OSDN Git Service
Justin Lebar [Thu, 9 Jun 2016 20:04:08 +0000 (20:04 +0000)]
[NVPTX] Add intrinsics for shfl instructions.
Summary:
Currently clang emits these instructions via inline (volatile) asm in
the CUDA headers. Switching to intrinsics will let the optimizer reason
across calls to these intrinsics.
Reviewers: tra
Subscribers: llvm-commits, jholewinski
Differential Revision: http://reviews.llvm.org/D21160
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272298
91177308-0d34-0410-b5e6-
96231b3b80d8
Sanjoy Das [Thu, 9 Jun 2016 19:58:30 +0000 (19:58 +0000)]
NFC cleanup of InitializePasses.h
- Alphabetically sort the initializeXXX calls (this was brought up in
D21115)
- Remove repeated function names from doxygen comments
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272297
91177308-0d34-0410-b5e6-
96231b3b80d8
Justin Lebar [Thu, 9 Jun 2016 19:49:26 +0000 (19:49 +0000)]
[NVPTX] Mark bar.sync intrinsic as convergent.
Summary:
__syncthreads, which corresponds to bar.sync 0, is already convergent.
This makes the more general bar.sync n likewise convergent.
Reviewers: tra
Subscribers: llvm-commits, jholewinski
Differential Revision: http://reviews.llvm.org/D21161
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272295
91177308-0d34-0410-b5e6-
96231b3b80d8
Easwaran Raman [Thu, 9 Jun 2016 19:44:46 +0000 (19:44 +0000)]
[PM] Port LCSSA to the new PM.
Differential Revision: http://reviews.llvm.org/D21090
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272294
91177308-0d34-0410-b5e6-
96231b3b80d8
Vedant Kumar [Thu, 9 Jun 2016 19:36:48 +0000 (19:36 +0000)]
Revert "[lit] Use os.devnull instead of named temp files"
This reverts commit r272290. It breaks a test that depends on being able
to seek the /dev/null equivalent on Windows:
http://bb.pgr.jp/builders/ninja-clang-x64-mingw64-RA/builds/11360
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272293
91177308-0d34-0410-b5e6-
96231b3b80d8
Wei Ding [Thu, 9 Jun 2016 19:17:15 +0000 (19:17 +0000)]
AMDGPU/SI: Fix 32-bit fdiv lowering
We were using the fast fdiv lowering for all division, implementation of
IEEE754 fdiv is added.
http://reviews.llvm.org/D20557
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272292
91177308-0d34-0410-b5e6-
96231b3b80d8
Vedant Kumar [Thu, 9 Jun 2016 18:38:41 +0000 (18:38 +0000)]
[lit] Use os.devnull instead of named temp files
Use os.devnull instead of tempfiles when substituting '/dev/null' on
Windows machines. This should make the bots just a bit speedier.
Thanks to Yunzhong Gao for testing this patch on Windows!
Differential Revision: http://reviews.llvm.org/D20549
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272290
91177308-0d34-0410-b5e6-
96231b3b80d8
Michael Kuperstein [Thu, 9 Jun 2016 18:03:15 +0000 (18:03 +0000)]
[LV] Use vector phis for some secondary induction variables
Previously, we materialized secondary vector IVs from the primary scalar IV,
by offseting the primary to match the correct start value, and then broadcasting
it - inside the loop body. Instead, we can use a real vector IV, like we do for
the primary.
This enables using vector IVs for secondary integer IVs whose type matches the
type of the primary.
Differential Revision: http://reviews.llvm.org/D20932
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272283
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Bieneman [Thu, 9 Jun 2016 17:22:02 +0000 (17:22 +0000)]
[CMake] Add LLVM_TOOLS_INSTALL_DIR to LLVMConfig.cmake
This is the more-correct fix to out-of-tree building. AddLLVM.cmake relies on this variable being set, so we should make sure it is set in LLVMConfig.cmake.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272279
91177308-0d34-0410-b5e6-
96231b3b80d8
Davide Italiano [Thu, 9 Jun 2016 17:06:01 +0000 (17:06 +0000)]
Also fix a typo. Need more coffee today.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272278
91177308-0d34-0410-b5e6-
96231b3b80d8
Davide Italiano [Thu, 9 Jun 2016 17:04:38 +0000 (17:04 +0000)]
Improve r272262, check that __stack_chk_guard is used.
Thanks to Rafael for the suggestion.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272277
91177308-0d34-0410-b5e6-
96231b3b80d8
Jan Vesely [Thu, 9 Jun 2016 16:04:00 +0000 (16:04 +0000)]
SelectionDAG: Implement expansion of {S,U}MIN/MAX in integer legalization
Fixes {u,}long_{min,max,clamp} opencl piglit regressions on EG.
Reviewers: arsenm
Differential Revision: http://reviews.llvm.org/D17898
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272272
91177308-0d34-0410-b5e6-
96231b3b80d8
Hans Wennborg [Thu, 9 Jun 2016 15:33:06 +0000 (15:33 +0000)]
PDB/Raw/Hash.h: try to fix VS2013 build
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272269
91177308-0d34-0410-b5e6-
96231b3b80d8
Haicheng Wu [Thu, 9 Jun 2016 15:24:29 +0000 (15:24 +0000)]
Reapply "[MBP] Reduce code size by running tail merging in MBP.""
This reapplies commit r271930, r271915, r271923. They hit a bug in
Thumb which is fixed in r272258 now.
The original message:
The code layout that TailMerging (inside BranchFolding) works on is not the
final layout optimized based on the branch probability. Generally, after
BlockPlacement, many new merging opportunities emerge.
This patch calls Tail Merging after MBP and calls MBP again if Tail Merging
merges anything.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272267
91177308-0d34-0410-b5e6-
96231b3b80d8
Ulrich Weigand [Thu, 9 Jun 2016 15:19:16 +0000 (15:19 +0000)]
[SystemZ] Enable long displacement constraints for inline ASM operands
This enables use of the 'S' constraint for inline ASM operands on
SystemZ, which allows for a memory reference with a signed 20-bit
immediate displacement. This patch includes corresponding documentation
and test case updates.
I've changed the 'T' constraint to match the new behavior for 'S', as
'T' also uses a long displacement (though index constraints are still
not implemented). I also changed 'm' to match the behavior for 'S' as
this will allow for a wider range of displacements for 'm', though
correct me if that's not the right decision.
Author: colpell
Differential Revision: http://reviews.llvm.org/D21097
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272266
91177308-0d34-0410-b5e6-
96231b3b80d8
Davide Italiano [Thu, 9 Jun 2016 15:16:58 +0000 (15:16 +0000)]
Move stackguard test to X86/ directory as it's not generic.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272264
91177308-0d34-0410-b5e6-
96231b3b80d8
Davide Italiano [Thu, 9 Jun 2016 14:23:38 +0000 (14:23 +0000)]
[CodeGen] Change getSDagStackGuard to get an internal sym.
Fixes a crash in the backend during an LTO build of rtld(1) in
FreeBSD.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272262
91177308-0d34-0410-b5e6-
96231b3b80d8
Hrvoje Varga [Thu, 9 Jun 2016 12:57:23 +0000 (12:57 +0000)]
[mips][microMIPS] Implement BOVC, BNVC, EXT, INS and JALRC instructions
Differential Revision: http://reviews.llvm.org/D11798
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272259
91177308-0d34-0410-b5e6-
96231b3b80d8
James Molloy [Thu, 9 Jun 2016 11:51:29 +0000 (11:51 +0000)]
[Thumb] A branch is not part of an IT block
ReplaceTailWithBranchTo assumed that if an instruction is predicated, it must be part of an IT block. This is not correct for conditional branches.
No testcase as this was triggered by the reverted patch r272017 - test coverage will occur when that patch is re-reverted and there is no known way to trigger this in the meantime.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272258
91177308-0d34-0410-b5e6-
96231b3b80d8
Igor Breger [Thu, 9 Jun 2016 11:46:55 +0000 (11:46 +0000)]
[AVX512] Remove masked_move/blendm intrinsic from back-end.
This is complement patch to D21060.
Differential Revision: http://reviews.llvm.org/D21174
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272257
91177308-0d34-0410-b5e6-
96231b3b80d8
Zlatko Buljan [Thu, 9 Jun 2016 11:15:53 +0000 (11:15 +0000)]
[mips][microMIPS] Add CodeGen support for SEL.*, SELEQZ, SELNEZ, SELEQZ.*, SELNEZ.* and CMP.condn.fmt instructions
Differential Revision: http://reviews.llvm.org/D20862
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272256
91177308-0d34-0410-b5e6-
96231b3b80d8
Sam Kolton [Thu, 9 Jun 2016 11:04:45 +0000 (11:04 +0000)]
[AMDGPU] Disassembler: Support for sdwa instructions
Reviewers: vpykhtin, tstellarAMD
Subscribers: arsenm, kzhuravl
Differential Revision: http://reviews.llvm.org/D21129
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272255
91177308-0d34-0410-b5e6-
96231b3b80d8
Diana Picus [Thu, 9 Jun 2016 10:31:05 +0000 (10:31 +0000)]
[llc] Remove exit-on-error flag from MIR tests (PR27770)
This is made possible by removing an assert in llc that assumed
MIRParser::parseLLVMModule would exit on error. MIRParser's documentation states
that it returns null if a parsing error occurs, so there's no reason to assert.
We can instead just fall through to where the check for a module is performed
and exit if it is null.
This commit is part of the clean-up after r269655.
Fixes PR27770
Differential Revision: http://reviews.llvm.org/D20371
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272254
91177308-0d34-0410-b5e6-
96231b3b80d8
Craig Topper [Thu, 9 Jun 2016 07:49:08 +0000 (07:49 +0000)]
[AVX512] Fix shuffle decode printing for several instructions with write masks. There are still more bugs here with UNPCK and PALIGN for sure. But these were the easiest ones to fix.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272252
91177308-0d34-0410-b5e6-
96231b3b80d8
James Molloy [Thu, 9 Jun 2016 07:39:08 +0000 (07:39 +0000)]
[Thumb] Select a BIC instead of AND if the immediate can be encoded more optimally negated
If an immediate is only used in an AND node, it is possible that the immediate can be more optimally materialized when negated. If this is the case, we can negate the immediate and use a BIC instead;
int i(int a) {
return a & 0xfffffeec;
}
Used to produce:
ldr r1, [CONSTPOOL]
ands r0, r1
CONSTPOOL: 0xfffffeec
And now produces:
movs r1, #255
adds r1, #20 ; Less costly immediate generation
bics r0, r1
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272251
91177308-0d34-0410-b5e6-
96231b3b80d8
Craig Topper [Thu, 9 Jun 2016 07:10:34 +0000 (07:10 +0000)]
[X86] Fix a test I failed to re-generate in r272249.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272250
91177308-0d34-0410-b5e6-
96231b3b80d8
Craig Topper [Thu, 9 Jun 2016 07:06:38 +0000 (07:06 +0000)]
[X86] Bring consistent naming to the SSE/AVX and AVX512 PALIGNR instructions. Then add shuffle decode printing for the EVEX forms which is made easier by having the naming structure more similar to other instructions.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272249
91177308-0d34-0410-b5e6-
96231b3b80d8
Craig Topper [Thu, 9 Jun 2016 07:06:33 +0000 (07:06 +0000)]
[X86] Fix bad comment in assert. NFC
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272248
91177308-0d34-0410-b5e6-
96231b3b80d8
Xinliang David Li [Thu, 9 Jun 2016 03:22:39 +0000 (03:22 +0000)]
Revert r272194 No need for it if loop Analysis Manager is used
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272243
91177308-0d34-0410-b5e6-
96231b3b80d8
Saleem Abdulrasool [Thu, 9 Jun 2016 03:14:32 +0000 (03:14 +0000)]
test: fix typo
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272242
91177308-0d34-0410-b5e6-
96231b3b80d8
Saleem Abdulrasool [Thu, 9 Jun 2016 02:56:40 +0000 (02:56 +0000)]
AArch64: support the `.arch` directive in the IAS
Add support to the AArch64 IAS for the `.arch` directive. This allows the
assembly input to use architectural functionality in part of a file. This is
used in existing code like BoringSSL.
Resolves PR26016!
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272241
91177308-0d34-0410-b5e6-
96231b3b80d8
Kostya Serebryany [Thu, 9 Jun 2016 01:20:35 +0000 (01:20 +0000)]
[libFuzzer] add one more OOM test, which we currently don't handle very well
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272240
91177308-0d34-0410-b5e6-
96231b3b80d8
Teresa Johnson [Thu, 9 Jun 2016 01:14:13 +0000 (01:14 +0000)]
[ThinLTO/gold] Enable summary-based internalization
Summary: Enable existing summary-based importing support in the gold-plugin.
Reviewers: mehdi_amini
Subscribers: llvm-commits, mehdi_amini
Differential Revision: http://reviews.llvm.org/D21080
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272239
91177308-0d34-0410-b5e6-
96231b3b80d8
Sanjoy Das [Thu, 9 Jun 2016 01:14:03 +0000 (01:14 +0000)]
Minor clean up in loopHasNoAbnormalExits; NFC
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272238
91177308-0d34-0410-b5e6-
96231b3b80d8
Sanjoy Das [Thu, 9 Jun 2016 01:13:59 +0000 (01:13 +0000)]
Be wary of abnormal exits from loop when exploiting UB
We can safely rely on a NoWrap add recurrence causing UB down the road
only if we know the loop does not have a exit expressed in a way that is
opaque to ScalarEvolution (e.g. by a function call that conditionally
calls exit(0)).
I believe with this change PR28012 is fixed.
Note: I had to change some llvm-lit tests in LoopReroll, since it looks
like they were depending on this incorrect behavior.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272237
91177308-0d34-0410-b5e6-
96231b3b80d8
Sanjoy Das [Thu, 9 Jun 2016 01:13:54 +0000 (01:13 +0000)]
Factor out a loopHasNoAbnormalExits; NFC
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272236
91177308-0d34-0410-b5e6-
96231b3b80d8
Richard Smith [Thu, 9 Jun 2016 00:53:21 +0000 (00:53 +0000)]
Search for llvm-symbolizer binary in the same directory as argv[0], before
looking for it along $PATH. This allows installs of LLVM tools outside of
$PATH to find the symbolizer and produce pretty backtraces if they crash.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272232
91177308-0d34-0410-b5e6-
96231b3b80d8
Reid Kleckner [Thu, 9 Jun 2016 00:29:00 +0000 (00:29 +0000)]
[codeview] Skip DIGlobalVariables with no variable
They have probably been discarded during optimization.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272231
91177308-0d34-0410-b5e6-
96231b3b80d8
Zachary Turner [Thu, 9 Jun 2016 00:21:37 +0000 (00:21 +0000)]
[pdbdump-fuzzer] Try to fix build errors in fuzzer.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272230
91177308-0d34-0410-b5e6-
96231b3b80d8
Rui Ueyama [Thu, 9 Jun 2016 00:10:19 +0000 (00:10 +0000)]
[pdbdump] Verify part of TPI hash streams.
TPI hash table contains a parallel array for the type records.
For each type record R, a hash value is calculated by `H(R) % NumBuckets`
where H is a hash function, and the result is stored to a bucket element.
H is TPI1::hashPrec function in microsoft-pdb repository.
Our hash function does not support all type record types yet.
Currently it supports only records for line number.
I'll extend it in a follow up patch.
The aim of verify the hash table is not only detect corrupted files.
It ensures that our understanding of how the hash values are calculated
is correct.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272229
91177308-0d34-0410-b5e6-
96231b3b80d8
Alina Sbirlea [Thu, 9 Jun 2016 00:08:15 +0000 (00:08 +0000)]
[cpu-detection] Add missing break statements in outer switches
Summary:
Break on all switch cases for outer and inner switches.
No functionality changed.
Reviewers: llvm-commits, sanjoy
Differential Revision: http://reviews.llvm.org/D21158
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272228
91177308-0d34-0410-b5e6-
96231b3b80d8
Quentin Colombet [Wed, 8 Jun 2016 23:27:46 +0000 (23:27 +0000)]
[MIR] Check that generic virtual registers get a size.
Without that check it was possible to write test cases where the size
was not specified and we ended up with weird asserts down the road,
because the default value (1) would not make sense.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272226
91177308-0d34-0410-b5e6-
96231b3b80d8
Rui Ueyama [Wed, 8 Jun 2016 23:15:09 +0000 (23:15 +0000)]
Function names should start with lowercase letters.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272225
91177308-0d34-0410-b5e6-
96231b3b80d8
Michael Zolotukhin [Wed, 8 Jun 2016 23:13:21 +0000 (23:13 +0000)]
[LoopSimplify] Preserve LCSSA when merging exit blocks.
Summary:
This fixes PR26682. Also add LCSSA as a preserved pass to LoopSimplify,
that looks correct to me and allows to write a test for the issue.
Reviewers: chandlerc, bogner, sanjoy
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D21112
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272224
91177308-0d34-0410-b5e6-
96231b3b80d8
Rui Ueyama [Wed, 8 Jun 2016 23:11:14 +0000 (23:11 +0000)]
[PDB] Move PDB functions to a separate file.
We are going to use the hash functions from TPI streams.
Differential Revision: http://reviews.llvm.org/D21142
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272223
91177308-0d34-0410-b5e6-
96231b3b80d8
Michael Zolotukhin [Wed, 8 Jun 2016 22:49:59 +0000 (22:49 +0000)]
[LoopUnroll] Check that DT is available before trying to verify it.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272221
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Bieneman [Wed, 8 Jun 2016 22:48:12 +0000 (22:48 +0000)]
[CMake] Cleanup version check for 2.8.11
We are always greater than CMake 2.8.11, so we don't need this check.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272220
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Bieneman [Wed, 8 Jun 2016 22:48:01 +0000 (22:48 +0000)]
[CMake] Cleanup version check for 2.8.12
Since we're always greater than 2.8.12, we don't need this check anymore.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272219
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Bieneman [Wed, 8 Jun 2016 22:38:30 +0000 (22:38 +0000)]
[CMake] Cleanup version checks for CMake 3.0
This just removes some redundant checks and updates warning text.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272218
91177308-0d34-0410-b5e6-
96231b3b80d8
Justin Bogner [Wed, 8 Jun 2016 22:36:37 +0000 (22:36 +0000)]
cmake: Simplify add_lit_testsuites
cmake 3.4 introduced LIST_DIRECTORIES to glob recurse, which can be
used to simplify this code greatly.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272217
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Bieneman [Wed, 8 Jun 2016 22:19:25 +0000 (22:19 +0000)]
[CMake] Cleanup uses of USES_TERMINAL
Now that we are on CMake 3.4.3 we no longer need a version check around this.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272211
91177308-0d34-0410-b5e6-
96231b3b80d8
Quentin Colombet [Wed, 8 Jun 2016 21:55:30 +0000 (21:55 +0000)]
[RegBankSelect] Print out the actual mapping of the operands.
This improves the debuggability of the pass.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272210
91177308-0d34-0410-b5e6-
96231b3b80d8
Quentin Colombet [Wed, 8 Jun 2016 21:55:29 +0000 (21:55 +0000)]
[RegBankSelect] Remove a debug print of a potentially dead instruction.
For complex rewrittings, which do not occur currently, the related
machine instruction may have been deleted in the process. Therefore, do
not try to print it after the mapping is applied.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272209
91177308-0d34-0410-b5e6-
96231b3b80d8
Quentin Colombet [Wed, 8 Jun 2016 21:55:26 +0000 (21:55 +0000)]
[RegisterBankInfo] Avoid code duplication in OperandsMapper for the computation of the end of range.
Refactor the code so that we do not compute in two different places the
end iterator for the range of new virtual registers for a given operand.
Although this refactoring was intended as NFC, this is not the case
because it actually fixes a bug where we were returning a range off by 1
(too long). Right now, this could not result in an actual bug because we
were accessing this range via the BreakDown size of the related operand.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272208
91177308-0d34-0410-b5e6-
96231b3b80d8
Quentin Colombet [Wed, 8 Jun 2016 21:55:23 +0000 (21:55 +0000)]
[RegisterBankInfo] Add dump/print methods for OperandsMapper.
Improve debuggability of the OperandsMapper helper class.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272207
91177308-0d34-0410-b5e6-
96231b3b80d8
Michael Zolotukhin [Wed, 8 Jun 2016 21:55:16 +0000 (21:55 +0000)]
[SLPVectorizer] Handle GEP with differing constant index types
Summary:
This fixes PR27617.
Bug description: The SLPVectorizer asserts on encountering GEPs with different index types, such as i8 and i64.
The patch includes a simple relaxation of the assert to allow constants being of different types, along with a regression test that will provoke the unrelaxed assert.
Reviewers: nadav, mzolotukhin
Subscribers: JesperAntonsson, llvm-commits, mzolotukhin
Differential Revision: http://reviews.llvm.org/D20685
Patch by Jesper Antonsson!
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272206
91177308-0d34-0410-b5e6-
96231b3b80d8
Davide Italiano [Wed, 8 Jun 2016 21:31:55 +0000 (21:31 +0000)]
[PM] Add missing caching of GlobalsAA to EarlyCSE.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272204
91177308-0d34-0410-b5e6-
96231b3b80d8
Dehao Chen [Wed, 8 Jun 2016 21:30:12 +0000 (21:30 +0000)]
Revive reviews.llvm.org/D12778 to handle forward-hot-prob and backward-hot-prob consistently.
Summary:
Consider the following diamond CFG:
A
/ \
B C
\/
D
Suppose A->B and A->C have probabilities 81% and 19%. In block-placement, A->B is called a hot edge and the final placement should be ABDC. However, the current implementation outputs ABCD. This is because when choosing the next block of B, it checks if Freq(C->D) > Freq(B->D) * 20%, which is true (if Freq(A) = 100, then Freq(B->D) = 81, Freq(C->D) = 19, and 19 > 81*20%=16.2). Actually, we should use 25% instead of 20% as the probability here, so that we have 19 < 81*25%=20.25, and the desired ABDC layout will be generated.
Reviewers: djasper, davidxl
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D20989
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272203
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Bieneman [Wed, 8 Jun 2016 21:22:03 +0000 (21:22 +0000)]
[CMake] Fixing a typo
This was called out on the list a long time ago and just got pointed out to me again. Need to fix it before I forget.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272201
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Bieneman [Wed, 8 Jun 2016 21:19:26 +0000 (21:19 +0000)]
[CMake] Support overriding binary install directory
This patch adds a new option LLVM_TOOLS_INSTALL_DIR which allows customizing the location executables and symlinks get installed to. This adds the functionality provided by autoconf's --bindir flag.
This patch is based on patches from and collaboration with Tony Kelman, and replaces http://reviews.llvm.org/D20934.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272200
91177308-0d34-0410-b5e6-
96231b3b80d8
Sanjay Patel [Wed, 8 Jun 2016 21:10:01 +0000 (21:10 +0000)]
[InstCombine] move fold of select of add/sub to helper function; NFCI
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272199
91177308-0d34-0410-b5e6-
96231b3b80d8
Reid Kleckner [Wed, 8 Jun 2016 20:34:29 +0000 (20:34 +0000)]
[DebugInfo] Add calling convention support for DWARF and CodeView
Summary:
Now DISubroutineType has a 'cc' field which should be a DW_CC_ enum. If
it is present and non-zero, the backend will emit it as a
DW_AT_calling_convention attribute. On the CodeView side, we translate
it to the appropriate enum for the LF_PROCEDURE record.
I added a new LLVM vendor specific enum to the list of DWARF calling
conventions. DWARF does not appear to attempt to standardize these, so I
assume it's OK to do this until we coordinate with GCC on how to emit
vectorcall convention functions.
Reviewers: dexonsmith, majnemer, aaboud, amccarth
Subscribers: mehdi_amini, llvm-commits
Differential Revision: http://reviews.llvm.org/D21114
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272197
91177308-0d34-0410-b5e6-
96231b3b80d8
Sanjay Patel [Wed, 8 Jun 2016 20:31:52 +0000 (20:31 +0000)]
[InstCombine] fix outdated comment, simplify logic; NFCI
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272196
91177308-0d34-0410-b5e6-
96231b3b80d8
Evgeny Stupachenko [Wed, 8 Jun 2016 20:21:24 +0000 (20:21 +0000)]
The patch set unroll disable pragma when unroll
with user specified count has been applied.
Summary:
Previously SetLoopAlreadyUnrolled() set the disable pragma only if
there was some loop metadata.
Now it set the pragma in all cases. This helps to prevent multiple
unroll when -unroll-count=N is given.
Reviewers: mzolotukhin
Differential Revision: http://reviews.llvm.org/D20765
From: Evgeny Stupachenko <evstupac@gmail.com>
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272195
91177308-0d34-0410-b5e6-
96231b3b80d8
Xinliang David Li [Wed, 8 Jun 2016 20:15:37 +0000 (20:15 +0000)]
[PM] Refector LoopAccessInfo analysis code
This is the preparation patch to port the analysis to new PM
Differential Revision: http://reviews.llvm.org/D20560
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272194
91177308-0d34-0410-b5e6-
96231b3b80d8
Sanjay Patel [Wed, 8 Jun 2016 20:09:04 +0000 (20:09 +0000)]
[InstCombine] reduce indent; NFC
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272193
91177308-0d34-0410-b5e6-
96231b3b80d8
Tim Shen [Wed, 8 Jun 2016 19:42:32 +0000 (19:42 +0000)]
[MemCpyOpt] Do not exchange llvm.lifetime.start and llvm.memcpy
Reviewers: iteratee
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D21087
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272192
91177308-0d34-0410-b5e6-
96231b3b80d8
Sanjay Patel [Wed, 8 Jun 2016 19:33:52 +0000 (19:33 +0000)]
[InstCombine] use copyIRFlags() ; NFCI
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272191
91177308-0d34-0410-b5e6-
96231b3b80d8
Benjamin Kramer [Wed, 8 Jun 2016 19:09:22 +0000 (19:09 +0000)]
Apply most suggestions of clang-tidy's performance-unnecessary-value-param
Avoids unnecessary copies. All changes audited & pass tests with asan.
No functional change intended.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272190
91177308-0d34-0410-b5e6-
96231b3b80d8
Adrian McCarthy [Wed, 8 Jun 2016 18:22:59 +0000 (18:22 +0000)]
Generate codeview for array type metadata.
Differential Revision: http://reviews.llvm.org/D21107
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272187
91177308-0d34-0410-b5e6-
96231b3b80d8
Benjamin Kramer [Wed, 8 Jun 2016 18:19:38 +0000 (18:19 +0000)]
[CodeView] Remove manual expansion of the default copy ctor.
It provides nothing over the default one but makes the class not
trivially copyable. No functionality change intended.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272186
91177308-0d34-0410-b5e6-
96231b3b80d8
George Burgess IV [Wed, 8 Jun 2016 17:56:35 +0000 (17:56 +0000)]
Attempt #2 to appease the buildbots.
MSVC calls the copy ctor on StratifiedSets for some reason. So,
undelete it.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272184
91177308-0d34-0410-b5e6-
96231b3b80d8
Reid Kleckner [Wed, 8 Jun 2016 17:50:29 +0000 (17:50 +0000)]
[codeview] Avoid emitting an empty file checksum table
Again, the Microsoft linker does not like empty substreams.
We still emit an empty string table if CodeView is enabled, but that
doesn't cause problems because it always contains at least one null
byte.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272183
91177308-0d34-0410-b5e6-
96231b3b80d8
Sanjoy Das [Wed, 8 Jun 2016 17:48:46 +0000 (17:48 +0000)]
[SCEV] Break out of loop if there is no more work to do
This is NFC as far as externally visible behavior is concerned, but will
keep us from spinning in the worklist traversal algorithm unnecessarily.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272182
91177308-0d34-0410-b5e6-
96231b3b80d8
Sanjoy Das [Wed, 8 Jun 2016 17:48:42 +0000 (17:48 +0000)]
[SCEV] Track no-abnormal-exits instead of no-throw calls
Absence of may-unwind calls is not enough to guarantee that a
UB-generating use of an add-rec poison in the loop latch will actually
cause UB. We also need to guard against calls that terminate the thread
or infinite loop themselves.
This partially addresses PR28012.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272181
91177308-0d34-0410-b5e6-
96231b3b80d8
Sanjoy Das [Wed, 8 Jun 2016 17:48:36 +0000 (17:48 +0000)]
Teach isGuarantdToTransferExecToSuccessor about debug info intrinsics
Calls to `@llvm.dbg.*` can be assumed to terminate.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272180
91177308-0d34-0410-b5e6-
96231b3b80d8
Sanjoy Das [Wed, 8 Jun 2016 17:48:31 +0000 (17:48 +0000)]
Fix a bug in SCEV's poison value propagation
The worklist algorithm introduced in rL271151 didn't check to see if the
direct users of the post-inc add recurrence propagates poison. This
change fixes the problem and makes the code structure more obvious.
Note for release managers: correctness wise, this bug wasn't a
regression introduced by rL271151 -- the behavior of SCEV around
post-inc add recurrences was strictly improved (in terms of correctness)
in rL271151.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272179
91177308-0d34-0410-b5e6-
96231b3b80d8
Quentin Colombet [Wed, 8 Jun 2016 17:39:47 +0000 (17:39 +0000)]
[RegBankSelect] Silence an unused variable warning in release mode.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272177
91177308-0d34-0410-b5e6-
96231b3b80d8
Quentin Colombet [Wed, 8 Jun 2016 17:39:43 +0000 (17:39 +0000)]
[RegBankSelect] Comment on how we could improve repairing with copies.
When repairing with a copy, instead of accounting for the cost of that
copy and actually inserting it, we may be able to use an alternative
source for the register to repair and just use it.
Make sure this is documented, so that we consider that opportunity at
some point.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272176
91177308-0d34-0410-b5e6-
96231b3b80d8
Zachary Turner [Wed, 8 Jun 2016 17:32:25 +0000 (17:32 +0000)]
[pdb] Fix build errors in PDB unit tests.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272174
91177308-0d34-0410-b5e6-
96231b3b80d8
George Burgess IV [Wed, 8 Jun 2016 17:27:14 +0000 (17:27 +0000)]
Try to appease buildbots.
r272064 apparently made them angry. This undoes some changes made in
r272064 (defaulting move ctors) to make them happy again.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272173
91177308-0d34-0410-b5e6-
96231b3b80d8
Zachary Turner [Wed, 8 Jun 2016 17:26:39 +0000 (17:26 +0000)]
[pdb] Handle stream index errors better.
Reviewed By: ruiu
Differential Revision: http://reviews.llvm.org/D21128
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272172
91177308-0d34-0410-b5e6-
96231b3b80d8
Rui Ueyama [Wed, 8 Jun 2016 16:54:31 +0000 (16:54 +0000)]
Remove a patch .rej file.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272171
91177308-0d34-0410-b5e6-
96231b3b80d8
Quentin Colombet [Wed, 8 Jun 2016 16:53:32 +0000 (16:53 +0000)]
[AArch64][RegisterBankInfo] G_OR are fine on either GPR or FPR.
Teach AArch64RegisterBankInfo that G_OR can be mapped on either GPR or
FPR for 64-bit or 32-bit values.
Add test cases demonstrating how this information is used to coalesce a
computation on a single register bank.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272170
91177308-0d34-0410-b5e6-
96231b3b80d8
Quentin Colombet [Wed, 8 Jun 2016 16:45:04 +0000 (16:45 +0000)]
[RegBankSelect] Use RegisterBankInfo applyMapping method.
The RegBankSelect pass can now rely on the target to do the remapping of
the instructions.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272169
91177308-0d34-0410-b5e6-
96231b3b80d8
Vedant Kumar [Wed, 8 Jun 2016 16:39:32 +0000 (16:39 +0000)]
[ProfileData] Update llvm's copy of InstrProfData.inc
The new version of the header introduces the INSTR_PROF_VISIBILITY
macro. See http://reviews.llvm.org/D21116 for more details.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272166
91177308-0d34-0410-b5e6-
96231b3b80d8
Quentin Colombet [Wed, 8 Jun 2016 16:39:21 +0000 (16:39 +0000)]
[RegisterBankInfo] Implement the method to apply a mapping.
Now, the target will be able to provide its how implementation to remap
an instruction. This open the way to crazier optimizations, but to
beginning with, we will be able to handle something else than the
default mapping.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272165
91177308-0d34-0410-b5e6-
96231b3b80d8
Quentin Colombet [Wed, 8 Jun 2016 16:30:55 +0000 (16:30 +0000)]
[RegBankSelect] Use the OperandMapper class to hold remap information.
Now that we have an entity that hold the remap information the
rewritting should be easier to do.
No functional changes.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272164
91177308-0d34-0410-b5e6-
96231b3b80d8
Quentin Colombet [Wed, 8 Jun 2016 16:24:55 +0000 (16:24 +0000)]
[RegBankSelect] Use const_iterator instead of iterator for repairReg.
The repairing code has no reason to change the source or destination of
the registers.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272163
91177308-0d34-0410-b5e6-
96231b3b80d8
Quentin Colombet [Wed, 8 Jun 2016 16:18:13 +0000 (16:18 +0000)]
[RegisterBankInfo] Introduce OperandsMapper class.
This helper class is used to encapsulate the necessary information
to remap an instruction.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272161
91177308-0d34-0410-b5e6-
96231b3b80d8
Quentin Colombet [Wed, 8 Jun 2016 16:12:19 +0000 (16:12 +0000)]
[Target] Introduce a generic opcode for bitwise OR: G_OR.
This G_OR is used in GlobalISel to represent bitwise OR.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272160
91177308-0d34-0410-b5e6-
96231b3b80d8
Quentin Colombet [Wed, 8 Jun 2016 15:49:23 +0000 (15:49 +0000)]
[RegBankSelect] Introduce a command line option to override the running mode.
When the command line option is set, it overrides any thing that the
target may have set. The rationale is that we get what we asked for.
Options are respectively regbankselect-fast and regbankselect-greedy for
fast and greedy mode.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272158
91177308-0d34-0410-b5e6-
96231b3b80d8
Quentin Colombet [Wed, 8 Jun 2016 15:40:32 +0000 (15:40 +0000)]
[RegBankSelect] Explain what it would take to support non-copy
repairing.
Copies are easy because we repair only when there is a mismatch. For
non-copy repairing, i.e., cases that involves breaking down or gathering
up the value, one of the operand may not have a register bank yet. Thus,
derivate a cost from that, requires more work.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272157
91177308-0d34-0410-b5e6-
96231b3b80d8
Oliver Stannard [Wed, 8 Jun 2016 15:26:34 +0000 (15:26 +0000)]
[ARM] MSR instructions implicitly set CPSR
The MSR instructions can write to the CPSR, but we did not model this
fact, so we could emit them in the middle of IT blocks, changing the
condition flags for later instructions in the block.
The tests use two calls to llvm.write_register.i32 because it is valid
to use these instructions at the end of an IT block, which if conversion
does do in some cases. With two calls, the first clobbers the flags, so
a branch has to be used to make the second one conditional.
Differential Revision: http://reviews.llvm.org/D21139
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272154
91177308-0d34-0410-b5e6-
96231b3b80d8
Daniel Dunbar [Wed, 8 Jun 2016 14:41:44 +0000 (14:41 +0000)]
[lit] Ensure we get bytes when reading redirected output files.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272147
91177308-0d34-0410-b5e6-
96231b3b80d8
Saleem Abdulrasool [Wed, 8 Jun 2016 14:30:00 +0000 (14:30 +0000)]
Support: correct AArch64 TargetParser implementation
The architecture enumeration is shared across ARM and AArch64. However, the
data is not. The code incorrectly would index into the array using the
architecture index which was offset by the ARMv7 architecture enumeration. We
do not have a marker for indicating the architectural family to which the
enumeration belongs so we cannot be clever about offsetting the index (at least
it is not immediately apparent to me). Instead, fall back to the tried-and-true
method of slowly iterating the array (its not a large array, so the impact of
this is not too high).
Because of the incorrect indexing, if we were lucky, we would crash, but usually
we would return an invalid StringRef. We did not have any tests for the AArch64
target parser previously;. Extend the previous tests I had added for ARM to
cover AArch64 for ensuring that we return expected StringRefs.
Take the opportunity to change some iterator types to references.
This work is needed to support parsing `.arch name` directives in the AArch64
target asm parser.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272145
91177308-0d34-0410-b5e6-
96231b3b80d8
Davide Italiano [Wed, 8 Jun 2016 13:56:59 +0000 (13:56 +0000)]
[PM] LoopSimplify. Remove unneeded pass dependencies. NFCI.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272140
91177308-0d34-0410-b5e6-
96231b3b80d8
Davide Italiano [Wed, 8 Jun 2016 13:32:23 +0000 (13:32 +0000)]
[PM/SimplifyCFG] Preserve GlobalsAA even if the IR is mutated.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272139
91177308-0d34-0410-b5e6-
96231b3b80d8
Vasileios Kalintiris [Wed, 8 Jun 2016 13:13:15 +0000 (13:13 +0000)]
[mips] Add a proper file header in MipsFastISel.cpp
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272138
91177308-0d34-0410-b5e6-
96231b3b80d8
Krzysztof Parzyszek [Wed, 8 Jun 2016 12:31:16 +0000 (12:31 +0000)]
[Hexagon] Modify HexagonExpandCondsets to handle subregisters
Also, switch to using functions from LiveIntervalAnalysis to update
live intervals, instead of performing the updates manually.
Re-committing r272045.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272135
91177308-0d34-0410-b5e6-
96231b3b80d8