Commit Graph

95449 Commits

Author SHA1 Message Date
Sanjoy Das 1f7b813e2b Remove duplicated code; NFC
ICmpInst::makeConstantRange does exactly the same thing as
ConstantRange::makeExactICmpRegion.

llvm-svn: 283059
2016-10-02 00:09:57 +00:00
Sanjoy Das 1b9cefcf03 [SCEV] Rely on ConstantRange instead of custom logic; NFCI
llvm-svn: 283058
2016-10-02 00:09:52 +00:00
Sanjoy Das 6ef69d97f5 [ConstantRange] Make getEquivalentICmp smarter
This change teaches getEquivalentICmp to be smarter about generating
ICMP_NE and ICMP_EQ predicates.

llvm-svn: 283057
2016-10-02 00:09:49 +00:00
Sanjoy Das 54e6a21dca [SCEV] Remove commented out code; NFC
llvm-svn: 283056
2016-10-02 00:09:45 +00:00
Simon Pilgrim 630dd6ff02 [X86][SSE] Cleaned up shuffle decode assertion messages
llvm-svn: 283050
2016-10-01 20:12:56 +00:00
Mehdi Amini 99d1b29503 Use StringRef for MemoryBuffer identifier API (NFC)
llvm-svn: 283043
2016-10-01 16:38:28 +00:00
Simon Pilgrim 5b0c15ddf7 Fix signed/unsigned warning
llvm-svn: 283041
2016-10-01 16:14:57 +00:00
Simon Pilgrim 1638d49f20 [X86][SSE] Add support for combining target shuffles to binary BLEND
We already had support for 1-input BLEND with zero - this adds support for 2-input BLEND as well.

llvm-svn: 283040
2016-10-01 16:04:28 +00:00
Mehdi Amini 7410717a62 Use StringRef in Registry API (NFC)
llvm-svn: 283039
2016-10-01 15:44:54 +00:00
Simon Pilgrim ae17cf20ce [X86][SSE] Always combine target shuffles to MOVSD/MOVSS
Now we can commute to BLENDPD/BLENDPS on SSE41+ targets if necessary, so simplify the combine matching where we can.

This required me to add a couple of scalar math movsd/moss fold patterns that hadn't been needed in the past.

llvm-svn: 283038
2016-10-01 15:33:01 +00:00
Simon Pilgrim ccdd1ff49b [X86][SSE] Enable commutation from MOVSD/MOVSS to BLENDPD/BLENDPS on SSE41+ targets
Instead of selecting between MOVSD/MOVSS and BLENDPD/BLENDPS at shuffle lowering by subtarget this will help us select the instruction based on actual commutation requirements.

We could possibly add BLENDPD/BLENDPS -> MOVSD/MOVSS commutation and MOVSD/MOVSS memory folding using a similar approach if it proves useful

I avoided adding AVX512 handling as I'm not sure when we should be making use of VBLENDPD/VBLENDPS on EVEX targets

llvm-svn: 283037
2016-10-01 14:26:11 +00:00
Nirav Dave e4c6153cf1 Revert "[MC] Prevent out of order HashDirective lexing in AsmLexer."
This reverts commit r282992 which appears to be causing an LTO test failure.

llvm-svn: 283034
2016-10-01 10:57:55 +00:00
Kostya Serebryany a5f1adab56 [libFuzzer] add fuzzer test for libxml2, finds https://bugzilla.gnome.org/show_bug.cgi?id=751631
llvm-svn: 283024
2016-10-01 07:37:40 +00:00
Kostya Serebryany d1f31d0a49 [libFuzzer] fix a recent bugs (buffer overflow)
llvm-svn: 283021
2016-10-01 07:13:25 +00:00
Craig Topper 5eb5ade894 [X86] Cleanup patterns for using VMOVDDUP for broadcasts.
-Remove OptForSize. Not all of the backend follows the same rules for creating broadcasts and there is no conflicting pattern.
-Don't stop selecting VEX VMOVDDUP when AVX512 is supported. We need VLX for EVEX VMOVDDUP.
-Only use VMOVDDUP for v2i64 broadcasts if AVX2 is not supported.

llvm-svn: 283020
2016-10-01 07:11:24 +00:00
Mehdi Amini 9af9a9d5f9 Revert "Use StringRef instead of raw pointer in TargetRegistry API (NFC)"
This reverts commit r283017. Creates an infinite loop somehow.

llvm-svn: 283019
2016-10-01 07:08:23 +00:00
Mehdi Amini 36d33fc109 Use StringRef instead of raw pointers in MCAsmInfo/MCInstrInfo APIs (NFC)
llvm-svn: 283018
2016-10-01 06:46:33 +00:00
Mehdi Amini cd354a659b Use StringRef instead of raw pointer in TargetRegistry API (NFC)
llvm-svn: 283017
2016-10-01 06:25:30 +00:00
Mehdi Amini 7419e940d2 Use StringRef instead of raw pointer in ExecutionEngine
llvm-svn: 283016
2016-10-01 06:22:04 +00:00
Craig Topper be351eea0c [AVX-512] Add EVEX versions of VPBROADCASTW patterns with truncated i32 loads.
llvm-svn: 283015
2016-10-01 06:01:23 +00:00
Mehdi Amini 48878ae579 Use StringRef in Datalayout API (NFC)
llvm-svn: 283013
2016-10-01 05:57:55 +00:00
Mehdi Amini f42ec7903f DIFlags: use StringRef instead of raw pointer (NFC)
llvm-svn: 283012
2016-10-01 05:57:50 +00:00
Mehdi Amini 217b246484 Revert "Use StringRef in Datalayout API (NFC)"
This reverts commit r283009. Bots are broken.

llvm-svn: 283011
2016-10-01 05:12:48 +00:00
Mehdi Amini 29baf9c0e1 Use StringRef in Datalayout API (NFC)
llvm-svn: 283009
2016-10-01 04:17:59 +00:00
Mehdi Amini f20abe53f4 Use StringRef in Pass Info/Support API (NFC)
llvm-svn: 283008
2016-10-01 04:03:30 +00:00
Mehdi Amini e11b745b66 Use StringRef in CommandLine Options handling (NFC)
llvm-svn: 283007
2016-10-01 03:43:20 +00:00
Mehdi Amini 9a72cd7b21 Use StringRef in TLI instead of raw pointer (NFC)
llvm-svn: 283005
2016-10-01 03:10:48 +00:00
Mehdi Amini 117296c0a0 Use StringRef in Pass/PassManager APIs (NFC)
llvm-svn: 283004
2016-10-01 02:56:57 +00:00
Mehdi Amini 86eeda8e20 Revert "AMDGPU: Don't use offen if it is 0"
This reverts commit r282999.
Tests are not passing: http://lab.llvm.org:8011/builders/clang-x86_64-linux-selfhost-modules/builds/20038

llvm-svn: 283003
2016-10-01 02:35:24 +00:00
Eric Christopher e8d141c675 Remove getTargetTriple and update all uses to use the Triple off
of the TargetMachine. NFC.

llvm-svn: 283002
2016-10-01 01:50:33 +00:00
Eric Christopher 364dbe06d3 Stop calling getTargetTriple off of the AsmPrinter and constructing a
TargetTriple, just grab it off of the TargetMachine. NFC.

llvm-svn: 283001
2016-10-01 01:50:29 +00:00
Eric Christopher 98983d0aff Remove TargetTriple from AArch64MCInstLower as it's used in few places
and can be pulled from the TargetMachine. NFC.

llvm-svn: 283000
2016-10-01 01:50:25 +00:00
Matt Arsenault 3070fdf798 AMDGPU: Don't use offen if it is 0
This removes many re-initializations of a base register to 0.

llvm-svn: 282999
2016-10-01 01:37:15 +00:00
Mehdi Amini 4cc259a469 Use StringRef in LTOCodegenerator (NFC)
llvm-svn: 282998
2016-10-01 01:18:23 +00:00
Mehdi Amini 05cfdd0800 Use StringRef in LTOModule implementation (NFC)
llvm-svn: 282997
2016-10-01 01:18:16 +00:00
Mehdi Amini b7fb124512 Use StringRef in Triple API (NFC)
llvm-svn: 282996
2016-10-01 01:16:22 +00:00
Kostya Serebryany d216922a80 [libFuzzer] implement the -shrink=1 option that tires to make elements of the corpus smaller, off by default
llvm-svn: 282995
2016-10-01 01:04:29 +00:00
Nirav Dave 9f2bd4e7ea [MC] Prevent out of order HashDirective lexing in AsmLexer.
To lex hash directives we peek ahead to find component tokens, create a
unified token, and unlex the peeked tokens so the parser does not need
to parse the tokens then. Make sure we do not to lex another hash
directive during peek operation.

This fixes PR28921.

Reviewers: rnk, loladiro

Subscribers: llvm-commits

Differential Revision: https://reviews.llvm.org/D24839

llvm-svn: 282992
2016-10-01 00:42:32 +00:00
Mehdi Amini 6610b01a27 [ASAN] Add the binder globals on Darwin to llvm.compiler.used to avoid LTO dead-stripping
The binder is in a specific section that "reverse" the edges in a
regular dead-stripping: the binder is live as long as a global it
references is live.

This is a big hammer that prevents LLVM from dead-stripping these,
while still allowing linker dead-stripping (with special knowledge
of the section).

Differential Revision: https://reviews.llvm.org/D24673

llvm-svn: 282988
2016-10-01 00:05:34 +00:00
Kostya Serebryany 90f8f36bca [libFuzzer] remove some experimental code
llvm-svn: 282983
2016-09-30 23:29:27 +00:00
Matthias Braun 298e007e99 ScheduleDAGInstrs: Cleanup, use range based for; NFC
llvm-svn: 282979
2016-09-30 23:08:07 +00:00
Kostya Serebryany 7022b94687 [libFuzzer] fix openssl fuzzer tests when running on a machine w/o openssl installed
llvm-svn: 282972
2016-09-30 22:35:08 +00:00
Kostya Serebryany e7e790bad6 [libFuzzer] remove unused option
llvm-svn: 282971
2016-09-30 22:29:57 +00:00
Reid Kleckner 9cb915b7be [SEH] Emit the parent frame offset label even if there are no funclets
This avoids errors about references to undefined local labels from
unreferenced filter functions.

Fixes (sort of) PR30431

llvm-svn: 282967
2016-09-30 22:10:12 +00:00
Quentin Colombet a6119958ff [AArch64][RegisterBankInfo] Use the helper functions for the checks
This makes sure the helper functions work as expected.

NFC.

llvm-svn: 282961
2016-09-30 21:46:21 +00:00
Quentin Colombet 7c3fa8e361 [AArch64][RegisterBankInfo] Rename getValueMappingIdx to getValueMapping
We don't return index, we return the actual ValueMapping.

NFC.

llvm-svn: 282960
2016-09-30 21:46:19 +00:00
Quentin Colombet b4afac7b32 [AArch64][RegisterBankInfo] Compress the ValueMapping table a bit.
We don't need to have singleton ValueMapping on their own, we can just
reuse one of the elements of the 3-ops mapping.
This allows even more code sharing.

NFC.

llvm-svn: 282959
2016-09-30 21:46:17 +00:00
Quentin Colombet 7fc5fe41c5 [AArch64][RegisterBankInfo] Refactor the code to access AArch64::ValMapping
Use a helper function to access ValMapping. This should make the code
easier to understand and maintain.

NFC.

llvm-svn: 282958
2016-09-30 21:46:15 +00:00
Quentin Colombet 15dc25bb3d [AArch64][RegisterBankInfo] Rename getRegBankIdx to getRegBankIdxOffset
The function name did not make it clear that the returned value was an
offset to apply to a register bank index.

NFC.

llvm-svn: 282957
2016-09-30 21:46:12 +00:00
Quentin Colombet b2308987ab [AArch64][RegisterBankInfo] Use the static opds mapping for alt mappings
Avoid to rely on the dynamically allocated operands mapping for the
alternative mapping.
NFC.

llvm-svn: 282956
2016-09-30 21:45:56 +00:00