Kostya Serebryany [Tue, 3 Mar 2015 02:45:21 +0000 (02:45 +0000)]
[asan] Since x32 has 32-bit pointers, it should use the same code sequence as ia32 for AddressSanitizer asm_flags test. Patch by H.J. Lu
llvm-svn: 231052
Sanjoy Das [Tue, 3 Mar 2015 01:59:38 +0000 (01:59 +0000)]
Revert "[ADT] fail-fast iterators for DenseMap"
This reverts commit r231035. It breaks clang.
llvm-svn: 231050
David Majnemer [Tue, 3 Mar 2015 01:50:05 +0000 (01:50 +0000)]
Sema: Properly initialize the thrown exception object
We would create the exception object with the wrong qualifiers, ensuring
that the wrong copy constructor would get called.
llvm-svn: 231049
Greg Clayton [Tue, 3 Mar 2015 01:40:46 +0000 (01:40 +0000)]
Fix a crasher where we might have a N_UNDF symbol with no name and this could cause ObjectFileMachO::ParseSymtab() to crash.
<rdar://problem/
19989491>
llvm-svn: 231048
Peter Collingbourne [Tue, 3 Mar 2015 01:39:13 +0000 (01:39 +0000)]
Add explicit type to empty std::set initializer to fix the libc++ build.
llvm-svn: 231047
Ahmed Bougacha [Tue, 3 Mar 2015 01:21:16 +0000 (01:21 +0000)]
[X86] Special-case 2x CMOV when custom-inserting.
This lets us avoid a few copies that are otherwise hard to get rid of.
The way this is done is, the custom-inserter looks at the following
instruction for another CMOV, and replaces both at the same time.
A previous version used a new CMOV2 opcode, but the custom inserter
is expected to be able to return a different basic block anyway, which
means it's OK - though far from ideal - to alter that block's contents.
Explicitly document that, in case it ever makes a difference.
Alternatives welcome!
Follow-up to r231045.
rdar://
19767934
Closes http://reviews.llvm.org/D8019
llvm-svn: 231046
Ahmed Bougacha [Tue, 3 Mar 2015 01:09:14 +0000 (01:09 +0000)]
[X86] Combine (cmov (and/or (setcc) (setcc))) into (cmov (cmov)).
Fold and/or of setcc's to double CMOV:
(CMOV F, T, ((cc1 | cc2) != 0)) -> (CMOV (CMOV F, T, cc1), T, cc2)
(CMOV F, T, ((cc1 & cc2) != 0)) -> (CMOV (CMOV T, F, !cc1), F, !cc2)
When we can't use the CMOV instruction, it might increase branch
mispredicts. When we can, or when there is no mispredict, this
improves throughput and reduces register pressure.
These can't be catched by generic combines, because the pattern can
appear when legalizing some instructions (such as fcmp une).
rdar://
19767934
http://reviews.llvm.org/D7634
llvm-svn: 231045
Reid Kleckner [Tue, 3 Mar 2015 01:04:39 +0000 (01:04 +0000)]
Fix cppeh breakage due to racing commits
llvm-svn: 231044
Peter Collingbourne [Tue, 3 Mar 2015 00:49:28 +0000 (00:49 +0000)]
LowerBitSets: Use byte arrays instead of bit sets to represent in-memory bit sets.
By loading from indexed offsets into a byte array and applying a mask, a
program can test bits from the bit set with a relatively short instruction
sequence. For example, suppose we have 15 bit sets to lay out:
A (16 bits), B (15 bits), C (14 bits), D (13 bits), E (12 bits),
F (11 bits), G (10 bits), H (9 bits), I (7 bits), J (6 bits), K (5 bits),
L (4 bits), M (3 bits), N (2 bits), O (1 bit)
These bits can be laid out in a 16-byte array like this:
Byte Offset
0123456789ABCDEF
Bit
7 HHHHHHHHHIIIIIII
6 GGGGGGGGGGJJJJJJ
5 FFFFFFFFFFFKKKKK
4 EEEEEEEEEEEELLLL
3 DDDDDDDDDDDDDMMM
2 CCCCCCCCCCCCCCNN
1 BBBBBBBBBBBBBBBO
0
AAAAAAAAAAAAAAAA
For example, to test bit X of A, we evaluate ((bits[X] & 1) != 0), or to
test bit X of I, we evaluate ((bits[9 + X] & 0x80) != 0). This can be done
in 1-2 machine instructions on x86, or 4-6 instructions on ARM.
This uses the LPT multiprocessor scheduling algorithm to lay out the bits
efficiently.
Saves ~450KB of instructions in a recent build of Chromium.
Differential Revision: http://reviews.llvm.org/D7954
llvm-svn: 231043
Andrew Kaylor [Tue, 3 Mar 2015 00:41:03 +0000 (00:41 +0000)]
Remap arguments and non-alloca values used by outlined C++ exception handlers.
Differential Revision: http://reviews.llvm.org/D7844
llvm-svn: 231042
Benjamin Kramer [Tue, 3 Mar 2015 00:17:09 +0000 (00:17 +0000)]
LoopIdiom: Give globals for memset_pattern16 private linkage.
There's really no reason to have them have entries in the symbol table
anymore. Old versions of ld64 had some bugs in this area but those have
been fixed long ago.
llvm-svn: 231041
Alexey Samsonov [Tue, 3 Mar 2015 00:14:32 +0000 (00:14 +0000)]
[Sanitizers] Remove duplication in sanitizer group definition. NFC.
There is no need to list sanitizers in both "UndefinedTrap" and
"Undefined" groups - it turns out using one group in a defintion
of another group "just works".
llvm-svn: 231040
Matthias Braun [Tue, 3 Mar 2015 00:12:08 +0000 (00:12 +0000)]
Fix typo in my last commit.
llvm-svn: 231039
Reid Kleckner [Tue, 3 Mar 2015 00:05:35 +0000 (00:05 +0000)]
WinEH: Run opt -instnamer over some cppeh tests and update CHECKs
In the future, we should run the output of clang through instnamer to
make it easier to manually edit test cases.
No functionality change.
llvm-svn: 231037
Fariborz Jahanian [Mon, 2 Mar 2015 23:58:02 +0000 (23:58 +0000)]
[SDK modernizer] Patch to fix type of the typed enums when
migrating to NS_ENUM typedef. rdar://
19994496
llvm-svn: 231036
Sanjoy Das [Mon, 2 Mar 2015 23:29:37 +0000 (23:29 +0000)]
[ADT] fail-fast iterators for DenseMap
Summary:
This patch is an attempt at making `DenseMapIterator`s "fail-fast".
Fail-fast iterators that have been invalidated due to insertion into
the host `DenseMap` deterministically trip an assert (in debug mode)
on access, instead of non-deterministically hitting memory corruption
issues.
Reviewers: dexonsmith, dberlin, ruiu, chandlerc
Reviewed By: chandlerc
Subscribers: yaron.keren, chandlerc, llvm-commits
Differential Revision: http://reviews.llvm.org/D7931
llvm-svn: 231035
Michael Zolotukhin [Mon, 2 Mar 2015 23:24:40 +0000 (23:24 +0000)]
TLI: Factor out sanitizeFunctionName. NFC.
llvm-svn: 231034
Rui Ueyama [Mon, 2 Mar 2015 23:03:33 +0000 (23:03 +0000)]
Make ArchiveLibraryFile::~ArchiveLibraryFile virtual.
"virtual" was present at a wrong place. FileArchive is a subclass of
ArchiveLibraryFile, and a FileArchive can be deleted through a
pointer of ArchiveLibraryFile. We want to make the destructor of the
base class virtual.
llvm-svn: 231033
Kuba Brecka [Mon, 2 Mar 2015 22:56:25 +0000 (22:56 +0000)]
Symbolizer refactoring: Make LibbacktraceSymbolizer adopt the SymbolizerTool interface
Reviewed at http://reviews.llvm.org/D7971
llvm-svn: 231032
Reid Kleckner [Mon, 2 Mar 2015 22:55:01 +0000 (22:55 +0000)]
Remove shell requirement from test/Modules/explicit-build-relpath.cpp
It only relies on 'cd', which the internal shell has now.
llvm-svn: 231031
Reid Kleckner [Mon, 2 Mar 2015 22:50:09 +0000 (22:50 +0000)]
Revert bad change from r231029
llvm-svn: 231030
Reid Kleckner [Mon, 2 Mar 2015 22:42:58 +0000 (22:42 +0000)]
Remove shell requirements from tests that use 'cd'
Modules and Tooling tests in particular tend to want to change the cwd,
so we were missing test coverage in this area on Windows. It should now
be easier to write such portable tests.
llvm-svn: 231029
Benjamin Kramer [Mon, 2 Mar 2015 22:20:22 +0000 (22:20 +0000)]
IndexedMap: Document why SmallVector<T, 0> is preferable here.
llvm-svn: 231028
Kuba Brecka [Mon, 2 Mar 2015 22:15:14 +0000 (22:15 +0000)]
Symbolizer refactoring: ExtractToken and friends
Reviewed at http://reviews.llvm.org/D7867
llvm-svn: 231027
Hans Wennborg [Mon, 2 Mar 2015 22:09:05 +0000 (22:09 +0000)]
clang-cl: Correctly ignore /openmp- (PR22748)
llvm-svn: 231026
Alexey Samsonov [Mon, 2 Mar 2015 22:03:52 +0000 (22:03 +0000)]
[Sanitizer] Run test/sanitizer_common lit tests on all supported architectures.
llvm-svn: 231025
Adrian Prantl [Mon, 2 Mar 2015 22:02:36 +0000 (22:02 +0000)]
Revert "Revert "For the dwarf expression code get the subtarget off of the current""
This reapplies r230990 without modifications.
llvm-svn: 231024
Adrian Prantl [Mon, 2 Mar 2015 22:02:33 +0000 (22:02 +0000)]
Refactor DebugLocDWARFExpression so it doesn't require access to the
TargetRegisterInfo. DebugLocEntry now holds a buffer with the raw bytes
of the pre-calculated DWARF expression.
Ought to be NFC, but it does slightly alter the output format of the
textual assembly.
This reapplies 230930 without the assertion in DebugLocEntry::finalize()
because not all Machine registers can be lowered into DWARF register
numbers and floating point constants cannot be expressed.
llvm-svn: 231023
Reid Kleckner [Mon, 2 Mar 2015 22:00:23 +0000 (22:00 +0000)]
cmake: Use MSVC atomics with clang-cl
llvm-svn: 231022
Jan Vesely [Mon, 2 Mar 2015 21:50:28 +0000 (21:50 +0000)]
Support: Use const pointers for reads.
Fixes tons of const-cast warnings.
Signed-off-by: Jan Vesely <jan.vesely@rutgers.edu>
Reviewed-by: Rui Ueyama <ruiu@google.com>
llvm-svn: 231021
Sanjoy Das [Mon, 2 Mar 2015 21:47:47 +0000 (21:47 +0000)]
Fix test case.
r230921 broke backend-optimization-failure.cpp: after
r230921, LLVM no longer emits an expression to compute 'Length - 1'
and this perturbs LoopSimplify enough to emit the warning on line 10
instead of line 9. This is a review request to fix the test case once
I re-land r230921.
llvm-svn: 231020
Fariborz Jahanian [Mon, 2 Mar 2015 21:46:17 +0000 (21:46 +0000)]
Attempt to fix buildbot.
llvm-svn: 231019
Sanjoy Das [Mon, 2 Mar 2015 21:41:07 +0000 (21:41 +0000)]
Revert some changes that were made to fix PR20680.
This re-lands change r230921. r230921 was reverted because it broke a
clang test; a checkin fixing the clang test will be commited shortly.
Summary:
As far as I can tell, the real bug causing the issue was fixed in
r230533. SCEVExpander should mark an increment operation as nuw or nsw
only if it can *prove* that the operation does not overflow. There
shouldn't be any situation where we have to do something different
because of no-wrap flags generated by SCEVExpander.
Revert "IndVarSimplify: Allow LFTR to fire more often"
This reverts commit
1ade0f0faa98877b688e0b9da58e876052c1e04e (SVN: 222213).
Revert "IndVarSimplify: Don't let LFTR compare against a poison value"
This reverts commit
c0f2b8b528d8a37b0a1522aae90af649d6357eb5 (SVN: 217102).
Reviewers: majnemer, atrick, spatel
Differential Revision: http://reviews.llvm.org/D7979
llvm-svn: 231018
Reid Kleckner [Mon, 2 Mar 2015 21:33:18 +0000 (21:33 +0000)]
lit: Add 'cd' support to the internal shell and port some tests
The internal shell was already threading around a 'cwd' parameter. We
just have to make it mutable so that we can update it as the test script
executes.
If the shell ever grows support for environment variable substitution,
we could also implement support for export.
llvm-svn: 231017
Rui Ueyama [Mon, 2 Mar 2015 21:19:12 +0000 (21:19 +0000)]
Use read{16,32,64}{le,be}() instead of *reinterpret_cast<u{little,big}{16,32,64}_t>().
llvm-svn: 231016
Benjamin Kramer [Mon, 2 Mar 2015 21:16:04 +0000 (21:16 +0000)]
SmallVector: Allow initialization and assignment from initializer_list.
Modeled after std::vector.
llvm-svn: 231015
Kuba Brecka [Mon, 2 Mar 2015 21:15:09 +0000 (21:15 +0000)]
Symbolizer refactoring: Move internals to separate files
Reviewed at http://reviews.llvm.org/D7972
llvm-svn: 231014
Renato Golin [Mon, 2 Mar 2015 21:02:43 +0000 (21:02 +0000)]
Allow compiler-rt build on Bitrig
This is the only patch Bitrig uses for compiler-rt. It adds support
for Bitrig/arm in the clear cache code.
Patch by Patrick Wildt.
llvm-svn: 231012
Michael Zolotukhin [Mon, 2 Mar 2015 20:50:08 +0000 (20:50 +0000)]
TLI: Use lambda. NFC.
llvm-svn: 231011
Zachary Turner [Mon, 2 Mar 2015 20:45:26 +0000 (20:45 +0000)]
[CMake] Refactor LLDB main CMake file.
The existing state of affairs was getting a little unwieldy.
All of LLDB's utility functions and initial configuration was in
the root CMake file. I split this up into 3 separate files and
moved them to relevant subfolders under cmake/modules.
Also, I deleted the add_lldb_definitions() function. It seemed
to be somewhat useless and did not serve any real purpose that
I was able to figure out.
llvm-svn: 231010
Matthias Braun [Mon, 2 Mar 2015 20:44:09 +0000 (20:44 +0000)]
Improve robustness of dependency-generation-crash.c test.
The test wants to provoke a failure when opening the output file.
Using chmod 0 on the output file does not work reliably on all
filesystems or when running the test as root.
Change the test to use a nonexistant directory instead.
Differential Revision: http://reviews.llvm.org/D7620
llvm-svn: 231009
Fariborz Jahanian [Mon, 2 Mar 2015 20:43:32 +0000 (20:43 +0000)]
[SDK modernizer]. Patch fixes driver's lack of
recognition of mernizer's -objcmt-migrate-property-dot-syntax
option. rdar://
19994452
llvm-svn: 231008
Michael Zolotukhin [Mon, 2 Mar 2015 20:43:24 +0000 (20:43 +0000)]
Make ToVectorTy static.
llvm-svn: 231007
Michael Zolotukhin [Mon, 2 Mar 2015 20:37:10 +0000 (20:37 +0000)]
Fix a copy-paste bug.
llvm-svn: 231006
Rui Ueyama [Mon, 2 Mar 2015 20:31:43 +0000 (20:31 +0000)]
Remove include/lld/Core/Endian.h and use llvm/Support/Endian.h instead.
llvm-svn: 231005
Adrian Prantl [Mon, 2 Mar 2015 20:01:54 +0000 (20:01 +0000)]
Revert "Refactor DebugLocDWARFExpression so it doesn't require access to the"
This reverts commit 230975 to investigate buildbot breakage.
llvm-svn: 231004
Adrian Prantl [Mon, 2 Mar 2015 20:01:47 +0000 (20:01 +0000)]
Revert "For the dwarf expression code get the subtarget off of the current"
This reverts commit 230990 because also reverting 230975.
llvm-svn: 231003
Rui Ueyama [Mon, 2 Mar 2015 20:00:15 +0000 (20:00 +0000)]
Support: Add {read,write}{16,32,64}{le,be} functions.
Add convenient functions for endian-aware reads/writes.
llvm-svn: 231002
Paul Robinson [Mon, 2 Mar 2015 19:52:42 +0000 (19:52 +0000)]
Remove useless .debug_macinfo section setup.
llvm-svn: 231001
Timur Iskhodzhanov [Mon, 2 Mar 2015 19:41:09 +0000 (19:41 +0000)]
[ASan/Win] Work around PR22545: call LLVM global_dtors in the MD atexit()
llvm-svn: 231000
Alexey Samsonov [Mon, 2 Mar 2015 19:34:27 +0000 (19:34 +0000)]
[MSan] Remove explicit -m64 from RUN lines.
Target-specific flags should usually be configured by CMake/lit.
llvm-svn: 230999
David Blaikie [Mon, 2 Mar 2015 19:34:11 +0000 (19:34 +0000)]
Change SystemZ large tests to use the existing long_tests property
(this is already used in Clang for a couple of tests)
Reviewers: uweigand
Differential Revision: http://reviews.llvm.org/D7965
llvm-svn: 230998
Alexey Samsonov [Mon, 2 Mar 2015 19:34:06 +0000 (19:34 +0000)]
[DFSan] Remove explicit -m64 from RUN lines.
Target-specific flags should usually be configured by CMake/lit.
llvm-svn: 230997
Alexey Samsonov [Mon, 2 Mar 2015 19:34:02 +0000 (19:34 +0000)]
[Sanitizer] Restrict open_memstream to x86_64. Fixes PR22714.
llvm-svn: 230996
Philip Reames [Mon, 2 Mar 2015 19:19:04 +0000 (19:19 +0000)]
[PerformanceTips] Document various items folks have suggested
This could stand to be expanded - patches welcome! - but let's at least write them down so they don't get forgotten.
llvm-svn: 230995
Filipe Cabecinhas [Mon, 2 Mar 2015 19:12:51 +0000 (19:12 +0000)]
Add -fexceptions for targets which need it
llvm-svn: 230994
Chaoren Lin [Mon, 2 Mar 2015 19:12:18 +0000 (19:12 +0000)]
Applying D7950 to four near copies of the file.
Summary:
Using std::thread/random instead of pthread/rand.
Two of them also suffered from the synchronization problem in D7916.
Reviewers: clayborg, zturner, ki.stfu
Subscribers: lldb-commits
Differential Revision: http://reviews.llvm.org/D7991
llvm-svn: 230993
Filipe Cabecinhas [Mon, 2 Mar 2015 19:11:53 +0000 (19:11 +0000)]
Add -fexceptions for targets where it isn't the default.
It still tests that objc++-cpp-output turns on -fcxx-exceptions and
-fobjc-exceptions by being a c++ and objc(++) file.
llvm-svn: 230992
Rafael Espindola [Mon, 2 Mar 2015 19:08:03 +0000 (19:08 +0000)]
Add r230655 back with a fix.
The issue is that now we have a diag handler during optimizations
and get forward every optimization remark, flooding stdout.
The same filtering should probably be done with or without a
custom handler, but for now just ignore remarks.
Original message:
gold-plugin: "Upgrade" debug info and handle its warnings.
The gold plugin never calls MaterializeModule, so any old debug info
was not deleted and could cause crashes.
Now that it is being "upgraded", the plugin also has to handle warnings
and create Modules with a nice id (it shows in the warning).
llvm-svn: 230991
Eric Christopher [Mon, 2 Mar 2015 19:01:47 +0000 (19:01 +0000)]
For the dwarf expression code get the subtarget off of the current
MachineFunction.
llvm-svn: 230990
Bob Wilson [Mon, 2 Mar 2015 19:01:14 +0000 (19:01 +0000)]
Add clang support for Objective-C application extensions.
This adds the -fapplication-extension option, along with the
ios_app_extension and macosx_app_extension availability attributes.
Patch by Ted Kremenek
llvm-svn: 230989
Juergen Ributzka [Mon, 2 Mar 2015 18:59:38 +0000 (18:59 +0000)]
Restore LLVMLinkModules C API until it is properly deprecated.
Add the enum "LLVMLinkerMode" back for backwards-compatibility and add the
linker mode parameter back to the "LLVMLinkModules" function. The paramter is
ignored and has no effect.
Patch provided by: Filip Pizlo
Reviewed by: Rafael and Sean
llvm-svn: 230988
Jan Vesely [Mon, 2 Mar 2015 18:56:52 +0000 (18:56 +0000)]
R600: Use c++11 style for loop
Signed-off-by: Jan Vesely <jan.vesely@rutgers.edu>
Reviewed-by: Tom Stellard <tom@stellard.net>
llvm-svn: 230987
Alexey Samsonov [Mon, 2 Mar 2015 18:55:46 +0000 (18:55 +0000)]
[Sanitizer] Fix StripPathPrefix function and improve test case.
llvm-svn: 230986
Paul Robinson [Mon, 2 Mar 2015 18:50:18 +0000 (18:50 +0000)]
Revert r230979, should apply to all X86 ELF.
llvm-svn: 230985
Filipe Cabecinhas [Mon, 2 Mar 2015 18:49:37 +0000 (18:49 +0000)]
Add -frtti and -fexceptions to tests that assume these are on.
Summary:
We now have targets that don't enable rtti/exceptions by default, and the
ASTMatchers tests are assuming that these features are on (e.g: They use
dynamic_cast or try).
Reviewers: klimek, thakis, djasper
Subscribers: klimek, cfe-commits
Differential Revision: http://reviews.llvm.org/D7892
llvm-svn: 230984
Zachary Turner [Mon, 2 Mar 2015 18:33:50 +0000 (18:33 +0000)]
[Windows] Fix getcwd() on Windows.
Windows doesn't define getcwd(), so this test executable failed
to compile.
Patch by Adrian McCarthy
Differential Revision: http://reviews.llvm.org/D7962
llvm-svn: 230983
Daniel Jasper [Mon, 2 Mar 2015 18:07:00 +0000 (18:07 +0000)]
clang-tidy: Update test as underlying problem apparently was fixed.
llvm-svn: 230982
Chris Bieneman [Mon, 2 Mar 2015 17:50:13 +0000 (17:50 +0000)]
Making the SO version major.minor instead of just major because ABI and API change between minor versions.
llvm-svn: 230981
Dmitry Vyukov [Mon, 2 Mar 2015 17:45:18 +0000 (17:45 +0000)]
asan: fix windows build after commit 230978
llvm-svn: 230980
Paul Robinson [Mon, 2 Mar 2015 17:44:52 +0000 (17:44 +0000)]
[PS4] Correct relocation for DWARF TLS references.
llvm-svn: 230979
Dmitry Vyukov [Mon, 2 Mar 2015 17:36:02 +0000 (17:36 +0000)]
asan: fix signal handling during stoptheworld
The problem is that without SA_RESTORER flag, kernel ignores the handler. So tracer actually did not setup any handler.
Add SA_RESTORER flag when setting up handlers.
Add a test that causes SIGSEGV in stoptheworld callback.
Move SignalContext from asan to sanitizer_common to print better diagnostics about signal in the tracer thread.
http://reviews.llvm.org/D8005
llvm-svn: 230978
Hans Wennborg [Mon, 2 Mar 2015 17:30:42 +0000 (17:30 +0000)]
export.sh: Make sure umask is not too restrictive (PR22742)
llvm-svn: 230977
Justin Bogner [Mon, 2 Mar 2015 17:26:43 +0000 (17:26 +0000)]
Detect malformed YAML sequence in yaml::Input::beginSequence()
When reading a yaml::SequenceTraits object, YAMLIO does not report an
error if the yaml item is not a sequence. Instead, YAMLIO reads an
empty sequence. For example:
---
seq:
foo: 1
bar: 2
...
If `seq` is a SequenceTraits object, then reading the above yaml will
yield `seq` as an empty sequence.
Fix this to report an error for the above mapping ("not a sequence")
Patch by William Fisher. Thanks!
llvm-svn: 230976
Adrian Prantl [Mon, 2 Mar 2015 17:21:06 +0000 (17:21 +0000)]
Refactor DebugLocDWARFExpression so it doesn't require access to the
TargetRegisterInfo. DebugLocEntry now holds a buffer with the raw bytes
of the pre-calculated DWARF expression.
Ought to be NFC, but it does slightly alter the output format of the
textual assembly.
This reapplies 230930 with a relaxed assertion in DebugLocEntry::finalize()
that allows for empty DWARF expressions for constant FP values.
llvm-svn: 230975
Sergey Matveev [Mon, 2 Mar 2015 17:13:38 +0000 (17:13 +0000)]
[asan] Demote "trying to poison/unpoison memory region" messages...
... to verbosity level 3. Because log spam.
llvm-svn: 230974
Benjamin Kramer [Mon, 2 Mar 2015 16:45:08 +0000 (16:45 +0000)]
Accidentaly inverted the condition again. Sorry.
llvm-svn: 230973
Benjamin Kramer [Mon, 2 Mar 2015 16:42:56 +0000 (16:42 +0000)]
Avoid assertion in MSVC 2013 debug builds.
llvm-svn: 230972
Benjamin Kramer [Mon, 2 Mar 2015 16:09:24 +0000 (16:09 +0000)]
CodeGen: Fix passing of classes with only one AVX vector member in AVX registers
isSingleElementStruct was a bit too tight in its definition of struct
so we got a mismatch between classify() and the actual code generation.
To make matters worse the code in GetByteVectorType still defaulted to
<2 x double> if it encountered a type it didn't know, making this a
silent miscompilation (PR22753).
Completely remove the "preferred type" stuff from GetByteVectorType and
make it fail an assertion if someone tries to use it with a type not
suitable for a vector register.
llvm-svn: 230971
Tom Stellard [Mon, 2 Mar 2015 15:29:41 +0000 (15:29 +0000)]
Implement step builtin
This has been tested with piglit, OpenCV, and the ocl conformance tests.
llvm-svn: 230970
Tom Stellard [Mon, 2 Mar 2015 15:29:39 +0000 (15:29 +0000)]
Implement smoothstep builtin v2
This has been tested with piglit, OpenCV, and the ocl conformance tests.
v2:
- Fix typo in smoothstep.h
llvm-svn: 230969
Tom Stellard [Mon, 2 Mar 2015 15:29:37 +0000 (15:29 +0000)]
Implement radians builtin v2
This has been tested with piglit, OpenCV, and the ocl conformance tests.
v2:
- Move to the common/ directory
llvm-svn: 230968
Tom Stellard [Mon, 2 Mar 2015 15:29:35 +0000 (15:29 +0000)]
Implement degrees builtin v2
This has been tested with piglit, OpenCV, and the ocl conformance tests.
v2:
- Move to the common/ directory
llvm-svn: 230967
Benjamin Kramer [Mon, 2 Mar 2015 15:24:41 +0000 (15:24 +0000)]
AsmWriter: Only print one space after the load type
Before: %x = load i32, i32* %i
After: %x = load i32, i32* %i
Purely cosmetic, so no new test case.
llvm-svn: 230966
Benjamin Kramer [Mon, 2 Mar 2015 15:24:36 +0000 (15:24 +0000)]
SLPVectorizer: Rewrite ArrayRef slice compare to be more idiomatic.
NFC intended.
llvm-svn: 230965
Benjamin Kramer [Mon, 2 Mar 2015 15:24:30 +0000 (15:24 +0000)]
TblGen: Remove copy of SmallVector::operator==. NFC intended.
llvm-svn: 230964
Robert Flack [Mon, 2 Mar 2015 15:14:50 +0000 (15:14 +0000)]
Reduce the number of components initialized by lldb-server to reduce static binary size.
Separate out the necessary component initialization for lldb-server such that the linker can greatly reduce the binary size. With this patch the size of lldb-server on my 64 bit linux release build drops from 46MB to 26MB.
Differential Revision: http://reviews.llvm.org/D7880
llvm-svn: 230963
Elena Demikhovsky [Mon, 2 Mar 2015 15:00:34 +0000 (15:00 +0000)]
AVX-512: Add assembly parser support for Rounding mode
By Asaf Badouh <asaf.badouh@intel.com>
llvm-svn: 230962
Timur Iskhodzhanov [Mon, 2 Mar 2015 14:49:44 +0000 (14:49 +0000)]
[ASan/Win] Rename a test.
I came up with the original name trying to test a different issue and forgot to rename afterwards
llvm-svn: 230961
Johannes Doerfert [Mon, 2 Mar 2015 14:06:01 +0000 (14:06 +0000)]
Do not model scalar accesses in non-affine subregions
If a scalar was defined and used only in a non-affine subregion we do
not need to model the accesses. However, if the scalar was defined
inside the region and escapes the region we have to model the access.
The same is true if the scalar was defined outside and used inside the
region.
llvm-svn: 230960
Johannes Doerfert [Mon, 2 Mar 2015 13:41:53 +0000 (13:41 +0000)]
[FIX] Make parallel codegen aware of region statements
llvm-svn: 230959
Pavel Labath [Mon, 2 Mar 2015 13:39:39 +0000 (13:39 +0000)]
Fix build breakage on win7-msvc caused by r230955
llvm-svn: 230958
Benjamin Kramer [Mon, 2 Mar 2015 13:16:28 +0000 (13:16 +0000)]
NVPTX: Remove dead code.
Fun fact: This file was never referenced since the initial checkin of
the NVPTX backend.
llvm-svn: 230957
Vasileios Kalintiris [Mon, 2 Mar 2015 12:47:32 +0000 (12:47 +0000)]
[mips] Optimize conditional moves where RHS is zero.
Summary:
When the RHS of a conditional move node is zero, we can utilize the $zero
register by inverting the conditional move instruction and by swapping the
order of its True/False operands.
Reviewers: dsanders
Differential Revision: http://reviews.llvm.org/D7945
llvm-svn: 230956
Pavel Labath [Mon, 2 Mar 2015 12:46:22 +0000 (12:46 +0000)]
Fix handling of backslashes in Args parsing
Summary:
Presently Args::SetCommandString allows quotes to be escaped with backslash. However, the
backslash itself is not removed from the argument, nor there is a way to escape the backslash
itself. This leads to surprising results:
"a b" c" -> 'a b', 'c' # Here we actually have an unterminated quote, but that is ignored
"a b\" c" -> 'a b\" c' # We try to escape the quote. That works but the backslash is not removed.
"a b\\" c" -> 'a b\\" c' # Escaping the backslash has no effect.
This change changes quote handling to be more shell-like:
- single quotes and backquotes are literal and there is no way to escape the closing quote or
anything else inside;
- inside double quotes you can use backslash to escape the closing quote and another backslash
- outside any quotes, you can use backslash to escape quotes, spaces and itself.
This makes the parsing more consistent with what the user is familiar and increases the
probability that pasting the command line from shell to the "process launch" command "just work".
Reviewers: clayborg
Subscribers: lldb-commits
Differential Revision: http://reviews.llvm.org/D7855
llvm-svn: 230955
Elena Demikhovsky [Mon, 2 Mar 2015 12:46:21 +0000 (12:46 +0000)]
AVX-512: Simplified MOV patterns, no functional changes.
llvm-svn: 230954
Andrew Wilkins [Mon, 2 Mar 2015 12:42:45 +0000 (12:42 +0000)]
[llgo] debug: create replaceable type through DIBuilder
Summary:
llgo was asserting in DebugInfo, which was interpreting
the temporary MDNodes we were creating as DIScopes instead
of DITypes (in DIScope::getRef).
This proposal changes llgo to use DIBuilder's
createReplaceableCompositeType method to create a DIType
that can be RAUW'd.
Reviewers: pcc
Reviewed By: pcc
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D7852
llvm-svn: 230953
Alexander Kornienko [Mon, 2 Mar 2015 12:39:18 +0000 (12:39 +0000)]
[clang-tidy] Refactor: Move llvm clang-tidy checks to namespace clang::tidy::llvm
clang-tidy checks are organized into modules. This refactoring moves the llvm
module checks to clang::tidy::llvm
http://reviews.llvm.org/D7995
Patch by Richard Thomson!
llvm-svn: 230952
Andrew Wilkins [Mon, 2 Mar 2015 12:27:04 +0000 (12:27 +0000)]
bindings/go: expose DIBuilder::createReplaceableCompositeType
Summary:
We extend the DIBuilder type, adding a method for creating
replaceable composite types. This is necessary for creating
debug info describing self-referential types.
Reviewers: pcc
Reviewed By: pcc
Subscribers: axw, llvm-commits
Differential Revision: http://reviews.llvm.org/D7851
llvm-svn: 230951
Alexander Kornienko [Mon, 2 Mar 2015 12:25:03 +0000 (12:25 +0000)]
[clang-tidy] Refactor: Move misc clang-tidy checks to namespace clang::tidy::misc
clang-tidy checks are organized into modules. This refactoring moves the misc
module checks into the namespace clang::tidy::misc
http://reviews.llvm.org/D7996
Patch by Richard Thomson!
llvm-svn: 230950