C++ -*-===// +//===- llvm/ADT/SparseMultiSet.h - Sparse multiset --------------*- C++ -*-===// // // The LLVM Compiler Infrastructure // @@ -101,7 +101,7 @@ class SparseMultiSet { unsigned Prev; unsigned Next; - SMSNode(ValueT D, unsigned P, unsigned N) : Data(D), Prev(P), Next(N) { } + SMSNode(ValueT D, unsigned P, unsigned N) : Data(D), Prev(P), Next(N) {} /// List tails have invalid Nexts. bool isTail() const { @@ -118,8 +118,8 @@ class SparseMultiSet { bool isValid() const { return Prev != INVALID; } }; - typedef typename KeyFunctorT::argument_type KeyT; - typedef SmallVector DenseT; + using KeyT = typename KeyFunctorT::argument_type; + using DenseT = SmallVector; DenseT Dense; SparseT *Sparse = nullptr; unsigned Universe = 0; @@ -183,12 +183,12 @@ class SparseMultiSet { } public: - typedef ValueT value_type; - typedef ValueT &reference; - typedef const ValueT &const_reference; - typedef ValueT *pointer; - typedef const ValueT *const_pointer; - typedef unsigned size_type; + using value_type = ValueT; + using reference = ValueT &; + using const_reference = const ValueT &; + using pointer = ValueT *; + using const_pointer = const ValueT *; + using size_type = unsigned; SparseMultiSet() = default; SparseMultiSet(const SparseMultiSet &) = delete; @@ -227,7 +227,7 @@ class SparseMultiSet { unsigned SparseIdx; iterator_base(SMSPtrTy P, unsigned I, unsigned SI) - : SMS(P), Idx(I), SparseIdx(SI) { } + : SMS(P), Idx(I), SparseIdx(SI) {} /// Whether our iterator has fallen outside our dense vector. bool isEnd() const { @@ -248,11 +248,11 @@ class SparseMultiSet { void setNext(unsigned N) { SMS->Dense[Idx].Next = N; } public: - typedef std::iterator super; - typedef typename super::value_type value_type; - typedef typename super::difference_type difference_type; - typedef typename super::pointer pointer; - typedef typename super::reference reference; + using super = std::iterator; + using value_type = typename super::value_type; + using difference_type = typename super::difference_type; + using pointer = typename super::pointer; + using reference = typename super::reference; reference operator*() const { assert(isKeyed() && SMS->sparseIndex(SMS->Dense[Idx].Data) == SparseIdx && @@ -308,11 +308,12 @@ class SparseMultiSet { return I; } }; - typedef iterator_base iterator; - typedef iterator_base const_iterator; + + using iterator = iterator_base; + using const_iterator = iterator_base; // Convenience types - typedef std::pair RangePair; + using RangePair = std::pair; /// Returns an iterator past this container. Note that such an iterator cannot /// be decremented, but will compare equal to other end iterators. diff --git a/include/llvm/ADT/SparseSet.h b/include/llvm/ADT/SparseSet.h index 00c18c743219..25ade8831922 100644 --- a/include/llvm/ADT/SparseSet.h +++ b/include/llvm/ADT/SparseSet.h @@ -1,4 +1,4 @@ -//===--- llvm/ADT/SparseSet.h - Sparse set ----------------------*- C++ -*-===// +//===- llvm/ADT/SparseSet.h - Sparse set ------------------------*- C++ -*-===// // // The LLVM Compiler Infrastructure // @@ -125,9 +125,9 @@ class SparseSet { !std::numeric_limits::is_signed, "SparseT must be an unsigned integer type"); - typedef typename KeyFunctorT::argument_type KeyT; - typedef SmallVector DenseT; - typedef unsigned size_type; + using KeyT = typename KeyFunctorT::argument_type; + using DenseT = SmallVector; + using size_type = unsigned; DenseT Dense; SparseT *Sparse = nullptr; unsigned Universe = 0; @@ -135,11 +135,11 @@ class SparseSet { SparseSetValFunctor ValIndexOf; public: - typedef ValueT value_type; - typedef ValueT &reference; - typedef const ValueT &const_reference; - typedef ValueT *pointer; - typedef const ValueT *const_pointer; + using value_type = ValueT; + using reference = ValueT &; + using const_reference = const ValueT &; + using pointer = ValueT *; + using const_pointer = const ValueT *; SparseSet() = default; SparseSet(const SparseSet &) = delete; @@ -168,8 +168,8 @@ class SparseSet { } // Import trivial vector stuff from DenseT. - typedef typename DenseT::iterator iterator; - typedef typename DenseT::const_iterator const_iterator; + using iterator = typename DenseT::iterator; + using const_iterator = typename DenseT::const_iterator; const_iterator begin() const { return Dense.begin(); } const_iterator end() const { return Dense.end(); } diff --git a/include/llvm/ADT/StringExtras.h b/include/llvm/ADT/StringExtras.h index 1c109be3bab3..e22a3f688c40 100644 --- a/include/llvm/ADT/StringExtras.h +++ b/include/llvm/ADT/StringExtras.h @@ -1,4 +1,4 @@ -//===-- llvm/ADT/StringExtras.h - Useful string functions -------*- C++ -*-===// +//===- llvm/ADT/StringExtras.h - Useful string functions --------*- C++ -*-===// // // The LLVM Compiler Infrastructure // @@ -15,12 +15,18 @@ #define LLVM_ADT_STRINGEXTRAS_H #include "llvm/ADT/StringRef.h" -#include "llvm/Support/DataTypes.h" #include +#include +#include +#include +#include +#include +#include namespace llvm { -class raw_ostream; + template class SmallVectorImpl; +class raw_ostream; /// hexdigit - Return the hexadecimal character for the /// given number \p X (which should be less than 16). @@ -128,7 +134,6 @@ static inline std::string utostr(uint64_t X, bool isNeg = false) { return std::string(BufPtr, std::end(Buffer)); } - static inline std::string itostr(int64_t X) { if (X < 0) return utostr(static_cast(-X), true); @@ -261,13 +266,14 @@ template inline size_t join_items_size(const A1 &A, Args &&... Items) { return join_one_item_size(A) + join_items_size(std::forward(Items)...); } -} + +} // end namespace detail /// Joins the strings in the range [Begin, End), adding Separator between /// the elements. template inline std::string join(IteratorT Begin, IteratorT End, StringRef Separator) { - typedef typename std::iterator_traits::iterator_category tag; + using tag = typename std::iterator_traits::iterator_category; return detail::join_impl(Begin, End, Separator, tag()); } @@ -295,6 +301,6 @@ inline std::string join_items(Sep Separator, Args &&... Items) { return Result; } -} // End llvm namespace +} // end namespace llvm -#endif +#endif // LLVM_ADT_STRINGEXTRAS_H diff --git a/include/llvm/ADT/StringMap.h b/include/llvm/ADT/StringMap.h index c36fda7d6906..d573148665a1 100644 --- a/include/llvm/ADT/StringMap.h +++ b/include/llvm/ADT/StringMap.h @@ -1,4 +1,4 @@ -//===--- StringMap.h - String Hash table map interface ----------*- C++ -*-===// +//===- StringMap.h - String Hash table map interface ------------*- C++ -*-===// // // The LLVM Compiler Infrastructure // @@ -16,25 +16,23 @@ #include "llvm/ADT/StringRef.h" #include "llvm/ADT/iterator.h" +#include "llvm/ADT/iterator_range.h" #include "llvm/Support/Allocator.h" #include "llvm/Support/PointerLikeTypeTraits.h" +#include #include #include #include #include #include -#include +#include #include namespace llvm { - template - class StringMapConstIterator; - template - class StringMapIterator; - template class StringMapKeyIterator; - template - class StringMapEntry; +template class StringMapConstIterator; +template class StringMapIterator; +template class StringMapKeyIterator; /// StringMapEntryBase - Shared base class of StringMapEntry instances. class StringMapEntryBase { @@ -53,17 +51,15 @@ class StringMapImpl { // Array of NumBuckets pointers to entries, null pointers are holes. // TheTable[NumBuckets] contains a sentinel value for easy iteration. Followed // by an array of the actual hash values as unsigned integers. - StringMapEntryBase **TheTable; - unsigned NumBuckets; - unsigned NumItems; - unsigned NumTombstones; + StringMapEntryBase **TheTable = nullptr; + unsigned NumBuckets = 0; + unsigned NumItems = 0; + unsigned NumTombstones = 0; unsigned ItemSize; protected: explicit StringMapImpl(unsigned itemSize) - : TheTable(nullptr), - // Initialize the map with zero buckets to allocation. - NumBuckets(0), NumItems(0), NumTombstones(0), ItemSize(itemSize) {} + : ItemSize(itemSize) {} StringMapImpl(StringMapImpl &&RHS) : TheTable(RHS.TheTable), NumBuckets(RHS.NumBuckets), NumItems(RHS.NumItems), NumTombstones(RHS.NumTombstones), @@ -225,9 +221,10 @@ class StringMap : public StringMapImpl { AllocatorTy Allocator; public: - typedef StringMapEntry MapEntryTy; + using MapEntryTy = StringMapEntry; StringMap() : StringMapImpl(static_cast(sizeof(MapEntryTy))) {} + explicit StringMap(unsigned InitialSize) : StringMapImpl(InitialSize, static_cast(sizeof(MapEntryTy))) {} @@ -248,12 +245,6 @@ class StringMap : public StringMapImpl { StringMap(StringMap &&RHS) : StringMapImpl(std::move(RHS)), Allocator(std::move(RHS.Allocator)) {} - StringMap &operator=(StringMap RHS) { - StringMapImpl::swap(RHS); - std::swap(Allocator, RHS.Allocator); - return *this; - } - StringMap(const StringMap &RHS) : StringMapImpl(static_cast(sizeof(MapEntryTy))), Allocator(RHS.Allocator) { @@ -289,16 +280,37 @@ class StringMap : public StringMapImpl { // not worthwhile. } + StringMap &operator=(StringMap RHS) { + StringMapImpl::swap(RHS); + std::swap(Allocator, RHS.Allocator); + return *this; + } + + ~StringMap() { + // Delete all the elements in the map, but don't reset the elements + // to default values. This is a copy of clear(), but avoids unnecessary + // work not required in the destructor. + if (!empty()) { + for (unsigned I = 0, E = NumBuckets; I != E; ++I) { + StringMapEntryBase *Bucket = TheTable[I]; + if (Bucket && Bucket != getTombstoneVal()) { + static_cast(Bucket)->Destroy(Allocator); + } + } + } + free(TheTable); + } + AllocatorTy &getAllocator() { return Allocator; } const AllocatorTy &getAllocator() const { return Allocator; } - typedef const char* key_type; - typedef ValueTy mapped_type; - typedef StringMapEntry value_type; - typedef size_t size_type; + using key_type = const char*; + using mapped_type = ValueTy; + using value_type = StringMapEntry; + using size_type = size_t; - typedef StringMapConstIterator const_iterator; - typedef StringMapIterator iterator; + using const_iterator = StringMapConstIterator; + using iterator = StringMapIterator; iterator begin() { return iterator(TheTable, NumBuckets == 0); @@ -313,7 +325,7 @@ class StringMap : public StringMapImpl { return const_iterator(TheTable+NumBuckets, true); } - llvm::iterator_range> keys() const { + iterator_range> keys() const { return make_range(StringMapKeyIterator(begin()), StringMapKeyIterator(end())); } @@ -433,21 +445,6 @@ class StringMap : public StringMapImpl { erase(I); return true; } - - ~StringMap() { - // Delete all the elements in the map, but don't reset the elements - // to default values. This is a copy of clear(), but avoids unnecessary - // work not required in the destructor. - if (!empty()) { - for (unsigned I = 0, E = NumBuckets; I != E; ++I) { - StringMapEntryBase *Bucket = TheTable[I]; - if (Bucket && Bucket != getTombstoneVal()) { - static_cast(Bucket)->Destroy(Allocator); - } - } - } - free(TheTable); - } }; template @@ -542,7 +539,6 @@ class StringMapKeyIterator public: StringMapKeyIterator() = default; - explicit StringMapKeyIterator(StringMapConstIterator Iter) : base(std::move(Iter)) {} diff --git a/include/llvm/ADT/StringRef.h b/include/llvm/ADT/StringRef.h index ce48f6d3bad3..4b25f56432df 100644 --- a/include/llvm/ADT/StringRef.h +++ b/include/llvm/ADT/StringRef.h @@ -1,4 +1,4 @@ -//===--- StringRef.h - Constant String Reference Wrapper --------*- C++ -*-===// +//===- StringRef.h - Constant String Reference Wrapper ----------*- C++ -*-===// // // The LLVM Compiler Infrastructure // @@ -15,16 +15,18 @@ #include "llvm/Support/Compiler.h" #include #include +#include #include #include +#include #include #include namespace llvm { - template - class SmallVectorImpl; + class APInt; class hash_code; + template class SmallVectorImpl; class StringRef; /// Helper functions for StringRef::getAsInteger. @@ -46,10 +48,11 @@ namespace llvm { /// general safe to store a StringRef. class StringRef { public: - typedef const char *iterator; - typedef const char *const_iterator; static const size_t npos = ~size_t(0); - typedef size_t size_type; + + using iterator = const char *; + using const_iterator = const char *; + using size_type = size_t; private: /// The start of the string, in an external buffer. @@ -906,6 +909,7 @@ namespace llvm { // StringRefs can be treated like a POD type. template struct isPodLike; template <> struct isPodLike { static const bool value = true; }; -} -#endif +} // end namespace llvm + +#endif // LLVM_ADT_STRINGREF_H diff --git a/include/llvm/ADT/StringSet.h b/include/llvm/ADT/StringSet.h index c32c2a497438..9af44c07df79 100644 --- a/include/llvm/ADT/StringSet.h +++ b/include/llvm/ADT/StringSet.h @@ -1,4 +1,4 @@ -//===--- StringSet.h - The LLVM Compiler Driver -----------------*- C++ -*-===// +//===- StringSet.h - The LLVM Compiler Driver -------------------*- C++ -*-===// // // The LLVM Compiler Infrastructure // @@ -15,13 +15,19 @@ #define LLVM_ADT_STRINGSET_H #include "llvm/ADT/StringMap.h" +#include "llvm/ADT/StringRef.h" +#include "llvm/Support/Allocator.h" +#include +#include +#include namespace llvm { /// StringSet - A wrapper for StringMap that provides set-like functionality. - template - class StringSet : public llvm::StringMap { - typedef llvm::StringMap base; + template + class StringSet : public StringMap { + using base = StringMap; + public: StringSet() = default; StringSet(std::initializer_list S) { @@ -40,6 +46,7 @@ namespace llvm { base::insert(std::make_pair(*It, '\0')); } }; -} + +} // end namespace llvm #endif // LLVM_ADT_STRINGSET_H diff --git a/include/llvm/ADT/TinyPtrVector.h b/include/llvm/ADT/TinyPtrVector.h index ca43b6046193..79740713f75b 100644 --- a/include/llvm/ADT/TinyPtrVector.h +++ b/include/llvm/ADT/TinyPtrVector.h @@ -30,9 +30,9 @@ namespace llvm { template class TinyPtrVector { public: - typedef SmallVector VecTy; - typedef typename VecTy::value_type value_type; - typedef PointerUnion PtrUnion; + using VecTy = SmallVector; + using value_type = typename VecTy::value_type; + using PtrUnion = PointerUnion; private: PtrUnion Val; @@ -167,10 +167,10 @@ class TinyPtrVector { return Val.template get()->size(); } - typedef EltTy *iterator; - typedef const EltTy *const_iterator; - typedef std::reverse_iterator reverse_iterator; - typedef std::reverse_iterator const_reverse_iterator; + using iterator = EltTy *; + using const_iterator = const EltTy *; + using reverse_iterator = std::reverse_iterator; + using const_reverse_iterator = std::reverse_iterator; iterator begin() { if (Val.template is()) diff --git a/include/llvm/ADT/UniqueVector.h b/include/llvm/ADT/UniqueVector.h index e1ab4b56023f..b17fb2392baf 100644 --- a/include/llvm/ADT/UniqueVector.h +++ b/include/llvm/ADT/UniqueVector.h @@ -1,4 +1,4 @@ -//===-- llvm/ADT/UniqueVector.h ---------------------------------*- C++ -*-===// +//===- llvm/ADT/UniqueVector.h ----------------------------------*- C++ -*-===// // // The LLVM Compiler Infrastructure // @@ -24,16 +24,15 @@ namespace llvm { /// Entries can be fetched using operator[] with the entry ID. template class UniqueVector { public: - typedef typename std::vector VectorType; - typedef typename VectorType::iterator iterator; - typedef typename VectorType::const_iterator const_iterator; + using VectorType = typename std::vector; + using iterator = typename VectorType::iterator; + using const_iterator = typename VectorType::const_iterator; private: // Map - Used to handle the correspondence of entry to ID. std::map Map; // Vector - ID ordered vector of entries. Entries can be indexed by ID - 1. - // VectorType Vector; public: @@ -68,7 +67,6 @@ template class UniqueVector { } /// operator[] - Returns a reference to the entry with the specified ID. - /// const T &operator[](unsigned ID) const { assert(ID-1 < size() && "ID is 0 or out of range!"); return Vector[ID - 1]; @@ -87,21 +85,18 @@ template class UniqueVector { const_iterator end() const { return Vector.end(); } /// size - Returns the number of entries in the vector. - /// size_t size() const { return Vector.size(); } /// empty - Returns true if the vector is empty. - /// bool empty() const { return Vector.empty(); } /// reset - Clears all the entries. - /// void reset() { Map.clear(); Vector.resize(0, 0); } }; -} // End of namespace llvm +} // end namespace llvm #endif // LLVM_ADT_UNIQUEVECTOR_H diff --git a/include/llvm/Analysis/ProfileSummaryInfo.h b/include/llvm/Analysis/ProfileSummaryInfo.h index c5f97083af4d..6aaabe1d1889 100644 --- a/include/llvm/Analysis/ProfileSummaryInfo.h +++ b/include/llvm/Analysis/ProfileSummaryInfo.h @@ -55,6 +55,21 @@ class ProfileSummaryInfo { ProfileSummaryInfo(ProfileSummaryInfo &&Arg) : M(Arg.M), Summary(std::move(Arg.Summary)) {} + /// \brief Returns true if profile summary is available. + bool hasProfileSummary() { return computeSummary(); } + + /// \brief Returns true if module \c M has sample profile. + bool hasSampleProfile() { + return hasProfileSummary() && + Summary->getKind() == ProfileSummary::PSK_Sample; + } + + /// \brief Returns true if module \c M has instrumentation profile. + bool hasInstrumentationProfile() { + return hasProfileSummary() && + Summary->getKind() == ProfileSummary::PSK_Instr; + } + /// Handle the invalidation of this information. /// /// When used as a result of \c ProfileSummaryAnalysis this method will be diff --git a/include/llvm/DebugInfo/CodeView/CVTypeVisitor.h b/include/llvm/DebugInfo/CodeView/CVTypeVisitor.h index f3122f0bf7f0..6d9f345755ab 100644 --- a/include/llvm/DebugInfo/CodeView/CVTypeVisitor.h +++ b/include/llvm/DebugInfo/CodeView/CVTypeVisitor.h @@ -28,7 +28,7 @@ class CVTypeVisitor { Error visitTypeRecord(CVType &Record, TypeIndex Index); Error visitTypeRecord(CVType &Record); - Error visitMemberRecord(CVMemberRecord &Record); + Error visitMemberRecord(CVMemberRecord Record); /// Visits the type records in Data. Sets the error flag on parse failures. Error visitTypeStream(const CVTypeArray &Types); @@ -47,6 +47,36 @@ class CVTypeVisitor { TinyPtrVector Handlers; }; +enum VisitorDataSource { + VDS_BytesPresent, // The record bytes are passed into the the visitation + // function. The algorithm should first deserialize them + // before passing them on through the pipeline. + VDS_BytesExternal // The record bytes are not present, and it is the + // responsibility of the visitor callback interface to + // supply the bytes. +}; + +Error visitTypeRecord(CVType &Record, TypeIndex Index, + TypeVisitorCallbacks &Callbacks, + VisitorDataSource Source = VDS_BytesPresent, + TypeServerHandler *TS = nullptr); +Error visitTypeRecord(CVType &Record, TypeVisitorCallbacks &Callbacks, + VisitorDataSource Source = VDS_BytesPresent, + TypeServerHandler *TS = nullptr); + +Error visitMemberRecord(CVMemberRecord Record, TypeVisitorCallbacks &Callbacks, + VisitorDataSource Source = VDS_BytesPresent); +Error visitMemberRecord(TypeLeafKind Kind, ArrayRef Record, + TypeVisitorCallbacks &Callbacks); + +Error visitMemberRecordStream(ArrayRef FieldList, + TypeVisitorCallbacks &Callbacks); + +Error visitTypeStream(const CVTypeArray &Types, TypeVisitorCallbacks &Callbacks, + TypeServerHandler *TS = nullptr); +Error visitTypeStream(CVTypeRange Types, TypeVisitorCallbacks &Callbacks, + TypeServerHandler *TS = nullptr); + } // end namespace codeview } // end namespace llvm diff --git a/include/llvm/DebugInfo/CodeView/RandomAccessTypeVisitor.h b/include/llvm/DebugInfo/CodeView/RandomAccessTypeVisitor.h index 35a8010f1163..21288df89be2 100644 --- a/include/llvm/DebugInfo/CodeView/RandomAccessTypeVisitor.h +++ b/include/llvm/DebugInfo/CodeView/RandomAccessTypeVisitor.h @@ -11,13 +11,10 @@ #define LLVM_DEBUGINFO_CODEVIEW_RANDOMACCESSTYPEVISITOR_H #include "llvm/ADT/TinyPtrVector.h" -#include "llvm/DebugInfo/CodeView/CVTypeVisitor.h" #include "llvm/DebugInfo/CodeView/TypeDatabase.h" #include "llvm/DebugInfo/CodeView/TypeDatabaseVisitor.h" -#include "llvm/DebugInfo/CodeView/TypeDeserializer.h" #include "llvm/DebugInfo/CodeView/TypeIndex.h" #include "llvm/DebugInfo/CodeView/TypeRecord.h" -#include "llvm/DebugInfo/CodeView/TypeVisitorCallbackPipeline.h" #include "llvm/Support/Error.h" namespace llvm { @@ -73,18 +70,6 @@ class RandomAccessTypeVisitor { /// The database visitor which adds new records to the database. TypeDatabaseVisitor DatabaseVisitor; - /// The deserializer which deserializes new records. - TypeDeserializer Deserializer; - - /// The visitation callback pipeline to use. By default this contains a - /// deserializer and a type database visitor. But the callback specified - /// in the constructor is also added. - TypeVisitorCallbackPipeline Pipeline; - - /// The visitor used to visit the internal pipeline for deserialization and - /// database maintenance. - CVTypeVisitor InternalVisitor; - /// A vector mapping type indices to type offset. For every record that has /// been visited, contains the absolute offset of that record in the record /// array. diff --git a/include/llvm/DebugInfo/DWARF/DWARFAttribute.h b/include/llvm/DebugInfo/DWARF/DWARFAttribute.h index 5919aaddea40..c3953b62d780 100644 --- a/include/llvm/DebugInfo/DWARF/DWARFAttribute.h +++ b/include/llvm/DebugInfo/DWARF/DWARFAttribute.h @@ -31,10 +31,10 @@ struct DWARFAttribute { dwarf::Attribute Attr; /// The form and value for this attribute. DWARFFormValue Value; - + DWARFAttribute(uint32_t O, dwarf::Attribute A = dwarf::Attribute(0), dwarf::Form F = dwarf::Form(0)) : Attr(A), Value(F) {} - + bool isValid() const { return Offset != 0 && Attr != dwarf::Attribute(0); } diff --git a/include/llvm/DebugInfo/DWARF/DWARFDebugArangeSet.h b/include/llvm/DebugInfo/DWARF/DWARFDebugArangeSet.h index 40eb7e9a8836..2d82104ea098 100644 --- a/include/llvm/DebugInfo/DWARF/DWARFDebugArangeSet.h +++ b/include/llvm/DebugInfo/DWARF/DWARFDebugArangeSet.h @@ -22,19 +22,19 @@ class raw_ostream; class DWARFDebugArangeSet { public: struct Header { - // The total length of the entries for that set, not including the length - // field itself. + /// The total length of the entries for that set, not including the length + /// field itself. uint32_t Length; - // The offset from the beginning of the .debug_info section of the - // compilation unit entry referenced by the table. + /// The offset from the beginning of the .debug_info section of the + /// compilation unit entry referenced by the table. uint32_t CuOffset; - // The DWARF version number. + /// The DWARF version number. uint16_t Version; - // The size in bytes of an address on the target architecture. For segmented - // addressing, this is the size of the offset portion of the address. + /// The size in bytes of an address on the target architecture. For segmented + /// addressing, this is the size of the offset portion of the address. uint8_t AddrSize; - // The size in bytes of a segment descriptor on the target architecture. - // If the target system uses a flat address space, this value is 0. + /// The size in bytes of a segment descriptor on the target architecture. + /// If the target system uses a flat address space, this value is 0. uint8_t SegSize; }; diff --git a/include/llvm/DebugInfo/DWARF/DWARFDebugAranges.h b/include/llvm/DebugInfo/DWARF/DWARFDebugAranges.h index c06771d6afb4..2237aa361d18 100644 --- a/include/llvm/DebugInfo/DWARF/DWARFDebugAranges.h +++ b/include/llvm/DebugInfo/DWARF/DWARFDebugAranges.h @@ -28,7 +28,7 @@ class DWARFDebugAranges { void clear(); void extract(DataExtractor DebugArangesData); - // Call appendRange multiple times and then call construct. + /// Call appendRange multiple times and then call construct. void appendRange(uint32_t CUOffset, uint64_t LowPC, uint64_t HighPC); void construct(); @@ -58,9 +58,9 @@ class DWARFDebugAranges { return LowPC < other.LowPC; } - uint64_t LowPC; // Start of address range. - uint32_t Length; // End of address range (not including this address). - uint32_t CUOffset; // Offset of the compile unit or die. + uint64_t LowPC; /// Start of address range. + uint32_t Length; /// End of address range (not including this address). + uint32_t CUOffset; /// Offset of the compile unit or die. }; struct RangeEndpoint { diff --git a/include/llvm/DebugInfo/DWARF/DWARFDebugRangeList.h b/include/llvm/DebugInfo/DWARF/DWARFDebugRangeList.h index 23a573b7a9fa..95ec1be62a79 100644 --- a/include/llvm/DebugInfo/DWARF/DWARFDebugRangeList.h +++ b/include/llvm/DebugInfo/DWARF/DWARFDebugRangeList.h @@ -33,31 +33,31 @@ typedef std::vector DWARFAddressRangesVector; class DWARFDebugRangeList { public: struct RangeListEntry { - // A beginning address offset. This address offset has the size of an - // address and is relative to the applicable base address of the - // compilation unit referencing this range list. It marks the beginning - // of an address range. + /// A beginning address offset. This address offset has the size of an + /// address and is relative to the applicable base address of the + /// compilation unit referencing this range list. It marks the beginning + /// of an address range. uint64_t StartAddress; - // An ending address offset. This address offset again has the size of - // an address and is relative to the applicable base address of the - // compilation unit referencing this range list. It marks the first - // address past the end of the address range. The ending address must - // be greater than or equal to the beginning address. + /// An ending address offset. This address offset again has the size of + /// an address and is relative to the applicable base address of the + /// compilation unit referencing this range list. It marks the first + /// address past the end of the address range. The ending address must + /// be greater than or equal to the beginning address. uint64_t EndAddress; - // The end of any given range list is marked by an end of list entry, - // which consists of a 0 for the beginning address offset - // and a 0 for the ending address offset. + /// The end of any given range list is marked by an end of list entry, + /// which consists of a 0 for the beginning address offset + /// and a 0 for the ending address offset. bool isEndOfListEntry() const { return (StartAddress == 0) && (EndAddress == 0); } - // A base address selection entry consists of: - // 1. The value of the largest representable address offset - // (for example, 0xffffffff when the size of an address is 32 bits). - // 2. An address, which defines the appropriate base address for - // use in interpreting the beginning and ending address offsets of - // subsequent entries of the location list. + /// A base address selection entry consists of: + /// 1. The value of the largest representable address offset + /// (for example, 0xffffffff when the size of an address is 32 bits). + /// 2. An address, which defines the appropriate base address for + /// use in interpreting the beginning and ending address offsets of + /// subsequent entries of the location list. bool isBaseAddressSelectionEntry(uint8_t AddressSize) const { assert(AddressSize == 4 || AddressSize == 8); if (AddressSize == 4) @@ -68,7 +68,7 @@ class DWARFDebugRangeList { }; private: - // Offset in .debug_ranges section. + /// Offset in .debug_ranges section. uint32_t Offset; uint8_t AddressSize; std::vector Entries; diff --git a/include/llvm/DebugInfo/DWARF/DWARFDie.h b/include/llvm/DebugInfo/DWARF/DWARFDie.h index ee06125ea278..ca94a90fabfc 100644 --- a/include/llvm/DebugInfo/DWARF/DWARFDie.h +++ b/include/llvm/DebugInfo/DWARF/DWARFDie.h @@ -24,10 +24,10 @@ #include namespace llvm { - + class DWARFUnit; class raw_ostream; - + //===----------------------------------------------------------------------===// /// Utility class that carries the DWARF compile/type unit and the debug info /// entry in an object. @@ -47,7 +47,7 @@ class DWARFDie { public: DWARFDie() = default; DWARFDie(DWARFUnit *Unit, const DWARFDebugInfoEntry * D) : U(Unit), Die(D) {} - + bool isValid() const { return U && Die; } explicit operator bool() const { return isValid(); } const DWARFDebugInfoEntry *getDebugInfoEntry() const { return Die; } @@ -68,7 +68,7 @@ class DWARFDie { assert(isValid() && "must check validity prior to calling"); return Die->getOffset(); } - + dwarf::Tag getTag() const { auto AbbrevDecl = getAbbreviationDeclarationPtr(); if (AbbrevDecl) @@ -80,7 +80,7 @@ class DWARFDie { assert(isValid() && "must check validity prior to calling"); return Die->hasChildren(); } - + /// Returns true for a valid DIE that terminates a sibling chain. bool isNULL() const { return getAbbreviationDeclarationPtr() == nullptr; @@ -97,13 +97,13 @@ class DWARFDie { /// \returns a valid DWARFDie instance if this object has a parent or an /// invalid DWARFDie instance if it doesn't. DWARFDie getParent() const; - + /// Get the sibling of this DIE object. /// /// \returns a valid DWARFDie instance if this object has a sibling or an /// invalid DWARFDie instance if it doesn't. DWARFDie getSibling() const; - + /// Get the first child of this DIE object. /// /// \returns a valid DWARFDie instance if this object has children or an @@ -113,7 +113,7 @@ class DWARFDie { return DWARFDie(U, Die + 1); return DWARFDie(); } - + /// Dump the DIE and all of its attributes to the supplied stream. /// /// \param OS the stream to use for output. @@ -121,7 +121,7 @@ class DWARFDie { /// children. /// \param indent the number of characters to indent each line that is output. void dump(raw_ostream &OS, unsigned recurseDepth, unsigned indent = 0) const; - + /// Extract the specified attribute from this DIE. /// /// Extract an attribute value from this DIE only. This call doesn't look @@ -132,7 +132,7 @@ class DWARFDie { /// \returns an optional DWARFFormValue that will have the form value if the /// attribute was successfully extracted. Optional find(dwarf::Attribute Attr) const; - + /// Extract the first value of any attribute in Attrs from this DIE. /// /// Extract the first attribute that matches from this DIE only. This call @@ -180,7 +180,7 @@ class DWARFDie { /// /// \returns anm optional absolute section offset value for the attribute. Optional getRangesBaseAttribute() const; - + /// Get the DW_AT_high_pc attribute value as an address. /// /// In DWARF version 4 and later the high PC can be encoded as an offset from @@ -196,7 +196,7 @@ class DWARFDie { /// Retrieves DW_AT_low_pc and DW_AT_high_pc from CU. /// Returns true if both attributes are present. bool getLowAndHighPC(uint64_t &LowPC, uint64_t &HighPC) const; - + /// Get the address ranges for this DIE. /// /// Get the hi/low PC range if both attributes are available or exrtracts the @@ -208,7 +208,7 @@ class DWARFDie { /// \returns a address range vector that might be empty if no address range /// information is available. DWARFAddressRangesVector getAddressRanges() const; - + /// Get all address ranges for any DW_TAG_subprogram DIEs in this DIE or any /// of its children. /// @@ -218,19 +218,19 @@ class DWARFDie { /// /// \param Ranges the addres range vector to fill in. void collectChildrenAddressRanges(DWARFAddressRangesVector &Ranges) const; - + bool addressRangeContainsAddress(const uint64_t Address) const; - + /// If a DIE represents a subprogram (or inlined subroutine), returns its /// mangled name (or short name, if mangled is missing). This name may be /// fetched from specification or abstract origin for this subprogram. /// Returns null if no name is found. const char *getSubroutineName(DINameKind Kind) const; - + /// Return the DIE name resolving DW_AT_sepcification or DW_AT_abstract_origin /// references if necessary. Returns null if no name is found. const char *getName(DINameKind Kind) const; - + /// Returns the declaration line (start line) for a DIE, assuming it specifies /// a subprogram. This may be fetched from specification or abstract origin /// for this subprogram by resolving DW_AT_sepcification or @@ -251,21 +251,21 @@ class DWARFDie { /// there is no DW_AT_GNU_discriminator attribute in this DIE. void getCallerFrame(uint32_t &CallFile, uint32_t &CallLine, uint32_t &CallColumn, uint32_t &CallDiscriminator) const; - + class attribute_iterator; /// Get an iterator range to all attributes in the current DIE only. /// /// \returns an iterator range for the attributes of the current DIE. iterator_range attributes() const; - + class iterator; - + iterator begin() const; iterator end() const; iterator_range children() const; }; - + class DWARFDie::attribute_iterator : public iterator_facade_base { @@ -275,7 +275,7 @@ class DWARFDie::attribute_iterator : DWARFAttribute AttrValue; /// The attribute index within the abbreviation declaration in Die. uint32_t Index; - + /// Update the attribute index and attempt to read the attribute value. If the /// attribute is able to be read, update AttrValue and the Index member /// variable. If the attribute value is not able to be read, an appropriate diff --git a/include/llvm/DebugInfo/DWARF/DWARFFormValue.h b/include/llvm/DebugInfo/DWARF/DWARFFormValue.h index f3516ebdecba..a30e0be9c3c3 100644 --- a/include/llvm/DebugInfo/DWARF/DWARFFormValue.h +++ b/include/llvm/DebugInfo/DWARF/DWARFFormValue.h @@ -49,9 +49,9 @@ class DWARFFormValue { const uint8_t *data = nullptr; }; - dwarf::Form Form; // Form for this value. - ValueType Value; // Contains all data for the form. - const DWARFUnit *U = nullptr; // Remember the DWARFUnit at extract time. + dwarf::Form Form; /// Form for this value. + ValueType Value; /// Contains all data for the form. + const DWARFUnit *U = nullptr; /// Remember the DWARFUnit at extract time. public: DWARFFormValue(dwarf::Form F = dwarf::Form(0)) : Form(F) {} @@ -72,11 +72,14 @@ class DWARFFormValue { const DWARFUnit *getUnit() const { return U; } void dump(raw_ostream &OS) const; - /// \brief extracts a value in data at offset *offset_ptr. + /// Extracts a value in \p Data at offset \p *OffsetPtr. /// /// The passed DWARFUnit is allowed to be nullptr, in which /// case no relocation processing will be performed and some /// kind of forms that depend on Unit information are disallowed. + /// \param Data The DataExtractor to use. + /// \param OffsetPtr The offset within DataExtractor where the data starts. + /// \param U The optional DWARFUnit supplying information for some forms. /// \returns whether the extraction succeeded. bool extractValue(const DataExtractor &Data, uint32_t *OffsetPtr, const DWARFUnit *U); diff --git a/include/llvm/DebugInfo/DWARF/DWARFGdbIndex.h b/include/llvm/DebugInfo/DWARF/DWARFGdbIndex.h index 7a52218663b9..8d1ac5c83c23 100644 --- a/include/llvm/DebugInfo/DWARF/DWARFGdbIndex.h +++ b/include/llvm/DebugInfo/DWARF/DWARFGdbIndex.h @@ -29,25 +29,25 @@ class DWARFGdbIndex { uint32_t ConstantPoolOffset; struct CompUnitEntry { - uint64_t Offset; // Offset of a CU in the .debug_info section. - uint64_t Length; // Length of that CU. + uint64_t Offset; /// Offset of a CU in the .debug_info section. + uint64_t Length; /// Length of that CU. }; SmallVector CuList; struct AddressEntry { - uint64_t LowAddress; // The low address. - uint64_t HighAddress; // The high address. - uint32_t CuIndex; // The CU index. + uint64_t LowAddress; /// The low address. + uint64_t HighAddress; /// The high address. + uint32_t CuIndex; /// The CU index. }; SmallVector AddressArea; struct SymTableEntry { - uint32_t NameOffset; // Offset of the symbol's name in the constant pool. - uint32_t VecOffset; // Offset of the CU vector in the constant pool. + uint32_t NameOffset; /// Offset of the symbol's name in the constant pool. + uint32_t VecOffset; /// Offset of the CU vector in the constant pool. }; SmallVector SymbolTable; - // Each value is CU index + attributes. + /// Each value is CU index + attributes. SmallVector>, 0> ConstantPoolVectors; diff --git a/include/llvm/DebugInfo/DWARF/DWARFRelocMap.h b/include/llvm/DebugInfo/DWARF/DWARFRelocMap.h index f1e03bb4c2e1..ec0397a0fb09 100644 --- a/include/llvm/DebugInfo/DWARF/DWARFRelocMap.h +++ b/include/llvm/DebugInfo/DWARF/DWARFRelocMap.h @@ -17,15 +17,14 @@ namespace llvm { struct RelocAddrEntry { - uint8_t Width; int64_t Value; }; -// In place of applying the relocations to the data we've read from disk we use -// a separate mapping table to the side and checking that at locations in the -// dwarf where we expect relocated values. This adds a bit of complexity to the -// dwarf parsing/extraction at the benefit of not allocating memory for the -// entire size of the debug info sections. +/// In place of applying the relocations to the data we've read from disk we use +/// a separate mapping table to the side and checking that at locations in the +/// dwarf where we expect relocated values. This adds a bit of complexity to the +/// dwarf parsing/extraction at the benefit of not allocating memory for the +/// entire size of the debug info sections. typedef DenseMap RelocAddrMap; } // end namespace llvm diff --git a/include/llvm/DebugInfo/DWARF/DWARFUnit.h b/include/llvm/DebugInfo/DWARF/DWARFUnit.h index 68e541bac73c..c15e27f36a8b 100644 --- a/include/llvm/DebugInfo/DWARF/DWARFUnit.h +++ b/include/llvm/DebugInfo/DWARF/DWARFUnit.h @@ -111,7 +111,7 @@ class DWARFUnitSection final : public SmallVector, 1>, class DWARFUnit { DWARFContext &Context; - // Section containing this DWARFUnit. + /// Section containing this DWARFUnit. const DWARFSection &InfoSection; const DWARFDebugAbbrev *Abbrev; @@ -133,12 +133,12 @@ class DWARFUnit { uint8_t UnitType; uint8_t AddrSize; uint64_t BaseAddr; - // The compile unit debug information entry items. + /// The compile unit debug information entry items. std::vector DieArray; - // Map from range's start address to end address and corresponding DIE. - // IntervalMap does not support range removal, as a result, we use the - // std::map::upper_bound for address range lookup. + /// Map from range's start address to end address and corresponding DIE. + /// IntervalMap does not support range removal, as a result, we use the + /// std::map::upper_bound for address range lookup. std::map> AddrDieMap; typedef iterator_range::iterator> die_iterator_range; @@ -189,7 +189,7 @@ class DWARFUnit { AddrOffsetSectionBase = Base; } - // Recursively update address to Die map. + /// Recursively update address to Die map. void updateAddressDieMap(DWARFDie Die); void setRangesSection(const DWARFSection *RS, uint32_t Base) { diff --git a/include/llvm/DebugInfo/PDB/Native/TpiStream.h b/include/llvm/DebugInfo/PDB/Native/TpiStream.h index 4579cbf4227b..c5549983ed43 100644 --- a/include/llvm/DebugInfo/PDB/Native/TpiStream.h +++ b/include/llvm/DebugInfo/PDB/Native/TpiStream.h @@ -51,6 +51,7 @@ class TpiStream { HashTable &getHashAdjusters(); codeview::CVTypeRange types(bool *HadError) const; + const codeview::CVTypeArray &typeArray() const { return TypeRecords; } Error commit(); diff --git a/include/llvm/IR/IntrinsicsPowerPC.td b/include/llvm/IR/IntrinsicsPowerPC.td index 64240a929782..6321bb81b8cb 100644 --- a/include/llvm/IR/IntrinsicsPowerPC.td +++ b/include/llvm/IR/IntrinsicsPowerPC.td @@ -1132,4 +1132,6 @@ def int_ppc_tsuspend : GCCBuiltin<"__builtin_tsuspend">, def int_ppc_ttest : GCCBuiltin<"__builtin_ttest">, Intrinsic<[llvm_i64_ty], [], []>; + +def int_ppc_cfence : Intrinsic<[], [llvm_anyint_ty], []>; } diff --git a/include/llvm/Target/GlobalISel/SelectionDAGCompat.td b/include/llvm/Target/GlobalISel/SelectionDAGCompat.td index a06c67fe814c..071ec2edb538 100644 --- a/include/llvm/Target/GlobalISel/SelectionDAGCompat.td +++ b/include/llvm/Target/GlobalISel/SelectionDAGCompat.td @@ -62,6 +62,7 @@ def : GINodeEquiv; def : GINodeEquiv; def : GINodeEquiv; def : GINodeEquiv; +def : GINodeEquiv; def : GINodeEquiv; // Specifies the GlobalISel equivalents for SelectionDAG's ComplexPattern. diff --git a/lib/Analysis/DependenceAnalysis.cpp b/lib/Analysis/DependenceAnalysis.cpp index a4672efeedd6..e4d58bf1b4eb 100644 --- a/lib/Analysis/DependenceAnalysis.cpp +++ b/lib/Analysis/DependenceAnalysis.cpp @@ -2984,7 +2984,7 @@ bool DependenceInfo::propagate(const SCEV *&Src, const SCEV *&Dst, SmallVectorImpl &Constraints, bool &Consistent) { bool Result = false; - for (int LI = Loops.find_first(); LI >= 0; LI = Loops.find_next(LI)) { + for (unsigned LI : Loops.set_bits()) { DEBUG(dbgs() << "\t Constraint[" << LI << "] is"); DEBUG(Constraints[LI].dump(dbgs())); if (Constraints[LI].isDistance()) @@ -3266,7 +3266,7 @@ bool DependenceInfo::tryDelinearize(Instruction *Src, Instruction *Dst, // For debugging purposes, dump a small bit vector to dbgs(). static void dumpSmallBitVector(SmallBitVector &BV) { dbgs() << "{"; - for (int VI = BV.find_first(); VI >= 0; VI = BV.find_next(VI)) { + for (unsigned VI : BV.set_bits()) { dbgs() << VI; if (BV.find_next(VI) >= 0) dbgs() << ' '; @@ -3506,7 +3506,7 @@ DependenceInfo::depends(Instruction *Src, Instruction *Dst, NewConstraint.setAny(SE); // test separable subscripts - for (int SI = Separable.find_first(); SI >= 0; SI = Separable.find_next(SI)) { + for (unsigned SI : Separable.set_bits()) { DEBUG(dbgs() << "testing subscript " << SI); switch (Pair[SI].Classification) { case Subscript::ZIV: @@ -3545,14 +3545,14 @@ DependenceInfo::depends(Instruction *Src, Instruction *Dst, SmallVector Constraints(MaxLevels + 1); for (unsigned II = 0; II <= MaxLevels; ++II) Constraints[II].setAny(SE); - for (int SI = Coupled.find_first(); SI >= 0; SI = Coupled.find_next(SI)) { + for (unsigned SI : Coupled.set_bits()) { DEBUG(dbgs() << "testing subscript group " << SI << " { "); SmallBitVector Group(Pair[SI].Group); SmallBitVector Sivs(Pairs); SmallBitVector Mivs(Pairs); SmallBitVector ConstrainedLevels(MaxLevels + 1); SmallVector PairsInGroup; - for (int SJ = Group.find_first(); SJ >= 0; SJ = Group.find_next(SJ)) { + for (unsigned SJ : Group.set_bits()) { DEBUG(dbgs() << SJ << " "); if (Pair[SJ].Classification == Subscript::SIV) Sivs.set(SJ); @@ -3564,7 +3564,7 @@ DependenceInfo::depends(Instruction *Src, Instruction *Dst, DEBUG(dbgs() << "}\n"); while (Sivs.any()) { bool Changed = false; - for (int SJ = Sivs.find_first(); SJ >= 0; SJ = Sivs.find_next(SJ)) { + for (unsigned SJ : Sivs.set_bits()) { DEBUG(dbgs() << "testing subscript " << SJ << ", SIV\n"); // SJ is an SIV subscript that's part of the current coupled group unsigned Level; @@ -3588,7 +3588,7 @@ DependenceInfo::depends(Instruction *Src, Instruction *Dst, DEBUG(dbgs() << " propagating\n"); DEBUG(dbgs() << "\tMivs = "); DEBUG(dumpSmallBitVector(Mivs)); - for (int SJ = Mivs.find_first(); SJ >= 0; SJ = Mivs.find_next(SJ)) { + for (unsigned SJ : Mivs.set_bits()) { // SJ is an MIV subscript that's part of the current coupled group DEBUG(dbgs() << "\tSJ = " << SJ << "\n"); if (propagate(Pair[SJ].Src, Pair[SJ].Dst, Pair[SJ].Loops, @@ -3622,7 +3622,7 @@ DependenceInfo::depends(Instruction *Src, Instruction *Dst, } // test & propagate remaining RDIVs - for (int SJ = Mivs.find_first(); SJ >= 0; SJ = Mivs.find_next(SJ)) { + for (unsigned SJ : Mivs.set_bits()) { if (Pair[SJ].Classification == Subscript::RDIV) { DEBUG(dbgs() << "RDIV test\n"); if (testRDIV(Pair[SJ].Src, Pair[SJ].Dst, Result)) @@ -3635,7 +3635,7 @@ DependenceInfo::depends(Instruction *Src, Instruction *Dst, // test remaining MIVs // This code is temporary. // Better to somehow test all remaining subscripts simultaneously. - for (int SJ = Mivs.find_first(); SJ >= 0; SJ = Mivs.find_next(SJ)) { + for (unsigned SJ : Mivs.set_bits()) { if (Pair[SJ].Classification == Subscript::MIV) { DEBUG(dbgs() << "MIV test\n"); if (testMIV(Pair[SJ].Src, Pair[SJ].Dst, Pair[SJ].Loops, Result)) @@ -3647,9 +3647,8 @@ DependenceInfo::depends(Instruction *Src, Instruction *Dst, // update Result.DV from constraint vector DEBUG(dbgs() << " updating\n"); - for (int SJ = ConstrainedLevels.find_first(); SJ >= 0; - SJ = ConstrainedLevels.find_next(SJ)) { - if (SJ > (int)CommonLevels) + for (unsigned SJ : ConstrainedLevels.set_bits()) { + if (SJ > CommonLevels) break; updateDirection(Result.DV[SJ - 1], Constraints[SJ]); if (Result.DV[SJ - 1].Direction == Dependence::DVEntry::NONE) @@ -3859,7 +3858,7 @@ const SCEV *DependenceInfo::getSplitIteration(const Dependence &Dep, NewConstraint.setAny(SE); // test separable subscripts - for (int SI = Separable.find_first(); SI >= 0; SI = Separable.find_next(SI)) { + for (unsigned SI : Separable.set_bits()) { switch (Pair[SI].Classification) { case Subscript::SIV: { unsigned Level; @@ -3886,12 +3885,12 @@ const SCEV *DependenceInfo::getSplitIteration(const Dependence &Dep, SmallVector Constraints(MaxLevels + 1); for (unsigned II = 0; II <= MaxLevels; ++II) Constraints[II].setAny(SE); - for (int SI = Coupled.find_first(); SI >= 0; SI = Coupled.find_next(SI)) { + for (unsigned SI : Coupled.set_bits()) { SmallBitVector Group(Pair[SI].Group); SmallBitVector Sivs(Pairs); SmallBitVector Mivs(Pairs); SmallBitVector ConstrainedLevels(MaxLevels + 1); - for (int SJ = Group.find_first(); SJ >= 0; SJ = Group.find_next(SJ)) { + for (unsigned SJ : Group.set_bits()) { if (Pair[SJ].Classification == Subscript::SIV) Sivs.set(SJ); else @@ -3899,7 +3898,7 @@ const SCEV *DependenceInfo::getSplitIteration(const Dependence &Dep, } while (Sivs.any()) { bool Changed = false; - for (int SJ = Sivs.find_first(); SJ >= 0; SJ = Sivs.find_next(SJ)) { + for (unsigned SJ : Sivs.set_bits()) { // SJ is an SIV subscript that's part of the current coupled group unsigned Level; const SCEV *SplitIter = nullptr; @@ -3914,7 +3913,7 @@ const SCEV *DependenceInfo::getSplitIteration(const Dependence &Dep, } if (Changed) { // propagate, possibly creating new SIVs and ZIVs - for (int SJ = Mivs.find_first(); SJ >= 0; SJ = Mivs.find_next(SJ)) { + for (unsigned SJ : Mivs.set_bits()) { // SJ is an MIV subscript that's part of the current coupled group if (propagate(Pair[SJ].Src, Pair[SJ].Dst, Pair[SJ].Loops, Constraints, Result.Consistent)) { diff --git a/lib/Analysis/InlineCost.cpp b/lib/Analysis/InlineCost.cpp index 44c14cb17c22..4702569126c6 100644 --- a/lib/Analysis/InlineCost.cpp +++ b/lib/Analysis/InlineCost.cpp @@ -669,21 +669,33 @@ void CallAnalyzer::updateThreshold(CallSite CS, Function &Callee) { Threshold = MaxIfValid(Threshold, Params.HintThreshold); if (PSI) { BlockFrequencyInfo *CallerBFI = GetBFI ? &((*GetBFI)(*Caller)) : nullptr; - if (PSI->isHotCallSite(CS, CallerBFI)) { - DEBUG(dbgs() << "Hot callsite.\n"); - Threshold = Params.HotCallSiteThreshold.getValue(); - } else if (PSI->isFunctionEntryHot(&Callee)) { - DEBUG(dbgs() << "Hot callee.\n"); - // If callsite hotness can not be determined, we may still know - // that the callee is hot and treat it as a weaker hint for threshold - // increase. - Threshold = MaxIfValid(Threshold, Params.HintThreshold); - } else if (PSI->isColdCallSite(CS, CallerBFI)) { - DEBUG(dbgs() << "Cold callsite.\n"); - Threshold = MinIfValid(Threshold, Params.ColdCallSiteThreshold); - } else if (PSI->isFunctionEntryCold(&Callee)) { - DEBUG(dbgs() << "Cold callee.\n"); - Threshold = MinIfValid(Threshold, Params.ColdThreshold); + // FIXME: After switching to the new passmanager, simplify the logic below + // by checking only the callsite hotness/coldness. The check for CallerBFI + // exists only because we do not have BFI available with the old PM. + // + // Use callee's hotness information only if we have no way of determining + // callsite's hotness information. Callsite hotness can be determined if + // sample profile is used (which adds hotness metadata to calls) or if + // caller's BlockFrequencyInfo is available. + if (CallerBFI || PSI->hasSampleProfile()) { + if (PSI->isHotCallSite(CS, CallerBFI)) { + DEBUG(dbgs() << "Hot callsite.\n"); + Threshold = Params.HotCallSiteThreshold.getValue(); + } else if (PSI->isColdCallSite(CS, CallerBFI)) { + DEBUG(dbgs() << "Cold callsite.\n"); + Threshold = MinIfValid(Threshold, Params.ColdCallSiteThreshold); + } + } else { + if (PSI->isFunctionEntryHot(&Callee)) { + DEBUG(dbgs() << "Hot callee.\n"); + // If callsite hotness can not be determined, we may still know + // that the callee is hot and treat it as a weaker hint for threshold + // increase. + Threshold = MaxIfValid(Threshold, Params.HintThreshold); + } else if (PSI->isFunctionEntryCold(&Callee)) { + DEBUG(dbgs() << "Cold callee.\n"); + Threshold = MinIfValid(Threshold, Params.ColdThreshold); + } } } } diff --git a/lib/Analysis/InstructionSimplify.cpp b/lib/Analysis/InstructionSimplify.cpp index 5728887cc1e9..5652248a60ce 100644 --- a/lib/Analysis/InstructionSimplify.cpp +++ b/lib/Analysis/InstructionSimplify.cpp @@ -1752,6 +1752,24 @@ static Value *SimplifyAndInst(Value *Op0, Value *Op1, const SimplifyQuery &Q, (A == Op0 || B == Op0)) return Op0; + // A mask that only clears known zeros of a shifted value is a no-op. + Value *X; + const APInt *Mask; + const APInt *ShAmt; + if (match(Op1, m_APInt(Mask))) { + // If all bits in the inverted and shifted mask are clear: + // and (shl X, ShAmt), Mask --> shl X, ShAmt + if (match(Op0, m_Shl(m_Value(X), m_APInt(ShAmt))) && + (~(*Mask)).lshr(*ShAmt).isNullValue()) + return Op0; + + // If all bits in the inverted and shifted mask are clear: + // and (lshr X, ShAmt), Mask --> lshr X, ShAmt + if (match(Op0, m_LShr(m_Value(X), m_APInt(ShAmt))) && + (~(*Mask)).shl(*ShAmt).isNullValue()) + return Op0; + } + // A & (-A) = A if A is a power of two or zero. if (match(Op0, m_Neg(m_Specific(Op1))) || match(Op1, m_Neg(m_Specific(Op0)))) { diff --git a/lib/Analysis/ProfileSummaryInfo.cpp b/lib/Analysis/ProfileSummaryInfo.cpp index 502f4205b689..12b86daa602b 100644 --- a/lib/Analysis/ProfileSummaryInfo.cpp +++ b/lib/Analysis/ProfileSummaryInfo.cpp @@ -75,7 +75,7 @@ ProfileSummaryInfo::getProfileCount(const Instruction *Inst, return None; assert((isa(Inst) || isa(Inst)) && "We can only get profile count for call/invoke instruction."); - if (computeSummary() && Summary->getKind() == ProfileSummary::PSK_Sample) { + if (hasSampleProfile()) { // In sample PGO mode, check if there is a profile metadata on the // instruction. If it is present, determine hotness solely based on that, // since the sampled entry count may not be accurate. diff --git a/lib/Analysis/ScalarEvolution.cpp b/lib/Analysis/ScalarEvolution.cpp index 800354d2f5b4..a746ddfd7a63 100644 --- a/lib/Analysis/ScalarEvolution.cpp +++ b/lib/Analysis/ScalarEvolution.cpp @@ -629,19 +629,19 @@ static int CompareSCEVComplexity( const SCEVAddRecExpr *LA = cast(LHS); const SCEVAddRecExpr *RA = cast(RHS); - // If there is a dominance relationship between the loops, sort by the - // dominance. Otherwise, sort by depth. We require such order in getAddExpr. + // There is always a dominance between two recs that are used by one SCEV, + // so we can safely sort recs by loop header dominance. We require such + // order in getAddExpr. const Loop *LLoop = LA->getLoop(), *RLoop = RA->getLoop(); if (LLoop != RLoop) { const BasicBlock *LHead = LLoop->getHeader(), *RHead = RLoop->getHeader(); assert(LHead != RHead && "Two loops share the same header?"); if (DT.dominates(LHead, RHead)) return 1; - else if (DT.dominates(RHead, LHead)) - return -1; - unsigned LDepth = LLoop->getLoopDepth(), RDepth = RLoop->getLoopDepth(); - if (LDepth != RDepth) - return (int)LDepth - (int)RDepth; + else + assert(DT.dominates(RHead, LHead) && + "No dominance between recurrences used by one SCEV?"); + return -1; } // Addrec complexity grows with operand count. @@ -2512,22 +2512,23 @@ const SCEV *ScalarEvolution::getAddExpr(SmallVectorImpl &Ops, SmallVector AddRecOps(AddRec->op_begin(), AddRec->op_end()); for (; OtherIdx != Ops.size() && isa(Ops[OtherIdx]); - ++OtherIdx) - if (const auto *OtherAddRec = dyn_cast(Ops[OtherIdx])) - if (OtherAddRec->getLoop() == AddRecLoop) { - for (unsigned i = 0, e = OtherAddRec->getNumOperands(); - i != e; ++i) { - if (i >= AddRecOps.size()) { - AddRecOps.append(OtherAddRec->op_begin()+i, - OtherAddRec->op_end()); - break; - } - SmallVector TwoOps = { - AddRecOps[i], OtherAddRec->getOperand(i)}; - AddRecOps[i] = getAddExpr(TwoOps, SCEV::FlagAnyWrap, Depth + 1); + ++OtherIdx) { + const auto *OtherAddRec = cast(Ops[OtherIdx]); + if (OtherAddRec->getLoop() == AddRecLoop) { + for (unsigned i = 0, e = OtherAddRec->getNumOperands(); + i != e; ++i) { + if (i >= AddRecOps.size()) { + AddRecOps.append(OtherAddRec->op_begin()+i, + OtherAddRec->op_end()); + break; } - Ops.erase(Ops.begin() + OtherIdx); --OtherIdx; + SmallVector TwoOps = { + AddRecOps[i], OtherAddRec->getOperand(i)}; + AddRecOps[i] = getAddExpr(TwoOps, SCEV::FlagAnyWrap, Depth + 1); } + Ops.erase(Ops.begin() + OtherIdx); --OtherIdx; + } + } // Step size has changed, so we cannot guarantee no self-wraparound. Ops[Idx] = getAddRecExpr(AddRecOps, AddRecLoop, SCEV::FlagAnyWrap); return getAddExpr(Ops, SCEV::FlagAnyWrap, Depth + 1); diff --git a/lib/CodeGen/AggressiveAntiDepBreaker.cpp b/lib/CodeGen/AggressiveAntiDepBreaker.cpp index 3a57772cc7f5..43b245c66400 100644 --- a/lib/CodeGen/AggressiveAntiDepBreaker.cpp +++ b/lib/CodeGen/AggressiveAntiDepBreaker.cpp @@ -128,8 +128,7 @@ AggressiveAntiDepBreaker::AggressiveAntiDepBreaker( } DEBUG(dbgs() << "AntiDep Critical-Path Registers:"); - DEBUG(for (int r = CriticalPathSet.find_first(); r != -1; - r = CriticalPathSet.find_next(r)) + DEBUG(for (unsigned r : CriticalPathSet.set_bits()) dbgs() << " " << TRI->getName(r)); DEBUG(dbgs() << '\n'); } @@ -571,7 +570,7 @@ bool AggressiveAntiDepBreaker::FindSuitableFreeRegisters( DEBUG({ dbgs() << " ::"; - for (int r = BV.find_first(); r != -1; r = BV.find_next(r)) + for (unsigned r : BV.set_bits()) dbgs() << " " << TRI->getName(r); dbgs() << "\n"; }); diff --git a/lib/CodeGen/AsmPrinter/CodeViewDebug.cpp b/lib/CodeGen/AsmPrinter/CodeViewDebug.cpp index 98163bffb60b..7d945690e9c3 100644 --- a/lib/CodeGen/AsmPrinter/CodeViewDebug.cpp +++ b/lib/CodeGen/AsmPrinter/CodeViewDebug.cpp @@ -501,7 +501,7 @@ void CodeViewDebug::emitTypeInformation() { Error E = Reader.readArray(Types, Reader.getLength()); if (!E) { TypeVisitorCallbacks C; - E = CVTypeVisitor(C).visitTypeStream(Types); + E = codeview::visitTypeStream(Types, C); } if (E) { logAllUnhandledErrors(std::move(E), errs(), "error: "); diff --git a/lib/CodeGen/AsmPrinter/DbgValueHistoryCalculator.cpp b/lib/CodeGen/AsmPrinter/DbgValueHistoryCalculator.cpp index 22fd7bb46056..20e1467b30c3 100644 --- a/lib/CodeGen/AsmPrinter/DbgValueHistoryCalculator.cpp +++ b/lib/CodeGen/AsmPrinter/DbgValueHistoryCalculator.cpp @@ -209,8 +209,7 @@ void llvm::calculateDbgValueHistory(const MachineFunction *MF, } else if (MO.isRegMask()) { // If this is a register mask operand, clobber all debug values in // non-CSRs. - for (int I = ChangingRegs.find_first(); I != -1; - I = ChangingRegs.find_next(I)) { + for (unsigned I : ChangingRegs.set_bits()) { // Don't consider SP to be clobbered by register masks. if (unsigned(I) != SP && TRI->isPhysicalRegister(I) && MO.clobbersPhysReg(I)) { diff --git a/lib/CodeGen/GlobalISel/IRTranslator.cpp b/lib/CodeGen/GlobalISel/IRTranslator.cpp index 811858f136eb..77dfb13ac1f2 100644 --- a/lib/CodeGen/GlobalISel/IRTranslator.cpp +++ b/lib/CodeGen/GlobalISel/IRTranslator.cpp @@ -1129,6 +1129,11 @@ void IRTranslator::finalizeFunction() { ValToVReg.clear(); FrameIndices.clear(); MachinePreds.clear(); + // MachineIRBuilder::DebugLoc can outlive the DILocation it holds. Clear it + // to avoid accessing free’d memory (in runOnMachineFunction) and to avoid + // destroying it twice (in ~IRTranslator() and ~LLVMContext()) + EntryBuilder = MachineIRBuilder(); + CurBuilder = MachineIRBuilder(); } bool IRTranslator::runOnMachineFunction(MachineFunction &CurMF) { diff --git a/lib/CodeGen/MachineVerifier.cpp b/lib/CodeGen/MachineVerifier.cpp index ab433273b189..b53b002f55a6 100644 --- a/lib/CodeGen/MachineVerifier.cpp +++ b/lib/CodeGen/MachineVerifier.cpp @@ -760,7 +760,7 @@ MachineVerifier::visitMachineBasicBlockBefore(const MachineBasicBlock *MBB) { const MachineFrameInfo &MFI = MF->getFrameInfo(); BitVector PR = MFI.getPristineRegs(*MF); - for (int I = PR.find_first(); I>0; I = PR.find_next(I)) { + for (unsigned I : PR.set_bits()) { for (MCSubRegIterator SubRegs(I, TRI, /*IncludeSelf=*/true); SubRegs.isValid(); ++SubRegs) regsLive.insert(*SubRegs); diff --git a/lib/CodeGen/RegAllocGreedy.cpp b/lib/CodeGen/RegAllocGreedy.cpp index 06500289c971..47d726f6da7a 100644 --- a/lib/CodeGen/RegAllocGreedy.cpp +++ b/lib/CodeGen/RegAllocGreedy.cpp @@ -285,8 +285,7 @@ class RAGreedy : public MachineFunctionPass, // Set B[i] = C for every live bundle where B[i] was NoCand. unsigned getBundles(SmallVectorImpl &B, unsigned C) { unsigned Count = 0; - for (int i = LiveBundles.find_first(); i >= 0; - i = LiveBundles.find_next(i)) + for (unsigned i : LiveBundles.set_bits()) if (B[i] == NoCand) { B[i] = C; Count++; @@ -1162,9 +1161,8 @@ bool RAGreedy::calcCompactRegion(GlobalSplitCandidate &Cand) { } DEBUG({ - for (int i = Cand.LiveBundles.find_first(); i>=0; - i = Cand.LiveBundles.find_next(i)) - dbgs() << " EB#" << i; + for (int i : Cand.LiveBundles.set_bits()) + dbgs() << " EB#" << i; dbgs() << ".\n"; }); return true; @@ -1482,8 +1480,7 @@ unsigned RAGreedy::calculateRegionSplitCost(LiveInterval &VirtReg, DEBUG({ dbgs() << ", total = "; MBFI->printBlockFreq(dbgs(), Cost) << " with bundles"; - for (int i = Cand.LiveBundles.find_first(); i>=0; - i = Cand.LiveBundles.find_next(i)) + for (int i : Cand.LiveBundles.set_bits()) dbgs() << " EB#" << i; dbgs() << ".\n"; }); diff --git a/lib/CodeGen/SelectionDAG/DAGCombiner.cpp b/lib/CodeGen/SelectionDAG/DAGCombiner.cpp index caf5cb497a71..0ccee175abfb 100644 --- a/lib/CodeGen/SelectionDAG/DAGCombiner.cpp +++ b/lib/CodeGen/SelectionDAG/DAGCombiner.cpp @@ -13087,14 +13087,28 @@ SDValue DAGCombiner::visitSTORE(SDNode *N) { } } - // If this is a store followed by a store with the same value to the same - // location, then the store is dead/noop. if (StoreSDNode *ST1 = dyn_cast(Chain)) { - if (ST1->getBasePtr() == Ptr && ST->getMemoryVT() == ST1->getMemoryVT() && - ST1->getValue() == Value && ST->isUnindexed() && !ST->isVolatile() && - ST1->isUnindexed() && !ST1->isVolatile()) { - // The store is dead, remove it. - return Chain; + if (ST->isUnindexed() && !ST->isVolatile() && ST1->isUnindexed() && + !ST1->isVolatile() && ST1->getBasePtr() == Ptr && + ST->getMemoryVT() == ST1->getMemoryVT()) { + // If this is a store followed by a store with the same value to the same + // location, then the store is dead/noop. + if (ST1->getValue() == Value) { + // The store is dead, remove it. + return Chain; + } + + // If this is a store who's preceeding store to the same location + // and no one other node is chained to that store we can effectively + // drop the store. Do not remove stores to undef as they may be used as + // data sinks. + if (OptLevel != CodeGenOpt::None && ST1->hasOneUse() && + !ST1->getBasePtr().isUndef()) { + // ST1 is fully overwritten and can be elided. Combine with it's chain + // value. + CombineTo(ST1, ST1->getChain()); + return SDValue(); + } } } diff --git a/lib/CodeGen/SpillPlacement.cpp b/lib/CodeGen/SpillPlacement.cpp index f10c98ef4e50..43cbf4add0f8 100644 --- a/lib/CodeGen/SpillPlacement.cpp +++ b/lib/CodeGen/SpillPlacement.cpp @@ -310,7 +310,7 @@ void SpillPlacement::addLinks(ArrayRef Links) { bool SpillPlacement::scanActiveBundles() { RecentPositive.clear(); - for (int n = ActiveNodes->find_first(); n>=0; n = ActiveNodes->find_next(n)) { + for (unsigned n : ActiveNodes->set_bits()) { update(n); // A node that must spill, or a node without any links is not going to // change its value ever again, so exclude it from iterations. @@ -365,7 +365,7 @@ SpillPlacement::finish() { // Write preferences back to ActiveNodes. bool Perfect = true; - for (int n = ActiveNodes->find_first(); n>=0; n = ActiveNodes->find_next(n)) + for (unsigned n : ActiveNodes->set_bits()) if (!nodes[n].preferReg()) { ActiveNodes->reset(n); Perfect = false; diff --git a/lib/CodeGen/StackColoring.cpp b/lib/CodeGen/StackColoring.cpp index f51d959a089a..86a16187fcb6 100644 --- a/lib/CodeGen/StackColoring.cpp +++ b/lib/CodeGen/StackColoring.cpp @@ -703,12 +703,10 @@ void StackColoring::calculateLiveIntervals(unsigned NumSlots) { // Create the interval of the blocks that we previously found to be 'alive'. BlockLifetimeInfo &MBBLiveness = BlockLiveness[&MBB]; - for (int pos = MBBLiveness.LiveIn.find_first(); pos != -1; - pos = MBBLiveness.LiveIn.find_next(pos)) { + for (unsigned pos : MBBLiveness.LiveIn.set_bits()) { Starts[pos] = Indexes->getMBBStartIdx(&MBB); } - for (int pos = MBBLiveness.LiveOut.find_first(); pos != -1; - pos = MBBLiveness.LiveOut.find_next(pos)) { + for (unsigned pos : MBBLiveness.LiveOut.set_bits()) { Finishes[pos] = Indexes->getMBBEndIdx(&MBB); } diff --git a/lib/CodeGen/TargetLoweringBase.cpp b/lib/CodeGen/TargetLoweringBase.cpp index 39aa946fa840..5f63fd4320bb 100644 --- a/lib/CodeGen/TargetLoweringBase.cpp +++ b/lib/CodeGen/TargetLoweringBase.cpp @@ -1312,7 +1312,7 @@ TargetLoweringBase::findRepresentativeClass(const TargetRegisterInfo *TRI, // Find the first legal register class with the largest spill size. const TargetRegisterClass *BestRC = RC; - for (int i = SuperRegRC.find_first(); i >= 0; i = SuperRegRC.find_next(i)) { + for (unsigned i : SuperRegRC.set_bits()) { const TargetRegisterClass *SuperRC = TRI->getRegClass(i); // We want the largest possible spill size. if (TRI->getSpillSize(*SuperRC) <= TRI->getSpillSize(*BestRC)) diff --git a/lib/CodeGen/TargetPassConfig.cpp b/lib/CodeGen/TargetPassConfig.cpp index e6c5d8753b83..9724cb074584 100644 --- a/lib/CodeGen/TargetPassConfig.cpp +++ b/lib/CodeGen/TargetPassConfig.cpp @@ -564,6 +564,14 @@ void TargetPassConfig::addISelPrepare() { addPass(createVerifierPass()); } +/// -regalloc=... command line option. +static FunctionPass *useDefaultRegisterAllocator() { return nullptr; } +static cl::opt > +RegAlloc("regalloc", + cl::init(&useDefaultRegisterAllocator), + cl::desc("Register allocator to use")); + /// Add the complete set of target-independent postISel code generator passes. /// /// This can be read as the standard order of major LLVM CodeGen stages. Stages @@ -625,8 +633,12 @@ void TargetPassConfig::addMachinePasses() { // including phi elimination and scheduling. if (getOptimizeRegAlloc()) addOptimizedRegAlloc(createRegAllocPass(true)); - else + else { + if (RegAlloc != &useDefaultRegisterAllocator && + RegAlloc != &createFastRegisterAllocator) + report_fatal_error("Must use fast (default) register allocator for unoptimized regalloc."); addFastRegAlloc(createRegAllocPass(false)); + } // Run post-ra passes. addPostRegAlloc(); @@ -759,19 +771,12 @@ MachinePassRegistry RegisterRegAlloc::Registry; /// A dummy default pass factory indicates whether the register allocator is /// overridden on the command line. static llvm::once_flag InitializeDefaultRegisterAllocatorFlag; -static FunctionPass *useDefaultRegisterAllocator() { return nullptr; } + static RegisterRegAlloc defaultRegAlloc("default", "pick register allocator based on -O option", useDefaultRegisterAllocator); -/// -regalloc=... command line option. -static cl::opt > -RegAlloc("regalloc", - cl::init(&useDefaultRegisterAllocator), - cl::desc("Register allocator to use")); - static void initializeDefaultRegisterAllocatorOnce() { RegisterRegAlloc::FunctionPassCtor Ctor = RegisterRegAlloc::getDefault(); @@ -781,7 +786,6 @@ static void initializeDefaultRegisterAllocatorOnce() { } } - /// Instantiate the default register allocator pass for this target for either /// the optimized or unoptimized allocation path. This will be added to the pass /// manager by addFastRegAlloc in the unoptimized case or addOptimizedRegAlloc diff --git a/lib/CodeGen/TargetRegisterInfo.cpp b/lib/CodeGen/TargetRegisterInfo.cpp index f6e4c17d514c..41ec082a24cf 100644 --- a/lib/CodeGen/TargetRegisterInfo.cpp +++ b/lib/CodeGen/TargetRegisterInfo.cpp @@ -50,8 +50,7 @@ bool TargetRegisterInfo::checkAllSuperRegsMarked(const BitVector &RegisterSet, ArrayRef Exceptions) const { // Check that all super registers of reserved regs are reserved as well. BitVector Checked(getNumRegs()); - for (int Reg = RegisterSet.find_first(); Reg>=0; - Reg = RegisterSet.find_next(Reg)) { + for (unsigned Reg : RegisterSet.set_bits()) { if (Checked[Reg]) continue; for (MCSuperRegIterator SR(Reg, this); SR.isValid(); ++SR) { diff --git a/lib/DebugInfo/CodeView/CVTypeDumper.cpp b/lib/DebugInfo/CodeView/CVTypeDumper.cpp index bcc8218d9446..02e1682f76e7 100644 --- a/lib/DebugInfo/CodeView/CVTypeDumper.cpp +++ b/lib/DebugInfo/CodeView/CVTypeDumper.cpp @@ -11,7 +11,6 @@ #include "llvm/DebugInfo/CodeView/CVTypeVisitor.h" #include "llvm/DebugInfo/CodeView/TypeDatabase.h" #include "llvm/DebugInfo/CodeView/TypeDatabaseVisitor.h" -#include "llvm/DebugInfo/CodeView/TypeDeserializer.h" #include "llvm/DebugInfo/CodeView/TypeRecord.h" #include "llvm/DebugInfo/CodeView/TypeVisitorCallbackPipeline.h" #include "llvm/Support/BinaryByteStream.h" @@ -21,38 +20,23 @@ using namespace llvm::codeview; Error CVTypeDumper::dump(const CVType &Record, TypeVisitorCallbacks &Dumper) { TypeDatabaseVisitor DBV(TypeDB); - TypeDeserializer Deserializer; TypeVisitorCallbackPipeline Pipeline; - Pipeline.addCallbackToPipeline(Deserializer); Pipeline.addCallbackToPipeline(DBV); Pipeline.addCallbackToPipeline(Dumper); - CVTypeVisitor Visitor(Pipeline); - if (Handler) - Visitor.addTypeServerHandler(*Handler); - CVType RecordCopy = Record; - if (auto EC = Visitor.visitTypeRecord(RecordCopy)) - return EC; - return Error::success(); + return codeview::visitTypeRecord(RecordCopy, Pipeline, VDS_BytesPresent, + Handler); } Error CVTypeDumper::dump(const CVTypeArray &Types, TypeVisitorCallbacks &Dumper) { TypeDatabaseVisitor DBV(TypeDB); - TypeDeserializer Deserializer; TypeVisitorCallbackPipeline Pipeline; - Pipeline.addCallbackToPipeline(Deserializer); Pipeline.addCallbackToPipeline(DBV); Pipeline.addCallbackToPipeline(Dumper); - CVTypeVisitor Visitor(Pipeline); - if (Handler) - Visitor.addTypeServerHandler(*Handler); - - if (auto EC = Visitor.visitTypeStream(Types)) - return EC; - return Error::success(); + return codeview::visitTypeStream(Types, Pipeline, Handler); } Error CVTypeDumper::dump(ArrayRef Data, TypeVisitorCallbacks &Dumper) { diff --git a/lib/DebugInfo/CodeView/CVTypeVisitor.cpp b/lib/DebugInfo/CodeView/CVTypeVisitor.cpp index b6ed0453d9c4..0f7f5f667790 100644 --- a/lib/DebugInfo/CodeView/CVTypeVisitor.cpp +++ b/lib/DebugInfo/CodeView/CVTypeVisitor.cpp @@ -59,13 +59,8 @@ static Expected deserializeTypeServerRecord(CVType &Record) { }; TypeServer2Record R(TypeRecordKind::TypeServer2); - TypeDeserializer Deserializer; StealTypeServerVisitor Thief(R); - TypeVisitorCallbackPipeline Pipeline; - Pipeline.addCallbackToPipeline(Deserializer); - Pipeline.addCallbackToPipeline(Thief); - CVTypeVisitor Visitor(Pipeline); - if (auto EC = Visitor.visitTypeRecord(Record)) + if (auto EC = visitTypeRecord(Record, Thief)) return std::move(EC); return R; @@ -178,7 +173,7 @@ static Error visitMemberRecord(CVMemberRecord &Record, return Error::success(); } -Error CVTypeVisitor::visitMemberRecord(CVMemberRecord &Record) { +Error CVTypeVisitor::visitMemberRecord(CVMemberRecord Record) { return ::visitMemberRecord(Record, Callbacks); } @@ -224,3 +219,93 @@ Error CVTypeVisitor::visitFieldListMemberStream(ArrayRef Data) { BinaryStreamReader SR(S); return visitFieldListMemberStream(SR); } + +namespace { +struct FieldListVisitHelper { + FieldListVisitHelper(TypeVisitorCallbacks &Callbacks, ArrayRef Data, + VisitorDataSource Source) + : Stream(Data, llvm::support::little), Reader(Stream), + Deserializer(Reader), + Visitor((Source == VDS_BytesPresent) ? Pipeline : Callbacks) { + if (Source == VDS_BytesPresent) { + Pipeline.addCallbackToPipeline(Deserializer); + Pipeline.addCallbackToPipeline(Callbacks); + } + } + + BinaryByteStream Stream; + BinaryStreamReader Reader; + FieldListDeserializer Deserializer; + TypeVisitorCallbackPipeline Pipeline; + CVTypeVisitor Visitor; +}; + +struct VisitHelper { + VisitHelper(TypeVisitorCallbacks &Callbacks, VisitorDataSource Source, + TypeServerHandler *TS) + : Visitor((Source == VDS_BytesPresent) ? Pipeline : Callbacks) { + if (TS) + Visitor.addTypeServerHandler(*TS); + if (Source == VDS_BytesPresent) { + Pipeline.addCallbackToPipeline(Deserializer); + Pipeline.addCallbackToPipeline(Callbacks); + } + } + + TypeDeserializer Deserializer; + TypeVisitorCallbackPipeline Pipeline; + CVTypeVisitor Visitor; +}; +} + +Error llvm::codeview::visitTypeRecord(CVType &Record, TypeIndex Index, + TypeVisitorCallbacks &Callbacks, + VisitorDataSource Source, + TypeServerHandler *TS) { + VisitHelper Helper(Callbacks, Source, TS); + return Helper.Visitor.visitTypeRecord(Record, Index); +} + +Error llvm::codeview::visitTypeRecord(CVType &Record, + TypeVisitorCallbacks &Callbacks, + VisitorDataSource Source, + TypeServerHandler *TS) { + VisitHelper Helper(Callbacks, Source, TS); + return Helper.Visitor.visitTypeRecord(Record); +} + +Error llvm::codeview::visitMemberRecordStream(ArrayRef FieldList, + TypeVisitorCallbacks &Callbacks) { + CVTypeVisitor Visitor(Callbacks); + return Visitor.visitFieldListMemberStream(FieldList); +} + +Error llvm::codeview::visitMemberRecord(CVMemberRecord Record, + TypeVisitorCallbacks &Callbacks, + VisitorDataSource Source) { + FieldListVisitHelper Helper(Callbacks, Record.Data, Source); + return Helper.Visitor.visitMemberRecord(Record); +} + +Error llvm::codeview::visitMemberRecord(TypeLeafKind Kind, + ArrayRef Record, + TypeVisitorCallbacks &Callbacks) { + CVMemberRecord R; + R.Data = Record; + R.Kind = Kind; + return visitMemberRecord(R, Callbacks, VDS_BytesPresent); +} + +Error llvm::codeview::visitTypeStream(const CVTypeArray &Types, + TypeVisitorCallbacks &Callbacks, + TypeServerHandler *TS) { + VisitHelper Helper(Callbacks, VDS_BytesPresent, TS); + return Helper.Visitor.visitTypeStream(Types); +} + +Error llvm::codeview::visitTypeStream(CVTypeRange Types, + TypeVisitorCallbacks &Callbacks, + TypeServerHandler *TS) { + VisitHelper Helper(Callbacks, VDS_BytesPresent, TS); + return Helper.Visitor.visitTypeStream(Types); +} diff --git a/lib/DebugInfo/CodeView/RandomAccessTypeVisitor.cpp b/lib/DebugInfo/CodeView/RandomAccessTypeVisitor.cpp index 4cb9acbe07d9..704d1131108a 100644 --- a/lib/DebugInfo/CodeView/RandomAccessTypeVisitor.cpp +++ b/lib/DebugInfo/CodeView/RandomAccessTypeVisitor.cpp @@ -9,6 +9,7 @@ #include "llvm/DebugInfo/CodeView/RandomAccessTypeVisitor.h" +#include "llvm/DebugInfo/CodeView/CVTypeVisitor.h" #include "llvm/DebugInfo/CodeView/TypeDatabase.h" #include "llvm/DebugInfo/CodeView/TypeServerHandler.h" #include "llvm/DebugInfo/CodeView/TypeVisitorCallbacks.h" @@ -20,9 +21,7 @@ RandomAccessTypeVisitor::RandomAccessTypeVisitor( const CVTypeArray &Types, uint32_t NumRecords, PartialOffsetArray PartialOffsets) : Database(NumRecords), Types(Types), DatabaseVisitor(Database), - InternalVisitor(Pipeline), PartialOffsets(PartialOffsets) { - Pipeline.addCallbackToPipeline(Deserializer); - Pipeline.addCallbackToPipeline(DatabaseVisitor); + PartialOffsets(PartialOffsets) { KnownOffsets.resize(Database.capacity()); } @@ -38,8 +37,7 @@ Error RandomAccessTypeVisitor::visitTypeIndex(TypeIndex TI, assert(Database.contains(TI)); auto &Record = Database.getTypeRecord(TI); - CVTypeVisitor V(Callbacks); - return V.visitTypeRecord(Record, TI); + return codeview::visitTypeRecord(Record, TI, Callbacks); } Error RandomAccessTypeVisitor::visitRangeForType(TypeIndex TI) { @@ -78,7 +76,7 @@ Error RandomAccessTypeVisitor::visitRange(TypeIndex Begin, uint32_t BeginOffset, while (Begin != End) { assert(!Database.contains(Begin)); - if (auto EC = InternalVisitor.visitTypeRecord(*RI, Begin)) + if (auto EC = codeview::visitTypeRecord(*RI, Begin, DatabaseVisitor)) return EC; KnownOffsets[Begin.toArrayIndex()] = BeginOffset; diff --git a/lib/DebugInfo/CodeView/TypeDumpVisitor.cpp b/lib/DebugInfo/CodeView/TypeDumpVisitor.cpp index 27a6e0987886..9485c9cfedff 100644 --- a/lib/DebugInfo/CodeView/TypeDumpVisitor.cpp +++ b/lib/DebugInfo/CodeView/TypeDumpVisitor.cpp @@ -216,8 +216,7 @@ Error TypeDumpVisitor::visitMemberEnd(CVMemberRecord &Record) { Error TypeDumpVisitor::visitKnownRecord(CVType &CVR, FieldListRecord &FieldList) { - CVTypeVisitor Visitor(*this); - if (auto EC = Visitor.visitFieldListMemberStream(FieldList.Data)) + if (auto EC = codeview::visitMemberRecordStream(FieldList.Data, *this)) return EC; return Error::success(); diff --git a/lib/DebugInfo/CodeView/TypeStreamMerger.cpp b/lib/DebugInfo/CodeView/TypeStreamMerger.cpp index aad20ae6dda1..51f24fa3f135 100644 --- a/lib/DebugInfo/CodeView/TypeStreamMerger.cpp +++ b/lib/DebugInfo/CodeView/TypeStreamMerger.cpp @@ -361,8 +361,7 @@ Error TypeStreamMerger::visitKnownRecord(CVType &, FieldListRecord &R) { // Visit the members inside the field list. HadUntranslatedMember = false; FieldListBuilder.begin(); - CVTypeVisitor Visitor(*this); - if (auto EC = Visitor.visitFieldListMemberStream(R.Data)) + if (auto EC = codeview::visitMemberRecordStream(R.Data, *this)) return EC; // Write the record if we translated all field list members. @@ -440,18 +439,9 @@ Error TypeStreamMerger::visitUnknownType(CVType &Rec) { Error TypeStreamMerger::mergeStream(const CVTypeArray &Types) { assert(IndexMap.empty()); - TypeVisitorCallbackPipeline Pipeline; LastError = Error::success(); - TypeDeserializer Deserializer; - Pipeline.addCallbackToPipeline(Deserializer); - Pipeline.addCallbackToPipeline(*this); - - CVTypeVisitor Visitor(Pipeline); - if (Handler) - Visitor.addTypeServerHandler(*Handler); - - if (auto EC = Visitor.visitTypeStream(Types)) + if (auto EC = codeview::visitTypeStream(Types, *this, Handler)) return EC; // If we found bad indices but no other errors, try doing another pass and see @@ -466,7 +456,8 @@ Error TypeStreamMerger::mergeStream(const CVTypeArray &Types) { IsSecondPass = true; NumBadIndices = 0; CurIndex = TypeIndex(TypeIndex::FirstNonSimpleIndex); - if (auto EC = Visitor.visitTypeStream(Types)) + + if (auto EC = codeview::visitTypeStream(Types, *this, Handler)) return EC; assert(NumBadIndices <= BadIndicesRemaining && diff --git a/lib/DebugInfo/DWARF/DWARFContext.cpp b/lib/DebugInfo/DWARF/DWARFContext.cpp index 59a060d143ff..61e75a2b56ab 100644 --- a/lib/DebugInfo/DWARF/DWARFContext.cpp +++ b/lib/DebugInfo/DWARF/DWARFContext.cpp @@ -1086,49 +1086,32 @@ DWARFContextInMemory::DWARFContextInMemory(const object::ObjectFile &Obj, continue; } + if (Section.relocation_begin() == Section.relocation_end()) + continue; + std::map AddrCache; - if (Section.relocation_begin() != Section.relocation_end()) { - uint64_t SectionSize = RelocatedSection->getSize(); - for (const RelocationRef &Reloc : Section.relocations()) { - // FIXME: it's not clear how to correctly handle scattered - // relocations. - if (isRelocScattered(Obj, Reloc)) - continue; + for (const RelocationRef &Reloc : Section.relocations()) { + // FIXME: it's not clear how to correctly handle scattered + // relocations. + if (isRelocScattered(Obj, Reloc)) + continue; - Expected SymAddrOrErr = - getSymbolAddress(Obj, Reloc, L, AddrCache); - if (!SymAddrOrErr) { - errs() << toString(SymAddrOrErr.takeError()) << '\n'; - continue; - } - - object::RelocVisitor V(Obj); - object::RelocToApply R(V.visit(Reloc.getType(), Reloc, *SymAddrOrErr)); - if (V.error()) { - SmallString<32> Name; - Reloc.getTypeName(Name); - errs() << "error: failed to compute relocation: " - << Name << "\n"; - continue; - } - uint64_t Address = Reloc.getOffset(); - if (Address + R.Width > SectionSize) { - errs() << "error: " << R.Width << "-byte relocation starting " - << Address << " bytes into section " << name << " which is " - << SectionSize << " bytes long.\n"; - continue; - } - if (R.Width > 8) { - errs() << "error: can't handle a relocation of more than 8 bytes at " - "a time.\n"; - continue; - } - DEBUG(dbgs() << "Writing " << format("%p", R.Value) - << " at " << format("%p", Address) - << " with width " << format("%d", R.Width) - << "\n"); - Map->insert({Address, {(uint8_t)R.Width, R.Value}}); + Expected SymAddrOrErr = + getSymbolAddress(Obj, Reloc, L, AddrCache); + if (!SymAddrOrErr) { + errs() << toString(SymAddrOrErr.takeError()) << '\n'; + continue; } + + object::RelocVisitor V(Obj); + object::RelocToApply R(V.visit(Reloc.getType(), Reloc, *SymAddrOrErr)); + if (V.error()) { + SmallString<32> Name; + Reloc.getTypeName(Name); + errs() << "error: failed to compute relocation: " << Name << "\n"; + continue; + } + Map->insert({Reloc.getOffset(), {R.Value}}); } } } diff --git a/lib/DebugInfo/PDB/Native/PDBTypeServerHandler.cpp b/lib/DebugInfo/PDB/Native/PDBTypeServerHandler.cpp index 629f3e80b0ed..cb783cf4fea7 100644 --- a/lib/DebugInfo/PDB/Native/PDBTypeServerHandler.cpp +++ b/lib/DebugInfo/PDB/Native/PDBTypeServerHandler.cpp @@ -55,9 +55,8 @@ PDBTypeServerHandler::handleInternal(PDBFile &File, auto ExpectedTpi = File.getPDBTpiStream(); if (!ExpectedTpi) return ExpectedTpi.takeError(); - CVTypeVisitor Visitor(Callbacks); - if (auto EC = Visitor.visitTypeStream(ExpectedTpi->types(nullptr))) + if (auto EC = codeview::visitTypeStream(ExpectedTpi->typeArray(), Callbacks)) return std::move(EC); return true; diff --git a/lib/ExecutionEngine/RuntimeDyld/RuntimeDyld.cpp b/lib/ExecutionEngine/RuntimeDyld/RuntimeDyld.cpp index e9a4b71c903d..ab86e5d6a0fd 100644 --- a/lib/ExecutionEngine/RuntimeDyld/RuntimeDyld.cpp +++ b/lib/ExecutionEngine/RuntimeDyld/RuntimeDyld.cpp @@ -705,7 +705,7 @@ RuntimeDyldImpl::emitSection(const ObjectFile &Obj, unsigned Alignment = (unsigned)Alignment64 & 0xffffffffL; unsigned PaddingSize = 0; unsigned StubBufSize = 0; - bool IsRequired = isRequiredForExecution(Section) || ProcessAllSections; + bool IsRequired = isRequiredForExecution(Section); bool IsVirtual = Section.isVirtual(); bool IsZeroInit = isZeroInit(Section); bool IsReadOnly = isReadOnlyData(Section); @@ -745,8 +745,8 @@ RuntimeDyldImpl::emitSection(const ObjectFile &Obj, Alignment = std::max(Alignment, getStubAlignment()); // Some sections, such as debug info, don't need to be loaded for execution. - // Leave those where they are. - if (IsRequired) { + // Process those only if explicitly requested. + if (IsRequired || ProcessAllSections) { Allocate = DataSize + PaddingSize + StubBufSize; if (!Allocate) Allocate = 1; @@ -790,6 +790,10 @@ RuntimeDyldImpl::emitSection(const ObjectFile &Obj, Sections.push_back( SectionEntry(Name, Addr, DataSize, Allocate, (uintptr_t)pData)); + // Debug info sections are linked as if their load address was zero + if (!IsRequired) + Sections.back().setLoadAddress(0); + if (Checker) Checker->registerSection(Obj.getFileName(), SectionID); diff --git a/lib/Support/CrashRecoveryContext.cpp b/lib/Support/CrashRecoveryContext.cpp index 98865f5e065e..bd38dd88201f 100644 --- a/lib/Support/CrashRecoveryContext.cpp +++ b/lib/Support/CrashRecoveryContext.cpp @@ -78,6 +78,9 @@ static bool gCrashRecoveryEnabled = false; static ManagedStatic> tlIsRecoveringFromCrash; +static void installExceptionOrSignalHandlers(); +static void uninstallExceptionOrSignalHandlers(); + CrashRecoveryContextCleanup::~CrashRecoveryContextCleanup() {} CrashRecoveryContext::~CrashRecoveryContext() { @@ -113,6 +116,23 @@ CrashRecoveryContext *CrashRecoveryContext::GetCurrent() { return CRCI->CRC; } +void CrashRecoveryContext::Enable() { + sys::ScopedLock L(*gCrashRecoveryContextMutex); + // FIXME: Shouldn't this be a refcount or something? + if (gCrashRecoveryEnabled) + return; + gCrashRecoveryEnabled = true; + installExceptionOrSignalHandlers(); +} + +void CrashRecoveryContext::Disable() { + sys::ScopedLock L(*gCrashRecoveryContextMutex); + if (!gCrashRecoveryEnabled) + return; + gCrashRecoveryEnabled = false; + uninstallExceptionOrSignalHandlers(); +} + void CrashRecoveryContext::registerCleanup(CrashRecoveryContextCleanup *cleanup) { if (!cleanup) @@ -140,30 +160,70 @@ CrashRecoveryContext::unregisterCleanup(CrashRecoveryContextCleanup *cleanup) { delete cleanup; } -#ifdef LLVM_ON_WIN32 +#if defined(_MSC_VER) +// If _MSC_VER is defined, we must have SEH. Use it if it's available. It's way +// better than VEH. Vectored exception handling catches all exceptions happening +// on the thread with installed exception handlers, so it can interfere with +// internal exception handling of other libraries on that thread. SEH works +// exactly as you would expect normal exception handling to work: it only +// catches exceptions if they would bubble out from the stack frame with __try / +// __except. -#include "Windows/WindowsSupport.h" +static void installExceptionOrSignalHandlers() {} +static void uninstallExceptionOrSignalHandlers() {} -// On Windows, we can make use of vectored exception handling to -// catch most crashing situations. Note that this does mean -// we will be alerted of exceptions *before* structured exception -// handling has the opportunity to catch it. But that isn't likely -// to cause problems because nowhere in the project is SEH being -// used. +bool CrashRecoveryContext::RunSafely(function_ref Fn) { + if (!gCrashRecoveryEnabled) { + Fn(); + return true; + } + + bool Result = true; + __try { + Fn(); + } __except (1) { // Catch any exception. + Result = false; + } + return Result; +} + +#else // !_MSC_VER + +#if defined(LLVM_ON_WIN32) +// This is a non-MSVC compiler, probably mingw gcc or clang without +// -fms-extensions. Use vectored exception handling (VEH). // -// Vectored exception handling is built on top of SEH, and so it -// works on a per-thread basis. +// On Windows, we can make use of vectored exception handling to catch most +// crashing situations. Note that this does mean we will be alerted of +// exceptions *before* structured exception handling has the opportunity to +// catch it. Unfortunately, this causes problems in practice with other code +// running on threads with LLVM crash recovery contexts, so we would like to +// eventually move away from VEH. +// +// Vectored works on a per-thread basis, which is an advantage over +// SetUnhandledExceptionFilter. SetUnhandledExceptionFilter also doesn't have +// any native support for chaining exception handlers, but VEH allows more than +// one. // // The vectored exception handler functionality was added in Windows // XP, so if support for older versions of Windows is required, // it will have to be added. -// -// If we want to support as far back as Win2k, we could use the -// SetUnhandledExceptionFilter API, but there's a risk of that -// being entirely overwritten (it's not a chain). + +#include "Windows/WindowsSupport.h" static LONG CALLBACK ExceptionHandler(PEXCEPTION_POINTERS ExceptionInfo) { + // DBG_PRINTEXCEPTION_WIDE_C is not properly defined on all supported + // compilers and platforms, so we define it manually. + constexpr ULONG DbgPrintExceptionWideC = 0x4001000AL; + switch (ExceptionInfo->ExceptionRecord->ExceptionCode) + { + case DBG_PRINTEXCEPTION_C: + case DbgPrintExceptionWideC: + case 0x406D1388: // set debugger thread name + return EXCEPTION_CONTINUE_EXECUTION; + } + // Lookup the current thread local recovery object. const CrashRecoveryContextImpl *CRCI = CurrentContext->get(); @@ -192,14 +252,7 @@ static LONG CALLBACK ExceptionHandler(PEXCEPTION_POINTERS ExceptionInfo) // non-NULL, valid VEH handles, or NULL. static sys::ThreadLocal sCurrentExceptionHandle; -void CrashRecoveryContext::Enable() { - sys::ScopedLock L(*gCrashRecoveryContextMutex); - - if (gCrashRecoveryEnabled) - return; - - gCrashRecoveryEnabled = true; - +static void installExceptionOrSignalHandlers() { // We can set up vectored exception handling now. We will install our // handler as the front of the list, though there's no assurances that // it will remain at the front (another call could install itself before @@ -208,14 +261,7 @@ void CrashRecoveryContext::Enable() { sCurrentExceptionHandle.set(handle); } -void CrashRecoveryContext::Disable() { - sys::ScopedLock L(*gCrashRecoveryContextMutex); - - if (!gCrashRecoveryEnabled) - return; - - gCrashRecoveryEnabled = false; - +static void uninstallExceptionOrSignalHandlers() { PVOID currentHandle = const_cast(sCurrentExceptionHandle.get()); if (currentHandle) { // Now we can remove the vectored exception handler from the chain @@ -226,7 +272,7 @@ void CrashRecoveryContext::Disable() { } } -#else +#else // !LLVM_ON_WIN32 // Generic POSIX implementation. // @@ -278,14 +324,7 @@ static void CrashRecoverySignalHandler(int Signal) { const_cast(CRCI)->HandleCrash(); } -void CrashRecoveryContext::Enable() { - sys::ScopedLock L(*gCrashRecoveryContextMutex); - - if (gCrashRecoveryEnabled) - return; - - gCrashRecoveryEnabled = true; - +static void installExceptionOrSignalHandlers() { // Setup the signal handler. struct sigaction Handler; Handler.sa_handler = CrashRecoverySignalHandler; @@ -297,20 +336,13 @@ void CrashRecoveryContext::Enable() { } } -void CrashRecoveryContext::Disable() { - sys::ScopedLock L(*gCrashRecoveryContextMutex); - - if (!gCrashRecoveryEnabled) - return; - - gCrashRecoveryEnabled = false; - +static void uninstallExceptionOrSignalHandlers() { // Restore the previous signal handlers. for (unsigned i = 0; i != NumSignals; ++i) sigaction(Signals[i], &PrevActions[i], nullptr); } -#endif +#endif // !LLVM_ON_WIN32 bool CrashRecoveryContext::RunSafely(function_ref Fn) { // If crash recovery is disabled, do nothing. @@ -328,6 +360,8 @@ bool CrashRecoveryContext::RunSafely(function_ref Fn) { return true; } +#endif // !_MSC_VER + void CrashRecoveryContext::HandleCrash() { CrashRecoveryContextImpl *CRCI = (CrashRecoveryContextImpl *) Impl; assert(CRCI && "Crash recovery context never initialized!"); diff --git a/lib/Support/Unix/Path.inc b/lib/Support/Unix/Path.inc index cdea09be41e0..fa28ba1b6ab6 100644 --- a/lib/Support/Unix/Path.inc +++ b/lib/Support/Unix/Path.inc @@ -103,16 +103,13 @@ #define STATVFS_F_FLAG(vfs) (vfs).f_flags #endif -#if defined(__FreeBSD__) || defined(__NetBSD__) -#include -#endif - using namespace llvm; namespace llvm { namespace sys { namespace fs { -#if defined(__Bitrig__) || defined(__OpenBSD__) || defined(__minix) || \ +#if defined(__FreeBSD__) || defined (__NetBSD__) || defined(__Bitrig__) || \ + defined(__OpenBSD__) || defined(__minix) || defined(__FreeBSD_kernel__) || \ defined(__linux__) || defined(__CYGWIN__) || defined(__DragonFly__) || \ defined(_AIX) static int @@ -167,7 +164,7 @@ getprogpath(char ret[PATH_MAX], const char *bin) free(pv); return nullptr; } -#endif // Bitrig || OpenBSD || minix || linux || CYGWIN || DragonFly || AIX +#endif // __FreeBSD__ || __NetBSD__ || __FreeBSD_kernel__ /// GetMainExecutable - Return the path to the main executable, given the /// value of argv[0] from program startup. @@ -183,24 +180,9 @@ std::string getMainExecutable(const char *argv0, void *MainAddr) { if (realpath(exe_path, link_path)) return link_path; } -#elif defined(__FreeBSD__) || defined(__FreeBSD_kernel__) || defined(__NetBSD__) - int mib[4]; - mib[0] = CTL_KERN; -#if defined(__FreeBSD__) || defined(__FreeBSD_kernel__) - mib[1] = KERN_PROC; - mib[2] = KERN_PROC_PATHNAME; - mib[3] = -1; -#else - mib[1] = KERN_PROC_ARGS; - mib[2] = -1; - mib[3] = KERN_PROC_PATHNAME; -#endif - char exe_path[PATH_MAX]; - size_t cb = sizeof(exe_path); - if (sysctl(mib, 4, exe_path, &cb, NULL, 0) == 0) - return exe_path; -#elif defined(__Bitrig__) || defined(__OpenBSD__) || defined(__minix) || \ - defined(__DragonFly__) || defined(_AIX) +#elif defined(__FreeBSD__) || defined (__NetBSD__) || defined(__Bitrig__) || \ + defined(__OpenBSD__) || defined(__minix) || defined(__DragonFly__) || \ + defined(__FreeBSD_kernel__) || defined(_AIX) char exe_path[PATH_MAX]; if (getprogpath(exe_path, argv0) != NULL) diff --git a/lib/Target/AArch64/AArch64FrameLowering.cpp b/lib/Target/AArch64/AArch64FrameLowering.cpp index dc916c034661..1aec602a2a36 100644 --- a/lib/Target/AArch64/AArch64FrameLowering.cpp +++ b/lib/Target/AArch64/AArch64FrameLowering.cpp @@ -1158,8 +1158,7 @@ void AArch64FrameLowering::determineCalleeSaves(MachineFunction &MF, } DEBUG(dbgs() << "*** determineCalleeSaves\nUsed CSRs:"; - for (int Reg = SavedRegs.find_first(); Reg != -1; - Reg = SavedRegs.find_next(Reg)) + for (unsigned Reg : SavedRegs.set_bits()) dbgs() << ' ' << PrintReg(Reg, RegInfo); dbgs() << "\n";); diff --git a/lib/Target/AArch64/AArch64ISelLowering.cpp b/lib/Target/AArch64/AArch64ISelLowering.cpp index 4f7c2e122390..1af36086ad90 100644 --- a/lib/Target/AArch64/AArch64ISelLowering.cpp +++ b/lib/Target/AArch64/AArch64ISelLowering.cpp @@ -553,7 +553,6 @@ AArch64TargetLowering::AArch64TargetLowering(const TargetMachine &TM, setTargetDAGCombine(ISD::INTRINSIC_VOID); setTargetDAGCombine(ISD::INTRINSIC_W_CHAIN); setTargetDAGCombine(ISD::INSERT_VECTOR_ELT); - setTargetDAGCombine(ISD::EXTRACT_VECTOR_ELT); MaxStoresPerMemset = MaxStoresPerMemsetOptSize = 8; MaxStoresPerMemcpy = MaxStoresPerMemcpyOptSize = 4; @@ -659,6 +658,19 @@ AArch64TargetLowering::AArch64TargetLowering(const TargetMachine &TM, setOperationAction(ISD::MUL, MVT::v4i32, Custom); setOperationAction(ISD::MUL, MVT::v2i64, Custom); + // Vector reductions + for (MVT VT : MVT::integer_valuetypes()) { + setOperationAction(ISD::VECREDUCE_ADD, VT, Custom); + setOperationAction(ISD::VECREDUCE_SMAX, VT, Custom); + setOperationAction(ISD::VECREDUCE_SMIN, VT, Custom); + setOperationAction(ISD::VECREDUCE_UMAX, VT, Custom); + setOperationAction(ISD::VECREDUCE_UMIN, VT, Custom); + } + for (MVT VT : MVT::fp_valuetypes()) { + setOperationAction(ISD::VECREDUCE_FMAX, VT, Custom); + setOperationAction(ISD::VECREDUCE_FMIN, VT, Custom); + } + setOperationAction(ISD::ANY_EXTEND, MVT::v4i32, Legal); setTruncStoreAction(MVT::v2i32, MVT::v2i16, Expand); // Likewise, narrowing and extending vector loads/stores aren't handled @@ -2606,6 +2618,14 @@ SDValue AArch64TargetLowering::LowerOperation(SDValue Op, return LowerMUL(Op, DAG); case ISD::INTRINSIC_WO_CHAIN: return LowerINTRINSIC_WO_CHAIN(Op, DAG); + case ISD::VECREDUCE_ADD: + case ISD::VECREDUCE_SMAX: + case ISD::VECREDUCE_SMIN: + case ISD::VECREDUCE_UMAX: + case ISD::VECREDUCE_UMIN: + case ISD::VECREDUCE_FMAX: + case ISD::VECREDUCE_FMIN: + return LowerVECREDUCE(Op, DAG); } } @@ -7128,6 +7148,47 @@ SDValue AArch64TargetLowering::LowerVSETCC(SDValue Op, return Cmp; } +static SDValue getReductionSDNode(unsigned Op, SDLoc DL, SDValue ScalarOp, + SelectionDAG &DAG) { + SDValue VecOp = ScalarOp.getOperand(0); + auto Rdx = DAG.getNode(Op, DL, VecOp.getSimpleValueType(), VecOp); + return DAG.getNode(ISD::EXTRACT_VECTOR_ELT, DL, ScalarOp.getValueType(), Rdx, + DAG.getConstant(0, DL, MVT::i64)); +} + +SDValue AArch64TargetLowering::LowerVECREDUCE(SDValue Op, + SelectionDAG &DAG) const { + SDLoc dl(Op); + switch (Op.getOpcode()) { + case ISD::VECREDUCE_ADD: + return getReductionSDNode(AArch64ISD::UADDV, dl, Op, DAG); + case ISD::VECREDUCE_SMAX: + return getReductionSDNode(AArch64ISD::SMAXV, dl, Op, DAG); + case ISD::VECREDUCE_SMIN: + return getReductionSDNode(AArch64ISD::SMINV, dl, Op, DAG); + case ISD::VECREDUCE_UMAX: + return getReductionSDNode(AArch64ISD::UMAXV, dl, Op, DAG); + case ISD::VECREDUCE_UMIN: + return getReductionSDNode(AArch64ISD::UMINV, dl, Op, DAG); + case ISD::VECREDUCE_FMAX: { + assert(Op->getFlags().hasNoNaNs() && "fmax vector reduction needs NoNaN flag"); + return DAG.getNode( + ISD::INTRINSIC_WO_CHAIN, dl, Op.getValueType(), + DAG.getConstant(Intrinsic::aarch64_neon_fmaxnmv, dl, MVT::i32), + Op.getOperand(0)); + } + case ISD::VECREDUCE_FMIN: { + assert(Op->getFlags().hasNoNaNs() && "fmin vector reduction needs NoNaN flag"); + return DAG.getNode( + ISD::INTRINSIC_WO_CHAIN, dl, Op.getValueType(), + DAG.getConstant(Intrinsic::aarch64_neon_fminnmv, dl, MVT::i32), + Op.getOperand(0)); + } + default: + llvm_unreachable("Unhandled reduction"); + } +} + /// getTgtMemIntrinsic - Represent NEON load and store intrinsics as /// MemIntrinsicNodes. The associated MachineMemOperands record the alignment /// specified in the intrinsic calls. @@ -9490,266 +9551,6 @@ static SDValue performSTORECombine(SDNode *N, return SDValue(); } -/// This function handles the log2-shuffle pattern produced by the -/// LoopVectorizer for the across vector reduction. It consists of -/// log2(NumVectorElements) steps and, in each step, 2^(s) elements -/// are reduced, where s is an induction variable from 0 to -/// log2(NumVectorElements). -static SDValue tryMatchAcrossLaneShuffleForReduction(SDNode *N, SDValue OpV, - unsigned Op, - SelectionDAG &DAG) { - EVT VTy = OpV->getOperand(0).getValueType(); - if (!VTy.isVector()) - return SDValue(); - - int NumVecElts = VTy.getVectorNumElements(); - if (Op == ISD::FMAXNUM || Op == ISD::FMINNUM) { - if (NumVecElts != 4) - return SDValue(); - } else { - if (NumVecElts != 4 && NumVecElts != 8 && NumVecElts != 16) - return SDValue(); - } - - int NumExpectedSteps = APInt(8, NumVecElts).logBase2(); - SDValue PreOp = OpV; - // Iterate over each step of the across vector reduction. - for (int CurStep = 0; CurStep != NumExpectedSteps; ++CurStep) { - SDValue CurOp = PreOp.getOperand(0); - SDValue Shuffle = PreOp.getOperand(1); - if (Shuffle.getOpcode() != ISD::VECTOR_SHUFFLE) { - // Try to swap the 1st and 2nd operand as add and min/max instructions - // are commutative. - CurOp = PreOp.getOperand(1); - Shuffle = PreOp.getOperand(0); - if (Shuffle.getOpcode() != ISD::VECTOR_SHUFFLE) - return SDValue(); - } - - // Check if the input vector is fed by the operator we want to handle, - // except the last step; the very first input vector is not necessarily - // the same operator we are handling. - if (CurOp.getOpcode() != Op && (CurStep != (NumExpectedSteps - 1))) - return SDValue(); - - // Check if it forms one step of the across vector reduction. - // E.g., - // %cur = add %1, %0 - // %shuffle = vector_shuffle %cur, <2, 3, u, u> - // %pre = add %cur, %shuffle - if (Shuffle.getOperand(0) != CurOp) - return SDValue(); - - int NumMaskElts = 1 << CurStep; - ArrayRef Mask = cast(Shuffle)->getMask(); - // Check mask values in each step. - // We expect the shuffle mask in each step follows a specific pattern - // denoted here by the form, where M is a sequence of integers - // starting from NumMaskElts, increasing by 1, and the number integers - // in M should be NumMaskElts. U is a sequence of UNDEFs and the number - // of undef in U should be NumVecElts - NumMaskElts. - // E.g., for <8 x i16>, mask values in each step should be : - // step 0 : <1,u,u,u,u,u,u,u> - // step 1 : <2,3,u,u,u,u,u,u> - // step 2 : <4,5,6,7,u,u,u,u> - for (int i = 0; i < NumVecElts; ++i) - if ((i < NumMaskElts && Mask[i] != (NumMaskElts + i)) || - (i >= NumMaskElts && !(Mask[i] < 0))) - return SDValue(); - - PreOp = CurOp; - } - unsigned Opcode; - bool IsIntrinsic = false; - - switch (Op) { - default: - llvm_unreachable("Unexpected operator for across vector reduction"); - case ISD::ADD: - Opcode = AArch64ISD::UADDV; - break; - case ISD::SMAX: - Opcode = AArch64ISD::SMAXV; - break; - case ISD::UMAX: - Opcode = AArch64ISD::UMAXV; - break; - case ISD::SMIN: - Opcode = AArch64ISD::SMINV; - break; - case ISD::UMIN: - Opcode = AArch64ISD::UMINV; - break; - case ISD::FMAXNUM: - Opcode = Intrinsic::aarch64_neon_fmaxnmv; - IsIntrinsic = true; - break; - case ISD::FMINNUM: - Opcode = Intrinsic::aarch64_neon_fminnmv; - IsIntrinsic = true; - break; - } - SDLoc DL(N); - - return IsIntrinsic - ? DAG.getNode(ISD::INTRINSIC_WO_CHAIN, DL, N->getValueType(0), - DAG.getConstant(Opcode, DL, MVT::i32), PreOp) - : DAG.getNode( - ISD::EXTRACT_VECTOR_ELT, DL, N->getValueType(0), - DAG.getNode(Opcode, DL, PreOp.getSimpleValueType(), PreOp), - DAG.getConstant(0, DL, MVT::i64)); -} - -/// Target-specific DAG combine for the across vector min/max reductions. -/// This function specifically handles the final clean-up step of the vector -/// min/max reductions produced by the LoopVectorizer. It is the log2-shuffle -/// pattern, which narrows down and finds the final min/max value from all -/// elements of the vector. -/// For example, for a <16 x i8> vector : -/// svn0 = vector_shuffle %0, undef<8,9,10,11,12,13,14,15,u,u,u,u,u,u,u,u> -/// %smax0 = smax %arr, svn0 -/// %svn1 = vector_shuffle %smax0, undef<4,5,6,7,u,u,u,u,u,u,u,u,u,u,u,u> -/// %smax1 = smax %smax0, %svn1 -/// %svn2 = vector_shuffle %smax1, undef<2,3,u,u,u,u,u,u,u,u,u,u,u,u,u,u> -/// %smax2 = smax %smax1, svn2 -/// %svn3 = vector_shuffle %smax2, undef<1,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u> -/// %sc = setcc %smax2, %svn3, gt -/// %n0 = extract_vector_elt %sc, #0 -/// %n1 = extract_vector_elt %smax2, #0 -/// %n2 = extract_vector_elt $smax2, #1 -/// %result = select %n0, %n1, n2 -/// becomes : -/// %1 = smaxv %0 -/// %result = extract_vector_elt %1, 0 -static SDValue -performAcrossLaneMinMaxReductionCombine(SDNode *N, SelectionDAG &DAG, - const AArch64Subtarget *Subtarget) { - if (!Subtarget->hasNEON()) - return SDValue(); - - SDValue N0 = N->getOperand(0); - SDValue IfTrue = N->getOperand(1); - SDValue IfFalse = N->getOperand(2); - - // Check if the SELECT merges up the final result of the min/max - // from a vector. - if (N0.getOpcode() != ISD::EXTRACT_VECTOR_ELT || - IfTrue.getOpcode() != ISD::EXTRACT_VECTOR_ELT || - IfFalse.getOpcode() != ISD::EXTRACT_VECTOR_ELT) - return SDValue(); - - // Expect N0 is fed by SETCC. - SDValue SetCC = N0.getOperand(0); - EVT SetCCVT = SetCC.getValueType(); - if (SetCC.getOpcode() != ISD::SETCC || !SetCCVT.isVector() || - SetCCVT.getVectorElementType() != MVT::i1) - return SDValue(); - - SDValue VectorOp = SetCC.getOperand(0); - unsigned Op = VectorOp->getOpcode(); - // Check if the input vector is fed by the operator we want to handle. - if (Op != ISD::SMAX && Op != ISD::UMAX && Op != ISD::SMIN && - Op != ISD::UMIN && Op != ISD::FMAXNUM && Op != ISD::FMINNUM) - return SDValue(); - - EVT VTy = VectorOp.getValueType(); - if (!VTy.isVector()) - return SDValue(); - - if (VTy.getSizeInBits() < 64) - return SDValue(); - - EVT EltTy = VTy.getVectorElementType(); - if (Op == ISD::FMAXNUM || Op == ISD::FMINNUM) { - if (EltTy != MVT::f32) - return SDValue(); - } else { - if (EltTy != MVT::i32 && EltTy != MVT::i16 && EltTy != MVT::i8) - return SDValue(); - } - - // Check if extracting from the same vector. - // For example, - // %sc = setcc %vector, %svn1, gt - // %n0 = extract_vector_elt %sc, #0 - // %n1 = extract_vector_elt %vector, #0 - // %n2 = extract_vector_elt $vector, #1 - if (!(VectorOp == IfTrue->getOperand(0) && - VectorOp == IfFalse->getOperand(0))) - return SDValue(); - - // Check if the condition code is matched with the operator type. - ISD::CondCode CC = cast(SetCC->getOperand(2))->get(); - if ((Op == ISD::SMAX && CC != ISD::SETGT && CC != ISD::SETGE) || - (Op == ISD::UMAX && CC != ISD::SETUGT && CC != ISD::SETUGE) || - (Op == ISD::SMIN && CC != ISD::SETLT && CC != ISD::SETLE) || - (Op == ISD::UMIN && CC != ISD::SETULT && CC != ISD::SETULE) || - (Op == ISD::FMAXNUM && CC != ISD::SETOGT && CC != ISD::SETOGE && - CC != ISD::SETUGT && CC != ISD::SETUGE && CC != ISD::SETGT && - CC != ISD::SETGE) || - (Op == ISD::FMINNUM && CC != ISD::SETOLT && CC != ISD::SETOLE && - CC != ISD::SETULT && CC != ISD::SETULE && CC != ISD::SETLT && - CC != ISD::SETLE)) - return SDValue(); - - // Expect to check only lane 0 from the vector SETCC. - if (!isNullConstant(N0.getOperand(1))) - return SDValue(); - - // Expect to extract the true value from lane 0. - if (!isNullConstant(IfTrue.getOperand(1))) - return SDValue(); - - // Expect to extract the false value from lane 1. - if (!isOneConstant(IfFalse.getOperand(1))) - return SDValue(); - - return tryMatchAcrossLaneShuffleForReduction(N, SetCC, Op, DAG); -} - -/// Target-specific DAG combine for the across vector add reduction. -/// This function specifically handles the final clean-up step of the vector -/// add reduction produced by the LoopVectorizer. It is the log2-shuffle -/// pattern, which adds all elements of a vector together. -/// For example, for a <4 x i32> vector : -/// %1 = vector_shuffle %0, <2,3,u,u> -/// %2 = add %0, %1 -/// %3 = vector_shuffle %2, <1,u,u,u> -/// %4 = add %2, %3 -/// %result = extract_vector_elt %4, 0 -/// becomes : -/// %0 = uaddv %0 -/// %result = extract_vector_elt %0, 0 -static SDValue -performAcrossLaneAddReductionCombine(SDNode *N, SelectionDAG &DAG, - const AArch64Subtarget *Subtarget) { - if (!Subtarget->hasNEON()) - return SDValue(); - SDValue N0 = N->getOperand(0); - SDValue N1 = N->getOperand(1); - - // Check if the input vector is fed by the ADD. - if (N0->getOpcode() != ISD::ADD) - return SDValue(); - - // The vector extract idx must constant zero because we only expect the final - // result of the reduction is placed in lane 0. - if (!isNullConstant(N1)) - return SDValue(); - - EVT VTy = N0.getValueType(); - if (!VTy.isVector()) - return SDValue(); - - EVT EltTy = VTy.getVectorElementType(); - if (EltTy != MVT::i32 && EltTy != MVT::i16 && EltTy != MVT::i8) - return SDValue(); - - if (VTy.getSizeInBits() < 64) - return SDValue(); - - return tryMatchAcrossLaneShuffleForReduction(N, N0, ISD::ADD, DAG); -} /// Target-specific DAG combine function for NEON load/store intrinsics /// to merge base address updates. @@ -10428,12 +10229,8 @@ SDValue AArch64TargetLowering::PerformDAGCombine(SDNode *N, return performBitcastCombine(N, DCI, DAG); case ISD::CONCAT_VECTORS: return performConcatVectorsCombine(N, DCI, DAG); - case ISD::SELECT: { - SDValue RV = performSelectCombine(N, DCI); - if (!RV.getNode()) - RV = performAcrossLaneMinMaxReductionCombine(N, DAG, Subtarget); - return RV; - } + case ISD::SELECT: + return performSelectCombine(N, DCI); case ISD::VSELECT: return performVSelectCombine(N, DCI.DAG); case ISD::LOAD: @@ -10455,8 +10252,6 @@ SDValue AArch64TargetLowering::PerformDAGCombine(SDNode *N, return performNVCASTCombine(N); case ISD::INSERT_VECTOR_ELT: return performPostLD1Combine(N, DCI, true); - case ISD::EXTRACT_VECTOR_ELT: - return performAcrossLaneAddReductionCombine(N, DAG, Subtarget); case ISD::INTRINSIC_VOID: case ISD::INTRINSIC_W_CHAIN: switch (cast(N->getOperand(1))->getZExtValue()) { @@ -10676,6 +10471,14 @@ void AArch64TargetLowering::ReplaceNodeResults( case ISD::BITCAST: ReplaceBITCASTResults(N, Results, DAG); return; + case ISD::VECREDUCE_ADD: + case ISD::VECREDUCE_SMAX: + case ISD::VECREDUCE_SMIN: + case ISD::VECREDUCE_UMAX: + case ISD::VECREDUCE_UMIN: + Results.push_back(LowerVECREDUCE(SDValue(N, 0), DAG)); + return; + case AArch64ISD::SADDV: ReplaceReductionResults(N, Results, DAG, ISD::ADD, AArch64ISD::SADDV); return; diff --git a/lib/Target/AArch64/AArch64ISelLowering.h b/lib/Target/AArch64/AArch64ISelLowering.h index 89db566c219c..ecc2517fb288 100644 --- a/lib/Target/AArch64/AArch64ISelLowering.h +++ b/lib/Target/AArch64/AArch64ISelLowering.h @@ -568,6 +568,7 @@ class AArch64TargetLowering : public TargetLowering { SDValue LowerVectorOR(SDValue Op, SelectionDAG &DAG) const; SDValue LowerCONCAT_VECTORS(SDValue Op, SelectionDAG &DAG) const; SDValue LowerFSINCOS(SDValue Op, SelectionDAG &DAG) const; + SDValue LowerVECREDUCE(SDValue Op, SelectionDAG &DAG) const; SDValue BuildSDIVPow2(SDNode *N, const APInt &Divisor, SelectionDAG &DAG, std::vector *Created) const override; diff --git a/lib/Target/AArch64/AArch64TargetTransformInfo.cpp b/lib/Target/AArch64/AArch64TargetTransformInfo.cpp index 7c6f55c06bce..43569af04347 100644 --- a/lib/Target/AArch64/AArch64TargetTransformInfo.cpp +++ b/lib/Target/AArch64/AArch64TargetTransformInfo.cpp @@ -769,3 +769,28 @@ unsigned AArch64TTIImpl::getMinPrefetchStride() { unsigned AArch64TTIImpl::getMaxPrefetchIterationsAhead() { return ST->getMaxPrefetchIterationsAhead(); } + +bool AArch64TTIImpl::useReductionIntrinsic(unsigned Opcode, Type *Ty, + TTI::ReductionFlags Flags) const { + assert(isa(Ty) && "Expected Ty to be a vector type"); + unsigned ScalarBits = Ty->getScalarSizeInBits(); + switch (Opcode) { + case Instruction::FAdd: + case Instruction::FMul: + case Instruction::And: + case Instruction::Or: + case Instruction::Xor: + case Instruction::Mul: + return false; + case Instruction::Add: + return ScalarBits * Ty->getVectorNumElements() >= 128; + case Instruction::ICmp: + return (ScalarBits < 64) && + (ScalarBits * Ty->getVectorNumElements() >= 128); + case Instruction::FCmp: + return Flags.NoNaN; + default: + llvm_unreachable("Unhandled reduction opcode"); + } + return false; +} diff --git a/lib/Target/AArch64/AArch64TargetTransformInfo.h b/lib/Target/AArch64/AArch64TargetTransformInfo.h index 280d97f3c502..d0299149c38c 100644 --- a/lib/Target/AArch64/AArch64TargetTransformInfo.h +++ b/lib/Target/AArch64/AArch64TargetTransformInfo.h @@ -145,6 +145,9 @@ class AArch64TTIImpl : public BasicTTIImplBase { bool shouldExpandReduction(const IntrinsicInst *II) const { return false; } + + bool useReductionIntrinsic(unsigned Opcode, Type *Ty, + TTI::ReductionFlags Flags) const; /// @} }; diff --git a/lib/Target/AMDGPU/AMDGPUISelDAGToDAG.cpp b/lib/Target/AMDGPU/AMDGPUISelDAGToDAG.cpp index 7c99752b881f..c3ac796a0a44 100644 --- a/lib/Target/AMDGPU/AMDGPUISelDAGToDAG.cpp +++ b/lib/Target/AMDGPU/AMDGPUISelDAGToDAG.cpp @@ -1707,10 +1707,38 @@ bool AMDGPUDAGToDAGISel::SelectVOP3PMods(SDValue In, SDValue &Src, // FIXME: Look for on separate components if (Src.getOpcode() == ISD::FNEG) { - Mods |= (SISrcMods::NEG | SISrcMods::NEG_HI); + Mods ^= (SISrcMods::NEG | SISrcMods::NEG_HI); Src = Src.getOperand(0); } + if (Src.getOpcode() == ISD::BUILD_VECTOR) { + unsigned VecMods = Mods; + + SDValue Lo = Src.getOperand(0); + SDValue Hi = Src.getOperand(1); + + if (Lo.getOpcode() == ISD::FNEG) { + Lo = Lo.getOperand(0); + Mods ^= SISrcMods::NEG; + } + + if (Hi.getOpcode() == ISD::FNEG) { + Hi = Hi.getOperand(0); + Mods ^= SISrcMods::NEG_HI; + } + + if (Lo == Hi && !isInlineImmediate(Lo.getNode())) { + // Really a scalar input. Just select from the low half of the register to + // avoid packing. + + Src = Lo; + SrcMods = CurDAG->getTargetConstant(Mods, SDLoc(In), MVT::i32); + return true; + } + + Mods = VecMods; + } + // Packed instructions do not have abs modifiers. // FIXME: Handle abs/neg of individual components. diff --git a/lib/Target/AMDGPU/AMDGPUSubtarget.h b/lib/Target/AMDGPU/AMDGPUSubtarget.h index bed7d326b3dd..e543cae07ada 100644 --- a/lib/Target/AMDGPU/AMDGPUSubtarget.h +++ b/lib/Target/AMDGPU/AMDGPUSubtarget.h @@ -289,6 +289,10 @@ class AMDGPUSubtarget : public AMDGPUGenSubtargetInfo { return getGeneration() >= GFX9; } + bool hasMin3Max3_16() const { + return getGeneration() >= GFX9; + } + bool hasCARRY() const { return (getGeneration() >= EVERGREEN); } diff --git a/lib/Target/AMDGPU/SIISelLowering.cpp b/lib/Target/AMDGPU/SIISelLowering.cpp index 48a14e4dbea2..286be355bc14 100644 --- a/lib/Target/AMDGPU/SIISelLowering.cpp +++ b/lib/Target/AMDGPU/SIISelLowering.cpp @@ -4491,7 +4491,8 @@ SDValue SITargetLowering::performMinMaxCombine(SDNode *N, if (Opc != AMDGPUISD::FMIN_LEGACY && Opc != AMDGPUISD::FMAX_LEGACY && - VT != MVT::f64) { + VT != MVT::f64 && + ((VT != MVT::f16 && VT != MVT::i16) || Subtarget->hasMin3Max3_16())) { // max(max(a, b), c) -> max3(a, b, c) // min(min(a, b), c) -> min3(a, b, c) if (Op0.getOpcode() == Opc && Op0.hasOneUse()) { diff --git a/lib/Target/AMDGPU/SIInstrInfo.cpp b/lib/Target/AMDGPU/SIInstrInfo.cpp index 065fd09eb356..38a16b525a75 100644 --- a/lib/Target/AMDGPU/SIInstrInfo.cpp +++ b/lib/Target/AMDGPU/SIInstrInfo.cpp @@ -765,7 +765,7 @@ void SIInstrInfo::storeRegToStackSlot(MachineBasicBlock &MBB, .addFrameIndex(FrameIndex) // addr .addMemOperand(MMO) .addReg(MFI->getScratchRSrcReg(), RegState::Implicit) - .addReg(MFI->getScratchWaveOffsetReg(), RegState::Implicit); + .addReg(MFI->getFrameOffsetReg(), RegState::Implicit); // Add the scratch resource registers as implicit uses because we may end up // needing them, and need to ensure that the reserved registers are // correctly handled. @@ -796,7 +796,7 @@ void SIInstrInfo::storeRegToStackSlot(MachineBasicBlock &MBB, .addReg(SrcReg, getKillRegState(isKill)) // data .addFrameIndex(FrameIndex) // addr .addReg(MFI->getScratchRSrcReg()) // scratch_rsrc - .addReg(MFI->getScratchWaveOffsetReg()) // scratch_offset + .addReg(MFI->getFrameOffsetReg()) // scratch_offset .addImm(0) // offset .addMemOperand(MMO); } @@ -869,7 +869,7 @@ void SIInstrInfo::loadRegFromStackSlot(MachineBasicBlock &MBB, .addFrameIndex(FrameIndex) // addr .addMemOperand(MMO) .addReg(MFI->getScratchRSrcReg(), RegState::Implicit) - .addReg(MFI->getScratchWaveOffsetReg(), RegState::Implicit); + .addReg(MFI->getFrameOffsetReg(), RegState::Implicit); if (ST.hasScalarStores()) { // m0 is used for offset to scalar stores if used to spill. @@ -892,10 +892,10 @@ void SIInstrInfo::loadRegFromStackSlot(MachineBasicBlock &MBB, unsigned Opcode = getVGPRSpillRestoreOpcode(SpillSize); BuildMI(MBB, MI, DL, get(Opcode), DestReg) - .addFrameIndex(FrameIndex) // vaddr - .addReg(MFI->getScratchRSrcReg()) // scratch_rsrc - .addReg(MFI->getScratchWaveOffsetReg()) // scratch_offset - .addImm(0) // offset + .addFrameIndex(FrameIndex) // vaddr + .addReg(MFI->getScratchRSrcReg()) // scratch_rsrc + .addReg(MFI->getFrameOffsetReg()) // scratch_offset + .addImm(0) // offset .addMemOperand(MMO); } diff --git a/lib/Target/AMDGPU/SIRegisterInfo.cpp b/lib/Target/AMDGPU/SIRegisterInfo.cpp index 8820e294562b..06cfc95be96a 100644 --- a/lib/Target/AMDGPU/SIRegisterInfo.cpp +++ b/lib/Target/AMDGPU/SIRegisterInfo.cpp @@ -654,11 +654,11 @@ bool SIRegisterInfo::spillSGPR(MachineBasicBlock::iterator MI, int64_t Offset = (ST.getWavefrontSize() * FrOffset) + (EltSize * i); if (Offset != 0) { BuildMI(*MBB, MI, DL, TII->get(AMDGPU::S_ADD_U32), OffsetReg) - .addReg(MFI->getScratchWaveOffsetReg()) + .addReg(MFI->getFrameOffsetReg()) .addImm(Offset); } else { BuildMI(*MBB, MI, DL, TII->get(AMDGPU::S_MOV_B32), OffsetReg) - .addReg(MFI->getScratchWaveOffsetReg()); + .addReg(MFI->getFrameOffsetReg()); } BuildMI(*MBB, MI, DL, TII->get(ScalarStoreOp)) @@ -715,11 +715,11 @@ bool SIRegisterInfo::spillSGPR(MachineBasicBlock::iterator MI, = MF->getMachineMemOperand(PtrInfo, MachineMemOperand::MOStore, EltSize, MinAlign(Align, EltSize * i)); BuildMI(*MBB, MI, DL, TII->get(AMDGPU::SI_SPILL_V32_SAVE)) - .addReg(TmpReg, RegState::Kill) // src - .addFrameIndex(Index) // vaddr - .addReg(MFI->getScratchRSrcReg()) // srrsrc - .addReg(MFI->getScratchWaveOffsetReg()) // soffset - .addImm(i * 4) // offset + .addReg(TmpReg, RegState::Kill) // src + .addFrameIndex(Index) // vaddr + .addReg(MFI->getScratchRSrcReg()) // srrsrc + .addReg(MFI->getFrameOffsetReg()) // soffset + .addImm(i * 4) // offset .addMemOperand(MMO); } } @@ -806,11 +806,11 @@ bool SIRegisterInfo::restoreSGPR(MachineBasicBlock::iterator MI, int64_t Offset = (ST.getWavefrontSize() * FrOffset) + (EltSize * i); if (Offset != 0) { BuildMI(*MBB, MI, DL, TII->get(AMDGPU::S_ADD_U32), OffsetReg) - .addReg(MFI->getScratchWaveOffsetReg()) + .addReg(MFI->getFrameOffsetReg()) .addImm(Offset); } else { BuildMI(*MBB, MI, DL, TII->get(AMDGPU::S_MOV_B32), OffsetReg) - .addReg(MFI->getScratchWaveOffsetReg()); + .addReg(MFI->getFrameOffsetReg()); } auto MIB = @@ -853,10 +853,10 @@ bool SIRegisterInfo::restoreSGPR(MachineBasicBlock::iterator MI, MinAlign(Align, EltSize * i)); BuildMI(*MBB, MI, DL, TII->get(AMDGPU::SI_SPILL_V32_RESTORE), TmpReg) - .addFrameIndex(Index) // vaddr - .addReg(MFI->getScratchRSrcReg()) // srsrc - .addReg(MFI->getScratchWaveOffsetReg()) // soffset - .addImm(i * 4) // offset + .addFrameIndex(Index) // vaddr + .addReg(MFI->getScratchRSrcReg()) // srsrc + .addReg(MFI->getFrameOffsetReg()) // soffset + .addImm(i * 4) // offset .addMemOperand(MMO); auto MIB = diff --git a/lib/Target/AMDGPU/VOP3Instructions.td b/lib/Target/AMDGPU/VOP3Instructions.td index ffa6c60d6b1f..c0b5069948fb 100644 --- a/lib/Target/AMDGPU/VOP3Instructions.td +++ b/lib/Target/AMDGPU/VOP3Instructions.td @@ -300,10 +300,19 @@ def V_AND_OR_B32 : VOP3Inst <"v_and_or_b32", VOP3_Profile>; def V_OR3_B32 : VOP3Inst <"v_or3_b32", VOP3_Profile>; def V_XAD_U32 : VOP3Inst <"v_xad_u32", VOP3_Profile>; + def V_MED3_F16 : VOP3Inst <"v_med3_f16", VOP3_Profile, AMDGPUfmed3>; def V_MED3_I16 : VOP3Inst <"v_med3_i16", VOP3_Profile, AMDGPUsmed3>; def V_MED3_U16 : VOP3Inst <"v_med3_u16", VOP3_Profile, AMDGPUumed3>; -} + +def V_MIN3_F16 : VOP3Inst <"v_min3_f16", VOP3_Profile, AMDGPUfmin3>; +def V_MIN3_I16 : VOP3Inst <"v_min3_i16", VOP3_Profile, AMDGPUsmin3>; +def V_MIN3_U16 : VOP3Inst <"v_min3_u16", VOP3_Profile, AMDGPUumin3>; + +def V_MAX3_F16 : VOP3Inst <"v_max3_f16", VOP3_Profile, AMDGPUfmax3>; +def V_MAX3_I16 : VOP3Inst <"v_max3_i16", VOP3_Profile, AMDGPUsmax3>; +def V_MAX3_U16 : VOP3Inst <"v_max3_u16", VOP3_Profile, AMDGPUumax3>; +} // End SubtargetPredicate = isGFX9 //===----------------------------------------------------------------------===// @@ -509,6 +518,15 @@ defm V_OR3_B32 : VOP3_Real_vi <0x202>; defm V_PACK_B32_F16 : VOP3_Real_vi <0x2a0>; defm V_XAD_U32 : VOP3_Real_vi <0x1f3>; + +defm V_MIN3_F16 : VOP3_Real_vi <0x1f4>; +defm V_MIN3_I16 : VOP3_Real_vi <0x1f5>; +defm V_MIN3_U16 : VOP3_Real_vi <0x1f6>; + +defm V_MAX3_F16 : VOP3_Real_vi <0x1f7>; +defm V_MAX3_I16 : VOP3_Real_vi <0x1f8>; +defm V_MAX3_U16 : VOP3_Real_vi <0x1f9>; + defm V_MED3_F16 : VOP3_Real_vi <0x1fa>; defm V_MED3_I16 : VOP3_Real_vi <0x1fb>; defm V_MED3_U16 : VOP3_Real_vi <0x1fc>; diff --git a/lib/Target/ARM/ARMInstructionSelector.cpp b/lib/Target/ARM/ARMInstructionSelector.cpp index 8c680cdf9b47..b1f059835ff5 100644 --- a/lib/Target/ARM/ARMInstructionSelector.cpp +++ b/lib/Target/ARM/ARMInstructionSelector.cpp @@ -345,25 +345,10 @@ bool ARMInstructionSelector::select(MachineInstr &I) const { I.setDesc(TII.get(COPY)); return selectCopy(I, TII, MRI, TRI, RBI); } - case G_ADD: case G_GEP: I.setDesc(TII.get(ARM::ADDrr)); MIB.add(predOps(ARMCC::AL)).add(condCodeOp()); break; - case G_SUB: - I.setDesc(TII.get(ARM::SUBrr)); - MIB.add(predOps(ARMCC::AL)).add(condCodeOp()); - break; - case G_MUL: - if (TII.getSubtarget().hasV6Ops()) { - I.setDesc(TII.get(ARM::MUL)); - } else { - assert(TII.getSubtarget().useMulOps() && "Unsupported target"); - I.setDesc(TII.get(ARM::MULv5)); - MIB->getOperand(0).setIsEarlyClobber(true); - } - MIB.add(predOps(ARMCC::AL)).add(condCodeOp()); - break; case G_FRAME_INDEX: // Add 0 to the given frame index and hope it will eventually be folded into // the user(s). diff --git a/lib/Target/ARM/Thumb1FrameLowering.cpp b/lib/Target/ARM/Thumb1FrameLowering.cpp index d0fd366ab9ed..1a17d4e33e4f 100644 --- a/lib/Target/ARM/Thumb1FrameLowering.cpp +++ b/lib/Target/ARM/Thumb1FrameLowering.cpp @@ -571,8 +571,7 @@ bool Thumb1FrameLowering::emitPopSpecialFixUp(MachineBasicBlock &MBB, GPRsNoLRSP.reset(ARM::LR); GPRsNoLRSP.reset(ARM::SP); GPRsNoLRSP.reset(ARM::PC); - for (int Register = GPRsNoLRSP.find_first(); Register != -1; - Register = GPRsNoLRSP.find_next(Register)) { + for (unsigned Register : GPRsNoLRSP.set_bits()) { if (!UsedRegs.contains(Register)) { // Remember the first pop-friendly register and exit. if (PopFriendly.test(Register)) { diff --git a/lib/Target/Mips/MipsDelaySlotFiller.cpp b/lib/Target/Mips/MipsDelaySlotFiller.cpp index ae58c26e145a..1597057ad63f 100644 --- a/lib/Target/Mips/MipsDelaySlotFiller.cpp +++ b/lib/Target/Mips/MipsDelaySlotFiller.cpp @@ -386,7 +386,7 @@ void RegDefsUses::setCallerSaved(const MachineInstr &MI) { void RegDefsUses::setUnallocatableRegs(const MachineFunction &MF) { BitVector AllocSet = TRI.getAllocatableSet(MF); - for (int R = AllocSet.find_first(); R != -1; R = AllocSet.find_next(R)) + for (unsigned R : AllocSet.set_bits()) for (MCRegAliasIterator AI(R, &TRI, false); AI.isValid(); ++AI) AllocSet.set(*AI); diff --git a/lib/Target/PowerPC/PPCFrameLowering.cpp b/lib/Target/PowerPC/PPCFrameLowering.cpp index 40bfe3a449f7..57a1d373c88c 100644 --- a/lib/Target/PowerPC/PPCFrameLowering.cpp +++ b/lib/Target/PowerPC/PPCFrameLowering.cpp @@ -1765,31 +1765,36 @@ void PPCFrameLowering::processFunctionBeforeFrameFinalized(MachineFunction &MF, // Check whether the frame pointer register is allocated. If so, make sure it // is spilled to the correct offset. if (needsFP(MF)) { - HasGPSaveArea = true; - int FI = PFI->getFramePointerSaveIndex(); assert(FI && "No Frame Pointer Save Slot!"); - MFI.setObjectOffset(FI, LowerBound + MFI.getObjectOffset(FI)); + // FP is R31/X31, so no need to update MinGPR/MinG8R. + HasGPSaveArea = true; } if (PFI->usesPICBase()) { - HasGPSaveArea = true; - int FI = PFI->getPICBasePointerSaveIndex(); assert(FI && "No PIC Base Pointer Save Slot!"); - MFI.setObjectOffset(FI, LowerBound + MFI.getObjectOffset(FI)); + + MinGPR = std::min(MinGPR, PPC::R30); + HasGPSaveArea = true; } const PPCRegisterInfo *RegInfo = Subtarget.getRegisterInfo(); if (RegInfo->hasBasePointer(MF)) { - HasGPSaveArea = true; - int FI = PFI->getBasePointerSaveIndex(); assert(FI && "No Base Pointer Save Slot!"); - MFI.setObjectOffset(FI, LowerBound + MFI.getObjectOffset(FI)); + + unsigned BP = RegInfo->getBaseRegister(MF); + if (PPC::G8RCRegClass.contains(BP)) { + MinG8R = std::min(MinG8R, BP); + HasG8SaveArea = true; + } else if (PPC::GPRCRegClass.contains(BP)) { + MinGPR = std::min(MinGPR, BP); + HasGPSaveArea = true; + } } // General register save area starts right below the Floating-point diff --git a/lib/Target/PowerPC/PPCISelLowering.cpp b/lib/Target/PowerPC/PPCISelLowering.cpp index 17bdd595da10..144aea850833 100644 --- a/lib/Target/PowerPC/PPCISelLowering.cpp +++ b/lib/Target/PowerPC/PPCISelLowering.cpp @@ -410,6 +410,11 @@ PPCTargetLowering::PPCTargetLowering(const PPCTargetMachine &TM, // To handle counter-based loop conditions. setOperationAction(ISD::INTRINSIC_W_CHAIN, MVT::i1, Custom); + setOperationAction(ISD::INTRINSIC_VOID, MVT::i8, Custom); + setOperationAction(ISD::INTRINSIC_VOID, MVT::i16, Custom); + setOperationAction(ISD::INTRINSIC_VOID, MVT::i32, Custom); + setOperationAction(ISD::INTRINSIC_VOID, MVT::Other, Custom); + // Comparisons that require checking two conditions. setCondCodeAction(ISD::SETULT, MVT::f32, Expand); setCondCodeAction(ISD::SETULT, MVT::f64, Expand); @@ -8184,6 +8189,26 @@ SDValue PPCTargetLowering::LowerINTRINSIC_WO_CHAIN(SDValue Op, return Flags; } +SDValue PPCTargetLowering::LowerINTRINSIC_VOID(SDValue Op, + SelectionDAG &DAG) const { + // SelectionDAGBuilder::visitTargetIntrinsic may insert one extra chain to + // the beginning of the argument list. + int ArgStart = isa(Op.getOperand(0)) ? 0 : 1; + SDLoc DL(Op); + switch (cast(Op.getOperand(ArgStart))->getZExtValue()) { + case Intrinsic::ppc_cfence: { + assert(Subtarget.isPPC64() && "Only 64-bit is supported for now."); + return SDValue(DAG.getMachineNode(PPC::CFENCE8, DL, MVT::Other, + DAG.getNode(ISD::ANY_EXTEND, DL, MVT::i64, + Op.getOperand(ArgStart + 1))), + 0); + } + default: + break; + } + return SDValue(); +} + SDValue PPCTargetLowering::LowerSIGN_EXTEND_INREG(SDValue Op, SelectionDAG &DAG) const { SDLoc dl(Op); @@ -8649,6 +8674,9 @@ SDValue PPCTargetLowering::LowerOperation(SDValue Op, SelectionDAG &DAG) const { // Frame & Return address. case ISD::RETURNADDR: return LowerRETURNADDR(Op, DAG); case ISD::FRAMEADDR: return LowerFRAMEADDR(Op, DAG); + + case ISD::INTRINSIC_VOID: + return LowerINTRINSIC_VOID(Op, DAG); } } @@ -8753,12 +8781,19 @@ Instruction *PPCTargetLowering::emitLeadingFence(IRBuilder<> &Builder, Instruction *PPCTargetLowering::emitTrailingFence(IRBuilder<> &Builder, Instruction *Inst, AtomicOrdering Ord) const { - if (Inst->hasAtomicLoad() && isAcquireOrStronger(Ord)) + if (Inst->hasAtomicLoad() && isAcquireOrStronger(Ord)) { + // See http://www.cl.cam.ac.uk/~pes20/cpp/cpp0xmappings.html and + // http://www.rdrop.com/users/paulmck/scalability/paper/N2745r.2011.03.04a.html + // and http://www.cl.cam.ac.uk/~pes20/cppppc/ for justification. + if (isa(Inst) && Subtarget.isPPC64()) + return Builder.CreateCall( + Intrinsic::getDeclaration( + Builder.GetInsertBlock()->getParent()->getParent(), + Intrinsic::ppc_cfence, {Inst->getType()}), + {Inst}); + // FIXME: Can use isync for rmw operation. return callIntrinsic(Builder, Intrinsic::ppc_lwsync); - // FIXME: this is too conservative, a dependent branch + isync is enough. - // See http://www.cl.cam.ac.uk/~pes20/cpp/cpp0xmappings.html and - // http://www.rdrop.com/users/paulmck/scalability/paper/N2745r.2011.03.04a.html - // and http://www.cl.cam.ac.uk/~pes20/cppppc/ for justification. + } return nullptr; } diff --git a/lib/Target/PowerPC/PPCISelLowering.h b/lib/Target/PowerPC/PPCISelLowering.h index 4fc744257262..acb77943b118 100644 --- a/lib/Target/PowerPC/PPCISelLowering.h +++ b/lib/Target/PowerPC/PPCISelLowering.h @@ -905,6 +905,7 @@ namespace llvm { SDValue LowerINSERT_VECTOR_ELT(SDValue Op, SelectionDAG &DAG) const; SDValue LowerEXTRACT_VECTOR_ELT(SDValue Op, SelectionDAG &DAG) const; SDValue LowerINTRINSIC_WO_CHAIN(SDValue Op, SelectionDAG &DAG) const; + SDValue LowerINTRINSIC_VOID(SDValue Op, SelectionDAG &DAG) const; SDValue LowerSCALAR_TO_VECTOR(SDValue Op, SelectionDAG &DAG) const; SDValue LowerSIGN_EXTEND_INREG(SDValue Op, SelectionDAG &DAG) const; SDValue LowerMUL(SDValue Op, SelectionDAG &DAG) const; diff --git a/lib/Target/PowerPC/PPCInstr64Bit.td b/lib/Target/PowerPC/PPCInstr64Bit.td index a8433919f0f3..a3f894c81a01 100644 --- a/lib/Target/PowerPC/PPCInstr64Bit.td +++ b/lib/Target/PowerPC/PPCInstr64Bit.td @@ -983,6 +983,10 @@ def LDgotTprelL: Pseudo<(outs g8rc:$rD), (ins s16imm64:$disp, g8rc_nox0:$reg), [(set i64:$rD, (PPCldGotTprelL tglobaltlsaddr:$disp, i64:$reg))]>, isPPC64; + +let isBarrier = 1, isPseudo = 1, Defs = [CR7], Itinerary = IIC_LdStSync in +def CFENCE8 : Pseudo<(outs), (ins g8rc:$cr), "#CFENCE8", []>; + def : Pat<(PPCaddTls i64:$in, tglobaltlsaddr:$g), (ADD8TLS $in, tglobaltlsaddr:$g)>; def ADDIStlsgdHA: Pseudo<(outs g8rc:$rD), (ins g8rc_nox0:$reg, s16imm64:$disp), diff --git a/lib/Target/PowerPC/PPCInstrInfo.cpp b/lib/Target/PowerPC/PPCInstrInfo.cpp index 790a8902b3d2..3afcec1248d5 100644 --- a/lib/Target/PowerPC/PPCInstrInfo.cpp +++ b/lib/Target/PowerPC/PPCInstrInfo.cpp @@ -1873,6 +1873,8 @@ PPCInstrInfo::getSerializableBitmaskMachineOperandTargetFlags() const { } bool PPCInstrInfo::expandPostRAPseudo(MachineInstr &MI) const { + auto &MBB = *MI.getParent(); + auto DL = MI.getDebugLoc(); switch (MI.getOpcode()) { case TargetOpcode::LOAD_STACK_GUARD: { assert(Subtarget.isTargetLinux() && @@ -1920,6 +1922,17 @@ bool PPCInstrInfo::expandPostRAPseudo(MachineInstr &MI) const { MI.setDesc(get(Opcode)); return true; } + case PPC::CFENCE8: { + auto Val = MI.getOperand(0).getReg(); + BuildMI(MBB, MI, DL, get(PPC::CMPW), PPC::CR7).addReg(Val).addReg(Val); + BuildMI(MBB, MI, DL, get(PPC::CTRL_DEP)) + .addImm(PPC::PRED_NE_MINUS) + .addReg(PPC::CR7) + .addImm(1); + MI.setDesc(get(PPC::ISYNC)); + MI.RemoveOperand(0); + return true; + } } return false; } diff --git a/lib/Target/PowerPC/PPCInstrInfo.td b/lib/Target/PowerPC/PPCInstrInfo.td index 1af5e7f28342..0766cfe4a987 100644 --- a/lib/Target/PowerPC/PPCInstrInfo.td +++ b/lib/Target/PowerPC/PPCInstrInfo.td @@ -1223,9 +1223,15 @@ let isBranch = 1, isTerminator = 1, hasCtrlDep = 1, PPC970_Unit = 7 in { // FIXME: should be able to write a pattern for PPCcondbranch, but can't use // a two-value operand where a dag node expects two operands. :( let isCodeGenOnly = 1 in { - def BCC : BForm<16, 0, 0, (outs), (ins pred:$cond, condbrtarget:$dst), - "b${cond:cc}${cond:pm} ${cond:reg}, $dst" - /*[(PPCcondbranch crrc:$crS, imm:$opc, bb:$dst)]*/>; + class BCC_class : BForm<16, 0, 0, (outs), (ins pred:$cond, condbrtarget:$dst), + "b${cond:cc}${cond:pm} ${cond:reg}, $dst" + /*[(PPCcondbranch crrc:$crS, imm:$opc, bb:$dst)]*/>; + def BCC : BCC_class; + + // The same as BCC, except that it's not a terminator. Used for introducing + // control flow dependency without creating new blocks. + let isTerminator = 0 in def CTRL_DEP : BCC_class; + def BCCA : BForm<16, 1, 0, (outs), (ins pred:$cond, abscondbrtarget:$dst), "b${cond:cc}a${cond:pm} ${cond:reg}, $dst">; diff --git a/lib/Target/SystemZ/SystemZTargetTransformInfo.cpp b/lib/Target/SystemZ/SystemZTargetTransformInfo.cpp index f56b238f91e6..6a3dc6799c43 100644 --- a/lib/Target/SystemZ/SystemZTargetTransformInfo.cpp +++ b/lib/Target/SystemZ/SystemZTargetTransformInfo.cpp @@ -325,6 +325,30 @@ int SystemZTTIImpl::getArithmeticInstrCost( unsigned ScalarBits = Ty->getScalarSizeInBits(); + // Div with a constant which is a power of 2 will be converted by + // DAGCombiner to use shifts. With vector shift-element instructions, a + // vector sdiv costs about as much as a scalar one. + const unsigned SDivCostEstimate = 4; + bool SDivPow2 = false; + bool UDivPow2 = false; + if ((Opcode == Instruction::SDiv || Opcode == Instruction::UDiv) && + Args.size() == 2) { + const ConstantInt *CI = nullptr; + if (const Constant *C = dyn_cast(Args[1])) { + if (C->getType()->isVectorTy()) + CI = dyn_cast_or_null(C->getSplatValue()); + else + CI = dyn_cast(C); + } + if (CI != nullptr && + (CI->getValue().isPowerOf2() || (-CI->getValue()).isPowerOf2())) { + if (Opcode == Instruction::SDiv) + SDivPow2 = true; + else + UDivPow2 = true; + } + } + if (Ty->isVectorTy()) { assert (ST->hasVector() && "getArithmeticInstrCost() called with vector type."); unsigned VF = Ty->getVectorNumElements(); @@ -333,10 +357,13 @@ int SystemZTTIImpl::getArithmeticInstrCost( // These vector operations are custom handled, but are still supported // with one instruction per vector, regardless of element size. if (Opcode == Instruction::Shl || Opcode == Instruction::LShr || - Opcode == Instruction::AShr) { + Opcode == Instruction::AShr || UDivPow2) { return NumVectors; } + if (SDivPow2) + return (NumVectors * SDivCostEstimate); + // These FP operations are supported with a single vector instruction for // double (base implementation assumes float generally costs 2). For // FP128, the scalar cost is 1, and there is no overhead since the values @@ -395,6 +422,11 @@ int SystemZTTIImpl::getArithmeticInstrCost( // 2 * ipm sequences ; xor ; shift ; compare return 7; + if (UDivPow2) + return 1; + if (SDivPow2) + return SDivCostEstimate; + // An extra extension for narrow types is needed. if ((Opcode == Instruction::SDiv || Opcode == Instruction::SRem)) // sext of op(s) for narrow types diff --git a/lib/Target/WebAssembly/WebAssemblyRegColoring.cpp b/lib/Target/WebAssembly/WebAssemblyRegColoring.cpp index 5fd4a8d1949e..ba39b6cdb568 100644 --- a/lib/Target/WebAssembly/WebAssemblyRegColoring.cpp +++ b/lib/Target/WebAssembly/WebAssemblyRegColoring.cpp @@ -140,8 +140,7 @@ bool WebAssemblyRegColoring::runOnMachineFunction(MachineFunction &MF) { // Check if it's possible to reuse any of the used colors. if (!MRI->isLiveIn(Old)) - for (int C(UsedColors.find_first()); C != -1; - C = UsedColors.find_next(C)) { + for (unsigned C : UsedColors.set_bits()) { if (MRI->getRegClass(SortedIntervals[C]->reg) != RC) continue; for (LiveInterval *OtherLI : Assignments[C]) diff --git a/lib/Target/WebAssembly/known_gcc_test_failures.txt b/lib/Target/WebAssembly/known_gcc_test_failures.txt index 8e8e5fd1eff1..54619589c341 100644 --- a/lib/Target/WebAssembly/known_gcc_test_failures.txt +++ b/lib/Target/WebAssembly/known_gcc_test_failures.txt @@ -33,9 +33,6 @@ built-in-setjmp.c pr60003.c # Error in the program / unsupported by Clang. -scal-to-vec1.c -scal-to-vec2.c -scal-to-vec3.c 20000822-1.c 20010209-1.c 20010605-1.c diff --git a/lib/Target/X86/X86.td b/lib/Target/X86/X86.td index 3a421fe77392..784c3a6557ff 100644 --- a/lib/Target/X86/X86.td +++ b/lib/Target/X86/X86.td @@ -235,8 +235,6 @@ def FeatureLEAUsesAG : SubtargetFeature<"lea-uses-ag", "LEAUsesAG", "true", "LEA instruction needs inputs at AG stage">; def FeatureSlowLEA : SubtargetFeature<"slow-lea", "SlowLEA", "true", "LEA instruction with certain arguments is slow">; -def FeatureSlow3OpsLEA : SubtargetFeature<"slow-3ops-lea", "Slow3OpsLEA", "true", - "LEA instruction with 3 ops or certain registers is slow">; def FeatureSlowIncDec : SubtargetFeature<"slow-incdec", "SlowIncDec", "true", "INC and DEC instructions are slower than ADD and SUB">; def FeatureSoftFloat @@ -482,7 +480,6 @@ def SNBFeatures : ProcessorFeatures<[], [ FeatureXSAVE, FeatureXSAVEOPT, FeatureLAHFSAHF, - FeatureSlow3OpsLEA, FeatureFastScalarFSQRT, FeatureFastSHLDRotate ]>; diff --git a/lib/Target/X86/X86FixupLEAs.cpp b/lib/Target/X86/X86FixupLEAs.cpp index 9f649dad8bc0..2cd4c1a3e7b3 100644 --- a/lib/Target/X86/X86FixupLEAs.cpp +++ b/lib/Target/X86/X86FixupLEAs.cpp @@ -27,26 +27,20 @@ #include "llvm/Target/TargetInstrInfo.h" using namespace llvm; -namespace llvm { -void initializeFixupLEAPassPass(PassRegistry &); -} - -#define FIXUPLEA_DESC "X86 LEA Fixup" -#define FIXUPLEA_NAME "x86-fixup-LEAs" - -#define DEBUG_TYPE FIXUPLEA_NAME +#define DEBUG_TYPE "x86-fixup-LEAs" STATISTIC(NumLEAs, "Number of LEA instructions created"); namespace { class FixupLEAPass : public MachineFunctionPass { enum RegUsageState { RU_NotUsed, RU_Write, RU_Read }; - + static char ID; /// \brief Loop over all of the instructions in the basic block /// replacing applicable instructions with LEA instructions, /// where appropriate. bool processBasicBlock(MachineFunction &MF, MachineFunction::iterator MFI); + StringRef getPassName() const override { return "X86 LEA Fixup"; } /// \brief Given a machine register, look for the instruction /// which writes it in the current basic block. If found, @@ -68,22 +62,6 @@ class FixupLEAPass : public MachineFunctionPass { void processInstructionForSLM(MachineBasicBlock::iterator &I, MachineFunction::iterator MFI); - - /// \brief Given a LEA instruction which is unprofitable - /// on SNB+ try to replace it with other instructions. - /// According to Intel's Optimization Reference Manual: - /// " For LEA instructions with three source operands and some specific - /// situations, instruction latency has increased to 3 cycles, and must - /// dispatch via port 1: - /// - LEA that has all three source operands: base, index, and offset - /// - LEA that uses base and index registers where the base is EBP, RBP, - /// or R13 - /// - LEA that uses RIP relative addressing mode - /// - LEA that uses 16-bit addressing mode " - /// This function currently handles the first 2 cases only. - MachineInstr *processInstrForSlow3OpLEA(MachineInstr &MI, - MachineFunction::iterator MFI); - /// \brief Look for LEAs that add 1 to reg or subtract 1 from reg /// and convert them to INC or DEC respectively. bool fixupIncDec(MachineBasicBlock::iterator &I, @@ -107,13 +85,7 @@ class FixupLEAPass : public MachineFunctionPass { MachineBasicBlock::iterator &MBBI) const; public: - static char ID; - - StringRef getPassName() const override { return FIXUPLEA_DESC; } - - FixupLEAPass() : MachineFunctionPass(ID) { - initializeFixupLEAPassPass(*PassRegistry::getPassRegistry()); - } + FixupLEAPass() : MachineFunctionPass(ID) {} /// \brief Loop over all of the basic blocks, /// replacing instructions by equivalent LEA instructions @@ -132,11 +104,8 @@ class FixupLEAPass : public MachineFunctionPass { bool OptIncDec; bool OptLEA; }; -} - char FixupLEAPass::ID = 0; - -INITIALIZE_PASS(FixupLEAPass, FIXUPLEA_NAME, FIXUPLEA_DESC, false, false) +} MachineInstr * FixupLEAPass::postRAConvertToLEA(MachineFunction::iterator &MFI, @@ -199,7 +168,7 @@ bool FixupLEAPass::runOnMachineFunction(MachineFunction &Func) { MF = &Func; const X86Subtarget &ST = Func.getSubtarget(); OptIncDec = !ST.slowIncDec() || Func.getFunction()->optForMinSize(); - OptLEA = ST.LEAusesAG() || ST.slowLEA() || ST.slow3OpsLEA(); + OptLEA = ST.LEAusesAG() || ST.slowLEA(); if (!OptLEA && !OptIncDec) return false; @@ -273,64 +242,9 @@ FixupLEAPass::searchBackwards(MachineOperand &p, MachineBasicBlock::iterator &I, return MachineBasicBlock::iterator(); } -static inline bool isLEA(const int Opcode) { - return Opcode == X86::LEA16r || Opcode == X86::LEA32r || - Opcode == X86::LEA64r || Opcode == X86::LEA64_32r; -} - -static inline bool isInefficientLEAReg(unsigned int Reg) { - return Reg == X86::EBP || Reg == X86::RBP || Reg == X86::R13; -} - -static inline bool isRegOperand(const MachineOperand &Op) { - return Op.isReg() && Op.getReg() != X86::NoRegister; -} -/// hasIneffecientLEARegs - LEA that uses base and index registers -/// where the base is EBP, RBP, or R13 -static inline bool hasInefficientLEABaseReg(const MachineOperand &Base, - const MachineOperand &Index) { - return Base.isReg() && isInefficientLEAReg(Base.getReg()) && - isRegOperand(Index); -} - -static inline bool hasLEAOffset(const MachineOperand &Offset) { - return (Offset.isImm() && Offset.getImm() != 0) || Offset.isGlobal(); -} - -// LEA instruction that has all three operands: offset, base and index -static inline bool isThreeOperandsLEA(const MachineOperand &Base, - const MachineOperand &Index, - const MachineOperand &Offset) { - return isRegOperand(Base) && isRegOperand(Index) && hasLEAOffset(Offset); -} - -static inline int getADDrrFromLEA(int LEAOpcode) { - switch (LEAOpcode) { - default: - llvm_unreachable("Unexpected LEA instruction"); - case X86::LEA16r: - return X86::ADD16rr; - case X86::LEA32r: - return X86::ADD32rr; - case X86::LEA64_32r: - case X86::LEA64r: - return X86::ADD64rr; - } -} - -static inline int getADDriFromLEA(int LEAOpcode, const MachineOperand &Offset) { - bool IsInt8 = Offset.isImm() && isInt<8>(Offset.getImm()); - switch (LEAOpcode) { - default: - llvm_unreachable("Unexpected LEA instruction"); - case X86::LEA16r: - return IsInt8 ? X86::ADD16ri8 : X86::ADD16ri; - case X86::LEA32r: - case X86::LEA64_32r: - return IsInt8 ? X86::ADD32ri8 : X86::ADD32ri; - case X86::LEA64r: - return IsInt8 ? X86::ADD64ri8 : X86::ADD64ri32; - } +static inline bool isLEA(const int opcode) { + return opcode == X86::LEA16r || opcode == X86::LEA32r || + opcode == X86::LEA64r || opcode == X86::LEA64_32r; } /// isLEASimpleIncOrDec - Does this LEA have one these forms: @@ -423,8 +337,8 @@ void FixupLEAPass::seekLEAFixup(MachineOperand &p, void FixupLEAPass::processInstructionForSLM(MachineBasicBlock::iterator &I, MachineFunction::iterator MFI) { MachineInstr &MI = *I; - const int Opcode = MI.getOpcode(); - if (!isLEA(Opcode)) + const int opcode = MI.getOpcode(); + if (!isLEA(opcode)) return; if (MI.getOperand(5).getReg() != 0 || !MI.getOperand(4).isImm() || !TII->isSafeToClobberEFLAGS(*MFI, I)) @@ -436,144 +350,55 @@ void FixupLEAPass::processInstructionForSLM(MachineBasicBlock::iterator &I, return; if (MI.getOperand(2).getImm() > 1) return; + int addrr_opcode, addri_opcode; + switch (opcode) { + default: + llvm_unreachable("Unexpected LEA instruction"); + case X86::LEA16r: + addrr_opcode = X86::ADD16rr; + addri_opcode = X86::ADD16ri; + break; + case X86::LEA32r: + addrr_opcode = X86::ADD32rr; + addri_opcode = X86::ADD32ri; + break; + case X86::LEA64_32r: + case X86::LEA64r: + addrr_opcode = X86::ADD64rr; + addri_opcode = X86::ADD64ri32; + break; + } DEBUG(dbgs() << "FixLEA: Candidate to replace:"; I->dump();); DEBUG(dbgs() << "FixLEA: Replaced by: ";); MachineInstr *NewMI = nullptr; + const MachineOperand &Dst = MI.getOperand(0); // Make ADD instruction for two registers writing to LEA's destination if (SrcR1 != 0 && SrcR2 != 0) { - const MCInstrDesc &ADDrr = TII->get(getADDrrFromLEA(Opcode)); - const MachineOperand &Src = MI.getOperand(SrcR1 == DstR ? 3 : 1); - NewMI = - BuildMI(*MFI, I, MI.getDebugLoc(), ADDrr, DstR).addReg(DstR).add(Src); + const MachineOperand &Src1 = MI.getOperand(SrcR1 == DstR ? 1 : 3); + const MachineOperand &Src2 = MI.getOperand(SrcR1 == DstR ? 3 : 1); + NewMI = BuildMI(*MF, MI.getDebugLoc(), TII->get(addrr_opcode)) + .add(Dst) + .add(Src1) + .add(Src2); + MFI->insert(I, NewMI); DEBUG(NewMI->dump();); } // Make ADD instruction for immediate if (MI.getOperand(4).getImm() != 0) { - const MCInstrDesc &ADDri = - TII->get(getADDriFromLEA(Opcode, MI.getOperand(4))); const MachineOperand &SrcR = MI.getOperand(SrcR1 == DstR ? 1 : 3); - NewMI = BuildMI(*MFI, I, MI.getDebugLoc(), ADDri, DstR) + NewMI = BuildMI(*MF, MI.getDebugLoc(), TII->get(addri_opcode)) + .add(Dst) .add(SrcR) .addImm(MI.getOperand(4).getImm()); + MFI->insert(I, NewMI); DEBUG(NewMI->dump();); } if (NewMI) { MFI->erase(I); - I = NewMI; + I = static_cast(NewMI); } } -MachineInstr * -FixupLEAPass::processInstrForSlow3OpLEA(MachineInstr &MI, - MachineFunction::iterator MFI) { - - const int LEAOpcode = MI.getOpcode(); - if (!isLEA(LEAOpcode)) - return nullptr; - - const MachineOperand &Dst = MI.getOperand(0); - const MachineOperand &Base = MI.getOperand(1); - const MachineOperand &Scale = MI.getOperand(2); - const MachineOperand &Index = MI.getOperand(3); - const MachineOperand &Offset = MI.getOperand(4); - const MachineOperand &Segment = MI.getOperand(5); - - if (!(isThreeOperandsLEA(Base, Index, Offset) || - hasInefficientLEABaseReg(Base, Index)) || - !TII->isSafeToClobberEFLAGS(*MFI, MI) || - Segment.getReg() != X86::NoRegister) - return nullptr; - - unsigned int DstR = Dst.getReg(); - unsigned int BaseR = Base.getReg(); - unsigned int IndexR = Index.getReg(); - unsigned SSDstR = - (LEAOpcode == X86::LEA64_32r) ? getX86SubSuperRegister(DstR, 64) : DstR; - bool IsScale1 = Scale.getImm() == 1; - bool IsInefficientBase = isInefficientLEAReg(BaseR); - bool IsInefficientIndex = isInefficientLEAReg(IndexR); - - // Skip these cases since it takes more than 2 instructions - // to replace the LEA instruction. - if (IsInefficientBase && SSDstR == BaseR && !IsScale1) - return nullptr; - if (LEAOpcode == X86::LEA64_32r && IsInefficientBase && - (IsInefficientIndex || !IsScale1)) - return nullptr; - - const DebugLoc DL = MI.getDebugLoc(); - const MCInstrDesc &ADDrr = TII->get(getADDrrFromLEA(LEAOpcode)); - const MCInstrDesc &ADDri = TII->get(getADDriFromLEA(LEAOpcode, Offset)); - - DEBUG(dbgs() << "FixLEA: Candidate to replace:"; MI.dump();); - DEBUG(dbgs() << "FixLEA: Replaced by: ";); - - // First try to replace LEA with one or two (for the 3-op LEA case) - // add instructions: - // 1.lea (%base,%index,1), %base => add %index,%base - // 2.lea (%base,%index,1), %index => add %base,%index - if (IsScale1 && (DstR == BaseR || DstR == IndexR)) { - const MachineOperand &Src = DstR == BaseR ? Index : Base; - MachineInstr *NewMI = - BuildMI(*MFI, MI, DL, ADDrr, DstR).addReg(DstR).add(Src); - DEBUG(NewMI->dump();); - // Create ADD instruction for the Offset in case of 3-Ops LEA. - if (hasLEAOffset(Offset)) { - NewMI = BuildMI(*MFI, MI, DL, ADDri, DstR).addReg(DstR).add(Offset); - DEBUG(NewMI->dump();); - } - return NewMI; - } - // If the base is inefficient try switching the index and base operands, - // otherwise just break the 3-Ops LEA inst into 2-Ops LEA + ADD instruction: - // lea offset(%base,%index,scale),%dst => - // lea (%base,%index,scale); add offset,%dst - if (!IsInefficientBase || (!IsInefficientIndex && IsScale1)) { - MachineInstr *NewMI = BuildMI(*MFI, MI, DL, TII->get(LEAOpcode)) - .add(Dst) - .add(IsInefficientBase ? Index : Base) - .add(Scale) - .add(IsInefficientBase ? Base : Index) - .addImm(0) - .add(Segment); - DEBUG(NewMI->dump();); - // Create ADD instruction for the Offset in case of 3-Ops LEA. - if (hasLEAOffset(Offset)) { - NewMI = BuildMI(*MFI, MI, DL, ADDri, DstR).addReg(DstR).add(Offset); - DEBUG(NewMI->dump();); - } - return NewMI; - } - // Handle the rest of the cases with inefficient base register: - assert(SSDstR != BaseR && "SSDstR == BaseR should be handled already!"); - assert(IsInefficientBase && "efficient base should be handled already!"); - - // lea (%base,%index,1), %dst => mov %base,%dst; add %index,%dst - if (IsScale1 && !hasLEAOffset(Offset)) { - TII->copyPhysReg(*MFI, MI, DL, DstR, BaseR, Base.isKill()); - DEBUG(MI.getPrevNode()->dump();); - - MachineInstr *NewMI = - BuildMI(*MFI, MI, DL, ADDrr, DstR).addReg(DstR).add(Index); - DEBUG(NewMI->dump();); - return NewMI; - } - // lea offset(%base,%index,scale), %dst => - // lea offset( ,%index,scale), %dst; add %base,%dst - MachineInstr *NewMI = BuildMI(*MFI, MI, DL, TII->get(LEAOpcode)) - .add(Dst) - .addReg(0) - .add(Scale) - .add(Index) - .add(Offset) - .add(Segment); - DEBUG(NewMI->dump();); - - NewMI = BuildMI(*MFI, MI, DL, ADDrr, DstR).addReg(DstR).add(Base); - DEBUG(NewMI->dump();); - return NewMI; -} - bool FixupLEAPass::processBasicBlock(MachineFunction &MF, MachineFunction::iterator MFI) { @@ -585,16 +410,8 @@ bool FixupLEAPass::processBasicBlock(MachineFunction &MF, if (OptLEA) { if (MF.getSubtarget().isSLM()) processInstructionForSLM(I, MFI); - - else { - if (MF.getSubtarget().slow3OpsLEA()) { - if (auto *NewMI = processInstrForSlow3OpLEA(*I, MFI)) { - MFI->erase(I); - I = NewMI; - } - } else - processInstruction(I, MFI); - } + else + processInstruction(I, MFI); } } return false; diff --git a/lib/Target/X86/X86InstructionSelector.cpp b/lib/Target/X86/X86InstructionSelector.cpp index de58d719acb4..5eb5ad52840a 100644 --- a/lib/Target/X86/X86InstructionSelector.cpp +++ b/lib/Target/X86/X86InstructionSelector.cpp @@ -19,6 +19,7 @@ #include "X86Subtarget.h" #include "X86TargetMachine.h" #include "llvm/CodeGen/GlobalISel/InstructionSelector.h" +#include "llvm/CodeGen/GlobalISel/Utils.h" #include "llvm/CodeGen/MachineBasicBlock.h" #include "llvm/CodeGen/MachineFunction.h" #include "llvm/CodeGen/MachineInstr.h" @@ -72,6 +73,9 @@ class X86InstructionSelector : public InstructionSelector { bool selectCmp(MachineInstr &I, MachineRegisterInfo &MRI, MachineFunction &MF) const; + bool selectUadde(MachineInstr &I, MachineRegisterInfo &MRI, + MachineFunction &MF) const; + const X86TargetMachine &TM; const X86Subtarget &STI; const X86InstrInfo &TII; @@ -243,6 +247,8 @@ bool X86InstructionSelector::select(MachineInstr &I) const { return true; if (selectCmp(I, MRI, MF)) return true; + if (selectUadde(I, MRI, MF)) + return true; return false; } @@ -564,6 +570,66 @@ bool X86InstructionSelector::selectCmp(MachineInstr &I, return true; } +bool X86InstructionSelector::selectUadde(MachineInstr &I, + MachineRegisterInfo &MRI, + MachineFunction &MF) const { + if (I.getOpcode() != TargetOpcode::G_UADDE) + return false; + + const unsigned DstReg = I.getOperand(0).getReg(); + const unsigned CarryOutReg = I.getOperand(1).getReg(); + const unsigned Op0Reg = I.getOperand(2).getReg(); + const unsigned Op1Reg = I.getOperand(3).getReg(); + unsigned CarryInReg = I.getOperand(4).getReg(); + + const LLT DstTy = MRI.getType(DstReg); + + if (DstTy != LLT::scalar(32)) + return false; + + // find CarryIn def instruction. + MachineInstr *Def = MRI.getVRegDef(CarryInReg); + while (Def->getOpcode() == TargetOpcode::G_TRUNC) { + CarryInReg = Def->getOperand(1).getReg(); + Def = MRI.getVRegDef(CarryInReg); + } + + unsigned Opcode; + if (Def->getOpcode() == TargetOpcode::G_UADDE) { + // carry set by prev ADD. + + BuildMI(*I.getParent(), I, I.getDebugLoc(), TII.get(X86::COPY), X86::EFLAGS) + .addReg(CarryInReg); + + if (!RBI.constrainGenericRegister(CarryInReg, X86::GR32RegClass, MRI)) + return false; + + Opcode = X86::ADC32rr; + } else if (auto val = getConstantVRegVal(CarryInReg, MRI)) { + // carry is constant, support only 0. + if (*val != 0) + return false; + + Opcode = X86::ADD32rr; + } else + return false; + + MachineInstr &AddInst = + *BuildMI(*I.getParent(), I, I.getDebugLoc(), TII.get(Opcode), DstReg) + .addReg(Op0Reg) + .addReg(Op1Reg); + + BuildMI(*I.getParent(), I, I.getDebugLoc(), TII.get(X86::COPY), CarryOutReg) + .addReg(X86::EFLAGS); + + if (!constrainSelectedInstRegOperands(AddInst, TII, TRI, RBI) || + !RBI.constrainGenericRegister(CarryOutReg, X86::GR32RegClass, MRI)) + return false; + + I.eraseFromParent(); + return true; +} + InstructionSelector * llvm::createX86InstructionSelector(const X86TargetMachine &TM, X86Subtarget &Subtarget, diff --git a/lib/Target/X86/X86LegalizerInfo.cpp b/lib/Target/X86/X86LegalizerInfo.cpp index cf26238c0239..8ce240714f17 100644 --- a/lib/Target/X86/X86LegalizerInfo.cpp +++ b/lib/Target/X86/X86LegalizerInfo.cpp @@ -59,6 +59,11 @@ void X86LegalizerInfo::setLegalizerInfo32bit() { for (auto Ty : {s8, s16, s32}) setAction({BinOp, Ty}, Legal); + for (unsigned Op : {G_UADDE}) { + setAction({Op, s32}, Legal); + setAction({Op, 1, s1}, Legal); + } + for (unsigned MemOp : {G_LOAD, G_STORE}) { for (auto Ty : {s8, s16, s32, p0}) setAction({MemOp, Ty}, Legal); diff --git a/lib/Target/X86/X86Subtarget.h b/lib/Target/X86/X86Subtarget.h index 02be95e2e556..de1514243aeb 100644 --- a/lib/Target/X86/X86Subtarget.h +++ b/lib/Target/X86/X86Subtarget.h @@ -253,11 +253,6 @@ class X86Subtarget final : public X86GenSubtargetInfo { /// True if the LEA instruction with certain arguments is slow bool SlowLEA; - /// True if the LEA instruction has all three source operands: base, index, - /// and offset or if the LEA instruction uses base and index registers where - /// the base is EBP, RBP,or R13 - bool Slow3OpsLEA; - /// True if INC and DEC instructions are slow when writing to flags bool SlowIncDec; @@ -495,7 +490,6 @@ class X86Subtarget final : public X86GenSubtargetInfo { bool callRegIndirect() const { return CallRegIndirect; } bool LEAusesAG() const { return LEAUsesAG; } bool slowLEA() const { return SlowLEA; } - bool slow3OpsLEA() const { return Slow3OpsLEA; } bool slowIncDec() const { return SlowIncDec; } bool hasCDI() const { return HasCDI; } bool hasPFI() const { return HasPFI; } diff --git a/lib/Target/X86/X86TargetMachine.cpp b/lib/Target/X86/X86TargetMachine.cpp index c6a90725d89c..9a82e6e50463 100644 --- a/lib/Target/X86/X86TargetMachine.cpp +++ b/lib/Target/X86/X86TargetMachine.cpp @@ -61,7 +61,6 @@ static cl::opt EnableMachineCombinerPass("x86-machine-combiner", namespace llvm { void initializeWinEHStatePassPass(PassRegistry &); -void initializeFixupLEAPassPass(PassRegistry &); void initializeX86ExecutionDepsFixPass(PassRegistry &); } // end namespace llvm @@ -76,7 +75,6 @@ extern "C" void LLVMInitializeX86Target() { initializeWinEHStatePassPass(PR); initializeFixupBWInstPassPass(PR); initializeEvexToVexInstPassPass(PR); - initializeFixupLEAPassPass(PR); initializeX86ExecutionDepsFixPass(PR); } diff --git a/lib/Target/X86/X86TargetTransformInfo.cpp b/lib/Target/X86/X86TargetTransformInfo.cpp index 80e18161a94b..8566bd91c89e 100644 --- a/lib/Target/X86/X86TargetTransformInfo.cpp +++ b/lib/Target/X86/X86TargetTransformInfo.cpp @@ -1392,6 +1392,16 @@ int X86TTIImpl::getIntrinsicInstrCost(Intrinsic::ID IID, Type *RetTy, // CTLZ: llvm\test\CodeGen\X86\vector-lzcnt-*.ll // CTPOP: llvm\test\CodeGen\X86\vector-popcnt-*.ll // CTTZ: llvm\test\CodeGen\X86\vector-tzcnt-*.ll + static const CostTblEntry AVX512BWCostTbl[] = { + { ISD::BITREVERSE, MVT::v8i64, 5 }, + { ISD::BITREVERSE, MVT::v16i32, 5 }, + { ISD::BITREVERSE, MVT::v32i16, 5 }, + { ISD::BITREVERSE, MVT::v64i8, 5 }, + }; + static const CostTblEntry AVX512CostTbl[] = { + { ISD::BITREVERSE, MVT::v8i64, 36 }, + { ISD::BITREVERSE, MVT::v16i32, 24 }, + }; static const CostTblEntry XOPCostTbl[] = { { ISD::BITREVERSE, MVT::v4i64, 4 }, { ISD::BITREVERSE, MVT::v8i32, 4 }, @@ -1550,6 +1560,14 @@ int X86TTIImpl::getIntrinsicInstrCost(Intrinsic::ID IID, Type *RetTy, MVT MTy = LT.second; // Attempt to lookup cost. + if (ST->hasBWI()) + if (const auto *Entry = CostTableLookup(AVX512BWCostTbl, ISD, MTy)) + return LT.first * Entry->Cost; + + if (ST->hasAVX512()) + if (const auto *Entry = CostTableLookup(AVX512CostTbl, ISD, MTy)) + return LT.first * Entry->Cost; + if (ST->hasXOP()) if (const auto *Entry = CostTableLookup(XOPCostTbl, ISD, MTy)) return LT.first * Entry->Cost; diff --git a/lib/Transforms/Coroutines/CoroFrame.cpp b/lib/Transforms/Coroutines/CoroFrame.cpp index 4480220f2cd4..417d57f7625b 100644 --- a/lib/Transforms/Coroutines/CoroFrame.cpp +++ b/lib/Transforms/Coroutines/CoroFrame.cpp @@ -347,6 +347,27 @@ static StructType *buildFrameType(Function &F, coro::Shape &Shape, return FrameTy; } +// We need to make room to insert a spill after initial PHIs, but before +// catchswitch instruction. Placing it before violates the requirement that +// catchswitch, like all other EHPads must be the first nonPHI in a block. +// +// Split away catchswitch into a separate block and insert in its place: +// +// cleanuppad cleanupret. +// +// cleanupret instruction will act as an insert point for the spill. +static Instruction *splitBeforeCatchSwitch(CatchSwitchInst *CatchSwitch) { + BasicBlock *CurrentBlock = CatchSwitch->getParent(); + BasicBlock *NewBlock = CurrentBlock->splitBasicBlock(CatchSwitch); + CurrentBlock->getTerminator()->eraseFromParent(); + + auto *CleanupPad = + CleanupPadInst::Create(CatchSwitch->getParentPad(), {}, "", CurrentBlock); + auto *CleanupRet = + CleanupReturnInst::Create(CleanupPad, NewBlock, CurrentBlock); + return CleanupRet; +} + // Replace all alloca and SSA values that are accessed across suspend points // with GetElementPointer from coroutine frame + loads and stores. Create an // AllocaSpillBB that will become the new entry block for the resume parts of @@ -437,8 +458,11 @@ static Instruction *insertSpills(SpillInfo &Spills, coro::Shape &Shape) { InsertPt = NewBB->getTerminator(); } else if (dyn_cast(CurrentValue)) { // Skip the PHINodes and EH pads instructions. - InsertPt = - &*cast(E.def())->getParent()->getFirstInsertionPt(); + BasicBlock *DefBlock = cast(E.def())->getParent(); + if (auto *CSI = dyn_cast(DefBlock->getTerminator())) + InsertPt = splitBeforeCatchSwitch(CSI); + else + InsertPt = &*DefBlock->getFirstInsertionPt(); } else { // For all other values, the spill is placed immediately after // the definition. diff --git a/lib/Transforms/InstCombine/InstCombineInternal.h b/lib/Transforms/InstCombine/InstCombineInternal.h index 1424f61fe701..f88a2c6acc3f 100644 --- a/lib/Transforms/InstCombine/InstCombineInternal.h +++ b/lib/Transforms/InstCombine/InstCombineInternal.h @@ -74,6 +74,27 @@ static inline unsigned getComplexity(Value *V) { return isa(V) ? (isa(V) ? 0 : 1) : 2; } +/// Predicate canonicalization reduces the number of patterns that need to be +/// matched by other transforms. For example, we may swap the operands of a +/// conditional branch or select to create a compare with a canonical (inverted) +/// predicate which is then more likely to be matched with other values. +static inline bool isCanonicalPredicate(CmpInst::Predicate Pred) { + switch (Pred) { + case CmpInst::ICMP_NE: + case CmpInst::ICMP_ULE: + case CmpInst::ICMP_SLE: + case CmpInst::ICMP_UGE: + case CmpInst::ICMP_SGE: + // TODO: There are 16 FCMP predicates. Should others be (not) canonical? + case CmpInst::FCMP_ONE: + case CmpInst::FCMP_OLE: + case CmpInst::FCMP_OGE: + return false; + default: + return true; + } +} + /// \brief Add one to a Constant static inline Constant *AddOne(Constant *C) { return ConstantExpr::getAdd(C, ConstantInt::get(C->getType(), 1)); diff --git a/lib/Transforms/InstCombine/InstructionCombining.cpp b/lib/Transforms/InstCombine/InstructionCombining.cpp index 65b1148cb03b..7ed9fd566b37 100644 --- a/lib/Transforms/InstCombine/InstructionCombining.cpp +++ b/lib/Transforms/InstCombine/InstructionCombining.cpp @@ -2210,37 +2210,17 @@ Instruction *InstCombiner::visitBranchInst(BranchInst &BI) { return &BI; } - // Canonicalize fcmp_one -> fcmp_oeq - FCmpInst::Predicate FPred; Value *Y; - if (match(&BI, m_Br(m_OneUse(m_FCmp(FPred, m_Value(X), m_Value(Y))), - TrueDest, FalseDest))) { - // TODO: Why are we only transforming these 3 predicates? - if (FPred == FCmpInst::FCMP_ONE || FPred == FCmpInst::FCMP_OLE || - FPred == FCmpInst::FCMP_OGE) { - FCmpInst *Cond = cast(BI.getCondition()); - Cond->setPredicate(FCmpInst::getInversePredicate(FPred)); - - // Swap Destinations and condition. - BI.swapSuccessors(); - Worklist.Add(Cond); - return &BI; - } - } - - // Canonicalize icmp_ne -> icmp_eq - ICmpInst::Predicate IPred; - if (match(&BI, m_Br(m_OneUse(m_ICmp(IPred, m_Value(X), m_Value(Y))), - TrueDest, FalseDest))) { - if (IPred == ICmpInst::ICMP_NE || IPred == ICmpInst::ICMP_ULE || - IPred == ICmpInst::ICMP_SLE || IPred == ICmpInst::ICMP_UGE || - IPred == ICmpInst::ICMP_SGE) { - ICmpInst *Cond = cast(BI.getCondition()); - Cond->setPredicate(ICmpInst::getInversePredicate(IPred)); - // Swap Destinations and condition. - BI.swapSuccessors(); - Worklist.Add(Cond); - return &BI; - } + // Canonicalize, for example, icmp_ne -> icmp_eq or fcmp_one -> fcmp_oeq. + CmpInst::Predicate Pred; + if (match(&BI, m_Br(m_OneUse(m_Cmp(Pred, m_Value(), m_Value())), TrueDest, + FalseDest)) && + !isCanonicalPredicate(Pred)) { + // Swap destinations and condition. + CmpInst *Cond = cast(BI.getCondition()); + Cond->setPredicate(CmpInst::getInversePredicate(Pred)); + BI.swapSuccessors(); + Worklist.Add(Cond); + return &BI; } return nullptr; @@ -3053,7 +3033,10 @@ static bool AddReachableCodeToWorklist(BasicBlock *BB, const DataLayout &DL, } } - InstrsForInstCombineWorklist.push_back(Inst); + // Skip processing debug intrinsics in InstCombine. Processing these call instructions + // consumes non-trivial amount of time and provides no value for the optimization. + if (!isa(Inst)) + InstrsForInstCombineWorklist.push_back(Inst); } // Recursively visit successors. If this is a branch or switch on a diff --git a/lib/Transforms/Scalar/LICM.cpp b/lib/Transforms/Scalar/LICM.cpp index 340c81fed0fd..37b9c4b1094e 100644 --- a/lib/Transforms/Scalar/LICM.cpp +++ b/lib/Transforms/Scalar/LICM.cpp @@ -546,7 +546,7 @@ static bool isLoadInvariantInLoop(LoadInst *LI, DominatorTree *DT, // If there are escaping uses of invariant.start instruction, the load maybe // non-invariant. if (!II || II->getIntrinsicID() != Intrinsic::invariant_start || - II->hasNUsesOrMore(1)) + !II->use_empty()) continue; unsigned InvariantSizeInBits = cast(II->getArgOperand(0))->getSExtValue() * 8; diff --git a/lib/Transforms/Scalar/LoopIdiomRecognize.cpp b/lib/Transforms/Scalar/LoopIdiomRecognize.cpp index 6693a26e8890..cb6223b070a6 100644 --- a/lib/Transforms/Scalar/LoopIdiomRecognize.cpp +++ b/lib/Transforms/Scalar/LoopIdiomRecognize.cpp @@ -1292,13 +1292,15 @@ bool LoopIdiomRecognize::recognizeAndInsertCTLZ() { BasicBlock *PH = CurLoop->getLoopPreheader(); Value *InitX = PhiX->getIncomingValueForBlock(PH); // If we check X != 0 before entering the loop we don't need a zero - // check in CTLZ intrinsic. - if (BasicBlock *PreCondBB = PH->getSinglePredecessor()) - if (BranchInst *PreCondBr = - dyn_cast(PreCondBB->getTerminator())) { - if (matchCondition(PreCondBr, PH) == InitX) - ZeroCheck = true; - } + // check in CTLZ intrinsic, but only if Cnt Phi is not used outside of the + // loop (if it is used we count CTLZ(X >> 1)). + if (!IsCntPhiUsedOutsideLoop) + if (BasicBlock *PreCondBB = PH->getSinglePredecessor()) + if (BranchInst *PreCondBr = + dyn_cast(PreCondBB->getTerminator())) { + if (matchCondition(PreCondBr, PH) == InitX) + ZeroCheck = true; + } // Check if CTLZ intrinsic is profitable. Assume it is always profitable // if we delete the loop (the loop has only 6 instructions): diff --git a/lib/Transforms/Scalar/LoopStrengthReduce.cpp b/lib/Transforms/Scalar/LoopStrengthReduce.cpp index ccedb98d7fa1..bd1f21c69eba 100644 --- a/lib/Transforms/Scalar/LoopStrengthReduce.cpp +++ b/lib/Transforms/Scalar/LoopStrengthReduce.cpp @@ -3902,8 +3902,7 @@ void LSRInstance::GenerateCrossUseConstantOffsets() { // Compute the difference between the two. int64_t Imm = (uint64_t)JImm - M->first; - for (int LUIdx = UsedByIndices.find_first(); LUIdx != -1; - LUIdx = UsedByIndices.find_next(LUIdx)) + for (unsigned LUIdx : UsedByIndices.set_bits()) // Make a memo of this use, offset, and register tuple. if (UniqueItems.insert(std::make_pair(LUIdx, Imm)).second) WorkItems.push_back(WorkItem(LUIdx, Imm, OrigReg)); diff --git a/lib/Transforms/Scalar/NewGVN.cpp b/lib/Transforms/Scalar/NewGVN.cpp index 5e0a705782ea..0e7572f8d2e5 100644 --- a/lib/Transforms/Scalar/NewGVN.cpp +++ b/lib/Transforms/Scalar/NewGVN.cpp @@ -642,6 +642,7 @@ class NewGVN { void updateProcessedCount(Value *V); void verifyMemoryCongruency() const; void verifyIterationSettled(Function &F); + void verifyStoreExpressions() const; bool singleReachablePHIPath(const MemoryAccess *, const MemoryAccess *) const; BasicBlock *getBlockForValue(Value *V) const; void deleteExpression(const Expression *E) const; @@ -2003,7 +2004,6 @@ void NewGVN::moveValueToNewCongruenceClass(Instruction *I, const Expression *E, // If it's not a memory use, set the MemoryAccess equivalence auto *InstMA = dyn_cast_or_null(MSSA->getMemoryAccess(I)); - bool InstWasMemoryLeader = InstMA && OldClass->getMemoryLeader() == InstMA; if (InstMA) moveMemoryToNewCongruenceClass(I, InstMA, OldClass, NewClass); ValueToClass[I] = NewClass; @@ -2029,31 +2029,6 @@ void NewGVN::moveValueToNewCongruenceClass(Instruction *I, const Expression *E, if (OldClass->getStoredValue()) OldClass->setStoredValue(nullptr); } - // If we destroy the old access leader and it's a store, we have to - // effectively destroy the congruence class. When it comes to scalars, - // anything with the same value is as good as any other. That means that - // one leader is as good as another, and as long as you have some leader for - // the value, you are good.. When it comes to *memory states*, only one - // particular thing really represents the definition of a given memory - // state. Once it goes away, we need to re-evaluate which pieces of memory - // are really still equivalent. The best way to do this is to re-value - // number things. The only way to really make that happen is to destroy the - // rest of the class. In order to effectively destroy the class, we reset - // ExpressionToClass for each by using the ValueToExpression mapping. The - // members later get marked as touched due to the leader change. We will - // create new congruence classes, and the pieces that are still equivalent - // will end back together in a new class. If this becomes too expensive, it - // is possible to use a versioning scheme for the congruence classes to - // avoid the expressions finding this old class. Note that the situation is - // different for memory phis, becuase they are evaluated anew each time, and - // they become equal not by hashing, but by seeing if all operands are the - // same (or only one is reachable). - if (OldClass->getStoreCount() > 0 && InstWasMemoryLeader) { - DEBUG(dbgs() << "Kicking everything out of class " << OldClass->getID() - << " because MemoryAccess leader changed"); - for (auto Member : *OldClass) - ExpressionToClass.erase(ValueToExpression.lookup(Member)); - } OldClass->setLeader(getNextValueLeader(OldClass)); OldClass->resetNextLeader(); markValueLeaderChangeTouched(OldClass); @@ -2062,7 +2037,6 @@ void NewGVN::moveValueToNewCongruenceClass(Instruction *I, const Expression *E, // Perform congruence finding on a given value numbering expression. void NewGVN::performCongruenceFinding(Instruction *I, const Expression *E) { - ValueToExpression[I] = E; // This is guaranteed to return something, since it will at least find // TOP. @@ -2132,6 +2106,18 @@ void NewGVN::performCongruenceFinding(Instruction *I, const Expression *E) { if (auto *CI = dyn_cast(I)) markPredicateUsersTouched(CI); } + // If we changed the class of the store, we want to ensure nothing finds the + // old store expression. In particular, loads do not compare against stored + // value, so they will find old store expressions (and associated class + // mappings) if we leave them in the table. + if (ClassChanged && isa(E)) { + auto *OldE = ValueToExpression.lookup(I); + // It could just be that the old class died. We don't want to erase it if we + // just moved classes. + if (OldE && isa(OldE) && !OldE->equals(*E)) + ExpressionToClass.erase(OldE); + } + ValueToExpression[I] = E; } // Process the fact that Edge (from, to) is reachable, including marking @@ -2651,6 +2637,30 @@ void NewGVN::verifyIterationSettled(Function &F) { #endif } +// Verify that for each store expression in the expression to class mapping, +// only the latest appears, and multiple ones do not appear. +// Because loads do not use the stored value when doing equality with stores, +// if we don't erase the old store expressions from the table, a load can find +// a no-longer valid StoreExpression. +void NewGVN::verifyStoreExpressions() const { +#ifndef NDEBUG + DenseSet> StoreExpressionSet; + for (const auto &KV : ExpressionToClass) { + if (auto *SE = dyn_cast(KV.first)) { + // Make sure a version that will conflict with loads is not already there + auto Res = + StoreExpressionSet.insert({SE->getOperand(0), SE->getMemoryLeader()}); + assert(Res.second && + "Stored expression conflict exists in expression table"); + auto *ValueExpr = ValueToExpression.lookup(SE->getStoreInst()); + assert(ValueExpr && ValueExpr->equals(*SE) && + "StoreExpression in ExpressionToClass is not latest " + "StoreExpression for value"); + } + } +#endif +} + // This is the main value numbering loop, it iterates over the initial touched // instruction set, propagating value numbers, marking things touched, etc, // until the set of touched instructions is completely empty. @@ -2668,8 +2678,7 @@ void NewGVN::iterateTouchedInstructions() { // TODO: As we hit a new block, we should push and pop equalities into a // table lookupOperandLeader can use, to catch things PredicateInfo // might miss, like edge-only equivalences. - for (int InstrNum = TouchedInstructions.find_first(); InstrNum != -1; - InstrNum = TouchedInstructions.find_next(InstrNum)) { + for (unsigned InstrNum : TouchedInstructions.set_bits()) { // This instruction was found to be dead. We don't bother looking // at it again. @@ -2776,6 +2785,7 @@ bool NewGVN::runGVN() { iterateTouchedInstructions(); verifyMemoryCongruency(); verifyIterationSettled(F); + verifyStoreExpressions(); Changed |= eliminateInstructions(F); diff --git a/lib/Transforms/Scalar/Reassociate.cpp b/lib/Transforms/Scalar/Reassociate.cpp index ef29d4141600..53320bff0883 100644 --- a/lib/Transforms/Scalar/Reassociate.cpp +++ b/lib/Transforms/Scalar/Reassociate.cpp @@ -1922,7 +1922,7 @@ Instruction *ReassociatePass::canonicalizeNegConstExpr(Instruction *I) { // User must be a binary operator with one or more uses. Instruction *User = I->user_back(); - if (!isa(User) || !User->hasNUsesOrMore(1)) + if (!isa(User) || User->use_empty()) return nullptr; unsigned UserOpcode = User->getOpcode(); diff --git a/lib/Transforms/Scalar/SimpleLoopUnswitch.cpp b/lib/Transforms/Scalar/SimpleLoopUnswitch.cpp index 4f608c97147d..b32a61a7e8f8 100644 --- a/lib/Transforms/Scalar/SimpleLoopUnswitch.cpp +++ b/lib/Transforms/Scalar/SimpleLoopUnswitch.cpp @@ -1,4 +1,4 @@ -//===-- SimpleLoopUnswitch.cpp - Hoist loop-invariant control flow --------===// +//===- SimpleLoopUnswitch.cpp - Hoist loop-invariant control flow ---------===// // // The LLVM Compiler Infrastructure // @@ -7,25 +7,41 @@ // //===----------------------------------------------------------------------===// -#include "llvm/Transforms/Scalar/SimpleLoopUnswitch.h" -#include "llvm/ADT/STLExtras.h" +#include "llvm/ADT/DenseMap.h" +#include "llvm/ADT/Sequence.h" +#include "llvm/ADT/SetVector.h" #include "llvm/ADT/SmallPtrSet.h" +#include "llvm/ADT/SmallVector.h" #include "llvm/ADT/Statistic.h" +#include "llvm/ADT/STLExtras.h" +#include "llvm/ADT/Twine.h" #include "llvm/Analysis/AssumptionCache.h" +#include "llvm/Analysis/LoopAnalysisManager.h" #include "llvm/Analysis/LoopInfo.h" #include "llvm/Analysis/LoopPass.h" +#include "llvm/IR/BasicBlock.h" +#include "llvm/IR/Constant.h" #include "llvm/IR/Constants.h" #include "llvm/IR/Dominators.h" #include "llvm/IR/Function.h" +#include "llvm/IR/InstrTypes.h" +#include "llvm/IR/Instruction.h" #include "llvm/IR/Instructions.h" -#include "llvm/Support/CommandLine.h" +#include "llvm/IR/Use.h" +#include "llvm/IR/Value.h" +#include "llvm/Pass.h" +#include "llvm/Support/Casting.h" #include "llvm/Support/Debug.h" +#include "llvm/Support/ErrorHandling.h" +#include "llvm/Support/GenericDomTree.h" #include "llvm/Support/raw_ostream.h" #include "llvm/Transforms/Utils/BasicBlockUtils.h" -#include "llvm/Transforms/Utils/Cloning.h" -#include "llvm/Transforms/Utils/Local.h" -#include "llvm/Transforms/Scalar/LoopPassManager.h" #include "llvm/Transforms/Utils/LoopUtils.h" +#include "llvm/Transforms/Scalar/SimpleLoopUnswitch.h" +#include +#include +#include +#include #define DEBUG_TYPE "simple-loop-unswitch" @@ -174,7 +190,7 @@ static void rewritePHINodesForUnswitchedExitBlock(BasicBlock &UnswitchedBB, // When the loop exit is directly unswitched we just need to update the // incoming basic block. We loop to handle weird cases with repeated // incoming blocks, but expect to typically only have one operand here. - for (auto i : llvm::seq(0, PN->getNumOperands())) { + for (auto i : seq(0, PN->getNumOperands())) { assert(PN->getIncomingBlock(i) == &OldExitingBB && "Found incoming block different from unique predecessor!"); PN->setIncomingBlock(i, &OldPH); @@ -688,9 +704,11 @@ PreservedAnalyses SimpleLoopUnswitchPass::run(Loop &L, LoopAnalysisManager &AM, } namespace { + class SimpleLoopUnswitchLegacyPass : public LoopPass { public: static char ID; // Pass ID, replacement for typeid + explicit SimpleLoopUnswitchLegacyPass() : LoopPass(ID) { initializeSimpleLoopUnswitchLegacyPassPass( *PassRegistry::getPassRegistry()); @@ -703,7 +721,8 @@ class SimpleLoopUnswitchLegacyPass : public LoopPass { getLoopAnalysisUsage(AU); } }; -} // namespace + +} // end anonymous namespace bool SimpleLoopUnswitchLegacyPass::runOnLoop(Loop *L, LPPassManager &LPM) { if (skipLoop(L)) diff --git a/test/Analysis/CostModel/SystemZ/div-pow2.ll b/test/Analysis/CostModel/SystemZ/div-pow2.ll new file mode 100644 index 000000000000..9ef2dd71e8fa --- /dev/null +++ b/test/Analysis/CostModel/SystemZ/div-pow2.ll @@ -0,0 +1,154 @@ +; RUN: opt < %s -cost-model -analyze -mtriple=systemz-unknown -mcpu=z13 | FileCheck %s + +; Scalar sdiv + +define i64 @fun0(i64 %a) { + %r = sdiv i64 %a, 2 + ret i64 %r +; CHECK: Cost Model: Found an estimated cost of 4 for instruction: %r = sdiv i64 %a, 2 +} + +define i64 @fun1(i64 %a) { + %r = sdiv i64 %a, -4 + ret i64 %r +; CHECK: Cost Model: Found an estimated cost of 4 for instruction: %r = sdiv i64 %a, -4 +} + +define i32 @fun2(i32 %a) { + %r = sdiv i32 %a, 8 + ret i32 %r +; CHECK: Cost Model: Found an estimated cost of 4 for instruction: %r = sdiv i32 %a, 8 +} + +define i32 @fun3(i32 %a) { + %r = sdiv i32 %a, -16 + ret i32 %r +; CHECK: Cost Model: Found an estimated cost of 4 for instruction: %r = sdiv i32 %a, -16 +} + +define i16 @fun4(i16 %a) { + %r = sdiv i16 %a, 32 + ret i16 %r +; CHECK: Cost Model: Found an estimated cost of 4 for instruction: %r = sdiv i16 %a, 32 +} + +define i16 @fun5(i16 %a) { + %r = sdiv i16 %a, -64 + ret i16 %r +; CHECK: Cost Model: Found an estimated cost of 4 for instruction: %r = sdiv i16 %a, -64 +} + +define i8 @fun6(i8 %a) { + %r = sdiv i8 %a, 64 + ret i8 %r +; CHECK: Cost Model: Found an estimated cost of 4 for instruction: %r = sdiv i8 %a, 64 +} + +define i8 @fun7(i8 %a) { + %r = sdiv i8 %a, -128 + ret i8 %r +; CHECK: Cost Model: Found an estimated cost of 4 for instruction: %r = sdiv i8 %a, -128 +} + + +; Vector sdiv + +define <2 x i64> @fun8(<2 x i64> %a) { + %r = sdiv <2 x i64> %a, + ret <2 x i64> %r +; CHECK: Cost Model: Found an estimated cost of 4 for instruction: %r = sdiv <2 x i64> %a, +} + +define <2 x i64> @fun9(<2 x i64> %a) { + %r = sdiv <2 x i64> %a, + ret <2 x i64> %r +; CHECK: Cost Model: Found an estimated cost of 4 for instruction: %r = sdiv <2 x i64> %a, +} + +define <4 x i32> @fun10(<4 x i32> %a) { + %r = sdiv <4 x i32> %a, + ret <4 x i32> %r +; CHECK: Cost Model: Found an estimated cost of 4 for instruction: %r = sdiv <4 x i32> %a, +} + +define <4 x i32> @fun11(<4 x i32> %a) { + %r = sdiv <4 x i32> %a, + ret <4 x i32> %r +; CHECK: Cost Model: Found an estimated cost of 4 for instruction: %r = sdiv <4 x i32> %a, @fun12(<8 x i16> %a) { + %r = sdiv <8 x i16> %a, + ret <8 x i16> %r +; CHECK: Cost Model: Found an estimated cost of 4 for instruction: %r = sdiv <8 x i16> %a, @fun13(<8 x i16> %a) { + %r = sdiv <8 x i16> %a, + ret <8 x i16> %r +; CHECK: Cost Model: Found an estimated cost of 4 for instruction: %r = sdiv <8 x i16> %a, @fun14(<16 x i8> %a) { + %r = sdiv <16 x i8> %a, + ret <16 x i8> %r +; CHECK: Cost Model: Found an estimated cost of 4 for instruction: %r = sdiv <16 x i8> %a, @fun15(<16 x i8> %a) { + %r = sdiv <16 x i8> %a, + ret <16 x i8> %r +; CHECK: Cost Model: Found an estimated cost of 4 for instruction: %r = sdiv <16 x i8> %a, @fun20(<2 x i64> %a) { + %r = udiv <2 x i64> %a, + ret <2 x i64> %r +; CHECK: Cost Model: Found an estimated cost of 1 for instruction: %r = udiv <2 x i64> %a, @fun21(<4 x i32> %a) { + %r = udiv <4 x i32> %a, + ret <4 x i32> %r +; CHECK: Cost Model: Found an estimated cost of 1 for instruction: %r = udiv <4 x i32> %a, @fun22(<8 x i16> %a) { + %r = udiv <8 x i16> %a, + ret <8 x i16> %r +; CHECK: Cost Model: Found an estimated cost of 1 for instruction: %r = udiv <8 x i16> %a, @fun23(<16 x i8> %a) { + %r = udiv <16 x i8> %a, + ret <16 x i8> %r +; CHECK: Cost Model: Found an estimated cost of 1 for instruction: %r = udiv <16 x i8> %a, @llvm.bitreverse.v8i32(<8 x i32>) declare <16 x i16> @llvm.bitreverse.v16i16(<16 x i16>) declare <32 x i8> @llvm.bitreverse.v32i8(<32 x i8>) +declare <8 x i64> @llvm.bitreverse.v8i64(<8 x i64>) +declare <16 x i32> @llvm.bitreverse.v16i32(<16 x i32>) +declare <32 x i16> @llvm.bitreverse.v32i16(<32 x i16>) +declare <64 x i8> @llvm.bitreverse.v64i8(<64 x i8>) + define <2 x i64> @var_bitreverse_v2i64(<2 x i64> %a) { ; CHECK: 'Cost Model Analysis' for function 'var_bitreverse_v2i64': ; SSE2: Found an estimated cost of 29 for instruction: %bitreverse ; SSE42: Found an estimated cost of 5 for instruction: %bitreverse ; AVX: Found an estimated cost of 5 for instruction: %bitreverse ; AVX2: Found an estimated cost of 5 for instruction: %bitreverse +; AVX512: Found an estimated cost of 5 for instruction: %bitreverse ; XOP: Found an estimated cost of 1 for instruction: %bitreverse %bitreverse = call <2 x i64> @llvm.bitreverse.v2i64(<2 x i64> %a) ret <2 x i64> %bitreverse @@ -81,17 +91,32 @@ define <4 x i64> @var_bitreverse_v4i64(<4 x i64> %a) { ; SSE42: Found an estimated cost of 10 for instruction: %bitreverse ; AVX: Found an estimated cost of 12 for instruction: %bitreverse ; AVX2: Found an estimated cost of 5 for instruction: %bitreverse +; AVX512: Found an estimated cost of 5 for instruction: %bitreverse ; XOP: Found an estimated cost of 4 for instruction: %bitreverse %bitreverse = call <4 x i64> @llvm.bitreverse.v4i64(<4 x i64> %a) ret <4 x i64> %bitreverse } +define <8 x i64> @var_bitreverse_v8i64(<8 x i64> %a) { +; CHECK: 'Cost Model Analysis' for function 'var_bitreverse_v8i64': +; SSE2: Found an estimated cost of 116 for instruction: %bitreverse +; SSE42: Found an estimated cost of 20 for instruction: %bitreverse +; AVX: Found an estimated cost of 24 for instruction: %bitreverse +; AVX2: Found an estimated cost of 10 for instruction: %bitreverse +; AVX512F: Found an estimated cost of 36 for instruction: %bitreverse +; AVX512BW: Found an estimated cost of 5 for instruction: %bitreverse +; XOP: Found an estimated cost of 8 for instruction: %bitreverse + %bitreverse = call <8 x i64> @llvm.bitreverse.v8i64(<8 x i64> %a) + ret <8 x i64> %bitreverse +} + define <4 x i32> @var_bitreverse_v4i32(<4 x i32> %a) { ; CHECK: 'Cost Model Analysis' for function 'var_bitreverse_v4i32': ; SSE2: Found an estimated cost of 27 for instruction: %bitreverse ; SSE42: Found an estimated cost of 5 for instruction: %bitreverse ; AVX: Found an estimated cost of 5 for instruction: %bitreverse ; AVX2: Found an estimated cost of 5 for instruction: %bitreverse +; AVX512: Found an estimated cost of 5 for instruction: %bitreverse ; XOP: Found an estimated cost of 1 for instruction: %bitreverse %bitreverse = call <4 x i32> @llvm.bitreverse.v4i32(<4 x i32> %a) ret <4 x i32> %bitreverse @@ -103,17 +128,32 @@ define <8 x i32> @var_bitreverse_v8i32(<8 x i32> %a) { ; SSE42: Found an estimated cost of 10 for instruction: %bitreverse ; AVX: Found an estimated cost of 12 for instruction: %bitreverse ; AVX2: Found an estimated cost of 5 for instruction: %bitreverse +; AVX512: Found an estimated cost of 5 for instruction: %bitreverse ; XOP: Found an estimated cost of 4 for instruction: %bitreverse %bitreverse = call <8 x i32> @llvm.bitreverse.v8i32(<8 x i32> %a) ret <8 x i32> %bitreverse } +define <16 x i32> @var_bitreverse_v16i32(<16 x i32> %a) { +; CHECK: 'Cost Model Analysis' for function 'var_bitreverse_v16i32': +; SSE2: Found an estimated cost of 108 for instruction: %bitreverse +; SSE42: Found an estimated cost of 20 for instruction: %bitreverse +; AVX: Found an estimated cost of 24 for instruction: %bitreverse +; AVX2: Found an estimated cost of 10 for instruction: %bitreverse +; AVX512F: Found an estimated cost of 24 for instruction: %bitreverse +; AVX512BW: Found an estimated cost of 5 for instruction: %bitreverse +; XOP: Found an estimated cost of 8 for instruction: %bitreverse + %bitreverse = call <16 x i32> @llvm.bitreverse.v16i32(<16 x i32> %a) + ret <16 x i32> %bitreverse +} + define <8 x i16> @var_bitreverse_v8i16(<8 x i16> %a) { ; CHECK: 'Cost Model Analysis' for function 'var_bitreverse_v8i16': ; SSE2: Found an estimated cost of 27 for instruction: %bitreverse ; SSE42: Found an estimated cost of 5 for instruction: %bitreverse ; AVX: Found an estimated cost of 5 for instruction: %bitreverse ; AVX2: Found an estimated cost of 5 for instruction: %bitreverse +; AVX512: Found an estimated cost of 5 for instruction: %bitreverse ; XOP: Found an estimated cost of 1 for instruction: %bitreverse %bitreverse = call <8 x i16> @llvm.bitreverse.v8i16(<8 x i16> %a) ret <8 x i16> %bitreverse @@ -125,17 +165,32 @@ define <16 x i16> @var_bitreverse_v16i16(<16 x i16> %a) { ; SSE42: Found an estimated cost of 10 for instruction: %bitreverse ; AVX: Found an estimated cost of 12 for instruction: %bitreverse ; AVX2: Found an estimated cost of 5 for instruction: %bitreverse +; AVX512: Found an estimated cost of 5 for instruction: %bitreverse ; XOP: Found an estimated cost of 4 for instruction: %bitreverse %bitreverse = call <16 x i16> @llvm.bitreverse.v16i16(<16 x i16> %a) ret <16 x i16> %bitreverse } +define <32 x i16> @var_bitreverse_v32i16(<32 x i16> %a) { +; CHECK: 'Cost Model Analysis' for function 'var_bitreverse_v32i16': +; SSE2: Found an estimated cost of 108 for instruction: %bitreverse +; SSE42: Found an estimated cost of 20 for instruction: %bitreverse +; AVX: Found an estimated cost of 24 for instruction: %bitreverse +; AVX2: Found an estimated cost of 10 for instruction: %bitreverse +; AVX512F: Found an estimated cost of 10 for instruction: %bitreverse +; AVX512BW: Found an estimated cost of 5 for instruction: %bitreverse +; XOP: Found an estimated cost of 8 for instruction: %bitreverse + %bitreverse = call <32 x i16> @llvm.bitreverse.v32i16(<32 x i16> %a) + ret <32 x i16> %bitreverse +} + define <16 x i8> @var_bitreverse_v16i8(<16 x i8> %a) { ; CHECK: 'Cost Model Analysis' for function 'var_bitreverse_v16i8': ; SSE2: Found an estimated cost of 20 for instruction: %bitreverse ; SSE42: Found an estimated cost of 5 for instruction: %bitreverse ; AVX: Found an estimated cost of 5 for instruction: %bitreverse ; AVX2: Found an estimated cost of 5 for instruction: %bitreverse +; AVX512: Found an estimated cost of 5 for instruction: %bitreverse ; XOP: Found an estimated cost of 1 for instruction: %bitreverse %bitreverse = call <16 x i8> @llvm.bitreverse.v16i8(<16 x i8> %a) ret <16 x i8> %bitreverse @@ -147,7 +202,21 @@ define <32 x i8> @var_bitreverse_v32i8(<32 x i8> %a) { ; SSE42: Found an estimated cost of 10 for instruction: %bitreverse ; AVX: Found an estimated cost of 12 for instruction: %bitreverse ; AVX2: Found an estimated cost of 5 for instruction: %bitreverse +; AVX512: Found an estimated cost of 5 for instruction: %bitreverse ; XOP: Found an estimated cost of 4 for instruction: %bitreverse %bitreverse = call <32 x i8> @llvm.bitreverse.v32i8(<32 x i8> %a) ret <32 x i8> %bitreverse } + +define <64 x i8> @var_bitreverse_v64i8(<64 x i8> %a) { +; CHECK: 'Cost Model Analysis' for function 'var_bitreverse_v64i8': +; SSE2: Found an estimated cost of 80 for instruction: %bitreverse +; SSE42: Found an estimated cost of 20 for instruction: %bitreverse +; AVX: Found an estimated cost of 24 for instruction: %bitreverse +; AVX2: Found an estimated cost of 10 for instruction: %bitreverse +; AVX512F: Found an estimated cost of 10 for instruction: %bitreverse +; AVX512BW: Found an estimated cost of 5 for instruction: %bitreverse +; XOP: Found an estimated cost of 8 for instruction: %bitreverse + %bitreverse = call <64 x i8> @llvm.bitreverse.v64i8(<64 x i8> %a) + ret <64 x i8> %bitreverse +} diff --git a/test/Analysis/CostModel/X86/ctbits-cost.ll b/test/Analysis/CostModel/X86/ctbits-cost.ll deleted file mode 100644 index aaf092c7b1d7..000000000000 --- a/test/Analysis/CostModel/X86/ctbits-cost.ll +++ /dev/null @@ -1,587 +0,0 @@ -; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -mcpu=pentium4 -cost-model -analyze | FileCheck %s -check-prefix=CHECK -check-prefix=SSE -check-prefix=SSE2 -check-prefix=NOPOPCNT -; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -mcpu=corei7 -cost-model -analyze | FileCheck %s -check-prefix=CHECK -check-prefix=SSE -check-prefix=SSE42 -check-prefix=POPCNT -; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -mcpu=corei7-avx -cost-model -analyze | FileCheck %s -check-prefix=CHECK -check-prefix=AVX -check-prefix=AVX1 -check-prefix=POPCNT -; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -mcpu=core-avx2 -cost-model -analyze | FileCheck %s -check-prefix=CHECK -check-prefix=AVX -check-prefix=AVX2 -check-prefix=POPCNT -; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -mcpu=bdver2 -cost-model -analyze | FileCheck %s -check-prefix=CHECK -check-prefix=AVX -check-prefix=AVX1 -check-prefix=POPCNT -; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -mcpu=bdver4 -cost-model -analyze | FileCheck %s -check-prefix=CHECK -check-prefix=AVX -check-prefix=AVX2 -check-prefix=POPCNT - -; Verify the cost of scalar population count instructions. - -declare i64 @llvm.ctpop.i64(i64) -declare i32 @llvm.ctpop.i32(i32) -declare i16 @llvm.ctpop.i16(i16) -declare i8 @llvm.ctpop.i8(i8) - -define i64 @var_ctpop_i64(i64 %a) { -; CHECK: 'Cost Model Analysis' for function 'var_ctpop_i64': -; NOPOPCNT: Found an estimated cost of 4 for instruction: %ctpop -; POPCNT: Found an estimated cost of 1 for instruction: %ctpop - %ctpop = call i64 @llvm.ctpop.i64(i64 %a) - ret i64 %ctpop -} - -define i32 @var_ctpop_i32(i32 %a) { -; CHECK: 'Cost Model Analysis' for function 'var_ctpop_i32': -; NOPOPCNT: Found an estimated cost of 4 for instruction: %ctpop -; POPCNT: Found an estimated cost of 1 for instruction: %ctpop - %ctpop = call i32 @llvm.ctpop.i32(i32 %a) - ret i32 %ctpop -} - -define i16 @var_ctpop_i16(i16 %a) { -; CHECK: 'Cost Model Analysis' for function 'var_ctpop_i16': -; NOPOPCNT: Found an estimated cost of 4 for instruction: %ctpop -; POPCNT: Found an estimated cost of 1 for instruction: %ctpop - %ctpop = call i16 @llvm.ctpop.i16(i16 %a) - ret i16 %ctpop -} - -define i8 @var_ctpop_i8(i8 %a) { -; CHECK: 'Cost Model Analysis' for function 'var_ctpop_i8': -; NOPOPCNT: Found an estimated cost of 4 for instruction: %ctpop -; POPCNT: Found an estimated cost of 1 for instruction: %ctpop - %ctpop = call i8 @llvm.ctpop.i8(i8 %a) - ret i8 %ctpop -} - -; Verify the cost of vector population count instructions. - -declare <2 x i64> @llvm.ctpop.v2i64(<2 x i64>) -declare <4 x i32> @llvm.ctpop.v4i32(<4 x i32>) -declare <8 x i16> @llvm.ctpop.v8i16(<8 x i16>) -declare <16 x i8> @llvm.ctpop.v16i8(<16 x i8>) - -declare <4 x i64> @llvm.ctpop.v4i64(<4 x i64>) -declare <8 x i32> @llvm.ctpop.v8i32(<8 x i32>) -declare <16 x i16> @llvm.ctpop.v16i16(<16 x i16>) -declare <32 x i8> @llvm.ctpop.v32i8(<32 x i8>) - -define <2 x i64> @var_ctpop_v2i64(<2 x i64> %a) { -; CHECK: 'Cost Model Analysis' for function 'var_ctpop_v2i64': -; SSE2: Found an estimated cost of 12 for instruction: %ctpop -; SSE42: Found an estimated cost of 7 for instruction: %ctpop -; AVX: Found an estimated cost of 7 for instruction: %ctpop - %ctpop = call <2 x i64> @llvm.ctpop.v2i64(<2 x i64> %a) - ret <2 x i64> %ctpop -} - -define <4 x i64> @var_ctpop_v4i64(<4 x i64> %a) { -; CHECK: 'Cost Model Analysis' for function 'var_ctpop_v4i64': -; SSE2: Found an estimated cost of 24 for instruction: %ctpop -; SSE42: Found an estimated cost of 14 for instruction: %ctpop -; AVX1: Found an estimated cost of 16 for instruction: %ctpop -; AVX2: Found an estimated cost of 7 for instruction: %ctpop - %ctpop = call <4 x i64> @llvm.ctpop.v4i64(<4 x i64> %a) - ret <4 x i64> %ctpop -} - -define <4 x i32> @var_ctpop_v4i32(<4 x i32> %a) { -; CHECK: 'Cost Model Analysis' for function 'var_ctpop_v4i32': -; SSE2: Found an estimated cost of 15 for instruction: %ctpop -; SSE42: Found an estimated cost of 11 for instruction: %ctpop -; AVX: Found an estimated cost of 11 for instruction: %ctpop - %ctpop = call <4 x i32> @llvm.ctpop.v4i32(<4 x i32> %a) - ret <4 x i32> %ctpop -} - -define <8 x i32> @var_ctpop_v8i32(<8 x i32> %a) { -; CHECK: 'Cost Model Analysis' for function 'var_ctpop_v8i32': -; SSE2: Found an estimated cost of 30 for instruction: %ctpop -; SSE42: Found an estimated cost of 22 for instruction: %ctpop -; AVX1: Found an estimated cost of 24 for instruction: %ctpop -; AVX2: Found an estimated cost of 11 for instruction: %ctpop - %ctpop = call <8 x i32> @llvm.ctpop.v8i32(<8 x i32> %a) - ret <8 x i32> %ctpop -} - -define <8 x i16> @var_ctpop_v8i16(<8 x i16> %a) { -; CHECK: 'Cost Model Analysis' for function 'var_ctpop_v8i16': -; SSE2: Found an estimated cost of 13 for instruction: %ctpop -; SSE42: Found an estimated cost of 9 for instruction: %ctpop -; AVX: Found an estimated cost of 9 for instruction: %ctpop - %ctpop = call <8 x i16> @llvm.ctpop.v8i16(<8 x i16> %a) - ret <8 x i16> %ctpop -} - -define <16 x i16> @var_ctpop_v16i16(<16 x i16> %a) { -; CHECK: 'Cost Model Analysis' for function 'var_ctpop_v16i16': -; SSE2: Found an estimated cost of 26 for instruction: %ctpop -; SSE42: Found an estimated cost of 18 for instruction: %ctpop -; AVX1: Found an estimated cost of 20 for instruction: %ctpop -; AVX2: Found an estimated cost of 9 for instruction: %ctpop - %ctpop = call <16 x i16> @llvm.ctpop.v16i16(<16 x i16> %a) - ret <16 x i16> %ctpop -} - -define <16 x i8> @var_ctpop_v16i8(<16 x i8> %a) { -; CHECK: 'Cost Model Analysis' for function 'var_ctpop_v16i8': -; SSE2: Found an estimated cost of 10 for instruction: %ctpop -; SSE42: Found an estimated cost of 6 for instruction: %ctpop -; AVX: Found an estimated cost of 6 for instruction: %ctpop - %ctpop = call <16 x i8> @llvm.ctpop.v16i8(<16 x i8> %a) - ret <16 x i8> %ctpop -} - -define <32 x i8> @var_ctpop_v32i8(<32 x i8> %a) { -; CHECK: 'Cost Model Analysis' for function 'var_ctpop_v32i8': -; SSE2: Found an estimated cost of 20 for instruction: %ctpop -; SSE42: Found an estimated cost of 12 for instruction: %ctpop -; AVX1: Found an estimated cost of 14 for instruction: %ctpop -; AVX2: Found an estimated cost of 6 for instruction: %ctpop - %ctpop = call <32 x i8> @llvm.ctpop.v32i8(<32 x i8> %a) - ret <32 x i8> %ctpop -} - -; Verify the cost of scalar leading zero count instructions. - -declare i64 @llvm.ctlz.i64(i64, i1) -declare i32 @llvm.ctlz.i32(i32, i1) -declare i16 @llvm.ctlz.i16(i16, i1) -declare i8 @llvm.ctlz.i8(i8, i1) - -define i64 @var_ctlz_i64(i64 %a) { -; CHECK: 'Cost Model Analysis' for function 'var_ctlz_i64': -; CHECK: Found an estimated cost of 1 for instruction: %ctlz - %ctlz = call i64 @llvm.ctlz.i64(i64 %a, i1 0) - ret i64 %ctlz -} - -define i64 @var_ctlz_i64u(i64 %a) { -; CHECK: 'Cost Model Analysis' for function 'var_ctlz_i64u': -; CHECK: Found an estimated cost of 1 for instruction: %ctlz - %ctlz = call i64 @llvm.ctlz.i64(i64 %a, i1 1) - ret i64 %ctlz -} - -define i32 @var_ctlz_i32(i32 %a) { -; CHECK: 'Cost Model Analysis' for function 'var_ctlz_i32': -; CHECK: Found an estimated cost of 1 for instruction: %ctlz - %ctlz = call i32 @llvm.ctlz.i32(i32 %a, i1 0) - ret i32 %ctlz -} - -define i32 @var_ctlz_i32u(i32 %a) { -; CHECK: 'Cost Model Analysis' for function 'var_ctlz_i32u': -; CHECK: Found an estimated cost of 1 for instruction: %ctlz - %ctlz = call i32 @llvm.ctlz.i32(i32 %a, i1 1) - ret i32 %ctlz -} - -define i16 @var_ctlz_i16(i16 %a) { -; CHECK: 'Cost Model Analysis' for function 'var_ctlz_i16': -; CHECK: Found an estimated cost of 1 for instruction: %ctlz - %ctlz = call i16 @llvm.ctlz.i16(i16 %a, i1 0) - ret i16 %ctlz -} - -define i16 @var_ctlz_i16u(i16 %a) { -; CHECK: 'Cost Model Analysis' for function 'var_ctlz_i16u': -; CHECK: Found an estimated cost of 1 for instruction: %ctlz - %ctlz = call i16 @llvm.ctlz.i16(i16 %a, i1 1) - ret i16 %ctlz -} - -define i8 @var_ctlz_i8(i8 %a) { -; CHECK: 'Cost Model Analysis' for function 'var_ctlz_i8': -; CHECK: Found an estimated cost of 1 for instruction: %ctlz - %ctlz = call i8 @llvm.ctlz.i8(i8 %a, i1 0) - ret i8 %ctlz -} - -define i8 @var_ctlz_i8u(i8 %a) { -; CHECK: 'Cost Model Analysis' for function 'var_ctlz_i8u': -; CHECK: Found an estimated cost of 1 for instruction: %ctlz - %ctlz = call i8 @llvm.ctlz.i8(i8 %a, i1 1) - ret i8 %ctlz -} - -; Verify the cost of vector leading zero count instructions. - -declare <2 x i64> @llvm.ctlz.v2i64(<2 x i64>, i1) -declare <4 x i32> @llvm.ctlz.v4i32(<4 x i32>, i1) -declare <8 x i16> @llvm.ctlz.v8i16(<8 x i16>, i1) -declare <16 x i8> @llvm.ctlz.v16i8(<16 x i8>, i1) - -declare <4 x i64> @llvm.ctlz.v4i64(<4 x i64>, i1) -declare <8 x i32> @llvm.ctlz.v8i32(<8 x i32>, i1) -declare <16 x i16> @llvm.ctlz.v16i16(<16 x i16>, i1) -declare <32 x i8> @llvm.ctlz.v32i8(<32 x i8>, i1) - -define <2 x i64> @var_ctlz_v2i64(<2 x i64> %a) { -; CHECK: 'Cost Model Analysis' for function 'var_ctlz_v2i64': -; SSE2: Found an estimated cost of 25 for instruction: %ctlz -; SSE42: Found an estimated cost of 23 for instruction: %ctlz -; AVX: Found an estimated cost of 23 for instruction: %ctlz - %ctlz = call <2 x i64> @llvm.ctlz.v2i64(<2 x i64> %a, i1 0) - ret <2 x i64> %ctlz -} - -define <2 x i64> @var_ctlz_v2i64u(<2 x i64> %a) { -; CHECK: 'Cost Model Analysis' for function 'var_ctlz_v2i64u': -; SSE2: Found an estimated cost of 25 for instruction: %ctlz -; SSE42: Found an estimated cost of 23 for instruction: %ctlz -; AVX: Found an estimated cost of 23 for instruction: %ctlz - %ctlz = call <2 x i64> @llvm.ctlz.v2i64(<2 x i64> %a, i1 1) - ret <2 x i64> %ctlz -} - -define <4 x i64> @var_ctlz_v4i64(<4 x i64> %a) { -; CHECK: 'Cost Model Analysis' for function 'var_ctlz_v4i64': -; SSE2: Found an estimated cost of 50 for instruction: %ctlz -; SSE42: Found an estimated cost of 46 for instruction: %ctlz -; AVX1: Found an estimated cost of 48 for instruction: %ctlz -; AVX2: Found an estimated cost of 23 for instruction: %ctlz - %ctlz = call <4 x i64> @llvm.ctlz.v4i64(<4 x i64> %a, i1 0) - ret <4 x i64> %ctlz -} - -define <4 x i64> @var_ctlz_v4i64u(<4 x i64> %a) { -; CHECK: 'Cost Model Analysis' for function 'var_ctlz_v4i64u': -; SSE2: Found an estimated cost of 50 for instruction: %ctlz -; SSE42: Found an estimated cost of 46 for instruction: %ctlz -; AVX1: Found an estimated cost of 48 for instruction: %ctlz -; AVX2: Found an estimated cost of 23 for instruction: %ctlz - %ctlz = call <4 x i64> @llvm.ctlz.v4i64(<4 x i64> %a, i1 1) - ret <4 x i64> %ctlz -} - -define <4 x i32> @var_ctlz_v4i32(<4 x i32> %a) { -; CHECK: 'Cost Model Analysis' for function 'var_ctlz_v4i32': -; SSE2: Found an estimated cost of 26 for instruction: %ctlz -; SSE42: Found an estimated cost of 18 for instruction: %ctlz -; AVX: Found an estimated cost of 18 for instruction: %ctlz - %ctlz = call <4 x i32> @llvm.ctlz.v4i32(<4 x i32> %a, i1 0) - ret <4 x i32> %ctlz -} - -define <4 x i32> @var_ctlz_v4i32u(<4 x i32> %a) { -; CHECK: 'Cost Model Analysis' for function 'var_ctlz_v4i32u': -; SSE2: Found an estimated cost of 26 for instruction: %ctlz -; SSE42: Found an estimated cost of 18 for instruction: %ctlz -; AVX: Found an estimated cost of 18 for instruction: %ctlz - %ctlz = call <4 x i32> @llvm.ctlz.v4i32(<4 x i32> %a, i1 1) - ret <4 x i32> %ctlz -} - -define <8 x i32> @var_ctlz_v8i32(<8 x i32> %a) { -; CHECK: 'Cost Model Analysis' for function 'var_ctlz_v8i32': -; SSE2: Found an estimated cost of 52 for instruction: %ctlz -; SSE42: Found an estimated cost of 36 for instruction: %ctlz -; AVX1: Found an estimated cost of 38 for instruction: %ctlz -; AVX2: Found an estimated cost of 18 for instruction: %ctlz - %ctlz = call <8 x i32> @llvm.ctlz.v8i32(<8 x i32> %a, i1 0) - ret <8 x i32> %ctlz -} - -define <8 x i32> @var_ctlz_v8i32u(<8 x i32> %a) { -; CHECK: 'Cost Model Analysis' for function 'var_ctlz_v8i32u': -; SSE2: Found an estimated cost of 52 for instruction: %ctlz -; SSE42: Found an estimated cost of 36 for instruction: %ctlz -; AVX1: Found an estimated cost of 38 for instruction: %ctlz -; AVX2: Found an estimated cost of 18 for instruction: %ctlz - %ctlz = call <8 x i32> @llvm.ctlz.v8i32(<8 x i32> %a, i1 1) - ret <8 x i32> %ctlz -} - -define <8 x i16> @var_ctlz_v8i16(<8 x i16> %a) { -; CHECK: 'Cost Model Analysis' for function 'var_ctlz_v8i16': -; SSE2: Found an estimated cost of 20 for instruction: %ctlz -; SSE42: Found an estimated cost of 14 for instruction: %ctlz -; AVX: Found an estimated cost of 14 for instruction: %ctlz - %ctlz = call <8 x i16> @llvm.ctlz.v8i16(<8 x i16> %a, i1 0) - ret <8 x i16> %ctlz -} - -define <8 x i16> @var_ctlz_v8i16u(<8 x i16> %a) { -; CHECK: 'Cost Model Analysis' for function 'var_ctlz_v8i16u': -; SSE2: Found an estimated cost of 20 for instruction: %ctlz -; SSE42: Found an estimated cost of 14 for instruction: %ctlz -; AVX: Found an estimated cost of 14 for instruction: %ctlz - %ctlz = call <8 x i16> @llvm.ctlz.v8i16(<8 x i16> %a, i1 1) - ret <8 x i16> %ctlz -} - -define <16 x i16> @var_ctlz_v16i16(<16 x i16> %a) { -; CHECK: 'Cost Model Analysis' for function 'var_ctlz_v16i16': -; SSE2: Found an estimated cost of 40 for instruction: %ctlz -; SSE42: Found an estimated cost of 28 for instruction: %ctlz -; AVX1: Found an estimated cost of 30 for instruction: %ctlz -; AVX2: Found an estimated cost of 14 for instruction: %ctlz - %ctlz = call <16 x i16> @llvm.ctlz.v16i16(<16 x i16> %a, i1 0) - ret <16 x i16> %ctlz -} - -define <16 x i16> @var_ctlz_v16i16u(<16 x i16> %a) { -; CHECK: 'Cost Model Analysis' for function 'var_ctlz_v16i16u': -; SSE2: Found an estimated cost of 40 for instruction: %ctlz -; SSE42: Found an estimated cost of 28 for instruction: %ctlz -; AVX1: Found an estimated cost of 30 for instruction: %ctlz -; AVX2: Found an estimated cost of 14 for instruction: %ctlz - %ctlz = call <16 x i16> @llvm.ctlz.v16i16(<16 x i16> %a, i1 1) - ret <16 x i16> %ctlz -} - -define <16 x i8> @var_ctlz_v16i8(<16 x i8> %a) { -; CHECK: 'Cost Model Analysis' for function 'var_ctlz_v16i8': -; SSE2: Found an estimated cost of 17 for instruction: %ctlz -; SSE42: Found an estimated cost of 9 for instruction: %ctlz -; AVX: Found an estimated cost of 9 for instruction: %ctlz - %ctlz = call <16 x i8> @llvm.ctlz.v16i8(<16 x i8> %a, i1 0) - ret <16 x i8> %ctlz -} - -define <16 x i8> @var_ctlz_v16i8u(<16 x i8> %a) { -; CHECK: 'Cost Model Analysis' for function 'var_ctlz_v16i8u': -; SSE2: Found an estimated cost of 17 for instruction: %ctlz -; SSE42: Found an estimated cost of 9 for instruction: %ctlz -; AVX: Found an estimated cost of 9 for instruction: %ctlz - %ctlz = call <16 x i8> @llvm.ctlz.v16i8(<16 x i8> %a, i1 1) - ret <16 x i8> %ctlz -} - -define <32 x i8> @var_ctlz_v32i8(<32 x i8> %a) { -; CHECK: 'Cost Model Analysis' for function 'var_ctlz_v32i8': -; SSE2: Found an estimated cost of 34 for instruction: %ctlz -; SSE42: Found an estimated cost of 18 for instruction: %ctlz -; AVX1: Found an estimated cost of 20 for instruction: %ctlz -; AVX2: Found an estimated cost of 9 for instruction: %ctlz - %ctlz = call <32 x i8> @llvm.ctlz.v32i8(<32 x i8> %a, i1 0) - ret <32 x i8> %ctlz -} - -define <32 x i8> @var_ctlz_v32i8u(<32 x i8> %a) { -; CHECK: 'Cost Model Analysis' for function 'var_ctlz_v32i8u': -; SSE2: Found an estimated cost of 34 for instruction: %ctlz -; SSE42: Found an estimated cost of 18 for instruction: %ctlz -; AVX1: Found an estimated cost of 20 for instruction: %ctlz -; AVX2: Found an estimated cost of 9 for instruction: %ctlz - %ctlz = call <32 x i8> @llvm.ctlz.v32i8(<32 x i8> %a, i1 1) - ret <32 x i8> %ctlz -} - -; Verify the cost of scalar trailing zero count instructions. - -declare i64 @llvm.cttz.i64(i64, i1) -declare i32 @llvm.cttz.i32(i32, i1) -declare i16 @llvm.cttz.i16(i16, i1) -declare i8 @llvm.cttz.i8(i8, i1) - -define i64 @var_cttz_i64(i64 %a) { -; CHECK: 'Cost Model Analysis' for function 'var_cttz_i64': -; CHECK: Found an estimated cost of 1 for instruction: %cttz - %cttz = call i64 @llvm.cttz.i64(i64 %a, i1 0) - ret i64 %cttz -} - -define i64 @var_cttz_i64u(i64 %a) { -; CHECK: 'Cost Model Analysis' for function 'var_cttz_i64u': -; CHECK: Found an estimated cost of 1 for instruction: %cttz - %cttz = call i64 @llvm.cttz.i64(i64 %a, i1 1) - ret i64 %cttz -} - -define i32 @var_cttz_i32(i32 %a) { -; CHECK: 'Cost Model Analysis' for function 'var_cttz_i32': -; CHECK: Found an estimated cost of 1 for instruction: %cttz - %cttz = call i32 @llvm.cttz.i32(i32 %a, i1 0) - ret i32 %cttz -} - -define i32 @var_cttz_i32u(i32 %a) { -; CHECK: 'Cost Model Analysis' for function 'var_cttz_i32u': -; CHECK: Found an estimated cost of 1 for instruction: %cttz - %cttz = call i32 @llvm.cttz.i32(i32 %a, i1 1) - ret i32 %cttz -} - -define i16 @var_cttz_i16(i16 %a) { -; CHECK: 'Cost Model Analysis' for function 'var_cttz_i16': -; CHECK: Found an estimated cost of 1 for instruction: %cttz - %cttz = call i16 @llvm.cttz.i16(i16 %a, i1 0) - ret i16 %cttz -} - -define i16 @var_cttz_i16u(i16 %a) { -; CHECK: 'Cost Model Analysis' for function 'var_cttz_i16u': -; CHECK: Found an estimated cost of 1 for instruction: %cttz - %cttz = call i16 @llvm.cttz.i16(i16 %a, i1 1) - ret i16 %cttz -} - -define i8 @var_cttz_i8(i8 %a) { -; CHECK: 'Cost Model Analysis' for function 'var_cttz_i8': -; CHECK: Found an estimated cost of 1 for instruction: %cttz - %cttz = call i8 @llvm.cttz.i8(i8 %a, i1 0) - ret i8 %cttz -} - -define i8 @var_cttz_i8u(i8 %a) { -; CHECK: 'Cost Model Analysis' for function 'var_cttz_i8u': -; CHECK: Found an estimated cost of 1 for instruction: %cttz - %cttz = call i8 @llvm.cttz.i8(i8 %a, i1 1) - ret i8 %cttz -} - -; Verify the cost of vector trailing zero count instructions. - -declare <2 x i64> @llvm.cttz.v2i64(<2 x i64>, i1) -declare <4 x i32> @llvm.cttz.v4i32(<4 x i32>, i1) -declare <8 x i16> @llvm.cttz.v8i16(<8 x i16>, i1) -declare <16 x i8> @llvm.cttz.v16i8(<16 x i8>, i1) - -declare <4 x i64> @llvm.cttz.v4i64(<4 x i64>, i1) -declare <8 x i32> @llvm.cttz.v8i32(<8 x i32>, i1) -declare <16 x i16> @llvm.cttz.v16i16(<16 x i16>, i1) -declare <32 x i8> @llvm.cttz.v32i8(<32 x i8>, i1) - -define <2 x i64> @var_cttz_v2i64(<2 x i64> %a) { -; CHECK: 'Cost Model Analysis' for function 'var_cttz_v2i64': -; SSE2: Found an estimated cost of 14 for instruction: %cttz -; SSE42: Found an estimated cost of 10 for instruction: %cttz -; AVX: Found an estimated cost of 10 for instruction: %cttz - %cttz = call <2 x i64> @llvm.cttz.v2i64(<2 x i64> %a, i1 0) - ret <2 x i64> %cttz -} - -define <2 x i64> @var_cttz_v2i64u(<2 x i64> %a) { -; CHECK: 'Cost Model Analysis' for function 'var_cttz_v2i64u': -; SSE2: Found an estimated cost of 14 for instruction: %cttz -; SSE42: Found an estimated cost of 10 for instruction: %cttz -; AVX: Found an estimated cost of 10 for instruction: %cttz - %cttz = call <2 x i64> @llvm.cttz.v2i64(<2 x i64> %a, i1 1) - ret <2 x i64> %cttz -} - -define <4 x i64> @var_cttz_v4i64(<4 x i64> %a) { -; CHECK: 'Cost Model Analysis' for function 'var_cttz_v4i64': -; SSE2: Found an estimated cost of 28 for instruction: %cttz -; SSE42: Found an estimated cost of 20 for instruction: %cttz -; AVX1: Found an estimated cost of 22 for instruction: %cttz -; AVX2: Found an estimated cost of 10 for instruction: %cttz - %cttz = call <4 x i64> @llvm.cttz.v4i64(<4 x i64> %a, i1 0) - ret <4 x i64> %cttz -} - -define <4 x i64> @var_cttz_v4i64u(<4 x i64> %a) { -; CHECK: 'Cost Model Analysis' for function 'var_cttz_v4i64u': -; SSE2: Found an estimated cost of 28 for instruction: %cttz -; SSE42: Found an estimated cost of 20 for instruction: %cttz -; AVX1: Found an estimated cost of 22 for instruction: %cttz -; AVX2: Found an estimated cost of 10 for instruction: %cttz - %cttz = call <4 x i64> @llvm.cttz.v4i64(<4 x i64> %a, i1 1) - ret <4 x i64> %cttz -} - -define <4 x i32> @var_cttz_v4i32(<4 x i32> %a) { -; CHECK: 'Cost Model Analysis' for function 'var_cttz_v4i32': -; SSE2: Found an estimated cost of 18 for instruction: %cttz -; SSE42: Found an estimated cost of 14 for instruction: %cttz -; AVX: Found an estimated cost of 14 for instruction: %cttz - %cttz = call <4 x i32> @llvm.cttz.v4i32(<4 x i32> %a, i1 0) - ret <4 x i32> %cttz -} - -define <4 x i32> @var_cttz_v4i32u(<4 x i32> %a) { -; CHECK: 'Cost Model Analysis' for function 'var_cttz_v4i32u': -; SSE2: Found an estimated cost of 18 for instruction: %cttz -; SSE42: Found an estimated cost of 14 for instruction: %cttz -; AVX: Found an estimated cost of 14 for instruction: %cttz - %cttz = call <4 x i32> @llvm.cttz.v4i32(<4 x i32> %a, i1 1) - ret <4 x i32> %cttz -} - -define <8 x i32> @var_cttz_v8i32(<8 x i32> %a) { -; CHECK: 'Cost Model Analysis' for function 'var_cttz_v8i32': -; SSE2: Found an estimated cost of 36 for instruction: %cttz -; SSE42: Found an estimated cost of 28 for instruction: %cttz -; AVX1: Found an estimated cost of 30 for instruction: %cttz -; AVX2: Found an estimated cost of 14 for instruction: %cttz - %cttz = call <8 x i32> @llvm.cttz.v8i32(<8 x i32> %a, i1 0) - ret <8 x i32> %cttz -} - -define <8 x i32> @var_cttz_v8i32u(<8 x i32> %a) { -; CHECK: 'Cost Model Analysis' for function 'var_cttz_v8i32u': -; SSE2: Found an estimated cost of 36 for instruction: %cttz -; SSE42: Found an estimated cost of 28 for instruction: %cttz -; AVX1: Found an estimated cost of 30 for instruction: %cttz -; AVX2: Found an estimated cost of 14 for instruction: %cttz - %cttz = call <8 x i32> @llvm.cttz.v8i32(<8 x i32> %a, i1 1) - ret <8 x i32> %cttz -} - -define <8 x i16> @var_cttz_v8i16(<8 x i16> %a) { -; CHECK: 'Cost Model Analysis' for function 'var_cttz_v8i16': -; SSE2: Found an estimated cost of 16 for instruction: %cttz -; SSE42: Found an estimated cost of 12 for instruction: %cttz -; AVX: Found an estimated cost of 12 for instruction: %cttz - %cttz = call <8 x i16> @llvm.cttz.v8i16(<8 x i16> %a, i1 0) - ret <8 x i16> %cttz -} - -define <8 x i16> @var_cttz_v8i16u(<8 x i16> %a) { -; CHECK: 'Cost Model Analysis' for function 'var_cttz_v8i16u': -; SSE2: Found an estimated cost of 16 for instruction: %cttz -; SSE42: Found an estimated cost of 12 for instruction: %cttz -; AVX: Found an estimated cost of 12 for instruction: %cttz - %cttz = call <8 x i16> @llvm.cttz.v8i16(<8 x i16> %a, i1 1) - ret <8 x i16> %cttz -} - -define <16 x i16> @var_cttz_v16i16(<16 x i16> %a) { -; CHECK: 'Cost Model Analysis' for function 'var_cttz_v16i16': -; SSE2: Found an estimated cost of 32 for instruction: %cttz -; SSE42: Found an estimated cost of 24 for instruction: %cttz -; AVX1: Found an estimated cost of 26 for instruction: %cttz -; AVX2: Found an estimated cost of 12 for instruction: %cttz - %cttz = call <16 x i16> @llvm.cttz.v16i16(<16 x i16> %a, i1 0) - ret <16 x i16> %cttz -} - -define <16 x i16> @var_cttz_v16i16u(<16 x i16> %a) { -; CHECK: 'Cost Model Analysis' for function 'var_cttz_v16i16u': -; SSE2: Found an estimated cost of 32 for instruction: %cttz -; SSE42: Found an estimated cost of 24 for instruction: %cttz -; AVX1: Found an estimated cost of 26 for instruction: %cttz -; AVX2: Found an estimated cost of 12 for instruction: %cttz - %cttz = call <16 x i16> @llvm.cttz.v16i16(<16 x i16> %a, i1 1) - ret <16 x i16> %cttz -} - -define <16 x i8> @var_cttz_v16i8(<16 x i8> %a) { -; CHECK: 'Cost Model Analysis' for function 'var_cttz_v16i8': -; SSE2: Found an estimated cost of 13 for instruction: %cttz -; SSE42: Found an estimated cost of 9 for instruction: %cttz -; AVX: Found an estimated cost of 9 for instruction: %cttz - %cttz = call <16 x i8> @llvm.cttz.v16i8(<16 x i8> %a, i1 0) - ret <16 x i8> %cttz -} - -define <16 x i8> @var_cttz_v16i8u(<16 x i8> %a) { -; CHECK: 'Cost Model Analysis' for function 'var_cttz_v16i8u': -; SSE2: Found an estimated cost of 13 for instruction: %cttz -; SSE42: Found an estimated cost of 9 for instruction: %cttz -; AVX: Found an estimated cost of 9 for instruction: %cttz - %cttz = call <16 x i8> @llvm.cttz.v16i8(<16 x i8> %a, i1 1) - ret <16 x i8> %cttz -} - -define <32 x i8> @var_cttz_v32i8(<32 x i8> %a) { -; CHECK: 'Cost Model Analysis' for function 'var_cttz_v32i8': -; SSE2: Found an estimated cost of 26 for instruction: %cttz -; SSE42: Found an estimated cost of 18 for instruction: %cttz -; AVX1: Found an estimated cost of 20 for instruction: %cttz -; AVX2: Found an estimated cost of 9 for instruction: %cttz - %cttz = call <32 x i8> @llvm.cttz.v32i8(<32 x i8> %a, i1 0) - ret <32 x i8> %cttz -} - -define <32 x i8> @var_cttz_v32i8u(<32 x i8> %a) { -; CHECK: 'Cost Model Analysis' for function 'var_cttz_v32i8u': -; SSE2: Found an estimated cost of 26 for instruction: %cttz -; SSE42: Found an estimated cost of 18 for instruction: %cttz -; AVX1: Found an estimated cost of 20 for instruction: %cttz -; AVX2: Found an estimated cost of 9 for instruction: %cttz - %cttz = call <32 x i8> @llvm.cttz.v32i8(<32 x i8> %a, i1 1) - ret <32 x i8> %cttz -} diff --git a/test/Analysis/CostModel/X86/ctlz.ll b/test/Analysis/CostModel/X86/ctlz.ll new file mode 100644 index 000000000000..2c97da15aee5 --- /dev/null +++ b/test/Analysis/CostModel/X86/ctlz.ll @@ -0,0 +1,233 @@ +; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -mcpu=pentium4 -cost-model -analyze | FileCheck %s -check-prefix=CHECK -check-prefix=SSE -check-prefix=SSE2 -check-prefix=NOPOPCNT +; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -mcpu=corei7 -cost-model -analyze | FileCheck %s -check-prefix=CHECK -check-prefix=SSE -check-prefix=SSE42 -check-prefix=POPCNT +; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -mcpu=corei7-avx -cost-model -analyze | FileCheck %s -check-prefix=CHECK -check-prefix=AVX -check-prefix=AVX1 -check-prefix=POPCNT +; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -mcpu=core-avx2 -cost-model -analyze | FileCheck %s -check-prefix=CHECK -check-prefix=AVX -check-prefix=AVX2 -check-prefix=POPCNT +; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -mcpu=bdver2 -cost-model -analyze | FileCheck %s -check-prefix=CHECK -check-prefix=AVX -check-prefix=AVX1 -check-prefix=POPCNT +; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -mcpu=bdver4 -cost-model -analyze | FileCheck %s -check-prefix=CHECK -check-prefix=AVX -check-prefix=AVX2 -check-prefix=POPCNT + +; Verify the cost of scalar leading zero count instructions. + +declare i64 @llvm.ctlz.i64(i64, i1) +declare i32 @llvm.ctlz.i32(i32, i1) +declare i16 @llvm.ctlz.i16(i16, i1) +declare i8 @llvm.ctlz.i8(i8, i1) + +define i64 @var_ctlz_i64(i64 %a) { +; CHECK: 'Cost Model Analysis' for function 'var_ctlz_i64': +; CHECK: Found an estimated cost of 1 for instruction: %ctlz + %ctlz = call i64 @llvm.ctlz.i64(i64 %a, i1 0) + ret i64 %ctlz +} + +define i64 @var_ctlz_i64u(i64 %a) { +; CHECK: 'Cost Model Analysis' for function 'var_ctlz_i64u': +; CHECK: Found an estimated cost of 1 for instruction: %ctlz + %ctlz = call i64 @llvm.ctlz.i64(i64 %a, i1 1) + ret i64 %ctlz +} + +define i32 @var_ctlz_i32(i32 %a) { +; CHECK: 'Cost Model Analysis' for function 'var_ctlz_i32': +; CHECK: Found an estimated cost of 1 for instruction: %ctlz + %ctlz = call i32 @llvm.ctlz.i32(i32 %a, i1 0) + ret i32 %ctlz +} + +define i32 @var_ctlz_i32u(i32 %a) { +; CHECK: 'Cost Model Analysis' for function 'var_ctlz_i32u': +; CHECK: Found an estimated cost of 1 for instruction: %ctlz + %ctlz = call i32 @llvm.ctlz.i32(i32 %a, i1 1) + ret i32 %ctlz +} + +define i16 @var_ctlz_i16(i16 %a) { +; CHECK: 'Cost Model Analysis' for function 'var_ctlz_i16': +; CHECK: Found an estimated cost of 1 for instruction: %ctlz + %ctlz = call i16 @llvm.ctlz.i16(i16 %a, i1 0) + ret i16 %ctlz +} + +define i16 @var_ctlz_i16u(i16 %a) { +; CHECK: 'Cost Model Analysis' for function 'var_ctlz_i16u': +; CHECK: Found an estimated cost of 1 for instruction: %ctlz + %ctlz = call i16 @llvm.ctlz.i16(i16 %a, i1 1) + ret i16 %ctlz +} + +define i8 @var_ctlz_i8(i8 %a) { +; CHECK: 'Cost Model Analysis' for function 'var_ctlz_i8': +; CHECK: Found an estimated cost of 1 for instruction: %ctlz + %ctlz = call i8 @llvm.ctlz.i8(i8 %a, i1 0) + ret i8 %ctlz +} + +define i8 @var_ctlz_i8u(i8 %a) { +; CHECK: 'Cost Model Analysis' for function 'var_ctlz_i8u': +; CHECK: Found an estimated cost of 1 for instruction: %ctlz + %ctlz = call i8 @llvm.ctlz.i8(i8 %a, i1 1) + ret i8 %ctlz +} + +; Verify the cost of vector leading zero count instructions. + +declare <2 x i64> @llvm.ctlz.v2i64(<2 x i64>, i1) +declare <4 x i32> @llvm.ctlz.v4i32(<4 x i32>, i1) +declare <8 x i16> @llvm.ctlz.v8i16(<8 x i16>, i1) +declare <16 x i8> @llvm.ctlz.v16i8(<16 x i8>, i1) + +declare <4 x i64> @llvm.ctlz.v4i64(<4 x i64>, i1) +declare <8 x i32> @llvm.ctlz.v8i32(<8 x i32>, i1) +declare <16 x i16> @llvm.ctlz.v16i16(<16 x i16>, i1) +declare <32 x i8> @llvm.ctlz.v32i8(<32 x i8>, i1) + +define <2 x i64> @var_ctlz_v2i64(<2 x i64> %a) { +; CHECK: 'Cost Model Analysis' for function 'var_ctlz_v2i64': +; SSE2: Found an estimated cost of 25 for instruction: %ctlz +; SSE42: Found an estimated cost of 23 for instruction: %ctlz +; AVX: Found an estimated cost of 23 for instruction: %ctlz + %ctlz = call <2 x i64> @llvm.ctlz.v2i64(<2 x i64> %a, i1 0) + ret <2 x i64> %ctlz +} + +define <2 x i64> @var_ctlz_v2i64u(<2 x i64> %a) { +; CHECK: 'Cost Model Analysis' for function 'var_ctlz_v2i64u': +; SSE2: Found an estimated cost of 25 for instruction: %ctlz +; SSE42: Found an estimated cost of 23 for instruction: %ctlz +; AVX: Found an estimated cost of 23 for instruction: %ctlz + %ctlz = call <2 x i64> @llvm.ctlz.v2i64(<2 x i64> %a, i1 1) + ret <2 x i64> %ctlz +} + +define <4 x i64> @var_ctlz_v4i64(<4 x i64> %a) { +; CHECK: 'Cost Model Analysis' for function 'var_ctlz_v4i64': +; SSE2: Found an estimated cost of 50 for instruction: %ctlz +; SSE42: Found an estimated cost of 46 for instruction: %ctlz +; AVX1: Found an estimated cost of 48 for instruction: %ctlz +; AVX2: Found an estimated cost of 23 for instruction: %ctlz + %ctlz = call <4 x i64> @llvm.ctlz.v4i64(<4 x i64> %a, i1 0) + ret <4 x i64> %ctlz +} + +define <4 x i64> @var_ctlz_v4i64u(<4 x i64> %a) { +; CHECK: 'Cost Model Analysis' for function 'var_ctlz_v4i64u': +; SSE2: Found an estimated cost of 50 for instruction: %ctlz +; SSE42: Found an estimated cost of 46 for instruction: %ctlz +; AVX1: Found an estimated cost of 48 for instruction: %ctlz +; AVX2: Found an estimated cost of 23 for instruction: %ctlz + %ctlz = call <4 x i64> @llvm.ctlz.v4i64(<4 x i64> %a, i1 1) + ret <4 x i64> %ctlz +} + +define <4 x i32> @var_ctlz_v4i32(<4 x i32> %a) { +; CHECK: 'Cost Model Analysis' for function 'var_ctlz_v4i32': +; SSE2: Found an estimated cost of 26 for instruction: %ctlz +; SSE42: Found an estimated cost of 18 for instruction: %ctlz +; AVX: Found an estimated cost of 18 for instruction: %ctlz + %ctlz = call <4 x i32> @llvm.ctlz.v4i32(<4 x i32> %a, i1 0) + ret <4 x i32> %ctlz +} + +define <4 x i32> @var_ctlz_v4i32u(<4 x i32> %a) { +; CHECK: 'Cost Model Analysis' for function 'var_ctlz_v4i32u': +; SSE2: Found an estimated cost of 26 for instruction: %ctlz +; SSE42: Found an estimated cost of 18 for instruction: %ctlz +; AVX: Found an estimated cost of 18 for instruction: %ctlz + %ctlz = call <4 x i32> @llvm.ctlz.v4i32(<4 x i32> %a, i1 1) + ret <4 x i32> %ctlz +} + +define <8 x i32> @var_ctlz_v8i32(<8 x i32> %a) { +; CHECK: 'Cost Model Analysis' for function 'var_ctlz_v8i32': +; SSE2: Found an estimated cost of 52 for instruction: %ctlz +; SSE42: Found an estimated cost of 36 for instruction: %ctlz +; AVX1: Found an estimated cost of 38 for instruction: %ctlz +; AVX2: Found an estimated cost of 18 for instruction: %ctlz + %ctlz = call <8 x i32> @llvm.ctlz.v8i32(<8 x i32> %a, i1 0) + ret <8 x i32> %ctlz +} + +define <8 x i32> @var_ctlz_v8i32u(<8 x i32> %a) { +; CHECK: 'Cost Model Analysis' for function 'var_ctlz_v8i32u': +; SSE2: Found an estimated cost of 52 for instruction: %ctlz +; SSE42: Found an estimated cost of 36 for instruction: %ctlz +; AVX1: Found an estimated cost of 38 for instruction: %ctlz +; AVX2: Found an estimated cost of 18 for instruction: %ctlz + %ctlz = call <8 x i32> @llvm.ctlz.v8i32(<8 x i32> %a, i1 1) + ret <8 x i32> %ctlz +} + +define <8 x i16> @var_ctlz_v8i16(<8 x i16> %a) { +; CHECK: 'Cost Model Analysis' for function 'var_ctlz_v8i16': +; SSE2: Found an estimated cost of 20 for instruction: %ctlz +; SSE42: Found an estimated cost of 14 for instruction: %ctlz +; AVX: Found an estimated cost of 14 for instruction: %ctlz + %ctlz = call <8 x i16> @llvm.ctlz.v8i16(<8 x i16> %a, i1 0) + ret <8 x i16> %ctlz +} + +define <8 x i16> @var_ctlz_v8i16u(<8 x i16> %a) { +; CHECK: 'Cost Model Analysis' for function 'var_ctlz_v8i16u': +; SSE2: Found an estimated cost of 20 for instruction: %ctlz +; SSE42: Found an estimated cost of 14 for instruction: %ctlz +; AVX: Found an estimated cost of 14 for instruction: %ctlz + %ctlz = call <8 x i16> @llvm.ctlz.v8i16(<8 x i16> %a, i1 1) + ret <8 x i16> %ctlz +} + +define <16 x i16> @var_ctlz_v16i16(<16 x i16> %a) { +; CHECK: 'Cost Model Analysis' for function 'var_ctlz_v16i16': +; SSE2: Found an estimated cost of 40 for instruction: %ctlz +; SSE42: Found an estimated cost of 28 for instruction: %ctlz +; AVX1: Found an estimated cost of 30 for instruction: %ctlz +; AVX2: Found an estimated cost of 14 for instruction: %ctlz + %ctlz = call <16 x i16> @llvm.ctlz.v16i16(<16 x i16> %a, i1 0) + ret <16 x i16> %ctlz +} + +define <16 x i16> @var_ctlz_v16i16u(<16 x i16> %a) { +; CHECK: 'Cost Model Analysis' for function 'var_ctlz_v16i16u': +; SSE2: Found an estimated cost of 40 for instruction: %ctlz +; SSE42: Found an estimated cost of 28 for instruction: %ctlz +; AVX1: Found an estimated cost of 30 for instruction: %ctlz +; AVX2: Found an estimated cost of 14 for instruction: %ctlz + %ctlz = call <16 x i16> @llvm.ctlz.v16i16(<16 x i16> %a, i1 1) + ret <16 x i16> %ctlz +} + +define <16 x i8> @var_ctlz_v16i8(<16 x i8> %a) { +; CHECK: 'Cost Model Analysis' for function 'var_ctlz_v16i8': +; SSE2: Found an estimated cost of 17 for instruction: %ctlz +; SSE42: Found an estimated cost of 9 for instruction: %ctlz +; AVX: Found an estimated cost of 9 for instruction: %ctlz + %ctlz = call <16 x i8> @llvm.ctlz.v16i8(<16 x i8> %a, i1 0) + ret <16 x i8> %ctlz +} + +define <16 x i8> @var_ctlz_v16i8u(<16 x i8> %a) { +; CHECK: 'Cost Model Analysis' for function 'var_ctlz_v16i8u': +; SSE2: Found an estimated cost of 17 for instruction: %ctlz +; SSE42: Found an estimated cost of 9 for instruction: %ctlz +; AVX: Found an estimated cost of 9 for instruction: %ctlz + %ctlz = call <16 x i8> @llvm.ctlz.v16i8(<16 x i8> %a, i1 1) + ret <16 x i8> %ctlz +} + +define <32 x i8> @var_ctlz_v32i8(<32 x i8> %a) { +; CHECK: 'Cost Model Analysis' for function 'var_ctlz_v32i8': +; SSE2: Found an estimated cost of 34 for instruction: %ctlz +; SSE42: Found an estimated cost of 18 for instruction: %ctlz +; AVX1: Found an estimated cost of 20 for instruction: %ctlz +; AVX2: Found an estimated cost of 9 for instruction: %ctlz + %ctlz = call <32 x i8> @llvm.ctlz.v32i8(<32 x i8> %a, i1 0) + ret <32 x i8> %ctlz +} + +define <32 x i8> @var_ctlz_v32i8u(<32 x i8> %a) { +; CHECK: 'Cost Model Analysis' for function 'var_ctlz_v32i8u': +; SSE2: Found an estimated cost of 34 for instruction: %ctlz +; SSE42: Found an estimated cost of 18 for instruction: %ctlz +; AVX1: Found an estimated cost of 20 for instruction: %ctlz +; AVX2: Found an estimated cost of 9 for instruction: %ctlz + %ctlz = call <32 x i8> @llvm.ctlz.v32i8(<32 x i8> %a, i1 1) + ret <32 x i8> %ctlz +} diff --git a/test/Analysis/CostModel/X86/ctpop.ll b/test/Analysis/CostModel/X86/ctpop.ll new file mode 100644 index 000000000000..f072cbaec492 --- /dev/null +++ b/test/Analysis/CostModel/X86/ctpop.ll @@ -0,0 +1,133 @@ +; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -mcpu=pentium4 -cost-model -analyze | FileCheck %s -check-prefix=CHECK -check-prefix=SSE -check-prefix=SSE2 -check-prefix=NOPOPCNT +; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -mcpu=corei7 -cost-model -analyze | FileCheck %s -check-prefix=CHECK -check-prefix=SSE -check-prefix=SSE42 -check-prefix=POPCNT +; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -mcpu=corei7-avx -cost-model -analyze | FileCheck %s -check-prefix=CHECK -check-prefix=AVX -check-prefix=AVX1 -check-prefix=POPCNT +; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -mcpu=core-avx2 -cost-model -analyze | FileCheck %s -check-prefix=CHECK -check-prefix=AVX -check-prefix=AVX2 -check-prefix=POPCNT +; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -mcpu=bdver2 -cost-model -analyze | FileCheck %s -check-prefix=CHECK -check-prefix=AVX -check-prefix=AVX1 -check-prefix=POPCNT +; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -mcpu=bdver4 -cost-model -analyze | FileCheck %s -check-prefix=CHECK -check-prefix=AVX -check-prefix=AVX2 -check-prefix=POPCNT + +; Verify the cost of scalar population count instructions. + +declare i64 @llvm.ctpop.i64(i64) +declare i32 @llvm.ctpop.i32(i32) +declare i16 @llvm.ctpop.i16(i16) +declare i8 @llvm.ctpop.i8(i8) + +define i64 @var_ctpop_i64(i64 %a) { +; CHECK: 'Cost Model Analysis' for function 'var_ctpop_i64': +; NOPOPCNT: Found an estimated cost of 4 for instruction: %ctpop +; POPCNT: Found an estimated cost of 1 for instruction: %ctpop + %ctpop = call i64 @llvm.ctpop.i64(i64 %a) + ret i64 %ctpop +} + +define i32 @var_ctpop_i32(i32 %a) { +; CHECK: 'Cost Model Analysis' for function 'var_ctpop_i32': +; NOPOPCNT: Found an estimated cost of 4 for instruction: %ctpop +; POPCNT: Found an estimated cost of 1 for instruction: %ctpop + %ctpop = call i32 @llvm.ctpop.i32(i32 %a) + ret i32 %ctpop +} + +define i16 @var_ctpop_i16(i16 %a) { +; CHECK: 'Cost Model Analysis' for function 'var_ctpop_i16': +; NOPOPCNT: Found an estimated cost of 4 for instruction: %ctpop +; POPCNT: Found an estimated cost of 1 for instruction: %ctpop + %ctpop = call i16 @llvm.ctpop.i16(i16 %a) + ret i16 %ctpop +} + +define i8 @var_ctpop_i8(i8 %a) { +; CHECK: 'Cost Model Analysis' for function 'var_ctpop_i8': +; NOPOPCNT: Found an estimated cost of 4 for instruction: %ctpop +; POPCNT: Found an estimated cost of 1 for instruction: %ctpop + %ctpop = call i8 @llvm.ctpop.i8(i8 %a) + ret i8 %ctpop +} + +; Verify the cost of vector population count instructions. + +declare <2 x i64> @llvm.ctpop.v2i64(<2 x i64>) +declare <4 x i32> @llvm.ctpop.v4i32(<4 x i32>) +declare <8 x i16> @llvm.ctpop.v8i16(<8 x i16>) +declare <16 x i8> @llvm.ctpop.v16i8(<16 x i8>) + +declare <4 x i64> @llvm.ctpop.v4i64(<4 x i64>) +declare <8 x i32> @llvm.ctpop.v8i32(<8 x i32>) +declare <16 x i16> @llvm.ctpop.v16i16(<16 x i16>) +declare <32 x i8> @llvm.ctpop.v32i8(<32 x i8>) + +define <2 x i64> @var_ctpop_v2i64(<2 x i64> %a) { +; CHECK: 'Cost Model Analysis' for function 'var_ctpop_v2i64': +; SSE2: Found an estimated cost of 12 for instruction: %ctpop +; SSE42: Found an estimated cost of 7 for instruction: %ctpop +; AVX: Found an estimated cost of 7 for instruction: %ctpop + %ctpop = call <2 x i64> @llvm.ctpop.v2i64(<2 x i64> %a) + ret <2 x i64> %ctpop +} + +define <4 x i64> @var_ctpop_v4i64(<4 x i64> %a) { +; CHECK: 'Cost Model Analysis' for function 'var_ctpop_v4i64': +; SSE2: Found an estimated cost of 24 for instruction: %ctpop +; SSE42: Found an estimated cost of 14 for instruction: %ctpop +; AVX1: Found an estimated cost of 16 for instruction: %ctpop +; AVX2: Found an estimated cost of 7 for instruction: %ctpop + %ctpop = call <4 x i64> @llvm.ctpop.v4i64(<4 x i64> %a) + ret <4 x i64> %ctpop +} + +define <4 x i32> @var_ctpop_v4i32(<4 x i32> %a) { +; CHECK: 'Cost Model Analysis' for function 'var_ctpop_v4i32': +; SSE2: Found an estimated cost of 15 for instruction: %ctpop +; SSE42: Found an estimated cost of 11 for instruction: %ctpop +; AVX: Found an estimated cost of 11 for instruction: %ctpop + %ctpop = call <4 x i32> @llvm.ctpop.v4i32(<4 x i32> %a) + ret <4 x i32> %ctpop +} + +define <8 x i32> @var_ctpop_v8i32(<8 x i32> %a) { +; CHECK: 'Cost Model Analysis' for function 'var_ctpop_v8i32': +; SSE2: Found an estimated cost of 30 for instruction: %ctpop +; SSE42: Found an estimated cost of 22 for instruction: %ctpop +; AVX1: Found an estimated cost of 24 for instruction: %ctpop +; AVX2: Found an estimated cost of 11 for instruction: %ctpop + %ctpop = call <8 x i32> @llvm.ctpop.v8i32(<8 x i32> %a) + ret <8 x i32> %ctpop +} + +define <8 x i16> @var_ctpop_v8i16(<8 x i16> %a) { +; CHECK: 'Cost Model Analysis' for function 'var_ctpop_v8i16': +; SSE2: Found an estimated cost of 13 for instruction: %ctpop +; SSE42: Found an estimated cost of 9 for instruction: %ctpop +; AVX: Found an estimated cost of 9 for instruction: %ctpop + %ctpop = call <8 x i16> @llvm.ctpop.v8i16(<8 x i16> %a) + ret <8 x i16> %ctpop +} + +define <16 x i16> @var_ctpop_v16i16(<16 x i16> %a) { +; CHECK: 'Cost Model Analysis' for function 'var_ctpop_v16i16': +; SSE2: Found an estimated cost of 26 for instruction: %ctpop +; SSE42: Found an estimated cost of 18 for instruction: %ctpop +; AVX1: Found an estimated cost of 20 for instruction: %ctpop +; AVX2: Found an estimated cost of 9 for instruction: %ctpop + %ctpop = call <16 x i16> @llvm.ctpop.v16i16(<16 x i16> %a) + ret <16 x i16> %ctpop +} + +define <16 x i8> @var_ctpop_v16i8(<16 x i8> %a) { +; CHECK: 'Cost Model Analysis' for function 'var_ctpop_v16i8': +; SSE2: Found an estimated cost of 10 for instruction: %ctpop +; SSE42: Found an estimated cost of 6 for instruction: %ctpop +; AVX: Found an estimated cost of 6 for instruction: %ctpop + %ctpop = call <16 x i8> @llvm.ctpop.v16i8(<16 x i8> %a) + ret <16 x i8> %ctpop +} + +define <32 x i8> @var_ctpop_v32i8(<32 x i8> %a) { +; CHECK: 'Cost Model Analysis' for function 'var_ctpop_v32i8': +; SSE2: Found an estimated cost of 20 for instruction: %ctpop +; SSE42: Found an estimated cost of 12 for instruction: %ctpop +; AVX1: Found an estimated cost of 14 for instruction: %ctpop +; AVX2: Found an estimated cost of 6 for instruction: %ctpop + %ctpop = call <32 x i8> @llvm.ctpop.v32i8(<32 x i8> %a) + ret <32 x i8> %ctpop +} diff --git a/test/Analysis/CostModel/X86/cttz.ll b/test/Analysis/CostModel/X86/cttz.ll new file mode 100644 index 000000000000..5d3c59b60232 --- /dev/null +++ b/test/Analysis/CostModel/X86/cttz.ll @@ -0,0 +1,233 @@ +; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -mcpu=pentium4 -cost-model -analyze | FileCheck %s -check-prefix=CHECK -check-prefix=SSE -check-prefix=SSE2 -check-prefix=NOPOPCNT +; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -mcpu=corei7 -cost-model -analyze | FileCheck %s -check-prefix=CHECK -check-prefix=SSE -check-prefix=SSE42 -check-prefix=POPCNT +; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -mcpu=corei7-avx -cost-model -analyze | FileCheck %s -check-prefix=CHECK -check-prefix=AVX -check-prefix=AVX1 -check-prefix=POPCNT +; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -mcpu=core-avx2 -cost-model -analyze | FileCheck %s -check-prefix=CHECK -check-prefix=AVX -check-prefix=AVX2 -check-prefix=POPCNT +; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -mcpu=bdver2 -cost-model -analyze | FileCheck %s -check-prefix=CHECK -check-prefix=AVX -check-prefix=AVX1 -check-prefix=POPCNT +; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -mcpu=bdver4 -cost-model -analyze | FileCheck %s -check-prefix=CHECK -check-prefix=AVX -check-prefix=AVX2 -check-prefix=POPCNT + +; Verify the cost of scalar trailing zero count instructions. + +declare i64 @llvm.cttz.i64(i64, i1) +declare i32 @llvm.cttz.i32(i32, i1) +declare i16 @llvm.cttz.i16(i16, i1) +declare i8 @llvm.cttz.i8(i8, i1) + +define i64 @var_cttz_i64(i64 %a) { +; CHECK: 'Cost Model Analysis' for function 'var_cttz_i64': +; CHECK: Found an estimated cost of 1 for instruction: %cttz + %cttz = call i64 @llvm.cttz.i64(i64 %a, i1 0) + ret i64 %cttz +} + +define i64 @var_cttz_i64u(i64 %a) { +; CHECK: 'Cost Model Analysis' for function 'var_cttz_i64u': +; CHECK: Found an estimated cost of 1 for instruction: %cttz + %cttz = call i64 @llvm.cttz.i64(i64 %a, i1 1) + ret i64 %cttz +} + +define i32 @var_cttz_i32(i32 %a) { +; CHECK: 'Cost Model Analysis' for function 'var_cttz_i32': +; CHECK: Found an estimated cost of 1 for instruction: %cttz + %cttz = call i32 @llvm.cttz.i32(i32 %a, i1 0) + ret i32 %cttz +} + +define i32 @var_cttz_i32u(i32 %a) { +; CHECK: 'Cost Model Analysis' for function 'var_cttz_i32u': +; CHECK: Found an estimated cost of 1 for instruction: %cttz + %cttz = call i32 @llvm.cttz.i32(i32 %a, i1 1) + ret i32 %cttz +} + +define i16 @var_cttz_i16(i16 %a) { +; CHECK: 'Cost Model Analysis' for function 'var_cttz_i16': +; CHECK: Found an estimated cost of 1 for instruction: %cttz + %cttz = call i16 @llvm.cttz.i16(i16 %a, i1 0) + ret i16 %cttz +} + +define i16 @var_cttz_i16u(i16 %a) { +; CHECK: 'Cost Model Analysis' for function 'var_cttz_i16u': +; CHECK: Found an estimated cost of 1 for instruction: %cttz + %cttz = call i16 @llvm.cttz.i16(i16 %a, i1 1) + ret i16 %cttz +} + +define i8 @var_cttz_i8(i8 %a) { +; CHECK: 'Cost Model Analysis' for function 'var_cttz_i8': +; CHECK: Found an estimated cost of 1 for instruction: %cttz + %cttz = call i8 @llvm.cttz.i8(i8 %a, i1 0) + ret i8 %cttz +} + +define i8 @var_cttz_i8u(i8 %a) { +; CHECK: 'Cost Model Analysis' for function 'var_cttz_i8u': +; CHECK: Found an estimated cost of 1 for instruction: %cttz + %cttz = call i8 @llvm.cttz.i8(i8 %a, i1 1) + ret i8 %cttz +} + +; Verify the cost of vector trailing zero count instructions. + +declare <2 x i64> @llvm.cttz.v2i64(<2 x i64>, i1) +declare <4 x i32> @llvm.cttz.v4i32(<4 x i32>, i1) +declare <8 x i16> @llvm.cttz.v8i16(<8 x i16>, i1) +declare <16 x i8> @llvm.cttz.v16i8(<16 x i8>, i1) + +declare <4 x i64> @llvm.cttz.v4i64(<4 x i64>, i1) +declare <8 x i32> @llvm.cttz.v8i32(<8 x i32>, i1) +declare <16 x i16> @llvm.cttz.v16i16(<16 x i16>, i1) +declare <32 x i8> @llvm.cttz.v32i8(<32 x i8>, i1) + +define <2 x i64> @var_cttz_v2i64(<2 x i64> %a) { +; CHECK: 'Cost Model Analysis' for function 'var_cttz_v2i64': +; SSE2: Found an estimated cost of 14 for instruction: %cttz +; SSE42: Found an estimated cost of 10 for instruction: %cttz +; AVX: Found an estimated cost of 10 for instruction: %cttz + %cttz = call <2 x i64> @llvm.cttz.v2i64(<2 x i64> %a, i1 0) + ret <2 x i64> %cttz +} + +define <2 x i64> @var_cttz_v2i64u(<2 x i64> %a) { +; CHECK: 'Cost Model Analysis' for function 'var_cttz_v2i64u': +; SSE2: Found an estimated cost of 14 for instruction: %cttz +; SSE42: Found an estimated cost of 10 for instruction: %cttz +; AVX: Found an estimated cost of 10 for instruction: %cttz + %cttz = call <2 x i64> @llvm.cttz.v2i64(<2 x i64> %a, i1 1) + ret <2 x i64> %cttz +} + +define <4 x i64> @var_cttz_v4i64(<4 x i64> %a) { +; CHECK: 'Cost Model Analysis' for function 'var_cttz_v4i64': +; SSE2: Found an estimated cost of 28 for instruction: %cttz +; SSE42: Found an estimated cost of 20 for instruction: %cttz +; AVX1: Found an estimated cost of 22 for instruction: %cttz +; AVX2: Found an estimated cost of 10 for instruction: %cttz + %cttz = call <4 x i64> @llvm.cttz.v4i64(<4 x i64> %a, i1 0) + ret <4 x i64> %cttz +} + +define <4 x i64> @var_cttz_v4i64u(<4 x i64> %a) { +; CHECK: 'Cost Model Analysis' for function 'var_cttz_v4i64u': +; SSE2: Found an estimated cost of 28 for instruction: %cttz +; SSE42: Found an estimated cost of 20 for instruction: %cttz +; AVX1: Found an estimated cost of 22 for instruction: %cttz +; AVX2: Found an estimated cost of 10 for instruction: %cttz + %cttz = call <4 x i64> @llvm.cttz.v4i64(<4 x i64> %a, i1 1) + ret <4 x i64> %cttz +} + +define <4 x i32> @var_cttz_v4i32(<4 x i32> %a) { +; CHECK: 'Cost Model Analysis' for function 'var_cttz_v4i32': +; SSE2: Found an estimated cost of 18 for instruction: %cttz +; SSE42: Found an estimated cost of 14 for instruction: %cttz +; AVX: Found an estimated cost of 14 for instruction: %cttz + %cttz = call <4 x i32> @llvm.cttz.v4i32(<4 x i32> %a, i1 0) + ret <4 x i32> %cttz +} + +define <4 x i32> @var_cttz_v4i32u(<4 x i32> %a) { +; CHECK: 'Cost Model Analysis' for function 'var_cttz_v4i32u': +; SSE2: Found an estimated cost of 18 for instruction: %cttz +; SSE42: Found an estimated cost of 14 for instruction: %cttz +; AVX: Found an estimated cost of 14 for instruction: %cttz + %cttz = call <4 x i32> @llvm.cttz.v4i32(<4 x i32> %a, i1 1) + ret <4 x i32> %cttz +} + +define <8 x i32> @var_cttz_v8i32(<8 x i32> %a) { +; CHECK: 'Cost Model Analysis' for function 'var_cttz_v8i32': +; SSE2: Found an estimated cost of 36 for instruction: %cttz +; SSE42: Found an estimated cost of 28 for instruction: %cttz +; AVX1: Found an estimated cost of 30 for instruction: %cttz +; AVX2: Found an estimated cost of 14 for instruction: %cttz + %cttz = call <8 x i32> @llvm.cttz.v8i32(<8 x i32> %a, i1 0) + ret <8 x i32> %cttz +} + +define <8 x i32> @var_cttz_v8i32u(<8 x i32> %a) { +; CHECK: 'Cost Model Analysis' for function 'var_cttz_v8i32u': +; SSE2: Found an estimated cost of 36 for instruction: %cttz +; SSE42: Found an estimated cost of 28 for instruction: %cttz +; AVX1: Found an estimated cost of 30 for instruction: %cttz +; AVX2: Found an estimated cost of 14 for instruction: %cttz + %cttz = call <8 x i32> @llvm.cttz.v8i32(<8 x i32> %a, i1 1) + ret <8 x i32> %cttz +} + +define <8 x i16> @var_cttz_v8i16(<8 x i16> %a) { +; CHECK: 'Cost Model Analysis' for function 'var_cttz_v8i16': +; SSE2: Found an estimated cost of 16 for instruction: %cttz +; SSE42: Found an estimated cost of 12 for instruction: %cttz +; AVX: Found an estimated cost of 12 for instruction: %cttz + %cttz = call <8 x i16> @llvm.cttz.v8i16(<8 x i16> %a, i1 0) + ret <8 x i16> %cttz +} + +define <8 x i16> @var_cttz_v8i16u(<8 x i16> %a) { +; CHECK: 'Cost Model Analysis' for function 'var_cttz_v8i16u': +; SSE2: Found an estimated cost of 16 for instruction: %cttz +; SSE42: Found an estimated cost of 12 for instruction: %cttz +; AVX: Found an estimated cost of 12 for instruction: %cttz + %cttz = call <8 x i16> @llvm.cttz.v8i16(<8 x i16> %a, i1 1) + ret <8 x i16> %cttz +} + +define <16 x i16> @var_cttz_v16i16(<16 x i16> %a) { +; CHECK: 'Cost Model Analysis' for function 'var_cttz_v16i16': +; SSE2: Found an estimated cost of 32 for instruction: %cttz +; SSE42: Found an estimated cost of 24 for instruction: %cttz +; AVX1: Found an estimated cost of 26 for instruction: %cttz +; AVX2: Found an estimated cost of 12 for instruction: %cttz + %cttz = call <16 x i16> @llvm.cttz.v16i16(<16 x i16> %a, i1 0) + ret <16 x i16> %cttz +} + +define <16 x i16> @var_cttz_v16i16u(<16 x i16> %a) { +; CHECK: 'Cost Model Analysis' for function 'var_cttz_v16i16u': +; SSE2: Found an estimated cost of 32 for instruction: %cttz +; SSE42: Found an estimated cost of 24 for instruction: %cttz +; AVX1: Found an estimated cost of 26 for instruction: %cttz +; AVX2: Found an estimated cost of 12 for instruction: %cttz + %cttz = call <16 x i16> @llvm.cttz.v16i16(<16 x i16> %a, i1 1) + ret <16 x i16> %cttz +} + +define <16 x i8> @var_cttz_v16i8(<16 x i8> %a) { +; CHECK: 'Cost Model Analysis' for function 'var_cttz_v16i8': +; SSE2: Found an estimated cost of 13 for instruction: %cttz +; SSE42: Found an estimated cost of 9 for instruction: %cttz +; AVX: Found an estimated cost of 9 for instruction: %cttz + %cttz = call <16 x i8> @llvm.cttz.v16i8(<16 x i8> %a, i1 0) + ret <16 x i8> %cttz +} + +define <16 x i8> @var_cttz_v16i8u(<16 x i8> %a) { +; CHECK: 'Cost Model Analysis' for function 'var_cttz_v16i8u': +; SSE2: Found an estimated cost of 13 for instruction: %cttz +; SSE42: Found an estimated cost of 9 for instruction: %cttz +; AVX: Found an estimated cost of 9 for instruction: %cttz + %cttz = call <16 x i8> @llvm.cttz.v16i8(<16 x i8> %a, i1 1) + ret <16 x i8> %cttz +} + +define <32 x i8> @var_cttz_v32i8(<32 x i8> %a) { +; CHECK: 'Cost Model Analysis' for function 'var_cttz_v32i8': +; SSE2: Found an estimated cost of 26 for instruction: %cttz +; SSE42: Found an estimated cost of 18 for instruction: %cttz +; AVX1: Found an estimated cost of 20 for instruction: %cttz +; AVX2: Found an estimated cost of 9 for instruction: %cttz + %cttz = call <32 x i8> @llvm.cttz.v32i8(<32 x i8> %a, i1 0) + ret <32 x i8> %cttz +} + +define <32 x i8> @var_cttz_v32i8u(<32 x i8> %a) { +; CHECK: 'Cost Model Analysis' for function 'var_cttz_v32i8u': +; SSE2: Found an estimated cost of 26 for instruction: %cttz +; SSE42: Found an estimated cost of 18 for instruction: %cttz +; AVX1: Found an estimated cost of 20 for instruction: %cttz +; AVX2: Found an estimated cost of 9 for instruction: %cttz + %cttz = call <32 x i8> @llvm.cttz.v32i8(<32 x i8> %a, i1 1) + ret <32 x i8> %cttz +} diff --git a/test/CodeGen/AArch64/aarch64-addv.ll b/test/CodeGen/AArch64/aarch64-addv.ll index 91797c062b88..e65992e9913d 100644 --- a/test/CodeGen/AArch64/aarch64-addv.ll +++ b/test/CodeGen/AArch64/aarch64-addv.ll @@ -1,18 +1,16 @@ ; RUN: llc < %s -mtriple=aarch64-eabi -aarch64-neon-syntax=generic | FileCheck %s +; Function Attrs: nounwind readnone +declare i64 @llvm.experimental.vector.reduce.add.i64.v2i64(<2 x i64>) +declare i32 @llvm.experimental.vector.reduce.add.i32.v4i32(<4 x i32>) +declare i16 @llvm.experimental.vector.reduce.add.i16.v8i16(<8 x i16>) +declare i8 @llvm.experimental.vector.reduce.add.i8.v16i8(<16 x i8>) + define i8 @add_B(<16 x i8>* %arr) { ; CHECK-LABEL: add_B ; CHECK: addv {{b[0-9]+}}, {{v[0-9]+}}.16b %bin.rdx = load <16 x i8>, <16 x i8>* %arr - %rdx.shuf0 = shufflevector <16 x i8> %bin.rdx, <16 x i8> undef, <16 x i32> - %bin.rdx0 = add <16 x i8> %bin.rdx, %rdx.shuf0 - %rdx.shuf = shufflevector <16 x i8> %bin.rdx0, <16 x i8> undef, <16 x i32> - %bin.rdx11 = add <16 x i8> %bin.rdx0, %rdx.shuf - %rdx.shuf12 = shufflevector <16 x i8> %bin.rdx11, <16 x i8> undef, <16 x i32> - %bin.rdx13 = add <16 x i8> %bin.rdx11, %rdx.shuf12 - %rdx.shuf13 = shufflevector <16 x i8> %bin.rdx13, <16 x i8> undef, <16 x i32> - %bin.rdx14 = add <16 x i8> %bin.rdx13, %rdx.shuf13 - %r = extractelement <16 x i8> %bin.rdx14, i32 0 + %r = call i8 @llvm.experimental.vector.reduce.add.i8.v16i8(<16 x i8> %bin.rdx) ret i8 %r } @@ -20,13 +18,7 @@ define i16 @add_H(<8 x i16>* %arr) { ; CHECK-LABEL: add_H ; CHECK: addv {{h[0-9]+}}, {{v[0-9]+}}.8h %bin.rdx = load <8 x i16>, <8 x i16>* %arr - %rdx.shuf = shufflevector <8 x i16> %bin.rdx, <8 x i16> undef, <8 x i32> - %bin.rdx11 = add <8 x i16> %bin.rdx, %rdx.shuf - %rdx.shuf12 = shufflevector <8 x i16> %bin.rdx11, <8 x i16> undef, <8 x i32> - %bin.rdx13 = add <8 x i16> %bin.rdx11, %rdx.shuf12 - %rdx.shuf13 = shufflevector <8 x i16> %bin.rdx13, <8 x i16> undef, <8 x i32> - %bin.rdx14 = add <8 x i16> %bin.rdx13, %rdx.shuf13 - %r = extractelement <8 x i16> %bin.rdx14, i32 0 + %r = call i16 @llvm.experimental.vector.reduce.add.i16.v8i16(<8 x i16> %bin.rdx) ret i16 %r } @@ -34,11 +26,7 @@ define i32 @add_S( <4 x i32>* %arr) { ; CHECK-LABEL: add_S ; CHECK: addv {{s[0-9]+}}, {{v[0-9]+}}.4s %bin.rdx = load <4 x i32>, <4 x i32>* %arr - %rdx.shuf = shufflevector <4 x i32> %bin.rdx, <4 x i32> undef, <4 x i32> - %bin.rdx11 = add <4 x i32> %bin.rdx, %rdx.shuf - %rdx.shuf12 = shufflevector <4 x i32> %bin.rdx11, <4 x i32> undef, <4 x i32> - %bin.rdx13 = add <4 x i32> %bin.rdx11, %rdx.shuf12 - %r = extractelement <4 x i32> %bin.rdx13, i32 0 + %r = call i32 @llvm.experimental.vector.reduce.add.i32.v4i32(<4 x i32> %bin.rdx) ret i32 %r } @@ -46,12 +34,12 @@ define i64 @add_D(<2 x i64>* %arr) { ; CHECK-LABEL: add_D ; CHECK-NOT: addv %bin.rdx = load <2 x i64>, <2 x i64>* %arr - %rdx.shuf0 = shufflevector <2 x i64> %bin.rdx, <2 x i64> undef, <2 x i32> - %bin.rdx0 = add <2 x i64> %bin.rdx, %rdx.shuf0 - %r = extractelement <2 x i64> %bin.rdx0, i32 0 + %r = call i64 @llvm.experimental.vector.reduce.add.i64.v2i64(<2 x i64> %bin.rdx) ret i64 %r } +declare i32 @llvm.experimental.vector.reduce.add.i32.v8i32(<8 x i32>) + define i32 @oversized_ADDV_256(i8* noalias nocapture readonly %arg1, i8* noalias nocapture readonly %arg2) { ; CHECK-LABEL: oversized_ADDV_256 ; CHECK: addv {{s[0-9]+}}, {{v[0-9]+}}.4s @@ -66,33 +54,16 @@ entry: %7 = icmp slt <8 x i32> %6, zeroinitializer %8 = sub nsw <8 x i32> zeroinitializer, %6 %9 = select <8 x i1> %7, <8 x i32> %8, <8 x i32> %6 - %rdx.shuf = shufflevector <8 x i32> %9, <8 x i32> undef, <8 x i32> - %bin.rdx = add <8 x i32> %9, %rdx.shuf - %rdx.shuf1 = shufflevector <8 x i32> %bin.rdx, <8 x i32> undef, <8 x i32> - %bin.rdx2 = add <8 x i32> %bin.rdx, %rdx.shuf1 - %rdx.shuf3 = shufflevector <8 x i32> %bin.rdx2, <8 x i32> undef, <8 x i32> - %bin.rdx4 = add <8 x i32> %bin.rdx2, %rdx.shuf3 - %10 = extractelement <8 x i32> %bin.rdx4, i32 0 - ret i32 %10 + %r = call i32 @llvm.experimental.vector.reduce.add.i32.v8i32(<8 x i32> %9) + ret i32 %r } +declare i32 @llvm.experimental.vector.reduce.add.i32.v16i32(<16 x i32>) + define i32 @oversized_ADDV_512(<16 x i32>* %arr) { ; CHECK-LABEL: oversized_ADDV_512 ; CHECK: addv {{s[0-9]+}}, {{v[0-9]+}}.4s %bin.rdx = load <16 x i32>, <16 x i32>* %arr - - %rdx.shuf0 = shufflevector <16 x i32> %bin.rdx, <16 x i32> undef, <16 x i32> - %bin.rdx0 = add <16 x i32> %bin.rdx, %rdx.shuf0 - - %rdx.shuf = shufflevector <16 x i32> %bin.rdx0, <16 x i32> undef, <16 x i32> - %bin.rdx11 = add <16 x i32> %bin.rdx0, %rdx.shuf - - %rdx.shuf12 = shufflevector <16 x i32> %bin.rdx11, <16 x i32> undef, <16 x i32> - %bin.rdx13 = add <16 x i32> %bin.rdx11, %rdx.shuf12 - - %rdx.shuf13 = shufflevector <16 x i32> %bin.rdx13, <16 x i32> undef, <16 x i32> - %bin.rdx14 = add <16 x i32> %bin.rdx13, %rdx.shuf13 - - %r = extractelement <16 x i32> %bin.rdx14, i32 0 + %r = call i32 @llvm.experimental.vector.reduce.add.i32.v16i32(<16 x i32> %bin.rdx) ret i32 %r } diff --git a/test/CodeGen/AArch64/aarch64-minmaxv.ll b/test/CodeGen/AArch64/aarch64-minmaxv.ll index 9a56cd6ae7c0..760a8f8419f9 100644 --- a/test/CodeGen/AArch64/aarch64-minmaxv.ll +++ b/test/CodeGen/AArch64/aarch64-minmaxv.ll @@ -2,344 +2,148 @@ target datalayout = "e-m:e-i64:64-i128:128-n32:64-S128" +declare i8 @llvm.experimental.vector.reduce.smax.i8.v16i8(<16 x i8>) +declare i16 @llvm.experimental.vector.reduce.smax.i16.v8i16(<8 x i16>) +declare i32 @llvm.experimental.vector.reduce.smax.i32.v4i32(<4 x i32>) +declare i8 @llvm.experimental.vector.reduce.umax.i8.v16i8(<16 x i8>) +declare i16 @llvm.experimental.vector.reduce.umax.i16.v8i16(<8 x i16>) +declare i32 @llvm.experimental.vector.reduce.umax.i32.v4i32(<4 x i32>) + +declare i8 @llvm.experimental.vector.reduce.smin.i8.v16i8(<16 x i8>) +declare i16 @llvm.experimental.vector.reduce.smin.i16.v8i16(<8 x i16>) +declare i32 @llvm.experimental.vector.reduce.smin.i32.v4i32(<4 x i32>) +declare i8 @llvm.experimental.vector.reduce.umin.i8.v16i8(<16 x i8>) +declare i16 @llvm.experimental.vector.reduce.umin.i16.v8i16(<8 x i16>) +declare i32 @llvm.experimental.vector.reduce.umin.i32.v4i32(<4 x i32>) + +declare float @llvm.experimental.vector.reduce.fmax.f32.v4f32(<4 x float>) +declare float @llvm.experimental.vector.reduce.fmin.f32.v4f32(<4 x float>) + ; CHECK-LABEL: smax_B ; CHECK: smaxv {{b[0-9]+}}, {{v[0-9]+}}.16b define i8 @smax_B(<16 x i8>* nocapture readonly %arr) { %arr.load = load <16 x i8>, <16 x i8>* %arr - %rdx.shuf = shufflevector <16 x i8> %arr.load, <16 x i8> undef, <16 x i32> - %rdx.minmax.cmp22 = icmp sgt <16 x i8> %arr.load, %rdx.shuf - %rdx.minmax.select23 = select <16 x i1> %rdx.minmax.cmp22, <16 x i8> %arr.load, <16 x i8> %rdx.shuf - %rdx.shuf24 = shufflevector <16 x i8> %rdx.minmax.select23, <16 x i8> undef, <16 x i32> - %rdx.minmax.cmp25 = icmp sgt <16 x i8> %rdx.minmax.select23, %rdx.shuf24 - %rdx.minmax.select26 = select <16 x i1> %rdx.minmax.cmp25, <16 x i8> %rdx.minmax.select23, <16 x i8> %rdx.shuf24 - %rdx.shuf27 = shufflevector <16 x i8> %rdx.minmax.select26, <16 x i8> undef, <16 x i32> - %rdx.minmax.cmp28 = icmp sgt <16 x i8> %rdx.minmax.select26, %rdx.shuf27 - %rdx.minmax.select29 = select <16 x i1> %rdx.minmax.cmp28, <16 x i8> %rdx.minmax.select26, <16 x i8> %rdx.shuf27 - %rdx.shuf30 = shufflevector <16 x i8> %rdx.minmax.select29, <16 x i8> undef, <16 x i32> - %rdx.minmax.cmp31 = icmp sgt <16 x i8> %rdx.minmax.select29, %rdx.shuf30 - %rdx.minmax.cmp31.elt = extractelement <16 x i1> %rdx.minmax.cmp31, i32 0 - %rdx.minmax.select29.elt = extractelement <16 x i8> %rdx.minmax.select29, i32 0 - %rdx.shuf30.elt = extractelement <16 x i8> %rdx.minmax.select29, i32 1 - %r = select i1 %rdx.minmax.cmp31.elt, i8 %rdx.minmax.select29.elt, i8 %rdx.shuf30.elt + %r = call i8 @llvm.experimental.vector.reduce.smax.i8.v16i8(<16 x i8> %arr.load) ret i8 %r } ; CHECK-LABEL: smax_H ; CHECK: smaxv {{h[0-9]+}}, {{v[0-9]+}}.8h define i16 @smax_H(<8 x i16>* nocapture readonly %arr) { - %rdx.minmax.select = load <8 x i16>, <8 x i16>* %arr - %rdx.shuf = shufflevector <8 x i16> %rdx.minmax.select, <8 x i16> undef, <8 x i32> - %rdx.minmax.cmp23 = icmp sgt <8 x i16> %rdx.minmax.select, %rdx.shuf - %rdx.minmax.select24 = select <8 x i1> %rdx.minmax.cmp23, <8 x i16> %rdx.minmax.select, <8 x i16> %rdx.shuf - %rdx.shuf25 = shufflevector <8 x i16> %rdx.minmax.select24, <8 x i16> undef, <8 x i32> - %rdx.minmax.cmp26 = icmp sgt <8 x i16> %rdx.minmax.select24, %rdx.shuf25 - %rdx.minmax.select27 = select <8 x i1> %rdx.minmax.cmp26, <8 x i16> %rdx.minmax.select24, <8 x i16> %rdx.shuf25 - %rdx.shuf28 = shufflevector <8 x i16> %rdx.minmax.select27, <8 x i16> undef, <8 x i32> - %rdx.minmax.cmp29 = icmp sgt <8 x i16> %rdx.minmax.select27, %rdx.shuf28 - %rdx.minmax.cmp29.elt = extractelement <8 x i1> %rdx.minmax.cmp29, i32 0 - %rdx.minmax.select27.elt = extractelement <8 x i16> %rdx.minmax.select27, i32 0 - %rdx.shuf28.elt = extractelement <8 x i16> %rdx.minmax.select27, i32 1 - %r = select i1 %rdx.minmax.cmp29.elt, i16 %rdx.minmax.select27.elt, i16 %rdx.shuf28.elt + %arr.load = load <8 x i16>, <8 x i16>* %arr + %r = call i16 @llvm.experimental.vector.reduce.smax.i16.v8i16(<8 x i16> %arr.load) ret i16 %r } ; CHECK-LABEL: smax_S ; CHECK: smaxv {{s[0-9]+}}, {{v[0-9]+}}.4s define i32 @smax_S(<4 x i32> * nocapture readonly %arr) { - %rdx.minmax.select = load <4 x i32>, <4 x i32>* %arr - %rdx.shuf = shufflevector <4 x i32> %rdx.minmax.select, <4 x i32> undef, <4 x i32> - %rdx.minmax.cmp18 = icmp sgt <4 x i32> %rdx.minmax.select, %rdx.shuf - %rdx.minmax.select19 = select <4 x i1> %rdx.minmax.cmp18, <4 x i32> %rdx.minmax.select, <4 x i32> %rdx.shuf - %rdx.shuf20 = shufflevector <4 x i32> %rdx.minmax.select19, <4 x i32> undef, <4 x i32> - %rdx.minmax.cmp21 = icmp sgt <4 x i32> %rdx.minmax.select19, %rdx.shuf20 - %rdx.minmax.cmp21.elt = extractelement <4 x i1> %rdx.minmax.cmp21, i32 0 - %rdx.minmax.select19.elt = extractelement <4 x i32> %rdx.minmax.select19, i32 0 - %rdx.shuf20.elt = extractelement <4 x i32> %rdx.minmax.select19, i32 1 - %r = select i1 %rdx.minmax.cmp21.elt, i32 %rdx.minmax.select19.elt, i32 %rdx.shuf20.elt + %arr.load = load <4 x i32>, <4 x i32>* %arr + %r = call i32 @llvm.experimental.vector.reduce.smax.i32.v4i32(<4 x i32> %arr.load) ret i32 %r } -; CHECK-LABEL: smax_D -; CHECK-NOT: smaxv -define i64 @smax_D(<2 x i64>* nocapture readonly %arr) { - %rdx.minmax.select = load <2 x i64>, <2 x i64>* %arr - %rdx.shuf = shufflevector <2 x i64> %rdx.minmax.select, <2 x i64> undef, <2 x i32> - %rdx.minmax.cmp18 = icmp sgt <2 x i64> %rdx.minmax.select, %rdx.shuf - %rdx.minmax.cmp18.elt = extractelement <2 x i1> %rdx.minmax.cmp18, i32 0 - %rdx.minmax.select.elt = extractelement <2 x i64> %rdx.minmax.select, i32 0 - %rdx.shuf.elt = extractelement <2 x i64> %rdx.minmax.select, i32 1 - %r = select i1 %rdx.minmax.cmp18.elt, i64 %rdx.minmax.select.elt, i64 %rdx.shuf.elt - ret i64 %r -} - - ; CHECK-LABEL: umax_B ; CHECK: umaxv {{b[0-9]+}}, {{v[0-9]+}}.16b define i8 @umax_B(<16 x i8>* nocapture readonly %arr) { - %rdx.minmax.select = load <16 x i8>, <16 x i8>* %arr - %rdx.shuf = shufflevector <16 x i8> %rdx.minmax.select, <16 x i8> undef, <16 x i32> - %rdx.minmax.cmp22 = icmp ugt <16 x i8> %rdx.minmax.select, %rdx.shuf - %rdx.minmax.select23 = select <16 x i1> %rdx.minmax.cmp22, <16 x i8> %rdx.minmax.select, <16 x i8> %rdx.shuf - %rdx.shuf24 = shufflevector <16 x i8> %rdx.minmax.select23, <16 x i8> undef, <16 x i32> - %rdx.minmax.cmp25 = icmp ugt <16 x i8> %rdx.minmax.select23, %rdx.shuf24 - %rdx.minmax.select26 = select <16 x i1> %rdx.minmax.cmp25, <16 x i8> %rdx.minmax.select23, <16 x i8> %rdx.shuf24 - %rdx.shuf27 = shufflevector <16 x i8> %rdx.minmax.select26, <16 x i8> undef, <16 x i32> - %rdx.minmax.cmp28 = icmp ugt <16 x i8> %rdx.minmax.select26, %rdx.shuf27 - %rdx.minmax.select29 = select <16 x i1> %rdx.minmax.cmp28, <16 x i8> %rdx.minmax.select26, <16 x i8> %rdx.shuf27 - %rdx.shuf30 = shufflevector <16 x i8> %rdx.minmax.select29, <16 x i8> undef, <16 x i32> - %rdx.minmax.cmp31 = icmp ugt <16 x i8> %rdx.minmax.select29, %rdx.shuf30 - %rdx.minmax.cmp31.elt = extractelement <16 x i1> %rdx.minmax.cmp31, i32 0 - %rdx.minmax.select29.elt = extractelement <16 x i8> %rdx.minmax.select29, i32 0 - %rdx.shuf30.elt = extractelement <16 x i8> %rdx.minmax.select29, i32 1 - %r = select i1 %rdx.minmax.cmp31.elt, i8 %rdx.minmax.select29.elt, i8 %rdx.shuf30.elt + %arr.load = load <16 x i8>, <16 x i8>* %arr + %r = call i8 @llvm.experimental.vector.reduce.umax.i8.v16i8(<16 x i8> %arr.load) ret i8 %r } ; CHECK-LABEL: umax_H ; CHECK: umaxv {{h[0-9]+}}, {{v[0-9]+}}.8h define i16 @umax_H(<8 x i16>* nocapture readonly %arr) { - %rdx.minmax.select = load <8 x i16>, <8 x i16>* %arr - %rdx.shuf = shufflevector <8 x i16> %rdx.minmax.select, <8 x i16> undef, <8 x i32> - %rdx.minmax.cmp23 = icmp ugt <8 x i16> %rdx.minmax.select, %rdx.shuf - %rdx.minmax.select24 = select <8 x i1> %rdx.minmax.cmp23, <8 x i16> %rdx.minmax.select, <8 x i16> %rdx.shuf - %rdx.shuf25 = shufflevector <8 x i16> %rdx.minmax.select24, <8 x i16> undef, <8 x i32> - %rdx.minmax.cmp26 = icmp ugt <8 x i16> %rdx.minmax.select24, %rdx.shuf25 - %rdx.minmax.select27 = select <8 x i1> %rdx.minmax.cmp26, <8 x i16> %rdx.minmax.select24, <8 x i16> %rdx.shuf25 - %rdx.shuf28 = shufflevector <8 x i16> %rdx.minmax.select27, <8 x i16> undef, <8 x i32> - %rdx.minmax.cmp29 = icmp ugt <8 x i16> %rdx.minmax.select27, %rdx.shuf28 - %rdx.minmax.cmp29.elt = extractelement <8 x i1> %rdx.minmax.cmp29, i32 0 - %rdx.minmax.select27.elt = extractelement <8 x i16> %rdx.minmax.select27, i32 0 - %rdx.shuf28.elt = extractelement <8 x i16> %rdx.minmax.select27, i32 1 - %r = select i1 %rdx.minmax.cmp29.elt, i16 %rdx.minmax.select27.elt, i16 %rdx.shuf28.elt + %arr.load = load <8 x i16>, <8 x i16>* %arr + %r = call i16 @llvm.experimental.vector.reduce.umax.i16.v8i16(<8 x i16> %arr.load) ret i16 %r } ; CHECK-LABEL: umax_S ; CHECK: umaxv {{s[0-9]+}}, {{v[0-9]+}}.4s define i32 @umax_S(<4 x i32>* nocapture readonly %arr) { - %rdx.minmax.select = load <4 x i32>, <4 x i32>* %arr - %rdx.shuf = shufflevector <4 x i32> %rdx.minmax.select, <4 x i32> undef, <4 x i32> - %rdx.minmax.cmp18 = icmp ugt <4 x i32> %rdx.minmax.select, %rdx.shuf - %rdx.minmax.select19 = select <4 x i1> %rdx.minmax.cmp18, <4 x i32> %rdx.minmax.select, <4 x i32> %rdx.shuf - %rdx.shuf20 = shufflevector <4 x i32> %rdx.minmax.select19, <4 x i32> undef, <4 x i32> - %rdx.minmax.cmp21 = icmp ugt <4 x i32> %rdx.minmax.select19, %rdx.shuf20 - %rdx.minmax.cmp21.elt = extractelement <4 x i1> %rdx.minmax.cmp21, i32 0 - %rdx.minmax.select19.elt = extractelement <4 x i32> %rdx.minmax.select19, i32 0 - %rdx.shuf20.elt = extractelement <4 x i32> %rdx.minmax.select19, i32 1 - %r = select i1 %rdx.minmax.cmp21.elt, i32 %rdx.minmax.select19.elt, i32 %rdx.shuf20.elt + %arr.load = load <4 x i32>, <4 x i32>* %arr + %r = call i32 @llvm.experimental.vector.reduce.umax.i32.v4i32(<4 x i32> %arr.load) ret i32 %r } -; CHECK-LABEL: umax_D -; CHECK-NOT: umaxv -define i64 @umax_D(<2 x i64>* nocapture readonly %arr) { - %rdx.minmax.select = load <2 x i64>, <2 x i64>* %arr - %rdx.shuf = shufflevector <2 x i64> %rdx.minmax.select, <2 x i64> undef, <2 x i32> - %rdx.minmax.cmp18 = icmp ugt <2 x i64> %rdx.minmax.select, %rdx.shuf - %rdx.minmax.cmp18.elt = extractelement <2 x i1> %rdx.minmax.cmp18, i32 0 - %rdx.minmax.select.elt = extractelement <2 x i64> %rdx.minmax.select, i32 0 - %rdx.shuf.elt = extractelement <2 x i64> %rdx.minmax.select, i32 1 - %r = select i1 %rdx.minmax.cmp18.elt, i64 %rdx.minmax.select.elt, i64 %rdx.shuf.elt - ret i64 %r -} - - ; CHECK-LABEL: smin_B ; CHECK: sminv {{b[0-9]+}}, {{v[0-9]+}}.16b define i8 @smin_B(<16 x i8>* nocapture readonly %arr) { - %rdx.minmax.select = load <16 x i8>, <16 x i8>* %arr - %rdx.shuf = shufflevector <16 x i8> %rdx.minmax.select, <16 x i8> undef, <16 x i32> - %rdx.minmax.cmp22 = icmp slt <16 x i8> %rdx.minmax.select, %rdx.shuf - %rdx.minmax.select23 = select <16 x i1> %rdx.minmax.cmp22, <16 x i8> %rdx.minmax.select, <16 x i8> %rdx.shuf - %rdx.shuf24 = shufflevector <16 x i8> %rdx.minmax.select23, <16 x i8> undef, <16 x i32> - %rdx.minmax.cmp25 = icmp slt <16 x i8> %rdx.minmax.select23, %rdx.shuf24 - %rdx.minmax.select26 = select <16 x i1> %rdx.minmax.cmp25, <16 x i8> %rdx.minmax.select23, <16 x i8> %rdx.shuf24 - %rdx.shuf27 = shufflevector <16 x i8> %rdx.minmax.select26, <16 x i8> undef, <16 x i32> - %rdx.minmax.cmp28 = icmp slt <16 x i8> %rdx.minmax.select26, %rdx.shuf27 - %rdx.minmax.select29 = select <16 x i1> %rdx.minmax.cmp28, <16 x i8> %rdx.minmax.select26, <16 x i8> %rdx.shuf27 - %rdx.shuf30 = shufflevector <16 x i8> %rdx.minmax.select29, <16 x i8> undef, <16 x i32> - %rdx.minmax.cmp31 = icmp slt <16 x i8> %rdx.minmax.select29, %rdx.shuf30 - %rdx.minmax.cmp31.elt = extractelement <16 x i1> %rdx.minmax.cmp31, i32 0 - %rdx.minmax.select29.elt = extractelement <16 x i8> %rdx.minmax.select29, i32 0 - %rdx.shuf30.elt = extractelement <16 x i8> %rdx.minmax.select29, i32 1 - %r = select i1 %rdx.minmax.cmp31.elt, i8 %rdx.minmax.select29.elt, i8 %rdx.shuf30.elt + %arr.load = load <16 x i8>, <16 x i8>* %arr + %r = call i8 @llvm.experimental.vector.reduce.smin.i8.v16i8(<16 x i8> %arr.load) ret i8 %r } ; CHECK-LABEL: smin_H ; CHECK: sminv {{h[0-9]+}}, {{v[0-9]+}}.8h define i16 @smin_H(<8 x i16>* nocapture readonly %arr) { - %rdx.minmax.select = load <8 x i16>, <8 x i16>* %arr - %rdx.shuf = shufflevector <8 x i16> %rdx.minmax.select, <8 x i16> undef, <8 x i32> - %rdx.minmax.cmp23 = icmp slt <8 x i16> %rdx.minmax.select, %rdx.shuf - %rdx.minmax.select24 = select <8 x i1> %rdx.minmax.cmp23, <8 x i16> %rdx.minmax.select, <8 x i16> %rdx.shuf - %rdx.shuf25 = shufflevector <8 x i16> %rdx.minmax.select24, <8 x i16> undef, <8 x i32> - %rdx.minmax.cmp26 = icmp slt <8 x i16> %rdx.minmax.select24, %rdx.shuf25 - %rdx.minmax.select27 = select <8 x i1> %rdx.minmax.cmp26, <8 x i16> %rdx.minmax.select24, <8 x i16> %rdx.shuf25 - %rdx.shuf28 = shufflevector <8 x i16> %rdx.minmax.select27, <8 x i16> undef, <8 x i32> - %rdx.minmax.cmp29 = icmp slt <8 x i16> %rdx.minmax.select27, %rdx.shuf28 - %rdx.minmax.cmp29.elt = extractelement <8 x i1> %rdx.minmax.cmp29, i32 0 - %rdx.minmax.select27.elt = extractelement <8 x i16> %rdx.minmax.select27, i32 0 - %rdx.shuf28.elt = extractelement <8 x i16> %rdx.minmax.select27, i32 1 - %r = select i1 %rdx.minmax.cmp29.elt, i16 %rdx.minmax.select27.elt, i16 %rdx.shuf28.elt + %arr.load = load <8 x i16>, <8 x i16>* %arr + %r = call i16 @llvm.experimental.vector.reduce.smin.i16.v8i16(<8 x i16> %arr.load) ret i16 %r } ; CHECK-LABEL: smin_S ; CHECK: sminv {{s[0-9]+}}, {{v[0-9]+}}.4s define i32 @smin_S(<4 x i32>* nocapture readonly %arr) { - %rdx.minmax.select = load <4 x i32>, <4 x i32>* %arr - %rdx.shuf = shufflevector <4 x i32> %rdx.minmax.select, <4 x i32> undef, <4 x i32> - %rdx.minmax.cmp18 = icmp slt <4 x i32> %rdx.minmax.select, %rdx.shuf - %rdx.minmax.select19 = select <4 x i1> %rdx.minmax.cmp18, <4 x i32> %rdx.minmax.select, <4 x i32> %rdx.shuf - %rdx.shuf20 = shufflevector <4 x i32> %rdx.minmax.select19, <4 x i32> undef, <4 x i32> - %rdx.minmax.cmp21 = icmp slt <4 x i32> %rdx.minmax.select19, %rdx.shuf20 - %rdx.minmax.cmp21.elt = extractelement <4 x i1> %rdx.minmax.cmp21, i32 0 - %rdx.minmax.select19.elt = extractelement <4 x i32> %rdx.minmax.select19, i32 0 - %rdx.shuf20.elt = extractelement <4 x i32> %rdx.minmax.select19, i32 1 - %r = select i1 %rdx.minmax.cmp21.elt, i32 %rdx.minmax.select19.elt, i32 %rdx.shuf20.elt + %arr.load = load <4 x i32>, <4 x i32>* %arr + %r = call i32 @llvm.experimental.vector.reduce.smin.i32.v4i32(<4 x i32> %arr.load) ret i32 %r } -; CHECK-LABEL: smin_D -; CHECK-NOT: sminv -define i64 @smin_D(<2 x i64>* nocapture readonly %arr) { - %rdx.minmax.select = load <2 x i64>, <2 x i64>* %arr - %rdx.shuf = shufflevector <2 x i64> %rdx.minmax.select, <2 x i64> undef, <2 x i32> - %rdx.minmax.cmp18 = icmp slt <2 x i64> %rdx.minmax.select, %rdx.shuf - %rdx.minmax.cmp18.elt = extractelement <2 x i1> %rdx.minmax.cmp18, i32 0 - %rdx.minmax.select.elt = extractelement <2 x i64> %rdx.minmax.select, i32 0 - %rdx.shuf.elt = extractelement <2 x i64> %rdx.minmax.select, i32 1 - %r = select i1 %rdx.minmax.cmp18.elt, i64 %rdx.minmax.select.elt, i64 %rdx.shuf.elt - ret i64 %r -} - - ; CHECK-LABEL: umin_B ; CHECK: uminv {{b[0-9]+}}, {{v[0-9]+}}.16b define i8 @umin_B(<16 x i8>* nocapture readonly %arr) { - %rdx.minmax.select = load <16 x i8>, <16 x i8>* %arr - %rdx.shuf = shufflevector <16 x i8> %rdx.minmax.select, <16 x i8> undef, <16 x i32> - %rdx.minmax.cmp22 = icmp ult <16 x i8> %rdx.minmax.select, %rdx.shuf - %rdx.minmax.select23 = select <16 x i1> %rdx.minmax.cmp22, <16 x i8> %rdx.minmax.select, <16 x i8> %rdx.shuf - %rdx.shuf24 = shufflevector <16 x i8> %rdx.minmax.select23, <16 x i8> undef, <16 x i32> - %rdx.minmax.cmp25 = icmp ult <16 x i8> %rdx.minmax.select23, %rdx.shuf24 - %rdx.minmax.select26 = select <16 x i1> %rdx.minmax.cmp25, <16 x i8> %rdx.minmax.select23, <16 x i8> %rdx.shuf24 - %rdx.shuf27 = shufflevector <16 x i8> %rdx.minmax.select26, <16 x i8> undef, <16 x i32> - %rdx.minmax.cmp28 = icmp ult <16 x i8> %rdx.minmax.select26, %rdx.shuf27 - %rdx.minmax.select29 = select <16 x i1> %rdx.minmax.cmp28, <16 x i8> %rdx.minmax.select26, <16 x i8> %rdx.shuf27 - %rdx.shuf30 = shufflevector <16 x i8> %rdx.minmax.select29, <16 x i8> undef, <16 x i32> - %rdx.minmax.cmp31 = icmp ult <16 x i8> %rdx.minmax.select29, %rdx.shuf30 - %rdx.minmax.cmp31.elt = extractelement <16 x i1> %rdx.minmax.cmp31, i32 0 - %rdx.minmax.select29.elt = extractelement <16 x i8> %rdx.minmax.select29, i32 0 - %rdx.shuf30.elt = extractelement <16 x i8> %rdx.minmax.select29, i32 1 - %r = select i1 %rdx.minmax.cmp31.elt, i8 %rdx.minmax.select29.elt, i8 %rdx.shuf30.elt + %arr.load = load <16 x i8>, <16 x i8>* %arr + %r = call i8 @llvm.experimental.vector.reduce.umin.i8.v16i8(<16 x i8> %arr.load) ret i8 %r } ; CHECK-LABEL: umin_H ; CHECK: uminv {{h[0-9]+}}, {{v[0-9]+}}.8h define i16 @umin_H(<8 x i16>* nocapture readonly %arr) { - %rdx.minmax.select = load <8 x i16>, <8 x i16>* %arr - %rdx.shuf = shufflevector <8 x i16> %rdx.minmax.select, <8 x i16> undef, <8 x i32> - %rdx.minmax.cmp23 = icmp ult <8 x i16> %rdx.minmax.select, %rdx.shuf - %rdx.minmax.select24 = select <8 x i1> %rdx.minmax.cmp23, <8 x i16> %rdx.minmax.select, <8 x i16> %rdx.shuf - %rdx.shuf25 = shufflevector <8 x i16> %rdx.minmax.select24, <8 x i16> undef, <8 x i32> - %rdx.minmax.cmp26 = icmp ult <8 x i16> %rdx.minmax.select24, %rdx.shuf25 - %rdx.minmax.select27 = select <8 x i1> %rdx.minmax.cmp26, <8 x i16> %rdx.minmax.select24, <8 x i16> %rdx.shuf25 - %rdx.shuf28 = shufflevector <8 x i16> %rdx.minmax.select27, <8 x i16> undef, <8 x i32> - %rdx.minmax.cmp29 = icmp ult <8 x i16> %rdx.minmax.select27, %rdx.shuf28 - %rdx.minmax.cmp29.elt = extractelement <8 x i1> %rdx.minmax.cmp29, i32 0 - %rdx.minmax.select27.elt = extractelement <8 x i16> %rdx.minmax.select27, i32 0 - %rdx.shuf28.elt = extractelement <8 x i16> %rdx.minmax.select27, i32 1 - %r = select i1 %rdx.minmax.cmp29.elt, i16 %rdx.minmax.select27.elt, i16 %rdx.shuf28.elt + %arr.load = load <8 x i16>, <8 x i16>* %arr + %r = call i16 @llvm.experimental.vector.reduce.umin.i16.v8i16(<8 x i16> %arr.load) ret i16 %r } ; CHECK-LABEL: umin_S ; CHECK: uminv {{s[0-9]+}}, {{v[0-9]+}}.4s define i32 @umin_S(<4 x i32>* nocapture readonly %arr) { - %rdx.minmax.select = load <4 x i32>, <4 x i32>* %arr - %rdx.shuf = shufflevector <4 x i32> %rdx.minmax.select, <4 x i32> undef, <4 x i32> - %rdx.minmax.cmp18 = icmp ult <4 x i32> %rdx.minmax.select, %rdx.shuf - %rdx.minmax.select19 = select <4 x i1> %rdx.minmax.cmp18, <4 x i32> %rdx.minmax.select, <4 x i32> %rdx.shuf - %rdx.shuf20 = shufflevector <4 x i32> %rdx.minmax.select19, <4 x i32> undef, <4 x i32> - %rdx.minmax.cmp21 = icmp ult <4 x i32> %rdx.minmax.select19, %rdx.shuf20 - %rdx.minmax.cmp21.elt = extractelement <4 x i1> %rdx.minmax.cmp21, i32 0 - %rdx.minmax.select19.elt = extractelement <4 x i32> %rdx.minmax.select19, i32 0 - %rdx.shuf20.elt = extractelement <4 x i32> %rdx.minmax.select19, i32 1 - %r = select i1 %rdx.minmax.cmp21.elt, i32 %rdx.minmax.select19.elt, i32 %rdx.shuf20.elt + %arr.load = load <4 x i32>, <4 x i32>* %arr + %r = call i32 @llvm.experimental.vector.reduce.umin.i32.v4i32(<4 x i32> %arr.load) ret i32 %r } -; CHECK-LABEL: umin_D -; CHECK-NOT: uminv -define i64 @umin_D(<2 x i64>* nocapture readonly %arr) { - %rdx.minmax.select = load <2 x i64>, <2 x i64>* %arr - %rdx.shuf = shufflevector <2 x i64> %rdx.minmax.select, <2 x i64> undef, <2 x i32> - %rdx.minmax.cmp18 = icmp ult <2 x i64> %rdx.minmax.select, %rdx.shuf - %rdx.minmax.cmp18.elt = extractelement <2 x i1> %rdx.minmax.cmp18, i32 0 - %rdx.minmax.select.elt = extractelement <2 x i64> %rdx.minmax.select, i32 0 - %rdx.shuf.elt = extractelement <2 x i64> %rdx.minmax.select, i32 1 - %r = select i1 %rdx.minmax.cmp18.elt, i64 %rdx.minmax.select.elt, i64 %rdx.shuf.elt - ret i64 %r -} - ; CHECK-LABEL: fmaxnm_S ; CHECK: fmaxnmv define float @fmaxnm_S(<4 x float>* nocapture readonly %arr) { - %rdx.minmax.select = load <4 x float>, <4 x float>* %arr - %rdx.shuf = shufflevector <4 x float> %rdx.minmax.select, <4 x float> undef, <4 x i32> - %rdx.minmax.cmp = fcmp fast oge <4 x float> %rdx.minmax.select, %rdx.shuf - %rdx.minmax.select1 = select <4 x i1> %rdx.minmax.cmp, <4 x float> %rdx.minmax.select, <4 x float> %rdx.shuf - %rdx.shuf1 = shufflevector <4 x float> %rdx.minmax.select1, <4 x float> undef, <4 x i32> - %rdx.minmax.cmp1 = fcmp fast oge <4 x float> %rdx.minmax.select1, %rdx.shuf1 - %rdx.minmax.cmp1.elt = extractelement <4 x i1> %rdx.minmax.cmp1, i32 0 - %rdx.minmax.select1.elt = extractelement <4 x float> %rdx.minmax.select1, i32 0 - %rdx.shuf1.elt = extractelement <4 x float> %rdx.minmax.select1, i32 1 - %r = select i1 %rdx.minmax.cmp1.elt, float %rdx.minmax.select1.elt, float %rdx.shuf1.elt + %arr.load = load <4 x float>, <4 x float>* %arr + %r = call nnan float @llvm.experimental.vector.reduce.fmax.f32.v4f32(<4 x float> %arr.load) ret float %r } ; CHECK-LABEL: fminnm_S ; CHECK: fminnmv define float @fminnm_S(<4 x float>* nocapture readonly %arr) { - %rdx.minmax.select = load <4 x float>, <4 x float>* %arr - %rdx.shuf = shufflevector <4 x float> %rdx.minmax.select, <4 x float> undef, <4 x i32> - %rdx.minmax.cmp = fcmp fast ole <4 x float> %rdx.minmax.select, %rdx.shuf - %rdx.minmax.select1 = select <4 x i1> %rdx.minmax.cmp, <4 x float> %rdx.minmax.select, <4 x float> %rdx.shuf - %rdx.shuf1 = shufflevector <4 x float> %rdx.minmax.select1, <4 x float> undef, <4 x i32> - %rdx.minmax.cmp1 = fcmp fast ole <4 x float> %rdx.minmax.select1, %rdx.shuf1 - %rdx.minmax.cmp1.elt = extractelement <4 x i1> %rdx.minmax.cmp1, i32 0 - %rdx.minmax.select1.elt = extractelement <4 x float> %rdx.minmax.select1, i32 0 - %rdx.shuf1.elt = extractelement <4 x float> %rdx.minmax.select1, i32 1 - %r = select i1 %rdx.minmax.cmp1.elt, float %rdx.minmax.select1.elt, float %rdx.shuf1.elt + %arr.load = load <4 x float>, <4 x float>* %arr + %r = call nnan float @llvm.experimental.vector.reduce.fmin.f32.v4f32(<4 x float> %arr.load) ret float %r } +declare i16 @llvm.experimental.vector.reduce.umax.i16.v16i16(<16 x i16>) + define i16 @oversized_umax_256(<16 x i16>* nocapture readonly %arr) { ; CHECK-LABEL: oversized_umax_256 ; CHECK: umax [[V0:v[0-9]+]].8h, {{v[0-9]+}}.8h, {{v[0-9]+}}.8h ; CHECK: umaxv {{h[0-9]+}}, [[V0]] - %rdx.minmax.select = load <16 x i16>, <16 x i16>* %arr - %rdx.shuf = shufflevector <16 x i16> %rdx.minmax.select, <16 x i16> undef, <16 x i32> - %rdx.minmax.cmp22 = icmp ugt <16 x i16> %rdx.minmax.select, %rdx.shuf - %rdx.minmax.select23 = select <16 x i1> %rdx.minmax.cmp22, <16 x i16> %rdx.minmax.select, <16 x i16> %rdx.shuf - %rdx.shuf24 = shufflevector <16 x i16> %rdx.minmax.select23, <16 x i16> undef, <16 x i32> - %rdx.minmax.cmp25 = icmp ugt <16 x i16> %rdx.minmax.select23, %rdx.shuf24 - %rdx.minmax.select26 = select <16 x i1> %rdx.minmax.cmp25, <16 x i16> %rdx.minmax.select23, <16 x i16> %rdx.shuf24 - %rdx.shuf27 = shufflevector <16 x i16> %rdx.minmax.select26, <16 x i16> undef, <16 x i32> - %rdx.minmax.cmp28 = icmp ugt <16 x i16> %rdx.minmax.select26, %rdx.shuf27 - %rdx.minmax.select29 = select <16 x i1> %rdx.minmax.cmp28, <16 x i16> %rdx.minmax.select26, <16 x i16> %rdx.shuf27 - %rdx.shuf30 = shufflevector <16 x i16> %rdx.minmax.select29, <16 x i16> undef, <16 x i32> - %rdx.minmax.cmp31 = icmp ugt <16 x i16> %rdx.minmax.select29, %rdx.shuf30 - %rdx.minmax.cmp31.elt = extractelement <16 x i1> %rdx.minmax.cmp31, i32 0 - %rdx.minmax.select29.elt = extractelement <16 x i16> %rdx.minmax.select29, i32 0 - %rdx.shuf30.elt = extractelement <16 x i16> %rdx.minmax.select29, i32 1 - %r = select i1 %rdx.minmax.cmp31.elt, i16 %rdx.minmax.select29.elt, i16 %rdx.shuf30.elt + %arr.load = load <16 x i16>, <16 x i16>* %arr + %r = call i16 @llvm.experimental.vector.reduce.umax.i16.v16i16(<16 x i16> %arr.load) ret i16 %r } +declare i32 @llvm.experimental.vector.reduce.umax.i32.v16i32(<16 x i32>) + define i32 @oversized_umax_512(<16 x i32>* nocapture readonly %arr) { ; CHECK-LABEL: oversized_umax_512 ; CHECK: umax v @@ -347,47 +151,23 @@ define i32 @oversized_umax_512(<16 x i32>* nocapture readonly %arr) { ; CHECK-NEXT: umax [[V0:v[0-9]+]].4s, {{v[0-9]+}}.4s, {{v[0-9]+}}.4s ; CHECK-NEXT: umaxv {{s[0-9]+}}, [[V0]] %arr.load = load <16 x i32>, <16 x i32>* %arr - %rdx.shuf = shufflevector <16 x i32> %arr.load, <16 x i32> undef, <16 x i32> - %rdx.minmax.cmp22 = icmp ugt <16 x i32> %arr.load, %rdx.shuf - %rdx.minmax.select23 = select <16 x i1> %rdx.minmax.cmp22, <16 x i32> %arr.load, <16 x i32> %rdx.shuf - %rdx.shuf24 = shufflevector <16 x i32> %rdx.minmax.select23, <16 x i32> undef, <16 x i32> - %rdx.minmax.cmp25 = icmp ugt <16 x i32> %rdx.minmax.select23, %rdx.shuf24 - %rdx.minmax.select26 = select <16 x i1> %rdx.minmax.cmp25, <16 x i32> %rdx.minmax.select23, <16 x i32> %rdx.shuf24 - %rdx.shuf27 = shufflevector <16 x i32> %rdx.minmax.select26, <16 x i32> undef, <16 x i32> - %rdx.minmax.cmp28 = icmp ugt <16 x i32> %rdx.minmax.select26, %rdx.shuf27 - %rdx.minmax.select29 = select <16 x i1> %rdx.minmax.cmp28, <16 x i32> %rdx.minmax.select26, <16 x i32> %rdx.shuf27 - %rdx.shuf30 = shufflevector <16 x i32> %rdx.minmax.select29, <16 x i32> undef, <16 x i32> - %rdx.minmax.cmp31 = icmp ugt <16 x i32> %rdx.minmax.select29, %rdx.shuf30 - %rdx.minmax.cmp31.elt = extractelement <16 x i1> %rdx.minmax.cmp31, i32 0 - %rdx.minmax.select29.elt = extractelement <16 x i32> %rdx.minmax.select29, i32 0 - %rdx.shuf30.elt = extractelement <16 x i32> %rdx.minmax.select29, i32 1 - %r = select i1 %rdx.minmax.cmp31.elt, i32 %rdx.minmax.select29.elt, i32 %rdx.shuf30.elt + %r = call i32 @llvm.experimental.vector.reduce.umax.i32.v16i32(<16 x i32> %arr.load) ret i32 %r } +declare i16 @llvm.experimental.vector.reduce.umin.i16.v16i16(<16 x i16>) + define i16 @oversized_umin_256(<16 x i16>* nocapture readonly %arr) { ; CHECK-LABEL: oversized_umin_256 ; CHECK: umin [[V0:v[0-9]+]].8h, {{v[0-9]+}}.8h, {{v[0-9]+}}.8h ; CHECK: uminv {{h[0-9]+}}, [[V0]] - %rdx.minmax.select = load <16 x i16>, <16 x i16>* %arr - %rdx.shuf = shufflevector <16 x i16> %rdx.minmax.select, <16 x i16> undef, <16 x i32> - %rdx.minmax.cmp22 = icmp ult <16 x i16> %rdx.minmax.select, %rdx.shuf - %rdx.minmax.select23 = select <16 x i1> %rdx.minmax.cmp22, <16 x i16> %rdx.minmax.select, <16 x i16> %rdx.shuf - %rdx.shuf24 = shufflevector <16 x i16> %rdx.minmax.select23, <16 x i16> undef, <16 x i32> - %rdx.minmax.cmp25 = icmp ult <16 x i16> %rdx.minmax.select23, %rdx.shuf24 - %rdx.minmax.select26 = select <16 x i1> %rdx.minmax.cmp25, <16 x i16> %rdx.minmax.select23, <16 x i16> %rdx.shuf24 - %rdx.shuf27 = shufflevector <16 x i16> %rdx.minmax.select26, <16 x i16> undef, <16 x i32> - %rdx.minmax.cmp28 = icmp ult <16 x i16> %rdx.minmax.select26, %rdx.shuf27 - %rdx.minmax.select29 = select <16 x i1> %rdx.minmax.cmp28, <16 x i16> %rdx.minmax.select26, <16 x i16> %rdx.shuf27 - %rdx.shuf30 = shufflevector <16 x i16> %rdx.minmax.select29, <16 x i16> undef, <16 x i32> - %rdx.minmax.cmp31 = icmp ult <16 x i16> %rdx.minmax.select29, %rdx.shuf30 - %rdx.minmax.cmp31.elt = extractelement <16 x i1> %rdx.minmax.cmp31, i32 0 - %rdx.minmax.select29.elt = extractelement <16 x i16> %rdx.minmax.select29, i32 0 - %rdx.shuf30.elt = extractelement <16 x i16> %rdx.minmax.select29, i32 1 - %r = select i1 %rdx.minmax.cmp31.elt, i16 %rdx.minmax.select29.elt, i16 %rdx.shuf30.elt + %arr.load = load <16 x i16>, <16 x i16>* %arr + %r = call i16 @llvm.experimental.vector.reduce.umin.i16.v16i16(<16 x i16> %arr.load) ret i16 %r } +declare i32 @llvm.experimental.vector.reduce.umin.i32.v16i32(<16 x i32>) + define i32 @oversized_umin_512(<16 x i32>* nocapture readonly %arr) { ; CHECK-LABEL: oversized_umin_512 ; CHECK: umin v @@ -395,47 +175,23 @@ define i32 @oversized_umin_512(<16 x i32>* nocapture readonly %arr) { ; CHECK-NEXT: umin [[V0:v[0-9]+]].4s, {{v[0-9]+}}.4s, {{v[0-9]+}}.4s ; CHECK-NEXT: uminv {{s[0-9]+}}, [[V0]] %arr.load = load <16 x i32>, <16 x i32>* %arr - %rdx.shuf = shufflevector <16 x i32> %arr.load, <16 x i32> undef, <16 x i32> - %rdx.minmax.cmp22 = icmp ult <16 x i32> %arr.load, %rdx.shuf - %rdx.minmax.select23 = select <16 x i1> %rdx.minmax.cmp22, <16 x i32> %arr.load, <16 x i32> %rdx.shuf - %rdx.shuf24 = shufflevector <16 x i32> %rdx.minmax.select23, <16 x i32> undef, <16 x i32> - %rdx.minmax.cmp25 = icmp ult <16 x i32> %rdx.minmax.select23, %rdx.shuf24 - %rdx.minmax.select26 = select <16 x i1> %rdx.minmax.cmp25, <16 x i32> %rdx.minmax.select23, <16 x i32> %rdx.shuf24 - %rdx.shuf27 = shufflevector <16 x i32> %rdx.minmax.select26, <16 x i32> undef, <16 x i32> - %rdx.minmax.cmp28 = icmp ult <16 x i32> %rdx.minmax.select26, %rdx.shuf27 - %rdx.minmax.select29 = select <16 x i1> %rdx.minmax.cmp28, <16 x i32> %rdx.minmax.select26, <16 x i32> %rdx.shuf27 - %rdx.shuf30 = shufflevector <16 x i32> %rdx.minmax.select29, <16 x i32> undef, <16 x i32> - %rdx.minmax.cmp31 = icmp ult <16 x i32> %rdx.minmax.select29, %rdx.shuf30 - %rdx.minmax.cmp31.elt = extractelement <16 x i1> %rdx.minmax.cmp31, i32 0 - %rdx.minmax.select29.elt = extractelement <16 x i32> %rdx.minmax.select29, i32 0 - %rdx.shuf30.elt = extractelement <16 x i32> %rdx.minmax.select29, i32 1 - %r = select i1 %rdx.minmax.cmp31.elt, i32 %rdx.minmax.select29.elt, i32 %rdx.shuf30.elt + %r = call i32 @llvm.experimental.vector.reduce.umin.i32.v16i32(<16 x i32> %arr.load) ret i32 %r } +declare i16 @llvm.experimental.vector.reduce.smax.i16.v16i16(<16 x i16>) + define i16 @oversized_smax_256(<16 x i16>* nocapture readonly %arr) { ; CHECK-LABEL: oversized_smax_256 ; CHECK: smax [[V0:v[0-9]+]].8h, {{v[0-9]+}}.8h, {{v[0-9]+}}.8h ; CHECK: smaxv {{h[0-9]+}}, [[V0]] %arr.load = load <16 x i16>, <16 x i16>* %arr - %rdx.shuf = shufflevector <16 x i16> %arr.load, <16 x i16> undef, <16 x i32> - %rdx.minmax.cmp22 = icmp sgt <16 x i16> %arr.load, %rdx.shuf - %rdx.minmax.select23 = select <16 x i1> %rdx.minmax.cmp22, <16 x i16> %arr.load, <16 x i16> %rdx.shuf - %rdx.shuf24 = shufflevector <16 x i16> %rdx.minmax.select23, <16 x i16> undef, <16 x i32> - %rdx.minmax.cmp25 = icmp sgt <16 x i16> %rdx.minmax.select23, %rdx.shuf24 - %rdx.minmax.select26 = select <16 x i1> %rdx.minmax.cmp25, <16 x i16> %rdx.minmax.select23, <16 x i16> %rdx.shuf24 - %rdx.shuf27 = shufflevector <16 x i16> %rdx.minmax.select26, <16 x i16> undef, <16 x i32> - %rdx.minmax.cmp28 = icmp sgt <16 x i16> %rdx.minmax.select26, %rdx.shuf27 - %rdx.minmax.select29 = select <16 x i1> %rdx.minmax.cmp28, <16 x i16> %rdx.minmax.select26, <16 x i16> %rdx.shuf27 - %rdx.shuf30 = shufflevector <16 x i16> %rdx.minmax.select29, <16 x i16> undef, <16 x i32> - %rdx.minmax.cmp31 = icmp sgt <16 x i16> %rdx.minmax.select29, %rdx.shuf30 - %rdx.minmax.cmp31.elt = extractelement <16 x i1> %rdx.minmax.cmp31, i32 0 - %rdx.minmax.select29.elt = extractelement <16 x i16> %rdx.minmax.select29, i32 0 - %rdx.shuf30.elt = extractelement <16 x i16> %rdx.minmax.select29, i32 1 - %r = select i1 %rdx.minmax.cmp31.elt, i16 %rdx.minmax.select29.elt, i16 %rdx.shuf30.elt + %r = call i16 @llvm.experimental.vector.reduce.smax.i16.v16i16(<16 x i16> %arr.load) ret i16 %r } +declare i32 @llvm.experimental.vector.reduce.smax.i32.v16i32(<16 x i32>) + define i32 @oversized_smax_512(<16 x i32>* nocapture readonly %arr) { ; CHECK-LABEL: oversized_smax_512 ; CHECK: smax v @@ -443,47 +199,23 @@ define i32 @oversized_smax_512(<16 x i32>* nocapture readonly %arr) { ; CHECK-NEXT: smax [[V0:v[0-9]+]].4s, {{v[0-9]+}}.4s, {{v[0-9]+}}.4s ; CHECK-NEXT: smaxv {{s[0-9]+}}, [[V0]] %arr.load = load <16 x i32>, <16 x i32>* %arr - %rdx.shuf = shufflevector <16 x i32> %arr.load, <16 x i32> undef, <16 x i32> - %rdx.minmax.cmp22 = icmp sgt <16 x i32> %arr.load, %rdx.shuf - %rdx.minmax.select23 = select <16 x i1> %rdx.minmax.cmp22, <16 x i32> %arr.load, <16 x i32> %rdx.shuf - %rdx.shuf24 = shufflevector <16 x i32> %rdx.minmax.select23, <16 x i32> undef, <16 x i32> - %rdx.minmax.cmp25 = icmp sgt <16 x i32> %rdx.minmax.select23, %rdx.shuf24 - %rdx.minmax.select26 = select <16 x i1> %rdx.minmax.cmp25, <16 x i32> %rdx.minmax.select23, <16 x i32> %rdx.shuf24 - %rdx.shuf27 = shufflevector <16 x i32> %rdx.minmax.select26, <16 x i32> undef, <16 x i32> - %rdx.minmax.cmp28 = icmp sgt <16 x i32> %rdx.minmax.select26, %rdx.shuf27 - %rdx.minmax.select29 = select <16 x i1> %rdx.minmax.cmp28, <16 x i32> %rdx.minmax.select26, <16 x i32> %rdx.shuf27 - %rdx.shuf30 = shufflevector <16 x i32> %rdx.minmax.select29, <16 x i32> undef, <16 x i32> - %rdx.minmax.cmp31 = icmp sgt <16 x i32> %rdx.minmax.select29, %rdx.shuf30 - %rdx.minmax.cmp31.elt = extractelement <16 x i1> %rdx.minmax.cmp31, i32 0 - %rdx.minmax.select29.elt = extractelement <16 x i32> %rdx.minmax.select29, i32 0 - %rdx.shuf30.elt = extractelement <16 x i32> %rdx.minmax.select29, i32 1 - %r = select i1 %rdx.minmax.cmp31.elt, i32 %rdx.minmax.select29.elt, i32 %rdx.shuf30.elt + %r = call i32 @llvm.experimental.vector.reduce.smax.i32.v16i32(<16 x i32> %arr.load) ret i32 %r } +declare i16 @llvm.experimental.vector.reduce.smin.i16.v16i16(<16 x i16>) + define i16 @oversized_smin_256(<16 x i16>* nocapture readonly %arr) { ; CHECK-LABEL: oversized_smin_256 ; CHECK: smin [[V0:v[0-9]+]].8h, {{v[0-9]+}}.8h, {{v[0-9]+}}.8h ; CHECK: sminv {{h[0-9]+}}, [[V0]] - %rdx.minmax.select = load <16 x i16>, <16 x i16>* %arr - %rdx.shuf = shufflevector <16 x i16> %rdx.minmax.select, <16 x i16> undef, <16 x i32> - %rdx.minmax.cmp22 = icmp slt <16 x i16> %rdx.minmax.select, %rdx.shuf - %rdx.minmax.select23 = select <16 x i1> %rdx.minmax.cmp22, <16 x i16> %rdx.minmax.select, <16 x i16> %rdx.shuf - %rdx.shuf24 = shufflevector <16 x i16> %rdx.minmax.select23, <16 x i16> undef, <16 x i32> - %rdx.minmax.cmp25 = icmp slt <16 x i16> %rdx.minmax.select23, %rdx.shuf24 - %rdx.minmax.select26 = select <16 x i1> %rdx.minmax.cmp25, <16 x i16> %rdx.minmax.select23, <16 x i16> %rdx.shuf24 - %rdx.shuf27 = shufflevector <16 x i16> %rdx.minmax.select26, <16 x i16> undef, <16 x i32> - %rdx.minmax.cmp28 = icmp slt <16 x i16> %rdx.minmax.select26, %rdx.shuf27 - %rdx.minmax.select29 = select <16 x i1> %rdx.minmax.cmp28, <16 x i16> %rdx.minmax.select26, <16 x i16> %rdx.shuf27 - %rdx.shuf30 = shufflevector <16 x i16> %rdx.minmax.select29, <16 x i16> undef, <16 x i32> - %rdx.minmax.cmp31 = icmp slt <16 x i16> %rdx.minmax.select29, %rdx.shuf30 - %rdx.minmax.cmp31.elt = extractelement <16 x i1> %rdx.minmax.cmp31, i32 0 - %rdx.minmax.select29.elt = extractelement <16 x i16> %rdx.minmax.select29, i32 0 - %rdx.shuf30.elt = extractelement <16 x i16> %rdx.minmax.select29, i32 1 - %r = select i1 %rdx.minmax.cmp31.elt, i16 %rdx.minmax.select29.elt, i16 %rdx.shuf30.elt + %arr.load = load <16 x i16>, <16 x i16>* %arr + %r = call i16 @llvm.experimental.vector.reduce.smin.i16.v16i16(<16 x i16> %arr.load) ret i16 %r } +declare i32 @llvm.experimental.vector.reduce.smin.i32.v16i32(<16 x i32>) + define i32 @oversized_smin_512(<16 x i32>* nocapture readonly %arr) { ; CHECK-LABEL: oversized_smin_512 ; CHECK: smin v @@ -491,20 +223,6 @@ define i32 @oversized_smin_512(<16 x i32>* nocapture readonly %arr) { ; CHECK-NEXT: smin [[V0:v[0-9]+]].4s, {{v[0-9]+}}.4s, {{v[0-9]+}}.4s ; CHECK-NEXT: sminv {{s[0-9]+}}, [[V0]] %arr.load = load <16 x i32>, <16 x i32>* %arr - %rdx.shuf = shufflevector <16 x i32> %arr.load, <16 x i32> undef, <16 x i32> - %rdx.minmax.cmp22 = icmp slt <16 x i32> %arr.load, %rdx.shuf - %rdx.minmax.select23 = select <16 x i1> %rdx.minmax.cmp22, <16 x i32> %arr.load, <16 x i32> %rdx.shuf - %rdx.shuf24 = shufflevector <16 x i32> %rdx.minmax.select23, <16 x i32> undef, <16 x i32> - %rdx.minmax.cmp25 = icmp slt <16 x i32> %rdx.minmax.select23, %rdx.shuf24 - %rdx.minmax.select26 = select <16 x i1> %rdx.minmax.cmp25, <16 x i32> %rdx.minmax.select23, <16 x i32> %rdx.shuf24 - %rdx.shuf27 = shufflevector <16 x i32> %rdx.minmax.select26, <16 x i32> undef, <16 x i32> - %rdx.minmax.cmp28 = icmp slt <16 x i32> %rdx.minmax.select26, %rdx.shuf27 - %rdx.minmax.select29 = select <16 x i1> %rdx.minmax.cmp28, <16 x i32> %rdx.minmax.select26, <16 x i32> %rdx.shuf27 - %rdx.shuf30 = shufflevector <16 x i32> %rdx.minmax.select29, <16 x i32> undef, <16 x i32> - %rdx.minmax.cmp31 = icmp slt <16 x i32> %rdx.minmax.select29, %rdx.shuf30 - %rdx.minmax.cmp31.elt = extractelement <16 x i1> %rdx.minmax.cmp31, i32 0 - %rdx.minmax.select29.elt = extractelement <16 x i32> %rdx.minmax.select29, i32 0 - %rdx.shuf30.elt = extractelement <16 x i32> %rdx.minmax.select29, i32 1 - %r = select i1 %rdx.minmax.cmp31.elt, i32 %rdx.minmax.select29.elt, i32 %rdx.shuf30.elt + %r = call i32 @llvm.experimental.vector.reduce.smin.i32.v16i32(<16 x i32> %arr.load) ret i32 %r } diff --git a/test/CodeGen/AArch64/arm64-vabs.ll b/test/CodeGen/AArch64/arm64-vabs.ll index c7b0c33550d0..ff7a0a8300e2 100644 --- a/test/CodeGen/AArch64/arm64-vabs.ll +++ b/test/CodeGen/AArch64/arm64-vabs.ll @@ -134,8 +134,10 @@ define <2 x i64> @uabdl2_2d(<4 x i32>* %A, <4 x i32>* %B) nounwind { ret <2 x i64> %tmp4 } -define i16 @uabdl8h_log2_shuffle(<16 x i8>* %a, <16 x i8>* %b) { -; CHECK-LABEL: uabdl8h_log2_shuffle +declare i16 @llvm.experimental.vector.reduce.add.i16.v16i16(<16 x i16>) + +define i16 @uabdl8h_rdx(<16 x i8>* %a, <16 x i8>* %b) { +; CHECK-LABEL: uabdl8h_rdx ; CHECK: uabdl2.8h ; CHECK: uabdl.8h %aload = load <16 x i8>, <16 x i8>* %a, align 1 @@ -146,20 +148,14 @@ define i16 @uabdl8h_log2_shuffle(<16 x i8>* %a, <16 x i8>* %b) { %abcmp = icmp slt <16 x i16> %abdiff, zeroinitializer %ababs = sub nsw <16 x i16> zeroinitializer, %abdiff %absel = select <16 x i1> %abcmp, <16 x i16> %ababs, <16 x i16> %abdiff - %rdx.shuf = shufflevector <16 x i16> %absel, <16 x i16> undef, <16 x i32> - %bin1.rdx = add <16 x i16> %absel, %rdx.shuf - %rdx.shufx = shufflevector <16 x i16> %bin1.rdx, <16 x i16> undef, <16 x i32> - %bin.rdx = add <16 x i16> %bin1.rdx, %rdx.shufx - %rdx.shuf136 = shufflevector <16 x i16> %bin.rdx, <16 x i16> undef, <16 x i32> - %bin.rdx137 = add <16 x i16> %bin.rdx, %rdx.shuf136 - %rdx.shuf138 = shufflevector <16 x i16> %bin.rdx137, <16 x i16> undef, <16 x i32> - %bin.rdx139 = add <16 x i16> %bin.rdx137, %rdx.shuf138 - %reduced_v = extractelement <16 x i16> %bin.rdx139, i16 0 + %reduced_v = call i16 @llvm.experimental.vector.reduce.add.i16.v16i16(<16 x i16> %absel) ret i16 %reduced_v } -define i32 @uabdl4s_log2_shuffle(<8 x i16>* %a, <8 x i16>* %b) { -; CHECK-LABEL: uabdl4s_log2_shuffle +declare i32 @llvm.experimental.vector.reduce.add.i32.v8i32(<8 x i32>) + +define i32 @uabdl4s_rdx(<8 x i16>* %a, <8 x i16>* %b) { +; CHECK-LABEL: uabdl4s_rdx ; CHECK: uabdl2.4s ; CHECK: uabdl.4s %aload = load <8 x i16>, <8 x i16>* %a, align 1 @@ -170,18 +166,14 @@ define i32 @uabdl4s_log2_shuffle(<8 x i16>* %a, <8 x i16>* %b) { %abcmp = icmp slt <8 x i32> %abdiff, zeroinitializer %ababs = sub nsw <8 x i32> zeroinitializer, %abdiff %absel = select <8 x i1> %abcmp, <8 x i32> %ababs, <8 x i32> %abdiff - %rdx.shuf = shufflevector <8 x i32> %absel, <8 x i32> undef, <8 x i32> - %bin.rdx = add <8 x i32> %absel, %rdx.shuf - %rdx.shuf136 = shufflevector <8 x i32> %bin.rdx, <8 x i32> undef, <8 x i32> - %bin.rdx137 = add <8 x i32> %bin.rdx, %rdx.shuf136 - %rdx.shuf138 = shufflevector <8 x i32> %bin.rdx137, <8 x i32> undef, <8 x i32> - %bin.rdx139 = add <8 x i32> %bin.rdx137, %rdx.shuf138 - %reduced_v = extractelement <8 x i32> %bin.rdx139, i32 0 + %reduced_v = call i32 @llvm.experimental.vector.reduce.add.i32.v8i32(<8 x i32> %absel) ret i32 %reduced_v } -define i64 @uabdl2d_log2_shuffle(<4 x i32>* %a, <4 x i32>* %b, i32 %h) { -; CHECK: uabdl2d_log2_shuffle +declare i64 @llvm.experimental.vector.reduce.add.i64.v4i64(<4 x i64>) + +define i64 @uabdl2d_rdx(<4 x i32>* %a, <4 x i32>* %b, i32 %h) { +; CHECK: uabdl2d_rdx ; CHECK: uabdl2.2d ; CHECK: uabdl.2d %aload = load <4 x i32>, <4 x i32>* %a, align 1 @@ -192,11 +184,7 @@ define i64 @uabdl2d_log2_shuffle(<4 x i32>* %a, <4 x i32>* %b, i32 %h) { %abcmp = icmp slt <4 x i64> %abdiff, zeroinitializer %ababs = sub nsw <4 x i64> zeroinitializer, %abdiff %absel = select <4 x i1> %abcmp, <4 x i64> %ababs, <4 x i64> %abdiff - %rdx.shuf136 = shufflevector <4 x i64> %absel, <4 x i64> undef, <4 x i32> - %bin.rdx137 = add <4 x i64> %absel, %rdx.shuf136 - %rdx.shuf138 = shufflevector <4 x i64> %bin.rdx137, <4 x i64> undef, <4 x i32> - %bin.rdx139 = add <4 x i64> %bin.rdx137, %rdx.shuf138 - %reduced_v = extractelement <4 x i64> %bin.rdx139, i16 0 + %reduced_v = call i64 @llvm.experimental.vector.reduce.add.i64.v4i64(<4 x i64> %absel) ret i64 %reduced_v } diff --git a/test/CodeGen/AArch64/ldst-zero.ll b/test/CodeGen/AArch64/ldst-zero.ll index 95b92ac70879..7d443a631f91 100644 --- a/test/CodeGen/AArch64/ldst-zero.ll +++ b/test/CodeGen/AArch64/ldst-zero.ll @@ -9,9 +9,9 @@ declare void @llvm.memset.p0i8.i64(i8* nocapture, i8, i64, i32, i1) ; Original test case which exhibited the bug define void @test1(%struct.tree_common* %t, i32 %code, i8* %type) { ; CHECK-LABEL: test1: -; CHECK: stp xzr, xzr, [x0, #8] -; CHECK: stp xzr, x2, [x0] -; CHECK: str w1, [x0, #16] +; CHECK-DAG: stp x2, xzr, [x0, #8] +; CHECK-DAG: str w1, [x0, #16] +; CHECK-DAG: str xzr, [x0] entry: %0 = bitcast %struct.tree_common* %t to i8* tail call void @llvm.memset.p0i8.i64(i8* %0, i8 0, i64 24, i32 8, i1 false) @@ -25,10 +25,8 @@ entry: ; Store to each struct element instead of using memset define void @test2(%struct.tree_common* %t, i32 %code, i8* %type) { ; CHECK-LABEL: test2: -; CHECK: stp xzr, xzr, [x0] -; CHECK: str wzr, [x0, #16] -; CHECK: str w1, [x0, #16] -; CHECK: str x2, [x0, #8] +; CHECK-DAG: str w1, [x0, #16] +; CHECK-DAG: stp xzr, x2, [x0] entry: %0 = getelementptr inbounds %struct.tree_common, %struct.tree_common* %t, i64 0, i32 0 %1 = getelementptr inbounds %struct.tree_common, %struct.tree_common* %t, i64 0, i32 1 @@ -44,9 +42,9 @@ entry: ; Vector store instead of memset define void @test3(%struct.tree_common* %t, i32 %code, i8* %type) { ; CHECK-LABEL: test3: -; CHECK: stp xzr, xzr, [x0, #8] -; CHECK: stp xzr, x2, [x0] -; CHECK: str w1, [x0, #16] +; CHECK-DAG: stp x2, xzr, [x0, #8] +; CHECK-DAG: str w1, [x0, #16] +; CHECK-DAG: str xzr, [x0] entry: %0 = bitcast %struct.tree_common* %t to <3 x i64>* store <3 x i64> zeroinitializer, <3 x i64>* %0, align 8 @@ -60,9 +58,8 @@ entry: ; Vector store, then store to vector elements define void @test4(<3 x i64>* %p, i64 %x, i64 %y) { ; CHECK-LABEL: test4: -; CHECK: stp xzr, xzr, [x0, #8] -; CHECK: stp xzr, x2, [x0] -; CHECK: str x1, [x0, #16] +; CHECK-DAG: stp x2, x1, [x0, #8] +; CHECK-DAG: str xzr, [x0] entry: store <3 x i64> zeroinitializer, <3 x i64>* %p, align 8 %0 = bitcast <3 x i64>* %p to i64* diff --git a/test/CodeGen/AArch64/misched-stp.ll b/test/CodeGen/AArch64/misched-stp.ll index 4ea481cae68e..1c9ea68834c2 100644 --- a/test/CodeGen/AArch64/misched-stp.ll +++ b/test/CodeGen/AArch64/misched-stp.ll @@ -1,20 +1,18 @@ ; REQUIRES: asserts -; RUN: llc < %s -mtriple=aarch64 -mcpu=cyclone -mattr=+use-aa -enable-misched -verify-misched -debug-only=misched -o - 2>&1 > /dev/null | FileCheck %s +; RUN: llc < %s -mtriple=aarch64 -mcpu=cyclone -mattr=+use-aa -enable-misched -verify-misched -o - | FileCheck %s ; Tests to check that the scheduler dependencies derived from alias analysis are ; correct when we have loads that have been split up so that they can later be ; merged into STP. -; CHECK: ********** MI Scheduling ********** -; CHECK: test_splat:BB#0 entry -; CHECK: SU({{[0-9]+}}): STRWui %vreg{{[0-9]+}}, %vreg{{[0-9]+}}, 3; mem:ST4[%3+8] -; CHECK: Successors: -; CHECK-NEXT: ord [[SU1:SU\([0-9]+\)]] -; CHECK: SU({{[0-9]+}}): STRWui %vreg{{[0-9]+}}, %vreg{{[0-9]+}}, 2; mem:ST4[%3+4] -; CHECK: Successors: -; CHECK-NEXT: ord [[SU2:SU\([0-9]+\)]] -; CHECK: [[SU1]]: STRWui %vreg{{[0-9]+}}, %vreg{{[0-9]+}}, 3; mem:ST4[%2] -; CHECK: [[SU2]]: STRWui %vreg{{[0-9]+}}, %vreg{{[0-9]+}}, 2; mem:ST4[%1] +; Now that overwritten stores are elided in SelectionDAG, dependencies +; are resolved and removed before MISCHED. Check that we have +; equivalent pair of stp calls as a baseline. + +; CHECK-LABEL: test_splat +; CHECK: ldr [[REG:w[0-9]+]], [x2] +; CHECK-DAG: stp w0, [[REG]], [x2, #12] +; CHECK-DAG: stp [[REG]], w1, [x2, #4] define void @test_splat(i32 %x, i32 %y, i32* %p) { entry: %val = load i32, i32* %p, align 4 @@ -35,16 +33,11 @@ entry: declare void @llvm.memset.p0i8.i64(i8* nocapture, i8, i64, i32, i1) %struct.tree_common = type { i8*, i8*, i32 } -; CHECK: ********** MI Scheduling ********** -; CHECK: test_zero:BB#0 entry -; CHECK: SU({{[0-9]+}}): STRXui %XZR, %vreg{{[0-9]+}}, 2; mem:ST8[%0+16] -; CHECK: Successors: -; CHECK-NEXT: ord [[SU3:SU\([0-9]+\)]] -; CHECK: SU({{[0-9]+}}): STRXui %XZR, %vreg{{[0-9]+}}, 1; mem:ST8[%0+8] -; CHECK: Successors: -; CHECK-NEXT: ord [[SU4:SU\([0-9]+\)]] -; CHECK: [[SU3]]: STRWui %vreg{{[0-9]+}}, %vreg{{[0-9]+}}, 4; mem:ST4[%code1] -; CHECK: [[SU4]]: STRXui %vreg{{[0-9]+}}, %vreg{{[0-9]+}}, 1; mem:ST8[%type2] +; CHECK-LABEL: test_zero +; CHECK-DAG: stp x2, xzr, [x0, #8] +; CHECK-DAG: str w1, [x0, #16] +; CHECK-DAG: str xzr, [x0] + define void @test_zero(%struct.tree_common* %t, i32 %code, i8* %type) { entry: %0 = bitcast %struct.tree_common* %t to i8* diff --git a/test/CodeGen/AMDGPU/fmax3.ll b/test/CodeGen/AMDGPU/fmax3.ll index a96eb5db9e2a..2e6d3f3c1e8f 100644 --- a/test/CodeGen/AMDGPU/fmax3.ll +++ b/test/CodeGen/AMDGPU/fmax3.ll @@ -1,39 +1,92 @@ -; RUN: llc -march=amdgcn -verify-machineinstrs < %s | FileCheck -check-prefix=SI %s -; RUN: llc -march=amdgcn -mcpu=tonga -mattr=-flat-for-global -verify-machineinstrs < %s | FileCheck -check-prefix=SI %s +; RUN: llc -march=amdgcn -verify-machineinstrs < %s | FileCheck -check-prefixes=GCN,SI %s +; RUN: llc -march=amdgcn -mcpu=tonga -mattr=-flat-for-global -verify-machineinstrs < %s | FileCheck -check-prefixes=GCN,VI %s +; RUN: llc -march=amdgcn -mcpu=gfx900 -mattr=-flat-for-global -verify-machineinstrs < %s | FileCheck -check-prefixes=GCN,GFX9 %s -declare float @llvm.maxnum.f32(float, float) nounwind readnone - -; SI-LABEL: {{^}}test_fmax3_olt_0: -; SI: buffer_load_dword [[REGC:v[0-9]+]] -; SI: buffer_load_dword [[REGB:v[0-9]+]] -; SI: buffer_load_dword [[REGA:v[0-9]+]] -; SI: v_max3_f32 [[RESULT:v[0-9]+]], [[REGC]], [[REGB]], [[REGA]] -; SI: buffer_store_dword [[RESULT]], -; SI: s_endpgm -define amdgpu_kernel void @test_fmax3_olt_0(float addrspace(1)* %out, float addrspace(1)* %aptr, float addrspace(1)* %bptr, float addrspace(1)* %cptr) nounwind { +; GCN-LABEL: {{^}}test_fmax3_olt_0_f32: +; GCN: buffer_load_dword [[REGC:v[0-9]+]] +; GCN: buffer_load_dword [[REGB:v[0-9]+]] +; GCN: buffer_load_dword [[REGA:v[0-9]+]] +; GCN: v_max3_f32 [[RESULT:v[0-9]+]], [[REGC]], [[REGB]], [[REGA]] +; GCN: buffer_store_dword [[RESULT]], +; GCN: s_endpgm +define amdgpu_kernel void @test_fmax3_olt_0_f32(float addrspace(1)* %out, float addrspace(1)* %aptr, float addrspace(1)* %bptr, float addrspace(1)* %cptr) #0 { %a = load volatile float, float addrspace(1)* %aptr, align 4 %b = load volatile float, float addrspace(1)* %bptr, align 4 %c = load volatile float, float addrspace(1)* %cptr, align 4 - %f0 = call float @llvm.maxnum.f32(float %a, float %b) nounwind readnone - %f1 = call float @llvm.maxnum.f32(float %f0, float %c) nounwind readnone + %f0 = call float @llvm.maxnum.f32(float %a, float %b) + %f1 = call float @llvm.maxnum.f32(float %f0, float %c) store float %f1, float addrspace(1)* %out, align 4 ret void } ; Commute operand of second fmax -; SI-LABEL: {{^}}test_fmax3_olt_1: -; SI: buffer_load_dword [[REGB:v[0-9]+]] -; SI: buffer_load_dword [[REGA:v[0-9]+]] -; SI: buffer_load_dword [[REGC:v[0-9]+]] -; SI: v_max3_f32 [[RESULT:v[0-9]+]], [[REGC]], [[REGB]], [[REGA]] -; SI: buffer_store_dword [[RESULT]], -; SI: s_endpgm -define amdgpu_kernel void @test_fmax3_olt_1(float addrspace(1)* %out, float addrspace(1)* %aptr, float addrspace(1)* %bptr, float addrspace(1)* %cptr) nounwind { +; GCN-LABEL: {{^}}test_fmax3_olt_1_f32: +; GCN: buffer_load_dword [[REGB:v[0-9]+]] +; GCN: buffer_load_dword [[REGA:v[0-9]+]] +; GCN: buffer_load_dword [[REGC:v[0-9]+]] +; GCN: v_max3_f32 [[RESULT:v[0-9]+]], [[REGC]], [[REGB]], [[REGA]] +; GCN: buffer_store_dword [[RESULT]], +; GCN: s_endpgm +define amdgpu_kernel void @test_fmax3_olt_1_f32(float addrspace(1)* %out, float addrspace(1)* %aptr, float addrspace(1)* %bptr, float addrspace(1)* %cptr) #0 { %a = load volatile float, float addrspace(1)* %aptr, align 4 %b = load volatile float, float addrspace(1)* %bptr, align 4 %c = load volatile float, float addrspace(1)* %cptr, align 4 - %f0 = call float @llvm.maxnum.f32(float %a, float %b) nounwind readnone - %f1 = call float @llvm.maxnum.f32(float %c, float %f0) nounwind readnone + %f0 = call float @llvm.maxnum.f32(float %a, float %b) + %f1 = call float @llvm.maxnum.f32(float %c, float %f0) store float %f1, float addrspace(1)* %out, align 4 ret void } + +; GCN-LABEL: {{^}}test_fmax3_olt_0_f16: +; GCN: buffer_load_ushort [[REGC:v[0-9]+]] +; GCN: buffer_load_ushort [[REGB:v[0-9]+]] +; GCN: buffer_load_ushort [[REGA:v[0-9]+]] + +; SI: v_max3_f32 [[RESULT_F32:v[0-9]+]], +; SI: v_cvt_f16_f32_e32 [[RESULT:v[0-9]+]], [[RESULT]] + +; VI: v_max_f16_e32 +; VI: v_max_f16_e32 [[RESULT:v[0-9]+]], + +; GFX9: v_max3_f16 [[RESULT:v[0-9]+]], [[REGC]], [[REGB]], [[REGA]] +; GCN: buffer_store_short [[RESULT]], +define amdgpu_kernel void @test_fmax3_olt_0_f16(half addrspace(1)* %out, half addrspace(1)* %aptr, half addrspace(1)* %bptr, half addrspace(1)* %cptr) #0 { + %a = load volatile half, half addrspace(1)* %aptr, align 2 + %b = load volatile half, half addrspace(1)* %bptr, align 2 + %c = load volatile half, half addrspace(1)* %cptr, align 2 + %f0 = call half @llvm.maxnum.f16(half %a, half %b) + %f1 = call half @llvm.maxnum.f16(half %f0, half %c) + store half %f1, half addrspace(1)* %out, align 2 + ret void +} + +; Commute operand of second fmax +; GCN-LABEL: {{^}}test_fmax3_olt_1_f16: +; GCN: buffer_load_ushort [[REGB:v[0-9]+]] +; GCN: buffer_load_ushort [[REGA:v[0-9]+]] +; GCN: buffer_load_ushort [[REGC:v[0-9]+]] + +; SI: v_max3_f32 [[RESULT_F32:v[0-9]+]], +; SI: v_cvt_f16_f32_e32 [[RESULT:v[0-9]+]], [[RESULT]] + +; VI: v_max_f16_e32 +; VI: v_max_f16_e32 [[RESULT:v[0-9]+]], + +; GFX9: v_max3_f16 [[RESULT:v[0-9]+]], [[REGC]], [[REGB]], [[REGA]] +; GCN: buffer_store_short [[RESULT]], +define amdgpu_kernel void @test_fmax3_olt_1_f16(half addrspace(1)* %out, half addrspace(1)* %aptr, half addrspace(1)* %bptr, half addrspace(1)* %cptr) #0 { + %a = load volatile half, half addrspace(1)* %aptr, align 2 + %b = load volatile half, half addrspace(1)* %bptr, align 2 + %c = load volatile half, half addrspace(1)* %cptr, align 2 + %f0 = call half @llvm.maxnum.f16(half %a, half %b) + %f1 = call half @llvm.maxnum.f16(half %c, half %f0) + store half %f1, half addrspace(1)* %out, align 2 + ret void +} + +declare i32 @llvm.amdgcn.workitem.id.x() #1 +declare float @llvm.maxnum.f32(float, float) #1 +declare half @llvm.maxnum.f16(half, half) #1 + +attributes #0 = { nounwind } +attributes #1 = { nounwind readnone speculatable } diff --git a/test/CodeGen/AMDGPU/fmin3.ll b/test/CodeGen/AMDGPU/fmin3.ll index 3183f77f090b..5fc5895c3ecb 100644 --- a/test/CodeGen/AMDGPU/fmin3.ll +++ b/test/CodeGen/AMDGPU/fmin3.ll @@ -1,40 +1,90 @@ -; RUN: llc -march=amdgcn -verify-machineinstrs < %s | FileCheck -check-prefix=SI %s -; RUN: llc -march=amdgcn -mcpu=tonga -mattr=-flat-for-global -verify-machineinstrs < %s | FileCheck -check-prefix=SI %s -; RUN: llc -march=amdgcn -mcpu=tonga -mattr=-flat-for-global -verify-machineinstrs < %s | FileCheck -check-prefix=SI %s +; RUN: llc -march=amdgcn -verify-machineinstrs < %s | FileCheck -check-prefixes=GCN,SI %s +; RUN: llc -march=amdgcn -mcpu=tonga -mattr=-flat-for-global -verify-machineinstrs < %s | FileCheck -check-prefixes=GCN,VI %s +; RUN: llc -march=amdgcn -mcpu=gfx900 -mattr=-flat-for-global -verify-machineinstrs < %s | FileCheck -check-prefixes=GCN,GFX9 %s -declare float @llvm.minnum.f32(float, float) nounwind readnone - -; SI-LABEL: {{^}}test_fmin3_olt_0: -; SI: buffer_load_dword [[REGC:v[0-9]+]] -; SI: buffer_load_dword [[REGB:v[0-9]+]] -; SI: buffer_load_dword [[REGA:v[0-9]+]] -; SI: v_min3_f32 [[RESULT:v[0-9]+]], [[REGC]], [[REGB]], [[REGA]] -; SI: buffer_store_dword [[RESULT]], -; SI: s_endpgm -define amdgpu_kernel void @test_fmin3_olt_0(float addrspace(1)* %out, float addrspace(1)* %aptr, float addrspace(1)* %bptr, float addrspace(1)* %cptr) nounwind { +; GCN-LABEL: {{^}}test_fmin3_olt_0_f32: +; GCN: buffer_load_dword [[REGC:v[0-9]+]] +; GCN: buffer_load_dword [[REGB:v[0-9]+]] +; GCN: buffer_load_dword [[REGA:v[0-9]+]] +; GCN: v_min3_f32 [[RESULT:v[0-9]+]], [[REGC]], [[REGB]], [[REGA]] +; GCN: buffer_store_dword [[RESULT]], +define amdgpu_kernel void @test_fmin3_olt_0_f32(float addrspace(1)* %out, float addrspace(1)* %aptr, float addrspace(1)* %bptr, float addrspace(1)* %cptr) #0 { %a = load volatile float, float addrspace(1)* %aptr, align 4 %b = load volatile float, float addrspace(1)* %bptr, align 4 %c = load volatile float, float addrspace(1)* %cptr, align 4 - %f0 = call float @llvm.minnum.f32(float %a, float %b) nounwind readnone - %f1 = call float @llvm.minnum.f32(float %f0, float %c) nounwind readnone + %f0 = call float @llvm.minnum.f32(float %a, float %b) + %f1 = call float @llvm.minnum.f32(float %f0, float %c) store float %f1, float addrspace(1)* %out, align 4 ret void } ; Commute operand of second fmin -; SI-LABEL: {{^}}test_fmin3_olt_1: -; SI: buffer_load_dword [[REGB:v[0-9]+]] -; SI: buffer_load_dword [[REGA:v[0-9]+]] -; SI: buffer_load_dword [[REGC:v[0-9]+]] -; SI: v_min3_f32 [[RESULT:v[0-9]+]], [[REGC]], [[REGB]], [[REGA]] -; SI: buffer_store_dword [[RESULT]], -; SI: s_endpgm -define amdgpu_kernel void @test_fmin3_olt_1(float addrspace(1)* %out, float addrspace(1)* %aptr, float addrspace(1)* %bptr, float addrspace(1)* %cptr) nounwind { +; GCN-LABEL: {{^}}test_fmin3_olt_1_f32: +; GCN: buffer_load_dword [[REGB:v[0-9]+]] +; GCN: buffer_load_dword [[REGA:v[0-9]+]] +; GCN: buffer_load_dword [[REGC:v[0-9]+]] +; GCN: v_min3_f32 [[RESULT:v[0-9]+]], [[REGC]], [[REGB]], [[REGA]] +; GCN: buffer_store_dword [[RESULT]], +define amdgpu_kernel void @test_fmin3_olt_1_f32(float addrspace(1)* %out, float addrspace(1)* %aptr, float addrspace(1)* %bptr, float addrspace(1)* %cptr) #0 { %a = load volatile float, float addrspace(1)* %aptr, align 4 %b = load volatile float, float addrspace(1)* %bptr, align 4 %c = load volatile float, float addrspace(1)* %cptr, align 4 - %f0 = call float @llvm.minnum.f32(float %a, float %b) nounwind readnone - %f1 = call float @llvm.minnum.f32(float %c, float %f0) nounwind readnone + %f0 = call float @llvm.minnum.f32(float %a, float %b) + %f1 = call float @llvm.minnum.f32(float %c, float %f0) store float %f1, float addrspace(1)* %out, align 4 ret void } + +; GCN-LABEL: {{^}}test_fmin3_olt_0_f16: +; GCN: buffer_load_ushort [[REGC:v[0-9]+]] +; GCN: buffer_load_ushort [[REGB:v[0-9]+]] +; GCN: buffer_load_ushort [[REGA:v[0-9]+]] + +; SI: v_min3_f32 [[RESULT_F32:v[0-9]+]], +; SI: v_cvt_f16_f32_e32 [[RESULT:v[0-9]+]], [[RESULT]] + +; VI: v_min_f16_e32 +; VI: v_min_f16_e32 [[RESULT:v[0-9]+]], + +; GFX9: v_min3_f16 [[RESULT:v[0-9]+]], [[REGC]], [[REGB]], [[REGA]] +; GCN: buffer_store_short [[RESULT]], +define amdgpu_kernel void @test_fmin3_olt_0_f16(half addrspace(1)* %out, half addrspace(1)* %aptr, half addrspace(1)* %bptr, half addrspace(1)* %cptr) #0 { + %a = load volatile half, half addrspace(1)* %aptr, align 2 + %b = load volatile half, half addrspace(1)* %bptr, align 2 + %c = load volatile half, half addrspace(1)* %cptr, align 2 + %f0 = call half @llvm.minnum.f16(half %a, half %b) + %f1 = call half @llvm.minnum.f16(half %f0, half %c) + store half %f1, half addrspace(1)* %out, align 2 + ret void +} + +; Commute operand of second fmin +; GCN-LABEL: {{^}}test_fmin3_olt_1_f16: +; GCN: buffer_load_ushort [[REGB:v[0-9]+]] +; GCN: buffer_load_ushort [[REGA:v[0-9]+]] +; GCN: buffer_load_ushort [[REGC:v[0-9]+]] + +; SI: v_min3_f32 [[RESULT_F32:v[0-9]+]], +; SI: v_cvt_f16_f32_e32 [[RESULT:v[0-9]+]], [[RESULT]] + +; VI: v_min_f16_e32 +; VI: v_min_f16_e32 [[RESULT:v[0-9]+]], + +; GFX9: v_min3_f16 [[RESULT:v[0-9]+]], [[REGC]], [[REGB]], [[REGA]] +; GCN: buffer_store_short [[RESULT]], +define amdgpu_kernel void @test_fmin3_olt_1_f16(half addrspace(1)* %out, half addrspace(1)* %aptr, half addrspace(1)* %bptr, half addrspace(1)* %cptr) #0 { + %a = load volatile half, half addrspace(1)* %aptr, align 2 + %b = load volatile half, half addrspace(1)* %bptr, align 2 + %c = load volatile half, half addrspace(1)* %cptr, align 2 + %f0 = call half @llvm.minnum.f16(half %a, half %b) + %f1 = call half @llvm.minnum.f16(half %c, half %f0) + store half %f1, half addrspace(1)* %out, align 2 + ret void +} + +declare i32 @llvm.amdgcn.workitem.id.x() #1 +declare float @llvm.minnum.f32(float, float) #1 +declare half @llvm.minnum.f16(half, half) #1 + +attributes #0 = { nounwind } +attributes #1 = { nounwind readnone speculatable } diff --git a/test/CodeGen/AMDGPU/global-constant.ll b/test/CodeGen/AMDGPU/global-constant.ll index 80acfcca7082..1898c8fb63ea 100644 --- a/test/CodeGen/AMDGPU/global-constant.ll +++ b/test/CodeGen/AMDGPU/global-constant.ll @@ -29,10 +29,10 @@ define amdgpu_kernel void @private_test(i32 %index, float addrspace(1)* %out) { %ptr = getelementptr [4 x float], [4 x float] addrspace(2) * @private1, i32 0, i32 %index %val = load float, float addrspace(2)* %ptr - store float %val, float addrspace(1)* %out + store volatile float %val, float addrspace(1)* %out %ptr2 = getelementptr [4 x float], [4 x float] addrspace(2) * @private2, i32 0, i32 %index %val2 = load float, float addrspace(2)* %ptr2 - store float %val2, float addrspace(1)* %out + store volatile float %val2, float addrspace(1)* %out ret void } diff --git a/test/CodeGen/AMDGPU/immv216.ll b/test/CodeGen/AMDGPU/immv216.ll index c15a30e3c540..96132d841997 100644 --- a/test/CodeGen/AMDGPU/immv216.ll +++ b/test/CodeGen/AMDGPU/immv216.ll @@ -288,9 +288,9 @@ define amdgpu_kernel void @commute_add_inline_imm_0.5_v2f16(<2 x half> addrspace } ; GCN-LABEL: {{^}}commute_add_literal_v2f16: -; GFX9: buffer_load_dword [[VAL:v[0-9]+]] -; GFX9: s_mov_b32 [[K:s[0-9]+]], 0x64006400 -; GFX9: v_pk_add_f16 [[REG:v[0-9]+]], [[K]], [[VAL]] +; GFX9-DAG: buffer_load_dword [[VAL:v[0-9]+]] +; GFX9-DAG: v_mov_b32_e32 [[K:v[0-9]+]], 0x6400{{$}} +; GFX9: v_pk_add_f16 [[REG:v[0-9]+]], [[K]], [[VAL]] op_sel_hi:[0,1]{{$}} ; GFX9: buffer_store_dword [[REG]] ; VI-DAG: v_mov_b32_e32 [[K:v[0-9]+]], 0x6400{{$}} diff --git a/test/CodeGen/AMDGPU/max3.ll b/test/CodeGen/AMDGPU/max3.ll index 4bb4fd46becd..46dcf8e340f4 100644 --- a/test/CodeGen/AMDGPU/max3.ll +++ b/test/CodeGen/AMDGPU/max3.ll @@ -1,41 +1,94 @@ -; RUN: llc -march=amdgcn < %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s +; RUN: llc -march=amdgcn < %s | FileCheck -check-prefixes=GCN,SI %s +; RUN: llc -march=amdgcn -mcpu=tonga < %s | FileCheck -check-prefixes=GCN,VI %s +; RUN: llc -march=amdgcn -mcpu=gfx900 < %s | FileCheck -check-prefixes=GCN,GFX9 %s -declare i32 @llvm.amdgcn.workitem.id.x() nounwind readnone - -; FUNC-LABEL: @v_test_imax3_sgt_i32 -; SI: v_max3_i32 -define amdgpu_kernel void @v_test_imax3_sgt_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %aptr, i32 addrspace(1)* %bptr, i32 addrspace(1)* %cptr) nounwind { - %tid = call i32 @llvm.amdgcn.workitem.id.x() nounwind readnone +; GCN-LABEL: {{^}}v_test_imax3_sgt_i32: +; GCN: v_max3_i32 +define amdgpu_kernel void @v_test_imax3_sgt_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %aptr, i32 addrspace(1)* %bptr, i32 addrspace(1)* %cptr) #0 { + %tid = call i32 @llvm.amdgcn.workitem.id.x() %gep0 = getelementptr i32, i32 addrspace(1)* %aptr, i32 %tid %gep1 = getelementptr i32, i32 addrspace(1)* %bptr, i32 %tid %gep2 = getelementptr i32, i32 addrspace(1)* %cptr, i32 %tid %outgep = getelementptr i32, i32 addrspace(1)* %out, i32 %tid - %a = load i32, i32 addrspace(1)* %gep0, align 4 - %b = load i32, i32 addrspace(1)* %gep1, align 4 - %c = load i32, i32 addrspace(1)* %gep2, align 4 + %a = load i32, i32 addrspace(1)* %gep0 + %b = load i32, i32 addrspace(1)* %gep1 + %c = load i32, i32 addrspace(1)* %gep2 %icmp0 = icmp sgt i32 %a, %b %i0 = select i1 %icmp0, i32 %a, i32 %b %icmp1 = icmp sgt i32 %i0, %c %i1 = select i1 %icmp1, i32 %i0, i32 %c - store i32 %i1, i32 addrspace(1)* %out, align 4 + store i32 %i1, i32 addrspace(1)* %out ret void } -; FUNC-LABEL: @v_test_umax3_ugt_i32 -; SI: v_max3_u32 -define amdgpu_kernel void @v_test_umax3_ugt_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %aptr, i32 addrspace(1)* %bptr, i32 addrspace(1)* %cptr) nounwind { - %tid = call i32 @llvm.amdgcn.workitem.id.x() nounwind readnone +; GCN-LABEL: {{^}}v_test_umax3_ugt_i32: +; GCN: v_max3_u32 +define amdgpu_kernel void @v_test_umax3_ugt_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %aptr, i32 addrspace(1)* %bptr, i32 addrspace(1)* %cptr) #0 { + %tid = call i32 @llvm.amdgcn.workitem.id.x() %gep0 = getelementptr i32, i32 addrspace(1)* %aptr, i32 %tid %gep1 = getelementptr i32, i32 addrspace(1)* %bptr, i32 %tid %gep2 = getelementptr i32, i32 addrspace(1)* %cptr, i32 %tid %outgep = getelementptr i32, i32 addrspace(1)* %out, i32 %tid - %a = load i32, i32 addrspace(1)* %gep0, align 4 - %b = load i32, i32 addrspace(1)* %gep1, align 4 - %c = load i32, i32 addrspace(1)* %gep2, align 4 + %a = load i32, i32 addrspace(1)* %gep0 + %b = load i32, i32 addrspace(1)* %gep1 + %c = load i32, i32 addrspace(1)* %gep2 %icmp0 = icmp ugt i32 %a, %b %i0 = select i1 %icmp0, i32 %a, i32 %b %icmp1 = icmp ugt i32 %i0, %c %i1 = select i1 %icmp1, i32 %i0, i32 %c - store i32 %i1, i32 addrspace(1)* %out, align 4 + store i32 %i1, i32 addrspace(1)* %out ret void } + +; GCN-LABEL: {{^}}v_test_imax3_sgt_i16: +; SI: v_max3_i32 + +; VI: v_max_i16 +; VI: v_max_i16 + +; GFX9: v_max3_i16 +define amdgpu_kernel void @v_test_imax3_sgt_i16(i16 addrspace(1)* %out, i16 addrspace(1)* %aptr, i16 addrspace(1)* %bptr, i16 addrspace(1)* %cptr) #0 { + %tid = call i32 @llvm.amdgcn.workitem.id.x() + %gep0 = getelementptr i16, i16 addrspace(1)* %aptr, i32 %tid + %gep1 = getelementptr i16, i16 addrspace(1)* %bptr, i32 %tid + %gep2 = getelementptr i16, i16 addrspace(1)* %cptr, i32 %tid + %outgep = getelementptr i16, i16 addrspace(1)* %out, i32 %tid + %a = load i16, i16 addrspace(1)* %gep0 + %b = load i16, i16 addrspace(1)* %gep1 + %c = load i16, i16 addrspace(1)* %gep2 + %icmp0 = icmp sgt i16 %a, %b + %i0 = select i1 %icmp0, i16 %a, i16 %b + %icmp1 = icmp sgt i16 %i0, %c + %i1 = select i1 %icmp1, i16 %i0, i16 %c + store i16 %i1, i16 addrspace(1)* %out + ret void +} + +; GCN-LABEL: {{^}}v_test_umax3_ugt_i16: +; SI: v_max3_u32 + +; VI: v_max_u16 +; VI: v_max_u16 + +; GFX9: v_max3_u16 +define amdgpu_kernel void @v_test_umax3_ugt_i16(i16 addrspace(1)* %out, i16 addrspace(1)* %aptr, i16 addrspace(1)* %bptr, i16 addrspace(1)* %cptr) #0 { + %tid = call i32 @llvm.amdgcn.workitem.id.x() + %gep0 = getelementptr i16, i16 addrspace(1)* %aptr, i32 %tid + %gep1 = getelementptr i16, i16 addrspace(1)* %bptr, i32 %tid + %gep2 = getelementptr i16, i16 addrspace(1)* %cptr, i32 %tid + %outgep = getelementptr i16, i16 addrspace(1)* %out, i32 %tid + %a = load i16, i16 addrspace(1)* %gep0 + %b = load i16, i16 addrspace(1)* %gep1 + %c = load i16, i16 addrspace(1)* %gep2 + %icmp0 = icmp ugt i16 %a, %b + %i0 = select i1 %icmp0, i16 %a, i16 %b + %icmp1 = icmp ugt i16 %i0, %c + %i1 = select i1 %icmp1, i16 %i0, i16 %c + store i16 %i1, i16 addrspace(1)* %out + ret void +} + +declare i32 @llvm.amdgcn.workitem.id.x() #1 + +attributes #0 = { nounwind } +attributes #1 = { nounwind readnone speculatable } diff --git a/test/CodeGen/AMDGPU/min3.ll b/test/CodeGen/AMDGPU/min3.ll index 59d5d2cdb1aa..e20fb81f2ecf 100644 --- a/test/CodeGen/AMDGPU/min3.ll +++ b/test/CodeGen/AMDGPU/min3.ll @@ -1,50 +1,50 @@ -; RUN: llc -march=amdgcn < %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s +; RUN: llc -march=amdgcn < %s | FileCheck -check-prefixes=GCN,SI %s +; RUN: llc -march=amdgcn -mcpu=tonga < %s | FileCheck -check-prefixes=GCN,VI %s +; RUN: llc -march=amdgcn -mcpu=gfx900 < %s | FileCheck -check-prefixes=GCN,GFX9 %s -declare i32 @llvm.amdgcn.workitem.id.x() nounwind readnone - -; FUNC-LABEL: @v_test_imin3_slt_i32 -; SI: v_min3_i32 -define amdgpu_kernel void @v_test_imin3_slt_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %aptr, i32 addrspace(1)* %bptr, i32 addrspace(1)* %cptr) nounwind { - %tid = call i32 @llvm.amdgcn.workitem.id.x() nounwind readnone +; GCN-LABEL: {{^}}v_test_imin3_slt_i32: +; GCN: v_min3_i32 +define amdgpu_kernel void @v_test_imin3_slt_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %aptr, i32 addrspace(1)* %bptr, i32 addrspace(1)* %cptr) #0 { + %tid = call i32 @llvm.amdgcn.workitem.id.x() %gep0 = getelementptr i32, i32 addrspace(1)* %aptr, i32 %tid %gep1 = getelementptr i32, i32 addrspace(1)* %bptr, i32 %tid %gep2 = getelementptr i32, i32 addrspace(1)* %cptr, i32 %tid %outgep = getelementptr i32, i32 addrspace(1)* %out, i32 %tid - %a = load i32, i32 addrspace(1)* %gep0, align 4 - %b = load i32, i32 addrspace(1)* %gep1, align 4 - %c = load i32, i32 addrspace(1)* %gep2, align 4 + %a = load i32, i32 addrspace(1)* %gep0 + %b = load i32, i32 addrspace(1)* %gep1 + %c = load i32, i32 addrspace(1)* %gep2 %icmp0 = icmp slt i32 %a, %b %i0 = select i1 %icmp0, i32 %a, i32 %b %icmp1 = icmp slt i32 %i0, %c %i1 = select i1 %icmp1, i32 %i0, i32 %c - store i32 %i1, i32 addrspace(1)* %outgep, align 4 + store i32 %i1, i32 addrspace(1)* %outgep ret void } -; FUNC-LABEL: @v_test_umin3_ult_i32 -; SI: v_min3_u32 -define amdgpu_kernel void @v_test_umin3_ult_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %aptr, i32 addrspace(1)* %bptr, i32 addrspace(1)* %cptr) nounwind { - %tid = call i32 @llvm.amdgcn.workitem.id.x() nounwind readnone +; GCN-LABEL: {{^}}v_test_umin3_ult_i32: +; GCN: v_min3_u32 +define amdgpu_kernel void @v_test_umin3_ult_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %aptr, i32 addrspace(1)* %bptr, i32 addrspace(1)* %cptr) #0 { + %tid = call i32 @llvm.amdgcn.workitem.id.x() %gep0 = getelementptr i32, i32 addrspace(1)* %aptr, i32 %tid %gep1 = getelementptr i32, i32 addrspace(1)* %bptr, i32 %tid %gep2 = getelementptr i32, i32 addrspace(1)* %cptr, i32 %tid %outgep = getelementptr i32, i32 addrspace(1)* %out, i32 %tid - %a = load i32, i32 addrspace(1)* %gep0, align 4 - %b = load i32, i32 addrspace(1)* %gep1, align 4 - %c = load i32, i32 addrspace(1)* %gep2, align 4 + %a = load i32, i32 addrspace(1)* %gep0 + %b = load i32, i32 addrspace(1)* %gep1 + %c = load i32, i32 addrspace(1)* %gep2 %icmp0 = icmp ult i32 %a, %b %i0 = select i1 %icmp0, i32 %a, i32 %b %icmp1 = icmp ult i32 %i0, %c %i1 = select i1 %icmp1, i32 %i0, i32 %c - store i32 %i1, i32 addrspace(1)* %outgep, align 4 + store i32 %i1, i32 addrspace(1)* %outgep ret void } -; FUNC-LABEL: @v_test_umin_umin_umin -; SI: v_min_i32 -; SI: v_min3_i32 -define amdgpu_kernel void @v_test_umin_umin_umin(i32 addrspace(1)* %out, i32 addrspace(1)* %aptr, i32 addrspace(1)* %bptr, i32 addrspace(1)* %cptr) nounwind { - %tid = call i32 @llvm.amdgcn.workitem.id.x() nounwind readnone +; GCN-LABEL: {{^}}v_test_umin_umin_umin: +; GCN: v_min_i32 +; GCN: v_min3_i32 +define amdgpu_kernel void @v_test_umin_umin_umin(i32 addrspace(1)* %out, i32 addrspace(1)* %aptr, i32 addrspace(1)* %bptr, i32 addrspace(1)* %cptr) #0 { + %tid = call i32 @llvm.amdgcn.workitem.id.x() %tid2 = mul i32 %tid, 2 %gep0 = getelementptr i32, i32 addrspace(1)* %aptr, i32 %tid %gep1 = getelementptr i32, i32 addrspace(1)* %bptr, i32 %tid @@ -57,10 +57,10 @@ define amdgpu_kernel void @v_test_umin_umin_umin(i32 addrspace(1)* %out, i32 add %outgep0 = getelementptr i32, i32 addrspace(1)* %out, i32 %tid %outgep1 = getelementptr i32, i32 addrspace(1)* %out, i32 %tid2 - %a = load i32, i32 addrspace(1)* %gep0, align 4 - %b = load i32, i32 addrspace(1)* %gep1, align 4 - %c = load i32, i32 addrspace(1)* %gep2, align 4 - %d = load i32, i32 addrspace(1)* %gep3, align 4 + %a = load i32, i32 addrspace(1)* %gep0 + %b = load i32, i32 addrspace(1)* %gep1 + %c = load i32, i32 addrspace(1)* %gep2 + %d = load i32, i32 addrspace(1)* %gep3 %icmp0 = icmp slt i32 %a, %b %i0 = select i1 %icmp0, i32 %a, i32 %b @@ -71,14 +71,14 @@ define amdgpu_kernel void @v_test_umin_umin_umin(i32 addrspace(1)* %out, i32 add %icmp2 = icmp slt i32 %i0, %i1 %i2 = select i1 %icmp2, i32 %i0, i32 %i1 - store i32 %i2, i32 addrspace(1)* %outgep1, align 4 + store i32 %i2, i32 addrspace(1)* %outgep1 ret void } -; FUNC-LABEL: @v_test_umin3_2_uses -; SI-NOT: v_min3 -define amdgpu_kernel void @v_test_umin3_2_uses(i32 addrspace(1)* %out, i32 addrspace(1)* %aptr, i32 addrspace(1)* %bptr, i32 addrspace(1)* %cptr) nounwind { - %tid = call i32 @llvm.amdgcn.workitem.id.x() nounwind readnone +; GCN-LABEL: {{^}}v_test_umin3_2_uses: +; GCN-NOT: v_min3 +define amdgpu_kernel void @v_test_umin3_2_uses(i32 addrspace(1)* %out, i32 addrspace(1)* %aptr, i32 addrspace(1)* %bptr, i32 addrspace(1)* %cptr) #0 { + %tid = call i32 @llvm.amdgcn.workitem.id.x() %tid2 = mul i32 %tid, 2 %gep0 = getelementptr i32, i32 addrspace(1)* %aptr, i32 %tid %gep1 = getelementptr i32, i32 addrspace(1)* %bptr, i32 %tid @@ -91,10 +91,10 @@ define amdgpu_kernel void @v_test_umin3_2_uses(i32 addrspace(1)* %out, i32 addrs %outgep0 = getelementptr i32, i32 addrspace(1)* %out, i32 %tid %outgep1 = getelementptr i32, i32 addrspace(1)* %out, i32 %tid2 - %a = load i32, i32 addrspace(1)* %gep0, align 4 - %b = load i32, i32 addrspace(1)* %gep1, align 4 - %c = load i32, i32 addrspace(1)* %gep2, align 4 - %d = load i32, i32 addrspace(1)* %gep3, align 4 + %a = load i32, i32 addrspace(1)* %gep0 + %b = load i32, i32 addrspace(1)* %gep1 + %c = load i32, i32 addrspace(1)* %gep2 + %d = load i32, i32 addrspace(1)* %gep3 %icmp0 = icmp slt i32 %a, %b %i0 = select i1 %icmp0, i32 %a, i32 %b @@ -105,7 +105,60 @@ define amdgpu_kernel void @v_test_umin3_2_uses(i32 addrspace(1)* %out, i32 addrs %icmp2 = icmp slt i32 %i0, %c %i2 = select i1 %icmp2, i32 %i0, i32 %c - store i32 %i2, i32 addrspace(1)* %outgep0, align 4 - store i32 %i0, i32 addrspace(1)* %outgep1, align 4 + store i32 %i2, i32 addrspace(1)* %outgep0 + store i32 %i0, i32 addrspace(1)* %outgep1 ret void } + +; GCN-LABEL: {{^}}v_test_imin3_slt_i16: +; SI: v_min3_i32 + +; VI: v_min_i16 +; VI: v_min_i16 + +; GFX9: v_min3_i16 +define amdgpu_kernel void @v_test_imin3_slt_i16(i16 addrspace(1)* %out, i16 addrspace(1)* %aptr, i16 addrspace(1)* %bptr, i16 addrspace(1)* %cptr) #0 { + %tid = call i32 @llvm.amdgcn.workitem.id.x() + %gep0 = getelementptr i16, i16 addrspace(1)* %aptr, i32 %tid + %gep1 = getelementptr i16, i16 addrspace(1)* %bptr, i32 %tid + %gep2 = getelementptr i16, i16 addrspace(1)* %cptr, i32 %tid + %outgep = getelementptr i16, i16 addrspace(1)* %out, i32 %tid + %a = load i16, i16 addrspace(1)* %gep0 + %b = load i16, i16 addrspace(1)* %gep1 + %c = load i16, i16 addrspace(1)* %gep2 + %icmp0 = icmp slt i16 %a, %b + %i0 = select i1 %icmp0, i16 %a, i16 %b + %icmp1 = icmp slt i16 %i0, %c + %i1 = select i1 %icmp1, i16 %i0, i16 %c + store i16 %i1, i16 addrspace(1)* %outgep + ret void +} + +; GCN-LABEL: {{^}}v_test_umin3_ult_i16: +; SI: v_min3_u32 + +; VI: v_min_u16 +; VI: v_min_u16 + +; GFX9: v_min3_u16 +define amdgpu_kernel void @v_test_umin3_ult_i16(i16 addrspace(1)* %out, i16 addrspace(1)* %aptr, i16 addrspace(1)* %bptr, i16 addrspace(1)* %cptr) #0 { + %tid = call i32 @llvm.amdgcn.workitem.id.x() + %gep0 = getelementptr i16, i16 addrspace(1)* %aptr, i32 %tid + %gep1 = getelementptr i16, i16 addrspace(1)* %bptr, i32 %tid + %gep2 = getelementptr i16, i16 addrspace(1)* %cptr, i32 %tid + %outgep = getelementptr i16, i16 addrspace(1)* %out, i32 %tid + %a = load i16, i16 addrspace(1)* %gep0 + %b = load i16, i16 addrspace(1)* %gep1 + %c = load i16, i16 addrspace(1)* %gep2 + %icmp0 = icmp ult i16 %a, %b + %i0 = select i1 %icmp0, i16 %a, i16 %b + %icmp1 = icmp ult i16 %i0, %c + %i1 = select i1 %icmp1, i16 %i0, i16 %c + store i16 %i1, i16 addrspace(1)* %outgep + ret void +} + +declare i32 @llvm.amdgcn.workitem.id.x() #1 + +attributes #0 = { nounwind } +attributes #1 = { nounwind readnone speculatable } diff --git a/test/CodeGen/AMDGPU/packed-op-sel.ll b/test/CodeGen/AMDGPU/packed-op-sel.ll new file mode 100644 index 000000000000..6ff0c54c33d0 --- /dev/null +++ b/test/CodeGen/AMDGPU/packed-op-sel.ll @@ -0,0 +1,266 @@ +; RUN: llc -march=amdgcn -mcpu=gfx900 -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefix=GCN -check-prefix=GFX9 %s + +; GCN-LABEL: {{^}}fma_vector_vector_scalar_lo: +; GCN: ds_read_b32 [[VEC0:v[0-9]+]] +; GCN: ds_read_b32 [[VEC1:v[0-9]+]] +; GCN: ds_read_u16 [[SCALAR0:v[0-9]+]] + +; GCN-NOT: pack +; GCN-NOT: and +; GCN-NOT: shl +; GCN-NOT: or + +; GCN: v_pk_fma_f16 v{{[0-9]+}}, [[VEC0]], [[VEC1]], [[SCALAR0]] op_sel_hi:[1,1,0]{{$}} +define amdgpu_kernel void @fma_vector_vector_scalar_lo(<2 x half> addrspace(1)* %out, <2 x half> addrspace(3)* %lds, half addrspace(3)* %arg2) #0 { +bb: + %lds.gep1 = getelementptr inbounds <2 x half>, <2 x half> addrspace(3)* %lds, i32 1 + + %vec0 = load volatile <2 x half>, <2 x half> addrspace(3)* %lds, align 4 + %vec1 = load volatile <2 x half>, <2 x half> addrspace(3)* %lds.gep1, align 4 + %scalar0 = load volatile half, half addrspace(3)* %arg2, align 2 + + %scalar0.vec = insertelement <2 x half> undef, half %scalar0, i32 0 + %scalar0.broadcast = shufflevector <2 x half> %scalar0.vec, <2 x half> undef, <2 x i32> zeroinitializer + + %result = tail call <2 x half> @llvm.fma.v2f16(<2 x half> %vec0, <2 x half> %vec1, <2 x half> %scalar0.broadcast) + store <2 x half> %result, <2 x half> addrspace(1)* %out, align 4 + ret void +} + +; Apply fneg to broadcasted vector +; GCN-LABEL: {{^}}fma_vector_vector_neg_broadcast_scalar_lo: +; GCN: ds_read_b32 [[VEC0:v[0-9]+]] +; GCN: ds_read_b32 [[VEC1:v[0-9]+]] +; GCN: ds_read_u16 [[SCALAR0:v[0-9]+]] + +; GCN-NOT: pack +; GCN-NOT: and +; GCN-NOT: shl +; GCN-NOT: or + +; GCN: v_pk_fma_f16 v{{[0-9]+}}, [[VEC0]], [[VEC1]], [[SCALAR0]] op_sel_hi:[1,1,0] neg_lo:[0,0,1] neg_hi:[0,0,1]{{$}} +define amdgpu_kernel void @fma_vector_vector_neg_broadcast_scalar_lo(<2 x half> addrspace(1)* %out, <2 x half> addrspace(3)* %lds, half addrspace(3)* %arg2) #0 { +bb: + %lds.gep1 = getelementptr inbounds <2 x half>, <2 x half> addrspace(3)* %lds, i32 1 + + %vec0 = load volatile <2 x half>, <2 x half> addrspace(3)* %lds, align 4 + %vec1 = load volatile <2 x half>, <2 x half> addrspace(3)* %lds.gep1, align 4 + %scalar0 = load volatile half, half addrspace(3)* %arg2, align 2 + + %scalar0.vec = insertelement <2 x half> undef, half %scalar0, i32 0 + %scalar0.broadcast = shufflevector <2 x half> %scalar0.vec, <2 x half> undef, <2 x i32> zeroinitializer + %neg.scalar0.broadcast = fsub <2 x half> , %scalar0.broadcast + + %result = tail call <2 x half> @llvm.fma.v2f16(<2 x half> %vec0, <2 x half> %vec1, <2 x half> %neg.scalar0.broadcast) + store <2 x half> %result, <2 x half> addrspace(1)* %out, align 4 + ret void +} + +; Apply fneg before broadcast +; GCN-LABEL: {{^}}fma_vector_vector_neg_scalar_lo: +; GCN: ds_read_b32 [[VEC0:v[0-9]+]] +; GCN: ds_read_b32 [[VEC1:v[0-9]+]] +; GCN: ds_read_u16 [[SCALAR0:v[0-9]+]] + +; GCN-NOT: pack +; GCN-NOT: and +; GCN-NOT: shl +; GCN-NOT: or + +; GCN: v_pk_fma_f16 v{{[0-9]+}}, [[VEC0]], [[VEC1]], [[SCALAR0]] op_sel_hi:[1,1,0] neg_lo:[0,0,1] neg_hi:[0,0,1]{{$}} +define amdgpu_kernel void @fma_vector_vector_neg_scalar_lo(<2 x half> addrspace(1)* %out, <2 x half> addrspace(3)* %lds, half addrspace(3)* %arg2) #0 { +bb: + %lds.gep1 = getelementptr inbounds <2 x half>, <2 x half> addrspace(3)* %lds, i32 1 + + %vec0 = load volatile <2 x half>, <2 x half> addrspace(3)* %lds, align 4 + %vec1 = load volatile <2 x half>, <2 x half> addrspace(3)* %lds.gep1, align 4 + %scalar0 = load volatile half, half addrspace(3)* %arg2, align 2 + + %neg.scalar0 = fsub half -0.0, %scalar0 + %neg.scalar0.vec = insertelement <2 x half> undef, half %neg.scalar0, i32 0 + %neg.scalar0.broadcast = shufflevector <2 x half> %neg.scalar0.vec, <2 x half> undef, <2 x i32> zeroinitializer + + %result = tail call <2 x half> @llvm.fma.v2f16(<2 x half> %vec0, <2 x half> %vec1, <2 x half> %neg.scalar0.broadcast) + store <2 x half> %result, <2 x half> addrspace(1)* %out, align 4 + ret void +} + +; Apply fneg before and after broadcast, and should cancel out. +; GCN-LABEL: {{^}}fma_vector_vector_neg_broadcast_neg_scalar_lo: +; GCN: ds_read_b32 [[VEC0:v[0-9]+]] +; GCN: ds_read_b32 [[VEC1:v[0-9]+]] +; GCN: ds_read_u16 [[SCALAR0:v[0-9]+]] + +; GCN-NOT: pack +; GCN-NOT: and +; GCN-NOT: shl +; GCN-NOT: or + +; GCN: v_pk_fma_f16 v{{[0-9]+}}, [[VEC0]], [[VEC1]], [[SCALAR0]] op_sel_hi:[1,1,0]{{$}} +define amdgpu_kernel void @fma_vector_vector_neg_broadcast_neg_scalar_lo(<2 x half> addrspace(1)* %out, <2 x half> addrspace(3)* %lds, half addrspace(3)* %arg2) #0 { +bb: + %lds.gep1 = getelementptr inbounds <2 x half>, <2 x half> addrspace(3)* %lds, i32 1 + + %vec0 = load volatile <2 x half>, <2 x half> addrspace(3)* %lds, align 4 + %vec1 = load volatile <2 x half>, <2 x half> addrspace(3)* %lds.gep1, align 4 + %scalar0 = load volatile half, half addrspace(3)* %arg2, align 2 + + %neg.scalar0 = fsub half -0.0, %scalar0 + %neg.scalar0.vec = insertelement <2 x half> undef, half %neg.scalar0, i32 0 + %neg.scalar0.broadcast = shufflevector <2 x half> %neg.scalar0.vec, <2 x half> undef, <2 x i32> zeroinitializer + %neg.neg.scalar0.broadcast = fsub <2 x half> , %neg.scalar0.broadcast + + %result = tail call <2 x half> @llvm.fma.v2f16(<2 x half> %vec0, <2 x half> %vec1, <2 x half> %neg.neg.scalar0.broadcast) + store <2 x half> %result, <2 x half> addrspace(1)* %out, align 4 + ret void +} + +; Add scalar, but negate low component +; GCN-LABEL: {{^}}fma_vector_vector_scalar_neg_lo: +; GCN: ds_read_b32 [[VEC0:v[0-9]+]] +; GCN: ds_read_b32 [[VEC1:v[0-9]+]] +; GCN: ds_read_u16 [[SCALAR0:v[0-9]+]] + +; GCN-NOT: pack +; GCN-NOT: and +; GCN-NOT: shl +; GCN-NOT: or + +; GCN: v_pk_fma_f16 v{{[0-9]+}}, [[VEC0]], [[VEC1]], [[SCALAR0]] op_sel_hi:[1,1,0] neg_lo:[0,0,1]{{$}} +define amdgpu_kernel void @fma_vector_vector_scalar_neg_lo(<2 x half> addrspace(1)* %out, <2 x half> addrspace(3)* %lds, half addrspace(3)* %arg2) #0 { +bb: + %lds.gep1 = getelementptr inbounds <2 x half>, <2 x half> addrspace(3)* %lds, i32 1 + + %vec0 = load volatile <2 x half>, <2 x half> addrspace(3)* %lds, align 4 + %vec1 = load volatile <2 x half>, <2 x half> addrspace(3)* %lds.gep1, align 4 + %scalar0 = load volatile half, half addrspace(3)* %arg2, align 2 + + %neg.scalar0 = fsub half -0.0, %scalar0 + %neg.scalar0.vec = insertelement <2 x half> undef, half %neg.scalar0, i32 0 + %neg.scalar0.scalar0 = insertelement <2 x half> %neg.scalar0.vec, half %scalar0, i32 1 + %result = tail call <2 x half> @llvm.fma.v2f16(<2 x half> %vec0, <2 x half> %vec1, <2 x half> %neg.scalar0.scalar0) + store <2 x half> %result, <2 x half> addrspace(1)* %out, align 4 + ret void +} + +; Add scalar, but negate high component +; GCN-LABEL: {{^}}fma_vector_vector_scalar_neg_hi: +; GCN: ds_read_b32 [[VEC0:v[0-9]+]] +; GCN: ds_read_b32 [[VEC1:v[0-9]+]] +; GCN: ds_read_u16 [[SCALAR0:v[0-9]+]] + +; GCN-NOT: pack +; GCN-NOT: and +; GCN-NOT: shl +; GCN-NOT: or + +; GCN: v_pk_fma_f16 v{{[0-9]+}}, [[VEC0]], [[VEC1]], [[SCALAR0]] op_sel_hi:[1,1,0] neg_hi:[0,0,1]{{$}} +define amdgpu_kernel void @fma_vector_vector_scalar_neg_hi(<2 x half> addrspace(1)* %out, <2 x half> addrspace(3)* %lds, half addrspace(3)* %arg2) #0 { +bb: + %lds.gep1 = getelementptr inbounds <2 x half>, <2 x half> addrspace(3)* %lds, i32 1 + + %vec0 = load volatile <2 x half>, <2 x half> addrspace(3)* %lds, align 4 + %vec1 = load volatile <2 x half>, <2 x half> addrspace(3)* %lds.gep1, align 4 + %scalar0 = load volatile half, half addrspace(3)* %arg2, align 2 + + %neg.scalar0 = fsub half -0.0, %scalar0 + %neg.scalar0.vec = insertelement <2 x half> undef, half %scalar0, i32 0 + %scalar0.neg.scalar0 = insertelement <2 x half> %neg.scalar0.vec, half %neg.scalar0, i32 1 + %result = tail call <2 x half> @llvm.fma.v2f16(<2 x half> %vec0, <2 x half> %vec1, <2 x half> %scalar0.neg.scalar0) + store <2 x half> %result, <2 x half> addrspace(1)* %out, align 4 + ret void +} + +; Apply fneg before broadcast with bitcast +; GCN-LABEL: {{^}}add_vector_neg_bitcast_scalar_lo: +; GCN: ds_read_b32 [[VEC0:v[0-9]+]] +; GCN: ds_read_u16 [[SCALAR0:v[0-9]+]] + +; GCN-NOT: pack +; GCN-NOT: and +; GCN-NOT: shl +; GCN-NOT: or + +; GCN: v_xor_b32_e32 [[NEG_SCALAR0:v[0-9]+]], 0x8000, [[SCALAR0]] +; GCN-NEXT: v_pk_add_u16 v{{[0-9]+}}, [[VEC0]], [[NEG_SCALAR0]] op_sel_hi:[1,0]{{$}} +define amdgpu_kernel void @add_vector_neg_bitcast_scalar_lo(<2 x i16> addrspace(1)* %out, <2 x i16> addrspace(3)* %lds, half addrspace(3)* %arg2) #0 { +bb: + %vec0 = load volatile <2 x i16>, <2 x i16> addrspace(3)* %lds, align 4 + %scalar0 = load volatile half, half addrspace(3)* %arg2, align 2 + %neg.scalar0 = fsub half -0.0, %scalar0 + %neg.scalar0.bc = bitcast half %neg.scalar0 to i16 + + %neg.scalar0.vec = insertelement <2 x i16> undef, i16 %neg.scalar0.bc, i32 0 + %neg.scalar0.broadcast = shufflevector <2 x i16> %neg.scalar0.vec, <2 x i16> undef, <2 x i32> zeroinitializer + + %result = add <2 x i16> %vec0, %neg.scalar0.broadcast + store <2 x i16> %result, <2 x i16> addrspace(1)* %out, align 4 + ret void +} + +; GCN-LABEL: {{^}}fma_vector_vector_scalar_lo_neg_scalar_hi: +; GCN: ds_read_b32 [[VEC0:v[0-9]+]] +; GCN: ds_read_b32 [[VEC1:v[0-9]+]] +; GCN: ds_read_u16 [[SCALAR0:v[0-9]+]] +; GCN: ds_read_u16 [[SCALAR1:v[0-9]+]] + +; FIXME: Remove and +; GCN: v_and_b32_e32 [[SCALAR0]], 0xffff, [[SCALAR0]] +; GCN: v_xor_b32_e32 [[SCALAR1]], 0x8000, [[SCALAR1]] +; GCN: v_lshl_or_b32 [[PACKED:v[0-9]+]], [[SCALAR1]], 16, [[SCALAR0]] + +; GCN: v_pk_fma_f16 v{{[0-9]+}}, [[VEC0]], [[VEC1]], [[PACKED]]{{$}} +define amdgpu_kernel void @fma_vector_vector_scalar_lo_neg_scalar_hi(<2 x half> addrspace(1)* %out, <2 x half> addrspace(3)* %lds, half addrspace(3)* %arg2) #0 { +bb: + %lds.gep1 = getelementptr inbounds <2 x half>, <2 x half> addrspace(3)* %lds, i32 1 + %arg2.gep = getelementptr inbounds half, half addrspace(3)* %arg2, i32 2 + + %vec0 = load volatile <2 x half>, <2 x half> addrspace(3)* %lds, align 4 + %vec1 = load volatile <2 x half>, <2 x half> addrspace(3)* %lds.gep1, align 4 + + %scalar0 = load volatile half, half addrspace(3)* %arg2, align 2 + %scalar1 = load volatile half, half addrspace(3)* %arg2.gep, align 2 + + %neg.scalar1 = fsub half -0.0, %scalar1 + %vec.ins0 = insertelement <2 x half> undef, half %scalar0, i32 0 + %vec2 = insertelement <2 x half> %vec.ins0, half %neg.scalar1, i32 1 + %result = tail call <2 x half> @llvm.fma.v2f16(<2 x half> %vec0, <2 x half> %vec1, <2 x half> %vec2) + store <2 x half> %result, <2 x half> addrspace(1)* %out, align 4 + ret void +} + +; GCN-LABEL: {{^}}fma_vector_vector_neg_scalar_lo_scalar_hi: +; GCN: ds_read_b32 [[VEC0:v[0-9]+]] +; GCN: ds_read_b32 [[VEC1:v[0-9]+]] +; GCN: ds_read_u16 [[SCALAR0:v[0-9]+]] +; GCN: ds_read_u16 [[SCALAR1:v[0-9]+]] + +; FIXME: Remove and +; GCN: v_and_b32_e32 [[SCALAR0]], 0xffff, [[SCALAR0]] +; GCN: v_lshl_or_b32 [[PACKED:v[0-9]+]], [[SCALAR1]], 16, [[SCALAR0]] + +; GCN: v_pk_fma_f16 v{{[0-9]+}}, [[VEC0]], [[VEC1]], [[PACKED]] neg_lo:[0,0,1] neg_hi:[0,0,1]{{$}} +define amdgpu_kernel void @fma_vector_vector_neg_scalar_lo_scalar_hi(<2 x half> addrspace(1)* %out, <2 x half> addrspace(3)* %lds, half addrspace(3)* %arg2) #0 { +bb: + %lds.gep1 = getelementptr inbounds <2 x half>, <2 x half> addrspace(3)* %lds, i32 1 + %arg2.gep = getelementptr inbounds half, half addrspace(3)* %arg2, i32 2 + + %vec0 = load volatile <2 x half>, <2 x half> addrspace(3)* %lds, align 4 + %vec1 = load volatile <2 x half>, <2 x half> addrspace(3)* %lds.gep1, align 4 + + %scalar0 = load volatile half, half addrspace(3)* %arg2, align 2 + %scalar1 = load volatile half, half addrspace(3)* %arg2.gep, align 2 + + %vec.ins0 = insertelement <2 x half> undef, half %scalar0, i32 0 + %vec2 = insertelement <2 x half> %vec.ins0, half %scalar1, i32 1 + %neg.vec2 = fsub <2 x half> , %vec2 + + %result = tail call <2 x half> @llvm.fma.v2f16(<2 x half> %vec0, <2 x half> %vec1, <2 x half> %neg.vec2) + store <2 x half> %result, <2 x half> addrspace(1)* %out, align 4 + ret void +} + +declare <2 x half> @llvm.fma.v2f16(<2 x half>, <2 x half>, <2 x half>) #1 + +attributes #0 = { nounwind } +attributes #1 = { nounwind readnone } diff --git a/test/CodeGen/ARM/2011-02-04-AntidepMultidef.ll b/test/CodeGen/ARM/2011-02-04-AntidepMultidef.ll index 8403dd991360..777eccb00b02 100644 --- a/test/CodeGen/ARM/2011-02-04-AntidepMultidef.ll +++ b/test/CodeGen/ARM/2011-02-04-AntidepMultidef.ll @@ -20,7 +20,7 @@ bb3: ; preds = %bb, %entry bb8: ; preds = %bb3 %1 = getelementptr inbounds i8, i8* %0, i32 0 - store i8 0, i8* %1, align 1 + store volatile i8 0, i8* %1, align 1 %2 = call i32 @ptou() nounwind ; CHECK: umull [[REGISTER:lr|r[0-9]+]], ; CHECK-NOT: [[REGISTER]], @@ -35,7 +35,7 @@ bb8: ; preds = %bb3 %7 = or i8 %6, 48 %8 = add i8 %6, 87 %iftmp.5.0.1 = select i1 %5, i8 %7, i8 %8 - store i8 %iftmp.5.0.1, i8* %p8, align 1 + store volatile i8 %iftmp.5.0.1, i8* %p8, align 1 ; CHECK: umull [[REGISTER:lr|r[0-9]+]], ; CHECK-NOT: [[REGISTER]], ; CHECK: {{lr|r[0-9]+}}, {{lr|r[0-9]+$}} @@ -49,7 +49,7 @@ bb8: ; preds = %bb3 %13 = or i8 %12, 48 %14 = add i8 %12, 87 %iftmp.5.0.2 = select i1 %11, i8 %13, i8 %14 - store i8 %iftmp.5.0.2, i8* %p8, align 1 + store volatile i8 %iftmp.5.0.2, i8* %p8, align 1 ; CHECK: umull [[REGISTER:lr|r[0-9]+]], ; CHECK-NOT: [[REGISTER]], ; CHECK: {{lr|r[0-9]+}}, {{lr|r[0-9]+$}} @@ -63,7 +63,7 @@ bb8: ; preds = %bb3 %19 = or i8 %18, 48 %20 = add i8 %18, 87 %iftmp.5.0.4 = select i1 %17, i8 %19, i8 %20 - store i8 %iftmp.5.0.4, i8* null, align 1 + store volatile i8 %iftmp.5.0.4, i8* null, align 1 ; CHECK: umull [[REGISTER:lr|r[0-9]+]], ; CHECK-NOT: [[REGISTER]], ; CHECK: {{lr|r[0-9]+}}, {{lr|r[0-9]+$}} @@ -74,7 +74,7 @@ bb8: ; preds = %bb3 %22 = urem i32 %21, 10 %23 = icmp ult i32 %22, 10 %iftmp.5.0.5 = select i1 %23, i8 0, i8 %val8 - store i8 %iftmp.5.0.5, i8* %p8, align 1 + store volatile i8 %iftmp.5.0.5, i8* %p8, align 1 ; CHECK: umull [[REGISTER:lr|r[0-9]+]], ; CHECK-NOT: [[REGISTER]], ; CHECK: {{lr|r[0-9]+}}, {{lr|r[0-9]+$}} @@ -88,7 +88,7 @@ bb8: ; preds = %bb3 %28 = or i8 %27, 48 %29 = add i8 %27, 87 %iftmp.5.0.6 = select i1 %26, i8 %28, i8 %29 - store i8 %iftmp.5.0.6, i8* %p8, align 1 + store volatile i8 %iftmp.5.0.6, i8* %p8, align 1 ; CHECK: umull [[REGISTER:lr|r[0-9]+]], ; CHECK-NOT: [[REGISTER]], ; CHECK: {{lr|r[0-9]+}}, {{lr|r[0-9]+$}} @@ -102,7 +102,7 @@ bb8: ; preds = %bb3 %34 = or i8 %33, 48 %35 = add i8 %33, 87 %iftmp.5.0.7 = select i1 %32, i8 %34, i8 %35 - store i8 %iftmp.5.0.7, i8* %p8, align 1 + store volatile i8 %iftmp.5.0.7, i8* %p8, align 1 ; CHECK: umull [[REGISTER:lr|r[0-9]+]], ; CHECK-NOT: [[REGISTER]], ; CHECK: {{lr|r[0-9]+}}, {{lr|r[0-9]+$}} @@ -116,7 +116,7 @@ bb8: ; preds = %bb3 %40 = or i8 %39, 48 %41 = add i8 %39, 87 %iftmp.5.0.8 = select i1 %38, i8 %40, i8 %41 - store i8 %iftmp.5.0.8, i8* null, align 1 + store volatile i8 %iftmp.5.0.8, i8* null, align 1 br label %bb46 bb46: ; preds = %bb3 diff --git a/test/CodeGen/ARM/2012-10-04-AAPCS-byval-align8.ll b/test/CodeGen/ARM/2012-10-04-AAPCS-byval-align8.ll index 2a5af6199a34..954860219d19 100644 --- a/test/CodeGen/ARM/2012-10-04-AAPCS-byval-align8.ll +++ b/test/CodeGen/ARM/2012-10-04-AAPCS-byval-align8.ll @@ -13,7 +13,7 @@ entry: ; CHECK: sub sp, sp, #12 ; CHECK: sub sp, sp, #4 ; CHECK: add r0, sp, #4 -; CHECK: stm sp, {r0, r1, r2, r3} +; CHECK: stmib sp, {r1, r2, r3} %g = alloca i8* %g1 = bitcast i8** %g to i8* call void @llvm.va_start(i8* %g1) diff --git a/test/CodeGen/ARM/dag-combine-ldst.ll b/test/CodeGen/ARM/dag-combine-ldst.ll index c1960ee6c6e9..077754ef013d 100644 --- a/test/CodeGen/ARM/dag-combine-ldst.ll +++ b/test/CodeGen/ARM/dag-combine-ldst.ll @@ -8,7 +8,7 @@ ; CHECK-LABEL: {{^}}main ; CHECK: mov [[TMP:r[0-9]+]], #0 ; CHECK-NEXT: str [[TMP]], [sp, #4] -; CHECK-NEXT: str [[TMP]], [sp] +; CHECK_O0: str [[TMP]], [sp] ; CHECK_O0: ldr [[TMP:r[0-9]+]], [sp] ; CHECK_O0-NEXT: add [[TMP]], [[TMP]], #2 ; CHECK_O1-NOT: ldr [[TMP:r[0-9]+]], [sp] diff --git a/test/CodeGen/MSP430/vararg.ll b/test/CodeGen/MSP430/vararg.ll index 6c8bceff5de9..a708b89cbd8f 100644 --- a/test/CodeGen/MSP430/vararg.ll +++ b/test/CodeGen/MSP430/vararg.ll @@ -25,7 +25,6 @@ define i16 @va_arg(i8* %vl) nounwind { entry: ; CHECK-LABEL: va_arg: %vl.addr = alloca i8*, align 2 -; CHECK: mov.w r12, 0(r1) store i8* %vl, i8** %vl.addr, align 2 ; CHECK: mov.w r12, [[REG:r[0-9]+]] ; CHECK-NEXT: add.w #2, [[REG]] diff --git a/test/CodeGen/Mips/msa/bmzi_bmnzi.ll b/test/CodeGen/Mips/msa/bmzi_bmnzi.ll index d1cb3c348c73..de62dcd69403 100644 --- a/test/CodeGen/Mips/msa/bmzi_bmnzi.ll +++ b/test/CodeGen/Mips/msa/bmzi_bmnzi.ll @@ -9,9 +9,9 @@ entry: %0 = load <16 x i8>, <16 x i8>* @llvm_mips_bmnzi_b_ARG1 %1 = load <16 x i8>, <16 x i8>* @llvm_mips_bmnzi_b_ARG2 %2 = tail call <16 x i8> @llvm.mips.bmnzi.b(<16 x i8> %0, <16 x i8> %1, i32 240) - store <16 x i8> %2, <16 x i8>* @llvm_mips_bmnzi_b_RES + store volatile <16 x i8> %2, <16 x i8>* @llvm_mips_bmnzi_b_RES %3 = tail call <16 x i8> @llvm.mips.bmnzi.b(<16 x i8> %0, <16 x i8> %1, i32 15) - store <16 x i8> %3, <16 x i8>* @llvm_mips_bmnzi_b_RES + store volatile <16 x i8> %3, <16 x i8>* @llvm_mips_bmnzi_b_RES %4 = tail call <16 x i8> @llvm.mips.bmnzi.b(<16 x i8> %0, <16 x i8> %1, i32 170) store <16 x i8> %4, <16 x i8>* @llvm_mips_bmnzi_b_RES ret void @@ -32,9 +32,9 @@ entry: %0 = load <16 x i8>, <16 x i8>* @llvm_mips_bmnzi_b_ARG1 %1 = load <16 x i8>, <16 x i8>* @llvm_mips_bmnzi_b_ARG2 %2 = tail call <16 x i8> @llvm.mips.bmzi.b(<16 x i8> %0, <16 x i8> %1, i32 240) - store <16 x i8> %2, <16 x i8>* @llvm_mips_bmnzi_b_RES + store volatile <16 x i8> %2, <16 x i8>* @llvm_mips_bmnzi_b_RES %3 = tail call <16 x i8> @llvm.mips.bmzi.b(<16 x i8> %0, <16 x i8> %1, i32 15) - store <16 x i8> %3, <16 x i8>* @llvm_mips_bmnzi_b_RES + store volatile <16 x i8> %3, <16 x i8>* @llvm_mips_bmnzi_b_RES %4 = tail call <16 x i8> @llvm.mips.bmzi.b(<16 x i8> %0, <16 x i8> %1, i32 170) store <16 x i8> %4, <16 x i8>* @llvm_mips_bmnzi_b_RES ret void diff --git a/test/CodeGen/PowerPC/atomic-2.ll b/test/CodeGen/PowerPC/atomic-2.ll index 18715ddb37c6..2039c1f57f17 100644 --- a/test/CodeGen/PowerPC/atomic-2.ll +++ b/test/CodeGen/PowerPC/atomic-2.ll @@ -1,7 +1,7 @@ -; RUN: llc < %s -march=ppc64 | FileCheck %s -check-prefix=CHECK -check-prefix=CHECK-BE -; RUN: llc < %s -march=ppc64le -mtriple=powerpc64le-unknown-linux-gnu | FileCheck %s -check-prefix=CHECK -check-prefix=CHECK-LE -; RUN: llc < %s -march=ppc64 -mcpu=pwr7 | FileCheck %s -; RUN: llc < %s -march=ppc64 -mcpu=pwr8 | FileCheck %s -check-prefix=CHECK-P8U +; RUN: llc < %s -ppc-asm-full-reg-names -march=ppc64 | FileCheck %s -check-prefix=CHECK -check-prefix=CHECK-BE +; RUN: llc < %s -ppc-asm-full-reg-names -march=ppc64le -mtriple=powerpc64le-unknown-linux-gnu | FileCheck %s -check-prefix=CHECK -check-prefix=CHECK-LE +; RUN: llc < %s -ppc-asm-full-reg-names -march=ppc64 -mcpu=pwr7 | FileCheck %s +; RUN: llc < %s -ppc-asm-full-reg-names -march=ppc64 -mcpu=pwr8 | FileCheck %s -check-prefix=CHECK-P8U define i64 @exchange_and_add(i64* %mem, i64 %val) nounwind { ; CHECK-LABEL: exchange_and_add: @@ -108,8 +108,10 @@ entry: ; CHECK: @atomic_load %tmp = load atomic i64, i64* %mem acquire, align 64 ; CHECK-NOT: ldarx -; CHECK: ld -; CHECK: lwsync +; CHECK: ld [[VAL:r[0-9]+]] +; CHECK: cmpw [[CR:cr[0-9]+]], [[VAL]], [[VAL]] +; CHECK: bne- [[CR]], .+4 +; CHECK: isync ret i64 %tmp } diff --git a/test/CodeGen/PowerPC/atomics-indexed.ll b/test/CodeGen/PowerPC/atomics-indexed.ll index 7a0dde034d68..cfe15f0061c4 100644 --- a/test/CodeGen/PowerPC/atomics-indexed.ll +++ b/test/CodeGen/PowerPC/atomics-indexed.ll @@ -10,16 +10,22 @@ define i8 @load_x_i8_seq_cst([100000 x i8]* %mem) { ; CHECK-LABEL: load_x_i8_seq_cst ; CHECK: sync -; CHECK: lbzx -; CHECK: lwsync +; CHECK: lbzx [[VAL:r[0-9]+]] +; CHECK-PPC32: lwsync +; CHECK-PPC64: cmpw [[CR:cr[0-9]+]], [[VAL]], [[VAL]] +; CHECK-PPC64: bne- [[CR]], .+4 +; CHECK-PPC64: isync %ptr = getelementptr inbounds [100000 x i8], [100000 x i8]* %mem, i64 0, i64 90000 %val = load atomic i8, i8* %ptr seq_cst, align 1 ret i8 %val } define i16 @load_x_i16_acquire([100000 x i16]* %mem) { ; CHECK-LABEL: load_x_i16_acquire -; CHECK: lhzx -; CHECK: lwsync +; CHECK: lhzx [[VAL:r[0-9]+]] +; CHECK-PPC32: lwsync +; CHECK-PPC64: cmpw [[CR:cr[0-9]+]], [[VAL]], [[VAL]] +; CHECK-PPC64: bne- [[CR]], .+4 +; CHECK-PPC64: isync %ptr = getelementptr inbounds [100000 x i16], [100000 x i16]* %mem, i64 0, i64 90000 %val = load atomic i16, i16* %ptr acquire, align 2 ret i16 %val diff --git a/test/CodeGen/PowerPC/atomics-regression.ll b/test/CodeGen/PowerPC/atomics-regression.ll index 9af82b625532..054d3a4146b0 100644 --- a/test/CodeGen/PowerPC/atomics-regression.ll +++ b/test/CodeGen/PowerPC/atomics-regression.ll @@ -23,7 +23,9 @@ define i8 @test2(i8* %ptr) { ; PPC64LE-LABEL: test2: ; PPC64LE: # BB#0: ; PPC64LE-NEXT: lbz 3, 0(3) -; PPC64LE-NEXT: lwsync +; PPC64LE-NEXT: cmpw 7, 3, 3 +; PPC64LE-NEXT: bne- 7, .+4 +; PPC64LE-NEXT: isync ; PPC64LE-NEXT: blr %val = load atomic i8, i8* %ptr acquire, align 1 ret i8 %val @@ -35,7 +37,9 @@ define i8 @test3(i8* %ptr) { ; PPC64LE-NEXT: sync ; PPC64LE-NEXT: ori 2, 2, 0 ; PPC64LE-NEXT: lbz 3, 0(3) -; PPC64LE-NEXT: lwsync +; PPC64LE-NEXT: cmpw 7, 3, 3 +; PPC64LE-NEXT: bne- 7, .+4 +; PPC64LE-NEXT: isync ; PPC64LE-NEXT: blr %val = load atomic i8, i8* %ptr seq_cst, align 1 ret i8 %val @@ -63,7 +67,9 @@ define i16 @test6(i16* %ptr) { ; PPC64LE-LABEL: test6: ; PPC64LE: # BB#0: ; PPC64LE-NEXT: lhz 3, 0(3) -; PPC64LE-NEXT: lwsync +; PPC64LE-NEXT: cmpw 7, 3, 3 +; PPC64LE-NEXT: bne- 7, .+4 +; PPC64LE-NEXT: isync ; PPC64LE-NEXT: blr %val = load atomic i16, i16* %ptr acquire, align 2 ret i16 %val @@ -75,7 +81,9 @@ define i16 @test7(i16* %ptr) { ; PPC64LE-NEXT: sync ; PPC64LE-NEXT: ori 2, 2, 0 ; PPC64LE-NEXT: lhz 3, 0(3) -; PPC64LE-NEXT: lwsync +; PPC64LE-NEXT: cmpw 7, 3, 3 +; PPC64LE-NEXT: bne- 7, .+4 +; PPC64LE-NEXT: isync ; PPC64LE-NEXT: blr %val = load atomic i16, i16* %ptr seq_cst, align 2 ret i16 %val @@ -103,7 +111,9 @@ define i32 @test10(i32* %ptr) { ; PPC64LE-LABEL: test10: ; PPC64LE: # BB#0: ; PPC64LE-NEXT: lwz 3, 0(3) -; PPC64LE-NEXT: lwsync +; PPC64LE-NEXT: cmpw 7, 3, 3 +; PPC64LE-NEXT: bne- 7, .+4 +; PPC64LE-NEXT: isync ; PPC64LE-NEXT: blr %val = load atomic i32, i32* %ptr acquire, align 4 ret i32 %val @@ -115,7 +125,9 @@ define i32 @test11(i32* %ptr) { ; PPC64LE-NEXT: sync ; PPC64LE-NEXT: ori 2, 2, 0 ; PPC64LE-NEXT: lwz 3, 0(3) -; PPC64LE-NEXT: lwsync +; PPC64LE-NEXT: cmpw 7, 3, 3 +; PPC64LE-NEXT: bne- 7, .+4 +; PPC64LE-NEXT: isync ; PPC64LE-NEXT: blr %val = load atomic i32, i32* %ptr seq_cst, align 4 ret i32 %val @@ -143,7 +155,9 @@ define i64 @test14(i64* %ptr) { ; PPC64LE-LABEL: test14: ; PPC64LE: # BB#0: ; PPC64LE-NEXT: ld 3, 0(3) -; PPC64LE-NEXT: lwsync +; PPC64LE-NEXT: cmpw 7, 3, 3 +; PPC64LE-NEXT: bne- 7, .+4 +; PPC64LE-NEXT: isync ; PPC64LE-NEXT: blr %val = load atomic i64, i64* %ptr acquire, align 8 ret i64 %val @@ -155,7 +169,9 @@ define i64 @test15(i64* %ptr) { ; PPC64LE-NEXT: sync ; PPC64LE-NEXT: ori 2, 2, 0 ; PPC64LE-NEXT: ld 3, 0(3) -; PPC64LE-NEXT: lwsync +; PPC64LE-NEXT: cmpw 7, 3, 3 +; PPC64LE-NEXT: bne- 7, .+4 +; PPC64LE-NEXT: isync ; PPC64LE-NEXT: blr %val = load atomic i64, i64* %ptr seq_cst, align 8 ret i64 %val @@ -9544,3 +9560,35 @@ define i64 @test559(i64* %ptr, i64 %val) { %ret = atomicrmw umin i64* %ptr, i64 %val singlethread seq_cst ret i64 %ret } + +; The second load should never be scheduled before isync. +define i32 @test_ordering0(i32* %ptr1, i32* %ptr2) { +; PPC64LE-LABEL: test_ordering0: +; PPC64LE: # BB#0: +; PPC64LE-NEXT: lwz 4, 0(3) +; PPC64LE-NEXT: cmpw 7, 4, 4 +; PPC64LE-NEXT: bne- 7, .+4 +; PPC64LE-NEXT: isync +; PPC64LE-NEXT: lwz 3, 0(3) +; PPC64LE-NEXT: add 3, 4, 3 +; PPC64LE-NEXT: blr + %val1 = load atomic i32, i32* %ptr1 acquire, align 4 + %val2 = load i32, i32* %ptr1 + %add = add i32 %val1, %val2 + ret i32 %add +} + +; The second store should never be scheduled before isync. +define i32 @test_ordering1(i32* %ptr1, i32 %val1, i32* %ptr2) { +; PPC64LE-LABEL: test_ordering1: +; PPC64LE: # BB#0: +; PPC64LE-NEXT: lwz 3, 0(3) +; PPC64LE-NEXT: cmpw 7, 3, 3 +; PPC64LE-NEXT: bne- 7, .+4 +; PPC64LE-NEXT: isync +; PPC64LE-NEXT: stw 4, 0(5) +; PPC64LE-NEXT: blr + %val2 = load atomic i32, i32* %ptr1 acquire, align 4 + store i32 %val1, i32* %ptr2 + ret i32 %val2 +} diff --git a/test/CodeGen/PowerPC/atomics.ll b/test/CodeGen/PowerPC/atomics.ll index 2e1eff0f634d..61d54534f5fd 100644 --- a/test/CodeGen/PowerPC/atomics.ll +++ b/test/CodeGen/PowerPC/atomics.ll @@ -25,9 +25,12 @@ define i16 @load_i16_monotonic(i16* %mem) { } define i32 @load_i32_acquire(i32* %mem) { ; CHECK-LABEL: load_i32_acquire -; CHECK: lwz +; CHECK: lwz [[VAL:r[0-9]+]] %val = load atomic i32, i32* %mem acquire, align 4 -; CHECK: lwsync +; CHECK-PPC32: lwsync +; CHECK-PPC64: cmpw [[CR:cr[0-9]+]], [[VAL]], [[VAL]] +; CHECK-PPC64: bne- [[CR]], .+4 +; CHECK-PPC64: isync ret i32 %val } define i64 @load_i64_seq_cst(i64* %mem) { @@ -35,9 +38,12 @@ define i64 @load_i64_seq_cst(i64* %mem) { ; CHECK: sync ; PPC32: __sync_ ; PPC64-NOT: __sync_ -; PPC64: ld +; PPC64: ld [[VAL:r[0-9]+]] %val = load atomic i64, i64* %mem seq_cst, align 8 -; CHECK: lwsync +; CHECK-PPC32: lwsync +; CHECK-PPC64: cmpw [[CR:cr[0-9]+]], [[VAL]], [[VAL]] +; CHECK-PPC64: bne- [[CR]], .+4 +; CHECK-PPC64: isync ret i64 %val } diff --git a/test/CodeGen/PowerPC/ppcf128sf.ll b/test/CodeGen/PowerPC/ppcf128sf.ll index 6804b551e572..fde7d48da7c2 100644 --- a/test/CodeGen/PowerPC/ppcf128sf.ll +++ b/test/CodeGen/PowerPC/ppcf128sf.ll @@ -14,19 +14,19 @@ entry: %0 = load ppc_fp128, ppc_fp128* @ld, align 16 %1 = load ppc_fp128, ppc_fp128* @ld2, align 16 %add = fadd ppc_fp128 %0, %1 - store ppc_fp128 %add, ppc_fp128* %c, align 16 + store volatile ppc_fp128 %add, ppc_fp128* %c, align 16 %2 = load ppc_fp128, ppc_fp128* @ld, align 16 %3 = load ppc_fp128, ppc_fp128* @ld2, align 16 %sub = fsub ppc_fp128 %2, %3 - store ppc_fp128 %sub, ppc_fp128* %c, align 16 + store volatile ppc_fp128 %sub, ppc_fp128* %c, align 16 %4 = load ppc_fp128, ppc_fp128* @ld, align 16 %5 = load ppc_fp128, ppc_fp128* @ld2, align 16 %mul = fmul ppc_fp128 %4, %5 - store ppc_fp128 %mul, ppc_fp128* %c, align 16 + store volatile ppc_fp128 %mul, ppc_fp128* %c, align 16 %6 = load ppc_fp128, ppc_fp128* @ld, align 16 %7 = load ppc_fp128, ppc_fp128* @ld2, align 16 %div = fdiv ppc_fp128 %6, %7 - store ppc_fp128 %div, ppc_fp128* %c, align 16 + store volatile ppc_fp128 %div, ppc_fp128* %c, align 16 ret void ; CHECK-LABEL: __gcc_qadd diff --git a/test/CodeGen/PowerPC/save-bp.ll b/test/CodeGen/PowerPC/save-bp.ll new file mode 100644 index 000000000000..1c7e19a1d5cb --- /dev/null +++ b/test/CodeGen/PowerPC/save-bp.ll @@ -0,0 +1,54 @@ +; RUN: llc -march=ppc64 -ppc-always-use-base-pointer < %s | FileCheck %s --check-prefix CHECK --check-prefix PPC64 +; RUN: llc -march=ppc32 -ppc-always-use-base-pointer < %s | FileCheck %s --check-prefix CHECK --check-prefix PPC32 +; RUN: llc -march=ppc32 -ppc-always-use-base-pointer -relocation-model pic < %s | FileCheck %s --check-prefix CHECK --check-prefix PPC32PIC + +; CHECK-LABEL: fred: + +; Check for saving/restoring frame pointer (X31) and base pointer (X30) +; on ppc64: +; PPC64: std 31, -8(1) +; PPC64: std 30, -16(1) +; PPC64: ld 31, -8(1) +; PPC64: ld 30, -16(1) + +; Check for saving/restoring frame pointer (R31) and base pointer (R30) +; on ppc32: +; PPC32: stwux 1, 1, 0 +; PPC32; addic 0, 0, -4 +; PPC32: stwx 31, 0, 0 +; PPC32: addic 0, 0, -4 +; PPC32: stwx 30, 0, 0 +; The restore sequence: +; PPC32: lwz 31, 0(1) +; PPC32: addic 30, 0, 8 +; PPC32: lwz 0, -4(31) +; PPC32: lwz 30, -8(31) +; PPC32: mr 1, 31 +; PPC32: mr 31, 0 + +; Check for saving/restoring frame pointer (R31) and base pointer (R29) +; on ppc32/pic. This is mostly the same as without pic, except that base +; pointer is in R29. +; PPC32PIC: stwux 1, 1, 0 +; PPC32PIC; addic 0, 0, -4 +; PPC32PIC: stwx 31, 0, 0 +; PPC32PIC: addic 0, 0, -8 +; PPC32PIC: stwx 29, 0, 0 +; The restore sequence: +; PPC32PIC: lwz 31, 0(1) +; PPC32PIC: addic 29, 0, 12 +; PPC32PIC: lwz 0, -4(31) +; PPC32PIC: lwz 29, -12(31) +; PPC32PIC: mr 1, 31 +; PPC32PIC: mr 31, 0 + + +target datalayout = "E-m:e-p:32:32-i64:64-n32" +target triple = "powerpc-unknown-freebsd" + +define i64 @fred() local_unnamed_addr #0 { +entry: + ret i64 0 +} + +attributes #0 = { norecurse readnone nounwind sspstrong "no-frame-pointer-elim"="true" "target-cpu"="ppc" } diff --git a/test/CodeGen/PowerPC/save-cr-ppc32svr4.ll b/test/CodeGen/PowerPC/save-cr-ppc32svr4.ll new file mode 100644 index 000000000000..9fabca186050 --- /dev/null +++ b/test/CodeGen/PowerPC/save-cr-ppc32svr4.ll @@ -0,0 +1,46 @@ +; RUN: llc -march=ppc32 -relocation-model pic < %s | FileCheck %s +; +; Make sure that the CR register is saved correctly on PPC32/SVR4. + +; CHECK-LABEL: fred: +; CHECK: stwu 1, -32(1) +; CHECK: stw 31, 28(1) +; CHECK: mr 31, 1 +; CHECK: stw 30, 24(1) +; CHECK: mfcr [[CR:[0-9]+]] +; CHECK: stw [[CR]], 20(31) + +target datalayout = "E-m:e-p:32:32-i64:64-n32" +target triple = "powerpc-unknown-freebsd" + +; Function Attrs: norecurse nounwind readnone sspstrong +define i64 @fred(double %a0) local_unnamed_addr #0 { +b1: + %v2 = fcmp olt double %a0, 0x43E0000000000000 + br i1 %v2, label %b3, label %b7 + +b3: ; preds = %b1 + %v4 = fcmp olt double %a0, 0xC3E0000000000000 + %v5 = fptosi double %a0 to i64 + %v6 = select i1 %v4, i64 -9223372036854775808, i64 %v5 + br label %b14 + +b7: ; preds = %b1 + %v8 = fcmp olt double %a0, 0x43F0000000000000 + br i1 %v8, label %b9, label %b11 + +b9: ; preds = %b7 + %v10 = fptoui double %a0 to i64 + br label %b14 + +b11: ; preds = %b7 + %v12 = fcmp ogt double %a0, 0.000000e+00 + %v13 = sext i1 %v12 to i64 + br label %b14 + +b14: ; preds = %b11, %b9, %b3 + %v15 = phi i64 [ %v6, %b3 ], [ %v10, %b9 ], [ %v13, %b11 ] + ret i64 %v15 +} + +attributes #0 = { norecurse nounwind readnone sspstrong "disable-tail-calls"="false" "less-precise-fpmad"="false" "no-frame-pointer-elim"="true" "no-frame-pointer-elim-non-leaf" "stack-protector-buffer-size"="8" "target-cpu"="ppc" } diff --git a/test/CodeGen/PowerPC/save-crbp-ppc32svr4.ll b/test/CodeGen/PowerPC/save-crbp-ppc32svr4.ll new file mode 100644 index 000000000000..b7b3c1ada965 --- /dev/null +++ b/test/CodeGen/PowerPC/save-crbp-ppc32svr4.ll @@ -0,0 +1,57 @@ +; RUN: llc -march=ppc32 -relocation-model pic < %s | FileCheck %s + +; CHECK-LABEL: fred +; CHECK: stwux 1, 1, 0 +; Save R31..R29 via R0: +; CHECK: addic 0, 0, -4 +; CHECK: stwx 31, 0, 0 +; CHECK: addic 0, 0, -4 +; CHECK: stwx 30, 0, 0 +; CHECK: addic 0, 0, -4 +; CHECK: stwx 29, 0, 0 +; Set R29 back to the value of R0 from before the updates: +; CHECK: addic 29, 0, 12 +; Save CR through R12 using R29 as the stack pointer (aligned base pointer). +; CHECK: mfcr 12 +; CHECK: stw 28, -16(29) +; CHECK: stw 12, -20(29) + +target datalayout = "E-m:e-p:32:32-i64:64-n32" +target triple = "powerpc-unknown-freebsd" + +; Function Attrs: norecurse readnone sspstrong +define i64 @fred(double %a0) local_unnamed_addr #0 { +b1: + %v2 = alloca i64, align 128 + store i64 0, i64* %v2 + %v3 = fcmp olt double %a0, 0x43E0000000000000 + br i1 %v3, label %b4, label %b8 + +b4: ; preds = %b1 + %v5 = fcmp olt double %a0, 0xC3E0000000000000 + %v6 = fptosi double %a0 to i64 + store i64 %v6, i64* %v2 + %v7 = select i1 %v5, i64 -9223372036854775808, i64 %v6 + br label %b15 + +b8: ; preds = %b1 + %v9 = fcmp olt double %a0, 0x43F0000000000000 + br i1 %v9, label %b10, label %b12 + +b10: ; preds = %b8 + %v11 = fptoui double %a0 to i64 + br label %b15 + +b12: ; preds = %b8 + %v13 = fcmp ogt double %a0, 0.000000e+00 + %v14 = sext i1 %v13 to i64 + br label %b15 + +b15: ; preds = %b12, %b10, %b4 + %v16 = phi i64 [ %v7, %b4 ], [ %v11, %b10 ], [ %v14, %b12 ] + %v17 = load i64, i64* %v2 + %v18 = add i64 %v17, %v16 + ret i64 %v18 +} + +attributes #0 = { norecurse readnone sspstrong "disable-tail-calls"="false" "less-precise-fpmad"="false" "no-frame-pointer-elim"="true" "no-frame-pointer-elim-non-leaf" "stack-protector-buffer-size"="8" "target-cpu"="ppc" } diff --git a/test/CodeGen/SPARC/32abi.ll b/test/CodeGen/SPARC/32abi.ll index 09e7a3a09d86..3807f84d4e92 100644 --- a/test/CodeGen/SPARC/32abi.ll +++ b/test/CodeGen/SPARC/32abi.ll @@ -25,17 +25,17 @@ define void @intarg(i8 %a0, ; %i0 i32 %a5, ; %i5 i32 signext %a6, ; [%fp+92] i8* %a7) { ; [%fp+96] - store i8 %a0, i8* %a4 - store i8 %a1, i8* %a4 + store volatile i8 %a0, i8* %a4 + store volatile i8 %a1, i8* %a4 %p16 = bitcast i8* %a4 to i16* - store i16 %a2, i16* %p16 + store volatile i16 %a2, i16* %p16 %p32 = bitcast i8* %a4 to i32* - store i32 %a3, i32* %p32 + store volatile i32 %a3, i32* %p32 %pp = bitcast i8* %a4 to i8** - store i8* %a4, i8** %pp - store i32 %a5, i32* %p32 - store i32 %a6, i32* %p32 - store i8* %a7, i8** %pp + store volatile i8* %a4, i8** %pp + store volatile i32 %a5, i32* %p32 + store volatile i32 %a6, i32* %p32 + store volatile i8* %a7, i8** %pp ret void } diff --git a/test/CodeGen/SPARC/64abi.ll b/test/CodeGen/SPARC/64abi.ll index b963be2e9853..771cc409554b 100644 --- a/test/CodeGen/SPARC/64abi.ll +++ b/test/CodeGen/SPARC/64abi.ll @@ -24,17 +24,17 @@ define void @intarg(i8 %a0, ; %i0 i32 %a5, ; %i5 i32 signext %a6, ; [%fp+BIAS+176] i8* %a7) { ; [%fp+BIAS+184] - store i8 %a0, i8* %a4 - store i8 %a1, i8* %a4 + store volatile i8 %a0, i8* %a4 + store volatile i8 %a1, i8* %a4 %p16 = bitcast i8* %a4 to i16* - store i16 %a2, i16* %p16 + store volatile i16 %a2, i16* %p16 %p32 = bitcast i8* %a4 to i32* - store i32 %a3, i32* %p32 + store volatile i32 %a3, i32* %p32 %pp = bitcast i8* %a4 to i8** - store i8* %a4, i8** %pp - store i32 %a5, i32* %p32 - store i32 %a6, i32* %p32 - store i8* %a7, i8** %pp + store volatile i8* %a4, i8** %pp + store volatile i32 %a5, i32* %p32 + store volatile i32 %a6, i32* %p32 + store volatile i8* %a7, i8** %pp ret void } @@ -316,7 +316,7 @@ define void @call_ret_i64_pair(i64* %i0) { %rv = call { i64, i64 } @ret_i64_pair(i32 undef, i32 undef, i64* undef, i64* undef) %e0 = extractvalue { i64, i64 } %rv, 0 - store i64 %e0, i64* %i0 + store volatile i64 %e0, i64* %i0 %e1 = extractvalue { i64, i64 } %rv, 1 store i64 %e1, i64* %i0 ret void diff --git a/test/CodeGen/SystemZ/swift-return.ll b/test/CodeGen/SystemZ/swift-return.ll index 69d0e979190c..977816f66bec 100644 --- a/test/CodeGen/SystemZ/swift-return.ll +++ b/test/CodeGen/SystemZ/swift-return.ll @@ -189,11 +189,11 @@ define void @consume_i1_ret() { %v6 = extractvalue { i1, i1, i1, i1 } %call, 2 %v7 = extractvalue { i1, i1, i1, i1 } %call, 3 %val = zext i1 %v3 to i32 - store i32 %val, i32* @var + store volatile i32 %val, i32* @var %val2 = zext i1 %v5 to i32 - store i32 %val2, i32* @var + store volatile i32 %val2, i32* @var %val3 = zext i1 %v6 to i32 - store i32 %val3, i32* @var + store volatile i32 %val3, i32* @var %val4 = zext i1 %v7 to i32 store i32 %val4, i32* @var ret void diff --git a/test/CodeGen/Thumb/stack-access.ll b/test/CodeGen/Thumb/stack-access.ll index 44217aba62d5..533559a67421 100644 --- a/test/CodeGen/Thumb/stack-access.ll +++ b/test/CodeGen/Thumb/stack-access.ll @@ -7,13 +7,13 @@ define void @test1(i8** %p) { %z = alloca i8, align 1 ; CHECK: add r1, sp, #8 ; CHECK: str r1, [r0] - store i8* %x, i8** %p, align 4 + store volatile i8* %x, i8** %p, align 4 ; CHECK: add r1, sp, #4 ; CHECK: str r1, [r0] - store i8* %y, i8** %p, align 4 + store volatile i8* %y, i8** %p, align 4 ; CHECK: mov r1, sp ; CHECK: str r1, [r0] - store i8* %z, i8** %p, align 4 + store volatile i8* %z, i8** %p, align 4 ret void } @@ -24,10 +24,10 @@ define void @test2([1024 x i8]** %p) { ; CHECK: add r1, sp, #1020 ; CHECK: adds r1, #4 ; CHECK: str r1, [r0] - store [1024 x i8]* %arr1, [1024 x i8]** %p, align 4 + store volatile [1024 x i8]* %arr1, [1024 x i8]** %p, align 4 ; CHECK: mov r1, sp ; CHECK: str r1, [r0] - store [1024 x i8]* %arr2, [1024 x i8]** %p, align 4 + store volatile [1024 x i8]* %arr2, [1024 x i8]** %p, align 4 ret void } diff --git a/test/CodeGen/Thumb2/ldr-str-imm12.ll b/test/CodeGen/Thumb2/ldr-str-imm12.ll index 3e4bd02097ad..c6d00d4c1e11 100644 --- a/test/CodeGen/Thumb2/ldr-str-imm12.ll +++ b/test/CodeGen/Thumb2/ldr-str-imm12.ll @@ -50,9 +50,9 @@ bb420: ; preds = %bb20, %bb20 ; CHECK: str{{(.w)?}} r{{[0-9]+}}, [sp ; CHECK: str{{(.w)?}} r{{[0-9]+}}, [sp ; CHECK: str{{(.w)?}} r{{[0-9]+}}, [sp - store %union.rec* null, %union.rec** @zz_hold, align 4 + store volatile %union.rec* null, %union.rec** @zz_hold, align 4 store %union.rec* null, %union.rec** @zz_res, align 4 - store %union.rec* %x, %union.rec** @zz_hold, align 4 + store volatile %union.rec* %x, %union.rec** @zz_hold, align 4 %0 = call %union.rec* @Manifest(%union.rec* undef, %union.rec* %env, %struct.STYLE* %style, %union.rec** %bthr, %union.rec** %fthr, %union.rec** %target, %union.rec** %crs, i32 %ok, i32 %need_expand, %union.rec** %enclose, i32 %fcr) nounwind ; <%union.rec*> [#uses=0] unreachable diff --git a/test/CodeGen/X86/GlobalISel/add-scalar.ll b/test/CodeGen/X86/GlobalISel/add-scalar.ll index 553bc2789ff0..85db1c0e7e7a 100644 --- a/test/CodeGen/X86/GlobalISel/add-scalar.ll +++ b/test/CodeGen/X86/GlobalISel/add-scalar.ll @@ -1,44 +1,94 @@ ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py ; RUN: llc -mtriple=x86_64-linux-gnu -global-isel < %s -o - | FileCheck %s --check-prefix=ALL --check-prefix=X64 +; RUN: llc -mtriple=i386-linux-gnu -global-isel < %s -o - | FileCheck %s --check-prefix=ALL --check-prefix=X32 define i64 @test_add_i64(i64 %arg1, i64 %arg2) { -; ALL-LABEL: test_add_i64: -; ALL: # BB#0: -; ALL-NEXT: leaq (%rsi,%rdi), %rax -; ALL-NEXT: retq +; X64-LABEL: test_add_i64: +; X64: # BB#0: +; X64-NEXT: leaq (%rsi,%rdi), %rax +; X64-NEXT: retq +; +; X32-LABEL: test_add_i64: +; X32: # BB#0: +; X32-NEXT: pushl %ebp +; X32-NEXT: .Lcfi0: +; X32-NEXT: .cfi_def_cfa_offset 8 +; X32-NEXT: .Lcfi1: +; X32-NEXT: .cfi_offset %ebp, -8 +; X32-NEXT: movl %esp, %ebp +; X32-NEXT: .Lcfi2: +; X32-NEXT: .cfi_def_cfa_register %ebp +; X32-NEXT: pushl %esi +; X32-NEXT: .Lcfi3: +; X32-NEXT: .cfi_offset %esi, -12 +; X32-NEXT: leal 8(%ebp), %ecx +; X32-NEXT: leal 12(%ebp), %esi +; X32-NEXT: leal 16(%ebp), %eax +; X32-NEXT: movl (%eax), %eax +; X32-NEXT: leal 20(%ebp), %edx +; X32-NEXT: movl (%edx), %edx +; X32-NEXT: addl (%ecx), %eax +; X32-NEXT: adcl (%esi), %edx +; X32-NEXT: popl %esi +; X32-NEXT: popl %ebp +; X32-NEXT: retl %ret = add i64 %arg1, %arg2 ret i64 %ret } define i32 @test_add_i32(i32 %arg1, i32 %arg2) { -; ALL-LABEL: test_add_i32: -; ALL: # BB#0: -; ALL-NEXT: # kill: %EDI %EDI %RDI -; ALL-NEXT: # kill: %ESI %ESI %RSI -; ALL-NEXT: leal (%rsi,%rdi), %eax -; ALL-NEXT: retq +; X64-LABEL: test_add_i32: +; X64: # BB#0: +; X64-NEXT: # kill: %EDI %EDI %RDI +; X64-NEXT: # kill: %ESI %ESI %RSI +; X64-NEXT: leal (%rsi,%rdi), %eax +; X64-NEXT: retq +; +; X32-LABEL: test_add_i32: +; X32: # BB#0: +; X32-NEXT: leal 4(%esp), %ecx +; X32-NEXT: leal 8(%esp), %eax +; X32-NEXT: movl (%eax), %eax +; X32-NEXT: addl (%ecx), %eax +; X32-NEXT: retl %ret = add i32 %arg1, %arg2 ret i32 %ret } define i16 @test_add_i16(i16 %arg1, i16 %arg2) { -; ALL-LABEL: test_add_i16: -; ALL: # BB#0: -; ALL-NEXT: # kill: %DI %DI %RDI -; ALL-NEXT: # kill: %SI %SI %RSI -; ALL-NEXT: leal (%rsi,%rdi), %eax -; ALL-NEXT: # kill: %AX %AX %EAX -; ALL-NEXT: retq +; X64-LABEL: test_add_i16: +; X64: # BB#0: +; X64-NEXT: # kill: %DI %DI %RDI +; X64-NEXT: # kill: %SI %SI %RSI +; X64-NEXT: leal (%rsi,%rdi), %eax +; X64-NEXT: # kill: %AX %AX %EAX +; X64-NEXT: retq +; +; X32-LABEL: test_add_i16: +; X32: # BB#0: +; X32-NEXT: leal 4(%esp), %ecx +; X32-NEXT: leal 8(%esp), %eax +; X32-NEXT: movzwl (%eax), %eax +; X32-NEXT: addw (%ecx), %ax +; X32-NEXT: retl %ret = add i16 %arg1, %arg2 ret i16 %ret } define i8 @test_add_i8(i8 %arg1, i8 %arg2) { -; ALL-LABEL: test_add_i8: -; ALL: # BB#0: -; ALL-NEXT: addb %dil, %sil -; ALL-NEXT: movl %esi, %eax -; ALL-NEXT: retq +; X64-LABEL: test_add_i8: +; X64: # BB#0: +; X64-NEXT: addb %dil, %sil +; X64-NEXT: movl %esi, %eax +; X64-NEXT: retq +; +; X32-LABEL: test_add_i8: +; X32: # BB#0: +; X32-NEXT: leal 4(%esp), %ecx +; X32-NEXT: leal 8(%esp), %eax +; X32-NEXT: movb (%eax), %al +; X32-NEXT: addb (%ecx), %al +; X32-NEXT: retl %ret = add i8 %arg1, %arg2 ret i8 %ret } diff --git a/test/CodeGen/X86/GlobalISel/legalize-add.mir b/test/CodeGen/X86/GlobalISel/legalize-add.mir index 22619cc71033..6a03388da947 100644 --- a/test/CodeGen/X86/GlobalISel/legalize-add.mir +++ b/test/CodeGen/X86/GlobalISel/legalize-add.mir @@ -1,40 +1,67 @@ -# RUN: llc -mtriple=x86_64-linux-gnu -global-isel -run-pass=legalizer %s -o - | FileCheck %s - +# RUN: llc -mtriple=x86_64-linux-gnu -global-isel -run-pass=legalizer %s -o - | FileCheck %s --check-prefix=ALL --check-prefix=X64 +# RUN: llc -mtriple=i386-linux-gnu -global-isel -run-pass=legalizer %s -o - | FileCheck %s --check-prefix=ALL --check-prefix=X32 --- | - ; ModuleID = '' - source_filename = "" - target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128" - target triple = "x86_64--linux-gnu" + define void @test_add_i32() { + ret void + } - define i32 @test_add_i32(i32 %arg1, i32 %arg2) { - %ret = add i32 %arg1, %arg2 - ret i32 %ret + define void @test_add_i64() { + ret void } ... --- name: test_add_i32 +# ALL-LABEL: name: test_add_i32 alignment: 4 legalized: false regBankSelected: false -selected: false -tracksRegLiveness: true registers: - { id: 0, class: _ } - { id: 1, class: _ } - { id: 2, class: _ } +# ALL: %0(s32) = IMPLICIT_DEF +# ALL-NEXT: %1(s32) = IMPLICIT_DEF +# ALL-NEXT: %2(s32) = G_ADD %0, %1 +# ALL-NEXT: RET 0 body: | bb.1 (%ir-block.0): - liveins: %edi, %esi - ; CHECK-LABEL: name: test_add_i32 - ; CHECK: [[VAL1:%.*]](s32) = COPY %edi - ; CHECK: [[VAL2:%.*]](s32) = COPY %esi - ; CHECK: [[RES:%.*]](s32) = G_ADD [[VAL1:%.*]], [[VAL2:%.*]] - - %0(s32) = COPY %edi - %1(s32) = COPY %esi + %0(s32) = IMPLICIT_DEF + %1(s32) = IMPLICIT_DEF %2(s32) = G_ADD %0, %1 - %eax = COPY %2(s32) - RET 0, implicit %eax + RET 0 + +... +--- +name: test_add_i64 +# ALL-LABEL: name: test_add_i64 +alignment: 4 +legalized: false +regBankSelected: false +registers: + - { id: 0, class: _ } + - { id: 1, class: _ } + - { id: 2, class: _ } +# X64: %0(s64) = IMPLICIT_DEF +# X64-NEXT: %1(s64) = IMPLICIT_DEF +# X64-NEXT: %2(s64) = G_ADD %0, %1 +# X64-NEXT: RET 0 +# +# X32: %0(s64) = IMPLICIT_DEF +# X32-NEXT: %1(s64) = IMPLICIT_DEF +# X32-NEXT: %3(s32), %4(s32) = G_UNMERGE_VALUES %0(s64) +# X32-NEXT: %5(s32), %6(s32) = G_UNMERGE_VALUES %1(s64) +# X32-NEXT: %12(s8) = G_CONSTANT i8 0 +# X32-NEXT: %7(s1) = G_TRUNC %12(s8) +# X32-NEXT: %8(s32), %9(s1) = G_UADDE %3, %5, %7 +# X32-NEXT: %10(s32), %11(s1) = G_UADDE %4, %6, %9 +# X32-NEXT: %2(s64) = G_MERGE_VALUES %8(s32), %10(s32) +# X32-NEXT: RET 0 +body: | + bb.1 (%ir-block.0): + %0(s64) = IMPLICIT_DEF + %1(s64) = IMPLICIT_DEF + %2(s64) = G_ADD %0, %1 + RET 0 ... diff --git a/test/CodeGen/X86/GlobalISel/regbankselect-X32.mir b/test/CodeGen/X86/GlobalISel/regbankselect-X32.mir new file mode 100644 index 000000000000..a115d1fa3255 --- /dev/null +++ b/test/CodeGen/X86/GlobalISel/regbankselect-X32.mir @@ -0,0 +1,36 @@ +# RUN: llc -mtriple=i386-linux-gnu -global-isel -run-pass=regbankselect %s -o - | FileCheck %s --check-prefix=CHECK --check-prefix=FAST +# RUN: llc -mtriple=i386-linux-gnu -global-isel -regbankselect-greedy -run-pass=regbankselect %s -o - | FileCheck %s --check-prefix=CHECK --check-prefix=GREEDY + +--- | + define void @test_uadde_i32() { + ret void + } + +... +--- +name: test_uadde_i32 +# CHECK-LABEL: name: test_uadde_i32 +alignment: 4 +legalized: true +regBankSelected: false +# CHECK: registers: +# CHECK-NEXT: - { id: 0, class: gpr } +# CHECK-NEXT: - { id: 1, class: gpr } +# CHECK-NEXT: - { id: 2, class: gpr } +# CHECK-NEXT: - { id: 3, class: gpr } +# CHECK-NEXT: - { id: 4, class: gpr } +registers: + - { id: 0, class: _ } + - { id: 1, class: _ } + - { id: 2, class: _ } + - { id: 3, class: _ } + - { id: 4, class: _ } +body: | + bb.0 (%ir-block.0): + %0(s32) = IMPLICIT_DEF + %1(s32) = IMPLICIT_DEF + %2(s1) = IMPLICIT_DEF + %3(s32), %4(s1) = G_UADDE %0, %1, %2 + RET 0 + +... diff --git a/test/CodeGen/X86/GlobalISel/select-add-x32.mir b/test/CodeGen/X86/GlobalISel/select-add-x32.mir new file mode 100644 index 000000000000..8710aaa61a21 --- /dev/null +++ b/test/CodeGen/X86/GlobalISel/select-add-x32.mir @@ -0,0 +1,63 @@ +# RUN: llc -mtriple=i386-linux-gnu -global-isel -run-pass=instruction-select -verify-machineinstrs %s -o - | FileCheck %s --check-prefix=X32 +--- | + define i64 @test_add_i64(i64 %a, i64 %b) { + %r = add i64 %a, %b + ret i64 %r + } + +... +--- +name: test_add_i64 +# X32-LABEL: name: test_add_i64 +alignment: 4 +legalized: true +regBankSelected: true +# X32: registers: +# X32-NEXT: - { id: 0, class: gr32 } +# X32-NEXT: - { id: 1, class: gr32 } +# X32-NEXT: - { id: 2, class: gr32 } +# X32-NEXT: - { id: 3, class: gr32 } +# X32-NEXT: - { id: 4, class: gpr } +# X32-NEXT: - { id: 5, class: gr32 } +# X32-NEXT: - { id: 6, class: gr32 } +# X32-NEXT: - { id: 7, class: gr32 } +# X32-NEXT: - { id: 8, class: gr32 } +# X32-NEXT: - { id: 9, class: gpr } +registers: + - { id: 0, class: gpr } + - { id: 1, class: gpr } + - { id: 2, class: gpr } + - { id: 3, class: gpr } + - { id: 4, class: gpr } + - { id: 5, class: gpr } + - { id: 6, class: gpr } + - { id: 7, class: gpr } + - { id: 8, class: gpr } + - { id: 9, class: gpr } +# X32: %0 = IMPLICIT_DEF +# X32-NEXT: %1 = IMPLICIT_DEF +# X32-NEXT: %2 = IMPLICIT_DEF +# X32-NEXT: %3 = IMPLICIT_DEF +# X32-NEXT: %5 = ADD32rr %0, %2, implicit-def %eflags +# X32-NEXT: %6 = COPY %eflags +# X32-NEXT: %eflags = COPY %6 +# X32-NEXT: %7 = ADC32rr %1, %3, implicit-def %eflags, implicit %eflags +# X32-NEXT: %8 = COPY %eflags +# X32-NEXT: %eax = COPY %5 +# X32-NEXT: %edx = COPY %7 +# X32-NEXT: RET 0, implicit %eax, implicit %edx +body: | + bb.0 (%ir-block.0): + %0(s32) = IMPLICIT_DEF + %1(s32) = IMPLICIT_DEF + %2(s32) = IMPLICIT_DEF + %3(s32) = IMPLICIT_DEF + %9(s8) = G_CONSTANT i8 0 + %4(s1) = G_TRUNC %9(s8) + %5(s32), %6(s1) = G_UADDE %0, %2, %4 + %7(s32), %8(s1) = G_UADDE %1, %3, %6 + %eax = COPY %5(s32) + %edx = COPY %7(s32) + RET 0, implicit %eax, implicit %edx + +... diff --git a/test/CodeGen/X86/arg-copy-elide.ll b/test/CodeGen/X86/arg-copy-elide.ll index b9a2eeeb7f8f..126f5a1c7976 100644 --- a/test/CodeGen/X86/arg-copy-elide.ll +++ b/test/CodeGen/X86/arg-copy-elide.ll @@ -253,9 +253,7 @@ entry: ; CHECK: calll _addrof_i32 ; CHECK: retl - ; Don't elide the copy when the alloca is escaped with a store. - define void @escape_with_store(i32 %x) { %x1 = alloca i32 %x2 = alloca i32* @@ -268,9 +266,8 @@ define void @escape_with_store(i32 %x) { } ; CHECK-LABEL: _escape_with_store: -; CHECK-DAG: movl {{.*}}(%esp), %[[reg:[^ ]*]] -; CHECK-DAG: movl $0, [[offs:[0-9]*]](%esp) -; CHECK: movl %[[reg]], [[offs]](%esp) +; CHECK: movl {{.*}}(%esp), %[[reg:[^ ]*]] +; CHECK: movl %[[reg]], [[offs:[0-9]*]](%esp) ; CHECK: calll _addrof_i32 diff --git a/test/CodeGen/X86/leaFixup32.mir b/test/CodeGen/X86/leaFixup32.mir deleted file mode 100644 index 70aac21c7ff2..000000000000 --- a/test/CodeGen/X86/leaFixup32.mir +++ /dev/null @@ -1,508 +0,0 @@ -# RUN: llc -run-pass x86-fixup-LEAs -mcpu=corei7-avx -o - %s | FileCheck %s ---- | - ; ModuleID = 'test/CodeGen/X86/fixup-lea.ll' - source_filename = "test/CodeGen/X86/fixup-lea.ll" - target datalayout = "e-m:e-p:32:32-f64:32:64-f80:32-n8:16:32-S128" - target triple = "i386" - ;generated using: llc -stop-after x86-pad-short-functions fixup-lea.ll > leaFinxup32.mir - - ;test2add_32: 3 operands LEA32r that can be replaced with 2 add instructions - ; where ADD32ri8 is chosen - define i32 @test2add_32() { - ret i32 0 - } - - ;test2add_ebp_32: 3 operands LEA32r that can be replaced with 2 add instructions - ; where the base is rbp/r13/ebp register - define i32 @test2add_ebp_32() { - ret i32 0 - } - - ;test1add_ebp_32: 2 operands LEA32r where base register is ebp and can be replaced - ; with an add instruction - define i32 @test1add_ebp_32() { - ret i32 0 - } - - ;testleaadd_32: 3 operands LEA32r that can be replaced with 1 lea 1 add instructions - define i32 @testleaadd_32() { - ret i32 0 - } - - ;testleaadd_ebp_32: 3 operands LEA32r that can be replaced with 1 lea 1 add instructions - ; where the base is ebp register - define i32 @testleaadd_ebp_32() { - ret i32 0 - } - - ;test1lea_ebp_32: 2 operands LEA32r wher base register is rbp/r13/ebp and can be replaced - ; with a lea instruction - define i32 @test1lea_ebp_32() { - ret i32 0 - } - - ;test2addi32_32: 3 operands LEA32r that can be replaced with 2 add instructions where ADD32ri32 - ; is chosen - define i32 @test2addi32_32() { - ret i32 0 - } - - ;test1mov1add_ebp_32: 2 operands LEA32r that can be replaced with 1 add 1 mov instructions - ; where the base is rbp/r13/ebp register - define i32 @test1mov1add_ebp_32() { - ret i32 0 - } - - ;testleaadd_ebp_index_32: 3 operands LEA32r that can be replaced with 1 lea 1 add instructions - ; where the base and the index are ebp register and there is offset - define i32 @testleaadd_ebp_index_32() { - ret i32 0 - } - - ;testleaadd_ebp_index2_32: 3 operands LEA32r that can be replaced with 1 lea 1 add instructions - ; where the base and the index are ebp register and there is scale - define i32 @testleaadd_ebp_index2_32() { - ret i32 0 - } - - ;test_skip_opt_32: 3 operands LEA32r that can not be replaced with 2 instructions - define i32 @test_skip_opt_32() { - ret i32 0 - } - - ;test_skip_eflags_32: LEA32r that cannot be replaced since its not safe to clobber eflags - define i32 @test_skip_eflags_32() { - ret i32 0 - } - -... ---- -name: test2add_32 -alignment: 4 -exposesReturnsTwice: false -legalized: false -regBankSelected: false -selected: false -tracksRegLiveness: true -liveins: - - { reg: '%eax' } - - { reg: '%ebp' } -frameInfo: - isFrameAddressTaken: false - isReturnAddressTaken: false - hasStackMap: false - hasPatchPoint: false - stackSize: 0 - offsetAdjustment: 0 - maxAlignment: 0 - adjustsStack: false - hasCalls: false - maxCallFrameSize: 0 - hasOpaqueSPAdjustment: false - hasVAStart: false - hasMustTailInVarArgFunc: false -body: | - bb.0 (%ir-block.0): - liveins: %eax, %ebp - ; CHECK: %eax = ADD32rr %eax, killed %ebp - ; CHECK: %eax = ADD32ri8 %eax, -5 - - %eax = LEA32r killed %eax, 1, killed %ebp, -5, _ - RETQ %eax - -... ---- -name: test2add_ebp_32 -alignment: 4 -exposesReturnsTwice: false -legalized: false -regBankSelected: false -selected: false -tracksRegLiveness: true -liveins: - - { reg: '%eax' } - - { reg: '%ebp' } -frameInfo: - isFrameAddressTaken: false - isReturnAddressTaken: false - hasStackMap: false - hasPatchPoint: false - stackSize: 0 - offsetAdjustment: 0 - maxAlignment: 0 - adjustsStack: false - hasCalls: false - maxCallFrameSize: 0 - hasOpaqueSPAdjustment: false - hasVAStart: false - hasMustTailInVarArgFunc: false -body: | - bb.0 (%ir-block.0): - liveins: %eax, %ebp - ; CHECK: %ebp = ADD32rr %ebp, killed %eax - ; CHECK: %ebp = ADD32ri8 %ebp, -5 - - %ebp = LEA32r killed %ebp, 1, killed %eax, -5, _ - RETQ %ebp - -... ---- -name: test1add_ebp_32 -alignment: 4 -exposesReturnsTwice: false -legalized: false -regBankSelected: false -selected: false -tracksRegLiveness: true -liveins: - - { reg: '%eax' } - - { reg: '%ebp' } -frameInfo: - isFrameAddressTaken: false - isReturnAddressTaken: false - hasStackMap: false - hasPatchPoint: false - stackSize: 0 - offsetAdjustment: 0 - maxAlignment: 0 - adjustsStack: false - hasCalls: false - maxCallFrameSize: 0 - hasOpaqueSPAdjustment: false - hasVAStart: false - hasMustTailInVarArgFunc: false -body: | - bb.0 (%ir-block.0): - liveins: %eax, %ebp - ; CHECK: %ebp = ADD32rr %ebp, killed %eax - - %ebp = LEA32r killed %ebp, 1, killed %eax, 0, _ - RETQ %ebp - -... ---- -name: testleaadd_32 -alignment: 4 -exposesReturnsTwice: false -legalized: false -regBankSelected: false -selected: false -tracksRegLiveness: true -liveins: - - { reg: '%eax' } - - { reg: '%ebp' } - - { reg: '%ebx' } -frameInfo: - isFrameAddressTaken: false - isReturnAddressTaken: false - hasStackMap: false - hasPatchPoint: false - stackSize: 0 - offsetAdjustment: 0 - maxAlignment: 0 - adjustsStack: false - hasCalls: false - maxCallFrameSize: 0 - hasOpaqueSPAdjustment: false - hasVAStart: false - hasMustTailInVarArgFunc: false -body: | - bb.0 (%ir-block.0): - liveins: %eax, %ebp, %esi - ; CHECK: %ebx = LEA32r killed %eax, 1, killed %ebp, 0 - ; CHECK: %ebx = ADD32ri8 %ebx, -5 - - %ebx = LEA32r killed %eax, 1, killed %ebp, -5, _ - RETQ %ebx - -... ---- -name: testleaadd_ebp_32 -alignment: 4 -exposesReturnsTwice: false -legalized: false -regBankSelected: false -selected: false -tracksRegLiveness: true -liveins: - - { reg: '%eax' } - - { reg: '%ebp' } - - { reg: '%ebx' } -frameInfo: - isFrameAddressTaken: false - isReturnAddressTaken: false - hasStackMap: false - hasPatchPoint: false - stackSize: 0 - offsetAdjustment: 0 - maxAlignment: 0 - adjustsStack: false - hasCalls: false - maxCallFrameSize: 0 - hasOpaqueSPAdjustment: false - hasVAStart: false - hasMustTailInVarArgFunc: false -body: | - bb.0 (%ir-block.0): - liveins: %eax, %ebp - ; CHECK: %ebx = LEA32r killed %eax, 1, killed %ebp, 0, _ - ; CHECK: %ebx = ADD32ri8 %ebx, -5 - - %ebx = LEA32r killed %ebp, 1, killed %eax, -5, _ - RETQ %ebx - -... ---- -name: test1lea_ebp_32 -alignment: 4 -exposesReturnsTwice: false -legalized: false -regBankSelected: false -selected: false -tracksRegLiveness: true -liveins: - - { reg: '%eax' } - - { reg: '%ebp' } - - { reg: '%ebx' } -frameInfo: - isFrameAddressTaken: false - isReturnAddressTaken: false - hasStackMap: false - hasPatchPoint: false - stackSize: 0 - offsetAdjustment: 0 - maxAlignment: 0 - adjustsStack: false - hasCalls: false - maxCallFrameSize: 0 - hasOpaqueSPAdjustment: false - hasVAStart: false - hasMustTailInVarArgFunc: false -body: | - bb.0 (%ir-block.0): - liveins: %eax, %ebp - ; CHECK: %ebx = LEA32r killed %eax, 1, killed %ebp, 0, _ - - %ebx = LEA32r killed %ebp, 1, killed %eax, 0, _ - RETQ %ebx - -... ---- -name: test2addi32_32 -alignment: 4 -exposesReturnsTwice: false -legalized: false -regBankSelected: false -selected: false -tracksRegLiveness: true -liveins: - - { reg: '%eax' } - - { reg: '%ebp' } -frameInfo: - isFrameAddressTaken: false - isReturnAddressTaken: false - hasStackMap: false - hasPatchPoint: false - stackSize: 0 - offsetAdjustment: 0 - maxAlignment: 0 - adjustsStack: false - hasCalls: false - maxCallFrameSize: 0 - hasOpaqueSPAdjustment: false - hasVAStart: false - hasMustTailInVarArgFunc: false -body: | - bb.0 (%ir-block.0): - liveins: %eax, %ebp - ; CHECK: %eax = ADD32rr %eax, killed %ebp - ; CHECK: %eax = ADD32ri %eax, 129 - - %eax = LEA32r killed %eax, 1, killed %ebp, 129, _ - RETQ %eax - -... ---- -name: test1mov1add_ebp_32 -alignment: 4 -exposesReturnsTwice: false -legalized: false -regBankSelected: false -selected: false -tracksRegLiveness: true -liveins: - - { reg: '%eax' } - - { reg: '%ebp' } -frameInfo: - isFrameAddressTaken: false - isReturnAddressTaken: false - hasStackMap: false - hasPatchPoint: false - stackSize: 0 - offsetAdjustment: 0 - maxAlignment: 0 - adjustsStack: false - hasCalls: false - maxCallFrameSize: 0 - hasOpaqueSPAdjustment: false - hasVAStart: false - hasMustTailInVarArgFunc: false -body: | - bb.0 (%ir-block.0): - liveins: %eax, %ebp, %ebx - ; CHECK: %ebx = MOV32rr killed %ebp - ; CHECK: %ebx = ADD32rr %ebx, killed %ebp - - %ebx = LEA32r killed %ebp, 1, killed %ebp, 0, _ - RETQ %ebx - -... ---- -name: testleaadd_ebp_index_32 -alignment: 4 -exposesReturnsTwice: false -legalized: false -regBankSelected: false -selected: false -tracksRegLiveness: true -liveins: - - { reg: '%ebx' } - - { reg: '%ebp' } -frameInfo: - isFrameAddressTaken: false - isReturnAddressTaken: false - hasStackMap: false - hasPatchPoint: false - stackSize: 0 - offsetAdjustment: 0 - maxAlignment: 0 - adjustsStack: false - hasCalls: false - maxCallFrameSize: 0 - hasOpaqueSPAdjustment: false - hasVAStart: false - hasMustTailInVarArgFunc: false -body: | - bb.0 (%ir-block.0): - liveins: %eax, %ebp, %ebx - ; CHECK: %ebx = LEA32r _, 1, killed %ebp, 5, _ - ; CHECK: %ebx = ADD32rr %ebx, killed %ebp - - %ebx = LEA32r killed %ebp, 1, killed %ebp, 5, _ - RETQ %ebx - -... ---- -name: testleaadd_ebp_index2_32 -alignment: 4 -exposesReturnsTwice: false -legalized: false -regBankSelected: false -selected: false -tracksRegLiveness: true -liveins: - - { reg: '%ebx' } - - { reg: '%ebp' } -frameInfo: - isFrameAddressTaken: false - isReturnAddressTaken: false - hasStackMap: false - hasPatchPoint: false - stackSize: 0 - offsetAdjustment: 0 - maxAlignment: 0 - adjustsStack: false - hasCalls: false - maxCallFrameSize: 0 - hasOpaqueSPAdjustment: false - hasVAStart: false - hasMustTailInVarArgFunc: false -body: | - bb.0 (%ir-block.0): - liveins: %eax, %ebp, %ebx - ; CHECK: %ebx = LEA32r _, 4, killed %ebp, 5, _ - ; CHECK: %ebx = ADD32rr %ebx, killed %ebp - - %ebx = LEA32r killed %ebp, 4, killed %ebp, 5, _ - RETQ %ebx - -... ---- -name: test_skip_opt_32 -alignment: 4 -exposesReturnsTwice: false -legalized: false -regBankSelected: false -selected: false -tracksRegLiveness: true -liveins: - - { reg: '%ebx' } - - { reg: '%ebp' } -frameInfo: - isFrameAddressTaken: false - isReturnAddressTaken: false - hasStackMap: false - hasPatchPoint: false - stackSize: 0 - offsetAdjustment: 0 - maxAlignment: 0 - adjustsStack: false - hasCalls: false - maxCallFrameSize: 0 - hasOpaqueSPAdjustment: false - hasVAStart: false - hasMustTailInVarArgFunc: false -body: | - bb.0 (%ir-block.0): - liveins: %eax, %ebp, %ebx - ; CHECK: %ebp = LEA32r killed %ebp, 4, killed %ebp, 0, _ - - %ebp = LEA32r killed %ebp, 4, killed %ebp, 0, _ - RETQ %ebp - -... ---- -name: test_skip_eflags_32 -alignment: 4 -exposesReturnsTwice: false -legalized: false -regBankSelected: false -selected: false -tracksRegLiveness: true -liveins: - - { reg: '%ebp' } - - { reg: '%eax' } -frameInfo: - isFrameAddressTaken: false - isReturnAddressTaken: false - hasStackMap: false - hasPatchPoint: false - stackSize: 0 - offsetAdjustment: 0 - maxAlignment: 0 - adjustsStack: false - hasCalls: false - maxCallFrameSize: 0 - hasOpaqueSPAdjustment: false - hasVAStart: false - hasMustTailInVarArgFunc: false -body: | - bb.0 (%ir-block.0): - liveins: %eax, %ebp, %ebx - ; CHECK: %ebx = LEA32r killed %eax, 4, killed %eax, 5, _ - ; CHECK: %ebp = LEA32r killed %ebx, 4, killed %ebx, 0, _ - ; CHECK: %ebp = ADD32ri8 %ebp, 5 - - CMP32rr %eax, killed %ebx, implicit-def %eflags - %ebx = LEA32r killed %eax, 4, killed %eax, 5, _ - JE_1 %bb.1, implicit %eflags - RETQ %ebx - bb.1: - liveins: %eax, %ebp, %ebx - %ebp = LEA32r killed %ebx, 4, killed %ebx, 5, _ - RETQ %ebp - -... - - - diff --git a/test/CodeGen/X86/leaFixup64.mir b/test/CodeGen/X86/leaFixup64.mir deleted file mode 100644 index 9b0058750598..000000000000 --- a/test/CodeGen/X86/leaFixup64.mir +++ /dev/null @@ -1,1041 +0,0 @@ -# RUN: llc -run-pass x86-fixup-LEAs -mcpu=corei7-avx -o - %s | FileCheck %s ---- | - ; ModuleID = 'lea-2.ll' - source_filename = "lea-2.ll" - target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128" - ;generated using: llc -stop-after x86-pad-short-functions lea-2.ll > leaFinxup64.mir - - ;testleaadd_64_32_1: 3 operands LEA64_32r cannot be replaced with 2 add instructions - ; but can be replaced with 1 lea + 1 add - define i32 @testleaadd_64_32_1() { - ret i32 0 - } - - ;testleaadd_rbp_64_32_1: 3 operands LEA64_32r cannot be replaced with 2 add instructions - ; where the base is rbp/r13/ebp register but it can be replaced with 1 lea + 1 add - define i32 @testleaadd_rbp_64_32_1() { - ret i32 0 - } - - ;test1lea_rbp_64_32_1: 2 operands LEA64_32r where base register is rbp/r13/ebp and can not - ; be replaced with an add instruction but can be replaced with 1 lea instruction - define i32 @test1lea_rbp_64_32_1() { - ret i32 0 - } - - ;test2add_64: 3 operands LEA64r that can be replaced with 2 add instructions - define i32 @test2add_64() { - ret i32 0 - } - - ;test2add_rbp_64: 3 operands LEA64r that can be replaced with 2 add instructions - ; where the base is rbp/r13/ebp register - define i32 @test2add_rbp_64() { - ret i32 0 - } - - ;test1add_rbp_64: 2 operands LEA64r where base register is rbp/r13/ebp and can be replaced - ; with an add instruction - define i32 @test1add_rbp_64() { - ret i32 0 - } - - ;testleaadd_64_32: 3 operands LEA64_32r that can be replaced with 1 lea 1 add instructions - define i32 @testleaadd_64_32() { - ret i32 0 - } - - ;testleaadd_rbp_64_32: 3 operands LEA64_32r that can be replaced with 1 lea 1 add instructions - ; where the base is rbp/r13/ebp register - define i32 @testleaadd_rbp_64_32() { - ret i32 0 - } - - ;test1lea_rbp_64_32: 2 operands LEA64_32r where base register is rbp/r13/ebp and can be replaced - ; with a lea instruction - define i32 @test1lea_rbp_64_32() { - ret i32 0 - } - - ;testleaadd_64: 3 operands LEA64r that can be replaced with 1 lea 1 add instructions - define i32 @testleaadd_64() { - ret i32 0 - } - - ;testleaadd_rbp_64: 3 operands LEA64r that can be replaced with 1 lea 1 add instructions - ; where the base is rbp/r13/ebp register - define i32 @testleaadd_rbp_64() { - ret i32 0 - } - - ;test1lea_rbp_64: 2 operands LEA64r wher base register is rbp/r13/ebp and can be replaced - ; with a lea instruction - define i32 @test1lea_rbp_64() { - ret i32 0 - } - - ;test8: dst = base & scale!=1, can't optimize - define i32 @test8() { - ret i32 0 - } - - ;testleaaddi32_64_32: 3 operands LEA64_32r that can be replaced with 1 lea + 1 add instructions where - ; ADD64ri32 is chosen - define i32 @testleaaddi32_64_32() { - ret i32 0 - } - - ;test1mov1add_rbp_64_32: 2 operands LEA64_32r cannot be replaced with 1 add 1 mov instructions - ; where the base is rbp/r13/ebp register - define i32 @test1mov1add_rbp_64_32() { - ret i32 0 - } - - ;testleaadd_rbp_index_64_32: 3 operands LEA64_32r that cannot replaced with 1 lea 1 add instructions - ; where the base and the index are ebp register and there is offset - define i32 @testleaadd_rbp_index_64_32() { - ret i32 0 - } - - ;testleaadd_rbp_index2_64_32: 3 operands LEA64_32r that cannot replaced with 1 lea 1 add instructions - ; where the base and the index are ebp register and there is scale - define i32 @testleaadd_rbp_index2_64_32() { - ret i32 0 - } - - ;test2addi32_64: 3 operands LEA64r that can be replaced with 2 add instructions where ADD64ri32 - ; is chosen - define i32 @test2addi32_64() { - ret i32 0 - } - - ;test1mov1add_rbp_64: 2 operands LEA64r that can be replaced with 1 add 1 mov instructions - ; where the base is rbp/r13/ebp register - define i32 @test1mov1add_rbp_64() { - ret i32 0 - } - - ;testleaadd_rbp_index_64: 3 operands LEA64r that can be replaced with 1 lea 1 add instructions - ; where the base and the index are ebp register and there is offset - define i32 @testleaadd_rbp_index_64() { - ret i32 0 - } - - ;testleaadd_rbp_index2_64: 3 operands LEA64r that can be replaced with 1 lea 1 add instructions - ; where the base and the index are ebp register and there is scale - define i32 @testleaadd_rbp_index2_64() { - ret i32 0 - } - - ;test_skip_opt_64: 3 operands LEA64r that can not be replaced with 2 instructions - define i32 @test_skip_opt_64() { - ret i32 0 - } - - ;test_skip_eflags_64: LEA64r that cannot be replaced since its not safe to clobber eflags - define i32 @test_skip_eflags_64() { - ret i32 0 - } - - ;test_skip_opt_64_32: 3 operands LEA64_32r that can not be replaced with 2 instructions - define i32 @test_skip_opt_64_32() { - ret i32 0 - } - - ;test_skip_eflags_64_32: LEA64_32r that cannot be replaced since its not safe to clobber eflags - define i32 @test_skip_eflags_64_32() { - ret i32 0 - } - - -... ---- -name: testleaadd_64_32_1 -alignment: 4 -exposesReturnsTwice: false -legalized: false -regBankSelected: false -selected: false -tracksRegLiveness: true -liveins: - - { reg: '%rax' } - - { reg: '%rbp' } -frameInfo: - isFrameAddressTaken: false - isReturnAddressTaken: false - hasStackMap: false - hasPatchPoint: false - stackSize: 0 - offsetAdjustment: 0 - maxAlignment: 0 - adjustsStack: false - hasCalls: false - maxCallFrameSize: 0 - hasOpaqueSPAdjustment: false - hasVAStart: false - hasMustTailInVarArgFunc: false -body: | - bb.0 (%ir-block.0): - liveins: %rax, %rbp - ; CHECK: %eax = LEA64_32r killed %rax, 1, killed %rbp, 0 - ; CHECK: %eax = ADD32ri8 %eax, -5 - - %eax = LEA64_32r killed %rax, 1, killed %rbp, -5, _ - RETQ %eax - -... ---- -name: testleaadd_rbp_64_32_1 -alignment: 4 -exposesReturnsTwice: false -legalized: false -regBankSelected: false -selected: false -tracksRegLiveness: true -liveins: - - { reg: '%rax' } - - { reg: '%rbp' } -frameInfo: - isFrameAddressTaken: false - isReturnAddressTaken: false - hasStackMap: false - hasPatchPoint: false - stackSize: 0 - offsetAdjustment: 0 - maxAlignment: 0 - adjustsStack: false - hasCalls: false - maxCallFrameSize: 0 - hasOpaqueSPAdjustment: false - hasVAStart: false - hasMustTailInVarArgFunc: false -body: | - bb.0 (%ir-block.0): - liveins: %rax, %rbp - ; CHECK: %ebp = LEA64_32r killed %rax, 1, killed %rbp, 0 - ; CHECK: %ebp = ADD32ri8 %ebp, -5 - - %ebp = LEA64_32r killed %rbp, 1, killed %rax, -5, _ - RETQ %ebp - -... ---- -name: test1lea_rbp_64_32_1 -alignment: 4 -exposesReturnsTwice: false -legalized: false -regBankSelected: false -selected: false -tracksRegLiveness: true -liveins: - - { reg: '%rax' } - - { reg: '%rbp' } -frameInfo: - isFrameAddressTaken: false - isReturnAddressTaken: false - hasStackMap: false - hasPatchPoint: false - stackSize: 0 - offsetAdjustment: 0 - maxAlignment: 0 - adjustsStack: false - hasCalls: false - maxCallFrameSize: 0 - hasOpaqueSPAdjustment: false - hasVAStart: false - hasMustTailInVarArgFunc: false -body: | - bb.0 (%ir-block.0): - liveins: %rax, %rbp - ; CHECK: %ebp = LEA64_32r killed %rax, 1, killed %rbp, 0 - - %ebp = LEA64_32r killed %rbp, 1, killed %rax, 0, _ - RETQ %ebp - -... ---- -name: test2add_64 -alignment: 4 -exposesReturnsTwice: false -legalized: false -regBankSelected: false -selected: false -tracksRegLiveness: true -liveins: - - { reg: '%rax' } - - { reg: '%rbp' } -frameInfo: - isFrameAddressTaken: false - isReturnAddressTaken: false - hasStackMap: false - hasPatchPoint: false - stackSize: 0 - offsetAdjustment: 0 - maxAlignment: 0 - adjustsStack: false - hasCalls: false - maxCallFrameSize: 0 - hasOpaqueSPAdjustment: false - hasVAStart: false - hasMustTailInVarArgFunc: false -body: | - bb.0 (%ir-block.0): - liveins: %rax, %rbp - ; CHECK: %rax = ADD64rr %rax, killed %rbp - ; CHECK: %rax = ADD64ri8 %rax, -5 - - %rax = LEA64r killed %rax, 1, killed %rbp, -5, _ - RETQ %eax - -... ---- -name: test2add_rbp_64 -alignment: 4 -exposesReturnsTwice: false -legalized: false -regBankSelected: false -selected: false -tracksRegLiveness: true -liveins: - - { reg: '%rax' } - - { reg: '%rbp' } -frameInfo: - isFrameAddressTaken: false - isReturnAddressTaken: false - hasStackMap: false - hasPatchPoint: false - stackSize: 0 - offsetAdjustment: 0 - maxAlignment: 0 - adjustsStack: false - hasCalls: false - maxCallFrameSize: 0 - hasOpaqueSPAdjustment: false - hasVAStart: false - hasMustTailInVarArgFunc: false -body: | - bb.0 (%ir-block.0): - liveins: %rax, %rbp - ; CHECK: %rbp = ADD64rr %rbp, killed %rax - ; CHECK: %rbp = ADD64ri8 %rbp, -5 - - %rbp = LEA64r killed %rbp, 1, killed %rax, -5, _ - RETQ %ebp - -... ---- -name: test1add_rbp_64 -alignment: 4 -exposesReturnsTwice: false -legalized: false -regBankSelected: false -selected: false -tracksRegLiveness: true -liveins: - - { reg: '%rax' } - - { reg: '%rbp' } -frameInfo: - isFrameAddressTaken: false - isReturnAddressTaken: false - hasStackMap: false - hasPatchPoint: false - stackSize: 0 - offsetAdjustment: 0 - maxAlignment: 0 - adjustsStack: false - hasCalls: false - maxCallFrameSize: 0 - hasOpaqueSPAdjustment: false - hasVAStart: false - hasMustTailInVarArgFunc: false -body: | - bb.0 (%ir-block.0): - liveins: %rax, %rbp - ; CHECK: %rbp = ADD64rr %rbp, killed %rax - - %rbp = LEA64r killed %rbp, 1, killed %rax, 0, _ - RETQ %ebp - -... ---- -name: testleaadd_64_32 -alignment: 4 -exposesReturnsTwice: false -legalized: false -regBankSelected: false -selected: false -tracksRegLiveness: true -liveins: - - { reg: '%rax' } - - { reg: '%rbp' } - - { reg: '%rbx' } -frameInfo: - isFrameAddressTaken: false - isReturnAddressTaken: false - hasStackMap: false - hasPatchPoint: false - stackSize: 0 - offsetAdjustment: 0 - maxAlignment: 0 - adjustsStack: false - hasCalls: false - maxCallFrameSize: 0 - hasOpaqueSPAdjustment: false - hasVAStart: false - hasMustTailInVarArgFunc: false -body: | - bb.0 (%ir-block.0): - liveins: %rax, %rbp - ; CHECK: %ebx = LEA64_32r killed %rax, 1, killed %rbp, 0, _ - ; CHECK: %ebx = ADD32ri8 %ebx, -5 - - %ebx = LEA64_32r killed %rax, 1, killed %rbp, -5, _ - RETQ %ebx - -... ---- -name: testleaadd_rbp_64_32 -alignment: 4 -exposesReturnsTwice: false -legalized: false -regBankSelected: false -selected: false -tracksRegLiveness: true -liveins: - - { reg: '%rax' } - - { reg: '%rbp' } - - { reg: '%rbx' } -frameInfo: - isFrameAddressTaken: false - isReturnAddressTaken: false - hasStackMap: false - hasPatchPoint: false - stackSize: 0 - offsetAdjustment: 0 - maxAlignment: 0 - adjustsStack: false - hasCalls: false - maxCallFrameSize: 0 - hasOpaqueSPAdjustment: false - hasVAStart: false - hasMustTailInVarArgFunc: false -body: | - bb.0 (%ir-block.0): - liveins: %rax, %rbp - ; CHECK: %ebx = LEA64_32r killed %rax, 1, killed %rbp, 0, _ - ; CHECK: %ebx = ADD32ri8 %ebx, -5 - - %ebx = LEA64_32r killed %rbp, 1, killed %rax, -5, _ - RETQ %ebx - -... ---- -name: test1lea_rbp_64_32 -alignment: 4 -exposesReturnsTwice: false -legalized: false -regBankSelected: false -selected: false -tracksRegLiveness: true -liveins: - - { reg: '%rax' } - - { reg: '%rbp' } - - { reg: '%rbx' } -frameInfo: - isFrameAddressTaken: false - isReturnAddressTaken: false - hasStackMap: false - hasPatchPoint: false - stackSize: 0 - offsetAdjustment: 0 - maxAlignment: 0 - adjustsStack: false - hasCalls: false - maxCallFrameSize: 0 - hasOpaqueSPAdjustment: false - hasVAStart: false - hasMustTailInVarArgFunc: false -body: | - bb.0 (%ir-block.0): - liveins: %rax, %rbp - ; CHECK: %ebx = LEA64_32r killed %rax, 1, killed %rbp, 0, _ - - %ebx = LEA64_32r killed %rbp, 1, killed %rax, 0, _ - RETQ %ebx - -... ---- -name: testleaadd_64 -alignment: 4 -exposesReturnsTwice: false -legalized: false -regBankSelected: false -selected: false -tracksRegLiveness: true -liveins: - - { reg: '%rax' } - - { reg: '%rbp' } - - { reg: '%rbx' } -frameInfo: - isFrameAddressTaken: false - isReturnAddressTaken: false - hasStackMap: false - hasPatchPoint: false - stackSize: 0 - offsetAdjustment: 0 - maxAlignment: 0 - adjustsStack: false - hasCalls: false - maxCallFrameSize: 0 - hasOpaqueSPAdjustment: false - hasVAStart: false - hasMustTailInVarArgFunc: false -body: | - bb.0 (%ir-block.0): - liveins: %rax, %rbp - ; CHECK: %rbx = LEA64r killed %rax, 1, killed %rbp, 0, _ - ; CHECK: %rbx = ADD64ri8 %rbx, -5 - - %rbx = LEA64r killed %rax, 1, killed %rbp, -5, _ - RETQ %ebx - -... ---- -name: testleaadd_rbp_64 -alignment: 4 -exposesReturnsTwice: false -legalized: false -regBankSelected: false -selected: false -tracksRegLiveness: true -liveins: - - { reg: '%rax' } - - { reg: '%rbp' } - - { reg: '%rbx' } -frameInfo: - isFrameAddressTaken: false - isReturnAddressTaken: false - hasStackMap: false - hasPatchPoint: false - stackSize: 0 - offsetAdjustment: 0 - maxAlignment: 0 - adjustsStack: false - hasCalls: false - maxCallFrameSize: 0 - hasOpaqueSPAdjustment: false - hasVAStart: false - hasMustTailInVarArgFunc: false -body: | - bb.0 (%ir-block.0): - liveins: %rax, %rbp - ; CHECK: %rbx = LEA64r killed %rax, 1, killed %rbp, 0, _ - ; CHECK: %rbx = ADD64ri8 %rbx, -5 - - %rbx = LEA64r killed %rbp, 1, killed %rax, -5, _ - RETQ %ebx - -... ---- -name: test1lea_rbp_64 -alignment: 4 -exposesReturnsTwice: false -legalized: false -regBankSelected: false -selected: false -tracksRegLiveness: true -liveins: - - { reg: '%rax' } - - { reg: '%rbp' } - - { reg: '%rbx' } -frameInfo: - isFrameAddressTaken: false - isReturnAddressTaken: false - hasStackMap: false - hasPatchPoint: false - stackSize: 0 - offsetAdjustment: 0 - maxAlignment: 0 - adjustsStack: false - hasCalls: false - maxCallFrameSize: 0 - hasOpaqueSPAdjustment: false - hasVAStart: false - hasMustTailInVarArgFunc: false -body: | - bb.0 (%ir-block.0): - liveins: %rax, %rbp - ; CHECK: %rbx = LEA64r killed %rax, 1, killed %rbp, 0, _ - - %rbx = LEA64r killed %rbp, 1, killed %rax, 0, _ - RETQ %ebx - -... ---- -name: test8 -alignment: 4 -exposesReturnsTwice: false -legalized: false -regBankSelected: false -selected: false -tracksRegLiveness: true -liveins: - - { reg: '%rdi' } - - { reg: '%rbp' } -frameInfo: - isFrameAddressTaken: false - isReturnAddressTaken: false - hasStackMap: false - hasPatchPoint: false - stackSize: 0 - offsetAdjustment: 0 - maxAlignment: 0 - adjustsStack: false - hasCalls: false - maxCallFrameSize: 0 - hasOpaqueSPAdjustment: false - hasVAStart: false - hasMustTailInVarArgFunc: false -body: | - bb.0 (%ir-block.0): - liveins: %rdi, %rbp - ; CHECK: %r12 = LEA64r _, 2, killed %r13, 5, _ - ; CHECK: %r12 = ADD64rr %r12, killed %rbp - %rbp = KILL %rbp, implicit-def %rbp - %r13 = KILL %rdi, implicit-def %r13 - %r12 = LEA64r killed %rbp, 2, killed %r13, 5, _ - RETQ %r12 - -... ---- -name: testleaaddi32_64_32 -alignment: 4 -exposesReturnsTwice: false -legalized: false -regBankSelected: false -selected: false -tracksRegLiveness: true -liveins: - - { reg: '%rax' } - - { reg: '%rbp' } -frameInfo: - isFrameAddressTaken: false - isReturnAddressTaken: false - hasStackMap: false - hasPatchPoint: false - stackSize: 0 - offsetAdjustment: 0 - maxAlignment: 0 - adjustsStack: false - hasCalls: false - maxCallFrameSize: 0 - hasOpaqueSPAdjustment: false - hasVAStart: false - hasMustTailInVarArgFunc: false -body: | - bb.0 (%ir-block.0): - liveins: %rax, %rbp - ; CHECK: %eax = LEA64_32r killed %rax, 1, killed %rbp, 0 - ; CHECK: %eax = ADD32ri %eax, 129 - - %eax = LEA64_32r killed %rax, 1, killed %rbp, 129, _ - RETQ %eax - -... ---- -name: test1mov1add_rbp_64_32 -alignment: 4 -exposesReturnsTwice: false -legalized: false -regBankSelected: false -selected: false -tracksRegLiveness: true -liveins: - - { reg: '%rax' } - - { reg: '%rbp' } -frameInfo: - isFrameAddressTaken: false - isReturnAddressTaken: false - hasStackMap: false - hasPatchPoint: false - stackSize: 0 - offsetAdjustment: 0 - maxAlignment: 0 - adjustsStack: false - hasCalls: false - maxCallFrameSize: 0 - hasOpaqueSPAdjustment: false - hasVAStart: false - hasMustTailInVarArgFunc: false -body: | - bb.0 (%ir-block.0): - liveins: %rax, %rbp, %rbx - ; CHECK: %ebx = LEA64_32r killed %rbp, 1, killed %rbp, 0, _ - - %ebx = LEA64_32r killed %rbp, 1, killed %rbp, 0, _ - RETQ %ebx - -... ---- -name: testleaadd_rbp_index_64_32 -alignment: 4 -exposesReturnsTwice: false -legalized: false -regBankSelected: false -selected: false -tracksRegLiveness: true -liveins: - - { reg: '%rbx' } - - { reg: '%rbp' } -frameInfo: - isFrameAddressTaken: false - isReturnAddressTaken: false - hasStackMap: false - hasPatchPoint: false - stackSize: 0 - offsetAdjustment: 0 - maxAlignment: 0 - adjustsStack: false - hasCalls: false - maxCallFrameSize: 0 - hasOpaqueSPAdjustment: false - hasVAStart: false - hasMustTailInVarArgFunc: false -body: | - bb.0 (%ir-block.0): - liveins: %rax, %rbp, %rbx - ; CHECK: %ebx = LEA64_32r killed %rbp, 1, killed %rbp, 5, _ - - %ebx = LEA64_32r killed %rbp, 1, killed %rbp, 5, _ - RETQ %ebx - -... ---- -name: testleaadd_rbp_index2_64_32 -alignment: 4 -exposesReturnsTwice: false -legalized: false -regBankSelected: false -selected: false -tracksRegLiveness: true -liveins: - - { reg: '%rbx' } - - { reg: '%rbp' } -frameInfo: - isFrameAddressTaken: false - isReturnAddressTaken: false - hasStackMap: false - hasPatchPoint: false - stackSize: 0 - offsetAdjustment: 0 - maxAlignment: 0 - adjustsStack: false - hasCalls: false - maxCallFrameSize: 0 - hasOpaqueSPAdjustment: false - hasVAStart: false - hasMustTailInVarArgFunc: false -body: | - bb.0 (%ir-block.0): - liveins: %eax, %ebp, %ebx - ; CHECK: %ebx = LEA64_32r killed %rbp, 4, killed %rbp, 5, _ - - %ebx = LEA64_32r killed %rbp, 4, killed %rbp, 5, _ - RETQ %ebx - -... ---- -name: test2addi32_64 -alignment: 4 -exposesReturnsTwice: false -legalized: false -regBankSelected: false -selected: false -tracksRegLiveness: true -liveins: - - { reg: '%rax' } - - { reg: '%rbp' } -frameInfo: - isFrameAddressTaken: false - isReturnAddressTaken: false - hasStackMap: false - hasPatchPoint: false - stackSize: 0 - offsetAdjustment: 0 - maxAlignment: 0 - adjustsStack: false - hasCalls: false - maxCallFrameSize: 0 - hasOpaqueSPAdjustment: false - hasVAStart: false - hasMustTailInVarArgFunc: false -body: | - bb.0 (%ir-block.0): - liveins: %rax, %rbp - ; CHECK: %rax = ADD64rr %rax, killed %rbp - ; CHECK: %rax = ADD64ri32 %rax, 129 - - %rax = LEA64r killed %rax, 1, killed %rbp, 129, _ - RETQ %eax - -... ---- -name: test1mov1add_rbp_64 -alignment: 4 -exposesReturnsTwice: false -legalized: false -regBankSelected: false -selected: false -tracksRegLiveness: true -liveins: - - { reg: '%rax' } - - { reg: '%rbp' } -frameInfo: - isFrameAddressTaken: false - isReturnAddressTaken: false - hasStackMap: false - hasPatchPoint: false - stackSize: 0 - offsetAdjustment: 0 - maxAlignment: 0 - adjustsStack: false - hasCalls: false - maxCallFrameSize: 0 - hasOpaqueSPAdjustment: false - hasVAStart: false - hasMustTailInVarArgFunc: false -body: | - bb.0 (%ir-block.0): - liveins: %rax, %rbp, %rbx - ; CHECK: %rbx = MOV64rr killed %rbp - ; CHECK: %rbx = ADD64rr %rbx, killed %rbp - - %rbx = LEA64r killed %rbp, 1, killed %rbp, 0, _ - RETQ %ebx - -... ---- -name: testleaadd_rbp_index_64 -alignment: 4 -exposesReturnsTwice: false -legalized: false -regBankSelected: false -selected: false -tracksRegLiveness: true -liveins: - - { reg: '%rbx' } - - { reg: '%rbp' } -frameInfo: - isFrameAddressTaken: false - isReturnAddressTaken: false - hasStackMap: false - hasPatchPoint: false - stackSize: 0 - offsetAdjustment: 0 - maxAlignment: 0 - adjustsStack: false - hasCalls: false - maxCallFrameSize: 0 - hasOpaqueSPAdjustment: false - hasVAStart: false - hasMustTailInVarArgFunc: false -body: | - bb.0 (%ir-block.0): - liveins: %rax, %rbp, %rbx - ; CHECK: %rbx = LEA64r _, 1, killed %rbp, 5, _ - ; CHECK: %rbx = ADD64rr %rbx, killed %rbp - - %rbx = LEA64r killed %rbp, 1, killed %rbp, 5, _ - RETQ %ebx - -... ---- -name: testleaadd_rbp_index2_64 -alignment: 4 -exposesReturnsTwice: false -legalized: false -regBankSelected: false -selected: false -tracksRegLiveness: true -liveins: - - { reg: '%rbx' } - - { reg: '%rbp' } -frameInfo: - isFrameAddressTaken: false - isReturnAddressTaken: false - hasStackMap: false - hasPatchPoint: false - stackSize: 0 - offsetAdjustment: 0 - maxAlignment: 0 - adjustsStack: false - hasCalls: false - maxCallFrameSize: 0 - hasOpaqueSPAdjustment: false - hasVAStart: false - hasMustTailInVarArgFunc: false -body: | - bb.0 (%ir-block.0): - liveins: %rax, %rbp, %rbx - ; CHECK: %rbx = LEA64r _, 4, killed %rbp, 5, _ - ; CHECK: %rbx = ADD64rr %rbx, killed %rbp - - %rbx = LEA64r killed %rbp, 4, killed %rbp, 5, _ - RETQ %ebx - -... ---- -name: test_skip_opt_64 -alignment: 4 -exposesReturnsTwice: false -legalized: false -regBankSelected: false -selected: false -tracksRegLiveness: true -liveins: - - { reg: '%rbx' } - - { reg: '%rbp' } -frameInfo: - isFrameAddressTaken: false - isReturnAddressTaken: false - hasStackMap: false - hasPatchPoint: false - stackSize: 0 - offsetAdjustment: 0 - maxAlignment: 0 - adjustsStack: false - hasCalls: false - maxCallFrameSize: 0 - hasOpaqueSPAdjustment: false - hasVAStart: false - hasMustTailInVarArgFunc: false -body: | - bb.0 (%ir-block.0): - liveins: %rax, %rbp, %rbx - ; CHECK: %rbp = LEA64r killed %rbp, 4, killed %rbp, 0, _ - - %rbp = LEA64r killed %rbp, 4, killed %rbp, 0, _ - RETQ %ebp - -... ---- -name: test_skip_eflags_64 -alignment: 4 -exposesReturnsTwice: false -legalized: false -regBankSelected: false -selected: false -tracksRegLiveness: true -liveins: - - { reg: '%rbp' } - - { reg: '%rax' } -frameInfo: - isFrameAddressTaken: false - isReturnAddressTaken: false - hasStackMap: false - hasPatchPoint: false - stackSize: 0 - offsetAdjustment: 0 - maxAlignment: 0 - adjustsStack: false - hasCalls: false - maxCallFrameSize: 0 - hasOpaqueSPAdjustment: false - hasVAStart: false - hasMustTailInVarArgFunc: false -body: | - bb.0 (%ir-block.0): - liveins: %rax, %rbp, %rbx - ; CHECK: %rbx = LEA64r killed %rax, 4, killed %rax, 5, _ - ; CHECK: %rbp = LEA64r killed %rbx, 4, killed %rbx, 0, _ - ; CHECK: %rbp = ADD64ri8 %rbp, 5 - - CMP64rr %rax, killed %rbx, implicit-def %eflags - %rbx = LEA64r killed %rax, 4, killed %rax, 5, _ - JE_1 %bb.1, implicit %eflags - RETQ %ebx - bb.1: - liveins: %rax, %rbp, %rbx - %rbp = LEA64r killed %rbx, 4, killed %rbx, 5, _ - RETQ %ebp - -... ---- -name: test_skip_opt_64_32 -alignment: 4 -exposesReturnsTwice: false -legalized: false -regBankSelected: false -selected: false -tracksRegLiveness: true -liveins: - - { reg: '%rbx' } - - { reg: '%rbp' } -frameInfo: - isFrameAddressTaken: false - isReturnAddressTaken: false - hasStackMap: false - hasPatchPoint: false - stackSize: 0 - offsetAdjustment: 0 - maxAlignment: 0 - adjustsStack: false - hasCalls: false - maxCallFrameSize: 0 - hasOpaqueSPAdjustment: false - hasVAStart: false - hasMustTailInVarArgFunc: false -body: | - bb.0 (%ir-block.0): - liveins: %rax, %rbp, %rbx - ; CHECK: %ebp = LEA64_32r killed %rbp, 4, killed %rbp, 0, _ - - %ebp = LEA64_32r killed %rbp, 4, killed %rbp, 0, _ - RETQ %ebp - -... ---- -name: test_skip_eflags_64_32 -alignment: 4 -exposesReturnsTwice: false -legalized: false -regBankSelected: false -selected: false -tracksRegLiveness: true -liveins: - - { reg: '%rbp' } - - { reg: '%rax' } -frameInfo: - isFrameAddressTaken: false - isReturnAddressTaken: false - hasStackMap: false - hasPatchPoint: false - stackSize: 0 - offsetAdjustment: 0 - maxAlignment: 0 - adjustsStack: false - hasCalls: false - maxCallFrameSize: 0 - hasOpaqueSPAdjustment: false - hasVAStart: false - hasMustTailInVarArgFunc: false -body: | - bb.0 (%ir-block.0): - liveins: %rax, %rbp, %rbx - ; CHECK: %ebx = LEA64_32r killed %rax, 4, killed %rax, 5, _ - ; CHECK: %ebp = LEA64_32r killed %rbx, 4, killed %rbx, 0, _ - ; CHECK: %ebp = ADD32ri8 %ebp, 5 - - CMP64rr %rax, killed %rbx, implicit-def %eflags - %ebx = LEA64_32r killed %rax, 4, killed %rax, 5, _ - JE_1 %bb.1, implicit %eflags - RETQ %ebx - bb.1: - liveins: %rax, %rbp, %rbx - %ebp = LEA64_32r killed %rbx, 4, killed %rbx, 5, _ - RETQ %ebp - -... - - - diff --git a/test/CodeGen/X86/nontemporal.ll b/test/CodeGen/X86/nontemporal.ll index 33d5caba597c..d49c88724331 100644 --- a/test/CodeGen/X86/nontemporal.ll +++ b/test/CodeGen/X86/nontemporal.ll @@ -9,33 +9,29 @@ define void @f(<4 x float> %A, i8* %B, <2 x double> %C, i32 %D, <2 x i64> %E, <4 ; X32-SSE: # BB#0: ; X32-SSE-NEXT: pushl %ebp ; X32-SSE-NEXT: movl %esp, %ebp -; X32-SSE-NEXT: pushl %esi ; X32-SSE-NEXT: andl $-16, %esp ; X32-SSE-NEXT: subl $16, %esp ; X32-SSE-NEXT: movl 72(%ebp), %eax ; X32-SSE-NEXT: movl 76(%ebp), %ecx -; X32-SSE-NEXT: movl 12(%ebp), %edx ; X32-SSE-NEXT: movdqa 56(%ebp), %xmm3 ; X32-SSE-NEXT: movdqa 40(%ebp), %xmm4 ; X32-SSE-NEXT: movdqa 24(%ebp), %xmm5 -; X32-SSE-NEXT: movl 8(%ebp), %esi -; X32-SSE-NEXT: addps .LCPI0_0, %xmm0 -; X32-SSE-NEXT: movntps %xmm0, (%esi) -; X32-SSE-NEXT: paddq .LCPI0_1, %xmm2 -; X32-SSE-NEXT: movntdq %xmm2, (%esi) -; X32-SSE-NEXT: addpd .LCPI0_2, %xmm1 -; X32-SSE-NEXT: movntpd %xmm1, (%esi) -; X32-SSE-NEXT: paddd .LCPI0_3, %xmm5 -; X32-SSE-NEXT: movntdq %xmm5, (%esi) -; X32-SSE-NEXT: paddw .LCPI0_4, %xmm4 -; X32-SSE-NEXT: movntdq %xmm4, (%esi) -; X32-SSE-NEXT: paddb .LCPI0_5, %xmm3 -; X32-SSE-NEXT: movntdq %xmm3, (%esi) -; X32-SSE-NEXT: movntil %edx, (%esi) -; X32-SSE-NEXT: movntil %ecx, 4(%esi) -; X32-SSE-NEXT: movntil %eax, (%esi) -; X32-SSE-NEXT: leal -4(%ebp), %esp -; X32-SSE-NEXT: popl %esi +; X32-SSE-NEXT: movl 8(%ebp), %edx +; X32-SSE-NEXT: addps {{\.LCPI.*}}, %xmm0 +; X32-SSE-NEXT: movntps %xmm0, (%edx) +; X32-SSE-NEXT: paddq {{\.LCPI.*}}, %xmm2 +; X32-SSE-NEXT: movntdq %xmm2, (%edx) +; X32-SSE-NEXT: addpd {{\.LCPI.*}}, %xmm1 +; X32-SSE-NEXT: movntpd %xmm1, (%edx) +; X32-SSE-NEXT: paddd {{\.LCPI.*}}, %xmm5 +; X32-SSE-NEXT: movntdq %xmm5, (%edx) +; X32-SSE-NEXT: paddw {{\.LCPI.*}}, %xmm4 +; X32-SSE-NEXT: movntdq %xmm4, (%edx) +; X32-SSE-NEXT: paddb {{\.LCPI.*}}, %xmm3 +; X32-SSE-NEXT: movntdq %xmm3, (%edx) +; X32-SSE-NEXT: movntil %ecx, 4(%edx) +; X32-SSE-NEXT: movntil %eax, (%edx) +; X32-SSE-NEXT: movl %ebp, %esp ; X32-SSE-NEXT: popl %ebp ; X32-SSE-NEXT: retl ; @@ -43,33 +39,29 @@ define void @f(<4 x float> %A, i8* %B, <2 x double> %C, i32 %D, <2 x i64> %E, <4 ; X32-AVX: # BB#0: ; X32-AVX-NEXT: pushl %ebp ; X32-AVX-NEXT: movl %esp, %ebp -; X32-AVX-NEXT: pushl %esi ; X32-AVX-NEXT: andl $-16, %esp ; X32-AVX-NEXT: subl $16, %esp ; X32-AVX-NEXT: movl 72(%ebp), %eax ; X32-AVX-NEXT: movl 76(%ebp), %ecx -; X32-AVX-NEXT: movl 12(%ebp), %edx ; X32-AVX-NEXT: vmovdqa 56(%ebp), %xmm3 ; X32-AVX-NEXT: vmovdqa 40(%ebp), %xmm4 ; X32-AVX-NEXT: vmovdqa 24(%ebp), %xmm5 -; X32-AVX-NEXT: movl 8(%ebp), %esi -; X32-AVX-NEXT: vaddps .LCPI0_0, %xmm0, %xmm0 -; X32-AVX-NEXT: vmovntps %xmm0, (%esi) -; X32-AVX-NEXT: vpaddq .LCPI0_1, %xmm2, %xmm0 -; X32-AVX-NEXT: vmovntdq %xmm0, (%esi) -; X32-AVX-NEXT: vaddpd .LCPI0_2, %xmm1, %xmm0 -; X32-AVX-NEXT: vmovntpd %xmm0, (%esi) -; X32-AVX-NEXT: vpaddd .LCPI0_3, %xmm5, %xmm0 -; X32-AVX-NEXT: vmovntdq %xmm0, (%esi) -; X32-AVX-NEXT: vpaddw .LCPI0_4, %xmm4, %xmm0 -; X32-AVX-NEXT: vmovntdq %xmm0, (%esi) -; X32-AVX-NEXT: vpaddb .LCPI0_5, %xmm3, %xmm0 -; X32-AVX-NEXT: vmovntdq %xmm0, (%esi) -; X32-AVX-NEXT: movntil %edx, (%esi) -; X32-AVX-NEXT: movntil %ecx, 4(%esi) -; X32-AVX-NEXT: movntil %eax, (%esi) -; X32-AVX-NEXT: leal -4(%ebp), %esp -; X32-AVX-NEXT: popl %esi +; X32-AVX-NEXT: movl 8(%ebp), %edx +; X32-AVX-NEXT: vaddps {{\.LCPI.*}}, %xmm0, %xmm0 +; X32-AVX-NEXT: vmovntps %xmm0, (%edx) +; X32-AVX-NEXT: vpaddq {{\.LCPI.*}}, %xmm2, %xmm0 +; X32-AVX-NEXT: vmovntdq %xmm0, (%edx) +; X32-AVX-NEXT: vaddpd {{\.LCPI.*}}, %xmm1, %xmm0 +; X32-AVX-NEXT: vmovntpd %xmm0, (%edx) +; X32-AVX-NEXT: vpaddd {{\.LCPI.*}}, %xmm5, %xmm0 +; X32-AVX-NEXT: vmovntdq %xmm0, (%edx) +; X32-AVX-NEXT: vpaddw {{\.LCPI.*}}, %xmm4, %xmm0 +; X32-AVX-NEXT: vmovntdq %xmm0, (%edx) +; X32-AVX-NEXT: vpaddb {{\.LCPI.*}}, %xmm3, %xmm0 +; X32-AVX-NEXT: vmovntdq %xmm0, (%edx) +; X32-AVX-NEXT: movntil %ecx, 4(%edx) +; X32-AVX-NEXT: movntil %eax, (%edx) +; X32-AVX-NEXT: movl %ebp, %esp ; X32-AVX-NEXT: popl %ebp ; X32-AVX-NEXT: retl ; diff --git a/test/CodeGen/X86/psubus.ll b/test/CodeGen/X86/psubus.ll index 35f96eda35e1..a1f1e084d330 100644 --- a/test/CodeGen/X86/psubus.ll +++ b/test/CodeGen/X86/psubus.ll @@ -1,219 +1,169 @@ ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py ; RUN: llc < %s -mtriple=x86_64-apple-macosx10.8.0 -mattr=+sse2 | FileCheck %s --check-prefix=SSE --check-prefix=SSE2 ; RUN: llc < %s -mtriple=x86_64-apple-macosx10.8.0 -mattr=+ssse3 | FileCheck %s --check-prefix=SSE --check-prefix=SSSE3 +; RUN: llc < %s -mtriple=x86_64-apple-macosx10.8.0 -mattr=+sse4.1 | FileCheck %s --check-prefix=SSE --check-prefix=SSE41 ; RUN: llc < %s -mtriple=x86_64-apple-macosx10.8.0 -mattr=+avx | FileCheck %s --check-prefix=AVX --check-prefix=AVX1 ; RUN: llc < %s -mtriple=x86_64-apple-macosx10.8.0 -mattr=+avx2 | FileCheck %s --check-prefix=AVX --check-prefix=AVX2 -define void @test1(i16* nocapture %head) nounwind { +define <8 x i16> @test1(<8 x i16> %x) nounwind { ; SSE-LABEL: test1: ; SSE: ## BB#0: ## %vector.ph -; SSE-NEXT: movdqu (%rdi), %xmm0 ; SSE-NEXT: psubusw {{.*}}(%rip), %xmm0 -; SSE-NEXT: movdqu %xmm0, (%rdi) ; SSE-NEXT: retq ; ; AVX-LABEL: test1: ; AVX: ## BB#0: ## %vector.ph -; AVX-NEXT: vmovdqu (%rdi), %xmm0 ; AVX-NEXT: vpsubusw {{.*}}(%rip), %xmm0, %xmm0 -; AVX-NEXT: vmovdqu %xmm0, (%rdi) ; AVX-NEXT: retq vector.ph: - %0 = getelementptr inbounds i16, i16* %head, i64 0 - %1 = bitcast i16* %0 to <8 x i16>* - %2 = load <8 x i16>, <8 x i16>* %1, align 2 - %3 = icmp slt <8 x i16> %2, zeroinitializer - %4 = xor <8 x i16> %2, - %5 = select <8 x i1> %3, <8 x i16> %4, <8 x i16> zeroinitializer - store <8 x i16> %5, <8 x i16>* %1, align 2 - ret void + %0 = icmp slt <8 x i16> %x, zeroinitializer + %1 = xor <8 x i16> %x, + %res = select <8 x i1> %0, <8 x i16> %1, <8 x i16> zeroinitializer + ret <8 x i16> %res } -define void @test2(i16* nocapture %head) nounwind { +define <8 x i16> @test2(<8 x i16> %x) nounwind { ; SSE-LABEL: test2: ; SSE: ## BB#0: ## %vector.ph -; SSE-NEXT: movdqu (%rdi), %xmm0 ; SSE-NEXT: psubusw {{.*}}(%rip), %xmm0 -; SSE-NEXT: movdqu %xmm0, (%rdi) ; SSE-NEXT: retq ; ; AVX-LABEL: test2: ; AVX: ## BB#0: ## %vector.ph -; AVX-NEXT: vmovdqu (%rdi), %xmm0 ; AVX-NEXT: vpsubusw {{.*}}(%rip), %xmm0, %xmm0 -; AVX-NEXT: vmovdqu %xmm0, (%rdi) ; AVX-NEXT: retq vector.ph: - %0 = getelementptr inbounds i16, i16* %head, i64 0 - %1 = bitcast i16* %0 to <8 x i16>* - %2 = load <8 x i16>, <8 x i16>* %1, align 2 - %3 = icmp ugt <8 x i16> %2, - %4 = add <8 x i16> %2, - %5 = select <8 x i1> %3, <8 x i16> %4, <8 x i16> zeroinitializer - store <8 x i16> %5, <8 x i16>* %1, align 2 - ret void + %0 = icmp ugt <8 x i16> %x, + %1 = add <8 x i16> %x, + %res = select <8 x i1> %0, <8 x i16> %1, <8 x i16> zeroinitializer + ret <8 x i16> %res } -define void @test3(i16* nocapture %head, i16 zeroext %w) nounwind { +define <8 x i16> @test3(<8 x i16> %x, i16 zeroext %w) nounwind { ; SSE-LABEL: test3: ; SSE: ## BB#0: ## %vector.ph -; SSE-NEXT: movd %esi, %xmm0 -; SSE-NEXT: pshuflw {{.*#+}} xmm0 = xmm0[0,0,0,0,4,5,6,7] -; SSE-NEXT: pshufd {{.*#+}} xmm0 = xmm0[0,0,1,1] -; SSE-NEXT: movdqu (%rdi), %xmm1 -; SSE-NEXT: psubusw %xmm0, %xmm1 -; SSE-NEXT: movdqu %xmm1, (%rdi) +; SSE-NEXT: movd %edi, %xmm1 +; SSE-NEXT: pshuflw {{.*#+}} xmm1 = xmm1[0,0,0,0,4,5,6,7] +; SSE-NEXT: pshufd {{.*#+}} xmm1 = xmm1[0,0,1,1] +; SSE-NEXT: psubusw %xmm1, %xmm0 ; SSE-NEXT: retq ; ; AVX1-LABEL: test3: ; AVX1: ## BB#0: ## %vector.ph -; AVX1-NEXT: vmovd %esi, %xmm0 -; AVX1-NEXT: vpshuflw {{.*#+}} xmm0 = xmm0[0,0,0,0,4,5,6,7] -; AVX1-NEXT: vpshufd {{.*#+}} xmm0 = xmm0[0,0,1,1] -; AVX1-NEXT: vmovdqu (%rdi), %xmm1 -; AVX1-NEXT: vpsubusw %xmm0, %xmm1, %xmm0 -; AVX1-NEXT: vmovdqu %xmm0, (%rdi) +; AVX1-NEXT: vmovd %edi, %xmm1 +; AVX1-NEXT: vpshuflw {{.*#+}} xmm1 = xmm1[0,0,0,0,4,5,6,7] +; AVX1-NEXT: vpshufd {{.*#+}} xmm1 = xmm1[0,0,1,1] +; AVX1-NEXT: vpsubusw %xmm1, %xmm0, %xmm0 ; AVX1-NEXT: retq ; ; AVX2-LABEL: test3: ; AVX2: ## BB#0: ## %vector.ph -; AVX2-NEXT: vmovd %esi, %xmm0 -; AVX2-NEXT: vpbroadcastw %xmm0, %xmm0 -; AVX2-NEXT: vmovdqu (%rdi), %xmm1 -; AVX2-NEXT: vpsubusw %xmm0, %xmm1, %xmm0 -; AVX2-NEXT: vmovdqu %xmm0, (%rdi) +; AVX2-NEXT: vmovd %edi, %xmm1 +; AVX2-NEXT: vpbroadcastw %xmm1, %xmm1 +; AVX2-NEXT: vpsubusw %xmm1, %xmm0, %xmm0 ; AVX2-NEXT: retq vector.ph: %0 = insertelement <8 x i16> undef, i16 %w, i32 0 %broadcast15 = shufflevector <8 x i16> %0, <8 x i16> undef, <8 x i32> zeroinitializer - %1 = getelementptr inbounds i16, i16* %head, i64 0 - %2 = bitcast i16* %1 to <8 x i16>* - %3 = load <8 x i16>, <8 x i16>* %2, align 2 - %4 = icmp ult <8 x i16> %3, %broadcast15 - %5 = sub <8 x i16> %3, %broadcast15 - %6 = select <8 x i1> %4, <8 x i16> zeroinitializer, <8 x i16> %5 - store <8 x i16> %6, <8 x i16>* %2, align 2 - ret void + %1 = icmp ult <8 x i16> %x, %broadcast15 + %2 = sub <8 x i16> %x, %broadcast15 + %res = select <8 x i1> %1, <8 x i16> zeroinitializer, <8 x i16> %2 + ret <8 x i16> %res } -define void @test4(i8* nocapture %head) nounwind { +define <16 x i8> @test4(<16 x i8> %x) nounwind { ; SSE-LABEL: test4: ; SSE: ## BB#0: ## %vector.ph -; SSE-NEXT: movdqu (%rdi), %xmm0 ; SSE-NEXT: psubusb {{.*}}(%rip), %xmm0 -; SSE-NEXT: movdqu %xmm0, (%rdi) ; SSE-NEXT: retq ; ; AVX-LABEL: test4: ; AVX: ## BB#0: ## %vector.ph -; AVX-NEXT: vmovdqu (%rdi), %xmm0 ; AVX-NEXT: vpsubusb {{.*}}(%rip), %xmm0, %xmm0 -; AVX-NEXT: vmovdqu %xmm0, (%rdi) ; AVX-NEXT: retq vector.ph: - %0 = getelementptr inbounds i8, i8* %head, i64 0 - %1 = bitcast i8* %0 to <16 x i8>* - %2 = load <16 x i8>, <16 x i8>* %1, align 1 - %3 = icmp slt <16 x i8> %2, zeroinitializer - %4 = xor <16 x i8> %2, - %5 = select <16 x i1> %3, <16 x i8> %4, <16 x i8> zeroinitializer - store <16 x i8> %5, <16 x i8>* %1, align 1 - ret void + %0 = icmp slt <16 x i8> %x, zeroinitializer + %1 = xor <16 x i8> %x, + %res = select <16 x i1> %0, <16 x i8> %1, <16 x i8> zeroinitializer + ret <16 x i8> %res } -define void @test5(i8* nocapture %head) nounwind { +define <16 x i8> @test5(<16 x i8> %x) nounwind { ; SSE-LABEL: test5: ; SSE: ## BB#0: ## %vector.ph -; SSE-NEXT: movdqu (%rdi), %xmm0 ; SSE-NEXT: psubusb {{.*}}(%rip), %xmm0 -; SSE-NEXT: movdqu %xmm0, (%rdi) ; SSE-NEXT: retq ; ; AVX-LABEL: test5: ; AVX: ## BB#0: ## %vector.ph -; AVX-NEXT: vmovdqu (%rdi), %xmm0 ; AVX-NEXT: vpsubusb {{.*}}(%rip), %xmm0, %xmm0 -; AVX-NEXT: vmovdqu %xmm0, (%rdi) ; AVX-NEXT: retq vector.ph: - %0 = getelementptr inbounds i8, i8* %head, i64 0 - %1 = bitcast i8* %0 to <16 x i8>* - %2 = load <16 x i8>, <16 x i8>* %1, align 1 - %3 = icmp ugt <16 x i8> %2, - %4 = add <16 x i8> %2, - %5 = select <16 x i1> %3, <16 x i8> %4, <16 x i8> zeroinitializer - store <16 x i8> %5, <16 x i8>* %1, align 1 - ret void + %0 = icmp ugt <16 x i8> %x, + %1 = add <16 x i8> %x, + %res = select <16 x i1> %0, <16 x i8> %1, <16 x i8> zeroinitializer + ret <16 x i8> %res } -define void @test6(i8* nocapture %head, i8 zeroext %w) nounwind { +define <16 x i8> @test6(<16 x i8> %x, i8 zeroext %w) nounwind { ; SSE2-LABEL: test6: ; SSE2: ## BB#0: ## %vector.ph -; SSE2-NEXT: movd %esi, %xmm0 -; SSE2-NEXT: punpcklbw {{.*#+}} xmm0 = xmm0[0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7] -; SSE2-NEXT: pshuflw {{.*#+}} xmm0 = xmm0[0,0,0,0,4,5,6,7] -; SSE2-NEXT: pshufd {{.*#+}} xmm0 = xmm0[0,0,1,1] -; SSE2-NEXT: movdqu (%rdi), %xmm1 -; SSE2-NEXT: psubusb %xmm0, %xmm1 -; SSE2-NEXT: movdqu %xmm1, (%rdi) +; SSE2-NEXT: movd %edi, %xmm1 +; SSE2-NEXT: punpcklbw {{.*#+}} xmm1 = xmm1[0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7] +; SSE2-NEXT: pshuflw {{.*#+}} xmm1 = xmm1[0,0,0,0,4,5,6,7] +; SSE2-NEXT: pshufd {{.*#+}} xmm1 = xmm1[0,0,1,1] +; SSE2-NEXT: psubusb %xmm1, %xmm0 ; SSE2-NEXT: retq ; ; SSSE3-LABEL: test6: ; SSSE3: ## BB#0: ## %vector.ph -; SSSE3-NEXT: movd %esi, %xmm0 -; SSSE3-NEXT: pxor %xmm1, %xmm1 -; SSSE3-NEXT: pshufb %xmm1, %xmm0 -; SSSE3-NEXT: movdqu (%rdi), %xmm1 -; SSSE3-NEXT: psubusb %xmm0, %xmm1 -; SSSE3-NEXT: movdqu %xmm1, (%rdi) +; SSSE3-NEXT: movd %edi, %xmm1 +; SSSE3-NEXT: pxor %xmm2, %xmm2 +; SSSE3-NEXT: pshufb %xmm2, %xmm1 +; SSSE3-NEXT: psubusb %xmm1, %xmm0 ; SSSE3-NEXT: retq ; +; SSE41-LABEL: test6: +; SSE41: ## BB#0: ## %vector.ph +; SSE41-NEXT: movd %edi, %xmm1 +; SSE41-NEXT: pxor %xmm2, %xmm2 +; SSE41-NEXT: pshufb %xmm2, %xmm1 +; SSE41-NEXT: psubusb %xmm1, %xmm0 +; SSE41-NEXT: retq +; ; AVX1-LABEL: test6: ; AVX1: ## BB#0: ## %vector.ph -; AVX1-NEXT: vmovd %esi, %xmm0 -; AVX1-NEXT: vpxor %xmm1, %xmm1, %xmm1 -; AVX1-NEXT: vpshufb %xmm1, %xmm0, %xmm0 -; AVX1-NEXT: vmovdqu (%rdi), %xmm1 -; AVX1-NEXT: vpsubusb %xmm0, %xmm1, %xmm0 -; AVX1-NEXT: vmovdqu %xmm0, (%rdi) +; AVX1-NEXT: vmovd %edi, %xmm1 +; AVX1-NEXT: vpxor %xmm2, %xmm2, %xmm2 +; AVX1-NEXT: vpshufb %xmm2, %xmm1, %xmm1 +; AVX1-NEXT: vpsubusb %xmm1, %xmm0, %xmm0 ; AVX1-NEXT: retq ; ; AVX2-LABEL: test6: ; AVX2: ## BB#0: ## %vector.ph -; AVX2-NEXT: vmovd %esi, %xmm0 -; AVX2-NEXT: vpbroadcastb %xmm0, %xmm0 -; AVX2-NEXT: vmovdqu (%rdi), %xmm1 -; AVX2-NEXT: vpsubusb %xmm0, %xmm1, %xmm0 -; AVX2-NEXT: vmovdqu %xmm0, (%rdi) +; AVX2-NEXT: vmovd %edi, %xmm1 +; AVX2-NEXT: vpbroadcastb %xmm1, %xmm1 +; AVX2-NEXT: vpsubusb %xmm1, %xmm0, %xmm0 ; AVX2-NEXT: retq vector.ph: %0 = insertelement <16 x i8> undef, i8 %w, i32 0 %broadcast15 = shufflevector <16 x i8> %0, <16 x i8> undef, <16 x i32> zeroinitializer - %1 = getelementptr inbounds i8, i8* %head, i64 0 - %2 = bitcast i8* %1 to <16 x i8>* - %3 = load <16 x i8>, <16 x i8>* %2, align 1 - %4 = icmp ult <16 x i8> %3, %broadcast15 - %5 = sub <16 x i8> %3, %broadcast15 - %6 = select <16 x i1> %4, <16 x i8> zeroinitializer, <16 x i8> %5 - store <16 x i8> %6, <16 x i8>* %2, align 1 - ret void + %1 = icmp ult <16 x i8> %x, %broadcast15 + %2 = sub <16 x i8> %x, %broadcast15 + %res = select <16 x i1> %1, <16 x i8> zeroinitializer, <16 x i8> %2 + ret <16 x i8> %res } -define void @test7(i16* nocapture %head) nounwind { +define <16 x i16> @test7(<16 x i16> %x) nounwind { ; SSE-LABEL: test7: ; SSE: ## BB#0: ## %vector.ph -; SSE-NEXT: movdqu (%rdi), %xmm0 -; SSE-NEXT: movdqu 16(%rdi), %xmm1 ; SSE-NEXT: movdqa {{.*#+}} xmm2 = [32768,32768,32768,32768,32768,32768,32768,32768] ; SSE-NEXT: psubusw %xmm2, %xmm0 ; SSE-NEXT: psubusw %xmm2, %xmm1 -; SSE-NEXT: movdqu %xmm1, 16(%rdi) -; SSE-NEXT: movdqu %xmm0, (%rdi) ; SSE-NEXT: retq ; ; AVX1-LABEL: test7: ; AVX1: ## BB#0: ## %vector.ph -; AVX1-NEXT: vmovdqu (%rdi), %ymm0 ; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm1 ; AVX1-NEXT: vpxor %xmm2, %xmm2, %xmm2 ; AVX1-NEXT: vpcmpgtw %xmm1, %xmm2, %xmm1 @@ -221,43 +171,29 @@ define void @test7(i16* nocapture %head) nounwind { ; AVX1-NEXT: vinsertf128 $1, %xmm1, %ymm2, %ymm1 ; AVX1-NEXT: vxorps {{.*}}(%rip), %ymm0, %ymm0 ; AVX1-NEXT: vandps %ymm0, %ymm1, %ymm0 -; AVX1-NEXT: vmovups %ymm0, (%rdi) -; AVX1-NEXT: vzeroupper ; AVX1-NEXT: retq ; ; AVX2-LABEL: test7: ; AVX2: ## BB#0: ## %vector.ph -; AVX2-NEXT: vmovdqu (%rdi), %ymm0 ; AVX2-NEXT: vpsubusw {{.*}}(%rip), %ymm0, %ymm0 -; AVX2-NEXT: vmovdqu %ymm0, (%rdi) -; AVX2-NEXT: vzeroupper ; AVX2-NEXT: retq vector.ph: - %0 = getelementptr inbounds i16, i16* %head, i64 0 - %1 = bitcast i16* %0 to <16 x i16>* - %2 = load <16 x i16>, <16 x i16>* %1, align 2 - %3 = icmp slt <16 x i16> %2, zeroinitializer - %4 = xor <16 x i16> %2, - %5 = select <16 x i1> %3, <16 x i16> %4, <16 x i16> zeroinitializer - store <16 x i16> %5, <16 x i16>* %1, align 2 - ret void + %0 = icmp slt <16 x i16> %x, zeroinitializer + %1 = xor <16 x i16> %x, + %res = select <16 x i1> %0, <16 x i16> %1, <16 x i16> zeroinitializer + ret <16 x i16> %res } -define void @test8(i16* nocapture %head) nounwind { +define <16 x i16> @test8(<16 x i16> %x) nounwind { ; SSE-LABEL: test8: ; SSE: ## BB#0: ## %vector.ph -; SSE-NEXT: movdqu (%rdi), %xmm0 -; SSE-NEXT: movdqu 16(%rdi), %xmm1 ; SSE-NEXT: movdqa {{.*#+}} xmm2 = [32767,32767,32767,32767,32767,32767,32767,32767] ; SSE-NEXT: psubusw %xmm2, %xmm0 ; SSE-NEXT: psubusw %xmm2, %xmm1 -; SSE-NEXT: movdqu %xmm1, 16(%rdi) -; SSE-NEXT: movdqu %xmm0, (%rdi) ; SSE-NEXT: retq ; ; AVX1-LABEL: test8: ; AVX1: ## BB#0: ## %vector.ph -; AVX1-NEXT: vmovdqu (%rdi), %ymm0 ; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm1 ; AVX1-NEXT: vmovdqa {{.*#+}} xmm2 = [32768,32768,32768,32768,32768,32768,32768,32768] ; AVX1-NEXT: vpxor %xmm2, %xmm1, %xmm3 @@ -271,48 +207,33 @@ define void @test8(i16* nocapture %head) nounwind { ; AVX1-NEXT: vpaddw %xmm3, %xmm0, %xmm0 ; AVX1-NEXT: vinsertf128 $1, %xmm1, %ymm0, %ymm0 ; AVX1-NEXT: vandps %ymm0, %ymm2, %ymm0 -; AVX1-NEXT: vmovups %ymm0, (%rdi) -; AVX1-NEXT: vzeroupper ; AVX1-NEXT: retq ; ; AVX2-LABEL: test8: ; AVX2: ## BB#0: ## %vector.ph -; AVX2-NEXT: vmovdqu (%rdi), %ymm0 ; AVX2-NEXT: vpsubusw {{.*}}(%rip), %ymm0, %ymm0 -; AVX2-NEXT: vmovdqu %ymm0, (%rdi) -; AVX2-NEXT: vzeroupper ; AVX2-NEXT: retq vector.ph: - %0 = getelementptr inbounds i16, i16* %head, i64 0 - %1 = bitcast i16* %0 to <16 x i16>* - %2 = load <16 x i16>, <16 x i16>* %1, align 2 - %3 = icmp ugt <16 x i16> %2, - %4 = add <16 x i16> %2, - %5 = select <16 x i1> %3, <16 x i16> %4, <16 x i16> zeroinitializer - store <16 x i16> %5, <16 x i16>* %1, align 2 - ret void - + %0 = icmp ugt <16 x i16> %x, + %1 = add <16 x i16> %x, + %res = select <16 x i1> %0, <16 x i16> %1, <16 x i16> zeroinitializer + ret <16 x i16> %res } -define void @test9(i16* nocapture %head, i16 zeroext %w) nounwind { +define <16 x i16> @test9(<16 x i16> %x, i16 zeroext %w) nounwind { ; SSE-LABEL: test9: ; SSE: ## BB#0: ## %vector.ph -; SSE-NEXT: movd %esi, %xmm0 -; SSE-NEXT: pshuflw {{.*#+}} xmm0 = xmm0[0,0,0,0,4,5,6,7] -; SSE-NEXT: pshufd {{.*#+}} xmm0 = xmm0[0,0,1,1] -; SSE-NEXT: movdqu (%rdi), %xmm1 -; SSE-NEXT: movdqu 16(%rdi), %xmm2 -; SSE-NEXT: psubusw %xmm0, %xmm1 -; SSE-NEXT: psubusw %xmm0, %xmm2 -; SSE-NEXT: movdqu %xmm2, 16(%rdi) -; SSE-NEXT: movdqu %xmm1, (%rdi) +; SSE-NEXT: movd %edi, %xmm2 +; SSE-NEXT: pshuflw {{.*#+}} xmm2 = xmm2[0,0,0,0,4,5,6,7] +; SSE-NEXT: pshufd {{.*#+}} xmm2 = xmm2[0,0,1,1] +; SSE-NEXT: psubusw %xmm2, %xmm0 +; SSE-NEXT: psubusw %xmm2, %xmm1 ; SSE-NEXT: retq ; ; AVX1-LABEL: test9: ; AVX1: ## BB#0: ## %vector.ph -; AVX1-NEXT: vmovdqu (%rdi), %ymm0 ; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm1 -; AVX1-NEXT: vmovd %esi, %xmm2 +; AVX1-NEXT: vmovd %edi, %xmm2 ; AVX1-NEXT: vpshuflw {{.*#+}} xmm2 = xmm2[0,0,0,0,4,5,6,7] ; AVX1-NEXT: vpshufd {{.*#+}} xmm2 = xmm2[0,0,1,1] ; AVX1-NEXT: vpsubw %xmm2, %xmm1, %xmm3 @@ -324,47 +245,33 @@ define void @test9(i16* nocapture %head, i16 zeroext %w) nounwind { ; AVX1-NEXT: vpcmpeqw %xmm2, %xmm0, %xmm0 ; AVX1-NEXT: vinsertf128 $1, %xmm1, %ymm0, %ymm0 ; AVX1-NEXT: vandps %ymm3, %ymm0, %ymm0 -; AVX1-NEXT: vmovups %ymm0, (%rdi) -; AVX1-NEXT: vzeroupper ; AVX1-NEXT: retq ; ; AVX2-LABEL: test9: ; AVX2: ## BB#0: ## %vector.ph -; AVX2-NEXT: vmovd %esi, %xmm0 -; AVX2-NEXT: vpbroadcastw %xmm0, %ymm0 -; AVX2-NEXT: vmovdqu (%rdi), %ymm1 -; AVX2-NEXT: vpsubusw %ymm0, %ymm1, %ymm0 -; AVX2-NEXT: vmovdqu %ymm0, (%rdi) -; AVX2-NEXT: vzeroupper +; AVX2-NEXT: vmovd %edi, %xmm1 +; AVX2-NEXT: vpbroadcastw %xmm1, %ymm1 +; AVX2-NEXT: vpsubusw %ymm1, %ymm0, %ymm0 ; AVX2-NEXT: retq vector.ph: %0 = insertelement <16 x i16> undef, i16 %w, i32 0 %broadcast15 = shufflevector <16 x i16> %0, <16 x i16> undef, <16 x i32> zeroinitializer - %1 = getelementptr inbounds i16, i16* %head, i64 0 - %2 = bitcast i16* %1 to <16 x i16>* - %3 = load <16 x i16>, <16 x i16>* %2, align 2 - %4 = icmp ult <16 x i16> %3, %broadcast15 - %5 = sub <16 x i16> %3, %broadcast15 - %6 = select <16 x i1> %4, <16 x i16> zeroinitializer, <16 x i16> %5 - store <16 x i16> %6, <16 x i16>* %2, align 2 - ret void + %1 = icmp ult <16 x i16> %x, %broadcast15 + %2 = sub <16 x i16> %x, %broadcast15 + %res = select <16 x i1> %1, <16 x i16> zeroinitializer, <16 x i16> %2 + ret <16 x i16> %res } -define void @test10(i8* nocapture %head) nounwind { +define <32 x i8> @test10(<32 x i8> %x) nounwind { ; SSE-LABEL: test10: ; SSE: ## BB#0: ## %vector.ph -; SSE-NEXT: movdqu (%rdi), %xmm0 -; SSE-NEXT: movdqu 16(%rdi), %xmm1 ; SSE-NEXT: movdqa {{.*#+}} xmm2 = [128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128] ; SSE-NEXT: psubusb %xmm2, %xmm0 ; SSE-NEXT: psubusb %xmm2, %xmm1 -; SSE-NEXT: movdqu %xmm1, 16(%rdi) -; SSE-NEXT: movdqu %xmm0, (%rdi) ; SSE-NEXT: retq ; ; AVX1-LABEL: test10: ; AVX1: ## BB#0: ## %vector.ph -; AVX1-NEXT: vmovdqu (%rdi), %ymm0 ; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm1 ; AVX1-NEXT: vpxor %xmm2, %xmm2, %xmm2 ; AVX1-NEXT: vpcmpgtb %xmm1, %xmm2, %xmm1 @@ -372,44 +279,29 @@ define void @test10(i8* nocapture %head) nounwind { ; AVX1-NEXT: vinsertf128 $1, %xmm1, %ymm2, %ymm1 ; AVX1-NEXT: vxorps {{.*}}(%rip), %ymm0, %ymm0 ; AVX1-NEXT: vandps %ymm0, %ymm1, %ymm0 -; AVX1-NEXT: vmovups %ymm0, (%rdi) -; AVX1-NEXT: vzeroupper ; AVX1-NEXT: retq ; ; AVX2-LABEL: test10: ; AVX2: ## BB#0: ## %vector.ph -; AVX2-NEXT: vmovdqu (%rdi), %ymm0 ; AVX2-NEXT: vpsubusb {{.*}}(%rip), %ymm0, %ymm0 -; AVX2-NEXT: vmovdqu %ymm0, (%rdi) -; AVX2-NEXT: vzeroupper ; AVX2-NEXT: retq vector.ph: - %0 = getelementptr inbounds i8, i8* %head, i64 0 - %1 = bitcast i8* %0 to <32 x i8>* - %2 = load <32 x i8>, <32 x i8>* %1, align 1 - %3 = icmp slt <32 x i8> %2, zeroinitializer - %4 = xor <32 x i8> %2, - %5 = select <32 x i1> %3, <32 x i8> %4, <32 x i8> zeroinitializer - store <32 x i8> %5, <32 x i8>* %1, align 1 - ret void - + %0 = icmp slt <32 x i8> %x, zeroinitializer + %1 = xor <32 x i8> %x, + %res = select <32 x i1> %0, <32 x i8> %1, <32 x i8> zeroinitializer + ret <32 x i8> %res } -define void @test11(i8* nocapture %head) nounwind { +define <32 x i8> @test11(<32 x i8> %x) nounwind { ; SSE-LABEL: test11: ; SSE: ## BB#0: ## %vector.ph -; SSE-NEXT: movdqu (%rdi), %xmm0 -; SSE-NEXT: movdqu 16(%rdi), %xmm1 ; SSE-NEXT: movdqa {{.*#+}} xmm2 = [127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127] ; SSE-NEXT: psubusb %xmm2, %xmm0 ; SSE-NEXT: psubusb %xmm2, %xmm1 -; SSE-NEXT: movdqu %xmm1, 16(%rdi) -; SSE-NEXT: movdqu %xmm0, (%rdi) ; SSE-NEXT: retq ; ; AVX1-LABEL: test11: ; AVX1: ## BB#0: ## %vector.ph -; AVX1-NEXT: vmovdqu (%rdi), %ymm0 ; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm1 ; AVX1-NEXT: vmovdqa {{.*#+}} xmm2 = [128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128] ; AVX1-NEXT: vpxor %xmm2, %xmm1, %xmm3 @@ -423,60 +315,51 @@ define void @test11(i8* nocapture %head) nounwind { ; AVX1-NEXT: vpaddb %xmm3, %xmm0, %xmm0 ; AVX1-NEXT: vinsertf128 $1, %xmm1, %ymm0, %ymm0 ; AVX1-NEXT: vandps %ymm0, %ymm2, %ymm0 -; AVX1-NEXT: vmovups %ymm0, (%rdi) -; AVX1-NEXT: vzeroupper ; AVX1-NEXT: retq ; ; AVX2-LABEL: test11: ; AVX2: ## BB#0: ## %vector.ph -; AVX2-NEXT: vmovdqu (%rdi), %ymm0 ; AVX2-NEXT: vpsubusb {{.*}}(%rip), %ymm0, %ymm0 -; AVX2-NEXT: vmovdqu %ymm0, (%rdi) -; AVX2-NEXT: vzeroupper ; AVX2-NEXT: retq vector.ph: - %0 = getelementptr inbounds i8, i8* %head, i64 0 - %1 = bitcast i8* %0 to <32 x i8>* - %2 = load <32 x i8>, <32 x i8>* %1, align 1 - %3 = icmp ugt <32 x i8> %2, - %4 = add <32 x i8> %2, - %5 = select <32 x i1> %3, <32 x i8> %4, <32 x i8> zeroinitializer - store <32 x i8> %5, <32 x i8>* %1, align 1 - ret void + %0 = icmp ugt <32 x i8> %x, + %1 = add <32 x i8> %x, + %res = select <32 x i1> %0, <32 x i8> %1, <32 x i8> zeroinitializer + ret <32 x i8> %res } -define void @test12(i8* nocapture %head, i8 zeroext %w) nounwind { +define <32 x i8> @test12(<32 x i8> %x, i8 zeroext %w) nounwind { ; SSE2-LABEL: test12: ; SSE2: ## BB#0: ## %vector.ph -; SSE2-NEXT: movd %esi, %xmm0 -; SSE2-NEXT: punpcklbw {{.*#+}} xmm0 = xmm0[0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7] -; SSE2-NEXT: pshuflw {{.*#+}} xmm0 = xmm0[0,0,0,0,4,5,6,7] -; SSE2-NEXT: pshufd {{.*#+}} xmm0 = xmm0[0,0,1,1] -; SSE2-NEXT: movdqu (%rdi), %xmm1 -; SSE2-NEXT: movdqu 16(%rdi), %xmm2 -; SSE2-NEXT: psubusb %xmm0, %xmm1 -; SSE2-NEXT: psubusb %xmm0, %xmm2 -; SSE2-NEXT: movdqu %xmm2, 16(%rdi) -; SSE2-NEXT: movdqu %xmm1, (%rdi) +; SSE2-NEXT: movd %edi, %xmm2 +; SSE2-NEXT: punpcklbw {{.*#+}} xmm2 = xmm2[0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7] +; SSE2-NEXT: pshuflw {{.*#+}} xmm2 = xmm2[0,0,0,0,4,5,6,7] +; SSE2-NEXT: pshufd {{.*#+}} xmm2 = xmm2[0,0,1,1] +; SSE2-NEXT: psubusb %xmm2, %xmm0 +; SSE2-NEXT: psubusb %xmm2, %xmm1 ; SSE2-NEXT: retq ; ; SSSE3-LABEL: test12: ; SSSE3: ## BB#0: ## %vector.ph -; SSSE3-NEXT: movd %esi, %xmm0 -; SSSE3-NEXT: pxor %xmm1, %xmm1 -; SSSE3-NEXT: pshufb %xmm1, %xmm0 -; SSSE3-NEXT: movdqu (%rdi), %xmm1 -; SSSE3-NEXT: movdqu 16(%rdi), %xmm2 -; SSSE3-NEXT: psubusb %xmm0, %xmm1 -; SSSE3-NEXT: psubusb %xmm0, %xmm2 -; SSSE3-NEXT: movdqu %xmm2, 16(%rdi) -; SSSE3-NEXT: movdqu %xmm1, (%rdi) +; SSSE3-NEXT: movd %edi, %xmm2 +; SSSE3-NEXT: pxor %xmm3, %xmm3 +; SSSE3-NEXT: pshufb %xmm3, %xmm2 +; SSSE3-NEXT: psubusb %xmm2, %xmm0 +; SSSE3-NEXT: psubusb %xmm2, %xmm1 ; SSSE3-NEXT: retq ; +; SSE41-LABEL: test12: +; SSE41: ## BB#0: ## %vector.ph +; SSE41-NEXT: movd %edi, %xmm2 +; SSE41-NEXT: pxor %xmm3, %xmm3 +; SSE41-NEXT: pshufb %xmm3, %xmm2 +; SSE41-NEXT: psubusb %xmm2, %xmm0 +; SSE41-NEXT: psubusb %xmm2, %xmm1 +; SSE41-NEXT: retq +; ; AVX1-LABEL: test12: ; AVX1: ## BB#0: ## %vector.ph -; AVX1-NEXT: vmovdqu (%rdi), %ymm0 -; AVX1-NEXT: vmovd %esi, %xmm1 +; AVX1-NEXT: vmovd %edi, %xmm1 ; AVX1-NEXT: vpxor %xmm2, %xmm2, %xmm2 ; AVX1-NEXT: vpshufb %xmm2, %xmm1, %xmm1 ; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm2 @@ -489,617 +372,675 @@ define void @test12(i8* nocapture %head, i8 zeroext %w) nounwind { ; AVX1-NEXT: vpcmpeqb %xmm1, %xmm0, %xmm0 ; AVX1-NEXT: vinsertf128 $1, %xmm2, %ymm0, %ymm0 ; AVX1-NEXT: vandps %ymm3, %ymm0, %ymm0 -; AVX1-NEXT: vmovups %ymm0, (%rdi) -; AVX1-NEXT: vzeroupper ; AVX1-NEXT: retq ; ; AVX2-LABEL: test12: ; AVX2: ## BB#0: ## %vector.ph -; AVX2-NEXT: vmovd %esi, %xmm0 -; AVX2-NEXT: vpbroadcastb %xmm0, %ymm0 -; AVX2-NEXT: vmovdqu (%rdi), %ymm1 -; AVX2-NEXT: vpsubusb %ymm0, %ymm1, %ymm0 -; AVX2-NEXT: vmovdqu %ymm0, (%rdi) -; AVX2-NEXT: vzeroupper +; AVX2-NEXT: vmovd %edi, %xmm1 +; AVX2-NEXT: vpbroadcastb %xmm1, %ymm1 +; AVX2-NEXT: vpsubusb %ymm1, %ymm0, %ymm0 ; AVX2-NEXT: retq vector.ph: %0 = insertelement <32 x i8> undef, i8 %w, i32 0 %broadcast15 = shufflevector <32 x i8> %0, <32 x i8> undef, <32 x i32> zeroinitializer - %1 = getelementptr inbounds i8, i8* %head, i64 0 - %2 = bitcast i8* %1 to <32 x i8>* - %3 = load <32 x i8>, <32 x i8>* %2, align 1 - %4 = icmp ult <32 x i8> %3, %broadcast15 - %5 = sub <32 x i8> %3, %broadcast15 - %6 = select <32 x i1> %4, <32 x i8> zeroinitializer, <32 x i8> %5 - store <32 x i8> %6, <32 x i8>* %2, align 1 - ret void + %1 = icmp ult <32 x i8> %x, %broadcast15 + %2 = sub <32 x i8> %x, %broadcast15 + %res = select <32 x i1> %1, <32 x i8> zeroinitializer, <32 x i8> %2 + ret <32 x i8> %res } -define void @test13(i16* nocapture %head, i32* nocapture %w) nounwind { +define <8 x i16> @test13(<8 x i16> %x, <8 x i32> %y) nounwind { ; SSE2-LABEL: test13: ; SSE2: ## BB#0: ## %vector.ph -; SSE2-NEXT: movdqu (%rdi), %xmm0 -; SSE2-NEXT: movdqu (%rsi), %xmm2 -; SSE2-NEXT: movdqu 16(%rsi), %xmm3 ; SSE2-NEXT: pxor %xmm4, %xmm4 -; SSE2-NEXT: movdqa %xmm0, %xmm1 -; SSE2-NEXT: punpcklwd {{.*#+}} xmm1 = xmm1[0],xmm4[0],xmm1[1],xmm4[1],xmm1[2],xmm4[2],xmm1[3],xmm4[3] +; SSE2-NEXT: movdqa %xmm0, %xmm3 +; SSE2-NEXT: punpcklwd {{.*#+}} xmm3 = xmm3[0],xmm4[0],xmm3[1],xmm4[1],xmm3[2],xmm4[2],xmm3[3],xmm4[3] ; SSE2-NEXT: punpckhwd {{.*#+}} xmm0 = xmm0[4],xmm4[4],xmm0[5],xmm4[5],xmm0[6],xmm4[6],xmm0[7],xmm4[7] ; SSE2-NEXT: movdqa {{.*#+}} xmm4 = [2147483648,2147483648,2147483648,2147483648] ; SSE2-NEXT: movdqa %xmm0, %xmm5 -; SSE2-NEXT: psubd %xmm3, %xmm0 -; SSE2-NEXT: pxor %xmm4, %xmm3 +; SSE2-NEXT: psubd %xmm2, %xmm0 +; SSE2-NEXT: pxor %xmm4, %xmm2 ; SSE2-NEXT: pxor %xmm4, %xmm5 -; SSE2-NEXT: pcmpgtd %xmm5, %xmm3 -; SSE2-NEXT: pshuflw {{.*#+}} xmm3 = xmm3[0,2,2,3,4,5,6,7] -; SSE2-NEXT: pshufhw {{.*#+}} xmm3 = xmm3[0,1,2,3,4,6,6,7] -; SSE2-NEXT: pshufd {{.*#+}} xmm3 = xmm3[0,2,2,3] -; SSE2-NEXT: movdqa %xmm2, %xmm5 -; SSE2-NEXT: pxor %xmm4, %xmm5 -; SSE2-NEXT: pxor %xmm1, %xmm4 -; SSE2-NEXT: pcmpgtd %xmm4, %xmm5 -; SSE2-NEXT: pshuflw {{.*#+}} xmm4 = xmm5[0,2,2,3,4,5,6,7] -; SSE2-NEXT: pshufhw {{.*#+}} xmm4 = xmm4[0,1,2,3,4,6,6,7] -; SSE2-NEXT: pshufd {{.*#+}} xmm4 = xmm4[0,2,2,3] -; SSE2-NEXT: punpcklqdq {{.*#+}} xmm4 = xmm4[0],xmm3[0] -; SSE2-NEXT: psubd %xmm2, %xmm1 +; SSE2-NEXT: pcmpgtd %xmm5, %xmm2 +; SSE2-NEXT: pshuflw {{.*#+}} xmm2 = xmm2[0,2,2,3,4,5,6,7] +; SSE2-NEXT: pshufhw {{.*#+}} xmm2 = xmm2[0,1,2,3,4,6,6,7] +; SSE2-NEXT: pshufd {{.*#+}} xmm5 = xmm2[0,2,2,3] +; SSE2-NEXT: movdqa %xmm1, %xmm2 +; SSE2-NEXT: pxor %xmm4, %xmm2 +; SSE2-NEXT: pxor %xmm3, %xmm4 +; SSE2-NEXT: pcmpgtd %xmm4, %xmm2 +; SSE2-NEXT: pshuflw {{.*#+}} xmm2 = xmm2[0,2,2,3,4,5,6,7] +; SSE2-NEXT: pshufhw {{.*#+}} xmm2 = xmm2[0,1,2,3,4,6,6,7] +; SSE2-NEXT: pshufd {{.*#+}} xmm2 = xmm2[0,2,2,3] +; SSE2-NEXT: punpcklqdq {{.*#+}} xmm2 = xmm2[0],xmm5[0] +; SSE2-NEXT: psubd %xmm1, %xmm3 ; SSE2-NEXT: pslld $16, %xmm0 ; SSE2-NEXT: psrad $16, %xmm0 -; SSE2-NEXT: pslld $16, %xmm1 -; SSE2-NEXT: psrad $16, %xmm1 -; SSE2-NEXT: packssdw %xmm0, %xmm1 -; SSE2-NEXT: pandn %xmm1, %xmm4 -; SSE2-NEXT: movdqu %xmm4, (%rdi) +; SSE2-NEXT: pslld $16, %xmm3 +; SSE2-NEXT: psrad $16, %xmm3 +; SSE2-NEXT: packssdw %xmm0, %xmm3 +; SSE2-NEXT: pandn %xmm3, %xmm2 +; SSE2-NEXT: movdqa %xmm2, %xmm0 ; SSE2-NEXT: retq ; ; SSSE3-LABEL: test13: ; SSSE3: ## BB#0: ## %vector.ph -; SSSE3-NEXT: movdqu (%rdi), %xmm0 -; SSSE3-NEXT: movdqu (%rsi), %xmm2 -; SSSE3-NEXT: movdqu 16(%rsi), %xmm3 ; SSSE3-NEXT: pxor %xmm4, %xmm4 -; SSSE3-NEXT: movdqa %xmm0, %xmm1 -; SSSE3-NEXT: punpcklwd {{.*#+}} xmm1 = xmm1[0],xmm4[0],xmm1[1],xmm4[1],xmm1[2],xmm4[2],xmm1[3],xmm4[3] +; SSSE3-NEXT: movdqa %xmm0, %xmm3 +; SSSE3-NEXT: punpcklwd {{.*#+}} xmm3 = xmm3[0],xmm4[0],xmm3[1],xmm4[1],xmm3[2],xmm4[2],xmm3[3],xmm4[3] ; SSSE3-NEXT: punpckhwd {{.*#+}} xmm0 = xmm0[4],xmm4[4],xmm0[5],xmm4[5],xmm0[6],xmm4[6],xmm0[7],xmm4[7] ; SSSE3-NEXT: movdqa {{.*#+}} xmm4 = [2147483648,2147483648,2147483648,2147483648] ; SSSE3-NEXT: movdqa %xmm0, %xmm5 -; SSSE3-NEXT: psubd %xmm3, %xmm0 -; SSSE3-NEXT: pxor %xmm4, %xmm3 -; SSSE3-NEXT: pxor %xmm4, %xmm5 -; SSSE3-NEXT: pcmpgtd %xmm5, %xmm3 -; SSSE3-NEXT: movdqa {{.*#+}} xmm5 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15] -; SSSE3-NEXT: pshufb %xmm5, %xmm3 +; SSSE3-NEXT: psubd %xmm2, %xmm0 ; SSSE3-NEXT: movdqa %xmm2, %xmm6 ; SSSE3-NEXT: pxor %xmm4, %xmm6 -; SSSE3-NEXT: pxor %xmm1, %xmm4 -; SSSE3-NEXT: pcmpgtd %xmm4, %xmm6 +; SSSE3-NEXT: pxor %xmm4, %xmm5 +; SSSE3-NEXT: pcmpgtd %xmm5, %xmm6 +; SSSE3-NEXT: movdqa {{.*#+}} xmm5 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15] ; SSSE3-NEXT: pshufb %xmm5, %xmm6 -; SSSE3-NEXT: punpcklqdq {{.*#+}} xmm6 = xmm6[0],xmm3[0] -; SSSE3-NEXT: psubd %xmm2, %xmm1 +; SSSE3-NEXT: movdqa %xmm1, %xmm2 +; SSSE3-NEXT: pxor %xmm4, %xmm2 +; SSSE3-NEXT: pxor %xmm3, %xmm4 +; SSSE3-NEXT: pcmpgtd %xmm4, %xmm2 +; SSSE3-NEXT: pshufb %xmm5, %xmm2 +; SSSE3-NEXT: punpcklqdq {{.*#+}} xmm2 = xmm2[0],xmm6[0] +; SSSE3-NEXT: psubd %xmm1, %xmm3 ; SSSE3-NEXT: pshufb %xmm5, %xmm0 -; SSSE3-NEXT: pshufb %xmm5, %xmm1 -; SSSE3-NEXT: punpcklqdq {{.*#+}} xmm1 = xmm1[0],xmm0[0] -; SSSE3-NEXT: pandn %xmm1, %xmm6 -; SSSE3-NEXT: movdqu %xmm6, (%rdi) +; SSSE3-NEXT: pshufb %xmm5, %xmm3 +; SSSE3-NEXT: punpcklqdq {{.*#+}} xmm3 = xmm3[0],xmm0[0] +; SSSE3-NEXT: pandn %xmm3, %xmm2 +; SSSE3-NEXT: movdqa %xmm2, %xmm0 ; SSSE3-NEXT: retq ; +; SSE41-LABEL: test13: +; SSE41: ## BB#0: ## %vector.ph +; SSE41-NEXT: pshufd {{.*#+}} xmm3 = xmm0[2,3,0,1] +; SSE41-NEXT: pmovzxwd {{.*#+}} xmm4 = xmm3[0],zero,xmm3[1],zero,xmm3[2],zero,xmm3[3],zero +; SSE41-NEXT: pmovzxwd {{.*#+}} xmm3 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero +; SSE41-NEXT: movdqa {{.*#+}} xmm5 = [2147483648,2147483648,2147483648,2147483648] +; SSE41-NEXT: movdqa %xmm3, %xmm6 +; SSE41-NEXT: psubd %xmm1, %xmm3 +; SSE41-NEXT: movdqa %xmm1, %xmm0 +; SSE41-NEXT: pxor %xmm5, %xmm0 +; SSE41-NEXT: pxor %xmm5, %xmm6 +; SSE41-NEXT: pcmpgtd %xmm6, %xmm0 +; SSE41-NEXT: movdqa {{.*#+}} xmm1 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15] +; SSE41-NEXT: pshufb %xmm1, %xmm0 +; SSE41-NEXT: movdqa %xmm2, %xmm6 +; SSE41-NEXT: pxor %xmm5, %xmm6 +; SSE41-NEXT: pxor %xmm4, %xmm5 +; SSE41-NEXT: pcmpgtd %xmm5, %xmm6 +; SSE41-NEXT: pshufb %xmm1, %xmm6 +; SSE41-NEXT: punpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm6[0] +; SSE41-NEXT: psubd %xmm2, %xmm4 +; SSE41-NEXT: pshufb %xmm1, %xmm3 +; SSE41-NEXT: pshufb %xmm1, %xmm4 +; SSE41-NEXT: punpcklqdq {{.*#+}} xmm3 = xmm3[0],xmm4[0] +; SSE41-NEXT: pandn %xmm3, %xmm0 +; SSE41-NEXT: retq +; ; AVX1-LABEL: test13: ; AVX1: ## BB#0: ## %vector.ph -; AVX1-NEXT: vmovdqu (%rsi), %ymm0 -; AVX1-NEXT: vpmovzxwd {{.*#+}} xmm1 = mem[0],zero,mem[1],zero,mem[2],zero,mem[3],zero -; AVX1-NEXT: vpmovzxwd {{.*#+}} xmm2 = mem[0],zero,mem[1],zero,mem[2],zero,mem[3],zero +; AVX1-NEXT: vpshufd {{.*#+}} xmm2 = xmm0[2,3,0,1] +; AVX1-NEXT: vpmovzxwd {{.*#+}} xmm2 = xmm2[0],zero,xmm2[1],zero,xmm2[2],zero,xmm2[3],zero +; AVX1-NEXT: vpmovzxwd {{.*#+}} xmm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero ; AVX1-NEXT: vmovdqa {{.*#+}} xmm3 = [2147483648,2147483648,2147483648,2147483648] -; AVX1-NEXT: vpxor %xmm3, %xmm2, %xmm4 -; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm5 +; AVX1-NEXT: vpxor %xmm3, %xmm0, %xmm4 +; AVX1-NEXT: vpxor %xmm3, %xmm1, %xmm5 +; AVX1-NEXT: vpcmpgtd %xmm4, %xmm5, %xmm4 +; AVX1-NEXT: vextractf128 $1, %ymm1, %xmm5 ; AVX1-NEXT: vpxor %xmm3, %xmm5, %xmm6 -; AVX1-NEXT: vpcmpgtd %xmm4, %xmm6, %xmm4 -; AVX1-NEXT: vpxor %xmm3, %xmm1, %xmm6 -; AVX1-NEXT: vpxor %xmm3, %xmm0, %xmm3 -; AVX1-NEXT: vpcmpgtd %xmm6, %xmm3, %xmm3 -; AVX1-NEXT: vpacksswb %xmm4, %xmm3, %xmm3 -; AVX1-NEXT: vpsubd %xmm0, %xmm1, %xmm0 -; AVX1-NEXT: vpsubd %xmm5, %xmm2, %xmm1 -; AVX1-NEXT: vmovdqa {{.*#+}} xmm2 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15] -; AVX1-NEXT: vpshufb %xmm2, %xmm1, %xmm1 -; AVX1-NEXT: vpshufb %xmm2, %xmm0, %xmm0 +; AVX1-NEXT: vpxor %xmm3, %xmm2, %xmm3 +; AVX1-NEXT: vpcmpgtd %xmm3, %xmm6, %xmm3 +; AVX1-NEXT: vpacksswb %xmm3, %xmm4, %xmm3 +; AVX1-NEXT: vpsubd %xmm5, %xmm2, %xmm2 +; AVX1-NEXT: vpsubd %xmm1, %xmm0, %xmm0 +; AVX1-NEXT: vmovdqa {{.*#+}} xmm1 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15] +; AVX1-NEXT: vpshufb %xmm1, %xmm0, %xmm0 +; AVX1-NEXT: vpshufb %xmm1, %xmm2, %xmm1 ; AVX1-NEXT: vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0] ; AVX1-NEXT: vpandn %xmm0, %xmm3, %xmm0 -; AVX1-NEXT: vmovdqu %xmm0, (%rdi) ; AVX1-NEXT: vzeroupper ; AVX1-NEXT: retq ; ; AVX2-LABEL: test13: ; AVX2: ## BB#0: ## %vector.ph -; AVX2-NEXT: vmovdqu (%rsi), %ymm0 -; AVX2-NEXT: vpmovzxwd {{.*#+}} ymm1 = mem[0],zero,mem[1],zero,mem[2],zero,mem[3],zero,mem[4],zero,mem[5],zero,mem[6],zero,mem[7],zero +; AVX2-NEXT: vpmovzxwd {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero +; AVX2-NEXT: vpbroadcastd {{.*}}(%rip), %ymm2 +; AVX2-NEXT: vpxor %ymm2, %ymm1, %ymm3 +; AVX2-NEXT: vpxor %ymm2, %ymm0, %ymm2 +; AVX2-NEXT: vpcmpgtd %ymm2, %ymm3, %ymm2 +; AVX2-NEXT: vextracti128 $1, %ymm2, %xmm3 +; AVX2-NEXT: vpacksswb %xmm3, %xmm2, %xmm2 +; AVX2-NEXT: vpsubd %ymm1, %ymm0, %ymm0 +; AVX2-NEXT: vpshufb {{.*#+}} ymm0 = ymm0[0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31] +; AVX2-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3] +; AVX2-NEXT: vpandn %xmm0, %xmm2, %xmm0 +; AVX2-NEXT: vzeroupper +; AVX2-NEXT: retq +vector.ph: + %lhs = zext <8 x i16> %x to <8 x i32> + %cond = icmp ult <8 x i32> %lhs, %y + %sub = sub <8 x i32> %lhs, %y + %trunc = trunc <8 x i32> %sub to <8 x i16> + %res = select <8 x i1> %cond, <8 x i16> zeroinitializer, <8 x i16> %trunc + ret <8 x i16> %res +} + +define <16 x i8> @test14(<16 x i8> %x, <16 x i32> %y) nounwind { +; SSE2-LABEL: test14: +; SSE2: ## BB#0: ## %vector.ph +; SSE2-NEXT: movdqa %xmm0, %xmm5 +; SSE2-NEXT: pxor %xmm0, %xmm0 +; SSE2-NEXT: movdqa %xmm5, %xmm6 +; SSE2-NEXT: punpcklbw {{.*#+}} xmm6 = xmm6[0],xmm0[0],xmm6[1],xmm0[1],xmm6[2],xmm0[2],xmm6[3],xmm0[3],xmm6[4],xmm0[4],xmm6[5],xmm0[5],xmm6[6],xmm0[6],xmm6[7],xmm0[7] +; SSE2-NEXT: movdqa %xmm6, %xmm8 +; SSE2-NEXT: punpcklwd {{.*#+}} xmm8 = xmm8[0],xmm0[0],xmm8[1],xmm0[1],xmm8[2],xmm0[2],xmm8[3],xmm0[3] +; SSE2-NEXT: punpckhwd {{.*#+}} xmm6 = xmm6[4],xmm0[4],xmm6[5],xmm0[5],xmm6[6],xmm0[6],xmm6[7],xmm0[7] +; SSE2-NEXT: punpckhbw {{.*#+}} xmm5 = xmm5[8],xmm0[8],xmm5[9],xmm0[9],xmm5[10],xmm0[10],xmm5[11],xmm0[11],xmm5[12],xmm0[12],xmm5[13],xmm0[13],xmm5[14],xmm0[14],xmm5[15],xmm0[15] +; SSE2-NEXT: movdqa %xmm5, %xmm10 +; SSE2-NEXT: punpcklwd {{.*#+}} xmm10 = xmm10[0],xmm0[0],xmm10[1],xmm0[1],xmm10[2],xmm0[2],xmm10[3],xmm0[3] +; SSE2-NEXT: punpckhwd {{.*#+}} xmm5 = xmm5[4],xmm0[4],xmm5[5],xmm0[5],xmm5[6],xmm0[6],xmm5[7],xmm0[7] +; SSE2-NEXT: movdqa {{.*#+}} xmm0 = [2147483648,2147483648,2147483648,2147483648] +; SSE2-NEXT: movdqa %xmm4, %xmm9 +; SSE2-NEXT: pxor %xmm0, %xmm9 +; SSE2-NEXT: psubd %xmm5, %xmm4 +; SSE2-NEXT: pxor %xmm0, %xmm5 +; SSE2-NEXT: pcmpgtd %xmm9, %xmm5 +; SSE2-NEXT: movdqa {{.*#+}} xmm9 = [255,255,255,255] +; SSE2-NEXT: pand %xmm9, %xmm5 +; SSE2-NEXT: movdqa %xmm3, %xmm7 +; SSE2-NEXT: pxor %xmm0, %xmm7 +; SSE2-NEXT: psubd %xmm10, %xmm3 +; SSE2-NEXT: pxor %xmm0, %xmm10 +; SSE2-NEXT: pcmpgtd %xmm7, %xmm10 +; SSE2-NEXT: pand %xmm9, %xmm10 +; SSE2-NEXT: packuswb %xmm5, %xmm10 +; SSE2-NEXT: movdqa %xmm2, %xmm5 +; SSE2-NEXT: pxor %xmm0, %xmm5 +; SSE2-NEXT: psubd %xmm6, %xmm2 +; SSE2-NEXT: pxor %xmm0, %xmm6 +; SSE2-NEXT: pcmpgtd %xmm5, %xmm6 +; SSE2-NEXT: pand %xmm9, %xmm6 +; SSE2-NEXT: movdqa %xmm1, %xmm5 +; SSE2-NEXT: pxor %xmm0, %xmm5 +; SSE2-NEXT: pxor %xmm8, %xmm0 +; SSE2-NEXT: pcmpgtd %xmm5, %xmm0 +; SSE2-NEXT: pand %xmm9, %xmm0 +; SSE2-NEXT: packuswb %xmm6, %xmm0 +; SSE2-NEXT: packuswb %xmm10, %xmm0 +; SSE2-NEXT: psubd %xmm8, %xmm1 +; SSE2-NEXT: pand %xmm9, %xmm4 +; SSE2-NEXT: pand %xmm9, %xmm3 +; SSE2-NEXT: packuswb %xmm4, %xmm3 +; SSE2-NEXT: pand %xmm9, %xmm2 +; SSE2-NEXT: pand %xmm9, %xmm1 +; SSE2-NEXT: packuswb %xmm2, %xmm1 +; SSE2-NEXT: packuswb %xmm3, %xmm1 +; SSE2-NEXT: pandn %xmm1, %xmm0 +; SSE2-NEXT: retq +; +; SSSE3-LABEL: test14: +; SSSE3: ## BB#0: ## %vector.ph +; SSSE3-NEXT: pxor %xmm7, %xmm7 +; SSSE3-NEXT: movdqa %xmm0, %xmm11 +; SSSE3-NEXT: punpcklbw {{.*#+}} xmm11 = xmm11[0],xmm7[0],xmm11[1],xmm7[1],xmm11[2],xmm7[2],xmm11[3],xmm7[3],xmm11[4],xmm7[4],xmm11[5],xmm7[5],xmm11[6],xmm7[6],xmm11[7],xmm7[7] +; SSSE3-NEXT: movdqa %xmm11, %xmm8 +; SSSE3-NEXT: punpcklwd {{.*#+}} xmm8 = xmm8[0],xmm7[0],xmm8[1],xmm7[1],xmm8[2],xmm7[2],xmm8[3],xmm7[3] +; SSSE3-NEXT: punpckhwd {{.*#+}} xmm11 = xmm11[4],xmm7[4],xmm11[5],xmm7[5],xmm11[6],xmm7[6],xmm11[7],xmm7[7] +; SSSE3-NEXT: punpckhbw {{.*#+}} xmm0 = xmm0[8],xmm7[8],xmm0[9],xmm7[9],xmm0[10],xmm7[10],xmm0[11],xmm7[11],xmm0[12],xmm7[12],xmm0[13],xmm7[13],xmm0[14],xmm7[14],xmm0[15],xmm7[15] +; SSSE3-NEXT: movdqa %xmm0, %xmm10 +; SSSE3-NEXT: punpcklwd {{.*#+}} xmm10 = xmm10[0],xmm7[0],xmm10[1],xmm7[1],xmm10[2],xmm7[2],xmm10[3],xmm7[3] +; SSSE3-NEXT: punpckhwd {{.*#+}} xmm0 = xmm0[4],xmm7[4],xmm0[5],xmm7[5],xmm0[6],xmm7[6],xmm0[7],xmm7[7] +; SSSE3-NEXT: movdqa {{.*#+}} xmm7 = [2147483648,2147483648,2147483648,2147483648] +; SSSE3-NEXT: movdqa %xmm4, %xmm9 +; SSSE3-NEXT: pxor %xmm7, %xmm9 +; SSSE3-NEXT: psubd %xmm0, %xmm4 +; SSSE3-NEXT: movdqa %xmm0, %xmm6 +; SSSE3-NEXT: pxor %xmm7, %xmm6 +; SSSE3-NEXT: pcmpgtd %xmm9, %xmm6 +; SSSE3-NEXT: movdqa {{.*#+}} xmm9 = +; SSSE3-NEXT: pshufb %xmm9, %xmm6 +; SSSE3-NEXT: movdqa %xmm3, %xmm5 +; SSSE3-NEXT: pxor %xmm7, %xmm5 +; SSSE3-NEXT: psubd %xmm10, %xmm3 +; SSSE3-NEXT: movdqa %xmm10, %xmm0 +; SSSE3-NEXT: pxor %xmm7, %xmm0 +; SSSE3-NEXT: pcmpgtd %xmm5, %xmm0 +; SSSE3-NEXT: pshufb %xmm9, %xmm0 +; SSSE3-NEXT: punpckldq {{.*#+}} xmm0 = xmm0[0],xmm6[0],xmm0[1],xmm6[1] +; SSSE3-NEXT: movdqa %xmm2, %xmm5 +; SSSE3-NEXT: pxor %xmm7, %xmm5 +; SSSE3-NEXT: psubd %xmm11, %xmm2 +; SSSE3-NEXT: pxor %xmm7, %xmm11 +; SSSE3-NEXT: pcmpgtd %xmm5, %xmm11 +; SSSE3-NEXT: movdqa {{.*#+}} xmm5 = <0,4,8,12,u,u,u,u,u,u,u,u,u,u,u,u> +; SSSE3-NEXT: pshufb %xmm5, %xmm11 +; SSSE3-NEXT: movdqa %xmm1, %xmm6 +; SSSE3-NEXT: pxor %xmm7, %xmm6 +; SSSE3-NEXT: pxor %xmm8, %xmm7 +; SSSE3-NEXT: pcmpgtd %xmm6, %xmm7 +; SSSE3-NEXT: pshufb %xmm5, %xmm7 +; SSSE3-NEXT: punpckldq {{.*#+}} xmm7 = xmm7[0],xmm11[0],xmm7[1],xmm11[1] +; SSSE3-NEXT: movsd {{.*#+}} xmm0 = xmm7[0],xmm0[1] +; SSSE3-NEXT: psubd %xmm8, %xmm1 +; SSSE3-NEXT: movdqa {{.*#+}} xmm5 = [255,0,0,0,255,0,0,0,255,0,0,0,255,0,0,0] +; SSSE3-NEXT: pand %xmm5, %xmm4 +; SSSE3-NEXT: pand %xmm5, %xmm3 +; SSSE3-NEXT: packuswb %xmm4, %xmm3 +; SSSE3-NEXT: pand %xmm5, %xmm2 +; SSSE3-NEXT: pand %xmm5, %xmm1 +; SSSE3-NEXT: packuswb %xmm2, %xmm1 +; SSSE3-NEXT: packuswb %xmm3, %xmm1 +; SSSE3-NEXT: andnpd %xmm1, %xmm0 +; SSSE3-NEXT: retq +; +; SSE41-LABEL: test14: +; SSE41: ## BB#0: ## %vector.ph +; SSE41-NEXT: movdqa %xmm0, %xmm5 +; SSE41-NEXT: pshufd {{.*#+}} xmm0 = xmm5[1,1,2,3] +; SSE41-NEXT: pmovzxbd {{.*#+}} xmm8 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero +; SSE41-NEXT: pmovzxbd {{.*#+}} xmm0 = xmm5[0],zero,zero,zero,xmm5[1],zero,zero,zero,xmm5[2],zero,zero,zero,xmm5[3],zero,zero,zero +; SSE41-NEXT: pshufd {{.*#+}} xmm6 = xmm5[2,3,0,1] +; SSE41-NEXT: pmovzxbd {{.*#+}} xmm9 = xmm6[0],zero,zero,zero,xmm6[1],zero,zero,zero,xmm6[2],zero,zero,zero,xmm6[3],zero,zero,zero +; SSE41-NEXT: pshufd {{.*#+}} xmm5 = xmm5[3,1,2,3] +; SSE41-NEXT: pmovzxbd {{.*#+}} xmm6 = xmm5[0],zero,zero,zero,xmm5[1],zero,zero,zero,xmm5[2],zero,zero,zero,xmm5[3],zero,zero,zero +; SSE41-NEXT: movdqa {{.*#+}} xmm5 = [2147483648,2147483648,2147483648,2147483648] +; SSE41-NEXT: movdqa %xmm4, %xmm7 +; SSE41-NEXT: pxor %xmm5, %xmm7 +; SSE41-NEXT: psubd %xmm6, %xmm4 +; SSE41-NEXT: pxor %xmm5, %xmm6 +; SSE41-NEXT: pcmpgtd %xmm7, %xmm6 +; SSE41-NEXT: movdqa {{.*#+}} xmm10 = +; SSE41-NEXT: pshufb %xmm10, %xmm6 +; SSE41-NEXT: movdqa %xmm3, %xmm7 +; SSE41-NEXT: pxor %xmm5, %xmm7 +; SSE41-NEXT: psubd %xmm9, %xmm3 +; SSE41-NEXT: pxor %xmm5, %xmm9 +; SSE41-NEXT: pcmpgtd %xmm7, %xmm9 +; SSE41-NEXT: pshufb %xmm10, %xmm9 +; SSE41-NEXT: punpckldq {{.*#+}} xmm9 = xmm9[0],xmm6[0],xmm9[1],xmm6[1] +; SSE41-NEXT: movdqa %xmm1, %xmm6 +; SSE41-NEXT: pxor %xmm5, %xmm6 +; SSE41-NEXT: psubd %xmm0, %xmm1 +; SSE41-NEXT: pxor %xmm5, %xmm0 +; SSE41-NEXT: pcmpgtd %xmm6, %xmm0 +; SSE41-NEXT: movdqa {{.*#+}} xmm6 = <0,4,8,12,u,u,u,u,u,u,u,u,u,u,u,u> +; SSE41-NEXT: pshufb %xmm6, %xmm0 +; SSE41-NEXT: movdqa %xmm2, %xmm7 +; SSE41-NEXT: pxor %xmm5, %xmm7 +; SSE41-NEXT: pxor %xmm8, %xmm5 +; SSE41-NEXT: pcmpgtd %xmm7, %xmm5 +; SSE41-NEXT: pshufb %xmm6, %xmm5 +; SSE41-NEXT: punpckldq {{.*#+}} xmm0 = xmm0[0],xmm5[0],xmm0[1],xmm5[1] +; SSE41-NEXT: pblendw {{.*#+}} xmm0 = xmm0[0,1,2,3],xmm9[4,5,6,7] +; SSE41-NEXT: psubd %xmm8, %xmm2 +; SSE41-NEXT: movdqa {{.*#+}} xmm5 = [255,0,0,0,255,0,0,0,255,0,0,0,255,0,0,0] +; SSE41-NEXT: pand %xmm5, %xmm4 +; SSE41-NEXT: pand %xmm5, %xmm3 +; SSE41-NEXT: packuswb %xmm4, %xmm3 +; SSE41-NEXT: pand %xmm5, %xmm1 +; SSE41-NEXT: pand %xmm5, %xmm2 +; SSE41-NEXT: packuswb %xmm2, %xmm1 +; SSE41-NEXT: packuswb %xmm3, %xmm1 +; SSE41-NEXT: pxor %xmm2, %xmm2 +; SSE41-NEXT: pblendvb %xmm0, %xmm2, %xmm1 +; SSE41-NEXT: movdqa %xmm1, %xmm0 +; SSE41-NEXT: retq +; +; AVX1-LABEL: test14: +; AVX1: ## BB#0: ## %vector.ph +; AVX1-NEXT: vpshufd {{.*#+}} xmm3 = xmm0[1,1,2,3] +; AVX1-NEXT: vpmovzxbd {{.*#+}} xmm8 = xmm3[0],zero,zero,zero,xmm3[1],zero,zero,zero,xmm3[2],zero,zero,zero,xmm3[3],zero,zero,zero +; AVX1-NEXT: vpmovzxbd {{.*#+}} xmm9 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero +; AVX1-NEXT: vpshufd {{.*#+}} xmm5 = xmm0[2,3,0,1] +; AVX1-NEXT: vpmovzxbd {{.*#+}} xmm10 = xmm5[0],zero,zero,zero,xmm5[1],zero,zero,zero,xmm5[2],zero,zero,zero,xmm5[3],zero,zero,zero +; AVX1-NEXT: vpshufd {{.*#+}} xmm0 = xmm0[3,1,2,3] +; AVX1-NEXT: vpmovzxbd {{.*#+}} xmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero +; AVX1-NEXT: vmovdqa {{.*#+}} xmm6 = [2147483648,2147483648,2147483648,2147483648] +; AVX1-NEXT: vextractf128 $1, %ymm2, %xmm7 +; AVX1-NEXT: vpxor %xmm6, %xmm7, %xmm3 +; AVX1-NEXT: vpxor %xmm6, %xmm0, %xmm4 +; AVX1-NEXT: vpcmpgtd %xmm3, %xmm4, %xmm3 +; AVX1-NEXT: vpxor %xmm6, %xmm2, %xmm4 +; AVX1-NEXT: vpxor %xmm6, %xmm10, %xmm5 +; AVX1-NEXT: vpcmpgtd %xmm4, %xmm5, %xmm4 +; AVX1-NEXT: vpacksswb %xmm3, %xmm4, %xmm11 +; AVX1-NEXT: vpxor %xmm6, %xmm1, %xmm4 +; AVX1-NEXT: vpxor %xmm6, %xmm9, %xmm5 +; AVX1-NEXT: vpcmpgtd %xmm4, %xmm5, %xmm4 +; AVX1-NEXT: vextractf128 $1, %ymm1, %xmm5 +; AVX1-NEXT: vpxor %xmm6, %xmm5, %xmm3 +; AVX1-NEXT: vpxor %xmm6, %xmm8, %xmm6 +; AVX1-NEXT: vpcmpgtd %xmm3, %xmm6, %xmm3 +; AVX1-NEXT: vpacksswb %xmm3, %xmm4, %xmm3 +; AVX1-NEXT: vpacksswb %xmm11, %xmm3, %xmm3 +; AVX1-NEXT: vpsubd %xmm8, %xmm5, %xmm4 +; AVX1-NEXT: vpsubd %xmm9, %xmm1, %xmm1 +; AVX1-NEXT: vpsubd %xmm10, %xmm2, %xmm2 +; AVX1-NEXT: vpsubd %xmm0, %xmm7, %xmm0 +; AVX1-NEXT: vmovdqa {{.*#+}} xmm5 = [255,0,0,0,255,0,0,0,255,0,0,0,255,0,0,0] +; AVX1-NEXT: vpand %xmm5, %xmm0, %xmm0 +; AVX1-NEXT: vpand %xmm5, %xmm2, %xmm2 +; AVX1-NEXT: vpackuswb %xmm0, %xmm2, %xmm0 +; AVX1-NEXT: vpand %xmm5, %xmm1, %xmm1 +; AVX1-NEXT: vpand %xmm5, %xmm4, %xmm2 +; AVX1-NEXT: vpackuswb %xmm2, %xmm1, %xmm1 +; AVX1-NEXT: vpackuswb %xmm0, %xmm1, %xmm0 +; AVX1-NEXT: vpxor %xmm1, %xmm1, %xmm1 +; AVX1-NEXT: vpblendvb %xmm3, %xmm1, %xmm0, %xmm0 +; AVX1-NEXT: vzeroupper +; AVX1-NEXT: retq +; +; AVX2-LABEL: test14: +; AVX2: ## BB#0: ## %vector.ph +; AVX2-NEXT: vpshufd {{.*#+}} xmm3 = xmm0[2,3,0,1] +; AVX2-NEXT: vpmovzxbd {{.*#+}} ymm3 = xmm3[0],zero,zero,zero,xmm3[1],zero,zero,zero,xmm3[2],zero,zero,zero,xmm3[3],zero,zero,zero,xmm3[4],zero,zero,zero,xmm3[5],zero,zero,zero,xmm3[6],zero,zero,zero,xmm3[7],zero,zero,zero +; AVX2-NEXT: vpmovzxbd {{.*#+}} ymm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero +; AVX2-NEXT: vpbroadcastd {{.*}}(%rip), %ymm4 +; AVX2-NEXT: vpxor %ymm4, %ymm1, %ymm5 +; AVX2-NEXT: vpxor %ymm4, %ymm0, %ymm6 +; AVX2-NEXT: vpcmpgtd %ymm5, %ymm6, %ymm5 +; AVX2-NEXT: vextracti128 $1, %ymm5, %xmm6 +; AVX2-NEXT: vpacksswb %xmm6, %xmm5, %xmm5 +; AVX2-NEXT: vpxor %ymm4, %ymm2, %ymm6 +; AVX2-NEXT: vpxor %ymm4, %ymm3, %ymm4 +; AVX2-NEXT: vpcmpgtd %ymm6, %ymm4, %ymm4 +; AVX2-NEXT: vextracti128 $1, %ymm4, %xmm6 +; AVX2-NEXT: vpacksswb %xmm6, %xmm4, %xmm4 +; AVX2-NEXT: vpacksswb %xmm4, %xmm5, %xmm4 +; AVX2-NEXT: vpsubd %ymm3, %ymm2, %ymm2 +; AVX2-NEXT: vpsubd %ymm0, %ymm1, %ymm0 +; AVX2-NEXT: vmovdqa {{.*#+}} ymm1 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31] +; AVX2-NEXT: vpshufb %ymm1, %ymm0, %ymm0 +; AVX2-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3] +; AVX2-NEXT: vmovdqa {{.*#+}} xmm3 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u> +; AVX2-NEXT: vpshufb %xmm3, %xmm0, %xmm0 +; AVX2-NEXT: vpshufb %ymm1, %ymm2, %ymm1 +; AVX2-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,2,2,3] +; AVX2-NEXT: vpshufb %xmm3, %xmm1, %xmm1 +; AVX2-NEXT: vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0] +; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1 +; AVX2-NEXT: vpblendvb %xmm4, %xmm1, %xmm0, %xmm0 +; AVX2-NEXT: vzeroupper +; AVX2-NEXT: retq +vector.ph: + %rhs = zext <16 x i8> %x to <16 x i32> + %cond = icmp ult <16 x i32> %y, %rhs + %sub = sub <16 x i32> %y, %rhs + %truncsub = trunc <16 x i32> %sub to <16 x i8> + %res = select <16 x i1> %cond, <16 x i8> zeroinitializer, <16 x i8> %truncsub + ret <16 x i8> %res +} + +define <8 x i16> @test15(<8 x i16> %x, <8 x i32> %y) nounwind { +; SSE2-LABEL: test15: +; SSE2: ## BB#0: ## %vector.ph +; SSE2-NEXT: pxor %xmm4, %xmm4 +; SSE2-NEXT: movdqa %xmm0, %xmm3 +; SSE2-NEXT: punpcklwd {{.*#+}} xmm3 = xmm3[0],xmm4[0],xmm3[1],xmm4[1],xmm3[2],xmm4[2],xmm3[3],xmm4[3] +; SSE2-NEXT: punpckhwd {{.*#+}} xmm0 = xmm0[4],xmm4[4],xmm0[5],xmm4[5],xmm0[6],xmm4[6],xmm0[7],xmm4[7] +; SSE2-NEXT: movdqa {{.*#+}} xmm4 = [2147483648,2147483648,2147483648,2147483648] +; SSE2-NEXT: movdqa %xmm0, %xmm5 +; SSE2-NEXT: psubd %xmm2, %xmm0 +; SSE2-NEXT: pxor %xmm4, %xmm2 +; SSE2-NEXT: pxor %xmm4, %xmm5 +; SSE2-NEXT: pcmpgtd %xmm2, %xmm5 +; SSE2-NEXT: pshuflw {{.*#+}} xmm2 = xmm5[0,2,2,3,4,5,6,7] +; SSE2-NEXT: pshufhw {{.*#+}} xmm2 = xmm2[0,1,2,3,4,6,6,7] +; SSE2-NEXT: pshufd {{.*#+}} xmm2 = xmm2[0,2,2,3] +; SSE2-NEXT: movdqa %xmm1, %xmm5 +; SSE2-NEXT: pxor %xmm4, %xmm5 +; SSE2-NEXT: pxor %xmm3, %xmm4 +; SSE2-NEXT: pcmpgtd %xmm5, %xmm4 +; SSE2-NEXT: pshuflw {{.*#+}} xmm4 = xmm4[0,2,2,3,4,5,6,7] +; SSE2-NEXT: pshufhw {{.*#+}} xmm4 = xmm4[0,1,2,3,4,6,6,7] +; SSE2-NEXT: pshufd {{.*#+}} xmm4 = xmm4[0,2,2,3] +; SSE2-NEXT: punpcklqdq {{.*#+}} xmm4 = xmm4[0],xmm2[0] +; SSE2-NEXT: psubd %xmm1, %xmm3 +; SSE2-NEXT: pslld $16, %xmm0 +; SSE2-NEXT: psrad $16, %xmm0 +; SSE2-NEXT: pslld $16, %xmm3 +; SSE2-NEXT: psrad $16, %xmm3 +; SSE2-NEXT: packssdw %xmm0, %xmm3 +; SSE2-NEXT: pand %xmm4, %xmm3 +; SSE2-NEXT: movdqa %xmm3, %xmm0 +; SSE2-NEXT: retq +; +; SSSE3-LABEL: test15: +; SSSE3: ## BB#0: ## %vector.ph +; SSSE3-NEXT: pxor %xmm4, %xmm4 +; SSSE3-NEXT: movdqa %xmm0, %xmm3 +; SSSE3-NEXT: punpcklwd {{.*#+}} xmm3 = xmm3[0],xmm4[0],xmm3[1],xmm4[1],xmm3[2],xmm4[2],xmm3[3],xmm4[3] +; SSSE3-NEXT: punpckhwd {{.*#+}} xmm0 = xmm0[4],xmm4[4],xmm0[5],xmm4[5],xmm0[6],xmm4[6],xmm0[7],xmm4[7] +; SSSE3-NEXT: movdqa {{.*#+}} xmm4 = [2147483648,2147483648,2147483648,2147483648] +; SSSE3-NEXT: movdqa %xmm0, %xmm5 +; SSSE3-NEXT: psubd %xmm2, %xmm0 +; SSSE3-NEXT: pxor %xmm4, %xmm2 +; SSSE3-NEXT: pxor %xmm4, %xmm5 +; SSSE3-NEXT: pcmpgtd %xmm2, %xmm5 +; SSSE3-NEXT: movdqa {{.*#+}} xmm2 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15] +; SSSE3-NEXT: pshufb %xmm2, %xmm5 +; SSSE3-NEXT: movdqa %xmm1, %xmm6 +; SSSE3-NEXT: pxor %xmm4, %xmm6 +; SSSE3-NEXT: pxor %xmm3, %xmm4 +; SSSE3-NEXT: pcmpgtd %xmm6, %xmm4 +; SSSE3-NEXT: pshufb %xmm2, %xmm4 +; SSSE3-NEXT: punpcklqdq {{.*#+}} xmm4 = xmm4[0],xmm5[0] +; SSSE3-NEXT: psubd %xmm1, %xmm3 +; SSSE3-NEXT: pshufb %xmm2, %xmm0 +; SSSE3-NEXT: pshufb %xmm2, %xmm3 +; SSSE3-NEXT: punpcklqdq {{.*#+}} xmm3 = xmm3[0],xmm0[0] +; SSSE3-NEXT: pand %xmm4, %xmm3 +; SSSE3-NEXT: movdqa %xmm3, %xmm0 +; SSSE3-NEXT: retq +; +; SSE41-LABEL: test15: +; SSE41: ## BB#0: ## %vector.ph +; SSE41-NEXT: pshufd {{.*#+}} xmm3 = xmm0[2,3,0,1] +; SSE41-NEXT: pmovzxwd {{.*#+}} xmm3 = xmm3[0],zero,xmm3[1],zero,xmm3[2],zero,xmm3[3],zero +; SSE41-NEXT: pmovzxwd {{.*#+}} xmm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero +; SSE41-NEXT: movdqa {{.*#+}} xmm4 = [2147483648,2147483648,2147483648,2147483648] +; SSE41-NEXT: movdqa %xmm0, %xmm5 +; SSE41-NEXT: psubd %xmm1, %xmm0 +; SSE41-NEXT: pxor %xmm4, %xmm1 +; SSE41-NEXT: pxor %xmm4, %xmm5 +; SSE41-NEXT: pcmpgtd %xmm1, %xmm5 +; SSE41-NEXT: movdqa {{.*#+}} xmm1 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15] +; SSE41-NEXT: pshufb %xmm1, %xmm5 +; SSE41-NEXT: movdqa %xmm2, %xmm6 +; SSE41-NEXT: pxor %xmm4, %xmm6 +; SSE41-NEXT: pxor %xmm3, %xmm4 +; SSE41-NEXT: pcmpgtd %xmm6, %xmm4 +; SSE41-NEXT: pshufb %xmm1, %xmm4 +; SSE41-NEXT: punpcklqdq {{.*#+}} xmm5 = xmm5[0],xmm4[0] +; SSE41-NEXT: psubd %xmm2, %xmm3 +; SSE41-NEXT: pshufb %xmm1, %xmm0 +; SSE41-NEXT: pshufb %xmm1, %xmm3 +; SSE41-NEXT: punpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm3[0] +; SSE41-NEXT: pand %xmm5, %xmm0 +; SSE41-NEXT: retq +; +; AVX1-LABEL: test15: +; AVX1: ## BB#0: ## %vector.ph +; AVX1-NEXT: vpshufd {{.*#+}} xmm2 = xmm0[2,3,0,1] +; AVX1-NEXT: vpmovzxwd {{.*#+}} xmm2 = xmm2[0],zero,xmm2[1],zero,xmm2[2],zero,xmm2[3],zero +; AVX1-NEXT: vpmovzxwd {{.*#+}} xmm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero +; AVX1-NEXT: vmovdqa {{.*#+}} xmm3 = [2147483648,2147483648,2147483648,2147483648] +; AVX1-NEXT: vpxor %xmm3, %xmm1, %xmm4 +; AVX1-NEXT: vpxor %xmm3, %xmm0, %xmm5 +; AVX1-NEXT: vpcmpgtd %xmm4, %xmm5, %xmm4 +; AVX1-NEXT: vextractf128 $1, %ymm1, %xmm5 +; AVX1-NEXT: vpxor %xmm3, %xmm5, %xmm6 +; AVX1-NEXT: vpxor %xmm3, %xmm2, %xmm3 +; AVX1-NEXT: vpcmpgtd %xmm6, %xmm3, %xmm3 +; AVX1-NEXT: vpacksswb %xmm3, %xmm4, %xmm3 +; AVX1-NEXT: vpsubd %xmm5, %xmm2, %xmm2 +; AVX1-NEXT: vpsubd %xmm1, %xmm0, %xmm0 +; AVX1-NEXT: vmovdqa {{.*#+}} xmm1 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15] +; AVX1-NEXT: vpshufb %xmm1, %xmm0, %xmm0 +; AVX1-NEXT: vpshufb %xmm1, %xmm2, %xmm1 +; AVX1-NEXT: vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0] +; AVX1-NEXT: vpand %xmm0, %xmm3, %xmm0 +; AVX1-NEXT: vzeroupper +; AVX1-NEXT: retq +; +; AVX2-LABEL: test15: +; AVX2: ## BB#0: ## %vector.ph +; AVX2-NEXT: vpmovzxwd {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero ; AVX2-NEXT: vpbroadcastd {{.*}}(%rip), %ymm2 ; AVX2-NEXT: vpxor %ymm2, %ymm1, %ymm3 ; AVX2-NEXT: vpxor %ymm2, %ymm0, %ymm2 ; AVX2-NEXT: vpcmpgtd %ymm3, %ymm2, %ymm2 ; AVX2-NEXT: vextracti128 $1, %ymm2, %xmm3 ; AVX2-NEXT: vpacksswb %xmm3, %xmm2, %xmm2 -; AVX2-NEXT: vpsubd %ymm0, %ymm1, %ymm0 -; AVX2-NEXT: vpshufb {{.*#+}} ymm0 = ymm0[0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31] -; AVX2-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3] -; AVX2-NEXT: vpandn %xmm0, %xmm2, %xmm0 -; AVX2-NEXT: vmovdqu %xmm0, (%rdi) -; AVX2-NEXT: vzeroupper -; AVX2-NEXT: retq -vector.ph: - %0 = getelementptr inbounds i16, i16* %head, i64 0 - %1 = bitcast i16* %0 to <8 x i16>* - %2 = load <8 x i16>, <8 x i16>* %1, align 2 - %3 = getelementptr inbounds i32, i32* %w, i64 0 - %4 = bitcast i32* %3 to <8 x i32>* - %5 = load <8 x i32>, <8 x i32>* %4, align 2 - %6 = zext <8 x i16> %2 to <8 x i32> - %7 = icmp ult <8 x i32> %6, %5 - %8 = sub <8 x i32> %6, %5 - %9 = trunc <8 x i32> %8 to <8 x i16> - %10 = select <8 x i1> %7, <8 x i16> zeroinitializer, <8 x i16> %9 - store <8 x i16> %10, <8 x i16>* %1, align 1 - ret void -} - -define void @test14(i8* nocapture %head, i32* nocapture %w) nounwind { -; SSE2-LABEL: test14: -; SSE2: ## BB#0: ## %vector.ph -; SSE2-NEXT: movdqu (%rdi), %xmm0 -; SSE2-NEXT: movdqu (%rsi), %xmm8 -; SSE2-NEXT: movdqu 16(%rsi), %xmm9 -; SSE2-NEXT: movdqu 32(%rsi), %xmm10 -; SSE2-NEXT: movdqu 48(%rsi), %xmm7 -; SSE2-NEXT: pxor %xmm3, %xmm3 -; SSE2-NEXT: movdqa %xmm0, %xmm1 -; SSE2-NEXT: punpcklbw {{.*#+}} xmm1 = xmm1[0],xmm3[0],xmm1[1],xmm3[1],xmm1[2],xmm3[2],xmm1[3],xmm3[3],xmm1[4],xmm3[4],xmm1[5],xmm3[5],xmm1[6],xmm3[6],xmm1[7],xmm3[7] -; SSE2-NEXT: movdqa %xmm1, %xmm2 -; SSE2-NEXT: punpcklwd {{.*#+}} xmm2 = xmm2[0],xmm3[0],xmm2[1],xmm3[1],xmm2[2],xmm3[2],xmm2[3],xmm3[3] -; SSE2-NEXT: punpckhwd {{.*#+}} xmm1 = xmm1[4],xmm3[4],xmm1[5],xmm3[5],xmm1[6],xmm3[6],xmm1[7],xmm3[7] -; SSE2-NEXT: punpckhbw {{.*#+}} xmm0 = xmm0[8],xmm3[8],xmm0[9],xmm3[9],xmm0[10],xmm3[10],xmm0[11],xmm3[11],xmm0[12],xmm3[12],xmm0[13],xmm3[13],xmm0[14],xmm3[14],xmm0[15],xmm3[15] -; SSE2-NEXT: movdqa %xmm0, %xmm6 -; SSE2-NEXT: punpcklwd {{.*#+}} xmm6 = xmm6[0],xmm3[0],xmm6[1],xmm3[1],xmm6[2],xmm3[2],xmm6[3],xmm3[3] -; SSE2-NEXT: punpckhwd {{.*#+}} xmm0 = xmm0[4],xmm3[4],xmm0[5],xmm3[5],xmm0[6],xmm3[6],xmm0[7],xmm3[7] -; SSE2-NEXT: movdqa {{.*#+}} xmm3 = [2147483648,2147483648,2147483648,2147483648] -; SSE2-NEXT: movdqa %xmm0, %xmm5 -; SSE2-NEXT: psubd %xmm7, %xmm0 -; SSE2-NEXT: pxor %xmm3, %xmm7 -; SSE2-NEXT: pxor %xmm3, %xmm5 -; SSE2-NEXT: pcmpgtd %xmm5, %xmm7 -; SSE2-NEXT: movdqa {{.*#+}} xmm5 = [255,255,255,255] -; SSE2-NEXT: pand %xmm5, %xmm7 -; SSE2-NEXT: movdqa %xmm6, %xmm4 -; SSE2-NEXT: psubd %xmm10, %xmm6 -; SSE2-NEXT: pxor %xmm3, %xmm10 -; SSE2-NEXT: pxor %xmm3, %xmm4 -; SSE2-NEXT: pcmpgtd %xmm4, %xmm10 -; SSE2-NEXT: pand %xmm5, %xmm10 -; SSE2-NEXT: packuswb %xmm7, %xmm10 -; SSE2-NEXT: movdqa %xmm1, %xmm4 -; SSE2-NEXT: psubd %xmm9, %xmm1 -; SSE2-NEXT: pxor %xmm3, %xmm9 -; SSE2-NEXT: pxor %xmm3, %xmm4 -; SSE2-NEXT: pcmpgtd %xmm4, %xmm9 -; SSE2-NEXT: pand %xmm5, %xmm9 -; SSE2-NEXT: movdqa %xmm8, %xmm4 -; SSE2-NEXT: pxor %xmm3, %xmm4 -; SSE2-NEXT: pxor %xmm2, %xmm3 -; SSE2-NEXT: pcmpgtd %xmm3, %xmm4 -; SSE2-NEXT: pand %xmm5, %xmm4 -; SSE2-NEXT: packuswb %xmm9, %xmm4 -; SSE2-NEXT: packuswb %xmm10, %xmm4 -; SSE2-NEXT: psubd %xmm8, %xmm2 -; SSE2-NEXT: pand %xmm5, %xmm0 -; SSE2-NEXT: pand %xmm5, %xmm6 -; SSE2-NEXT: packuswb %xmm0, %xmm6 -; SSE2-NEXT: pand %xmm5, %xmm1 -; SSE2-NEXT: pand %xmm5, %xmm2 -; SSE2-NEXT: packuswb %xmm1, %xmm2 -; SSE2-NEXT: packuswb %xmm6, %xmm2 -; SSE2-NEXT: pandn %xmm2, %xmm4 -; SSE2-NEXT: movdqu %xmm4, (%rdi) -; SSE2-NEXT: retq -; -; SSSE3-LABEL: test14: -; SSSE3: ## BB#0: ## %vector.ph -; SSSE3-NEXT: movdqu (%rdi), %xmm0 -; SSSE3-NEXT: movdqu (%rsi), %xmm8 -; SSSE3-NEXT: movdqu 16(%rsi), %xmm9 -; SSSE3-NEXT: movdqu 32(%rsi), %xmm10 -; SSSE3-NEXT: movdqu 48(%rsi), %xmm7 -; SSSE3-NEXT: pxor %xmm3, %xmm3 -; SSSE3-NEXT: movdqa %xmm0, %xmm1 -; SSSE3-NEXT: punpcklbw {{.*#+}} xmm1 = xmm1[0],xmm3[0],xmm1[1],xmm3[1],xmm1[2],xmm3[2],xmm1[3],xmm3[3],xmm1[4],xmm3[4],xmm1[5],xmm3[5],xmm1[6],xmm3[6],xmm1[7],xmm3[7] -; SSSE3-NEXT: movdqa %xmm1, %xmm2 -; SSSE3-NEXT: punpcklwd {{.*#+}} xmm2 = xmm2[0],xmm3[0],xmm2[1],xmm3[1],xmm2[2],xmm3[2],xmm2[3],xmm3[3] -; SSSE3-NEXT: punpckhwd {{.*#+}} xmm1 = xmm1[4],xmm3[4],xmm1[5],xmm3[5],xmm1[6],xmm3[6],xmm1[7],xmm3[7] -; SSSE3-NEXT: punpckhbw {{.*#+}} xmm0 = xmm0[8],xmm3[8],xmm0[9],xmm3[9],xmm0[10],xmm3[10],xmm0[11],xmm3[11],xmm0[12],xmm3[12],xmm0[13],xmm3[13],xmm0[14],xmm3[14],xmm0[15],xmm3[15] -; SSSE3-NEXT: movdqa %xmm0, %xmm6 -; SSSE3-NEXT: punpcklwd {{.*#+}} xmm6 = xmm6[0],xmm3[0],xmm6[1],xmm3[1],xmm6[2],xmm3[2],xmm6[3],xmm3[3] -; SSSE3-NEXT: punpckhwd {{.*#+}} xmm0 = xmm0[4],xmm3[4],xmm0[5],xmm3[5],xmm0[6],xmm3[6],xmm0[7],xmm3[7] -; SSSE3-NEXT: movdqa {{.*#+}} xmm3 = [2147483648,2147483648,2147483648,2147483648] -; SSSE3-NEXT: movdqa %xmm0, %xmm5 -; SSSE3-NEXT: psubd %xmm7, %xmm0 -; SSSE3-NEXT: pxor %xmm3, %xmm7 -; SSSE3-NEXT: pxor %xmm3, %xmm5 -; SSSE3-NEXT: pcmpgtd %xmm5, %xmm7 -; SSSE3-NEXT: movdqa {{.*#+}} xmm5 = -; SSSE3-NEXT: pshufb %xmm5, %xmm7 -; SSSE3-NEXT: movdqa %xmm6, %xmm4 -; SSSE3-NEXT: psubd %xmm10, %xmm6 -; SSSE3-NEXT: pxor %xmm3, %xmm10 -; SSSE3-NEXT: pxor %xmm3, %xmm4 -; SSSE3-NEXT: pcmpgtd %xmm4, %xmm10 -; SSSE3-NEXT: pshufb %xmm5, %xmm10 -; SSSE3-NEXT: punpckldq {{.*#+}} xmm10 = xmm10[0],xmm7[0],xmm10[1],xmm7[1] -; SSSE3-NEXT: movdqa %xmm1, %xmm4 -; SSSE3-NEXT: psubd %xmm9, %xmm1 -; SSSE3-NEXT: pxor %xmm3, %xmm9 -; SSSE3-NEXT: pxor %xmm3, %xmm4 -; SSSE3-NEXT: pcmpgtd %xmm4, %xmm9 -; SSSE3-NEXT: movdqa {{.*#+}} xmm4 = <0,4,8,12,u,u,u,u,u,u,u,u,u,u,u,u> -; SSSE3-NEXT: pshufb %xmm4, %xmm9 -; SSSE3-NEXT: movdqa %xmm8, %xmm5 -; SSSE3-NEXT: pxor %xmm3, %xmm5 -; SSSE3-NEXT: pxor %xmm2, %xmm3 -; SSSE3-NEXT: pcmpgtd %xmm3, %xmm5 -; SSSE3-NEXT: pshufb %xmm4, %xmm5 -; SSSE3-NEXT: punpckldq {{.*#+}} xmm5 = xmm5[0],xmm9[0],xmm5[1],xmm9[1] -; SSSE3-NEXT: movsd {{.*#+}} xmm10 = xmm5[0],xmm10[1] -; SSSE3-NEXT: psubd %xmm8, %xmm2 -; SSSE3-NEXT: movdqa {{.*#+}} xmm3 = [255,0,0,0,255,0,0,0,255,0,0,0,255,0,0,0] -; SSSE3-NEXT: pand %xmm3, %xmm0 -; SSSE3-NEXT: pand %xmm3, %xmm6 -; SSSE3-NEXT: packuswb %xmm0, %xmm6 -; SSSE3-NEXT: pand %xmm3, %xmm1 -; SSSE3-NEXT: pand %xmm3, %xmm2 -; SSSE3-NEXT: packuswb %xmm1, %xmm2 -; SSSE3-NEXT: packuswb %xmm6, %xmm2 -; SSSE3-NEXT: andnpd %xmm2, %xmm10 -; SSSE3-NEXT: movupd %xmm10, (%rdi) -; SSSE3-NEXT: retq -; -; AVX1-LABEL: test14: -; AVX1: ## BB#0: ## %vector.ph -; AVX1-NEXT: vmovdqu (%rsi), %ymm0 -; AVX1-NEXT: vmovdqu 32(%rsi), %ymm1 -; AVX1-NEXT: vpmovzxbd {{.*#+}} xmm8 = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero -; AVX1-NEXT: vpmovzxbd {{.*#+}} xmm9 = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero -; AVX1-NEXT: vpmovzxbd {{.*#+}} xmm10 = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero -; AVX1-NEXT: vpmovzxbd {{.*#+}} xmm5 = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero -; AVX1-NEXT: vmovdqa {{.*#+}} xmm6 = [2147483648,2147483648,2147483648,2147483648] -; AVX1-NEXT: vpxor %xmm6, %xmm5, %xmm7 -; AVX1-NEXT: vextractf128 $1, %ymm1, %xmm2 -; AVX1-NEXT: vpxor %xmm6, %xmm2, %xmm3 -; AVX1-NEXT: vpcmpgtd %xmm7, %xmm3, %xmm3 -; AVX1-NEXT: vpxor %xmm6, %xmm10, %xmm7 -; AVX1-NEXT: vpxor %xmm6, %xmm1, %xmm4 -; AVX1-NEXT: vpcmpgtd %xmm7, %xmm4, %xmm4 -; AVX1-NEXT: vpacksswb %xmm3, %xmm4, %xmm11 -; AVX1-NEXT: vpxor %xmm6, %xmm9, %xmm4 -; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm7 -; AVX1-NEXT: vpxor %xmm6, %xmm7, %xmm3 -; AVX1-NEXT: vpcmpgtd %xmm4, %xmm3, %xmm3 -; AVX1-NEXT: vpxor %xmm6, %xmm8, %xmm4 -; AVX1-NEXT: vpxor %xmm6, %xmm0, %xmm6 -; AVX1-NEXT: vpcmpgtd %xmm4, %xmm6, %xmm4 -; AVX1-NEXT: vpacksswb %xmm3, %xmm4, %xmm3 -; AVX1-NEXT: vpacksswb %xmm11, %xmm3, %xmm3 -; AVX1-NEXT: vpsubd %xmm0, %xmm8, %xmm0 -; AVX1-NEXT: vpsubd %xmm7, %xmm9, %xmm4 -; AVX1-NEXT: vpsubd %xmm1, %xmm10, %xmm1 -; AVX1-NEXT: vpsubd %xmm2, %xmm5, %xmm2 -; AVX1-NEXT: vmovdqa {{.*#+}} xmm5 = [255,0,0,0,255,0,0,0,255,0,0,0,255,0,0,0] -; AVX1-NEXT: vpand %xmm5, %xmm2, %xmm2 -; AVX1-NEXT: vpand %xmm5, %xmm1, %xmm1 -; AVX1-NEXT: vpackuswb %xmm2, %xmm1, %xmm1 -; AVX1-NEXT: vpand %xmm5, %xmm4, %xmm2 -; AVX1-NEXT: vpand %xmm5, %xmm0, %xmm0 -; AVX1-NEXT: vpackuswb %xmm2, %xmm0, %xmm0 -; AVX1-NEXT: vpackuswb %xmm1, %xmm0, %xmm0 -; AVX1-NEXT: vpxor %xmm1, %xmm1, %xmm1 -; AVX1-NEXT: vpblendvb %xmm3, %xmm1, %xmm0, %xmm0 -; AVX1-NEXT: vmovdqu %xmm0, (%rdi) -; AVX1-NEXT: vzeroupper -; AVX1-NEXT: retq -; -; AVX2-LABEL: test14: -; AVX2: ## BB#0: ## %vector.ph -; AVX2-NEXT: vmovdqu (%rsi), %ymm0 -; AVX2-NEXT: vmovdqu 32(%rsi), %ymm1 -; AVX2-NEXT: vpmovzxbd {{.*#+}} ymm2 = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero,mem[4],zero,zero,zero,mem[5],zero,zero,zero,mem[6],zero,zero,zero,mem[7],zero,zero,zero -; AVX2-NEXT: vpmovzxbd {{.*#+}} ymm3 = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero,mem[4],zero,zero,zero,mem[5],zero,zero,zero,mem[6],zero,zero,zero,mem[7],zero,zero,zero -; AVX2-NEXT: vpbroadcastd {{.*}}(%rip), %ymm4 -; AVX2-NEXT: vpxor %ymm4, %ymm3, %ymm5 -; AVX2-NEXT: vpxor %ymm4, %ymm1, %ymm6 -; AVX2-NEXT: vpcmpgtd %ymm5, %ymm6, %ymm5 -; AVX2-NEXT: vextracti128 $1, %ymm5, %xmm6 -; AVX2-NEXT: vpacksswb %xmm6, %xmm5, %xmm5 -; AVX2-NEXT: vpxor %ymm4, %ymm2, %ymm6 -; AVX2-NEXT: vpxor %ymm4, %ymm0, %ymm4 -; AVX2-NEXT: vpcmpgtd %ymm6, %ymm4, %ymm4 -; AVX2-NEXT: vextracti128 $1, %ymm4, %xmm6 -; AVX2-NEXT: vpacksswb %xmm6, %xmm4, %xmm4 -; AVX2-NEXT: vpacksswb %xmm5, %xmm4, %xmm4 -; AVX2-NEXT: vpsubd %ymm0, %ymm2, %ymm0 -; AVX2-NEXT: vpsubd %ymm1, %ymm3, %ymm1 -; AVX2-NEXT: vmovdqa {{.*#+}} ymm2 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31] -; AVX2-NEXT: vpshufb %ymm2, %ymm1, %ymm1 -; AVX2-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,2,2,3] -; AVX2-NEXT: vmovdqa {{.*#+}} xmm3 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u> -; AVX2-NEXT: vpshufb %xmm3, %xmm1, %xmm1 -; AVX2-NEXT: vpshufb %ymm2, %ymm0, %ymm0 -; AVX2-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3] -; AVX2-NEXT: vpshufb %xmm3, %xmm0, %xmm0 -; AVX2-NEXT: vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0] -; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1 -; AVX2-NEXT: vpblendvb %xmm4, %xmm1, %xmm0, %xmm0 -; AVX2-NEXT: vmovdqu %xmm0, (%rdi) -; AVX2-NEXT: vzeroupper -; AVX2-NEXT: retq -vector.ph: - %0 = getelementptr inbounds i8, i8* %head, i64 0 - %1 = bitcast i8* %0 to <16 x i8>* - %2 = load <16 x i8>, <16 x i8>* %1, align 2 - %3 = getelementptr inbounds i32, i32* %w, i64 0 - %4 = bitcast i32* %3 to <16 x i32>* - %5 = load <16 x i32>, <16 x i32>* %4, align 2 - %6 = zext <16 x i8> %2 to <16 x i32> - %7 = icmp ult <16 x i32> %6, %5 - %8 = sub <16 x i32> %6, %5 - %9 = trunc <16 x i32> %8 to <16 x i8> - %10 = select <16 x i1> %7, <16 x i8> zeroinitializer, <16 x i8> %9 - store <16 x i8> %10, <16 x i8>* %1, align 1 - ret void -} - -define void @test15(i16* nocapture %head, i32* nocapture %w) nounwind { -; SSE2-LABEL: test15: -; SSE2: ## BB#0: ## %vector.ph -; SSE2-NEXT: movdqu (%rdi), %xmm0 -; SSE2-NEXT: movdqu (%rsi), %xmm2 -; SSE2-NEXT: movdqu 16(%rsi), %xmm3 -; SSE2-NEXT: pxor %xmm4, %xmm4 -; SSE2-NEXT: movdqa %xmm0, %xmm1 -; SSE2-NEXT: punpcklwd {{.*#+}} xmm1 = xmm1[0],xmm4[0],xmm1[1],xmm4[1],xmm1[2],xmm4[2],xmm1[3],xmm4[3] -; SSE2-NEXT: punpckhwd {{.*#+}} xmm0 = xmm0[4],xmm4[4],xmm0[5],xmm4[5],xmm0[6],xmm4[6],xmm0[7],xmm4[7] -; SSE2-NEXT: movdqa {{.*#+}} xmm4 = [2147483648,2147483648,2147483648,2147483648] -; SSE2-NEXT: movdqa %xmm0, %xmm5 -; SSE2-NEXT: psubd %xmm3, %xmm0 -; SSE2-NEXT: pxor %xmm4, %xmm3 -; SSE2-NEXT: pxor %xmm4, %xmm5 -; SSE2-NEXT: pcmpgtd %xmm3, %xmm5 -; SSE2-NEXT: pshuflw {{.*#+}} xmm3 = xmm5[0,2,2,3,4,5,6,7] -; SSE2-NEXT: pshufhw {{.*#+}} xmm3 = xmm3[0,1,2,3,4,6,6,7] -; SSE2-NEXT: pshufd {{.*#+}} xmm3 = xmm3[0,2,2,3] -; SSE2-NEXT: movdqa %xmm2, %xmm5 -; SSE2-NEXT: pxor %xmm4, %xmm5 -; SSE2-NEXT: pxor %xmm1, %xmm4 -; SSE2-NEXT: pcmpgtd %xmm5, %xmm4 -; SSE2-NEXT: pshuflw {{.*#+}} xmm4 = xmm4[0,2,2,3,4,5,6,7] -; SSE2-NEXT: pshufhw {{.*#+}} xmm4 = xmm4[0,1,2,3,4,6,6,7] -; SSE2-NEXT: pshufd {{.*#+}} xmm4 = xmm4[0,2,2,3] -; SSE2-NEXT: punpcklqdq {{.*#+}} xmm4 = xmm4[0],xmm3[0] -; SSE2-NEXT: psubd %xmm2, %xmm1 -; SSE2-NEXT: pslld $16, %xmm0 -; SSE2-NEXT: psrad $16, %xmm0 -; SSE2-NEXT: pslld $16, %xmm1 -; SSE2-NEXT: psrad $16, %xmm1 -; SSE2-NEXT: packssdw %xmm0, %xmm1 -; SSE2-NEXT: pand %xmm4, %xmm1 -; SSE2-NEXT: movdqu %xmm1, (%rdi) -; SSE2-NEXT: retq -; -; SSSE3-LABEL: test15: -; SSSE3: ## BB#0: ## %vector.ph -; SSSE3-NEXT: movdqu (%rdi), %xmm0 -; SSSE3-NEXT: movdqu (%rsi), %xmm2 -; SSSE3-NEXT: movdqu 16(%rsi), %xmm4 -; SSSE3-NEXT: pxor %xmm3, %xmm3 -; SSSE3-NEXT: movdqa %xmm0, %xmm1 -; SSSE3-NEXT: punpcklwd {{.*#+}} xmm1 = xmm1[0],xmm3[0],xmm1[1],xmm3[1],xmm1[2],xmm3[2],xmm1[3],xmm3[3] -; SSSE3-NEXT: punpckhwd {{.*#+}} xmm0 = xmm0[4],xmm3[4],xmm0[5],xmm3[5],xmm0[6],xmm3[6],xmm0[7],xmm3[7] -; SSSE3-NEXT: movdqa {{.*#+}} xmm3 = [2147483648,2147483648,2147483648,2147483648] -; SSSE3-NEXT: movdqa %xmm0, %xmm5 -; SSSE3-NEXT: psubd %xmm4, %xmm0 -; SSSE3-NEXT: pxor %xmm3, %xmm4 -; SSSE3-NEXT: pxor %xmm3, %xmm5 -; SSSE3-NEXT: pcmpgtd %xmm4, %xmm5 -; SSSE3-NEXT: movdqa {{.*#+}} xmm4 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15] -; SSSE3-NEXT: pshufb %xmm4, %xmm5 -; SSSE3-NEXT: movdqa %xmm2, %xmm6 -; SSSE3-NEXT: pxor %xmm3, %xmm6 -; SSSE3-NEXT: pxor %xmm1, %xmm3 -; SSSE3-NEXT: pcmpgtd %xmm6, %xmm3 -; SSSE3-NEXT: pshufb %xmm4, %xmm3 -; SSSE3-NEXT: punpcklqdq {{.*#+}} xmm3 = xmm3[0],xmm5[0] -; SSSE3-NEXT: psubd %xmm2, %xmm1 -; SSSE3-NEXT: pshufb %xmm4, %xmm0 -; SSSE3-NEXT: pshufb %xmm4, %xmm1 -; SSSE3-NEXT: punpcklqdq {{.*#+}} xmm1 = xmm1[0],xmm0[0] -; SSSE3-NEXT: pand %xmm3, %xmm1 -; SSSE3-NEXT: movdqu %xmm1, (%rdi) -; SSSE3-NEXT: retq -; -; AVX1-LABEL: test15: -; AVX1: ## BB#0: ## %vector.ph -; AVX1-NEXT: vmovdqu (%rsi), %ymm0 -; AVX1-NEXT: vpmovzxwd {{.*#+}} xmm1 = mem[0],zero,mem[1],zero,mem[2],zero,mem[3],zero -; AVX1-NEXT: vpmovzxwd {{.*#+}} xmm2 = mem[0],zero,mem[1],zero,mem[2],zero,mem[3],zero -; AVX1-NEXT: vmovdqa {{.*#+}} xmm3 = [2147483648,2147483648,2147483648,2147483648] -; AVX1-NEXT: vpxor %xmm3, %xmm2, %xmm4 -; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm5 -; AVX1-NEXT: vpxor %xmm3, %xmm5, %xmm6 -; AVX1-NEXT: vpcmpgtd %xmm6, %xmm4, %xmm4 -; AVX1-NEXT: vpxor %xmm3, %xmm1, %xmm6 -; AVX1-NEXT: vpxor %xmm3, %xmm0, %xmm3 -; AVX1-NEXT: vpcmpgtd %xmm3, %xmm6, %xmm3 -; AVX1-NEXT: vpacksswb %xmm4, %xmm3, %xmm3 -; AVX1-NEXT: vpsubd %xmm0, %xmm1, %xmm0 -; AVX1-NEXT: vpsubd %xmm5, %xmm2, %xmm1 -; AVX1-NEXT: vmovdqa {{.*#+}} xmm2 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15] -; AVX1-NEXT: vpshufb %xmm2, %xmm1, %xmm1 -; AVX1-NEXT: vpshufb %xmm2, %xmm0, %xmm0 -; AVX1-NEXT: vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0] -; AVX1-NEXT: vpand %xmm0, %xmm3, %xmm0 -; AVX1-NEXT: vmovdqu %xmm0, (%rdi) -; AVX1-NEXT: vzeroupper -; AVX1-NEXT: retq -; -; AVX2-LABEL: test15: -; AVX2: ## BB#0: ## %vector.ph -; AVX2-NEXT: vmovdqu (%rsi), %ymm0 -; AVX2-NEXT: vpmovzxwd {{.*#+}} ymm1 = mem[0],zero,mem[1],zero,mem[2],zero,mem[3],zero,mem[4],zero,mem[5],zero,mem[6],zero,mem[7],zero -; AVX2-NEXT: vpbroadcastd {{.*}}(%rip), %ymm2 -; AVX2-NEXT: vpxor %ymm2, %ymm0, %ymm3 -; AVX2-NEXT: vpxor %ymm2, %ymm1, %ymm2 -; AVX2-NEXT: vpcmpgtd %ymm3, %ymm2, %ymm2 -; AVX2-NEXT: vextracti128 $1, %ymm2, %xmm3 -; AVX2-NEXT: vpacksswb %xmm3, %xmm2, %xmm2 -; AVX2-NEXT: vpsubd %ymm0, %ymm1, %ymm0 +; AVX2-NEXT: vpsubd %ymm1, %ymm0, %ymm0 ; AVX2-NEXT: vpshufb {{.*#+}} ymm0 = ymm0[0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31] ; AVX2-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3] ; AVX2-NEXT: vpand %xmm0, %xmm2, %xmm0 -; AVX2-NEXT: vmovdqu %xmm0, (%rdi) ; AVX2-NEXT: vzeroupper ; AVX2-NEXT: retq vector.ph: - %0 = getelementptr inbounds i16, i16* %head, i64 0 - %1 = bitcast i16* %0 to <8 x i16>* - %2 = load <8 x i16>, <8 x i16>* %1, align 2 - %3 = getelementptr inbounds i32, i32* %w, i64 0 - %4 = bitcast i32* %3 to <8 x i32>* - %5 = load <8 x i32>, <8 x i32>* %4, align 2 - %6 = zext <8 x i16> %2 to <8 x i32> - %7 = icmp ugt <8 x i32> %6, %5 - %8 = sub <8 x i32> %6, %5 - %9 = trunc <8 x i32> %8 to <8 x i16> - %10 = select <8 x i1> %7, <8 x i16> %9, <8 x i16> zeroinitializer - store <8 x i16> %10, <8 x i16>* %1, align 1 - ret void + %lhs = zext <8 x i16> %x to <8 x i32> + %cond = icmp ugt <8 x i32> %lhs, %y + %sub = sub <8 x i32> %lhs, %y + %truncsub = trunc <8 x i32> %sub to <8 x i16> + %res = select <8 x i1> %cond, <8 x i16> %truncsub, <8 x i16> zeroinitializer + ret <8 x i16> %res } -define void @test16(i16* nocapture %head, i32* nocapture %w) nounwind { +define <8 x i16> @test16(<8 x i16> %x, <8 x i32> %y) nounwind { ; SSE2-LABEL: test16: ; SSE2: ## BB#0: ## %vector.ph -; SSE2-NEXT: movdqu (%rdi), %xmm0 -; SSE2-NEXT: movdqu (%rsi), %xmm2 -; SSE2-NEXT: movdqu 16(%rsi), %xmm3 ; SSE2-NEXT: pxor %xmm4, %xmm4 -; SSE2-NEXT: movdqa %xmm0, %xmm1 -; SSE2-NEXT: punpcklwd {{.*#+}} xmm1 = xmm1[0],xmm4[0],xmm1[1],xmm4[1],xmm1[2],xmm4[2],xmm1[3],xmm4[3] +; SSE2-NEXT: movdqa %xmm0, %xmm3 +; SSE2-NEXT: punpcklwd {{.*#+}} xmm3 = xmm3[0],xmm4[0],xmm3[1],xmm4[1],xmm3[2],xmm4[2],xmm3[3],xmm4[3] ; SSE2-NEXT: punpckhwd {{.*#+}} xmm0 = xmm0[4],xmm4[4],xmm0[5],xmm4[5],xmm0[6],xmm4[6],xmm0[7],xmm4[7] ; SSE2-NEXT: movdqa {{.*#+}} xmm4 = [2147483648,2147483648,2147483648,2147483648] ; SSE2-NEXT: movdqa %xmm0, %xmm5 -; SSE2-NEXT: psubd %xmm3, %xmm0 -; SSE2-NEXT: pxor %xmm4, %xmm3 +; SSE2-NEXT: psubd %xmm2, %xmm0 +; SSE2-NEXT: pxor %xmm4, %xmm2 ; SSE2-NEXT: pxor %xmm4, %xmm5 -; SSE2-NEXT: pcmpgtd %xmm3, %xmm5 -; SSE2-NEXT: pshuflw {{.*#+}} xmm3 = xmm5[0,2,2,3,4,5,6,7] -; SSE2-NEXT: pshufhw {{.*#+}} xmm3 = xmm3[0,1,2,3,4,6,6,7] -; SSE2-NEXT: pshufd {{.*#+}} xmm3 = xmm3[0,2,2,3] -; SSE2-NEXT: movdqa %xmm2, %xmm5 +; SSE2-NEXT: pcmpgtd %xmm2, %xmm5 +; SSE2-NEXT: pshuflw {{.*#+}} xmm2 = xmm5[0,2,2,3,4,5,6,7] +; SSE2-NEXT: pshufhw {{.*#+}} xmm2 = xmm2[0,1,2,3,4,6,6,7] +; SSE2-NEXT: pshufd {{.*#+}} xmm2 = xmm2[0,2,2,3] +; SSE2-NEXT: movdqa %xmm1, %xmm5 ; SSE2-NEXT: pxor %xmm4, %xmm5 -; SSE2-NEXT: pxor %xmm1, %xmm4 +; SSE2-NEXT: pxor %xmm3, %xmm4 ; SSE2-NEXT: pcmpgtd %xmm5, %xmm4 ; SSE2-NEXT: pshuflw {{.*#+}} xmm4 = xmm4[0,2,2,3,4,5,6,7] ; SSE2-NEXT: pshufhw {{.*#+}} xmm4 = xmm4[0,1,2,3,4,6,6,7] ; SSE2-NEXT: pshufd {{.*#+}} xmm4 = xmm4[0,2,2,3] -; SSE2-NEXT: punpcklqdq {{.*#+}} xmm4 = xmm4[0],xmm3[0] -; SSE2-NEXT: psubd %xmm2, %xmm1 +; SSE2-NEXT: punpcklqdq {{.*#+}} xmm4 = xmm4[0],xmm2[0] +; SSE2-NEXT: psubd %xmm1, %xmm3 ; SSE2-NEXT: pslld $16, %xmm0 ; SSE2-NEXT: psrad $16, %xmm0 -; SSE2-NEXT: pslld $16, %xmm1 -; SSE2-NEXT: psrad $16, %xmm1 -; SSE2-NEXT: packssdw %xmm0, %xmm1 -; SSE2-NEXT: pand %xmm4, %xmm1 -; SSE2-NEXT: movdqu %xmm1, (%rdi) +; SSE2-NEXT: pslld $16, %xmm3 +; SSE2-NEXT: psrad $16, %xmm3 +; SSE2-NEXT: packssdw %xmm0, %xmm3 +; SSE2-NEXT: pand %xmm4, %xmm3 +; SSE2-NEXT: movdqa %xmm3, %xmm0 ; SSE2-NEXT: retq ; ; SSSE3-LABEL: test16: ; SSSE3: ## BB#0: ## %vector.ph -; SSSE3-NEXT: movdqu (%rdi), %xmm0 -; SSSE3-NEXT: movdqu (%rsi), %xmm2 -; SSSE3-NEXT: movdqu 16(%rsi), %xmm4 -; SSSE3-NEXT: pxor %xmm3, %xmm3 -; SSSE3-NEXT: movdqa %xmm0, %xmm1 -; SSSE3-NEXT: punpcklwd {{.*#+}} xmm1 = xmm1[0],xmm3[0],xmm1[1],xmm3[1],xmm1[2],xmm3[2],xmm1[3],xmm3[3] -; SSSE3-NEXT: punpckhwd {{.*#+}} xmm0 = xmm0[4],xmm3[4],xmm0[5],xmm3[5],xmm0[6],xmm3[6],xmm0[7],xmm3[7] -; SSSE3-NEXT: movdqa {{.*#+}} xmm3 = [2147483648,2147483648,2147483648,2147483648] +; SSSE3-NEXT: pxor %xmm4, %xmm4 +; SSSE3-NEXT: movdqa %xmm0, %xmm3 +; SSSE3-NEXT: punpcklwd {{.*#+}} xmm3 = xmm3[0],xmm4[0],xmm3[1],xmm4[1],xmm3[2],xmm4[2],xmm3[3],xmm4[3] +; SSSE3-NEXT: punpckhwd {{.*#+}} xmm0 = xmm0[4],xmm4[4],xmm0[5],xmm4[5],xmm0[6],xmm4[6],xmm0[7],xmm4[7] +; SSSE3-NEXT: movdqa {{.*#+}} xmm4 = [2147483648,2147483648,2147483648,2147483648] ; SSSE3-NEXT: movdqa %xmm0, %xmm5 -; SSSE3-NEXT: psubd %xmm4, %xmm0 +; SSSE3-NEXT: psubd %xmm2, %xmm0 +; SSSE3-NEXT: pxor %xmm4, %xmm2 +; SSSE3-NEXT: pxor %xmm4, %xmm5 +; SSSE3-NEXT: pcmpgtd %xmm2, %xmm5 +; SSSE3-NEXT: movdqa {{.*#+}} xmm2 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15] +; SSSE3-NEXT: pshufb %xmm2, %xmm5 +; SSSE3-NEXT: movdqa %xmm1, %xmm6 +; SSSE3-NEXT: pxor %xmm4, %xmm6 ; SSSE3-NEXT: pxor %xmm3, %xmm4 -; SSSE3-NEXT: pxor %xmm3, %xmm5 -; SSSE3-NEXT: pcmpgtd %xmm4, %xmm5 -; SSSE3-NEXT: movdqa {{.*#+}} xmm4 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15] -; SSSE3-NEXT: pshufb %xmm4, %xmm5 -; SSSE3-NEXT: movdqa %xmm2, %xmm6 -; SSSE3-NEXT: pxor %xmm3, %xmm6 -; SSSE3-NEXT: pxor %xmm1, %xmm3 -; SSSE3-NEXT: pcmpgtd %xmm6, %xmm3 -; SSSE3-NEXT: pshufb %xmm4, %xmm3 -; SSSE3-NEXT: punpcklqdq {{.*#+}} xmm3 = xmm3[0],xmm5[0] -; SSSE3-NEXT: psubd %xmm2, %xmm1 -; SSSE3-NEXT: pshufb %xmm4, %xmm0 -; SSSE3-NEXT: pshufb %xmm4, %xmm1 -; SSSE3-NEXT: punpcklqdq {{.*#+}} xmm1 = xmm1[0],xmm0[0] -; SSSE3-NEXT: pand %xmm3, %xmm1 -; SSSE3-NEXT: movdqu %xmm1, (%rdi) +; SSSE3-NEXT: pcmpgtd %xmm6, %xmm4 +; SSSE3-NEXT: pshufb %xmm2, %xmm4 +; SSSE3-NEXT: punpcklqdq {{.*#+}} xmm4 = xmm4[0],xmm5[0] +; SSSE3-NEXT: psubd %xmm1, %xmm3 +; SSSE3-NEXT: pshufb %xmm2, %xmm0 +; SSSE3-NEXT: pshufb %xmm2, %xmm3 +; SSSE3-NEXT: punpcklqdq {{.*#+}} xmm3 = xmm3[0],xmm0[0] +; SSSE3-NEXT: pand %xmm4, %xmm3 +; SSSE3-NEXT: movdqa %xmm3, %xmm0 ; SSSE3-NEXT: retq ; +; SSE41-LABEL: test16: +; SSE41: ## BB#0: ## %vector.ph +; SSE41-NEXT: pshufd {{.*#+}} xmm3 = xmm0[2,3,0,1] +; SSE41-NEXT: pmovzxwd {{.*#+}} xmm3 = xmm3[0],zero,xmm3[1],zero,xmm3[2],zero,xmm3[3],zero +; SSE41-NEXT: pmovzxwd {{.*#+}} xmm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero +; SSE41-NEXT: movdqa {{.*#+}} xmm4 = [2147483648,2147483648,2147483648,2147483648] +; SSE41-NEXT: movdqa %xmm0, %xmm5 +; SSE41-NEXT: psubd %xmm1, %xmm0 +; SSE41-NEXT: pxor %xmm4, %xmm1 +; SSE41-NEXT: pxor %xmm4, %xmm5 +; SSE41-NEXT: pcmpgtd %xmm1, %xmm5 +; SSE41-NEXT: movdqa {{.*#+}} xmm1 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15] +; SSE41-NEXT: pshufb %xmm1, %xmm5 +; SSE41-NEXT: movdqa %xmm2, %xmm6 +; SSE41-NEXT: pxor %xmm4, %xmm6 +; SSE41-NEXT: pxor %xmm3, %xmm4 +; SSE41-NEXT: pcmpgtd %xmm6, %xmm4 +; SSE41-NEXT: pshufb %xmm1, %xmm4 +; SSE41-NEXT: punpcklqdq {{.*#+}} xmm5 = xmm5[0],xmm4[0] +; SSE41-NEXT: psubd %xmm2, %xmm3 +; SSE41-NEXT: pshufb %xmm1, %xmm0 +; SSE41-NEXT: pshufb %xmm1, %xmm3 +; SSE41-NEXT: punpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm3[0] +; SSE41-NEXT: pand %xmm5, %xmm0 +; SSE41-NEXT: retq +; ; AVX1-LABEL: test16: ; AVX1: ## BB#0: ## %vector.ph -; AVX1-NEXT: vmovdqu (%rsi), %ymm0 -; AVX1-NEXT: vpmovzxwd {{.*#+}} xmm1 = mem[0],zero,mem[1],zero,mem[2],zero,mem[3],zero -; AVX1-NEXT: vpmovzxwd {{.*#+}} xmm2 = mem[0],zero,mem[1],zero,mem[2],zero,mem[3],zero +; AVX1-NEXT: vpshufd {{.*#+}} xmm2 = xmm0[2,3,0,1] +; AVX1-NEXT: vpmovzxwd {{.*#+}} xmm2 = xmm2[0],zero,xmm2[1],zero,xmm2[2],zero,xmm2[3],zero +; AVX1-NEXT: vpmovzxwd {{.*#+}} xmm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero ; AVX1-NEXT: vmovdqa {{.*#+}} xmm3 = [2147483648,2147483648,2147483648,2147483648] -; AVX1-NEXT: vpxor %xmm3, %xmm2, %xmm4 -; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm5 +; AVX1-NEXT: vpxor %xmm3, %xmm1, %xmm4 +; AVX1-NEXT: vpxor %xmm3, %xmm0, %xmm5 +; AVX1-NEXT: vpcmpgtd %xmm4, %xmm5, %xmm4 +; AVX1-NEXT: vextractf128 $1, %ymm1, %xmm5 ; AVX1-NEXT: vpxor %xmm3, %xmm5, %xmm6 -; AVX1-NEXT: vpcmpgtd %xmm6, %xmm4, %xmm4 -; AVX1-NEXT: vpxor %xmm3, %xmm1, %xmm6 -; AVX1-NEXT: vpxor %xmm3, %xmm0, %xmm3 -; AVX1-NEXT: vpcmpgtd %xmm3, %xmm6, %xmm3 -; AVX1-NEXT: vpacksswb %xmm4, %xmm3, %xmm3 -; AVX1-NEXT: vpsubd %xmm0, %xmm1, %xmm0 -; AVX1-NEXT: vpsubd %xmm5, %xmm2, %xmm1 -; AVX1-NEXT: vmovdqa {{.*#+}} xmm2 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15] -; AVX1-NEXT: vpshufb %xmm2, %xmm1, %xmm1 -; AVX1-NEXT: vpshufb %xmm2, %xmm0, %xmm0 +; AVX1-NEXT: vpxor %xmm3, %xmm2, %xmm3 +; AVX1-NEXT: vpcmpgtd %xmm6, %xmm3, %xmm3 +; AVX1-NEXT: vpacksswb %xmm3, %xmm4, %xmm3 +; AVX1-NEXT: vpsubd %xmm5, %xmm2, %xmm2 +; AVX1-NEXT: vpsubd %xmm1, %xmm0, %xmm0 +; AVX1-NEXT: vmovdqa {{.*#+}} xmm1 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15] +; AVX1-NEXT: vpshufb %xmm1, %xmm0, %xmm0 +; AVX1-NEXT: vpshufb %xmm1, %xmm2, %xmm1 ; AVX1-NEXT: vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0] ; AVX1-NEXT: vpand %xmm0, %xmm3, %xmm0 -; AVX1-NEXT: vmovdqu %xmm0, (%rdi) ; AVX1-NEXT: vzeroupper ; AVX1-NEXT: retq ; ; AVX2-LABEL: test16: ; AVX2: ## BB#0: ## %vector.ph -; AVX2-NEXT: vmovdqu (%rsi), %ymm0 -; AVX2-NEXT: vpmovzxwd {{.*#+}} ymm1 = mem[0],zero,mem[1],zero,mem[2],zero,mem[3],zero,mem[4],zero,mem[5],zero,mem[6],zero,mem[7],zero +; AVX2-NEXT: vpmovzxwd {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero ; AVX2-NEXT: vpbroadcastd {{.*}}(%rip), %ymm2 -; AVX2-NEXT: vpxor %ymm2, %ymm0, %ymm3 -; AVX2-NEXT: vpxor %ymm2, %ymm1, %ymm2 +; AVX2-NEXT: vpxor %ymm2, %ymm1, %ymm3 +; AVX2-NEXT: vpxor %ymm2, %ymm0, %ymm2 ; AVX2-NEXT: vpcmpgtd %ymm3, %ymm2, %ymm2 ; AVX2-NEXT: vextracti128 $1, %ymm2, %xmm3 ; AVX2-NEXT: vpacksswb %xmm3, %xmm2, %xmm2 -; AVX2-NEXT: vpsubd %ymm0, %ymm1, %ymm0 +; AVX2-NEXT: vpsubd %ymm1, %ymm0, %ymm0 ; AVX2-NEXT: vpshufb {{.*#+}} ymm0 = ymm0[0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31] ; AVX2-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3] ; AVX2-NEXT: vpand %xmm0, %xmm2, %xmm0 -; AVX2-NEXT: vmovdqu %xmm0, (%rdi) ; AVX2-NEXT: vzeroupper ; AVX2-NEXT: retq vector.ph: - %0 = getelementptr inbounds i16, i16* %head, i64 0 - %1 = bitcast i16* %0 to <8 x i16>* - %2 = load <8 x i16>, <8 x i16>* %1, align 2 - %3 = getelementptr inbounds i32, i32* %w, i64 0 - %4 = bitcast i32* %3 to <8 x i32>* - %5 = load <8 x i32>, <8 x i32>* %4, align 2 - %6 = zext <8 x i16> %2 to <8 x i32> - %7 = icmp ult <8 x i32> %5, %6 - %8 = sub <8 x i32> %6, %5 - %9 = trunc <8 x i32> %8 to <8 x i16> - %10 = select <8 x i1> %7, <8 x i16> %9, <8 x i16> zeroinitializer - store <8 x i16> %10, <8 x i16>* %1, align 1 - ret void + %lhs = zext <8 x i16> %x to <8 x i32> + %cond = icmp ult <8 x i32> %y, %lhs + %sub = sub <8 x i32> %lhs, %y + %truncsub = trunc <8 x i32> %sub to <8 x i16> + %res = select <8 x i1> %cond, <8 x i16> %truncsub, <8 x i16> zeroinitializer + ret <8 x i16> %res } diff --git a/test/CodeGen/X86/store-narrow.ll b/test/CodeGen/X86/store-narrow.ll index 16f152d169d3..5e9e1e364fef 100644 --- a/test/CodeGen/X86/store-narrow.ll +++ b/test/CodeGen/X86/store-narrow.ll @@ -134,10 +134,7 @@ entry: @g_16 = internal global i32 -1 ; X64-LABEL: test8: -; X64-NEXT: movl _g_16(%rip), %eax -; X64-NEXT: movl $0, _g_16(%rip) -; X64-NEXT: orl $1, %eax -; X64-NEXT: movl %eax, _g_16(%rip) +; X64-NEXT: orb $1, _g_16(%rip) ; X64-NEXT: ret define void @test8() nounwind { %tmp = load i32, i32* @g_16 diff --git a/test/CodeGen/X86/swift-return.ll b/test/CodeGen/X86/swift-return.ll index 60e33e62b4ad..0ea176d5d82f 100644 --- a/test/CodeGen/X86/swift-return.ll +++ b/test/CodeGen/X86/swift-return.ll @@ -184,11 +184,11 @@ define void @consume_i1_ret() { %v6 = extractvalue { i1, i1, i1, i1 } %call, 2 %v7 = extractvalue { i1, i1, i1, i1 } %call, 3 %val = zext i1 %v3 to i32 - store i32 %val, i32* @var + store volatile i32 %val, i32* @var %val2 = zext i1 %v5 to i32 - store i32 %val2, i32* @var + store volatile i32 %val2, i32* @var %val3 = zext i1 %v6 to i32 - store i32 %val3, i32* @var + store volatile i32 %val3, i32* @var %val4 = zext i1 %v7 to i32 store i32 %val4, i32* @var ret void diff --git a/test/CodeGen/X86/win32-spill-xmm.ll b/test/CodeGen/X86/win32-spill-xmm.ll index 0db97cfe20f0..c6b163b88b24 100644 --- a/test/CodeGen/X86/win32-spill-xmm.ll +++ b/test/CodeGen/X86/win32-spill-xmm.ll @@ -20,7 +20,7 @@ declare void @bar(<16 x float> %a, i32 %b) ; Check that proper alignment of spilled vector does not affect vargs ; CHECK-LABEL: vargs_not_affected -; CHECK: leal 28(%ebp), %eax +; CHECK: movl 28(%ebp), %eax define i32 @vargs_not_affected(<4 x float> %v, i8* %f, ...) { entry: %ap = alloca i8*, align 4 diff --git a/test/CodeGen/X86/win64_sibcall.ll b/test/CodeGen/X86/win64_sibcall.ll index 4bba0e1e0acd..42dd4d31ca9f 100644 --- a/test/CodeGen/X86/win64_sibcall.ll +++ b/test/CodeGen/X86/win64_sibcall.ll @@ -12,8 +12,8 @@ entry: ; LINUX: movq $0, -8(%rsp) %this = alloca %Object addrspace(1)* - store %Object addrspace(1)* null, %Object addrspace(1)** %this - store %Object addrspace(1)* %param0, %Object addrspace(1)** %this + store volatile %Object addrspace(1)* null, %Object addrspace(1)** %this + store volatile %Object addrspace(1)* %param0, %Object addrspace(1)** %this br label %0 ;