Commits · d3e5cb77e4cbba5966479c62d970eb28414995a2 · Roger Ferrer / llvm-epi · GitLab

Feb 16, 2017
- AMDGPU: Remove llvm.SI.sendmsg · d3e5cb77
  Matt Arsenault authored Feb 16, 2017
  d3e5cb77
- AMDGPU: Remove SI_fs_constant and SI_fs_interp intrinsics · d2c8a337
  Matt Arsenault authored Feb 16, 2017
  d2c8a337
Feb 15, 2017
- AMDGPU: Consolidate sendmsg/sendmsghalt handling and tests · a78ca62c
  Matt Arsenault authored Feb 15, 2017
  a78ca62c
- Codegen: Make chains from trellis-shaped CFGs · 7fbec9bd
  Kyle Butt authored Feb 15, 2017
  7fbec9bd
- [AMDGPU] Revert failed scheduling · 582a5237
  Stanislav Mekhanoshin authored Feb 15, 2017
  582a5237
- [AMDGPU] Fix MaxWorkGroupsPerCU for large workgroups · 19f98c6a
  Stanislav Mekhanoshin authored Feb 15, 2017
  19f98c6a
Feb 14, 2017
- Revert "[AMDGPU] Fix for SIMachineScheduler crash. SI Scheduler should track" · 9f61feac
  Alexander Timofeev authored Feb 14, 2017
  9f61feac
Feb 10, 2017
- AMDGPU : Add trap handler support. · 205bfdb3
  Wei Ding authored Feb 10, 2017
  205bfdb3
- [AMDGPU] Override PSet for M0 · 6dec2431
  Stanislav Mekhanoshin authored Feb 10, 2017
  6dec2431
Feb 09, 2017
- AMDGPU: Add pass to expand memcpy/memmove/memset · 0699ef39
  Matt Arsenault authored Feb 09, 2017
  0699ef39
- [AMDGPU] Calculate number of min/max SGPRs/VGPRs for WavesPerEU instead of using switch statement · fd871377
  Konstantin Zhuravlyov authored Feb 09, 2017
  fd871377
Feb 08, 2017
- [AMDGPU] Add target information that is required by tools to metadata · 9f89ede1
  Konstantin Zhuravlyov authored Feb 08, 2017
  9f89ede1
- AMDGPU: Enable InferAddressSpaces · 417e0072
  Matt Arsenault authored Feb 08, 2017
  417e0072
Feb 07, 2017
- [AMDGPU] Fix for SIMachineScheduler crash. SI Scheduler should track · a3dace36
  Alexander Timofeev authored Feb 07, 2017
  a3dace36
- [AMDGPU] Lower null pointers in static variable initializer · 8f844f39
  Yaxun Liu authored Feb 07, 2017
  8f844f39
Feb 04, 2017
- [RegisterCoalescer] Do not call getInstructionIndex with DBG_VALUE · 9809b105
  Brendon Cahoon authored Feb 04, 2017
  9809b105
Feb 03, 2017
- AMDGPU: Cleanup scalar_to_vector test · 5c8510fd
  Matt Arsenault authored Feb 03, 2017
  5c8510fd
- AMDGPU: Set MCAsmInfo::PointerSize · 1fa5eacf
  Matt Arsenault authored Feb 03, 2017
  1fa5eacf
- AMDGPU: Fold fneg into fmin/fmax_legacy · e1b59530
  Matt Arsenault authored Feb 03, 2017
  e1b59530
- AMDGPU: Fold fneg into fminnum/fmaxnum · 2511c031
  Matt Arsenault authored Feb 03, 2017
  2511c031
- llvm-readobj: fix next note entry calculation and print unknown note types · 803b141f
  Konstantin Zhuravlyov authored Feb 02, 2017
  803b141f
- AMDGPU: Check if users of fneg can fold mods · a8fcfadf
  Matt Arsenault authored Feb 02, 2017
  a8fcfadf
Feb 02, 2017
- Revert "In visitSTORE, always use FindBetterChain, rather than only when UseAA is enabled." · 93f9d5ce
  Nirav Dave authored Feb 02, 2017
  93f9d5ce
- In visitSTORE, always use FindBetterChain, rather than only when UseAA is enabled. · 4442667f
  Nirav Dave authored Feb 02, 2017
  4442667f
- AMDGPU: Use source modifiers with f16->f32 conversions · 9dba9bd4
  Matt Arsenault authored Feb 02, 2017
  9dba9bd4
Feb 01, 2017
- [AMDGPU] Account workgroup size in LDS occupancy limits · 2b913b1f
  Stanislav Mekhanoshin authored Feb 01, 2017
  2b913b1f
- AMDGPU: Improve nsw/nuw/exact when promoting uniform i16 ops · d59e6404
  Matt Arsenault authored Feb 01, 2017
  d59e6404
- CodeGen: Allow small copyable blocks to "break" the CFG. · b15c0667
  Kyle Butt authored Jan 31, 2017
  b15c0667
Jan 31, 2017
- AMDGPU: Use source mods with fcanonicalize · d5d78510
  Matt Arsenault authored Jan 31, 2017
  d5d78510
- AMDGPU/SI: Fix inst-select-load-smrd.mir on some builds · 124f5cc8
  Tom Stellard authored Jan 31, 2017
  124f5cc8
- [DAGCombine] require UnsafeFPMath for re-association of addition · 8813d5d2
  Nicolai Haehnle authored Jan 31, 2017
  8813d5d2
- AMDGPU: Generalize matching of v_med3_f32 · f84e5d9a
  Matt Arsenault authored Jan 31, 2017
  f84e5d9a
Jan 30, 2017
- Re-commit AMDGPU/GlobalISel: Add support for simple shaders · ca16621b
  Tom Stellard authored Jan 30, 2017
  ca16621b
- [AMDGPU] Internalize non-kernel symbols · a3b72798
  Stanislav Mekhanoshin authored Jan 30, 2017
  a3b72798
- AMDGPU: Undo sub x, c -> add x, -c canonicalization · af635240
  Matt Arsenault authored Jan 30, 2017
  af635240
- AMDGPU: Make i32 uaddo/usubo legal · ee3f0acf
  Matt Arsenault authored Jan 30, 2017
  ee3f0acf
- DAG: Fold fneg into compare with constant into the constant · 32e6bfa2
  Matt Arsenault authored Jan 30, 2017
  32e6bfa2
- Revert "AMDGPU/GlobalISel: Add support for simple shaders" · 7a19d56f
  Tom Stellard authored Jan 30, 2017
  7a19d56f
- AMDGPU/GlobalISel: Add support for simple shaders · e48f60ae
  Tom Stellard authored Jan 30, 2017
  e48f60ae
- DAG: Constant fold fp16_to_fp/fp16_to_fp · 0c687390
  Matt Arsenault authored Jan 30, 2017
  0c687390