Commits · d35d8da7d6ac6c08578ec0569b072292631691e0 · Roger Ferrer / llvm-epi · GitLab

Mar 29, 2021
- [AMDGPU] Mark additional VOP3 as commutable · d35d8da7
  Joe Nash authored Mar 23, 2021
  d35d8da7
- [PrologEpilogInserter][AMDGPU] Only adjust offset for emergency spill slots if the stack grows down · 489ca73a
  Roger Ferrer authored Mar 29, 2021
  489ca73a
- [AMDGPU] Extend gfx10 test coverage. NFC. · b082e6f8
  Petar Avramovic authored Mar 29, 2021
  b082e6f8
Mar 26, 2021
- [AMDGPU] Use reductions instead of scans in the atomic optimizer · 9d08f276
  Jay Foad authored Mar 19, 2021
  9d08f276
Mar 24, 2021
- Revert "[Passes] Add relative lookup table converter pass" · 5fbe1fdf
  Gulfem Savrun Yeniceri authored Mar 24, 2021
  5fbe1fdf
- [Passes] Add relative lookup table converter pass · 5fd001a5
  Gulfem Savrun Yeniceri authored Dec 29, 2020
  5fd001a5
- AMDGPU: Add target id and code object v4 support · f4ace637
  Konstantin Zhuravlyov authored Mar 24, 2021
  f4ace637
- [AMDGPU] SIOptimizeExecMaskingPreRA should check constant bus constraint when folds EXEC copy · dccf83ac
  alex-t authored Mar 18, 2021
  dccf83ac
Mar 23, 2021
- GlobalISel: Lower funnel shifts · b24436ac
  Matt Arsenault authored Mar 19, 2020
  
  b24436ac
- [AMDGPU] Use non-compressed exports in a test. NFC. · d42f63be
  Jay Foad authored Mar 23, 2021
  d42f63be
- [GlobalISel][AMDGPU] Lower G_UMULO/G_SMULO · d0e5422e
  Pushpinder Singh authored Mar 09, 2021
  d0e5422e
- [AMDGPU] Only unbundle memory accesses in SIMemoryLegalizer · 64db6b8d
  Carl Ritson authored Mar 23, 2021
  64db6b8d
- Revert "[Passes] Add relative lookup table converter pass" · e3a6d70c
  Gulfem Savrun Yeniceri authored Mar 23, 2021
  e3a6d70c
Mar 22, 2021
- [Passes] Add relative lookup table converter pass · 78a65cd9
  Gulfem Savrun Yeniceri authored Dec 29, 2020
  78a65cd9
- GlobalISel: Handle G_BUILD_VECTOR in isKnownToBeAPowerOfTwo · c34819af
  Matt Arsenault authored Jan 06, 2021
  
  c34819af
- AMDGPU: Allow tail calls for amdgpu_gfx functions · 1dd23c6d
  Matt Arsenault authored Mar 14, 2021
  
  1dd23c6d
Mar 21, 2021
- AMDGPU/GlobalISel: Enable CSE in pre-legalizer combiner · 6314a727
  Matt Arsenault authored Mar 20, 2021
  
  6314a727
Mar 20, 2021
- [AMDGPU] Rename SIInsertSkips Pass · fe5f4c39
  Carl Ritson authored Mar 20, 2021
  fe5f4c39
- [AMDGPU] Merge SIRemoveShortExecBranches into SIPreEmitPeephole · 5df2af8b
  Carl Ritson authored Mar 20, 2021
  5df2af8b
- [AMDGPU] Allow index optimisation in SIPreEmitPeephole for bundles · b76c0902
  Carl Ritson authored Mar 20, 2021
  b76c0902
Mar 19, 2021
- [AMDGPU] Rationalize some check prefixes and use more common prefixes. NFC. · 87248e85
  Jay Foad authored Mar 19, 2021
  
  87248e85
- [AMDGPU] Remove weird target triples from tests. NFC. · 5df52f77
  Jay Foad authored Mar 19, 2021
  
  5df52f77
- [DAG] computeKnownBits - add ISD::MULHS/MULHU/SMUL_LOHI/UMUL_LOHI handling · 9d2df964
  Simon Pilgrim authored Mar 19, 2021
  9d2df964
- [AMDGPU] Add atomic optimizer nouse tests · b8616e40
  Jay Foad authored Mar 18, 2021
  b8616e40
- [AMDGPU] Remove duplicate test functions. NFC. · 685335a0
  Jay Foad authored Mar 18, 2021
  
  685335a0
Mar 18, 2021
- [AMDGPU] Remove cpol, tfe, and swz from MUBUF patterns · edd6da10
  Stanislav Mekhanoshin authored Mar 15, 2021
  edd6da10
- [amdgpu] Update med3 combine to skip i64 · 253f804d
  Jon Chesterfield authored Mar 18, 2021
  253f804d
- [AMDGPU] Add some gfx1010 test coverage. NFC. · 078b338b
  Jay Foad authored Mar 18, 2021
  
  078b338b
- GlobalISel: Preserve source value information for outgoing byval args · b9a03849
  Matt Arsenault authored Mar 14, 2021
  b9a03849
- GlobalISel: Insert memcpy for outgoing byval arguments · 61f834cc
  Matt Arsenault authored Mar 12, 2021
  61f834cc
- [AMDGPU] Regenerate atomic_optimizations_global_pointer.ll tests · 388fbefb
  Simon Pilgrim authored Mar 18, 2021
  
  388fbefb
Mar 17, 2021
- [DAG] TargetLowering::isBinOp() - add ISD::SSUBSAT/USUBSAT · cfc256ba
  Simon Pilgrim authored Mar 17, 2021
  cfc256ba
- Revert rG3b635253ddd0106c88051cff3540d8eb90bee22f "[AMDGPU] Regenerate wave32.ll test checks" · 4a687405
  Simon Pilgrim authored Mar 17, 2021
  4a687405
- [AMDGPU] Regenerate wave32.ll test checks · 3b635253
  Simon Pilgrim authored Mar 17, 2021
  3b635253
Mar 15, 2021
- [AMDGPU] Fix copyPhysReg to not produce unalined vgpr access · bc27a318
  Stanislav Mekhanoshin authored Mar 12, 2021
  bc27a318
- [AMDGPU] Use single cache policy operand · 3bffb1cd
  Stanislav Mekhanoshin authored Feb 08, 2021
  3bffb1cd
- [amdgpu] Implement lower function LDS pass · 13e49dce
  Jon Chesterfield authored Mar 15, 2021
  13e49dce
- [AMDGPU] Fix shortfalls in WQM marking · 13877db2
  Carl Ritson authored Mar 15, 2021
  13877db2
Mar 13, 2021
- Reland [SCEV] Improve modelling for (null) pointer constants · 78b8ce40
  Roman Lebedev authored Mar 13, 2021
  78b8ce40
- Temporairly evert "[SCEV] Improve modelling for (null) pointer constants" · 329aeb5d
  Roman Lebedev authored Mar 13, 2021
  329aeb5d