platform/upstream/llvm.git
5 years agoAMDGPU: Add ds append/consume builtins
Matt Arsenault [Mon, 28 Jan 2019 23:59:18 +0000 (23:59 +0000)]
AMDGPU: Add ds append/consume builtins

llvm-svn: 352443

5 years ago[WebAssembly] Expand BUILD_PAIR nodes
Thomas Lively [Mon, 28 Jan 2019 23:44:31 +0000 (23:44 +0000)]
[WebAssembly] Expand BUILD_PAIR nodes

Reviewers: aheejin

Subscribers: dschuff, sbc100, jgravelle-google, hiraditya, sunfish

Differential Revision: https://reviews.llvm.org/D57276

llvm-svn: 352442

5 years ago[ThinLTO] Add option to dump per-module summary dot graph
Teresa Johnson [Mon, 28 Jan 2019 23:43:26 +0000 (23:43 +0000)]
[ThinLTO] Add option to dump per-module summary dot graph

Summary:
I found that there currently isn't a way to invoke exportToDot from
the command line for a per-module summary index, and therefore no
testing of that case. Add an internal option and use it to test dumping
of per module summary indexes.

In particular, I am looking at fixing the limitation that causes the
aliasee GUID in the per-module summary to be 0, and want to be able to
test that change.

Reviewers: evgeny777

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

Differential Revision: https://reviews.llvm.org/D57206

llvm-svn: 352441

5 years agoDemanded elements support for vector GEPs
Philip Reames [Mon, 28 Jan 2019 23:24:49 +0000 (23:24 +0000)]
Demanded elements support for vector GEPs

GEPs can produce either scalar or vector results. If we're extracting only a subset of the vector lanes, simplifying the operands is helpful in eliminating redundant computation, and (eventually) allowing further optimizations

Differential Revision: https://reviews.llvm.org/D57177

llvm-svn: 352440

5 years ago[docs] Fix a couple spelling errors.
Eli Friedman [Mon, 28 Jan 2019 23:03:41 +0000 (23:03 +0000)]
[docs] Fix a couple spelling errors.

llvm-svn: 352439

5 years ago[ThinLTO] Refine reachability check to fix compile time increase
Teresa Johnson [Mon, 28 Jan 2019 22:27:05 +0000 (22:27 +0000)]
[ThinLTO] Refine reachability check to fix compile time increase

Summary:
A recent fix to the ThinLTO whole program dead code elimination (D56117)
increased the thin link time on a large MSAN'ed binary by 2x.
It's likely that the time increased elsewhere, but was more noticeable
here since it was already large and ended up timing out.

That change made it so we would repeatedly scan all copies of linkonce
symbols for liveness every time they were encountered during the graph
traversal. This was needed since we only mark one copy of an aliasee as
live when we encounter a live alias. This patch fixes the issue in a
more efficient manner by simply proactively visiting the aliasee (thus
marking all copies live) when we encounter a live alias.

Two notes: One, this requires a hash table lookup (finding the aliasee
summary in the index based on aliasee GUID). However, the impact of this
seems to be small compared to the original pre-D56117 thin link time. It
could be addressed if we keep the aliasee ValueInfo in the alias summary
instead of the aliasee GUID, which I am exploring in a separate patch.

Second, we only populate the aliasee GUID field when reading summaries
from bitcode (whether we are reading individual summaries and merging on
the fly to form the compiled index, or reading in a serialized combined
index). Thankfully, that's currently the only way we can get to this
code as we don't yet support reading summaries from LLVM assembly
directly into a tool that performs the thin link (they must be converted
to bitcode first). I added a FIXME, however I have the fix under test
already. The easiest fix is to simply populate this field always, which
isn't hard, but more likely the change I am exploring to store the
ValueInfo instead as described above will subsume this. I don't want to
hold up the regression fix for this though.

Reviewers: trentxintong

Subscribers: mehdi_amini, inglorion, dexonsmith, llvm-commits

Differential Revision: https://reviews.llvm.org/D57203

llvm-svn: 352438

5 years ago[CGP] auto-generate complete checks for add overflow tests; NFC
Sanjay Patel [Mon, 28 Jan 2019 22:07:37 +0000 (22:07 +0000)]
[CGP] auto-generate complete checks for add overflow tests; NFC

llvm-svn: 352437

5 years ago[ASTImporter] Fix handling of overriden methods during ASTImport
Shafik Yaghmour [Mon, 28 Jan 2019 21:55:33 +0000 (21:55 +0000)]
[ASTImporter] Fix handling of overriden methods during ASTImport

Summary:
When importing classes we may add a CXXMethodDecl more than once to a CXXRecordDecl when handling overrides. This patch will fix the cases we currently know about and handle the case where we are only dealing with declarations.

Differential Revision: https://reviews.llvm.org/D56936

llvm-svn: 352436

5 years agoAttempt to fix build failure with GCC 5.4.
Rui Ueyama [Mon, 28 Jan 2019 21:45:50 +0000 (21:45 +0000)]
Attempt to fix build failure with GCC 5.4.

llvm-svn: 352435

5 years agoRevert "Make Type::GetByteSize optional (NFC)"
Adrian Prantl [Mon, 28 Jan 2019 21:44:35 +0000 (21:44 +0000)]
Revert "Make Type::GetByteSize optional (NFC)"

This reverts commit r352394 because it broke three windows-specific tests.

llvm-svn: 352434

5 years agoRecommit r352255 "[SelectionDAG][X86] Don't use SEXTLOAD for promoting masked loads...
Craig Topper [Mon, 28 Jan 2019 21:38:47 +0000 (21:38 +0000)]
Recommit r352255 "[SelectionDAG][X86] Don't use SEXTLOAD for promoting masked loads in the type legalizer"

This did not cause the buildbot failure it was previously reverted for.

Original commit message:

I'm not sure why we were using SEXTLOAD. EXTLOAD seems more appropriate since we don't care about the upper bits.

This patch changes this and then modifies the X86 post legalization combine to emit a extending shuffle instead of a sign_extend_vector_inreg. Could maybe use an any_extend_vector_inre

On AVX512 targets I think we might be able to use a masked vpmovzx and not have to expand this at all.

llvm-svn: 352433

5 years ago[RuntimeDyld] load all sections with ProcessAllSections
Yonghong Song [Mon, 28 Jan 2019 21:35:23 +0000 (21:35 +0000)]
[RuntimeDyld] load all sections with ProcessAllSections

This patch tried to address the following use case.
  . bcc (https://github.com/iovisor/bcc) utilizes llvm JIT to
    compile for BTF target.
  . with -g, .BTF and .BTF.ext sections (BPF debug info)
    will be generated by LLVM.
  . .BTF does not have relocations and .BTF.ext has some
    relocations.
  . With ProcessAllSections, .BTF.ext is loaded by JIT dynamic linker
    and is available to application. But .BTF is not loaded.

The bcc application needs both .BTF.ext and .BTF for debugging
purpose, and .BTF is not loaded. This patch addressed this issue
by iterating over all sections and loading any missing
sections, after symbol/relocation processing in loadObjectImpl().

Signed-off-by: Yonghong Song <yhs@fb.com>
Differential Revision: https://reviews.llvm.org/D55943

llvm-svn: 352432

5 years ago[ARM] Deduplicate table generated CC analysis code
Reid Kleckner [Mon, 28 Jan 2019 21:28:43 +0000 (21:28 +0000)]
[ARM] Deduplicate table generated CC analysis code

Create ARMCallingConv.cpp and emit code for calling convention analysis
from there.

llvm-svn: 352431

5 years ago[AArch64] Include AArch64GenCallingConv.inc once
Reid Kleckner [Mon, 28 Jan 2019 21:28:40 +0000 (21:28 +0000)]
[AArch64] Include AArch64GenCallingConv.inc once

Summary:
Avoids duplicating generated static helpers for calling convention
analysis.

This also means you can modify AArch64CallingConv.td without recompiling
the AArch64ISelLowering.cpp monolith, so it provides faster incremental
rebuilds.

Saves 12K in llc.exe, but adds a new object file, which is large.

Reviewers: efriedma, t.p.northover

Subscribers: mgorny, javed.absar, kristof.beyls, hiraditya, llvm-commits

Differential Revision: https://reviews.llvm.org/D56948

llvm-svn: 352430

5 years ago[GlobalISel][AArch64] Add legalization for G_FLOG
Jessica Paquette [Mon, 28 Jan 2019 21:27:23 +0000 (21:27 +0000)]
[GlobalISel][AArch64] Add legalization for G_FLOG

This adds support for legalizing G_FLOG into a RTLib call.

It adds a legalizer test, and updates the existing floating point tests.

https://reviews.llvm.org/D57347

llvm-svn: 352429

5 years agolld/coff: Make assoc comdat diag a bit more detailed
Nico Weber [Mon, 28 Jan 2019 21:16:15 +0000 (21:16 +0000)]
lld/coff: Make assoc comdat diag a bit more detailed

Many different sections can have the same name, so include the indices of the
sections mentioned in the diagnostic too.

I'm debugging something I can't repro locally, maybe this will help.

llvm-svn: 352428

5 years agoRevert "[libunwind] Drop the dependency on <algorithm>, add placement new inline"
Petr Hosek [Mon, 28 Jan 2019 20:55:12 +0000 (20:55 +0000)]
Revert "[libunwind] Drop the dependency on <algorithm>, add placement new inline"

This reverts commit r352384: this broke on ARM as UnwindCursor.hpp
still has some C++ library dependencies.

llvm-svn: 352427

5 years ago[libc++] Use runtime rather then compile-time glibc version check
Petr Hosek [Mon, 28 Jan 2019 20:39:50 +0000 (20:39 +0000)]
[libc++] Use runtime rather then compile-time glibc version check

glibc supports versioning, so it's possible to build against older
version and run against newer version. This is sometimes relied on
in practice, e.g. in Fuchsia build we build against older sysroot
(equivalent to Ubuntu Trusty) to cover the broadest possible range
of host systems, but that doesn't necessarily match the system that
binary is going to run on which may have newer version, in which case
the compile test used in curr_symbol is going to fail. Using runtime
check is more reliable.

Differential Revision: https://reviews.llvm.org/D56702

llvm-svn: 352425

5 years ago[InstCombine] add another saturating uadd test (no undefs); NFC
Sanjay Patel [Mon, 28 Jan 2019 20:37:18 +0000 (20:37 +0000)]
[InstCombine] add another saturating uadd test (no undefs); NFC

I forgot that our undef matching hasn't been completed in the previous commit.

llvm-svn: 352424

5 years ago[InstCombine] add tests for saturating uadd with constant; NFC
Sanjay Patel [Mon, 28 Jan 2019 20:32:48 +0000 (20:32 +0000)]
[InstCombine] add tests for saturating uadd with constant; NFC

llvm-svn: 352423

5 years agoAMDGPU: Add DS append/consume intrinsics
Matt Arsenault [Mon, 28 Jan 2019 20:14:49 +0000 (20:14 +0000)]
AMDGPU: Add DS append/consume intrinsics

Since these pass the pointer in m0 unlike other DS instructions, these
need to worry about whether the address is uniform or not. This
assumes the address is dynamically uniform, and just uses
readfirstlane to get a copy into an SGPR.

I don't know if these have the same 16-bit add for the addressing mode
offset problem on SI or not, but I've just assumed they do.

Also includes some misc. changes to avoid test differences between the
LDS and GDS versions.

llvm-svn: 352422

5 years ago[OPENMP][NVPTX]Emit service debug variable for NVPTX.
Alexey Bataev [Mon, 28 Jan 2019 20:03:02 +0000 (20:03 +0000)]
[OPENMP][NVPTX]Emit service debug variable for NVPTX.

In case of the empty module, the ptxas tool may emit error message about
empty debug info sections. This patch fixes this bug.

llvm-svn: 352421

5 years agogn build: Add get.py script to download prebuilt gn, make gn.py run downloaded gn...
Nico Weber [Mon, 28 Jan 2019 19:54:41 +0000 (19:54 +0000)]
gn build: Add get.py script to download prebuilt gn, make gn.py run downloaded gn if gn is not on PATH

Prebuilts are available for x86_64 Linux, macOS, Windows. The script always
pulls the latest GN version.

Differential Revision: https://reviews.llvm.org/D57256

llvm-svn: 352420

5 years agogn build: Make cmake sync script work on Windows if git is a bat file
Nico Weber [Mon, 28 Jan 2019 19:53:52 +0000 (19:53 +0000)]
gn build: Make cmake sync script work on Windows if git is a bat file

Differential Revision: https://reviews.llvm.org/D57338

llvm-svn: 352419

5 years ago[GlobalISel][AArch64] Add instruction selection support for @llvm.log10
Jessica Paquette [Mon, 28 Jan 2019 19:53:14 +0000 (19:53 +0000)]
[GlobalISel][AArch64] Add instruction selection support for @llvm.log10

This adds instruction selection support for @llvm.log10 in AArch64. It teaches
GISel to lower it to a library call, updates the relevant tests, and adds a
legalizer test for log10.

https://reviews.llvm.org/D57341

llvm-svn: 352418

5 years agoRemove unimplemented function
Adrian Prantl [Mon, 28 Jan 2019 19:38:08 +0000 (19:38 +0000)]
Remove unimplemented function

Looks like this was an unintended sideeffect of r124250.

Differential Revision: https://reviews.llvm.org/D57272

llvm-svn: 352417

5 years ago[AliasSetTracker] Cleanup more comments. [NFCI]
Alina Sbirlea [Mon, 28 Jan 2019 19:38:03 +0000 (19:38 +0000)]
[AliasSetTracker] Cleanup more comments. [NFCI]

llvm-svn: 352416

5 years agogn build: Fix `lld-link: unknown flag: -fuse-ld=lld` warnings on Windows
Nico Weber [Mon, 28 Jan 2019 19:32:52 +0000 (19:32 +0000)]
gn build: Fix `lld-link: unknown flag: -fuse-ld=lld` warnings on Windows

Fixes a minor regression from r351248.

While here, also make it possible to opt out of lld by saying
use_lld=false when clang_base_path is set. (use_lld still defaults to
true if clang_base_path is set.)

llvm-svn: 352415

5 years ago[MC] Do not consider .ifdef/.ifndef as a use
Scott Linder [Mon, 28 Jan 2019 19:32:08 +0000 (19:32 +0000)]
[MC] Do not consider .ifdef/.ifndef as a use

This is allowed by GAS and seems correct.

Differential Revision: https://reviews.llvm.org/D55439

llvm-svn: 352414

5 years agoELF: Set sh_info on RelaIplt to point to the IgotPlt output section.
Peter Collingbourne [Mon, 28 Jan 2019 19:29:41 +0000 (19:29 +0000)]
ELF: Set sh_info on RelaIplt to point to the IgotPlt output section.

Previously we were setting it to the GotPlt output section, which is
incorrect on ARM where this section is in .got. In static binaries
this can lead to sh_info being set to -1 (because there is no .got.plt)
which results in various tools rejecting the output file.

Differential Revision: https://reviews.llvm.org/D57274

llvm-svn: 352413

5 years ago[AArch64] Add 'apple-latest' CPU alias
Francis Visoiu Mistrih [Mon, 28 Jan 2019 19:27:33 +0000 (19:27 +0000)]
[AArch64] Add 'apple-latest' CPU alias

The 'apple-latest' alias is supposed to provide a CPU that contains the
latest Apple processor model supported by LLVM.

This is supposed to be used by tools like lldb to provide a target that
supports most of the CPU features.

For now, this is mapped to Cyclone.

Differential Revision: https://reviews.llvm.org/D56384

llvm-svn: 352412

5 years agoRevert "[CMake] Use __libc_start_main rather than fopen when checking for C library"
Petr Hosek [Mon, 28 Jan 2019 19:26:41 +0000 (19:26 +0000)]
Revert "[CMake] Use __libc_start_main rather than fopen when checking for C library"

This reverts commit r352341: it broke the build on macOS which doesn't
seem to provide __libc_start_main in its C library.

llvm-svn: 352411

5 years ago[GlobalISel] Add ISel support for @llvm.lifetime.start and @llvm.lifetime.end
Jessica Paquette [Mon, 28 Jan 2019 19:22:29 +0000 (19:22 +0000)]
[GlobalISel] Add ISel support for @llvm.lifetime.start and @llvm.lifetime.end

This adds ISel support for lifetime markers in opt levels above O0.

It also updates the arm64-irtranslator test, and updates some AArch64 tests that
use them for added coverage.

It also adds a testcase taken from the X86 codegen tests which verified a bug
caused by lifetime markers + stack colouring in the past. This is intended to
make sure that GISel doesn't re-introduce the bug.

(This is basically a straight copy from what SelectionDAG does in
SelectionDAGBuilder.cpp)

https://reviews.llvm.org/D57187

llvm-svn: 352410

5 years ago[CodeGen][X86] Expand UADDSAT to NOT+UMIN+ADD
Nikita Popov [Mon, 28 Jan 2019 19:19:09 +0000 (19:19 +0000)]
[CodeGen][X86] Expand UADDSAT to NOT+UMIN+ADD

Followup to D56636, this time handling the UADDSAT case by expanding
uadd.sat(a, b) to umin(a, ~b) + b.

Differential Revision: https://reviews.llvm.org/D56869

llvm-svn: 352409

5 years ago[CodeExtractor] Add support for the `swifterror` attribute
Vedant Kumar [Mon, 28 Jan 2019 19:13:37 +0000 (19:13 +0000)]
[CodeExtractor] Add support for the `swifterror` attribute

When passing a `swifterror` argument or alloca as an input to an
extraction region, mark the input parameter `swifterror`.

llvm-svn: 352408

5 years agoRefactoring. NFC.
Rui Ueyama [Mon, 28 Jan 2019 19:11:52 +0000 (19:11 +0000)]
Refactoring. NFC.

llvm-svn: 352407

5 years ago[AliasSetTracker] Cleanup comments. [NFCI]
Alina Sbirlea [Mon, 28 Jan 2019 19:01:32 +0000 (19:01 +0000)]
[AliasSetTracker] Cleanup comments. [NFCI]

llvm-svn: 352406

5 years agoCorrect help text for --unwindlib options.
Sterling Augustine [Mon, 28 Jan 2019 18:56:44 +0000 (18:56 +0000)]
Correct help text for --unwindlib options.

llvm-svn: 352405

5 years agoAdd --unwindlib=[libgcc|compiler-rt] to parallel --rtlib=
Sterling Augustine [Mon, 28 Jan 2019 18:56:43 +0000 (18:56 +0000)]
Add --unwindlib=[libgcc|compiler-rt] to parallel --rtlib=

Summary:
"clang++ hello.cc --rtlib=compiler-rt"

now works without specifying additional unwind or exception
handling libraries.

Reviewers: rsmith

Subscribers: srhines, dberris, aheejin, llvm-commits

Differential Revision: https://reviews.llvm.org/D57128

llvm-svn: 352404

5 years agoRevert "[AST] Introduce GenericSelectionExpr::Association"
Bruno Ricci [Mon, 28 Jan 2019 18:40:26 +0000 (18:40 +0000)]
Revert "[AST] Introduce GenericSelectionExpr::Association"

This breaks GCC 4.8.4. Reported by email by Hans Wennborg.

llvm-svn: 352403

5 years ago[GlobalISel][AArch64] Add instruction selection support for G_FCOS and G_FSIN
Jessica Paquette [Mon, 28 Jan 2019 18:34:18 +0000 (18:34 +0000)]
[GlobalISel][AArch64] Add instruction selection support for G_FCOS and G_FSIN

This contains all of the legalizer changes from D57197 necessary to select
G_FCOS and G_FSIN. It also updates several existing IR tests in
test/CodeGen/AArch64 that verify that we correctly lower the G_FCOS and G_FSIN
instructions.

https://reviews.llvm.org/D57197
3/3

llvm-svn: 352402

5 years ago[GlobalISel][AArch64] Add IRTranslator support for G_FCOS and G_FSIN
Jessica Paquette [Mon, 28 Jan 2019 18:34:17 +0000 (18:34 +0000)]
[GlobalISel][AArch64] Add IRTranslator support for G_FCOS and G_FSIN

This adds IRTranslator support for the G_FCOS and G_FSIN generic instructions.

https://reviews.llvm.org/D57197
2/3

llvm-svn: 352401

5 years ago[GlobalISel] Add G_FSIN and G_FCOS generic instructions
Jessica Paquette [Mon, 28 Jan 2019 18:34:16 +0000 (18:34 +0000)]
[GlobalISel] Add G_FSIN and G_FCOS generic instructions

This introduces generic instrutions for floating point sin and cos, G_FCOS and
G_FSIN. It updates the tests, etc.

https://reviews.llvm.org/D57197
1/3

llvm-svn: 352400

5 years ago[AliasSetTracker] Update signature to aliasesPointer [NFCI].
Alina Sbirlea [Mon, 28 Jan 2019 18:30:05 +0000 (18:30 +0000)]
[AliasSetTracker] Update signature to aliasesPointer [NFCI].

llvm-svn: 352399

5 years ago[testsuite] Remove unused seven module imports.
Jonas Devlieghere [Mon, 28 Jan 2019 18:07:41 +0000 (18:07 +0000)]
[testsuite] Remove unused seven module imports.

llvm-svn: 352398

5 years ago[testsuite] Remove trailing characters from command output.
Jonas Devlieghere [Mon, 28 Jan 2019 18:07:28 +0000 (18:07 +0000)]
[testsuite] Remove trailing characters from command output.

When running the test suite on macOS with Python 3 we noticed a
difference in behavior between Python 2 and Python 3 for
seven.get_command_output. The output contained a newline with Python 3,
but not for Python 2. This resulted in an invalid SDK path passed to the
compiler.

Differential revision: https://reviews.llvm.org/D57275

llvm-svn: 352397

5 years ago[NFC] TLI query with default(on) behavior wrt DAG combines for fmin/fmax target control
Michael Berg [Mon, 28 Jan 2019 18:03:08 +0000 (18:03 +0000)]
[NFC] TLI query with default(on) behavior wrt DAG combines for fmin/fmax target control

llvm-svn: 352396

5 years ago[libFuzzer][Windows] Use dllexport for all declarations in FuzzerInterface.h
Jonathan Metzman [Mon, 28 Jan 2019 17:51:13 +0000 (17:51 +0000)]
[libFuzzer][Windows] Use dllexport for all declarations in FuzzerInterface.h

Summary:
Use dllexport for all declarations in FuzzerInterface.h Use it for clang
even though clang supports default visibility attribute to prevent a
warning from being thrown when LLVMFuzzerMutate is defined with dllexport.
This makes `FUZZER_INTERFACE_VISIBILITY` (FuzzerInterface.h) consistent with
`ATTRIBUTE_INTERFACE` (FuzzerDefs.h) when using clang on Windows.

Reviewers: vitalybuka, morehouse

Reviewed By: morehouse

Differential Revision: https://reviews.llvm.org/D57305

llvm-svn: 352395

5 years agoMake Type::GetByteSize optional (NFC)
Adrian Prantl [Mon, 28 Jan 2019 17:49:33 +0000 (17:49 +0000)]
Make Type::GetByteSize optional (NFC)

This is a continuation of my quest to make the size 0 a supported value.

Differential Revision: https://reviews.llvm.org/D57273

llvm-svn: 352394

5 years ago[SimpleLoopUnswitch] Early check exit for trivial unswitch with MemorySSA.
Alina Sbirlea [Mon, 28 Jan 2019 17:48:45 +0000 (17:48 +0000)]
[SimpleLoopUnswitch] Early check exit for trivial unswitch with MemorySSA.

Summary:
If MemorySSA is avaiable, we can skip checking all instructions if block has any Defs.
(volatile loads are also Defs).
We still need to check all instructions for "canThrow", even if no Defs are found.

Reviewers: chandlerc

Subscribers: sanjoy, jlebar, Prazek, george.burgess.iv, llvm-commits

Differential Revision: https://reviews.llvm.org/D57129

llvm-svn: 352393

5 years ago[fuzzer][afl] Remove AFL_DRIVER_EXTRA_STATS_FILENAME
Jonathan Metzman [Mon, 28 Jan 2019 17:15:49 +0000 (17:15 +0000)]
[fuzzer][afl] Remove AFL_DRIVER_EXTRA_STATS_FILENAME

Summary:
Remove this feature as it is unused, buggy, and not worth correcting
since the forkserver makes it difficult.

Reviewers: morehouse, jfb

Reviewed By: morehouse

Differential Revision: https://reviews.llvm.org/D57308

llvm-svn: 352392

5 years agoAdd -fapply-global-visibility-to-externs for -cc1
Scott Linder [Mon, 28 Jan 2019 17:12:19 +0000 (17:12 +0000)]
Add -fapply-global-visibility-to-externs for -cc1

Introduce an option to request global visibility settings be applied to
declarations without a definition or an explicit visibility, rather than
the existing behavior of giving these default visibility. When the
visibility of all or most extern definitions are known this allows for
the same optimisations -fvisibility permits without updating source code
to annotate all declarations.

Differential Revision: https://reviews.llvm.org/D56868

llvm-svn: 352391

5 years ago[clang][OpenMP] OMPFlushClause is synthetic, no such clause exists
Roman Lebedev [Mon, 28 Jan 2019 17:04:11 +0000 (17:04 +0000)]
[clang][OpenMP] OMPFlushClause is synthetic, no such clause exists

Summary:
As discussed in https://reviews.llvm.org/D57112#inline-506781,
'flush' clause does not exist in the OpenMP spec, it can not be
specified, and `OMPFlushClause` class is just a helper class.

Therefore `OPENMP_CLAUSE()` in `clang/Basic/OpenMPKinds.def`
should not contain 'flush' "clause".

I have simply removed the `OPENMP_CLAUSE(flush, OMPFlushClause)`
from `clang/Basic/OpenMPKinds.def`, grepped for `OPENMP_CLAUSE`
and added `OPENMP_CLAUSE(flush, OMPFlushClause)` back to the **every**
place where `OPENMP_CLAUSE` is defined and `clang/Basic/OpenMPKinds.def`
is then included.

So as-is, this patch is a NFC. Possibly, some of these
`OPENMP_CLAUSE(flush, OMPFlushClause)` should be dropped,
i don't really know.

Test plan: `ninja check-clang`

Reviewers: ABataev

Reviewed By: ABataev

Subscribers: guansong, arphaman, cfe-commits

Tags: #openmp

Differential Revision: https://reviews.llvm.org/D57280

llvm-svn: 352390

5 years ago[X86][AVX] Remove lowerShuffleByMerging128BitLanes 2-lane restriction
Simon Pilgrim [Mon, 28 Jan 2019 17:02:35 +0000 (17:02 +0000)]
[X86][AVX] Remove lowerShuffleByMerging128BitLanes 2-lane restriction

First step towards adding support for 64-bit unary "sublane" handling (a bit like lowerShuffleAsRepeatedMaskAndLanePermute).

This allows us to add lowerV64I8Shuffle handling.

llvm-svn: 352389

5 years ago[CMake] Add code signing for lldb-server on iOS
Stefan Granitz [Mon, 28 Jan 2019 16:57:23 +0000 (16:57 +0000)]
[CMake] Add code signing for lldb-server on iOS

llvm-svn: 352388

5 years ago[CMake] Dump LLDB.framework tools at configuration time
Stefan Granitz [Mon, 28 Jan 2019 16:57:19 +0000 (16:57 +0000)]
[CMake] Dump LLDB.framework tools at configuration time

llvm-svn: 352387

5 years ago[LangRef] Mention vector support for bitreverse/bswap intrinsics (PR38012)
Simon Pilgrim [Mon, 28 Jan 2019 16:56:38 +0000 (16:56 +0000)]
[LangRef] Mention vector support for bitreverse/bswap intrinsics (PR38012)

Differential Revision: https://reviews.llvm.org/D57309

llvm-svn: 352386

5 years ago[libunwind] Drop the dependency on <algorithm>, add placement new inline
Petr Hosek [Mon, 28 Jan 2019 16:44:56 +0000 (16:44 +0000)]
[libunwind] Drop the dependency on <algorithm>, add placement new inline

We haven't eliminated C++ library dependency altogether in D57251,
UnwindCursor.hpp had an unused dependency on <algorithm> which was
pulling in other C++ headers. Removing that dependency also revealed
(correctly) that we need our own global placement new declaration. Now
libunwind should be independent of the C++ library.

Differential Revision: https://reviews.llvm.org/D57262

llvm-svn: 352384

5 years ago[llvm-objdump] - Restore a piece of code removed by mistake in r352366.
George Rimar [Mon, 28 Jan 2019 16:36:12 +0000 (16:36 +0000)]
[llvm-objdump] - Restore a piece of code removed by mistake in r352366.

Seems when committed the r352366
("[llvm-objdump] - Print LMAs when dumping section headers.")
I resolved merge conflict incorrectly and removed this piece by mistake.

Bots did not catch this yet, seems they are slow today,
but the `X86/adjust-vma.test` test case fails locally for me without that.

llvm-svn: 352383

5 years ago[CMake] Quick-Fix targets don't exist when building against LLVM install-tree with...
Stefan Granitz [Mon, 28 Jan 2019 16:15:27 +0000 (16:15 +0000)]
[CMake] Quick-Fix targets don't exist when building against LLVM install-tree with LLDB_INCLUDE_TESTS=ON

The issue came up during release testing for LLVM 8: https://bugs.llvm.org/show_bug.cgi?id=40443

Differential Revision: https://reviews.llvm.org/D57233

llvm-svn: 352382

5 years ago[CMake] Quick-Fix FileCheck target does not exist when building against LLVM install...
Stefan Granitz [Mon, 28 Jan 2019 16:14:57 +0000 (16:14 +0000)]
[CMake] Quick-Fix FileCheck target does not exist when building against LLVM install-tree with COMPILER_RT_INCLUDE_TESTS=ON

The issue came up during release testing for LLVM 8: https://bugs.llvm.org/show_bug.cgi?id=40443

Differential Revision: https://reviews.llvm.org/D57224

llvm-svn: 352381

5 years ago[x86] allow more shuffle splitting to avoid vpermps (PR40434)
Sanjay Patel [Mon, 28 Jan 2019 15:51:34 +0000 (15:51 +0000)]
[x86] allow more shuffle splitting to avoid vpermps (PR40434)

This is tricky to make optimal: sometimes we're better off using
a single wider op, but other times it makes more sense to combine
a narrow ops to achieve the same result.

This solves the case from:
https://bugs.llvm.org/show_bug.cgi?id=40434

There's potentially a similar change for vectors with 64-bit elements,
but it needs adjustments similar to rL352333 to avoid creating infinite
loops.

llvm-svn: 352380

5 years ago[llvm-objdump] - Update test after r352366. NFC.
George Rimar [Mon, 28 Jan 2019 15:49:41 +0000 (15:49 +0000)]
[llvm-objdump] - Update test after r352366. NFC.

Change the column name.

llvm-svn: 352379

5 years agoVERSION_GREATER_EQUAL not supported in llvm cmake.
Ranjeet Singh [Mon, 28 Jan 2019 15:48:07 +0000 (15:48 +0000)]
VERSION_GREATER_EQUAL not supported in llvm cmake.

Patch https://reviews.llvm.org/D56329 caused build failures for me when
building on Windows because of the use of cmake operator
'VERSION_GREATER_EQUAL' which isn't supported in older versions of cmake. The
llvm website states that minimum required version of cmake for building llvm is
3.4.3 https://llvm.org/docs/CMake.html

Differential Revision: https://reviews.llvm.org/D57326

llvm-svn: 352378

5 years agoRemove no longer needed Arm specific words in the LICENSE.txt file.
Arnaud A. de Grandmaison [Mon, 28 Jan 2019 15:42:58 +0000 (15:42 +0000)]
Remove no longer needed Arm specific words in the LICENSE.txt file.

As the codebase is now under the Apache 2.0 license with LLVM
Exceptions, and all Arm's contributions, past or future, are under that
new license, this Arm specific words in LICENSE.txt are no longer
needed.

llvm-svn: 352377

5 years agoRemove no longer needed Arm specific LICENSE.TXT file.
Arnaud A. de Grandmaison [Mon, 28 Jan 2019 15:38:01 +0000 (15:38 +0000)]
Remove no longer needed Arm specific LICENSE.TXT file.

As the codebase is now under the Apache 2.0 license with LLVM
Exceptions, and all Arm's contributions, past or future, are under that
new license, this Arm specific LICENSE.TXT is no longer needed, thus
removing it.

llvm-svn: 352376

5 years ago[cmake] Fix get_llvm_lit_path() to respect LLVM_EXTERNAL_LIT always
Michal Gorny [Mon, 28 Jan 2019 15:16:03 +0000 (15:16 +0000)]
[cmake] Fix get_llvm_lit_path() to respect LLVM_EXTERNAL_LIT always

Refactor the get_llvm_lit_path() logic to respect LLVM_EXTERNAL_LIT,
and require the fallback to be defined explicitly
as LLVM_DEFAULT_EXTERNAL_LIT. This fixes building libcxx standalone
after r346888.

The old logic was using LLVM_EXTERNAL_LIT both as user-defined cache
variable and an optional pre-definition of default value from caller
(e.g. libcxx). It included a hack to make this work by assigning
the value back and forth but it was fragile and stopped working
in libcxx.

The new logic is simpler and more transparent. Default value is
provided in a separate variable, and used only when user-specified
variable is empty (i.e. not overriden).

Differential Revision: https://reviews.llvm.org/D57282

llvm-svn: 352374

5 years ago[obj2yaml] - Dump the sh_entsize section field.
George Rimar [Mon, 28 Jan 2019 15:05:10 +0000 (15:05 +0000)]
[obj2yaml] - Dump the sh_entsize section field.

I faced with the fact that obj2yaml does not dump the sh_entsize field.
A problem arose when I tried to dump ELF versioning sections.

This is close to what D50235 did, but D50235 did the change for yaml2obj, and now
I had to do the same for obj2yaml.

Differential revision: https://reviews.llvm.org/D57229

llvm-svn: 352373

5 years ago[LLD][ELF] - Update tests after r352366.
George Rimar [Mon, 28 Jan 2019 15:03:47 +0000 (15:03 +0000)]
[LLD][ELF] - Update tests after r352366.

r352366 "[llvm-objdump] - Print LMAs when dumping section headers." changed the format of
llvm-objdump output. We have to update the LLD tests.

llvm-svn: 352372

5 years ago[llvm-objcopy] Fix crash when writing empty binary output
Jordan Rupprecht [Mon, 28 Jan 2019 15:02:40 +0000 (15:02 +0000)]
[llvm-objcopy] Fix crash when writing empty binary output

Summary: When using llvm-objcopy -O binary and the resulting file will be empty (e.g. removing the only section that would be written, or using --only-keep with a section that doesn't exist/isn't SHF_ALLOC), we crash because FileOutputBuffer expects Size > 0. Add a regression test, and change Buffer to open/truncate the output file in this case.

Reviewers: alexshap, jhenderson, jakehehrlich, espindola

Reviewed By: alexshap, jhenderson

Subscribers: jfb, llvm-commits, emaste, arichardson

Differential Revision: https://reviews.llvm.org/D56806

llvm-svn: 352371

5 years ago[mips] Support for +abs2008 attribute
Aleksandar Beserminji [Mon, 28 Jan 2019 14:59:30 +0000 (14:59 +0000)]
[mips] Support for +abs2008 attribute

Instruction abs.[ds] is not generating correct result when working
with NaNs for revisions prior mips32r6 and mips64r6.

To generate a sequence which always produce a correct result, but also
to allow user more control on how his code is compiled, attribute
+abs2008 is added, so user can choose legacy or 2008.

By default legacy mode is used on revisions prior R6. Mips32r6 and
mips64r6 use abs2008 mode by default.

Differential Revision: https://reviews.llvm.org/D35983

llvm-svn: 352370

5 years ago[AST] Introduce GenericSelectionExpr::Association
Bruno Ricci [Mon, 28 Jan 2019 14:18:11 +0000 (14:18 +0000)]
[AST] Introduce GenericSelectionExpr::Association

Introduce a new class GenericSelectionExpr::Association which bundle together
an association expression and its TypeSourceInfo.

An iterator GenericSelectionExpr::AssociationIterator is additionally added to
make it possible to iterate over ranges of Associations. This iterator is a
kind of proxy iterator which abstract over how exactly the expressions and the
TypeSourceInfos are stored.

Differential Revision: https://reviews.llvm.org/D57106

Reviewed By: aaron.ballman

Reviewers: aaron.ballman, steveire, dblaikie, mclow.lists
llvm-svn: 352369

5 years ago[clangd] Index main-file macros (bug 39761)
Haojian Wu [Mon, 28 Jan 2019 14:11:49 +0000 (14:11 +0000)]
[clangd] Index main-file macros (bug 39761)

Patch by Nathan Ridge!

Differential Revision: https://reviews.llvm.org/D55739

llvm-svn: 352367

5 years ago[llvm-objdump] - Print LMAs when dumping section headers.
George Rimar [Mon, 28 Jan 2019 14:11:35 +0000 (14:11 +0000)]
[llvm-objdump] - Print LMAs when dumping section headers.

When --section-headers is used, GNU objdump prints both LMA and VMA for sections.
llvm-objdump does not do that what makes it's output be slightly inconsistent.

Patch teaches llvm-objdump to print LMA/VMA for ELF file formats.
The behavior for other formats remains unchanged.

Differential revision: https://reviews.llvm.org/D57146

llvm-svn: 352366

5 years ago[clang-tidy] Fix a build error.
Haojian Wu [Mon, 28 Jan 2019 14:07:45 +0000 (14:07 +0000)]
[clang-tidy] Fix a build error.

llvm-svn: 352364

5 years ago[clang-tidy] Add the abseil-duration-addition check
Hyrum Wright [Mon, 28 Jan 2019 14:03:09 +0000 (14:03 +0000)]
[clang-tidy] Add the abseil-duration-addition check

Differential Revision: https://reviews.llvm.org/D57185

llvm-svn: 352362

5 years ago[clangd] Suggest adding missing includes for incomplete type diagnostics.
Eric Liu [Mon, 28 Jan 2019 14:01:55 +0000 (14:01 +0000)]
[clangd] Suggest adding missing includes for incomplete type diagnostics.

Summary:
This enables clangd to intercept compiler diagnostics and attach fixes (e.g. by
querying index). This patch adds missing includes for incomplete types e.g.
member access into class with only forward declaration. This would allow adding
missing includes for user-typed symbol names that are missing declarations
(e.g. typos) in the future.

Reviewers: sammccall

Reviewed By: sammccall

Subscribers: mgorny, ilya-biryukov, javed.absar, MaskRay, jkorous, mgrang, arphaman, kadircet, cfe-commits

Differential Revision: https://reviews.llvm.org/D56903

llvm-svn: 352361

5 years ago[AMDGPU] Add interpolation builtins
Tim Corringham [Mon, 28 Jan 2019 13:50:37 +0000 (13:50 +0000)]
[AMDGPU] Add interpolation builtins

Summary:
Added builtins for the interpolation intrinsics, and related LIT
test.

Reviewers: arsenm, tpr, dstuttard, #amdgpu

Reviewed By: arsenm, #amdgpu

Subscribers: kzhuravl, wdng, nhaehnle, yaxunl, dstuttard, tpr, t-tye, cfe-commits

Differential Revision: https://reviews.llvm.org/D46871

llvm-svn: 352358

5 years ago[AMDGPU] Add intrinsics for 16 bit interpolation
Tim Corringham [Mon, 28 Jan 2019 13:48:59 +0000 (13:48 +0000)]
[AMDGPU] Add intrinsics for 16 bit interpolation

Summary:
Added the intrinsics llvm.amdgcn.interp.p1.f16() and
llvm.amdgcn.interp.p2.f16() and related LIT test.

The p1 intrinsic generates code appropriate for both 16 and 32
bank LDS.

Reviewers: #amdgpu, dstuttard, arsenm, tpr

Reviewed By: #amdgpu, arsenm

Subscribers: jvesely, mgorny, arsenm, kzhuravl, wdng, nhaehnle, yaxunl, dstuttard, tpr, t-tye, llvm-commits

Differential Revision: https://reviews.llvm.org/D46754

llvm-svn: 352357

5 years ago[opaque pointer types] Remove GraphTraits specialization for Type.
James Y Knight [Mon, 28 Jan 2019 13:25:57 +0000 (13:25 +0000)]
[opaque pointer types] Remove GraphTraits specialization for Type.

The only caller has been deleted in r352076, and I'd like to minimize
the amount of code walking Type hierarchies generically, to make it
easier to identify code depending on pointee types.

llvm-svn: 352353

5 years ago[MIPS GlobalISel] Select sub
Petar Avramovic [Mon, 28 Jan 2019 12:10:17 +0000 (12:10 +0000)]
[MIPS GlobalISel] Select sub

Lower G_USUBO and G_USUBE. Add narrowScalar for G_SUB.
Legalize and select G_SUB for MIPS 32.

Differential Revision: https://reviews.llvm.org/D53416

llvm-svn: 352351

5 years ago[DebugInfo][DAG] Avoid re-ordering of DBG_VALUEs
Jeremy Morse [Mon, 28 Jan 2019 12:08:31 +0000 (12:08 +0000)]
[DebugInfo][DAG] Avoid re-ordering of DBG_VALUEs

This patch improves the placement of DBG_VALUEs when by SelectionDAG, which
as documented in PR40427 can go very wrong. At the core of this is
ProcessSourceNode, which assumes the last instruction in a BB is the start
of the last processed IR instruction, which isn't always true.

Instead, use a helper function to call InstrEmitter::EmitNode, that records
before-and-after iterators and determines the first of any new instruction
created during emission. This is passed to ProcessSourceNode, which can
then make more elightened decisions about ordering for DBG_VALUE placement.

Differential revision: https://reviews.llvm.org/D57163

llvm-svn: 352350

5 years agoRename getTypeQualifiers to getMethodQualifiers.
Anastasia Stulova [Mon, 28 Jan 2019 11:37:49 +0000 (11:37 +0000)]
Rename getTypeQualifiers to getMethodQualifiers.

Use more descriptive name for the method qualifiers getter.

Differential Revision: https://reviews.llvm.org/D56792

llvm-svn: 352349

5 years ago[llvm-objdump] - Fix comment. NFC.
George Rimar [Mon, 28 Jan 2019 10:48:54 +0000 (10:48 +0000)]
[llvm-objdump] - Fix comment. NFC.

This was mentioned by James Henderson
in review for https://reviews.llvm.org/D57051.

llvm-svn: 352348

5 years ago[llvm-objdump] - Implement the --adjust-vma option.
George Rimar [Mon, 28 Jan 2019 10:44:01 +0000 (10:44 +0000)]
[llvm-objdump] - Implement the --adjust-vma option.

GNU objdump's help says: "--adjust-vma: Add OFFSET to all displayed section addresses"
In real life what it does is a bit more complicated
(and IMO not always reasonable. For example, GNU objdump prints not only VMA, but also LMA
for sections. And with --adjust-vma it adjusts LMA, but only when a section has relocations.
llvm-objsump does not seem to support printing LMAs yet, but GNU's logic anyways does not
make sense for me here).

This patch tries to adjust VMA. I tried to implement a reasonable approach.
I am not adjusting sections that are not allocatable. As, for example, adjusting debug sections
VA's and rel[a] sections VA's should not make sense. This behavior seems to be GNU compatible.

Differential revision: https://reviews.llvm.org/D57051

llvm-svn: 352347

5 years ago[ARM GlobalISel] Support integer division for Thumb2
Diana Picus [Mon, 28 Jan 2019 10:37:30 +0000 (10:37 +0000)]
[ARM GlobalISel] Support integer division for Thumb2

Support G_SDIV, G_UDIV, G_SREM and G_UREM.

The only significant difference between arm and thumb mode is that we
need to check a different subtarget feature.

llvm-svn: 352346

5 years ago[AST] Add structural eq tests for template args
Gabor Marton [Mon, 28 Jan 2019 10:01:11 +0000 (10:01 +0000)]
[AST] Add structural eq tests for template args

Summary:
New tests added to verify equivalency of templates when their
parameters are different.

Reviewers: a_sidorin, shafik

Subscribers: rnkovacs, dkrupp, Szelethus, gamesh411, cfe-commits

Differential Revision: https://reviews.llvm.org/D57235

llvm-svn: 352345

5 years ago[X86] Add new variadic avx512 compress/expand intrinsics that use vXi1 types for...
Craig Topper [Mon, 28 Jan 2019 07:03:10 +0000 (07:03 +0000)]
[X86] Add new variadic avx512 compress/expand intrinsics that use vXi1 types for the mask argument.

Custom lower the builtins to these intrinsics. This enables the middle end to optimize out bitcasts for the masks.

llvm-svn: 352344

5 years ago[X86] Add new variadic avx512 compress/expand intrinsics that use vXi1 types for...
Craig Topper [Mon, 28 Jan 2019 07:03:03 +0000 (07:03 +0000)]
[X86] Add new variadic avx512 compress/expand intrinsics that use vXi1 types for the mask argument.

Remove and autoupgrade the old intrinsics

llvm-svn: 352343

5 years ago[X86] Add vbmi2 compressstore and expandload tests that aren't fast-isel tests.
Craig Topper [Mon, 28 Jan 2019 05:42:39 +0000 (05:42 +0000)]
[X86] Add vbmi2 compressstore and expandload tests that aren't fast-isel tests.

These got removed when we autoupgraded to target independent intrinsics, but we didn't have coverage anywhere else. The avx512f/avx512vl versions do have coverage.

Also move some tests back from the upgrade file that aren't really upgraded.

llvm-svn: 352342

5 years ago[CMake] Use __libc_start_main rather than fopen when checking for C library
Petr Hosek [Mon, 28 Jan 2019 04:12:54 +0000 (04:12 +0000)]
[CMake] Use __libc_start_main rather than fopen when checking for C library

The check_library_exists CMake uses a custom symbol definition. This
is a problem when checking for C library symbols because Clang
recognizes many of them as builtins, and returns the
-Wbuiltin-requires-header (or -Wincompatible-library-redeclaration)
error. When building with -Werror which is the default, this causes
the check_library_exists check fail making the build think that C
library isn't available.

To avoid this issue, we should use a symbol that isn't recognized by
Clang and wouldn't cause the same issue. __libc_start_main seems like
reasonable choice that fits the bill.

Differential Revision: https://reviews.llvm.org/D57142

llvm-svn: 352341

5 years ago[AArch64][GlobalISel] Teach RBS about G_FNEG default mapping.
Amara Emerson [Mon, 28 Jan 2019 03:21:14 +0000 (03:21 +0000)]
[AArch64][GlobalISel] Teach RBS about G_FNEG default mapping.

llvm-svn: 352340

5 years agoupdate upcoming meeting issue status
Eric Fiselier [Mon, 28 Jan 2019 02:59:01 +0000 (02:59 +0000)]
update upcoming meeting issue status

llvm-svn: 352339

5 years ago[AArch64][GlobalISel] Add some missing vector support for FP arithmetic ops.
Amara Emerson [Mon, 28 Jan 2019 02:28:22 +0000 (02:28 +0000)]
[AArch64][GlobalISel] Add some missing vector support for FP arithmetic ops.

Moved the fneg lowering legalization test from AArch64 to X86, as we want to
specify that it's already legal.

llvm-svn: 352338

5 years ago[AArch64][GlobalISel] Add some vector support for fp <-> int conversions.
Amara Emerson [Mon, 28 Jan 2019 02:27:59 +0000 (02:27 +0000)]
[AArch64][GlobalISel] Add some vector support for fp <-> int conversions.

Some unrelated, but benign, test changes as well due to the test update script.

llvm-svn: 352337

5 years ago[LLD][COFF] Partial sections
Alexandre Ganea [Mon, 28 Jan 2019 01:45:35 +0000 (01:45 +0000)]
[LLD][COFF] Partial sections

Persist (input) sections that make up an OutputSection. This is a supporting patch for the upcoming D54802.

Differential Revision: https://reviews.llvm.org/D55293

llvm-svn: 352336

5 years agoMake more allocator methods work correctly with in an out-of-process mode.
Dan Liew [Sun, 27 Jan 2019 22:45:11 +0000 (22:45 +0000)]
Make more allocator methods work correctly with in an out-of-process mode.

Summary:
This makes `GetBlockBegin()` and `GetBlockBeginFastLocked()` work correctly with `RemoteAddressSpaceView`.

This has a knock on effect of also making the `PointerIsMine()` and
`GetMetaData()` methods behave correctly when `RemoteAddressSpaceView`
is used to instantiate the allocators.

This will be used by future out-of-process allocator enumeration
patches.

rdar://problem/45284065

Reviewers: kcc, vitalybuka, dvyukov, cryptoad, eugenis, george.karpenkov, yln

Subscribers: #sanitizers, llvm-commits, kubamracek

Differential Revision: https://reviews.llvm.org/D56964

llvm-svn: 352335

5 years agoGlobalISel: Don't reduce elements for atomic load/store
Matt Arsenault [Sun, 27 Jan 2019 22:36:24 +0000 (22:36 +0000)]
GlobalISel: Don't reduce elements for atomic load/store

This is invalid for the same reason as in the narrowScalar handling
for load.

llvm-svn: 352334

5 years ago[x86] add restriction for lowering to vpermps
Sanjay Patel [Sun, 27 Jan 2019 21:53:33 +0000 (21:53 +0000)]
[x86] add restriction for lowering to vpermps

This transform was added with rL351346, and we had
an escape for shufps, but we also want one for
unpckps vs. vpermps because vpermps doesn't take
an immediate shuffle index operand.

llvm-svn: 352333

5 years agoGlobalISel: Factor fewerElementVectors into separate functions
Matt Arsenault [Sun, 27 Jan 2019 21:53:09 +0000 (21:53 +0000)]
GlobalISel: Factor fewerElementVectors into separate functions

llvm-svn: 352332