release_80 branch point. Afterwards, we will merge the rest of the
changes in the actual release_80 branch.
PR: 236062
MFC after: 1 month
X-MFC-With: r344779
Fix inline assembler constraint validation
The current constraint logic is both too lax and too strict. It fails
for input outside the [INT_MIN..INT_MAX] range, but it also
implicitly accepts 0 as value when it should not. Adjust logic to
handle both correctly.
Differential Revision: https://reviews.llvm.org/D58649
Pull in r355491 from upstream clang trunk (by Hans Wennborg):
Inline asm constraints: allow ICE-like pointers for the "n"
constraint (PR40890)
Apparently GCC allows this, and there's code relying on it (see bug).
The idea is to allow expression that would have been allowed if they
were cast to int. So I based the code on how such a cast would be
done (the CK_PointerToIntegral case in
IntExprEvaluator::VisitCastExpr()).
Differential Revision: https://reviews.llvm.org/D58821
These should fix assertions and errors when using the inline assembly
"n" constraint in certain ways.
In case of devel/valgrind, a pointer was used as the input for the
constraint, which lead to "Assertion failed: (isInt() && "Invalid
accessor"), function getInt".
In case of math/secp256k1, a very large integer value was used as input
for the constraint, which lead to "error: value '4624529908474429119'
out of range for constraint 'n'".
PR: 236216, 236194
MFC after: 1 month
X-MFC-With: r344779
not occur with gcc 7 or later). This should prevent the following error
from breaking the head-amd64-gcc CI builds:
In file included from /workspace/src/contrib/llvm/tools/lldb/source/API/SBMemoryRegionInfo.cpp:14:0:
/workspace/src/contrib/llvm/tools/lldb/include/lldb/Target/MemoryRegionInfo.h:128:54: error: 'template<class _InputIterator> lldb_private::MemoryRegionInfos::MemoryRegionInfos(_InputIterator, _InputIterator, const allocator_type&)' inherited from 'std::__1::vector<lldb_private::MemoryRegionInfo>'
using std::vector<lldb_private::MemoryRegionInfo>::vector;
^~~~~~
/workspace/src/contrib/llvm/tools/lldb/include/lldb/Target/MemoryRegionInfo.h:128:54: error: conflicts with version inherited from 'std::__1::vector<lldb_private::MemoryRegionInfo>'
Reported by: CI
[X86] Fix tls variable lowering issue with large code model
Summary:
The problem here is the lowering for tls variable. Below is the DAG
for the code. SelectionDAG has 11 nodes:
t0: ch = EntryToken
t8: i64,ch = load<(load 8 from `i8 addrspace(257)* null`,
addrspace 257)> t0, Constant:i64<0>, undef:i64
t10: i64 = X86ISD::WrapperRIP TargetGlobalTLSAddress:i64<i32*
@x> 0 [TF=10]
t11: i64,ch = load<(load 8 from got)> t0, t10, undef:i64
t12: i64 = add t8, t11
t4: i32,ch = load<(dereferenceable load 4 from @x)> t0, t12,
undef:i64
t6: ch = CopyToReg t0, Register:i32 %0, t4
And when mcmodel is large, below instruction can NOT be folded.
t10: i64 = X86ISD::WrapperRIP TargetGlobalTLSAddress:i64<i32* @x> 0
[TF=10]
t11: i64,ch = load<(load 8 from got)> t0, t10, undef:i64
So "t11: i64,ch = load<(load 8 from got)> t0, t10, undef:i64" is
lowered to " Morphed node: t11: i64,ch = MOV64rm<Mem:(load 8 from
got)> t10, TargetConstant:i8<1>, Register:i64 $noreg,
TargetConstant:i32<0>, Register:i32 $noreg, t0"
When llvm start to lower "t10: i64 = X86ISD::WrapperRIP
TargetGlobalTLSAddress:i64<i32* @x> 0 [TF=10]", it fails.
The patch is to fold the load and X86ISD::WrapperRIP.
Fixes PR26906
Patch by LuoYuanke
Reviewers: craig.topper, rnk, annita.zhang, wxiao3
Reviewed By: rnk
Subscribers: llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D58336
This should fix "fatal error: error in backend: Cannot select" messages
when compiling <ctype.h> functions using -mcmodel=large.
Reported by: phk
PR: 233143
MFC after: 3 days
Recommit r353293 "[LLD][ELF] - Set DF_STATIC_TLS flag for i386 target."
With the following changes:
1) Compilation fix:
std::atomic<bool> HasStaticTlsModel = false; ->
std::atomic<bool> HasStaticTlsModel{false};
2) Adjusted the comment in code.
Initial commit message:
DF_STATIC_TLS flag indicates that the shared object or executable
contains code using a static thread-local storage scheme.
Patch checks if IE/LE relocations were used to check if the code uses
a static model. If so it sets the DF_STATIC_TLS flag.
Differential revision: https://reviews.llvm.org/D57749
Pull in r353378 from upstream lld trunk (by George Rimar):
[LLD][ELF] - Set DF_STATIC_TLS flag for X64 target
This is the same as D57749, but for x64 target.
"ELF Handling For Thread-Local Storage" p41 says
(https://www.akkadia.org/drepper/tls.pdf):
R_X86_64_GOTTPOFF relocation is used for IE TLS models.
Hence if linker sees this relocation we should add DF_STATIC_TLS flag.
Differential revision: https://reviews.llvm.org/D57821
This adds support to lld for the DF_STATIC_TLS flag in shared objects,
which signals to the dynamic linker that the shared object requires
static thread local storage.
See also: https://reviews.freebsd.org/D19072
MFC after: 1 week
[MC] Make symbol version errors non-fatal
We stil don't have a source location, which is pretty lame, but at
least we won't tell the user to file a clang bug report anymore.
Fixes PR40712
This will make errors for symbols with @@ versions that are not defined
non-fatal. For example:
void f(void)
{
__asm__(".symver foo,bar@@baz");
}
will now result in:
error: versioned symbol bar@@baz must be defined
instead of clang crashing with a diagnostic report.
PR: 234671
Upstream PR: https://bugs.llvm.org/show_bug.cgi?id=40712
MFC after: 3 days
[ARM] Make PerformSHLSimplify add nodes to the DAG worklist correctly.
Intentionally excluding nodes from the DAGCombine worklist is likely
to lead to weird optimizations and infinite loops, so it's generally
a bad idea.
To avoid the infinite loops, fix DAGCombine to use the
isDesirableToCommuteWithShift target hook before performing the
transforms in question, and implement the target hook in the ARM
backend disable the transforms in question.
Fixes https://bugs.llvm.org/show_bug.cgi?id=38530 . (I don't have a
reduced testcase for that bug. But we should have sufficient test
coverage for PerformSHLSimplify given that we're not playing weird
tricks with the worklist. I can try to bugpoint it if necessary,
though.)
Differential Revision: https://reviews.llvm.org/D50667
This should fix a possible hang when compiling sys/dev/nxge/if_nxge.c
(which exists now only in the stable/11 branch) for arm.
[X86] Add FPSW as a Def on some FP instructions that were missing it.
Pull in r352608 from upstream llvm trunk (by Craig Topper):
[X86] Remove a couple places where we unnecessarily pass 0 to the
EmitPriority of some FP instruction aliases. NFC
As far as I can tell we already won't emit these aliases due to an
operand count check in the tablegen code. Removing these because I
couldn't make sense of the inconsistency between fadd and fmul from
reading the code.
I checked the AsmMatcher and AsmWriter files before and after this
change and there were no differences.
Pull in r353015 from upstream llvm trunk (by Craig Topper):
[X86] Print %st(0) as %st when its implicit to the instruction.
Continue printing it as %st(0) when its encoded in the instruction.
This is a step back from the change I made in r352985. This appears
to be more consistent with gcc and objdump behavior.
Pull in r353061 from upstream llvm trunk (by Craig Topper):
[X86] Print all register forms of x87 fadd/fsub/fdiv/fmul as having
two arguments where on is %st.
All of these instructions consume one encoded register and the other
register is %st. They either write the result to %st or the encoded
register. Previously we printed both arguments when the encoded
register was written. And we printed one argument when the result was
written to %st. For the stack popping forms the encoded register is
always the destination and we didn't print both operands. This was
inconsistent with gcc and objdump and just makes the output assembly
code harder to read.
This patch changes things to always print both operands making us
consistent with gcc and objdump. The parser should still be able to
handle the single register forms just as it did before. This also
matches the GNU assembler behavior.
Pull in r353141 from upstream llvm trunk (by Craig Topper):
[X86] Connect the default fpsr and dirflag clobbers in inline
assembly to the registers we have defined for them.
Summary:
We don't currently map these constraints to physical register numbers
so they don't make it to the MachineIR representation of inline
assembly.
This could have problems for proper dependency tracking in the
machine schedulers though I don't have a test case that shows that.
Reviewers: rnk
Reviewed By: rnk
Subscribers: eraman, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D57641
Pull in r353489 from upstream llvm trunk (by Craig Topper):
[X86] Add FPCW as a register and start using it as an implicit use on
floating point instructions.
Summary:
FPCW contains the rounding mode control which we manipulate to
implement fp to integer conversion by changing the roudning mode,
storing the value to the stack, and then changing the rounding mode
back. Because we didn't model FPCW and its dependency chain, other
instructions could be scheduled into the middle of the sequence.
This patch introduces the register and adds it as an implciit def of
FLDCW and implicit use of the FP binary arithmetic instructions and
store instructions. There are more instructions that need to be
updated, but this is a good start. I believe this fixes at least the
reduced test case from PR40529.
Reviewers: RKSimon, spatel, rnk, efriedma, andrew.w.kaylor
Subscribers: dim, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D57735
These should fix a problem in clang 7.0 where it would sometimes emit
long double floating point instructions in a slightly wrong order,
leading to failures in our libm tests. In particular, the cbrt_test
test case 'cbrtl_powl' and the trig_test test case 'reduction'.
Reported by: lwhsu
PR: 234040
Upstream PR: https://bugs.llvm.org/show_bug.cgi?id=40206
[X86] Add FPSW as a Def on some FP instructions that were missing it.
Pull in r353141 from upstream llvm trunk (by Craig Topper):
[X86] Connect the default fpsr and dirflag clobbers in inline
assembly to the registers we have defined for them.
Summary:
We don't currently map these constraints to physical register numbers
so they don't make it to the MachineIR representation of inline
assembly.
This could have problems for proper dependency tracking in the
machine schedulers though I don't have a test case that shows that.
Reviewers: rnk
Reviewed By: rnk
Subscribers: eraman, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D57641
Pull in r353489 from upstream llvm trunk (by Craig Topper):
[X86] Add FPCW as a register and start using it as an implicit use on
floating point instructions.
Summary:
FPCW contains the rounding mode control which we manipulate to
implement fp to integer conversion by changing the roudning mode,
storing the value to the stack, and then changing the rounding mode
back. Because we didn't model FPCW and its dependency chain, other
instructions could be scheduled into the middle of the sequence.
This patch introduces the register and adds it as an implciit def of
FLDCW and implicit use of the FP binary arithmetic instructions and
store instructions. There are more instructions that need to be
updated, but this is a good start. I believe this fixes at least the
reduced test case from PR40529.
Reviewers: RKSimon, spatel, rnk, efriedma, andrew.w.kaylor
Subscribers: dim, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D57735
These should fix a problem in clang 7.0 where it would sometimes emit
long double floating point instructions in a slightly wrong order,
leading to failures in our libm tests. In particular, the cbrt_test
test case 'cbrtl_powl' and the trig_test test case 'reduction'.
Also bump __FreeBSD_cc_version, to be able to detect this in our test
suite.
Reported by: lwhsu
PR: 234040
Upstream PR: https://bugs.llvm.org/show_bug.cgi?id=40206
MFC after: 1 week
Refactoring. NFC.
Pull in r352435 from upstream lld trunk (by Rui Ueyama):
Attempt to fix build failure with GCC 5.4.
Pull in r352482 from upstream lld trunk (by George Rimar):
[ELF] - Remove dead `readBfdName` declaration. NFC.
`readBfdName` was removed recently.
Pull in r352606 from upstream lld trunk (by me):
Recognize FreeBSD specific BFD names in OUTPUT_FORMAT
Summary:
After rLLD344952 ("Add OUTPUT_FORMAT linker script directive
support"), using BFD names such as `elf64-x86-64-freebsd` the
`OUTPUT_FORMAT` linker script command does not work anymore,
resulting in errors like:
```
ld: error: /home/dim/src/clang800-import/stand/efi/loader/arch/amd64/ldscript.amd64:2: unknown output format name: elf64-x86-64-freebsd
>>> OUTPUT_FORMAT("elf64-x86-64-freebsd", "elf64-x86-64-freebsd", "elf64-x86-64-freebsd")
>>> ^
```
To fix this, recognize a `-freebsd` suffix in BFD names, and also set
`Configuration::OSABI` to `ELFOSABI_FREEBSD` for those cases.
Add and/or update several test cases to check for the correct results
of these new `OUTPUT_FORMAT` arguments.
Reviewers: ruiu, atanasyan, grimar, hokein, emaste, espindola
Reviewed By: ruiu
Subscribers: nemanjai, javed.absar, arichardson, krytarowski, kristof.beyls, kbarton, llvm-commits
Differential Revision: https://reviews.llvm.org/D57283
FreeBSD previously defaulted to DWARF 2 because several tools (gdb,
ctfconvert, etc.) did not support later versions. These have either
been fixed or are deprecated.
Note that gdb 6 still exists but has been moved out of $PATH into
/usr/libexec and is intended only for use by crashinfo(8). The kernel
build sets the DWARF version explicitly via -gdwarf2, so this should
have no effect there.
PR: 234887 [exp-run]
Reviewed by: markj
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D17930
Revert "Revert r342183 "[DAGCombine] Fix crash when store merging
created an extract_subvector with invalid index.""
Fixed the assertion failure.
Differential Revision: https://reviews.llvm.org/D51831
This fixes 'Assertion failed: ((VT.getVectorNumElements() +
N2C->getZExtValue() <= N1.getValueType().getVectorNumElements()) &&
"Extract subvector overflow!"), function getNode' when building the
multimedia/aom port (with AVX2 enabled).
Reported by: jbeich
PR: 234480
MFC after: 6 weeks
X-MFC-With: r341825
7.0.1 release r349250. There were no functional changes since the 7.0.1
rc3 import.
PR: 230240, 230355
Relnotes: yes
MFC after: 2 months
X-MFC-With: r341825
Don't mess up RelIplt symbols during relocatable processing
Summary:
During upgrading of the FreeBSD source tree with lld 7.0.0, I noticed
that it started complaining about crt1.o having an "index past the
end of the symbol table".
Such a symbol table looks approximately like this, viewed with
readelf -s (note the Ndx field being messed up):
Symbol table '.symtab' contains 4 entries:
Num: Value Size Type Bind Vis Ndx Name
0: 00000000 0 NOTYPE LOCAL DEFAULT UND
1: 00000000 0 SECTION LOCAL DEFAULT 1
2: 00000000 0 NOTYPE WEAK HIDDEN RSV[0xffff] __rel_iplt_end
3: 00000000 0 NOTYPE WEAK HIDDEN RSV[0xffff] __rel_iplt_start
At first, it seemed that recent ifunc relocation work had caused this:
<https://reviews.freebsd.org/rS339351>, but it turned out that it was
due to incorrect processing of the object files by lld, when using -r
(a.k.a. --relocatable).
Bisecting showed that rL324421 ("Convert a use of Config->Static") was
the commit where this new behavior began. Simply reverting it solved
the issue, and the __rel_iplt symbols had an index of UND again.
Looking at Rafael's commit message, I think he simply missed the
possibility of --relocatable being in effect, so I have added an
additional check for it.
I also added a simple regression test case.
Reviewers: grimar, ruiu, emaste, espindola
Reviewed By: ruiu
Subscribers: arichardson, krytarowski, llvm-commits
Differential Revision: https://reviews.llvm.org/D53515
This fixes a problem in lld where it places incorrect indexes for ifunc
related symbols in crt1.o and friends, making it impossible to link most
normal programs with it.
ELF spec says that for SHT_REL and SHT_RELA sh_link should reference the
associated string table and sh_info should reference the "section to
which the relocation applies." ELF Tool Chain's elfcopy / strip use
this (in part) to control whether or not the relocation entry is copied
to the output.
LLVM PR 37538 https://bugs.llvm.org/show_bug.cgi?id=37538
Approved by: re (kib)
Obtained from: llvm r344226 (backported for 6.0)
Previously Clang required ifunc resolution functions to take no
arguments, presumably because GCC documented ifunc resolvers as taking
no arguments. However, GCC accepts resolvers accepting arguments, and
our rtld passes CPU ID information (cpuid, hwcap, etc.) to ifunc
resolvers. Just remove the check from the in-tree compiler for our in-
tree compiler; a different (per-OS) approach may be required upstream.
Reported by: mjg
Approved by: re (rgrimes)
MFC after: 1 week
Relnotes: Yes
Sponsored by: The FreeBSD Foundation
[ELF] - Allow LLD to produce file symbols.
This is for PR36716 and
this enables emitting STT_FILE symbols.
Output size affect is minor:
lld binary size changes from 52,883,408 to 52,949,400
clang binary size changes from 83,136,456 to 83,219,600
Differential revision: https://reviews.llvm.org/D45261
This fixes a regression in lld that made it stop emitting STT_FILE
symbols, which ctfmerge relies upon to uniquify function table entries
that reference STB_LOCAL symbols. Consequently, ctfmerge stopped
emitting entries for static functions into the function table, and
dtrace no longer gets type info for them.
Approved by: re (kib)
Reported by: markj
PR: 230444
MFC after: 3 days
[CodeGen] Initialize large arrays by copying from a global
Currently, clang compiles explicit initializers for array elements
into series of store instructions. For large arrays of built-in types
this results in bloated output code and significant amount of time
spent on the instruction selection phase. This patch fixes the issue
by initializing such arrays with global constants that store the
binary image of the initializer.
Differential Revision: https://reviews.llvm.org/D43181
This should fix a compiler hang (and excessive memory usage) while
building the science/rmg port.
Approved by: re (kib)
Reported by: yuri@tsoft.com
See also: https://bugs.llvm.org/show_bug.cgi?id=38798
MFC after: 3 days
Initialize LiveRegs once in BranchFolder::mergeCommonTails
This should fix '(TRI && "LivePhysRegs is not initialized."' assertions
when building the lang/qt5-qml port in certain configurations.
Approved by: re (kib)
Reported by: Piotr Kubaj <pkubaj@anongoth.pl>
PR: 231355
MFC after: 3 days
The current kernel ifunc implementation creates a PLT entry for each
ifunc definition. ifunc calls therefore consist of a call to the
PLT entry followed by an indirect jump. The jump target is written
during boot when the kernel linker resolves R_[*]_IRELATIVE relocations.
This implementation is defined by requirements for userland code, where
text relocations are avoided. This requirement is not present for the
kernel, so the implementation has avoidable overhead (namely, an extra
indirect jump per call).
Address this for now by adding a special option to the static linker
to inhibit PLT creation for ifuncs. Instead, relocations to ifunc call
sites are passed through to the output file, so the kernel linker can
enumerate such call sites and apply PC-relative relocations directly
to the text section. Thus the overhead of an ifunc call becomes exactly
the same as that of an ordinary function call. This option is only for
use by the kernel and will not work for regular programs.
The final form of this optimization is up for debate; for now, this
change is simple and static enough to be acceptable as an interim
solution.
Reviewed by: emaste
Discussed with: arichardson, dim
MFC after: 1 month
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D16748
Adjust MaxAtomicInlineWidth for i386/i486 targets.
This is to fix the bug reported in
https://bugs.llvm.org/show_bug.cgi?id=34347#c6. Currently, all
MaxAtomicInlineWidth of x86-32 targets are set to 64. However, i386
doesn't support any cmpxchg related instructions. i486 only supports
cmpxchg. So in this patch MaxAtomicInlineWidth is reset as follows:
For i386, the MaxAtomicInlineWidth should be 0 because no cmpxchg is
supported. For i486, the MaxAtomicInlineWidth should be 32 because
it supports cmpxchg. For others 32 bits x86 cpu, the
MaxAtomicInlineWidth should be 64 because of cmpxchg8b.
Differential Revision: https://reviews.llvm.org/D42154
This should fix buildworld on i386, because of our system libraries
missing __atomic_load_8, and possibly other 64 bit atomic functions, for
that architecture.
We should really fix that at some point, but since we have been actually
using cmpxchg8b for years now, it does not seem to matter much...
[x86] Fix a really subtle miscompile due to a somewhat glaring bug in
EFLAGS copy lowering.
If you have a branch of LLVM, you may want to cherrypick this. It is
extremely unlikely to hit this case empirically, but it will likely
manifest as an "impossible" branch being taken somewhere, and will be
... very hard to debug.
Hitting this requires complex conditions living across complex
control flow combined with some interesting memory (non-stack)
initialized with the results of a comparison. Also, because you have
to arrange for an EFLAGS copy to be in *just* the right place, almost
anything you do to the code will hide the bug. I was unable to reduce
anything remotely resembling a "good" test case from the place where
I hit it, and so instead I have constructed synthetic MIR testing
that directly exercises the bug in question (as well as the good
behavior for completeness).
The issue is that we would mistakenly assume any SETcc with a valid
condition and an initial operand that was a register and a virtual
register at that to be a register *defining* SETcc...
It isn't though....
This would in turn cause us to test some other bizarre register,
typically the base pointer of some memory. Now, testing this register
and using that to branch on doesn't make any sense. It even fails the
machine verifier (if you are running it) due to the wrong register
class. But it will make it through LLVM, assemble, and it *looks*
fine... But wow do you get a very unsual and surprising branch taken
in your actual code.
The fix is to actually check what kind of SETcc instruction we're
dealing with. Because there are a bunch of them, I just test the
may-store bit in the instruction. I've also added an assert for
sanity that ensure we are, in fact, *defining* the register operand.
=D
Noticed by: kib
MFC after: 1 week
The Tag_ABI_VFP_args build attribute controls the procedure call
standard used for floating point parameters on ARM. The values are:
0 - Base AAPCS (FP Parameters passed in Core (Integer) registers
1 - VFP AAPCS (FP Parameters passed in FP registers)
2 - Toolchain specific (Neither Base or VFP)
3 - Compatible with all (No use of floating point parameters)
If the Tag_ABI_VFP_args build attribute is missing it has an implicit
value of 0.
We use the attribute in two ways:
* Detect a clash in calling convention between Base, VFP and Toolchain.
we follow ld.bfd's lead and do not error if there is a clash between an
implicit Base AAPCS caused by a missing attribute. Many projects
including the hard-float (VFP AAPCS) version of glibc contain assembler
files that do not use floating point but do not have Tag_ABI_VFP_args.
* Set the EF_ARM_ABI_FLOAT_SOFT or EF_ARM_ABI_FLOAT_HARD ELF header flag
for Base or VFP AAPCS respectively. This flag is used by some ELF
loaders.
References:
* Addenda to, and Errata in, the ABI for the ARM Architecture for
Tag_ABI_VFP_args
* Elf for the ARM Architecture for ELF header flags
Fixes LLVM PR36009
PR: 229050
Obtained from: llvm r338377 by Peter Smith
The LLD implementation of Tag_ABI_VFP_args needs to check the rarely
seen values of 3 (toolchain specific) and 4 compatible with both Base
and VFP. Add the missing enumeration values so that LLD can refer to
them without having to use the raw numbers.
Obtained from: llvm r338373 by Peter Smith
The ELF for the Arm architecture document defines, for EF_ARM_EABI_VER5
and above, the flags EF_ARM_ABI_FLOAT_HARD and EF_ARM_ABI_FLOAT_SOFT.
These have been defined to be compatible with the existing
EF_ARM_VFP_FLOAT and EF_ARM_SOFT_FLOAT used by gcc for
EF_ARM_EABI_UNKNOWN.
This patch adds the flags in addition to the existing ones so that any
code depending on the old names will still work.
Obtained from: llvm r338370 by Peter Smith
[ELF] Update addends in non-allocatable sections for REL targets when
creating a relocatable output.
LLVM PR: 37735
LLVM Differential Revision: https://reviews.llvm.org/D48929
PR: 225128
Obtained from: LLVM r336799 by Igor Kudrin
Request init/fini array on FreeBSD 12 and later
Summary:
It seems a bad idea to change the default in the middle of a release
branch due to possible changes in global ctor / dtor ordering between
.ctors and .init_array. With FreeBSD 11.0's release imminent lets
change the default now for FreeBSD 12 (the current development
stream) and later.
FreeBSD rtld has supported .init_array / .fini_array for many years.
As of Jan 1 2017 all supported FreeBSD releases and branches will
have support.
Reviewers: dim, brooks, arichardson
Reviewed By: dim, brooks, arichardson
Subscribers: bsdjhb, krytarowski, emaste, cfe-commits
Differential Revision: https://reviews.llvm.org/D24867
Requested by: jhb
MFC after: 3 days
o Restore floating-point registers in jumpto()
These are required to native cross build GCC and GDB
(both do require libc++ and libunwind).
These are not tested.
Sponsored by: DARPA, AFRL
PeepholeOpt cleanup/refactor; NFC
- Less unnecessary use of `auto`
- Add early `using RegSubRegPair(AndIdx) =` to avoid countless
`TargetInstrInfo::` qualifications.
- Use references instead of pointers where possible.
- Remove unused parameters.
- Rewrite the CopyRewriter class hierarchy:
- Pull out uncoalescable copy rewriting functionality into
PeepholeOptimizer class.
- Use an abstract base class to make it clear that rewriters are
independent.
- Remove unnecessary \brief in doxygen comments.
- Remove unused constructor and method from ValueTracker.
- Replace UseAdvancedTracking of ValueTracker with DisableAdvCopyOpt
use.
Even though upstream marked this as "No Functional Change", it does
contain some functional changes, and these fix a compiler hang for one
particular source file in the devel/godot port.
PR: 228261
MFC after: 3 days
A non-alloc note section should not have a PT_NOTE program header.
Found while linking ghc (Haskell compiler) with lld on FreeBSD. Haskell
emits a .debug-ghc-link-info note section (as the name suggests, it
contains link info) as a SHT_NOTE section without SHF_ALLOC set.
For this case ld.bfd does not emit a PT_NOTE segment for
.debug-ghc-link-info. lld previously emitted a PT_NOTE with p_vaddr = 0
and FreeBSD's rtld segfaulted when trying to parse a note at address 0.
LLVM PR: https://llvm.org/pr37361
LLVM review: https://reviews.llvm.org/D46623
PR: 226872
Reviewed by: dim
Sponsored by: The FreeBSD Foundation
The target is not necessarily a FreeBSD binary - for example, it may be
a Linux binary running under the linuxulator. Basic ptrace (live)
debugging already worked in this case, except for the assertion.
Sponsored by: Turing Robotic Industries Inc.
[X86] In X86FlagsCopyLowering, when rewriting a memory setcc we need
to emit an explicit MOV8mr instruction.
Previously the code only knew how to handle setcc to a register.
This should fix a crash in the chromium build.
This fixes various assertion failures while building ports targeting
i386:
* www/firefox: isReg() && "This is not a register operand!"
* www/iridium, www/qt5-webengine: (I.atEnd() || std::next(I) ==
def_instr_end()) && "getVRegDef assumes a single definition or no
definition"
* devel/powerpc64-gcc: FromReg != ToReg && "Cannot replace a reg with
itself"
Reported by: jbeich
PR: 225330, 227686, 227698, 227699
MFC after: 1 week
X-MFC-With: r332833
EFLAGS copy that lives out of a basic block!" errors on i386.
Pull in r325446 from upstream clang trunk (by me):
[X86] Add 'sahf' CPU feature to frontend
Summary:
Make clang accept `-msahf` (and `-mno-sahf`) flags to activate the
`+sahf` feature for the backend, for bug 36028 (Incorrect use of
pushf/popf enables/disables interrupts on amd64 kernels). This was
originally submitted in bug 36037 by Jonathan Looney
<jonlooney@gmail.com>.
As described there, GCC also uses `-msahf` for this feature, and the
backend already recognizes the `+sahf` feature. All that is needed is
to teach clang to pass this on to the backend.
The mapping of feature support onto CPUs may not be complete; rather,
it was chosen to match LLVM's idea of which CPUs support this feature
(see lib/Target/X86/X86.td).
I also updated the affected test case (CodeGen/attr-target-x86.c) to
match the emitted output.
Reviewers: craig.topper, coby, efriedma, rsmith
Reviewed By: craig.topper
Subscribers: emaste, cfe-commits
Differential Revision: https://reviews.llvm.org/D43394
Pull in r328944 from upstream llvm trunk (by Chandler Carruth):
[x86] Expose more of the condition conversion routines in the public
API for X86's instruction information. I've now got a second patch
under review that needs these same APIs. This bit is nicely
orthogonal and obvious, so landing it. NFC.
Pull in r329414 from upstream llvm trunk (by Craig Topper):
[X86] Merge itineraries for CLC, CMC, and STC.
These are very simple flag setting instructions that appear to only
be a single uop. They're unlikely to need this separation.
Pull in r329657 from upstream llvm trunk (by Chandler Carruth):
[x86] Introduce a pass to begin more systematically fixing PR36028
and similar issues.
The key idea is to lower COPY nodes populating EFLAGS by scanning the
uses of EFLAGS and introducing dedicated code to preserve the
necessary state in a GPR. In the vast majority of cases, these uses
are cmovCC and jCC instructions. For such cases, we can very easily
save and restore the necessary information by simply inserting a
setCC into a GPR where the original flags are live, and then testing
that GPR directly to feed the cmov or conditional branch.
However, things are a bit more tricky if arithmetic is using the
flags. This patch handles the vast majority of cases that seem to
come up in practice: adc, adcx, adox, rcl, and rcr; all without
taking advantage of partially preserved EFLAGS as LLVM doesn't
currently model that at all.
There are a large number of operations that techinaclly observe
EFLAGS currently but shouldn't in this case -- they typically are
using DF. Currently, they will not be handled by this approach.
However, I have never seen this issue come up in practice. It is
already pretty rare to have these patterns come up in practical code
with LLVM. I had to resort to writing MIR tests to cover most of the
logic in this pass already. I suspect even with its current amount
of coverage of arithmetic users of EFLAGS it will be a significant
improvement over the current use of pushf/popf. It will also produce
substantially faster code in most of the common patterns.
This patch also removes all of the old lowering for EFLAGS copies,
and the hack that forced us to use a frame pointer when EFLAGS copies
were found anywhere in a function so that the dynamic stack
adjustment wasn't a problem. None of this is needed as we now lower
all of these copies directly in MI and without require stack
adjustments.
Lots of thanks to Reid who came up with several aspects of this
approach, and Craig who helped me work out a couple of things
tripping me up while working on this.
Differential Revision: https://reviews.llvm.org/D45146
Pull in r329673 from upstream llvm trunk (by Chandler Carruth):
[x86] Model the direction flag (DF) separately from the rest of
EFLAGS.
This cleans up a number of operations that only claimed te use EFLAGS
due to using DF. But no instructions which we think of us setting
EFLAGS actually modify DF (other than things like popf) and so this
needlessly creates uses of EFLAGS that aren't really there.
In fact, DF is so restrictive it is pretty easy to model. Only STD,
CLD, and the whole-flags writes (WRFLAGS and POPF) need to model
this.
I've also somewhat cleaned up some of the flag management instruction
definitions to be in the correct .td file.
Adding this extra register also uncovered a failure to use the
correct datatype to hold X86 registers, and I've corrected that as
necessary here.
Differential Revision: https://reviews.llvm.org/D45154
Pull in r330264 from upstream llvm trunk (by Chandler Carruth):
[x86] Fix PR37100 by teaching the EFLAGS copy lowering to rewrite
uses across basic blocks in the limited cases where it is very
straight forward to do so.
This will also be useful for other places where we do some limited
EFLAGS propagation across CFG edges and need to handle copy rewrites
afterward. I think this is rapidly approaching the maximum we can and
should be doing here. Everything else begins to require either heroic
analysis to prove how to do PHI insertion manually, or somehow
managing arbitrary PHI-ing of EFLAGS with general PHI insertion.
Neither of these seem at all promising so if those cases come up,
we'll almost certainly need to rewrite the parts of LLVM that produce
those patterns.
We do now require dominator trees in order to reliably diagnose
patterns that would require PHI nodes. This is a bit unfortunate but
it seems better than the completely mysterious crash we would get
otherwise.
Differential Revision: https://reviews.llvm.org/D45673
Together, these should ensure clang does not use pushf/popf sequences to
save and restore flags, avoiding problems with unrelated flags (such as
the interrupt flag) being restored unexpectedly.
Requested by: jtl
PR: 225330
MFC after: 1 week
[X86] Add 'sahf' CPU feature to frontend
Summary:
Make clang accept `-msahf` (and `-mno-sahf`) flags to activate the
`+sahf` feature for the backend, for bug 36028 (Incorrect use of
pushf/popf enables/disables interrupts on amd64 kernels). This was
originally submitted in bug 36037 by Jonathan Looney
<jonlooney@gmail.com>.
As described there, GCC also uses `-msahf` for this feature, and the
backend already recognizes the `+sahf` feature. All that is needed is
to teach clang to pass this on to the backend.
The mapping of feature support onto CPUs may not be complete; rather,
it was chosen to match LLVM's idea of which CPUs support this feature
(see lib/Target/X86/X86.td).
I also updated the affected test case (CodeGen/attr-target-x86.c) to
match the emitted output.
Reviewers: craig.topper, coby, efriedma, rsmith
Reviewed By: craig.topper
Subscribers: emaste, cfe-commits
Differential Revision: https://reviews.llvm.org/D43394
Pull in r328944 from upstream llvm trunk (by Chandler Carruth):
[x86] Expose more of the condition conversion routines in the public
API for X86's instruction information. I've now got a second patch
under review that needs these same APIs. This bit is nicely
orthogonal and obvious, so landing it. NFC.
Pull in r329414 from upstream llvm trunk (by Craig Topper):
[X86] Merge itineraries for CLC, CMC, and STC.
These are very simple flag setting instructions that appear to only
be a single uop. They're unlikely to need this separation.
Pull in r329657 from upstream llvm trunk (by Chandler Carruth):
[x86] Introduce a pass to begin more systematically fixing PR36028
and similar issues.
The key idea is to lower COPY nodes populating EFLAGS by scanning the
uses of EFLAGS and introducing dedicated code to preserve the
necessary state in a GPR. In the vast majority of cases, these uses
are cmovCC and jCC instructions. For such cases, we can very easily
save and restore the necessary information by simply inserting a
setCC into a GPR where the original flags are live, and then testing
that GPR directly to feed the cmov or conditional branch.
However, things are a bit more tricky if arithmetic is using the
flags. This patch handles the vast majority of cases that seem to
come up in practice: adc, adcx, adox, rcl, and rcr; all without
taking advantage of partially preserved EFLAGS as LLVM doesn't
currently model that at all.
There are a large number of operations that techinaclly observe
EFLAGS currently but shouldn't in this case -- they typically are
using DF. Currently, they will not be handled by this approach.
However, I have never seen this issue come up in practice. It is
already pretty rare to have these patterns come up in practical code
with LLVM. I had to resort to writing MIR tests to cover most of the
logic in this pass already. I suspect even with its current amount
of coverage of arithmetic users of EFLAGS it will be a significant
improvement over the current use of pushf/popf. It will also produce
substantially faster code in most of the common patterns.
This patch also removes all of the old lowering for EFLAGS copies,
and the hack that forced us to use a frame pointer when EFLAGS copies
were found anywhere in a function so that the dynamic stack
adjustment wasn't a problem. None of this is needed as we now lower
all of these copies directly in MI and without require stack
adjustments.
Lots of thanks to Reid who came up with several aspects of this
approach, and Craig who helped me work out a couple of things
tripping me up while working on this.
Differential Revision: https://reviews.llvm.org/D45146
Pull in r329673 from upstream llvm trunk (by Chandler Carruth):
[x86] Model the direction flag (DF) separately from the rest of
EFLAGS.
This cleans up a number of operations that only claimed te use EFLAGS
due to using DF. But no instructions which we think of us setting
EFLAGS actually modify DF (other than things like popf) and so this
needlessly creates uses of EFLAGS that aren't really there.
In fact, DF is so restrictive it is pretty easy to model. Only STD,
CLD, and the whole-flags writes (WRFLAGS and POPF) need to model
this.
I've also somewhat cleaned up some of the flag management instruction
definitions to be in the correct .td file.
Adding this extra register also uncovered a failure to use the
correct datatype to hold X86 registers, and I've corrected that as
necessary here.
Differential Revision: https://reviews.llvm.org/D45154
Together, these should ensure clang does not use pushf/popf sequences to
save and restore flags, avoiding problems with unrelated flags (such as
the interrupt flag) being restored unexpectedly.
Requested by: jtl
PR: 225330
MFC after: 1 week
Strip @VER suffices from the LTO output.
This fixes pr36623.
The problem is that we have to parse versions out of names before LTO
so that LTO can use that information.
When we get the LTO produced .o files, we replace the previous symbols
with the LTO produced ones, but they still have @ in their names.
We could just trim the name directly, but calling parseSymbolVersion
to do it is simpler.
This is a follow-up to r331366, since we discovered that lld could
append version strings to symbols twice, when using Link Time
Optimization.
MFC after: 3 months
X-MFC-With: r327952
Don't treat .symver as a regular alias definition.
This patch starts simplifying the handling of .symver.
For now it just moves the responsibility for creating an alias down to
the streamer. With that the asm streamer can pass a .symver unchanged,
which is nice since gas cannot parse "foo@bar = zed".
In a followup I hope to move the handling down to the writer so that
we don't need special hacks for avoiding breaking names with @@@ on
windows.
Pull in r327160 from upstream llvm trunk (by Rafael Espindola):
Delay creating an alias for @@@.
With this we only create an alias for @@@ once we know if it should
use @ or @@. This avoids last minutes renames and hacks to handle MS
names.
This only handles the ELF writer. LTO still has issues with @@@
aliases.
Pull in r327928 from upstream llvm trunk (by Vitaly Buka):
Object: Move attribute calculation into RecordStreamer. NFC
Summary: Preparation for D44274
Reviewers: pcc, espindola
Subscribers: hiraditya
Differential Revision: https://reviews.llvm.org/D44276
Pull in r327930 from upstream llvm trunk (by Vitaly Buka):
Object: Fix handling of @@@ in .symver directive
Summary:
name@@@nodename is going to be replaced with name@@nodename if symbols is
defined in the assembled file, or name@nodename if undefined.
https://sourceware.org/binutils/docs/as/Symver.html
Fixes PR36623
Reviewers: pcc, espindola
Subscribers: mehdi_amini, hiraditya
Differential Revision: https://reviews.llvm.org/D44274
Together, these changes fix handling of @@@ in .symver directives when
doing Link Time Optimization.
Reported by: Shawn Webb <shawn.webb@hardenedbsd.org>
MFC after: 3 months
X-MFC-With: r327952
This is originally based on a patch from David Chisnall for soft-float
N64 but has since been updated to support O32, N32, and hard-float ABIs.
The soft-float O32, N32, and N64 support has been committed upstream.
The hard-float changes are still in review upstream.
Enable LLVM_LIBUNWIND on mips when building with a suitable (C+11-capable)
toolchain. This has been tested with external GCC for all ABIs and
O32 and N64 with clang.
Reviewed by: emaste
Obtained from: CheriBSD (original N64 patch)
Sponsored by: DARPA / AFRL
Differential Revision: https://reviews.freebsd.org/D14701
[CodeGen] Fix TBAA info for accesses to members of base classes
Resolves:
Bug 35724 - regression (r315984): fatal error: error in backend:
Broken function found (Did not see access type in access path!)
https://bugs.llvm.org/show_bug.cgi?id=35724
Differential Revision: https://reviews.llvm.org/D41547
This fixes "Did not see access type in access path" fatal errors when
building the devel/gdb port (version 8.1).
Reported by: jbeich
PR: 226658
MFC after: 3 months
X-MFC-With: r327952
[ConstantFolding, InstSimplify] Handle more vector GEPs
This patch addresses some additional cases where the compiler crashes
upon encountering vector GEPs. This should fix PR36116.
Differential Revision: https://reviews.llvm.org/D44219
Reference: https://bugs.llvm.org/show_bug.cgi?id=36116
This fixes an assertion when building the emulators/snes9x port.
Reported by: jbeich
PR: 225471
MFC after: 3 months
X-MFC-With: r327952
[ARM] Fix for PR36577
Don't PerformSHLSimplify if the given node is used by a node that
also uses a constant because we may get stuck in an infinite combine
loop.
bugzilla: https://bugs.llvm.org/show_bug.cgi?id=36577
Patch by Sam Parker.
Differential Revision: https://reviews.llvm.org/D44097
This fixes a hang when compiling one particular file in java/openjdk8
for armv6 and armv7.
Reported by: swills
PR: 226388
PR36157: When injecting an implicit function declaration in C89, find
the right DeclContext rather than injecting it wherever we happen to
be.
This avoids creating functions whose DeclContext is a struct or
similar.
This fixes assertion failures when parsing certain not-completely-valid
struct declarations.
Reported by: ae
PR: 225862
MFC after: 3 months
X-MFC-With: r327952
Fix for #31362 - ms_abi is implemented incorrectly for values >=16
bytes.
Summary:
This patch is a fix for following issue:
https://bugs.llvm.org/show_bug.cgi?id=31362 The problem was caused by
front end lowering C calling conventions without taking into account
calling conventions enforced by attribute. In this case win64cc was
no correctly lowered on targets other than Windows.
Reviewed By: rnk (Reid Kleckner)
Differential Revision: https://reviews.llvm.org/D43016
Author: belickim <mateusz.belicki@intel.com>
This fixes clang 6.0.0 assertions when building the emulators/wine and
emulators/wine-devel ports, and should also make it use the correct
Windows calling conventions. Bump __FreeBSD_version to make the fix
easy to detect.
PR: 224863
MFC after: 3 months
X-MFC-With: r327952
6.0.0 (branches/release_60 r324090).
This introduces retpoline support, with the -mretpoline flag. The
upstream initial commit message (r323155 by Chandler Carruth) contains
quite a bit of explanation. Quoting:
Introduce the "retpoline" x86 mitigation technique for variant #2 of
the speculative execution vulnerabilities disclosed today,
specifically identified by CVE-2017-5715, "Branch Target Injection",
and is one of the two halves to Spectre.
Summary:
First, we need to explain the core of the vulnerability. Note that
this is a very incomplete description, please see the Project Zero
blog post for details:
https://googleprojectzero.blogspot.com/2018/01/reading-privileged-memory-with-side.html
The basis for branch target injection is to direct speculative
execution of the processor to some "gadget" of executable code by
poisoning the prediction of indirect branches with the address of
that gadget. The gadget in turn contains an operation that provides a
side channel for reading data. Most commonly, this will look like a
load of secret data followed by a branch on the loaded value and then
a load of some predictable cache line. The attacker then uses timing
of the processors cache to determine which direction the branch took
*in the speculative execution*, and in turn what one bit of the
loaded value was. Due to the nature of these timing side channels and
the branch predictor on Intel processors, this allows an attacker to
leak data only accessible to a privileged domain (like the kernel)
back into an unprivileged domain.
The goal is simple: avoid generating code which contains an indirect
branch that could have its prediction poisoned by an attacker. In
many cases, the compiler can simply use directed conditional branches
and a small search tree. LLVM already has support for lowering
switches in this way and the first step of this patch is to disable
jump-table lowering of switches and introduce a pass to rewrite
explicit indirectbr sequences into a switch over integers.
However, there is no fully general alternative to indirect calls. We
introduce a new construct we call a "retpoline" to implement indirect
calls in a non-speculatable way. It can be thought of loosely as a
trampoline for indirect calls which uses the RET instruction on x86.
Further, we arrange for a specific call->ret sequence which ensures
the processor predicts the return to go to a controlled, known
location. The retpoline then "smashes" the return address pushed onto
the stack by the call with the desired target of the original
indirect call. The result is a predicted return to the next
instruction after a call (which can be used to trap speculative
execution within an infinite loop) and an actual indirect branch to
an arbitrary address.
On 64-bit x86 ABIs, this is especially easily done in the compiler by
using a guaranteed scratch register to pass the target into this
device. For 32-bit ABIs there isn't a guaranteed scratch register
and so several different retpoline variants are introduced to use a
scratch register if one is available in the calling convention and to
otherwise use direct stack push/pop sequences to pass the target
address.
This "retpoline" mitigation is fully described in the following blog
post: https://support.google.com/faqs/answer/7625886
We also support a target feature that disables emission of the
retpoline thunk by the compiler to allow for custom thunks if users
want them. These are particularly useful in environments like
kernels that routinely do hot-patching on boot and want to hot-patch
their thunk to different code sequences. They can write this custom
thunk and use `-mretpoline-external-thunk` *in addition* to
`-mretpoline`. In this case, on x86-64 thu thunk names must be:
```
__llvm_external_retpoline_r11
```
or on 32-bit:
```
__llvm_external_retpoline_eax
__llvm_external_retpoline_ecx
__llvm_external_retpoline_edx
__llvm_external_retpoline_push
```
And the target of the retpoline is passed in the named register, or in
the case of the `push` suffix on the top of the stack via a `pushl`
instruction.
There is one other important source of indirect branches in x86 ELF
binaries: the PLT. These patches also include support for LLD to
generate PLT entries that perform a retpoline-style indirection.
The only other indirect branches remaining that we are aware of are
from precompiled runtimes (such as crt0.o and similar). The ones we
have found are not really attackable, and so we have not focused on
them here, but eventually these runtimes should also be replicated for
retpoline-ed configurations for completeness.
For kernels or other freestanding or fully static executables, the
compiler switch `-mretpoline` is sufficient to fully mitigate this
particular attack. For dynamic executables, you must compile *all*
libraries with `-mretpoline` and additionally link the dynamic
executable and all shared libraries with LLD and pass `-z
retpolineplt` (or use similar functionality from some other linker).
We strongly recommend also using `-z now` as non-lazy binding allows
the retpoline-mitigated PLT to be substantially smaller.
When manually apply similar transformations to `-mretpoline` to the
Linux kernel we observed very small performance hits to applications
running typic al workloads, and relatively minor hits (approximately
2%) even for extremely syscall-heavy applications. This is largely
due to the small number of indirect branches that occur in
performance sensitive paths of the kernel.
When using these patches on statically linked applications,
especially C++ applications, you should expect to see a much more
dramatic performance hit. For microbenchmarks that are switch,
indirect-, or virtual-call heavy we have seen overheads ranging from
10% to 50%.
However, real-world workloads exhibit substantially lower performance
impact. Notably, techniques such as PGO and ThinLTO dramatically
reduce the impact of hot indirect calls (by speculatively promoting
them to direct calls) and allow optimized search trees to be used to
lower switches. If you need to deploy these techniques in C++
applications, we *strongly* recommend that you ensure all hot call
targets are statically linked (avoiding PLT indirection) and use both
PGO and ThinLTO. Well tuned servers using all of these techniques saw
5% - 10% overhead from the use of retpoline.
We will add detailed documentation covering these components in
subsequent patches, but wanted to make the core functionality
available as soon as possible. Happy for more code review, but we'd
really like to get these patches landed and backported ASAP for
obvious reasons. We're planning to backport this to both 6.0 and 5.0
release streams and get a 5.0 release with just this cherry picked
ASAP for distros and vendors.
This patch is the work of a number of people over the past month:
Eric, Reid, Rui, and myself. I'm mailing it out as a single commit
due to the time sensitive nature of landing this and the need to
backport it. Huge thanks to everyone who helped out here, and
everyone at Intel who helped out in discussions about how to craft
this. Also, credit goes to Paul Turner (at Google, but not an LLVM
contributor) for much of the underlying retpoline design.
Reviewers: echristo, rnk, ruiu, craig.topper, DavidKreitzer
Subscribers: sanjoy, emaste, mcrosier, mgorny, mehdi_amini, hiraditya, llvm-commits
Differential Revision: https://reviews.llvm.org/D41723
MFC after: 3 months
X-MFC-With: r327952
PR: 224669