Jinsong Ji [Tue, 19 Feb 2019 21:25:13 +0000 (21:25 +0000)]
PowerPC: Fix typos in comments
llvm-svn: 354382
Andrew Scheidecker [Tue, 19 Feb 2019 21:21:54 +0000 (21:21 +0000)]
[ConstantFold] Fix misfolding fcmp of a ConstantExpr NaN with itself.
The code incorrectly inferred that the relationship of a constant expression
to itself is FCMP_OEQ (ordered and equal), when it's actually FCMP_UEQ
(unordered *or* equal). This change corrects that, and adds some more limited
folds that can be done in this case.
Differential revision: https://reviews.llvm.org/D51216
llvm-svn: 354381
Andrew Scheidecker [Tue, 19 Feb 2019 21:03:20 +0000 (21:03 +0000)]
[ConstantFold] Fix misfolding of icmp with a bitcast FP second operand.
In the process of trying to eliminate the bitcast, this was producing a
malformed icmp with FP operands.
Differential revision: https://reviews.llvm.org/D51215
llvm-svn: 354380
Vedant Kumar [Tue, 19 Feb 2019 20:45:00 +0000 (20:45 +0000)]
[llvm-cov] Add support for gcov --hash-filenames option
The patch adds support for --hash-filenames to llvm-cov. This option adds md5
hash of the source path to the name of the generated .gcov file. The option is
crucial for cases where you have multiple files with the same name but can't
use --preserve-paths as resulting filenames exceed the limit.
from gcov(1):
```
-x
--hash-filenames
By default, gcov uses the full pathname of the source files to to
create an output filename. This can lead to long filenames that
can overflow filesystem limits. This option creates names of the
form source-file##md5.gcov, where the source-file component is
the final filename part and the md5 component is calculated from
the full mangled name that would have been used otherwise.
```
Patch by Igor Ignatev!
Differential Revision: https://reviews.llvm.org/D58370
llvm-svn: 354379
Andrew Scheidecker [Tue, 19 Feb 2019 20:38:51 +0000 (20:38 +0000)]
Testing commit access
llvm-svn: 354378
Vitaly Buka [Tue, 19 Feb 2019 20:36:52 +0000 (20:36 +0000)]
[msan] Remove cxa_atexit_race.cc
Summary:
The goal of the test to check that msan does not crash when code is racy on __cxa_atexit. Original crash was caused by race condition in the glibc. With
the msan patch the msan does not crashes however the race is still there and the test triggers it.
Because the test relies on triggering of undefined behavior results are not
very predictable and it may occasionally crashes or hangs.
I don't see how to reasonably improve the test, so I remove it.
Reviewers: eugenis, peter.smith
Subscribers: jfb, jdoerfert, #sanitizers, llvm-commits
Tags: #sanitizers, #llvm
Differential Revision: https://reviews.llvm.org/D58396
llvm-svn: 354377
Craig Topper [Tue, 19 Feb 2019 20:12:20 +0000 (20:12 +0000)]
[X86] Don't consider functions ABI compatible for ArgumentPromotion pass if they view 512-bit vectors differently.
The use of the -mprefer-vector-width=256 command line option mixed with functions
using vector intrinsics can create situations where one function thinks 512 vectors
are legal, but another fucntion does not.
If a 512 bit vector is passed between them via a pointer, its possible ArgumentPromotion
might try to pass by value instead. This will result in type legalization for the two
functions handling the 512 bit vector differently leading to runtime failures.
Had the 512 bit vector been passed by value from clang codegen, both functions would
have been tagged with a min-legal-vector-width=512 function attribute. That would
make them be legalized the same way.
I observed this issue in 32-bit mode where a union containing a 512 bit vector was
being passed by a function that used intrinsics to one that did not. The caller
ended up passing in zmm0 and the callee tried to read it from ymm0 and ymm1.
The fix implemented here is just to consider it a mismatch if two functions
would handle 512 bit differently without looking at the types that are being
considered. This is the easist and safest fix, but it can be improved in the future.
Differential Revision: https://reviews.llvm.org/D58390
llvm-svn: 354376
Matthew Voss [Tue, 19 Feb 2019 19:46:08 +0000 (19:46 +0000)]
Revert "Revert "[llvm-objdump] Allow short options without arguments to be grouped""
- Tests that use multiple short switches now test them grouped and ungrouped.
- Ensure the output of ungrouped and grouped variants is identical
Differential Revision: https://reviews.llvm.org/D57904
llvm-svn: 354375
Daniel Sanders [Tue, 19 Feb 2019 19:45:03 +0000 (19:45 +0000)]
Fix builds for older macOS deployment targets after r354365
Surprisingly, check_symbol_exists is not sufficient. The macOS linker checks the
called functions against a compatibility list for the given deployment target
and check_symbol_exists doesn't trigger this check as it never calls the
function.
This fixes the GreenDragon bots where the deployment target is 10.9
llvm-svn: 354374
Kostya Serebryany [Tue, 19 Feb 2019 19:28:08 +0000 (19:28 +0000)]
[sanitizers] add a regression test for the bug fixed in r354366
llvm-svn: 354373
Jonathan Peyton [Tue, 19 Feb 2019 19:00:29 +0000 (19:00 +0000)]
[OpenMP] Remove XFAIL for cancellation tests using gcc
llvm-svn: 354370
Jonathan Peyton [Tue, 19 Feb 2019 18:51:11 +0000 (18:51 +0000)]
[OpenMP 5.0] Add omp_get_supported_active_levels()
This patch adds the new 5.0 API function omp_get_supported_active_levels().
Patch by Terry Wilmarth
Differential Revision: https://reviews.llvm.org/D58211
llvm-svn: 354368
Jonathan Peyton [Tue, 19 Feb 2019 18:47:57 +0000 (18:47 +0000)]
[OpenMP] Adding GOMP compatible cancellation
Remove fatal error messages from the cancellation API for GOMP
Add __kmp_barrier_gomp_cancel() to implement cancellation of parallel regions.
This new function uses the linear barrier algorithm with a cancellable
nonsleepable wait loop.
Differential Revision: https://reviews.llvm.org/D57969
llvm-svn: 354367
Kostya Serebryany [Tue, 19 Feb 2019 18:43:24 +0000 (18:43 +0000)]
[sanitizer] fix a memory safety bug (!!!) in sanitizer suppressions code, discovered by Aaron Jacobs
llvm-svn: 354366
Daniel Sanders [Tue, 19 Feb 2019 18:18:31 +0000 (18:18 +0000)]
Annotate timeline in Instruments with passes and other timed regions.
Summary:
Instruments is a useful tool for finding performance issues in LLVM but it can
be difficult to identify regions of interest on the timeline that we can use
to filter the profiler or allocations instrument. Xcode 10 and the latest
macOS/iOS/etc. added support for the os_signpost() API which allows us to
annotate the timeline with information that's meaningful to LLVM.
This patch causes timer start and end events to emit signposts. When used with
-time-passes, this causes the passes to be annotated on the Instruments timeline.
In addition to visually showing the duration of passes on the timeline, it also
allows us to filter the profile and allocations instrument down to an individual
pass allowing us to find the issues within that pass without being drowned out
by the noise from other parts of the compiler.
Using this in conjunction with the Time Profiler (in high frequency mode) and
the Allocations instrument is how I found the SparseBitVector that should have
been a BitVector and the DenseMap that could be replaced by a sorted vector a
couple months ago. I added NamedRegionTimers to TableGen and used the resulting
annotations to identify the slow portions of the Register Info Emitter. Some of
these were placed according to educated guesses while others were placed
according to hot functions from a previous profile. From there I filtered the
profile to a slow portion and the aforementioned issues stood out in the
profile.
To use this feature enable LLVM_SUPPORT_XCODE_SIGNPOSTS in CMake and run the
compiler under Instruments with -time-passes like so:
instruments -t 'Time Profiler' bin/llc -time-passes -o - input.ll'
Then open the resulting trace in Instruments.
There was a talk at WWDC 2018 that explained the feature which can be found at
https://developer.apple.com/videos/play/wwdc2018/405/ if you'd like to know
more about it.
Reviewers: bogner
Reviewed By: bogner
Subscribers: jdoerfert, mgorny, kristina, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D52954
llvm-svn: 354365
Jordan Rupprecht [Tue, 19 Feb 2019 18:14:44 +0000 (18:14 +0000)]
[libObject][NFC] Use sys::path::convert_to_slash.
Summary: As suggested in rL353995
Reviewers: compnerd
Reviewed By: compnerd
Subscribers: hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D58298
llvm-svn: 354364
Simon Pilgrim [Tue, 19 Feb 2019 18:05:42 +0000 (18:05 +0000)]
[X86][SSE] Generalize X86ISD::BLENDI support to more value types
D42042 introduced the ability for the ExecutionDomainFixPass to more easily change between BLENDPD/BLENDPS/PBLENDW as the domains required.
With this ability, we can avoid most bitcasts/scaling in the DAG that was occurring with X86ISD::BLENDI lowering/combining, blend with the vXi32/vXi64 vectors directly and use isel patterns to lower to the float vector equivalent vectors.
This helps the shuffle combining and SimplifyDemandedVectorElts be more aggressive as we lose track of fewer UNDEF elements than when we go up/down through bitcasts.
I've introduced a basic blend(bitcast(x),bitcast(y)) -> bitcast(blend(x,y)) fold, there are more generalizations I can do there (e.g. widening/scaling and handling the tricky v16i16 repeated mask case).
The vector-reduce-smin/smax regressions will be fixed in a future improvement to SimplifyDemandedBits to peek through bitcasts and support X86ISD::BLENDV.
Reapplied after reversion at rL353699 - AVX2 isel fix was applied at rL354358, additional test at rL354360/rL354361
Differential Revision: https://reviews.llvm.org/D57888
llvm-svn: 354363
Serge Guelton [Tue, 19 Feb 2019 18:03:47 +0000 (18:03 +0000)]
[NFC] Remove unused headers in Optional.h
llvm-svn: 354362
Simon Pilgrim [Tue, 19 Feb 2019 17:57:36 +0000 (17:57 +0000)]
Fix stupid assembly comment typo
llvm-svn: 354361
Simon Pilgrim [Tue, 19 Feb 2019 17:56:14 +0000 (17:56 +0000)]
[X86][SSE] Add pblendw commuted load test case
Reduced test case for the regression caused in D57888/rL353610
llvm-svn: 354360
Nikita Popov [Tue, 19 Feb 2019 17:37:55 +0000 (17:37 +0000)]
[SDAG] Use shift amount type in MULO promotion; NFC
Directly use the correct shift amount type if it is possible, and
future-proof the code against vectors. The added test makes sure that
bitwidths that do not fit into the shift amount type do not assert.
Split out from D57997.
llvm-svn: 354359
Simon Pilgrim [Tue, 19 Feb 2019 17:23:55 +0000 (17:23 +0000)]
[X86][AVX2] Hide VPBLENDD instructions behind AVX2 predicate
This was the cause of the regression in D57888 - the commuted load pattern wasn't hidden by the predicate so once we enabled v4i32 blends on SSE41+ targets then isel was incorrectly matched against AVX2+ instructions.
llvm-svn: 354358
Craig Topper [Tue, 19 Feb 2019 17:16:23 +0000 (17:16 +0000)]
[X86] Bugfix for nullptr check by klocwork
klocwork critical issues in CG files:
Patch by Xiang Zhang (xiangzhangllvm)
Differential Revision: https://reviews.llvm.org/D58363
llvm-svn: 354357
Craig Topper [Tue, 19 Feb 2019 17:13:40 +0000 (17:13 +0000)]
X86AsmParser AVX-512: Return error instead of hitting assert
When parsing a sequence of tokens beginning with {, it will hit an assert and crash if the token afterwards is not an identifier. Instead of this, return a more verbose error as seen elsewhere in the function.
Patch by Brandon Jones (BrandonTJones)
Differential Revision: https://reviews.llvm.org/D57375
llvm-svn: 354356
Craig Topper [Tue, 19 Feb 2019 17:05:11 +0000 (17:05 +0000)]
[X86] Filter out tuning feature flags and a few ISA feature flags when checking for function inline compatibility.
Tuning flags don't have any effect on the available instructions so aren't a good reason to prevent inlining.
There are also some ISA flags that don't have any intrinsics our ABI requirements that we can exclude. I've put only the most basic ones like cmpxchg16b and lahfsahf. These are interesting because they aren't present in all 64-bit CPUs, but we have codegen workarounds when they aren't present.
Loosening these checks can help with scenarios where a caller has a more specific CPU than a callee. The default tuning flags on our generic 'x86-64' CPU can currently make it inline compatible with other CPUs. I've also added an example test for 'nocona' and 'prescott' where 'nocona' is just a 64-bit capable version of 'prescott' but in 32-bit mode they should be completely compatible.
I've based the implementation here of the similar code in AMDGPU.
Differential Revision: https://reviews.llvm.org/D58371
llvm-svn: 354355
Matt Arsenault [Tue, 19 Feb 2019 17:03:09 +0000 (17:03 +0000)]
GlobalISel: Implement moreElementsVector for select
llvm-svn: 354354
Hans Wennborg [Tue, 19 Feb 2019 17:00:34 +0000 (17:00 +0000)]
index.rst: Remove bb-chapuni from list of IRC bots
llvm-svn: 354353
Hans Wennborg [Tue, 19 Feb 2019 17:00:29 +0000 (17:00 +0000)]
index.rst: Remove Dragonegg link
llvm-svn: 354352
Hans Wennborg [Tue, 19 Feb 2019 16:58:25 +0000 (16:58 +0000)]
Remove extraneous space in MSVC-style diagnostic output
There was an extra space between the file location and the diagnostic
message:
/tmp/a.c(1,12): warning: unused parameter 'unused'
the tests didn't catch this due to FileCheck not running in --strict-whitespace mode.
Reported by Marco: http://lists.llvm.org/pipermail/cfe-dev/2019-February/061326.html
Differential revision: https://reviews.llvm.org/D58377
llvm-svn: 354351
Petr Hosek [Tue, 19 Feb 2019 16:53:33 +0000 (16:53 +0000)]
[compiler-rt][CMake] Set project for the custom libc++
This is another follow up to r354212 which is broken on Darwin when
cross-compiling runtimes to Linux when it ignores the -fuse-ld=lld
linker flag and attempts to use the host linker when performing the
compiler identification. Upon investigation, I noticed that setting
the project with appropriate list of languages makes the error go
away and it shouldn't hurt either.
Differential Revision: https://reviews.llvm.org/D58372
llvm-svn: 354350
Ilya Biryukov [Tue, 19 Feb 2019 16:50:37 +0000 (16:50 +0000)]
[clangd] Add an option in the code to not display number of fixes
Summary:
Only to the APIs, which are used by our embedders.
We do not plan to add a user-facing option for this.
Reviewers: sammccall, ioeric
Reviewed By: sammccall
Subscribers: MaskRay, jkorous, arphaman, kadircet, cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D58387
llvm-svn: 354349
Matt Arsenault [Tue, 19 Feb 2019 16:44:22 +0000 (16:44 +0000)]
GlobalISel: Implement moreElementsVector for G_EXTRACT source
llvm-svn: 354348
Michael Kruse [Tue, 19 Feb 2019 16:38:20 +0000 (16:38 +0000)]
[OpenMP 5.0] Parsing/sema support for map clause with mapper modifier.
This patch implements the parsing and sema support for OpenMP map
clauses with potential user-defined mapper attached. User defined mapper
is a new feature in OpenMP 5.0. A map clause can have an explicit or
implicit associated mapper, which instructs the compiler to generate
extra data mapping. An example is shown below:
struct S { int len; int *d; };
#pragma omp declare mapper(id: struct S s) map(s, s.d[0:s.len])
struct S ss;
#pragma omp target map(mapper(id) tofrom: ss) // use the mapper with name 'id' to map ss
Contributed-by: Lingda Li <lildmh@gmail.com>
Differential Revision: https://reviews.llvm.org/D58074
llvm-svn: 354347
Simon Pilgrim [Tue, 19 Feb 2019 16:33:17 +0000 (16:33 +0000)]
[X86][AVX] Update VBROADCAST folds to always use v2i64 X86vzload
The VBROADCAST combines and SimplifyDemandedVectorElts improvements mean that we now more consistently use shorter (128-bit) X86vzload input operands.
Follow up to D58053
llvm-svn: 354346
Matt Arsenault [Tue, 19 Feb 2019 16:30:19 +0000 (16:30 +0000)]
GlobalISel: Implement moreElementsVector for bit ops
llvm-svn: 354345
James Henderson [Tue, 19 Feb 2019 16:22:21 +0000 (16:22 +0000)]
[yaml2obj][obj2yaml] Remove section type range markers from allowed mappings and support hex values
yaml2obj/obj2yaml previously supported SHT_LOOS, SHT_HIOS, and
SHT_LOPROC for section types. These are simply values that delineate a
range and don't really make sense as valid values. For example if a
section has type value 0x70000000, obj2yaml shouldn't print this value
as SHT_LOPROC. Additionally, this was missing the three other range
markers (SHT_HIPROC, SHT_LOUSER and SHT_HIUSER).
This change removes these three range markers. It also adds support for
specifying the type as an integer, to allow section types that LLVM
doesn't know about.
Reviewed by: grimar
Differential Revision: https://reviews.llvm.org/D58383
llvm-svn: 354344
Simon Pilgrim [Tue, 19 Feb 2019 16:20:09 +0000 (16:20 +0000)]
Cast from SDValue directly instead of superfluous getNode(). NFCI.
llvm-svn: 354343
Matt Arsenault [Tue, 19 Feb 2019 16:10:16 +0000 (16:10 +0000)]
GlobalISel: Verify g_insert
llvm-svn: 354342
Simon Atanasyan [Tue, 19 Feb 2019 16:00:07 +0000 (16:00 +0000)]
[mips] Fix indentations in the test. NFC
llvm-svn: 354341
Simon Pilgrim [Tue, 19 Feb 2019 15:57:09 +0000 (15:57 +0000)]
[X86][AVX] EltsFromConsecutiveLoads - Add BROADCAST lowering support
This patch adds scalar/subvector BROADCAST handling to EltsFromConsecutiveLoads.
It mainly shows codegen changes to 32-bit code which failed to handle i64 loads, although 64-bit code is also using this new path to more efficiently combine to a broadcast load.
Differential Revision: https://reviews.llvm.org/D58053
llvm-svn: 354340
George Rimar [Tue, 19 Feb 2019 15:29:25 +0000 (15:29 +0000)]
[ELF] - Update test case after yaml2obj format change.
yaml2obj was changed in r354338("[yaml2obj][obj2yaml] - Support SHT_GNU_versym (.gnu.version) section.")
llvm-svn: 354339
George Rimar [Tue, 19 Feb 2019 15:29:07 +0000 (15:29 +0000)]
[yaml2obj][obj2yaml] - Support SHT_GNU_versym (.gnu.version) section.
This patch adds support for parsing dumping the .gnu.version section.
Description of the section is: https://refspecs.linuxfoundation.org/LSB_1.3.0/gLSB/gLSB/symversion.html#SYMVERTBL
Differential revision: https://reviews.llvm.org/D58280
llvm-svn: 354338
Alexey Bader [Tue, 19 Feb 2019 15:19:06 +0000 (15:19 +0000)]
[OpenCL] Change type of block pointer for OpenCL
Summary:
For some reason OpenCL blocks in LLVM IR are represented as function pointers.
These pointers do not point to any real function and never get called. Actually
they point to some structure, which in turn contains pointer to the real block
invoke function.
This patch changes represntation of OpenCL blocks in LLVM IR from function
pointers to pointers to `%struct.__block_literal_generic`.
Such representation allows to avoid unnecessary bitcasts and simplifies
further processing (e.g. translation to SPIR-V ) of the module for targets
which do not support function pointers.
Patch by: Alexey Sotkin.
Reviewers: Anastasia, yaxunl, svenvh
Reviewed By: Anastasia
Subscribers: alexbatashev, cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D58277
llvm-svn: 354337
Simon Atanasyan [Tue, 19 Feb 2019 15:17:43 +0000 (15:17 +0000)]
Convert the eh-frame-hdr-abs-fde.s test to use x86 as a target architecture
The test checks common functionality. Let's use `x86` (generic LLD target)
as a target architecture.
llvm-svn: 354336
George Rimar [Tue, 19 Feb 2019 14:53:48 +0000 (14:53 +0000)]
Recommit r354328, r354329 "[obj2yaml][yaml2obj] - Add support of parsing/dumping of the .gnu.version_r section."
Fix:
Replace
assert(!IO.getContext() && "The IO context is initialized already");
with
assert(IO.getContext() && "The IO context is not initialized");
(this was introduced in r354329, where I tried to quickfix the darwin BB
and seems copypasted the assert from the wrong place).
Original commit message:
The section is described here:
https://refspecs.linuxfoundation.org/LSB_1.3.0/gLSB/gLSB/symverrqmts.html
Patch just teaches obj2yaml/yaml2obj to dump and parse such sections.
We did the finalization of string tables very late,
and I had to move the logic to make it a bit earlier.
That was needed in this patch since .gnu.version_r adds strings to .dynstr.
This might also be useful for implementing other special sections.
Everything else changed in this patch seems to be straightforward.
Differential revision: https://reviews.llvm.org/D58119
llvm-svn: 354335
Alex Bradbury [Tue, 19 Feb 2019 14:42:00 +0000 (14:42 +0000)]
[RISCV][NFC] Move some std::string to StringRef
llvm-svn: 354333
George Rimar [Tue, 19 Feb 2019 14:38:25 +0000 (14:38 +0000)]
Revert r354328, r354329 "[obj2yaml][yaml2obj] - Add support of parsing/dumping of the .gnu.version_r section."
Something went wrong. Bots are unhappy:
http://lab.llvm.org:8011/builders/llvm-clang-lld-x86_64-scei-ps4-ubuntu-fast/builds/44113/steps/test/logs/stdio
llvm-svn: 354332
Eric Liu [Tue, 19 Feb 2019 14:32:22 +0000 (14:32 +0000)]
[clangd] Handle unresolved scope specifier when fixing includes.
Summary:
In the following examples, "clangd" is unresolved, and the fixer will try to fix
include for `clang::clangd`; however, clang::clangd::X is usually intended. So
when handling a qualifier that is unresolved, we change the unresolved name and
scopes so that the fixer will fix "clang::clangd::X" in the following example.
```
namespace clang {
clangd::X
~~~~~~
}
// or
clang::clangd::X
~~~~~~
```
Reviewers: sammccall
Reviewed By: sammccall
Subscribers: ilya-biryukov, MaskRay, jkorous, arphaman, kadircet, jdoerfert, cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D58185
llvm-svn: 354330
George Rimar [Tue, 19 Feb 2019 14:22:10 +0000 (14:22 +0000)]
Fix BB after r354328.
Bot:
http://lab.llvm.org:8011/builders/lld-x86_64-darwin13/builds/30188/steps/build_Lld/logs/stdio
Error:
/Users/buildslave/as-bldslv9_new/lld-x86_64-darwin13/llvm.src/lib/ObjectYAML/ELFYAML.cpp:1013:15: error: unused variable 'Object' [-Werror,-Wunused-variable]
const auto *Object = static_cast<ELFYAML::Object *>(IO.getContext());
^
/Users/buildslave/as-bldslv9_new/lld-x86_64-darwin13/llvm.src/lib/ObjectYAML/ELFYAML.cpp:1023:15: error: unused variable 'Object' [-Werror,-Wunused-variable]
const auto *Object = static_cast<ELFYAML::Object *>(IO.getContext());
Fix:
change
const auto *Object = static_cast<ELFYAML::Object *>(IO.getContext());
assert(Object && "The IO context is not initialized");
to
assert(!IO.getContext() && "The IO context is initialized already");
llvm-svn: 354329
George Rimar [Tue, 19 Feb 2019 14:03:14 +0000 (14:03 +0000)]
[obj2yaml][yaml2obj] - Add support of parsing/dumping of the .gnu.version_r section.
The section is described here:
https://refspecs.linuxfoundation.org/LSB_1.3.0/gLSB/gLSB/symverrqmts.html
Patch just teaches obj2yaml/yaml2obj to dump and parse such sections.
We did the finalization of string tables very late,
and I had to move the logic to make it a bit earlier.
That was needed in this patch since .gnu.version_r adds strings to .dynstr.
This might also be useful for implementing other special sections.
Everything else changed in this patch seems to be straightforward.
Differential revision: https://reviews.llvm.org/D58119
llvm-svn: 354328
Pavel Labath [Tue, 19 Feb 2019 13:52:31 +0000 (13:52 +0000)]
Revert "minidump: Add ability to attach (breakpad) symbol files to placeholder modules"
This reverts r354263, because it uncovered a problem in handling of the
minidumps with conflicting UUIDs. If a minidump contains two files with
the same UUID, we will not create to placeholder modules for them, but
instead reuse the first one for the second instance. This creates a
problem because these modules have their load address hardcoded in them
(and I've added an assert to verify that).
Technically this is not a problem with this patch, as the same issue
existed in the previous implementation, but it did not have the assert
which would diagnose that. Nonetheless, I am reverting this until I
figure out what's the best course of action in this situation.
llvm-svn: 354324
Alex Bradbury [Tue, 19 Feb 2019 13:47:19 +0000 (13:47 +0000)]
[RISCV] Re-organise calling convention tests
Re-organise calling convention tests to prepare for ilp32f and ilp32d hard
float ABI tests. It's also clear that we need to introduce similar tests for
lp64.
llvm-svn: 354323
George Rimar [Tue, 19 Feb 2019 13:27:06 +0000 (13:27 +0000)]
Fix BB after r354319 "[yaml2obj] - Do not skip zeroes blocks if there are relocations against them."
Fix: move the test to x86 folder.
Seems it is needed, because llvm-objdump invocation used in test has -D (disasm) flag.
BB: http://lab.llvm.org:8011/builders/clang-hexagon-elf/builds/23016
/local/buildbot/slaves/hexagon-build-02/clang-hexagon-elf/stage1/bin/llvm-objdump:
error: '/local/buildbot/slaves/hexagon-build-02/clang-hexagon-elf/stage1/test/tools/llvm-objdump/Output/disasm-zeroes-relocations.test.tmp':
can't find target: : error: unable to get target for 'x86_64--', see --version and --triple.
.
llvm-svn: 354322
George Rimar [Tue, 19 Feb 2019 12:38:36 +0000 (12:38 +0000)]
[yaml2obj] - Do not skip zeroes blocks if there are relocations against them.
This is for -D -reloc combination.
With this patch, we do not skip the zero bytes that have a relocation against
them when -reloc is used. If -reloc is not used, then the behavior will be the same.
Differential revision: https://reviews.llvm.org/D58174
llvm-svn: 354319
George Rimar [Tue, 19 Feb 2019 12:15:04 +0000 (12:15 +0000)]
[yaml2obj] - Do not ignore explicit addresses for .dynsym and .dynstr
This fixes https://bugs.llvm.org/show_bug.cgi?id=40339
Previously if the addresses were set in YAML they were ignored for
.dynsym and .dynstr sections. The patch fixes that.
Differential revision: https://reviews.llvm.org/D58168
llvm-svn: 354318
Diana Picus [Tue, 19 Feb 2019 11:34:36 +0000 (11:34 +0000)]
Fix obsolete comment. NFC
Both files mentioned in the comment now include TargetOpcodes.def. Just
mention that directly.
llvm-svn: 354316
Simon Atanasyan [Tue, 19 Feb 2019 11:15:28 +0000 (11:15 +0000)]
[mips] Remove trailing whitespaces from the test case. NFC
llvm-svn: 354315
Max Kazantsev [Tue, 19 Feb 2019 11:14:05 +0000 (11:14 +0000)]
[NFC] API for signaling that the current loop is being deleted
We are planning to be able to delete the current loop in LoopSimplifyCFG
in the future. Add API to notify the loop pass manager that it happened.
llvm-svn: 354314
Max Kazantsev [Tue, 19 Feb 2019 11:13:58 +0000 (11:13 +0000)]
[NFC] Store loop header in a local to keep it available after the loop is deleted
llvm-svn: 354313
Simon Atanasyan [Tue, 19 Feb 2019 11:11:12 +0000 (11:11 +0000)]
[mips] Create LA25 thunks for MIPS R6 code
MIPS R6 code uses the `R_MIPS_PC26_S2` relocation for calls which might
cross boundaries of non-PIC-to-PIC code. We need to create a LA25 thunks
for that case.
llvm-svn: 354312
Simon Atanasyan [Tue, 19 Feb 2019 10:36:58 +0000 (10:36 +0000)]
[MIPS] Handle cross-mode (regular <-> microMIPS) jumps
The patch solves two tasks:
1. MIPS ABI allows to mix regular and microMIPS code and perform
cross-mode jumps. Linker needs to detect such cases and replace
jump/branch instructions by their cross-mode equivalents.
2. Other tools like dunamic linkers need to recognize cases when dynamic
table entries, e_entry field of an ELF header etc point to microMIPS
symbol. Linker should provide such information.
The first task is implemented in the `MIPS<ELFT>::relocateOne()` method.
New routine `fixupCrossModeJump` detects ISA mode change, checks and
replaces an instruction.
The main problem is how to recognize that relocation target is microMIPS
symbol. For absolute and section symbols compiler or assembler set the
less-significant bit of the symbol's value or sum of the symbol's value
and addend. And this bit signals to linker about microMIPS code. For
global symbols compiler cannot do the same trick because other tools like,
for example, disassembler wants to know an actual position of the symbol.
So compiler sets STO_MIPS_MICROMIPS flag in the `st_other` field.
In `MIPS<ELFT>::relocateOne()` method we have a symbol's value only and
cannot access any symbol's attributes. To pass type of the symbol
(regular/microMIPS) to that routine as well as other places where we
write a symbol value as-is (.dynamic section, `Elf_Ehdr::e_entry` field
etc) we set when necessary a less-significant bit in the `getSymVA`
function.
Differential revision: https://reviews.llvm.org/D40147
llvm-svn: 354311
Diana Picus [Tue, 19 Feb 2019 10:26:47 +0000 (10:26 +0000)]
[ARM GlobalISel] Support G_PHI for Thumb2
Same as arm mode.
llvm-svn: 354310
Pavel Labath [Tue, 19 Feb 2019 08:25:25 +0000 (08:25 +0000)]
Fix vscode tests for python3
encode/decode the data before sending it over the socket. Since (AFAICT)
the vscode protocol (unlike the gdb-remote one) is fully textual, using
the utf8 codec here is appropriate.
llvm-svn: 354308
Pavel Labath [Tue, 19 Feb 2019 08:12:41 +0000 (08:12 +0000)]
[gui] Simplify SourceFileWindowDelegate::WindowDelegateDraw
instead of printf-ing into a buffer, and them using that buffer as a
format string, simply use the appropriate indirect format string.
This also fixes a -Wformat-truncation warning with gcc.
llvm-svn: 354307
Fangrui Song [Tue, 19 Feb 2019 05:16:52 +0000 (05:16 +0000)]
[Dominators] Fix and optimize edge insertion of depth-based search
Summary:
After (x,y) is inserted, depth-based search finds all affected v that satisfies:
depth(nca(x,y))+1 < depth(v) && there exists a path P from y to v where every w on P satisfies depth(v) <= depth(w)
This reduces to a widest path problem (maximizing the depth of the
minimum vertex in the path) which can be solved by a modified version of
Dijkstra with a bucket queue (named depth-based search in the paper).
The algorithm visits vertices in decreasing order of bucket number.
However, the current code misused priority_queue to extract them in
increasing order. I cannot think of a failing scenario but it surely may
process vertices more than once due to the local usage of Processed.
This patch fixes this bug and simplifies/optimizes the code a bit. Also
add more comments.
Reviewers: kuhar
Reviewed By: kuhar
Subscribers: kristina, jdoerfert, llvm-commits, NutshellySima, brzycki
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D58349
llvm-svn: 354306
Craig Topper [Tue, 19 Feb 2019 03:04:14 +0000 (03:04 +0000)]
[X86] Remove command line strings from the ProcIntel* features.
These should always follow the CPU string. There's no reason to control them independently.
llvm-svn: 354304
Jonas Devlieghere [Tue, 19 Feb 2019 01:20:49 +0000 (01:20 +0000)]
[lldb-instr] Test that we ignore existing macros.
Although the functionality was already present, it wasn't tested.
llvm-svn: 354303
Jonas Devlieghere [Tue, 19 Feb 2019 01:04:31 +0000 (01:04 +0000)]
[lldb-instr] Wrap returns of struct/classes in LLDB_RECORD_RESULT
The instrumentation framework requires return values of custom classes
and structs to be wrapped in the LLDB_RECORD_RESULT macro.
llvm-svn: 354301
Jonas Devlieghere [Tue, 19 Feb 2019 00:08:10 +0000 (00:08 +0000)]
Disable TestModuleCXX.test on Windows
Importing cxx modules doesn't seem to work on Windows:
error: a.out :: Class 'tagARRAYDESC' has a member 'tdescElem' of type
'tagTYPEDESC' which does not have a complete definition.
error: a.out :: Class 'tagPARAMDESCEX' has a member 'varDefaultValue' of type
'tagVARIANT' which does not have a complete definition.
llvm-svn: 354300
Jessica Paquette [Mon, 18 Feb 2019 23:33:24 +0000 (23:33 +0000)]
[GlobalISel][AArch64] Legalize + select some llvm.ctlz.* intrinsics
Legalize/select llvm.ctlz.*
Add select-ctlz to show that we actually select them. Update arm64-clrsb.ll and
arm64-vclz.ll to show that we perform valid transformations in optimized builds,
and document where GISel can improve.
Differential Revision: https://reviews.llvm.org/D58155
llvm-svn: 354299
Sanjay Patel [Mon, 18 Feb 2019 23:33:05 +0000 (23:33 +0000)]
[CGP] form usub with overflow from sub+icmp
The motivating x86 cases for forming the intrinsic are shown in PR31754 and PR40487:
https://bugs.llvm.org/show_bug.cgi?id=31754
https://bugs.llvm.org/show_bug.cgi?id=40487
..and those are shown in the IR test file and x86 codegen file.
Matching the usubo pattern is harder than uaddo because we have 2 independent values rather than a def-use.
This adds a TLI hook that should preserve the existing behavior for uaddo formation, but disables usubo
formation by default. Only x86 overrides that setting for now although other targets will likely benefit
by forming usbuo too.
Differential Revision: https://reviews.llvm.org/D57789
llvm-svn: 354298
Davide Italiano [Mon, 18 Feb 2019 23:18:14 +0000 (23:18 +0000)]
[lldbtest] Fix some code to be compatible between py2 and py3.
llvm-svn: 354297
Jan Korous [Mon, 18 Feb 2019 23:12:29 +0000 (23:12 +0000)]
[clang][test] Fix FileManagerTest.getFileDontOpenRealPath for Windows
llvm-svn: 354296
Changpeng Fang [Mon, 18 Feb 2019 23:00:26 +0000 (23:00 +0000)]
AMDGPU: Use MachineInstr::mayAlias to replace areMemAccessesTriviallyDisjoint in LoadStoreOptimizer pass.
Summary:
This is to fix a memory dependence bug in LoadStoreOptimizer.
Reviewers:
arsenm, rampitec
Differential Revision:
https://reviews.llvm.org/D58295
llvm-svn: 354295
Jonas Devlieghere [Mon, 18 Feb 2019 22:59:52 +0000 (22:59 +0000)]
[lldb-instr] Add constructor and move test into lit/tools
The test had an implicit constructor for the Foo struct. Also, as the
instrumentation doesn't have to be reproducer specific, I moved the
tests into the lit/tools directory.
llvm-svn: 354294
Matt Arsenault [Mon, 18 Feb 2019 22:39:27 +0000 (22:39 +0000)]
GlobalISel: Implement widenScalar for g_extract scalar results
llvm-svn: 354293
Matt Arsenault [Mon, 18 Feb 2019 22:39:22 +0000 (22:39 +0000)]
GlobalISel: Make buildExtract use DstOp/SrcOp
llvm-svn: 354292
Jan Korous [Mon, 18 Feb 2019 22:33:40 +0000 (22:33 +0000)]
Reland "[clang][FileManager] fillRealPathName even if we aren't opening the file"
This reverts commit
e2bb3121fd4ab5b01f9ec1d2e3e9877db9c6a54c.
+ fixed test for Windows
llvm-svn: 354291
Julian Lettner [Mon, 18 Feb 2019 22:29:17 +0000 (22:29 +0000)]
[LSan][Darwin][NFC] Add comment explaining test failure
llvm-svn: 354290
Jason Molenda [Mon, 18 Feb 2019 22:18:11 +0000 (22:18 +0000)]
One more fix while I'm looking at this - remove the
unused IsSBProcess method, and have IsFBSProcess
return false if we don't have API that we can use to
make that determination, so we'll try other API
if we can.
llvm-svn: 354289
Jason Molenda [Mon, 18 Feb 2019 22:13:34 +0000 (22:13 +0000)]
Ah, misunderstood Jonas' feedback - fix this so we'll
do the right thing when both API are available. We
want to try both of them if the first one fails.
llvm-svn: 354288
Jason Molenda [Mon, 18 Feb 2019 22:10:23 +0000 (22:10 +0000)]
Clean up an unused variable warning when building this for
mac native.
llvm-svn: 354287
Davide Italiano [Mon, 18 Feb 2019 21:53:02 +0000 (21:53 +0000)]
[Python3] Fix TestObjCMethods.py to work with py2 and 3.
llvm-svn: 354286
Petr Hosek [Mon, 18 Feb 2019 20:58:06 +0000 (20:58 +0000)]
[libcxxabi][CMake] Drop unused HandleOutOfTreeLLVM include
This include doesn't seem to be needed for the standalone build (it's
not being used by libc++ build either), but introduces unnecessary
dependency because HandleOutOfTreeLLVM performs checks that require
a working C++ library. We shouldn't require a working C++ library to
build libc++abi or libc++ (it's what we're building after all).
Differential Revision: https://reviews.llvm.org/D58333
llvm-svn: 354284
Jonas Devlieghere [Mon, 18 Feb 2019 20:31:18 +0000 (20:31 +0000)]
[Reproducers] Make clang use lldb's VFS.
In r353906 we hooked up clang and lldb's reproducer infrastructure to
capture files used by clang. This patch adds the necessary logic to have
clang reuse the files from lldb's reproducer during replay.
Differential revision: https://reviews.llvm.org/D58309
llvm-svn: 354283
Julian Lettner [Mon, 18 Feb 2019 18:47:49 +0000 (18:47 +0000)]
[Sanitizer] On Darwin `__sanitizer_print_stack_trace` only prints topmost frame
In compiler-rt we have the notion of a `fast` and a `slow` stack
unwinder. Darwin currently only supports the fast unwinder.
From reading the code, my understanding is that
`BufferedStackTrace::Unwind` can be called with `bp=0, stack_top=0,
stack_bottom=0, request_fast_unwind=false`. If
`request_fast_unwind=true`, then we alos need to supply bp, stack_top,
and stack_bottom.
However, `BufferedStackTrace::Unwind` uses
`StackTrace::WillUseFastUnwind` which will adapt `request_fast_unwind`
if the requested unwinder is not supported. On Darwin, the result is
that we don't pass actual values for bp, stack_top, and stack_bottom,
but end up using the fast unwinder. The tests then fail because we only
print the topmost stack frame.
This patch adds a check to `WillUseFastUnwind` at the point of usage to
avoid the mismatch between `request_fast_unwind` and what `Unwind`
actually does. I am also interested in cleaning up the
`request_fast_unwind` machinery so this patch just the simplest thing
possible so I can enable the tests.
Reviewers: vitalybuka, vsk
Differential Revision: https://reviews.llvm.org/D58156
llvm-svn: 354282
Matt Arsenault [Mon, 18 Feb 2019 17:01:09 +0000 (17:01 +0000)]
GlobalISel: Fix double count of offset for irregular vector breakdowns
Fixes cases with odd vectors that break into multiple requested size
pieces.
llvm-svn: 354280
Sanjay Patel [Mon, 18 Feb 2019 16:46:12 +0000 (16:46 +0000)]
[x86] split more v8f32/v8i32 shuffles in lowering
Similar to D57867 - this is a small patch with lots of test diffs.
With half-vector-width narrowing potential, using an extract + 128-bit vshufps
is a win because it replaces a 256-bit shuffle with a 128-bit shufle.
This seems like it should be a win even for targets with 'fast-variable-shuffle',
but we are intentionally deferring that to an independent change to make sure
that is true.
Differential Revision: https://reviews.llvm.org/D58181
llvm-svn: 354279
Pavel Labath [Mon, 18 Feb 2019 16:05:26 +0000 (16:05 +0000)]
Un-XFAIL TestLinuxCore for windows
It turns out all that was needed to get this test passing was to fix the
python3 incompatibility.
llvm-svn: 354278
Sanjay Patel [Mon, 18 Feb 2019 16:04:22 +0000 (16:04 +0000)]
Revert "[InstCombine] reduce even more unsigned saturated add with 'not' op"
This reverts commit
079b610c29b4a428b3ae7b64dbac0378facf6632.
Bots are failing after this change on a stage 2 compile of clang.
llvm-svn: 354277
Sanjay Patel [Mon, 18 Feb 2019 15:21:39 +0000 (15:21 +0000)]
[InstCombine] reduce even more unsigned saturated add with 'not' op
We want to use the sum in the icmp to allow matching with
m_UAddWithOverflow and eliminate the 'not'. This is discussed
in D51929 and is another step towards solving PR14613:
https://bugs.llvm.org/show_bug.cgi?id=14613
Name: uaddsat, -1 fval
%notx = xor i32 %x, -1
%a = add i32 %x, %y
%c = icmp ugt i32 %notx, %y
%r = select i1 %c, i32 %a, i32 -1
=>
%a = add i32 %x, %y
%c2 = icmp ugt i32 %y, %a
%r = select i1 %c2, i32 -1, i32 %a
Name: uaddsat, -1 fval + ult
%notx = xor i32 %x, -1
%a = add i32 %x, %y
%c = icmp ult i32 %y, %notx
%r = select i1 %c, i32 %a, i32 -1
=>
%a = add i32 %x, %y
%c2 = icmp ugt i32 %y, %a
%r = select i1 %c2, i32 -1, i32 %a
https://rise4fun.com/Alive/nTp
llvm-svn: 354276
Pavel Labath [Mon, 18 Feb 2019 15:04:33 +0000 (15:04 +0000)]
Fix TestLinuxCore for python3
- dictionaries don't have iteritems()
- division returns floats
llvm-svn: 354273
Kadir Cetinkaya [Mon, 18 Feb 2019 14:23:19 +0000 (14:23 +0000)]
[clangd] Add tests for template specializations
Subscribers: ilya-biryukov, ioeric, MaskRay, jkorous, arphaman, cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D58190
llvm-svn: 354272
Andrea Di Biagio [Mon, 18 Feb 2019 14:15:25 +0000 (14:15 +0000)]
[MCA] Correctly update register definitions in the PRF after move elimination.
This patch fixes a bug where register writes performed by optimizable register
moves were sometimes wrongly treated like partial register updates. Before this
patch, llvm-mca wrongly predicted a 1.50 IPC for test reg-move-elimination-6.s
(added by this patch). With this patch, llvm-mca correctly updates the register
defintions in the PRF, and the IPC for that test is now correctly reported as 2.
llvm-svn: 354271
George Rimar [Mon, 18 Feb 2019 13:58:12 +0000 (13:58 +0000)]
[llvm-readobj] - Simplify .gnu.version_d dumping.
This is similar to D58048.
Instead of scanning the dynamic table to read the
DT_VERDEFNUM, we could take it from the sh_info field.
(https://docs.oracle.com/cd/E19683-01/816-1386/chapter6-94076/index.html)
The patch does this.
llvm-svn: 354270
Eric Liu [Mon, 18 Feb 2019 13:12:10 +0000 (13:12 +0000)]
[clangd] Cache include fixes for diagnostics caused by the same unresolved name or incomplete type.
Summary:
Multiple diagnostics can be caused by the same unresolved name or incomplete type,
especially if the code is copy-pasted without #includes. The cache can avoid making
repetitive index requests, and thus reduce latency and allow more diagnostics to be
fixed (we limit the number of index requests for each parse).
Reviewers: sammccall
Reviewed By: sammccall
Subscribers: ilya-biryukov, MaskRay, jkorous, arphaman, kadircet, jdoerfert, cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D58239
llvm-svn: 354268
Gabor Marton [Mon, 18 Feb 2019 13:09:27 +0000 (13:09 +0000)]
[ASTImporter] Find previous friend function template
Reviewers: a_sidorin, shafik, a.sidorin
Subscribers: rnkovacs, dkrupp, Szelethus, gamesh411, cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D57910
llvm-svn: 354267
Alexander Kornienko [Mon, 18 Feb 2019 12:50:35 +0000 (12:50 +0000)]
[clang-tidy] Fix links in docs.
llvm-svn: 354266
Oliver Stannard [Mon, 18 Feb 2019 12:39:47 +0000 (12:39 +0000)]
[ARM] Add pre-defined macros for ROPI and RWPI
This adds ACLE-defined macros to test for code being compiled in the ROPI and
RWPI position-independence modes.
Differential revision: https://reviews.llvm.org/D23610
llvm-svn: 354265
Serge Guelton [Mon, 18 Feb 2019 12:07:12 +0000 (12:07 +0000)]
[NFC] Make Optional<T> trivially copyable when T is trivially copyable
This is a follow-up to r354246 and a reimplementation of https://reviews.llvm.org/D57097?id=186600
that should not trigger any UB thanks to the use of an union.
This may still be subject to the problem solved by std::launder, but I'm unsure how it interacts whith union.
/me plans to revert if this triggers any relevant bot failure. At least this validates in Release mode with
clang 6.0.1 and gcc 4.8.5.
llvm-svn: 354264