[CodeGenPrepare] Cut off FindAllMemoryUses if there are too many uses.
This avoids excessive compile time. The case I'm looking at is
Function.cpp from an old version of LLVM that still had the giant
memcmp string matcher in it. Before r308322 this compiled in about 2
minutes, after it, clang takes infinite* time to compile it. With
this patch we're at 5 min, which is still bad but this is a
pathological case.
The cut off at 20 uses was chosen by looking at other cut-offs in LLVM
for user scanning. It's probably too high, but does the job and is
very unlikely to regress anything.
Fixes PR33900.
* I'm impatient and aborted after 15 minutes, on the bug report it was
killed after 2h.
Pull in r308986 from upstream llvm trunk (by Simon Pilgrim):
[X86][CGP] Reduce memcmp() expansion to 2 load pairs (PR33914)
D35067/rL308322 attempted to support up to 4 load pairs for memcmp
inlining which resulted in regressions for some optimized libc memcmp
implementations (PR33914).
Until we can match these more optimal cases, this patch reduces the
memcmp expansion to a maximum of 2 load pairs (which matches what we
do for -Os).
This patch should be considered for the 5.0.0 release branch as well
Differential Revision: https://reviews.llvm.org/D35830
These fix a hang (or extremely long compile time) when building older
LLVM ports.
Reported by: antoine
PR: 219139
Given an empty pattern (i.e. grep "" A B), bsdgrep(1) would previously exit()
with the appropriate exit code upon encountering an empty file. Likely intended
as an optimization, but this behavior is technically incorrect since an empty
pattern should match every line.
PR: 220924
Reviewed by: emaste, cem (earlier version), ngie
Approved by: emaste (mentor)
Differential Revision: https://reviews.freebsd.org/D11698
PR32034: Evaluate _Atomic(T) in-place when T is a class or array type.
This is necessary in order for the evaluation of an _Atomic
initializer for those types to have an associated object, which an
initializer for class or array type needs.
This fixes an assertion when building recent versions of LinuxCNC.
Reported by: trasz
PR: 220883
MFC after: 1 week
BREs recently became prematurely sensitive to the branching operator, which
outright broke expressions that used it instead of failing silently. Test
that \| is matching a literal | for the time being.
Reviewed by: cem, emaste, ngie
Approved by: emaste (mentor)
Differential Revision: https://reviews.freebsd.org/D11577
Previously these were shown only for MIPS objects.
Obtained from: ELF Tool Chain r3564
MFC after: 1 week
MFC with: r321045
Sponsored by: The FreeBSD Foundation
Some non-processor-specific DT_* values overlap the range DT_LOPROC to
DT_HIPROC. Handle common ones first, then the processor-specific ones.
Obtained from: ELF Tool Chain r3563
Sponsored by: The FreeBSD Foundation
Create libdl.so.1 as a filter for libc.so.7 which exports public dl*
functions. The functions are resolved from the rtld instead, the goal
of creating library is to avoid errors from the static linker due to
missed libdl. For static binaries, an empty .o is compiled into
libdl.a so that static binaries still get dl stubs from libc.a.
Right now lld cannot create filter objects, disable libdl on arm64
when binutils are not used.
Reviewed by: bdrewery, dim (previos version); emaste
Exp run: PR 220525, done by antoine
Sponsored by: The FreeBSD Foundation
MFC after: 1 month
Differential revision: https://reviews.freebsd.org/D11504
ATF cleanup routines run in separate processes from the tests themselves, so
they can't share global variables.
Also, setdomainname_test needs to be is_exclusive because the test cases
access a global resource.
PR: 219967
Reviewed by: ngie
MFC after: 3 weeks
Sponsored by: Spectra Logic Corp
Differential Revision: https://reviews.freebsd.org/D11188
the limit to 32MB instead.
Require user=root and memory=64MB+ first so one can be reasonably sure that
the test will function appropriately.
MFC after: 1 month
MFC with: r320726
PR: 220502
Script commands are processed before unused synthetic sections are
removed. Therefore, if a linker script matches one of these sections
it'll get emitted as an empty output section because the logic for
removing unused synthetic sections ignores script commands which
could have already matched and captured one of these sections. This
patch fixes that by also removing the unused synthetic sections from
the script commands.
Discussed with: dim
Obtained from: LLVM r307037
Sponsored by: The FreeBSD Foundation
This LLD commit was intended to fix one case of symbol versioning
(LLVM PR28414) but broke FreeBSD buildworld. It has been reverted
upstream in LLVM r306996.
Discussed with: dim
The summary of changes is as follows..
Generic changes::
- Added configure support [2].
- Check for lchmod filesystem support with create_file(..); for
testcases that require lchmod, skip the testcase -- otherwise
use chmod directly [1].
- Added Travis CI integration [2].
- Added utimensat testcases [1].
Linux support::
- Fixed Linux support to pass on later supported versions of
Fedora/Ubuntu [2].
- Conditionally enable posix_fallocate(2) support [2].
OSX support::
- Fixed compilation on OSX [2].
- Added partial OSX support (the test run isn't fully green yet)
[2].
MFC after: 2 months
Obtained from: https://github.com/pjd/pjdfstest/tree/0.1
Relnotes: yes
Submitted by: asomers [1], ngie [2]
Tested with: UFS, ZFS
This update is primarily bug fixes in C++ symbol demangling, including:
- rvalue reference
- builtin type auto and decltype(auto)
- revamped support for function return types
- formatting fixes
- omit void when its the only param
- ref-qualifiers and others in function types
- type qualifiers in pointer-to-member function types
- incorrect handling regarding CV-qualifiers in function types
- ref-qualifier found in nested-name
- properly handle <name> ::= <substitute><template-args>
- make sure that nested function name is not a substitute candidate
- correctly handle expression in template args
- skip unknown substitution abbreviations
MFC after: 4 days
valid when adding a pool (ippool -A), not when removing a pool
(ippool -R). It is a command line syntax error if specifying a seed (-S)
is specified when emoving a pool (-R).
The declaration was already inactive when INET6 was enabled
and it causes a build error in the other case because of
a conflict with the correct definition in stdlib.h.
Discussed with: dim, ume
MFC after: 2 weeks
