Commits · d42f76fd368ac885289836c0bce1d7ca6ad7348a · Roger Ferrer / llvm-epi · GitLab

Sep 07, 2021
- [Analysis, Target, Transforms] Construct SmallVector with iterator ranges (NFC) · 5648f717
  Kazu Hirata authored Sep 07, 2021
  
  5648f717
- [MC] Use local MCSubtargetInfo in writeNops · e63455d5
  Peter Smith authored Aug 09, 2021
  e63455d5
- [amdgpu] Enable selection of `s_cselect_b64`. · 640beb38
  Michael Liao authored Aug 30, 2021
  640beb38
- [AMDGPU][GlobalISel] Legalize G_MUL for non-standard types · 6c4b634d
  Mirko Brkusanin authored Sep 07, 2021
  6c4b634d
- [AMDGPU][GlobalISel] Legalization of G_ROTL and G_ROTR · 5263bf58
  Mirko Brkusanin authored Sep 07, 2021
  5263bf58
- [AMDGPU][GlobalISel] Legalize memcpy family of intrinsics · 36527cbe
  Mirko Brkusanin authored Sep 07, 2021
  36527cbe
Sep 03, 2021
- [AMDGPU] Small cleanup in optimizeCompareInstr. NFC. · d0c06471
  Stanislav Mekhanoshin authored Sep 03, 2021
  
  d0c06471
- AMDGPU: Remove FeatureLocalMemorySize0 · 79bcd4a7
  Matt Arsenault authored Sep 02, 2021
  79bcd4a7
- [AMDGPU] Process any power of 2 in optimizeCompareInstr · 78fbd1aa
  Stanislav Mekhanoshin authored Sep 02, 2021
  78fbd1aa
- [AMDGPU] Fold immediates in the optimizeCompareInstr · 2cfda6a6
  Stanislav Mekhanoshin authored Sep 02, 2021
  2cfda6a6
Sep 02, 2021
- [AMDGPU] Use S_BITCMP0_* to replace AND in optimizeCompareInstr · 832c87b4
  Stanislav Mekhanoshin authored Sep 01, 2021
  832c87b4
- [AMDGPU] Add merging into S_BUFFER_LOAD_DWORDX8_IMM · 30d6c39b
  Piotr Sobczak authored Aug 30, 2021
  30d6c39b
- [AMDGPU] Use S_BITCMP1_* to replace AND in optimizeCompareInstr · f3645c79
  Stanislav Mekhanoshin authored Sep 01, 2021
  f3645c79
- [AMDGPU] Introduce optimizeCompareInstr · bf77b112
  Stanislav Mekhanoshin authored Aug 31, 2021
  bf77b112
Sep 01, 2021
- [AMDGPU] enable scalar compare in truncate selection · e3cbf1d4
  alex-t authored Sep 01, 2021
  e3cbf1d4
- [NFC] Remove redundant code added in 04ce2de3 · b9b419a1
  Arthur Eubanks authored Sep 01, 2021
  
  b9b419a1
- Revert "Revert "Disable ReplaceLDS pass, patch up tests to match"" · 97688bfd
  hsmahesha authored Sep 01, 2021
  97688bfd
- Revert "Disable ReplaceLDS pass, patch up tests to match" · 5ae6804d
  hsmahesha authored Sep 01, 2021
  5ae6804d
- [AMDGPU] Introduce RC flags for vector register classes · 4dab1528
  Christudasan Devadasan authored Aug 27, 2021
  4dab1528
- [AMDGPU] Split entry basic block after alloca instructions. · 98f47131
  hsmahesha authored Sep 01, 2021
  98f47131
Aug 31, 2021
- [AMDGPU] Enable ds_min/ds_max on more subtargets · c9683926
  Joe Nash authored Aug 31, 2021
  c9683926
Aug 30, 2021
- [llvm] Remove redundant calls to str() and c_str() (NFC) · c50faffb
  Kazu Hirata authored Aug 30, 2021
  c50faffb
Aug 27, 2021
- AMDGPU: Remove implicit argument attributes when introducing new calls · 04ce2de3
  Matt Arsenault authored Aug 13, 2021
  04ce2de3
- AMDGPU: Invert AMDGPUAttributor · 088cc636
  Matt Arsenault authored Aug 11, 2021
  088cc636
- AMDGPU: Restrict attributor transforms · 46d82e73
  Matt Arsenault authored Aug 13, 2021
  46d82e73
- AMDGPU: Remove hacky attribute deduction from AMDGPUAttributor · cf32d61a
  Matt Arsenault authored Aug 11, 2021
  cf32d61a
- AMDGPU: Stop inferring use of llvm.amdgcn.kernarg.segment.ptr · 98d7aa43
  Matt Arsenault authored Aug 13, 2021
  98d7aa43
- AMDGPU: Fix crashing on kernel declarations when lowering LDS · ce51c5d4
  Matt Arsenault authored Aug 26, 2021
  ce51c5d4
Aug 26, 2021
- [AMDGPU] Invert partial vgpr to agpr spill lane order · 827dd17e
  Stanislav Mekhanoshin authored Aug 25, 2021
  827dd17e
- [AMDGPU] Remove dead and broken ComplexPatterns · 8f89e2f6
  Jessica Clarke authored Aug 26, 2021
  8f89e2f6
Aug 25, 2021
- [AMDGPU] Avoid assert for saved FP · 11b7ee97
  Stanislav Mekhanoshin authored Aug 03, 2021
  11b7ee97
- [AMDGPU] Divergence-driven compare operations instruction selection · ed0f4415
  alex-t authored Jul 15, 2021
  ed0f4415
- [AMDGPU] Support global_atomic_fmin/max on gfx10 · e381833b
  Joe Nash authored Aug 24, 2021
  e381833b
- [NFC][AMDGPU] Reduce includes dependencies. · 48958d02
  Daniil Fukalov authored Aug 23, 2021
  48958d02
- [MachineCopyPropagation] Check CrossCopyRegClass for cross-class copys · 549f6a81
  Vang Thao authored Aug 12, 2021
  549f6a81
Aug 24, 2021
- [MCA] Adding an AMDGPUCustomBehaviour implementation. · e4ebfb57
  Patrick Holland authored Jul 22, 2021
  e4ebfb57
Aug 23, 2021
- [AMDGPU] Try to fix a GCC 11 warning · 7a967d90
  Jay Foad authored Aug 23, 2021
  7a967d90
Aug 20, 2021
- [AMDGPU] Fix too many constants with flat scratch · f3fe44fa
  Sebastian Neubauer authored Aug 19, 2021
  f3fe44fa
- [Remarks] [AMDGPU] Emit optimization remarks for atomics generating hardware instructions · 508b0669
  Anshil Gandhi authored Aug 19, 2021
  508b0669
Aug 19, 2021
- [NFC] Replace some attribute methods that use confusing indexes · 44a3241f
  Arthur Eubanks authored Aug 19, 2021
  
  44a3241f