Summary:
When creating `ScopeTops` info for `try` ~ `catch` ~ `end_try`, we
should create not only `end_try` -> `try` mapping but also `catch` ->
`try` mapping as well. If this is not created, `block` and `end_block`
markers later added may span across an existing `catch`, resulting in
the incorrect code like:
```
try
block --| (X)
catch |
end_block --|
end_try
```
Reviewers: dschuff
Subscribers: sunfish, sbc100, jgravelle-google, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D58605
llvm-svn: 354945
Currently the symbol for MSPropertyDecl has kind `SymbolKind::Unknown`
which can trip up various indexing tools.
rdar://problem/46764224
Reviewers: akyrtzi, benlangmuir, jkorous
Reviewed By: jkorous
Subscribers: dexonsmith, cfe-commits, jkorous, jdoerfert, arphaman
Differential Revision: https://reviews.llvm.org/D57628
llvm-svn: 354942
DWARFFormValues can be created from a data extractor or by passing its
value directly. Until now this was done by member functions that
modified an existing object's internal state. This patch replaces a
subset of these methods with static method that return a new
DWARFFormValue.
llvm-svn: 354941
Those tests fail when linking against a new dylib but running against
macosx10.7. I believe this is caused by a duplicate definition of the
RTTI for exception classes in libc++.dylib and libc++abi.dylib, but
this matter still needs some investigation.
This issue was not caught previously because all the tests always linked
against the same dylib used for running (because LIT made it impossible
to do otherwise before r349171).
rdar://problem/46809586
llvm-svn: 354940
Summary:
This removes unnecessary instructions after TRY marker placement. There
are two cases:
- `end`/`end_block` can be removed if they overlap with `try`/`end_try`
and they have the same return types.
- `br` right before `catch` that branches to after `end_try` can be
deleted.
Reviewers: dschuff
Subscribers: sbc100, jgravelle-google, sunfish, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D58591
llvm-svn: 354939
Summary:
This behavior was originally added in rL252264 (git commit 76a7f365da)
in order to be extra careful with handling platforms like watchos and tvos.
However, as far as triples go, those two (and others) are treated as OSes and
not environments, so that should not really apply here.
Additionally, this behavior is incorrect and can lead to incorrect ArchSpecs.
Because android is specified as an environment and not an OS, not propogating
the environment can lead to modules and targets being misidentified.
Differential Revision: https://reviews.llvm.org/D58664
llvm-svn: 354938
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
llvm-svn: 354937
Since there is no "Load-and-Test-High" instruction, the 32 bit load of a
register to be compared with 0 can only be implemented with LT if the virtual
GRX32 register ends up in a low part (GR32 register).
This patch detects these cases and passes the GR32 registers (low parts) as
(soft) hints in getRegAllocationHints().
Review: Ulrich Weigand.
llvm-svn: 354935
Summary:
These functions should always return the opposite of the
`Triple{Environment,OS,Vendor}WasSpecified` functions. Unspecified unknown is
the same as unspecified, which is why one set of functions should give us what
we want. It's possible to have specified unknown, which is why we can't just
rely on checking the enum values of vendor/os/environment. We must also ensure
that the names of these are empty and not "unknown".
Differential Revision: https://reviews.llvm.org/D58653
llvm-svn: 354933
Splitting can make sanitizer errors harder to understand, as the
trapping instruction may not be in the function where the bug was
detected.
rdar://48142697
llvm-svn: 354931
Current PGO profile counts are not context sensitive. The branch probabilities
for the inlined functions are kept the same for all call-sites, and they might
be very different from the actual branch probabilities. These suboptimal
profiles can greatly affect some downstream optimizations, in particular for
the machine basic block placement optimization.
In this patch, we propose to have a post-inline PGO instrumentation/use pass,
which we called Context Sensitive PGO (CSPGO). For the users who want the best
possible performance, they can perform a second round of PGO instrument/use on
the top of the regular PGO. They will have two sets of profile counts. The
first pass profile will be manly for inline, indirect-call promotion, and
CGSCC simplification pass optimizations. The second pass profile is for
post-inline optimizations and code-gen optimizations.
A typical usage:
// Regular PGO instrumentation and generate pass1 profile.
> clang -O2 -fprofile-generate source.c -o gen
> ./gen
> llvm-profdata merge default.*profraw -o pass1.profdata
// CSPGO instrumentation.
> clang -O2 -fprofile-use=pass1.profdata -fcs-profile-generate -o gen2
> ./gen2
// Merge two sets of profiles
> llvm-profdata merge default.*profraw pass1.profdata -o profile.profdata
// Use the combined profile. Pass manager will invoke two PGO use passes.
> clang -O2 -fprofile-use=profile.profdata -o use
This change touches many components in the compiler. The reviewed patch
(D54175) will committed in phrases.
Differential Revision: https://reviews.llvm.org/D54175
llvm-svn: 354930
MSVC header files using vectorcall to differentiate overloaded functions, which
causes failure for AMDGPU target. This is because clang does not check function
calling convention based on function target.
This patch checks calling convention using the proper target info.
Differential Revision: https://reviews.llvm.org/D57716
llvm-svn: 354929
statements.
If the assembler instruction is not generated and the delayed diagnostic
is emitted, we may end up with extra warning message for variables used
in the asm statement. Since the asm statement is not built, the
variables may be left non-referenced and it may produce a warning about
a use of the non-initialized variables.
llvm-svn: 354928
For 'cascadelake' this is adding a 'avx512vnni' feature check to the 0x55 skylake-avx512 model check. These CPUs use the same model number and only differ in the stepping number. But the feature flag is simpler than collecting all the stepping numbers.
For 'znver2' this is just syncing with LLVM's Host.cpp.
llvm-svn: 354927
SITargetLowering::reassociateScalarOps() does not touch constants
so that DAGCombiner::ReassociateOps() does not revert the combine.
However a global address is not a ConstantSDNode.
Switched to the method used by DAGCombiner::ReassociateOps() itself
to detect constants.
Differential Revision: https://reviews.llvm.org/D58695
llvm-svn: 354926
Summary:
The MS C++ ABI has no constructor variants, but it has destructor
variants, so we should move the deleting destructor variant check
outside the check for "does the ABI have constructor variants".
Fixes PR37561, so basic code coverage works on Windows with C++.
Reviewers: vsk
Subscribers: jdoerfert, cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D58691
llvm-svn: 354924
Pass dummy '.' as format string for Timer() rather than an empty string,
in order to silence gcc warnings about empty format string
(-Wformat-zero-length). The actual format string is irrelevant
to the test in question.
Differential Revision: https://reviews.llvm.org/D58680
llvm-svn: 354922
Original implementation can't correctly handle __m256 and __m512 types
passed by reference through stack. This patch fixes it.
Patch by Wei Xiao!
Differential Revision: https://reviews.llvm.org/D57643
llvm-svn: 354921
Set LD_LIBRARY_PATH or local platform's equivalent of it when running
the 'Suite' tests. This is necessary when running tests inside build
tree with BUILD_SHARED_LIBS enabled, in order to make the LLDB modules
load freshly built LLVM libraries.
The code is copied from clang (test/Unit/lit.cfg). SHLIBDIR
substitution is added to site-config (already present in top-level LLDB
site-config) to future-proof this into supporting stand-alone builds
with shared LLDB libraries.
Differential Revision: https://reviews.llvm.org/D58610
llvm-svn: 354920
Summary:
When running the standalone main on a large corpus, I eventually get a
EMFILE error ("Too many open files").
Patch by Paul Chaignon
Reviewers: kcc, vitalybuka
Reviewed By: vitalybuka
Subscribers: lebedev.ri, #sanitizers, llvm-commits
Tags: #llvm, #sanitizers
Differential Revision: https://reviews.llvm.org/D58622
llvm-svn: 354918
Summary:
The gold linker allowed you to output the ELF files after LTO was run. It did
it by using the 'obj-path' option. This replicates that behavior.
Reviewers: espindola, ruiu, MaskRay, pcc
Reviewed By: MaskRay, pcc
Subscribers: grimar, emaste, inglorion, arichardson, steven_wu, dexonsmith, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D56046
llvm-svn: 354917
Summary:
This allows ASTs to be merged when they contain ChooseExpr (the GNU
__builtin_choose_expr construction). This is needed, for example, for
cross-CTU analysis of C code that makes use of __builtin_choose_expr.
The node is already supported in the AST, but it didn't have a matcher
in ASTMatchers. So, this change adds the matcher and adds support to
ASTImporter.
This was originally reviewed and approved in
https://reviews.llvm.org/D58292 and submitted as r354832. It was
reverted in r354839 due to failures on the Windows CI builds.
This version fixes the test failures on Windows, which were caused by
differences in template expansion between versions of clang on different
OSes. The version of clang built with MSVC and running on Windows never
expands the template in the C++ test in ImportExpr.ImportChooseExpr in
clang/unittests/AST/ASTImporter.cpp, but the version on Linux does for
the empty arguments and -fms-compatibility.
So, this version of the patch drops the C++ test for
__builtin_choose_expr, since that version was written to catch
regressions of the logic for isConditionTrue() in the AST import code
for ChooseExpr, and those regressions are also caught by
ASTImporterOptionSpecificTestBase.ImportChooseExpr, which does work on
Windows.
Reviewers: shafik, a_sidorin, martong, aaron.ballman, rnk, a.sidorin
Subscribers: cfe-commits, jdoerfert, rnkovacs, aaron.ballman
Tags: #clang
Differential Revision: https://reviews.llvm.org/D58663
llvm-svn: 354916
Check that we do not crash if a parallelism group is explicitly set to
None. Permits usage of the following pattern.
[lit.common.cfg]
lit_config.parallelism_groups['my_group'] = None
if <condition>:
lit_config.parallelism_groups['my_group'] = 3
[project/lit.cfg]
config.parallelism_group = 'my_group'
Reviewers: rnk
Differential Revision: https://reviews.llvm.org/D58305
llvm-svn: 354912
Fix https://bugs.llvm.org/show_bug.cgi?id=38583: Describe
how memcpy/memmove/memset behave when len=0. Also fix
some fallout from when the alignment parameter was
replaced by an attribute.
This closes PR38583.
Patch by RalfJung (Ralf)
Differential Revision: https://reviews.llvm.org/D57600
llvm-svn: 354911
The previous sort comparator was not deterministic, i.e. in some
situations it would be possible for lhs < rhs && rhs < lhs. This was
discovered by an STL assertion in a Windows debug build of llvm-tblgen.
Differential Revision: https://reviews.llvm.org/D58687
llvm-svn: 354910
Summary:
The current error message can cause confusion if llvm-config can't be
executed for reasons other than "not found". In my example, cross
compiling generated an llvm-config binary for aarch64 which couldn't be
executed natively. Instead of telling me that the error was with the
file architecture, it reports the file as not being present.
Reviewers: pcc
Subscribers: dberris, javed.absar, kristof.beyls, jdoerfert, #sanitizers, llvm-commits
Tags: #sanitizers, #llvm
Differential Revision: https://reviews.llvm.org/D58647
llvm-svn: 354907
As discussed on llvm-dev:
http://lists.llvm.org/pipermail/llvm-dev/2019-February/130491.html
We can't remove the compare+select in the general case because
we are treating funnel shift like a standard instruction (as
opposed to a special instruction like select/phi).
That means that if one of the operands of the funnel shift is
poison, the result is poison regardless of whether we know that
the operand is actually unused based on the instruction's
particular semantics.
The motivating case for this transform is the more specific
rotate op (rather than funnel shift), and we are preserving the
fold for that case because there is no chance of introducing
extra poison when there is no anonymous extra operand to the
funnel shift.
llvm-svn: 354905
This patch enables the following
1) AMD family 17h "znver2" tune flag (-march, -mcpu).
2) ISAs that are enabled for "znver2" architecture.
3) For the time being, it uses the znver1 scheduler model.
4) Tests are updated.
5) This patch is the clang counterpart to D58343
Reviewers: craig.topper
Tags: #clang
Differential Revision: https://reviews.llvm.org/D58344
llvm-svn: 354899
Julian Lettner