summaryrefslogtreecommitdiffstats
path: root/llvm/lib
Commit message (Expand)AuthorAgeFilesLines
* IfConversion: Fix branch predication bug.Kyle Butt2016-08-291-20/+61
* Use store operation to poison allocas for lifetime analysis.Vitaly Buka2016-08-291-59/+106
* [asan] Separate calculation of ShadowBytes from calculating ASanStackFrameLayoutVitaly Buka2016-08-292-33/+57
* [SimplifyCFG] Hoisting invalidates metadataDavid Majnemer2016-08-291-2/+8
* [AArch64] Adjust the scheduling model for Exynos M1.Evandro Menezes2016-08-291-4/+14
* [StatepointsForGC] Rematerialize in the presence of PHIsAnna Thomas2016-08-291-0/+35
* [LTO] Remove extraneous outputTeresa Johnson2016-08-291-1/+0
* [Constant] remove fdiv and frem from canTrap()Sanjay Patel2016-08-291-2/+0
* [Coroutines] Part 9: Add cleanup subfunction.Gor Nishanov2016-08-298-66/+141
* [TargetLowering] remove fdiv and frem from canOpTrap() (PR29114)Sanjay Patel2016-08-291-4/+0
* Do not use MRI::getMaxLaneMaskForVReg as a mask covering whole registerKrzysztof Parzyszek2016-08-292-7/+5
* AMDGPU/SI: Improve register allocation hints for sopk instructionsTom Stellard2016-08-291-0/+1
* Use the correct ctor/dtor section for dynamic-no-pic.Rafael Espindola2016-08-291-1/+1
* Move code only used by codegen out of MC. NFC.Rafael Espindola2016-08-292-45/+55
* Fix -Wunused-but-set-variable warning.Haojian Wu2016-08-291-4/+0
* AMDGPU/SI: Query AA, if available, in areMemAccessesTriviallyDisjoint()Tom Stellard2016-08-291-0/+11
* Fixed a bug in type legalizer for masked gather.Igor Breger2016-08-291-1/+9
* [AVX512] In some cases KORTEST instruction may be used instead of ZEXT + TEST...Igor Breger2016-08-292-5/+23
* [InstructionSelect] NumBlocks isn't defined in DEBUG build.Haojian Wu2016-08-291-1/+1
* [X86] Don't lower FABS/FNEG masking directly to a ConstantPool load. Just cre...Craig Topper2016-08-291-9/+4
* [AVX-512] Always use v8i64 when converting 512-bit FAND/FOR/FXOR/FANDN to int...Craig Topper2016-08-291-5/+3
* [AVX-512] Add support for selecting 512-bit VPABSB/VPABSW when BWI is available.Craig Topper2016-08-282-2/+19
* [AVX-512] Add patterns for selecting 128/256-bit EVEX VPABS instructions.Craig Topper2016-08-282-2/+37
* [InstCombine] use m_APInt to allow icmp (and X, Y), C folds for splat constan...Sanjay Patel2016-08-281-32/+34
* [X86][AVX512] Only combine EVEX targets shuffles to shuffles of the same numb...Simon Pilgrim2016-08-281-4/+12
* [PowerPC] Implement lowering for atomicrmw min/max/umin/umaxHal Finkel2016-08-284-5/+152
* [Loop Vectorizer] Fixed memory confilict checks.Elena Demikhovsky2016-08-281-3/+29
* [AVX-512] Promote AND/OR/XOR to v2i64/v4i64/v8i64 even when we have AVX512F/A...Craig Topper2016-08-282-18/+124
* [X86] Rename PABSB/D/W instructions to be consistent with SSE/AVX instruction...Craig Topper2016-08-282-40/+40
* AMDGPU/R600: Enable Load combineJan Vesely2016-08-271-0/+1
* [X86] Rename predicate function that detects if requires one of the REX.B, RE...Craig Topper2016-08-271-15/+16
* [X86] Keep looping over operands looking for byte registers even if we alread...Craig Topper2016-08-271-5/+4
* [X86] Include XMM/YMM/ZMM16-23 in X86II::isX86_64ExtendedReg. This feels more...Craig Topper2016-08-272-8/+4
* [X86] Don't allow DR8-DR15 to be assembled in 32-bit mode. Add missing test f...Craig Topper2016-08-271-0/+2
* [X86] Remove stale comment about FixupBWInsts pass being off by default. NFCCraig Topper2016-08-271-2/+0
* [AVX-512] Allow EVEX encoding unordered/ordered/equal/notequal VCMPPS/PD/SS/S...Craig Topper2016-08-272-8/+28
* [X86] Enable FR32/FR64 cmpeq/cmpne/cmpunord/cmpord to be commuted.Craig Topper2016-08-272-0/+9
* [AVX-512] Add load folding for EVEX vcmpps/pd/ss/sd.Craig Topper2016-08-271-0/+8
* [LTO] Don't create a new common unless merged has different sizeTeresa Johnson2016-08-271-3/+4
* AMDGPU: Mark sched model completeMatt Arsenault2016-08-271-1/+1
* AMDGPU: Remove unneeded implicit exec uses/defsMatt Arsenault2016-08-272-40/+48
* GVN-hoist: invalidate MD cache (PR29144)Sebastian Pop2016-08-271-0/+2
* [RegBankSelect] Do not abort when the target wants to fall back.Quentin Colombet2016-08-271-17/+48
* [InstructionSelect] Do not abort when the target wants to fall back.Quentin Colombet2016-08-271-7/+28
* [MachineLegalize] Do not abort when the target wants to fall back.Quentin Colombet2016-08-272-6/+26
* AMDGPU: Select mulhi 24-bit instructionsMatt Arsenault2016-08-277-20/+163
* AMDGPU: Move cndmask pseudo to be isel pseudoMatt Arsenault2016-08-273-23/+31
* AMDGPU: Fix sched type for branchesMatt Arsenault2016-08-271-1/+1
* AMDGPU: Remove register operand from si_mask_branchMatt Arsenault2016-08-272-5/+3
* AMDGPU: Improve error reporting for maximum branch distanceMatt Arsenault2016-08-271-30/+61
OpenPOWER on IntegriCloud