Commits · e508d643cfd583b92dfd5484e27fb280cf0f3c60 · Roger Ferrer / llvm-epi · GitLab

May 24, 2020
- [X86][AVX] Fold extract_subvector(broadcast(x),c) -> extract_subvector(broadcast(x),0) iff c != 0 · e508d643
  Simon Pilgrim authored May 24, 2020
  e508d643
- [Pass Manager] remove EarlyCSE as clean-up for VectorCombine · 57bb4787
  Sanjay Patel authored May 24, 2020
  57bb4787
- [LV] Either get invariant condition OR vector condition. · 0deab8a5
  Florian Hahn authored May 24, 2020
  0deab8a5
- [PhaseOrdering] adjust test to use default alias analysis with new pass manager; NFC · d43fac05
  Sanjay Patel authored May 24, 2020
  d43fac05
- [X86] SimplifyMultipleUseDemandedBitsForTargetNode - add initial X86ISD::VSRAI handling. · 1e7865d9
  Simon Pilgrim authored May 24, 2020
  1e7865d9
- AMDGPU.h - reduce TargetMachine.h include. NFC. · 71bed820
  Simon Pilgrim authored May 24, 2020
  71bed820
- [PowerPC] Add some InstAlias definitions · 86e3abc9
  Kang Zhang authored May 24, 2020
  86e3abc9
- [InstCombine] fold FP trunc into exact itofp · c048a02b
  Sanjay Patel authored May 24, 2020
  c048a02b
- [PatternMatch] abbreviate vector inst matchers; NFC · 7eed772a
  Sanjay Patel authored May 23, 2020
  7eed772a
- AMDGPUInstPrinter.cpp - add CommandLine.h include. NFC. · b05b69e0
  Simon Pilgrim authored May 24, 2020
  b05b69e0
- [VPlan] Use VPUser for VPWidenSelectRecipe operands (NFC). · 15224408
  Florian Hahn authored May 24, 2020
  15224408
- AMDGPUTargetObjectFile.h - remove unnecessary includes. NFC. · 725b3463
  Simon Pilgrim authored May 24, 2020
  725b3463
- AMDGPULibFunc - fix include order. NFC. · a6502560
  Simon Pilgrim authored May 24, 2020
  a6502560
- LoopSimplify.h - reduce unnecessary includes to forward declarations. NFC. · 510b0f42
  Simon Pilgrim authored May 24, 2020
  
  510b0f42
- X86Subtarget.h - remove unnecessary TargetMachine.h include. NFC. · d0f2a8a0
  Simon Pilgrim authored May 24, 2020
  d0f2a8a0
- [X86] Pull out repeated DemandedBits signmask variable. NFC. · 478f2ce5
  Simon Pilgrim authored May 24, 2020
  478f2ce5
- [TargetLowering] Improve expandFunnelShift shift amount masking · 16031067
  Simon Pilgrim authored May 24, 2020
  16031067
- [X86] Move CONCAT_VECTORS/INSERT_SUBVECTOR actions inside loop. NFC. · ffb36721
  Simon Pilgrim authored May 24, 2020
  ffb36721
- X86TargetMachine.h - remove unnecessary X86Subtarget forward declaration. NFC. · 04d32d7a
  Simon Pilgrim authored May 24, 2020
  04d32d7a
- [X86][AVX] Call SimplifyDemandedBits on MaskedLoadSDNode with non-boolean masks · 8310c9b7
  Simon Pilgrim authored May 24, 2020
  8310c9b7
- [X86] Add family/model for Intel Comet Lake CPUs for -march=native and function multiversioning · 2bb822bc
  Craig Topper authored May 19, 2020
  2bb822bc
- [X86] Fix typo in comment. NFC · 79401230
  Craig Topper authored May 23, 2020
  
  79401230
- [X86] Improve i8 + 'slow' i16 funnel shift codegen · cc65a7a5
  Simon Pilgrim authored May 24, 2020
  cc65a7a5
- [AArch64][GlobalISel] When generating SUBS for compares, don't write to wzr/xzr. · 99660217
  Amara Emerson authored May 22, 2020
  99660217
May 23, 2020
- [TLI][PowerPC] Introduce TLI query to check if MULH is cheaper than MUL + SHIFT · b631f86a
  Amy Kwan authored May 23, 2020
  b631f86a
- [CFIInstrInserter] Delete unneeded checks · de172ef6
  Fangrui Song authored May 23, 2020
  
  de172ef6
- [ValueTracking] Use assumptions in computeConstantRange. · 8d041811
  Florian Hahn authored May 22, 2020
  8d041811
- [DwarfEHPrepare] Don't prune unreachable resumes at optnone · 2833c46f
  Nikita Popov authored May 21, 2020
  2833c46f
- TargetLowering.h - remove unnecessary TargetMachine.h include. NFC · fe0006c8
  Simon Pilgrim authored May 23, 2020
  fe0006c8
- SimplifyCFG: Clean up optforfuzzing implementation · cdd006ee
  Matt Arsenault authored May 20, 2020
  cdd006ee
- AMDGPU: Refine rcp/rsq intrinsic folding for modern FP rules · 27fe841a
  Matt Arsenault authored May 21, 2020
  27fe841a
- AMDGPU: Implement isConstantPhysReg · 76e3dd0a
  Matt Arsenault authored May 20, 2020
  76e3dd0a
- AMDGPU: Define mode register · 2e82667f
  Matt Arsenault authored Jan 03, 2020
  2e82667f
- Silence warning from unit test · 286ca0f7
  Matt Arsenault authored May 22, 2020
  286ca0f7
- TableGen: Don't reconstruct CodeGenDAGTarget · 421a40b3
  Matt Arsenault authored May 23, 2020
  421a40b3
- [yaml2obj] - Move "repeated section/fill name" check earlier. · 304b0ed4
  Georgii Rymar authored May 14, 2020
  304b0ed4
- [yaml2obj] - Add a technical prefix for each unnamed chunk. · 38c5d6f7
  Georgii Rymar authored May 14, 2020
  38c5d6f7
- Revert "Added a new IRCanonicalizer pass." · 335de55f
  Michal Paszkowski authored May 23, 2020
  335de55f
- Revert "[gn build] Port 14d35853 " · fc12ead8
  Michal Paszkowski authored May 23, 2020
  fc12ead8
- [gn build] Port 14d35853 · a0c7108b
  LLVM GN Syncbot authored May 23, 2020
  
  a0c7108b