platform/upstream/llvm.git
6 years agoHandle zero-length debug directory entries.
Nico Weber [Wed, 5 Sep 2018 18:01:04 +0000 (18:01 +0000)]
Handle zero-length debug directory entries.

Part of https://reviews.llvm.org/D51652 (tests will be in the lld repo)

llvm-svn: 341485

6 years agoTest Commit for git-svn-cleanup comment.
Erich Keane [Wed, 5 Sep 2018 17:14:21 +0000 (17:14 +0000)]
Test Commit for git-svn-cleanup comment.

Removes the class name for the Expr class, which isn't necessary.

llvm-svn: 341484

6 years ago[OPENMP][NVPTX] Disable runtime-type info for CUDA devices.
Alexey Bataev [Wed, 5 Sep 2018 17:10:30 +0000 (17:10 +0000)]
[OPENMP][NVPTX] Disable runtime-type info for CUDA devices.

RTTI is not supported by the NVPTX target.

llvm-svn: 341483

6 years agoHold GIL while allocating memory for PythonString.
Tatyana Krasnukha [Wed, 5 Sep 2018 17:07:29 +0000 (17:07 +0000)]
Hold GIL while allocating memory for PythonString.

Summary:
Swig wraps C++ code into SWIG_PYTHON_THREAD_BEGIN_ALLOW; ...  SWIG_PYTHON_THREAD_END_ALLOW;
Thus, LLDB crashes with "Fatal Python error: Python memory allocator called without holding the GIL" when calls an lldb_SB***___str__ function.

Reviewers: clayborg

Reviewed By: clayborg

Differential Revision: https://reviews.llvm.org/D51569

llvm-svn: 341482

6 years ago[DAGCombiner] try to convert pow(x, 0.25) to sqrt(sqrt(x))
Sanjay Patel [Wed, 5 Sep 2018 17:01:56 +0000 (17:01 +0000)]
[DAGCombiner] try to convert pow(x, 0.25) to sqrt(sqrt(x))

This was proposed as an IR transform in D49306, but it was not clearly justifiable as a canonicalization.
Here, we only do the transform when the target tells us that sqrt can be lowered with inline code.

This is the basic case. Some potential enhancements are in the TODO comments:

1. Generalize the transform for other exponents (allow more than 2 sqrt calcs if that's really cheaper).
2. If we have less fast-math-flags, generate code to avoid -0.0 and/or INF.
3. Allow the transform when optimizing/minimizing size (might require a target hook to get that right).

Note that by default, x86 converts single-precision sqrt calcs into sqrt reciprocal estimate with
refinement. That codegen is controlled by CPU attributes and can be manually overridden. We have plenty
of test coverage for that already, so I didn't bother to include extra testing for that here. AArch uses
its full-precision ops in all cases (not sure if that's the intended behavior or not, but that should
also be covered by existing tests).

Differential Revision: https://reviews.llvm.org/D51630

llvm-svn: 341481

6 years ago[hwasan] deflake a test
Kostya Serebryany [Wed, 5 Sep 2018 16:09:53 +0000 (16:09 +0000)]
[hwasan] deflake a test

llvm-svn: 341480

6 years ago[XRay] Use a function to return a constant
Dean Michael Berris [Wed, 5 Sep 2018 16:03:21 +0000 (16:03 +0000)]
[XRay] Use a function to return a constant

We do this instead of using static constexpr char arrays because MSVC
2015 cannot handle the constant initialisation of those along with the
out-of-line storage declaration.

This is a follow-up to D51672.

llvm-svn: 341479

6 years ago[Hexagon] Ignore unnamed globals in HexagonConstExtenders
Krzysztof Parzyszek [Wed, 5 Sep 2018 15:54:44 +0000 (15:54 +0000)]
[Hexagon] Ignore unnamed globals in HexagonConstExtenders

This replaces r337723. The global list in the module can be huge with LTO,
plus the module can change between different invocations of the pass, so
there is no easy way to deterministically cache the ordering (especially
in the presence of multiple threads).

llvm-svn: 341478

6 years ago[XRay] Use `const` instead of `constexpr` in storage declaration
Dean Michael Berris [Wed, 5 Sep 2018 15:19:43 +0000 (15:19 +0000)]
[XRay] Use `const` instead of `constexpr` in storage declaration

This should make MSVC happy with the storage provided for static
constexpr character arrays out-of-line.

Follow-up to D51672.

llvm-svn: 341477

6 years ago[Sema] Store MacroInfo in CodeCompletionResult for macro results.
Eric Liu [Wed, 5 Sep 2018 14:59:17 +0000 (14:59 +0000)]
[Sema] Store MacroInfo in CodeCompletionResult for macro results.

Summary:
This provides information about the macro definition. For example, it
can be used to compute macro USRs.

Reviewers: sammccall

Subscribers: cfe-commits

Differential Revision: https://reviews.llvm.org/D51675

llvm-svn: 341476

6 years agoFix arm_neon.h and arm_fp16.h generation for compiling with std=c89
Diogo N. Sampaio [Wed, 5 Sep 2018 14:56:21 +0000 (14:56 +0000)]
Fix arm_neon.h and arm_fp16.h generation for compiling with std=c89

Summary:
The inline attribute is not valid for C standard 89. Replace the argument in the generation of header files with __inline, as well adding tests for both header files.

Reviewers: pbarrio, SjoerdMeijer, javed.absar, t.p.northover

Subscribers: t.p.northover, kristof.beyls, chrib, cfe-commits

Differential Revision: https://reviews.llvm.org/D51683

test/Headers/arm-fp16-header.c
test/Headers/arm-neon-header.c
utils/TableGen/NeonEmitter.cpp

llvm-svn: 341475

6 years agoEnable symbolize-deadlock.test on Windows
Max Moroz [Wed, 5 Sep 2018 14:43:38 +0000 (14:43 +0000)]
Enable symbolize-deadlock.test on Windows

Summary:
Remove unneeded #include of <unistd.h> from SymbolizeDeadlock.cpp to
enable compilation on Windows and enable symbolize-deadlock.test on
Windows.

Patch by Jonathan Metzman (@metzman).

Reviewers: Dor1s

Reviewed By: Dor1s

Subscribers: kcc, delcypher, llvm-commits, #sanitizers

Differential Revision: https://reviews.llvm.org/D51685

llvm-svn: 341473

6 years agoEnable DWARF accelerator tables by default when tuning for lldb (-glldb => -gpubnames)
Pavel Labath [Wed, 5 Sep 2018 14:38:44 +0000 (14:38 +0000)]
Enable DWARF accelerator tables by default when tuning for lldb (-glldb => -gpubnames)

Summary:
DWARF v5 accelerator tables provide a considerable performance
improvement for lldb and will make the default -glldb behavior same on
all targets (right now we emit apple tables on apple targets, but these
are not controlled by -gpubnames, only by -glldb).

Reviewers: dblaikie

Subscribers: probinson, clayborg, JDevlieghere, aprantl, cfe-commits

Differential Revision: https://reviews.llvm.org/D51576

llvm-svn: 341472

6 years ago[www] Updated outdated download links [NFC]
Raphael Isemann [Wed, 5 Sep 2018 14:27:48 +0000 (14:27 +0000)]
[www] Updated outdated download links [NFC]

Also added ubuntu and Arch Linux links.

llvm-svn: 341471

6 years ago[benchmark] Fix 32-bit build failure
Kirill Bobyrev [Wed, 5 Sep 2018 14:15:39 +0000 (14:15 +0000)]
[benchmark] Fix 32-bit build failure

This patch applies upstream commit:
https://github.com/google/benchmark/commit/f0901417c89d123474e6b91365029cfe32cf89dc

Tim Northover pointed out that benchmark build might be broken on 32-bit
macOS. This commit by Roman Lebedev (lebedev.ri) resolves the issue.

Reviewed By: lebedev.ri

Differential Revision: https://reviews.llvm.org/D51677

llvm-svn: 341469

6 years ago[ControlHeightReduction] Remove unused includes
Benjamin Kramer [Wed, 5 Sep 2018 13:51:05 +0000 (13:51 +0000)]
[ControlHeightReduction] Remove unused includes

Also clang-format them.

llvm-svn: 341468

6 years ago[clangd] Fix type/variable name conflict on some compilers
Sam McCall [Wed, 5 Sep 2018 13:22:11 +0000 (13:22 +0000)]
[clangd] Fix type/variable name conflict on some compilers

llvm-svn: 341467

6 years ago[clangd] Fix references.test assertions
Sam McCall [Wed, 5 Sep 2018 13:17:51 +0000 (13:17 +0000)]
[clangd] Fix references.test assertions

llvm-svn: 341466

6 years ago[clangd] make zlib compression optional for binary format
Sam McCall [Wed, 5 Sep 2018 13:17:47 +0000 (13:17 +0000)]
[clangd] make zlib compression optional for binary format

llvm-svn: 341465

6 years ago[llvm-strip] Support stripping multiple input files
Jordan Rupprecht [Wed, 5 Sep 2018 13:10:03 +0000 (13:10 +0000)]
[llvm-strip] Support stripping multiple input files

Summary:
Allow strip to be called on multiple input files, which is interpreted as stripping N files in place. Using multiple input files is incompatible with -o.

To allow this, create a `DriverConfig` struct which just wraps a list of `CopyConfigs`. objcopy will only ever have a single `CopyConfig`, but strip will have N (where N >= 1) CopyConfigs.

Reviewers: alexshap, jakehehrlich

Reviewed By: alexshap, jakehehrlich

Subscribers: MaskRay, jakehehrlich, llvm-commits

Differential Revision: https://reviews.llvm.org/D51660

llvm-svn: 341464

6 years ago[clangd] Sort GoToDefinition results.
Haojian Wu [Wed, 5 Sep 2018 12:00:15 +0000 (12:00 +0000)]
[clangd] Sort GoToDefinition results.

Summary:
GoToDefinition returns all declaration results (implicit/explicit) that are
in the same location, and the results are returned in arbitrary order.

Some LSP clients defaultly take the first result as the final result, which
might present a bad result (implicit decl) to users.

This patch ranks the result based on whether the declarations are
referenced explicitly/implicitly. We put explicit declarations first.

This also improves the "hover" (which just take the first result) feature
in some cases.

Reviewers: ilya-biryukov

Subscribers: kadircet, ioeric, MaskRay, jkorous, mgrang, arphaman, cfe-commits

Differential Revision: https://reviews.llvm.org/D50438

llvm-svn: 341463

6 years ago[clangd] Add xrefs LSP boilerplate implementation.
Sam McCall [Wed, 5 Sep 2018 11:53:07 +0000 (11:53 +0000)]
[clangd] Add xrefs LSP boilerplate implementation.

Reviewers: ilya-biryukov, ioeric

Subscribers: MaskRay, jkorous, arphaman, cfe-commits

Differential Revision: https://reviews.llvm.org/D50896

llvm-svn: 341462

6 years ago[Aggressive InstCombine] Move C bindings to their own header file.
Benjamin Kramer [Wed, 5 Sep 2018 11:41:12 +0000 (11:41 +0000)]
[Aggressive InstCombine] Move C bindings to their own header file.

llvm-svn: 341461

6 years agoFix -Wdocumentation warning. NFCI.
Simon Pilgrim [Wed, 5 Sep 2018 10:44:03 +0000 (10:44 +0000)]
Fix -Wdocumentation warning. NFCI.

llvm-svn: 341460

6 years ago[clangd] Avoid enum class+enumValN to avoid GCC bug(?), and use consistent style.
Sam McCall [Wed, 5 Sep 2018 10:39:58 +0000 (10:39 +0000)]
[clangd] Avoid enum class+enumValN to avoid GCC bug(?), and use consistent style.

llvm-svn: 341459

6 years ago[clangd] Implement findReferences function
Sam McCall [Wed, 5 Sep 2018 10:33:36 +0000 (10:33 +0000)]
[clangd] Implement findReferences function

clangd will use findReferences to provide LSP's reference feature.

llvm-svn: 341458

6 years ago[DebugInfo] Normalize common kinds of DWARF sub-expressions.
Jonas Devlieghere [Wed, 5 Sep 2018 10:18:36 +0000 (10:18 +0000)]
[DebugInfo] Normalize common kinds of DWARF sub-expressions.

Normalize common kinds of DWARF sub-expressions to make debug info
encoding a bit more compact:

  DW_OP_constu [X < 32] -> DW_OP_litX
  DW_OP_constu [all ones] -> DW_OP_lit0, DW_OP_not (64-bit only)

Differential revision: https://reviews.llvm.org/D51640

llvm-svn: 341457

6 years ago[NFC] Add test on full IV widening
Max Kazantsev [Wed, 5 Sep 2018 10:10:59 +0000 (10:10 +0000)]
[NFC] Add test on full IV widening

llvm-svn: 341456

6 years ago[VFS] Cache the current working directory for the real FS.
Eric Liu [Wed, 5 Sep 2018 09:45:27 +0000 (09:45 +0000)]
[VFS] Cache the current working directory for the real FS.

Reviewers: sammccall

Subscribers: cfe-commits

Differential Revision: https://reviews.llvm.org/D51641

llvm-svn: 341455

6 years agoRemove FrameAccess struct from hasLoadFromStackSlot
Sander de Smalen [Wed, 5 Sep 2018 08:59:50 +0000 (08:59 +0000)]
Remove FrameAccess struct from hasLoadFromStackSlot

This removes the FrameAccess struct that was added to the interface
in D51537, since the PseudoValue from the MachineMemoryOperand
can be safely casted to a FixedStackPseudoSourceValue.

Reviewers: MatzeB, thegameg, javed.absar

Reviewed By: thegameg

Differential Revision: https://reviews.llvm.org/D51617

llvm-svn: 341454

6 years ago[clangd] Fix typo. NFC
Fangrui Song [Wed, 5 Sep 2018 08:01:37 +0000 (08:01 +0000)]
[clangd] Fix typo. NFC

llvm-svn: 341452

6 years ago[clangd] Fix buildbot failures on older compilers from r341375
Sam McCall [Wed, 5 Sep 2018 07:52:49 +0000 (07:52 +0000)]
[clangd] Fix buildbot failures on older compilers from r341375

llvm-svn: 341451

6 years agoclang-format: Fix formatting C++ namespaces with preceding 'inline' or 'export' specifier
Sam McCall [Wed, 5 Sep 2018 07:44:02 +0000 (07:44 +0000)]
clang-format: Fix formatting C++ namespaces with preceding 'inline' or 'export' specifier

This fixes formatting namespaces with preceding 'inline' and 'export' (Modules TS) specifiers.

This change fixes namespaces not being identified as such with preceding 'inline' or 'export' specifiers.

Motivation: I was experimenting with the Modules TS (-fmodules-ts) and found it would be useful if clang-format would correctly format 'export namespace'. While making the changes, I noticed that similar issues still exist with 'inline namespace', and addressed them as well.

Patch by Marco Elver!

Reviewers: klimek, djasper, owenpan, sammccall

Reviewed By: owenpan, sammccall

Subscribers: owenpan, cfe-commits

Differential Revision: https://reviews.llvm.org/D51036

llvm-svn: 341450

6 years ago[clangd] Tune macro quality scoring for code completion.
Eric Liu [Wed, 5 Sep 2018 07:40:38 +0000 (07:40 +0000)]
[clangd] Tune macro quality scoring for code completion.

x0.2 seems to be too much penalty, macros might be wanted in some cases;
changing to 0.5x instead. The tuning didn't affect ranking for non-macro
completions.

llvm-svn: 341449

6 years ago[libomptaret][test] Announce compiler features
Jonas Hahnfeld [Wed, 5 Sep 2018 07:26:00 +0000 (07:26 +0000)]
[libomptaret][test] Announce compiler features

This is a follow-up to r341371: The new test for PR38704 doesn't
work with Clang 6.0. It uses an UNSUPPORTED: clang-6, but that
hasn't worked because the compiler features weren't known to lit.

llvm-svn: 341448

6 years ago[XRay] Add a RecordPrinter visitor for FDR Records
Dean Michael Berris [Wed, 5 Sep 2018 06:57:23 +0000 (06:57 +0000)]
[XRay] Add a RecordPrinter visitor for FDR Records

Summary:
This change adds a `RecordPrinter` type which does some basic text
serialization of the FDR record instances. This is one component of the
tool we're building to dump the records from an FDR mode log as-is.

This is a small part of D50441.

Reviewers: eizan, kpw

Subscribers: mgorny, hiraditya, llvm-commits

Differential Revision: https://reviews.llvm.org/D51672

llvm-svn: 341447

6 years ago[DebugInfo] Fix bug in LiveDebugVariables.
Hsiangkai Wang [Wed, 5 Sep 2018 05:58:53 +0000 (05:58 +0000)]
[DebugInfo] Fix bug in LiveDebugVariables.

In lib/CodeGen/LiveDebugVariables.cpp, it uses std::prev(MBBI) to
get DebugValue's SlotIndex. However, the previous instruction may be
also a debug instruction. It could not use a debug instruction to query
SlotIndex in mi2iMap.

Scan all debug instructions and use the first debug instruction to query
SlotIndex for following debug instructions. Only handle DBG_VALUE in
handleDebugValue().

Differential Revision: https://reviews.llvm.org/D50621

llvm-svn: 341446

6 years ago[Xray] Darwin providing defined value for weak symbols to fix linkage issue
David Carlier [Wed, 5 Sep 2018 05:18:34 +0000 (05:18 +0000)]
[Xray] Darwin providing defined value for weak symbols to fix linkage issue

- Temporary hack to make the buildbot failure stop on Darwin.

llvm-svn: 341445

6 years agoPrevent unsigned overflow.
Richard Trieu [Wed, 5 Sep 2018 04:19:15 +0000 (04:19 +0000)]
Prevent unsigned overflow.

The sum of the weights is caculated in an APInt, which has a width smaller than
64.  In certain cases, the sum of the widths would overflow when calculations
are done inside an APInt, but would not if done with uint64_t.  Since the
values will be passed as uint64_t in the function call anyways, do all the math
in 64 bits.  Also added an assert in case the probabilities overflow 64 bits.

llvm-svn: 341444

6 years agoFix -Wunused-function in release build after rL341386
Fangrui Song [Wed, 5 Sep 2018 03:10:20 +0000 (03:10 +0000)]
Fix -Wunused-function in release build after rL341386

llvm-svn: 341443

6 years agoAdd glibc_prereq to platform limits mmsghdr
Brian Cain [Wed, 5 Sep 2018 02:15:54 +0000 (02:15 +0000)]
Add glibc_prereq to platform limits mmsghdr

sendmmsg requires glibc >= 2.14.

Fixes PR38589.

Review: https://reviews.llvm.org/D51538
llvm-svn: 341442

6 years ago[hwasan] revert r341435 as it breaks the bot on aarch64
Kostya Serebryany [Wed, 5 Sep 2018 01:29:08 +0000 (01:29 +0000)]
[hwasan] revert r341435 as it breaks the bot on aarch64

llvm-svn: 341441

6 years ago[hwasan] fix colored output
Kostya Serebryany [Wed, 5 Sep 2018 01:27:48 +0000 (01:27 +0000)]
[hwasan] fix colored output

llvm-svn: 341440

6 years ago[WebAssembly] clang-format (NFC)
Heejin Ahn [Wed, 5 Sep 2018 01:27:38 +0000 (01:27 +0000)]
[WebAssembly] clang-format (NFC)

Summary: This patch runs clang-format on all wasm-only files.

Reviewers: aardappel, dschuff, sunfish, tlively

Subscribers: MatzeB, sbc100, jgravelle-google, llvm-commits

Differential Revision: https://reviews.llvm.org/D51447

llvm-svn: 341439

6 years ago[hwasan] print thread IDs when reporting a bug (also had to fix pthread_create on...
Kostya Serebryany [Wed, 5 Sep 2018 01:16:50 +0000 (01:16 +0000)]
[hwasan] print thread IDs when reporting a bug (also had to fix pthread_create on Linux)

llvm-svn: 341438

6 years agoAllow all supportable non-type attributes to be used with #pragma clang attribute.
Richard Smith [Wed, 5 Sep 2018 00:28:57 +0000 (00:28 +0000)]
Allow all supportable non-type attributes to be used with #pragma clang attribute.

Summary:
We previously disallowed use of undocumented attributes with #pragma clang
attribute, but the justification for doing so was weak and it prevented many
reasonable use cases.

Reviewers: aaron.ballman, arphaman

Subscribers: cfe-commits, rnk, benlangmuir, dexonsmith, erik.pilkington

Differential Revision: https://reviews.llvm.org/D51507

llvm-svn: 341437

6 years agoFix -Wunused-private-variable on non-Windows
Reid Kleckner [Wed, 5 Sep 2018 00:18:05 +0000 (00:18 +0000)]
Fix -Wunused-private-variable on non-Windows

I didn't want to use ifdefs in headers, but I'll do it to pacify a
warning.

llvm-svn: 341436

6 years ago[hwasan] use real TLS on linux to store the current thread -- this way we can call...
Kostya Serebryany [Wed, 5 Sep 2018 00:17:23 +0000 (00:17 +0000)]
[hwasan] use real TLS on linux to store the current thread -- this way we can call t->Destroy in __hwasan_thread_exit, same as on Android

llvm-svn: 341435

6 years ago[Windows] Convert from UTF-8 to UTF-16 when writing to a Windows console
Reid Kleckner [Wed, 5 Sep 2018 00:08:56 +0000 (00:08 +0000)]
[Windows] Convert from UTF-8 to UTF-16 when writing to a Windows console

Summary:
Calling WriteConsoleW is the most reliable way to print Unicode
characters to a Windows console.

If binary data gets printed to the console, attempting to re-encode it
shouldn't be a problem, since garbage in can produce garbage out.

This breaks printing strings in the local codepage, which WriteConsoleA
knows how to handle. For example, this can happen when user source code
is encoded with the local codepage, and an LLVM tool quotes it while
emitting a caret diagnostic. This is unfortunate, but well-behaved tools
should validate that their input is UTF-8 and escape non-UTF-8
characters before sending them to raw_fd_ostream. Clang already does
this, but not all LLVM tools do this.

One drawback to the current implementation is printing a string a byte
at a time doesn't work. Consider this LLVM code:
  for (char C : MyStr) outs() << C;

Because outs() is now unbuffered, we wil try to convert each byte to
UTF-16, which will fail. However, this already didn't work, so I think
we may as well update callers that do that as we find them to print
complete portions of strings. You can see a real example of this in my
patch to SourceMgr.cpp

Fixes PR38669 and PR36267.

Reviewers: zturner, efriedma

Subscribers: llvm-commits, hiraditya

Differential Revision: https://reviews.llvm.org/D51558

llvm-svn: 341433

6 years ago[hwasan] simplify the code, NFC
Kostya Serebryany [Wed, 5 Sep 2018 00:01:45 +0000 (00:01 +0000)]
[hwasan] simplify the code, NFC

llvm-svn: 341432

6 years ago[hwasan] remove stale Thread:: data members. While doing so noticed that GetThreadSta...
Kostya Serebryany [Tue, 4 Sep 2018 23:57:09 +0000 (23:57 +0000)]
[hwasan] remove stale Thread:: data members. While doing so noticed that GetThreadStackAndTls was always called with 'at_initialization=true', fixed that.

llvm-svn: 341431

6 years ago[hwasan] add a unique id to a thread and add debug prints for thread creation/destruction
Kostya Serebryany [Tue, 4 Sep 2018 23:26:08 +0000 (23:26 +0000)]
[hwasan] add a unique id to a thread and add debug prints for thread creation/destruction

llvm-svn: 341428

6 years ago[InstCombine] fix xor-or-xor fold to check uses and handle commutes
Sanjay Patel [Tue, 4 Sep 2018 23:22:13 +0000 (23:22 +0000)]
[InstCombine] fix xor-or-xor fold to check uses and handle commutes

I'm probably missing some way to use m_Deferred to remove the code
duplication, but that can be a follow-up.

The improvement in demand_shrink_nsw.ll is an example of missing
the fold because the pattern matching was deficient. I didn't try
to follow the bits in that test, but Alive says it's correct:
https://rise4fun.com/Alive/ugc

llvm-svn: 341426

6 years agoChange TestCompletion to only ever look inside of BaseDir
Frederic Riss [Tue, 4 Sep 2018 23:09:49 +0000 (23:09 +0000)]
Change TestCompletion to only ever look inside of BaseDir

TestCompletion was failing quite frequently on our Linux bots. Some tracing
revealed that when we are iterating BaseDir we are not getting all the entries.
More specifically, we are sometimes missing the entry corresponding to the
TestCompletion directory that the first test in DirCompletionAbsolute is
looking for. BaseDir is the directory where lit is creating all the temporary
files. The semantics of opendir/readdir are unclear when it comes to iterating
over a directory that changes contents, but it seems like on Linux you might
fail to list an entry even if it was there before opendir and is still present
throughout the iteration. Changing the test to only look inside of the test-
specific directory seems to fix the instability.

This commit also removes some assertions that were added to try to track down
this issue.

llvm-svn: 341425

6 years ago[InstCombine] update tests checks; NFC
Sanjay Patel [Tue, 4 Sep 2018 23:08:23 +0000 (23:08 +0000)]
[InstCombine] update tests checks; NFC

llvm-svn: 341424

6 years ago[WebAssembly] Fixed stale assert message in WebAssemblyMCInstLower
Wouter van Oortmerssen [Tue, 4 Sep 2018 22:59:05 +0000 (22:59 +0000)]
[WebAssembly] Fixed stale assert message in WebAssemblyMCInstLower

llvm-svn: 341423

6 years ago[ThinLTO] Fix memory corruption in ThinLTOCodeGenerator when CodeGenOnly was specified
Steven Wu [Tue, 4 Sep 2018 22:54:17 +0000 (22:54 +0000)]
[ThinLTO] Fix memory corruption in ThinLTOCodeGenerator when CodeGenOnly was specified

Summary:
Issue occurs when doing ThinLTO with CodeGenOnly flag.
TMBuilder.TheTriple is assigned to by multiple threads in an unsafe way resulting in double-free of std::string memory.

Pseudocode:
if (CodeGenOnly) {
  // Perform only parallel codegen and return.
  ThreadPool Pool;
  int count = 0;
  for (auto &ModuleBuffer : Modules) {
    Pool.async([&](int count) {
    ...
      /// Now call OutputBuffer = codegen(*TheModule);
      /// Which turns into initTMBuilder(moduleTMBuilder, Triple(TheModule.getTargetTriple()));
      /// Which turns into

      TMBuilder.TheTriple = std::move(TheTriple);   // std::string = "....."
      /// So, basically std::string assignment to same string on multiple threads = memory corruption

  }

  return;
}

Patch by Alex Borcan

Reviewers: llvm-commits, steven_wu

Reviewed By: steven_wu

Subscribers: mehdi_amini, inglorion, eraman, steven_wu, dexonsmith, llvm-commits

Differential Revision: https://reviews.llvm.org/D51651

llvm-svn: 341422

6 years ago[ODRHash] Extend hash to support all Type's.
Richard Trieu [Tue, 4 Sep 2018 22:53:19 +0000 (22:53 +0000)]
[ODRHash] Extend hash to support all Type's.

llvm-svn: 341421

6 years ago[sanitizer] optimize internal_memset for the most performance critical case (16-byte...
Kostya Serebryany [Tue, 4 Sep 2018 22:43:30 +0000 (22:43 +0000)]
[sanitizer] optimize internal_memset for the most performance critical case (16-byte-aligned)

llvm-svn: 341420

6 years ago[llvm-strip] Allow copying relocation sections without symbol tables.
Jordan Rupprecht [Tue, 4 Sep 2018 22:28:49 +0000 (22:28 +0000)]
[llvm-strip] Allow copying relocation sections without symbol tables.

Summary:
Fixes the error "Link field value 0 in section .rela.plt is invalid" when copying/stripping certain binaries. Minimal repro:

```
$ cat /tmp/a.c
int main() { return 0; }
$ clang -static /tmp/a.c -o /tmp/a
$ llvm-strip /tmp/a -o /tmp/b
llvm-strip: error: Link field value 0 in section .rela.plt is invalid.
```

Reviewers: jakehehrlich, alexshap

Subscribers: llvm-commits

Differential Revision: https://reviews.llvm.org/D51493

llvm-svn: 341419

6 years agoRevert r341373, since it fails on some targets.
Tim Shen [Tue, 4 Sep 2018 22:20:11 +0000 (22:20 +0000)]
Revert r341373, since it fails on some targets.

Differential Revision: https://reviews.llvm.org/D51354

llvm-svn: 341418

6 years agoRevert "Revert r341269: [Constant Hoisting] Hoisting Constant GEP Expressions"
Zhaoshi Zheng [Tue, 4 Sep 2018 22:17:03 +0000 (22:17 +0000)]
Revert "Revert r341269: [Constant Hoisting] Hoisting Constant GEP Expressions"

Reland r341269. Use std::stable_sort when sorting constant condidates.

Reverting commit, r341365:

  Revert r341269: [Constant Hoisting] Hoisting Constant GEP Expressions

  One of the tests is failing 50% of the time when expensive checks are
  enabled. Not sure how deep the problem is so just reverting while the
  author can investigate so that the bots stop repeatedly failing and
  blaming things incorrectly. Will respond with details on the original
  commit.

Original commit, r341269:

  [Constant Hoisting] Hoisting Constant GEP Expressions

  Leverage existing logic in constant hoisting pass to transform constant GEP
  expressions sharing the same base global variable. Multi-dimensional GEPs are
  rewritten into single-dimensional GEPs.

  https://reviews.llvm.org/D51396

Differential Revision: https://reviews.llvm.org/D51654

llvm-svn: 341417

6 years ago[LV] First order recurrence phis should not be treated as uniform
Anna Thomas [Tue, 4 Sep 2018 22:12:23 +0000 (22:12 +0000)]
[LV] First order recurrence phis should not be treated as uniform

This is fix for PR38786.
First order recurrence phis were incorrectly treated as uniform,
which caused them to be vectorized as uniform instructions.

Patch by Ayal Zaks and Orivej Desh!

Reviewed by: Anna

Differential Revision: https://reviews.llvm.org/D51639

llvm-svn: 341416

6 years ago[InstCombine] add tests for xor-or-xor fold; NFC
Sanjay Patel [Tue, 4 Sep 2018 22:10:23 +0000 (22:10 +0000)]
[InstCombine] add tests for xor-or-xor fold; NFC

There are 2 bugs shown here that were untested before:
1. We fail to perform the fold in 1/2 the possible commuted variants.
2. When the fold is done, it disregards extra uses.

llvm-svn: 341415

6 years ago[WebAssembly][NFC] Add colon to label in test
Thomas Lively [Tue, 4 Sep 2018 21:51:32 +0000 (21:51 +0000)]
[WebAssembly][NFC] Add colon to label in test

llvm-svn: 341414

6 years ago[AMDGPU] Legalize VGPR Rsrc operands for MUBUF instructions
Scott Linder [Tue, 4 Sep 2018 21:50:47 +0000 (21:50 +0000)]
[AMDGPU] Legalize VGPR Rsrc operands for MUBUF instructions

Emit a waterfall loop in the general case for a potentially-divergent Rsrc
operand. When practical, avoid this by using Addr64 instructions.

Differential Revision: https://reviews.llvm.org/D50982

llvm-svn: 341413

6 years agoFix a memory leak after rL341386.
Hiroshi Yamauchi [Tue, 4 Sep 2018 21:28:22 +0000 (21:28 +0000)]
Fix a memory leak after rL341386.

Reviewers: davidxl

Reviewed By: davidxl

Subscribers: llvm-commits

Differential Revision: https://reviews.llvm.org/D51658

llvm-svn: 341412

6 years ago[WebAssembly][NFC] Fix formatting and tests
Thomas Lively [Tue, 4 Sep 2018 21:26:17 +0000 (21:26 +0000)]
[WebAssembly][NFC] Fix formatting and tests

Summary: Small fixes

Reviewers: aheejin, dschuff

Subscribers: sbc100, jgravelle-google, sunfish, llvm-commits

Differential Revision: https://reviews.llvm.org/D51656

llvm-svn: 341411

6 years ago[InstCombine] make ((X & C) ^ C) form consistent for vectors
Sanjay Patel [Tue, 4 Sep 2018 21:17:14 +0000 (21:17 +0000)]
[InstCombine] make ((X & C) ^ C) form consistent for vectors

It would be better to create a 'not' here, but that's not possible yet.

llvm-svn: 341410

6 years ago[Hexagon] Don't packetize new-value stores with any other stores
Krzysztof Parzyszek [Tue, 4 Sep 2018 21:07:27 +0000 (21:07 +0000)]
[Hexagon] Don't packetize new-value stores with any other stores

llvm-svn: 341409

6 years agoWhen a relocation to an undefined symbol is an R_X86_64_PC32, an input
Sterling Augustine [Tue, 4 Sep 2018 21:06:59 +0000 (21:06 +0000)]
When a relocation to an undefined symbol is an R_X86_64_PC32, an input
section will not have an input file. Don't crash under those circumstances.

Neither clang nor llvm-mc generates R_X86_64_PC32 relocations due to
https://reviews.llvm.org/D43383, which makes it hard to write a test case.
However, gcc does generate such relocations. I want to get a fix in now,
but will figure out a way to actually exercise this code path as soon
as I can.

llvm-svn: 341408

6 years ago[NFC] Improve clarity in emitInstrCountChangedRemark
Jessica Paquette [Tue, 4 Sep 2018 21:03:43 +0000 (21:03 +0000)]
[NFC] Improve clarity in emitInstrCountChangedRemark

Add a "CouldOnlyImpactOneFunction" bool that's true when we pass in a function.

Just cleaning up a little bit, since I'm going to add in the per-function
remarks soon from D51467.

llvm-svn: 341407

6 years agoNFC: expand memset inline arm64 coverage
JF Bastien [Tue, 4 Sep 2018 21:02:00 +0000 (21:02 +0000)]
NFC: expand memset inline arm64 coverage

I'm looking at some codegen optimization in this area and want to make sure I understand the current codegen and don't regress it. This patch simply expands the two existing tests to capture more of the current code generation when it comes to heap-based and stack-based small memset on arm64. The tested code is already pretty good, notably when it comes to using STP, FP stores, FP immediate generation, and folding one of the stores into a stack spill when possible. The uses of STUR could be improved, and some more pairing could occur. Straying from bzero patterns currently yield suboptimal code, and I expect a variety of small changes could make things way better.

llvm-svn: 341406

6 years ago[InstCombine] simplify code for xor folds; NFCI
Sanjay Patel [Tue, 4 Sep 2018 21:00:13 +0000 (21:00 +0000)]
[InstCombine] simplify code for xor folds; NFCI

This is just a cleanup step. The TODO comments show
what is wrong with the 'and' version of the fold.
Fixing this should be part of recommitting:
rL300977

llvm-svn: 341405

6 years ago[CMake] Don't use -rtlib=compiler-rt with -nodefaultlibs.
Charles Davis [Tue, 4 Sep 2018 20:57:50 +0000 (20:57 +0000)]
[CMake] Don't use -rtlib=compiler-rt with -nodefaultlibs.

Summary:
This switch only has an effect at link time. It changes the default
compiler support library to `compiler-rt`. With `-nodefaultlibs`, this
library won't get linked anyway; Clang actually warns about that.

Reviewers: mstorsjo, rnk

Subscribers: dberris, mgorny, christof, cfe-commits

Differential Revision: https://reviews.llvm.org/D51645

llvm-svn: 341404

6 years ago[COFF] Allow exporting all symbols from system libraries specfied with -wholearchive:
Martin Storsjo [Tue, 4 Sep 2018 20:56:56 +0000 (20:56 +0000)]
[COFF] Allow exporting all symbols from system libraries specfied with -wholearchive:

When building a shared libc++.dll, it pulls in libc++abi.a statically
with the --wholearchive flag. If such a build is done with
--export-all-symbols, it's reasonable to assume that everything
from that library also should be exported with the same rules as normal
local object files, even though we normally avoid autoexporting things
from libc++abi.a in other cases when linking a DLL (user code).

Differential Revision: https://reviews.llvm.org/D51529

llvm-svn: 341403

6 years ago[MinGW] Move code for indicating "potentially not DSO local" into shouldAssumeDSOLoca...
Martin Storsjo [Tue, 4 Sep 2018 20:56:28 +0000 (20:56 +0000)]
[MinGW] Move code for indicating "potentially not DSO local" into shouldAssumeDSOLocal. NFC.

On Windows, if shouldAssumeDSOLocal returns false, it's either a
dllimport reference, or a reference that we should treat as non-local
and create a stub for.

Clean up AArch64Subtarget::ClassifyGlobalReference a little while
touching the flag handling relating to dllimport.

Differential Revision: https://reviews.llvm.org/D51590

llvm-svn: 341402

6 years ago[MinGW] [AArch64] Add stubs for potential automatic dllimported variables
Martin Storsjo [Tue, 4 Sep 2018 20:56:21 +0000 (20:56 +0000)]
[MinGW] [AArch64] Add stubs for potential automatic dllimported variables

The runtime pseudo relocations can't handle the AArch64 format PC
relative addressing in adrp+add/ldr pairs. By using stubs, the potentially
dllimported addresses can be touched up by the runtime pseudo relocation
framework.

Differential Revision: https://reviews.llvm.org/D51452

llvm-svn: 341401

6 years agoFix unused variable warning
Reid Kleckner [Tue, 4 Sep 2018 20:34:47 +0000 (20:34 +0000)]
Fix unused variable warning

llvm-svn: 341400

6 years ago[libomptarget] Remove `Devices` from `RTLInfoTy`
Sergey Dmitriev [Tue, 4 Sep 2018 20:23:09 +0000 (20:23 +0000)]
[libomptarget] Remove `Devices` from `RTLInfoTy`

This patch removes unused field `Devices` from `RTLInfoTy`.

Differential Revision: https://reviews.llvm.org/D51653

llvm-svn: 341399

6 years ago[SimpleLoopUnswitch] remove a chain of dead blocks at once
Fedor Sergeev [Tue, 4 Sep 2018 20:19:41 +0000 (20:19 +0000)]
[SimpleLoopUnswitch] remove a chain of dead blocks at once

Recent change to deleteDeadBlocksFromLoop was not enough to
fix all the problems related to dead blocks after nontrivial
unswitching of switches.

We need to delete all the dead blocks that were created during
unswitching, otherwise we will keep having problems with phi's
or dead blocks.

This change removes all the dead blocks that are reachable from the loop,
not trying to track whether these blocks are newly created by unswitching
or not. While not completely correct, we are unlikely to get loose but
reachable dead blocks that do not belong to our loop nest.

It does fix all the failures currently known, in particular PR38778.

Reviewed By: asbirlea
Differential Revision: https://reviews.llvm.org/D51519

llvm-svn: 341398

6 years ago[Codegen] Fix test added in rL341380: AArch64SelectionDAGTest needs to link against...
Jordan Rupprecht [Tue, 4 Sep 2018 20:19:17 +0000 (20:19 +0000)]
[Codegen] Fix test added in rL341380: AArch64SelectionDAGTest needs to link against Analysis lib.

llvm-svn: 341397

6 years agoSet console mode when -fansi-escape-codes is enabled
David Bolvansky [Tue, 4 Sep 2018 19:23:05 +0000 (19:23 +0000)]
Set console mode when -fansi-escape-codes is enabled

Summary:
Windows console now supports supports ANSI escape codes, but we need to enable it using SetConsoleMode with ENABLE_VIRTUAL_TERMINAL_PROCESSING flag.

Fixes https://bugs.llvm.org/show_bug.cgi?id=38817

Tested on Windows 10, screenshot:
https://i.imgur.com/bqYq0Uy.png

Reviewers: zturner, chandlerc

Reviewed By: zturner

Subscribers: llvm-commits

Differential Revision: https://reviews.llvm.org/D51611

llvm-svn: 341396

6 years ago[CMake] Provide a custom target to install LLVM libraries
Petr Hosek [Tue, 4 Sep 2018 19:10:37 +0000 (19:10 +0000)]
[CMake] Provide a custom target to install LLVM libraries

This simplifies installing all LLVM libraries when doing component
build; now you can include llvm-libraries in distribution components.

Differential Revision: https://reviews.llvm.org/D51603

llvm-svn: 341395

6 years ago[InstCombine] improve xor+and/or tests
Sanjay Patel [Tue, 4 Sep 2018 19:06:46 +0000 (19:06 +0000)]
[InstCombine] improve xor+and/or tests

The tests attempted to check for commuted variants
of these folds, but complexity-based canonicalization
meant we had no coverage for at least 1/2 of the cases.

Also, the folds correctly check hasOneUse(), but there
was no coverage for that.

llvm-svn: 341394

6 years agoAMDGPU: Fix DAG divergence not reporting flat loads
Matt Arsenault [Tue, 4 Sep 2018 18:58:19 +0000 (18:58 +0000)]
AMDGPU: Fix DAG divergence not reporting flat loads

Match behavior in DAG of r340343

llvm-svn: 341393

6 years agoDAG: Factor out helper function for odd vector sizes
Matt Arsenault [Tue, 4 Sep 2018 18:47:43 +0000 (18:47 +0000)]
DAG: Factor out helper function for odd vector sizes

llvm-svn: 341392

6 years agoFix build failures after rL341386.
Hiroshi Yamauchi [Tue, 4 Sep 2018 18:10:54 +0000 (18:10 +0000)]
Fix build failures after rL341386.

Reviewers: davidxl

Reviewed By: davidxl

Subscribers: llvm-commits

Differential Revision: https://reviews.llvm.org/D51647

llvm-svn: 341391

6 years agoclang-cl: Pass /Brepro to linker if it was passed to the compiler
Nico Weber [Tue, 4 Sep 2018 18:00:14 +0000 (18:00 +0000)]
clang-cl: Pass /Brepro to linker if it was passed to the compiler

Differential Revision: https://reviews.llvm.org/D51635

llvm-svn: 341390

6 years ago[WebAssembly] Fix operand rewriting in inline asm lowering.
Dan Gohman [Tue, 4 Sep 2018 17:46:12 +0000 (17:46 +0000)]
[WebAssembly] Fix operand rewriting in inline asm lowering.

Use MachineOperand::ChangeToImmediate rather than reassigning
MachineOperands to new values created from MachineOperand::CreateImm,
so that their parent pointers are preserved.

This fixes "Instruction has operand with wrong parent set" errors
reported by the MachineVerifier.

llvm-svn: 341389

6 years ago[CMake] Remove variable reference that isn't used.
Charles Davis [Tue, 4 Sep 2018 17:40:26 +0000 (17:40 +0000)]
[CMake] Remove variable reference that isn't used.

Summary:
This variable is never defined, so its value is always empty. Since
`libunwind` is needed to build the C++ ABI library in the first place,
it should never be linked to the C++ ABI library anyway.

Reviewers: mstorsjo, rnk

Subscribers: mgorny, christof, cfe-commits

Differential Revision: https://reviews.llvm.org/D51644

llvm-svn: 341388

6 years agoTerminate debugger if an assert was hit
David Bolvansky [Tue, 4 Sep 2018 17:19:15 +0000 (17:19 +0000)]
Terminate debugger if an assert was hit

Reviewers: JDevlieghere, teemperor, #lldb

Reviewed By: JDevlieghere

Subscribers: clayborg, lemo, lldb-commits

Differential Revision: https://reviews.llvm.org/D51604

llvm-svn: 341387

6 years ago[PGO] Control Height Reduction
Hiroshi Yamauchi [Tue, 4 Sep 2018 17:19:13 +0000 (17:19 +0000)]
[PGO] Control Height Reduction

Summary:
Control height reduction merges conditional blocks of code and reduces the
number of conditional branches in the hot path based on profiles.

if (hot_cond1) { // Likely true.
  do_stg_hot1();
}
if (hot_cond2) { // Likely true.
  do_stg_hot2();
}

->

if (hot_cond1 && hot_cond2) { // Hot path.
  do_stg_hot1();
  do_stg_hot2();
} else { // Cold path.
  if (hot_cond1) {
    do_stg_hot1();
  }
  if (hot_cond2) {
    do_stg_hot2();
  }
}

This speeds up some internal benchmarks up to ~30%.

Reviewers: davidxl

Reviewed By: davidxl

Subscribers: xbolva00, dmgreen, mehdi_amini, llvm-commits, mgorny

Differential Revision: https://reviews.llvm.org/D50591

llvm-svn: 341386

6 years ago[libFuzzer] Enable tests on Windows
Matt Morehouse [Tue, 4 Sep 2018 17:08:47 +0000 (17:08 +0000)]
[libFuzzer] Enable tests on Windows

Summary:
Enable tests on Windows and make check-fuzzer pass on it.  Make
check-fuzzer pass on Windows by fixing libFuzzer, fixing tests, and by
disabling tests on Windows. Most of these are disabled temporarily as
support for the tests and the features they test will be added
incrementally.  Other tests will not be enabled since they require
things that are not on Windows (eg: afl_driver tests).  Every test
that was explicitly disabled on Windows has a comment explaining why
(unless obvious like merge-posix.test).

The lit.cfg file was modified to support running tests on windows.
fuzzer-dirs.test was fixed by making the Windows implementation print
the same error message as the posix version.
merge-control-file.test was fixed by making the test binary end with
the ".exe" extension (on all platforms).

Patch By: metzman

Reviewers: morehouse

Reviewed By: morehouse

Subscribers: srhines, mgorny

Differential Revision: https://reviews.llvm.org/D51549

llvm-svn: 341385

6 years ago[TSan] Update assembly test again.
Matt Morehouse [Tue, 4 Sep 2018 17:04:30 +0000 (17:04 +0000)]
[TSan] Update assembly test again.

Previous commit incorrectly updated the read1 case.

llvm-svn: 341384

6 years ago[TSan] Update assembly code check.
Matt Morehouse [Tue, 4 Sep 2018 16:34:26 +0000 (16:34 +0000)]
[TSan] Update assembly code check.

The new assembly makes our benchmark faster, so it should be safe to
update the check.

llvm-svn: 341381

6 years ago[CodeGen] Fix remaining zext() assertions in SelectionDAG
Scott Linder [Tue, 4 Sep 2018 16:33:34 +0000 (16:33 +0000)]
[CodeGen] Fix remaining zext() assertions in SelectionDAG

Fix remaining cases not committed in https://reviews.llvm.org/D49574

Differential Revision: https://reviews.llvm.org/D50659

llvm-svn: 341380

6 years ago[MachO] Fix inconsistency between error messages when validating LC_DYSYMTAB
Francis Visoiu Mistrih [Tue, 4 Sep 2018 16:31:53 +0000 (16:31 +0000)]
[MachO] Fix inconsistency between error messages when validating LC_DYSYMTAB

llvm-svn: 341379

6 years ago[MachO] Fix LC_DYSYMTAB validation for external symbols
Francis Visoiu Mistrih [Tue, 4 Sep 2018 16:31:48 +0000 (16:31 +0000)]
[MachO] Fix LC_DYSYMTAB validation for external symbols

We were validating the same index (ilocalsym) twice, while iextdefsym
was never validated.

llvm-svn: 341378

6 years ago[dwarfdump] Improve -diff option by hiding more data.
Jonas Devlieghere [Tue, 4 Sep 2018 16:21:37 +0000 (16:21 +0000)]
[dwarfdump] Improve -diff option by hiding more data.

The -diff option makes it easy to diff dwarf by hiding addresses and
offsets. However not all of them were hidden, which should be fixed by
this patch.

Differential revision: https://reviews.llvm.org/D51593

llvm-svn: 341377