Commits · d6acd0196b3378bdeb5193053e290d7194c4f72d · Roger Ferrer / llvm-epi · GitLab

Jan 05, 2023
- clang/AMDGPU: Remove flat-address-space from feature map · 81849497
  Matt Arsenault authored Nov 28, 2022
  81849497
- [CodeGenOpenCL] Convert some tests to opaque pointers (NFC) · aae20a74
  Nikita Popov authored Jan 05, 2023
  
  aae20a74
Dec 30, 2022
- AMDGPU/clang: Remove target features from address space test builtins · e630d9b2
  Matt Arsenault authored Nov 28, 2022
  e630d9b2
Dec 29, 2022
- AMDGPU/clang: Add builtins for llvm.amdgcn.ballot · f4bcd7f5
  Matt Arsenault authored Jun 17, 2020
  f4bcd7f5
Dec 16, 2022
- Revert "[SROA] `isVectorPromotionViable()`: memory intrinsics operate on vectors of bytes (take 3)" · 96d3c826
  Roman Lebedev authored Dec 16, 2022
  96d3c826
- [SROA] `isVectorPromotionViable()`: memory intrinsics operate on vectors of bytes (take 3) · cfd594f8
  Roman Lebedev authored Dec 09, 2022
  cfd594f8
Dec 12, 2022
- [Clang] Convert various tests to opaque pointers (NFC) · 9466b491
  Nikita Popov authored Dec 12, 2022
  9466b491
Dec 07, 2022
- [AMDGPU] Annotate the intrinsics to be default and nocallback · f6e3a89c
  Johannes Doerfert authored Oct 04, 2022
  f6e3a89c
Dec 06, 2022
- [Clang][CodeGen] Use poison instead of undef for extra argument in __builtin_amdgcn_mov_dpp [NFC] · 481170cb
  Manuel Brito authored Dec 06, 2022
  481170cb
Nov 29, 2022
- Revert "enable code-object-version=5" · ca856fff
  Ron Lieberman authored Nov 29, 2022
  ca856fff
- enable code-object-version=5 · d882ba7a
  Ron Lieberman authored Nov 29, 2022
  
  d882ba7a
- [OpenCL] Remove arm-integer-dot-product extension pragmas · b0e4897a
  Sven van Haastregt authored Nov 29, 2022
  b0e4897a
Nov 28, 2022
- [AMDGPU] Intrinsic to expose s_wait_event for export ready · 7940888c
  David Stuttard authored Nov 17, 2022
  7940888c
Nov 26, 2022
- Revert "[SROA] `isVectorPromotionViable()`: memory intrinsics operate on vectors of bytes (take 2)" · 25f01d59
  Roman Lebedev authored Nov 27, 2022
  25f01d59
- [SROA] `isVectorPromotionViable()`: memory intrinsics operate on vectors of bytes (take 2) · 3c4d2a03
  Roman Lebedev authored Nov 26, 2022
  3c4d2a03
Nov 25, 2022
- Drop redundant pipe to opt -instnamer in clang tests · 54ad4d2d
  Alex Richardson authored Nov 24, 2022
  54ad4d2d
Nov 23, 2022
- Revert "[SROA] `isVectorPromotionViable()`: memory intrinsics operate on vectors of bytes" · 5cfc22ca
  Benjamin Kramer authored Nov 23, 2022
  5cfc22ca
- [SROA] `isVectorPromotionViable()`: memory intrinsics operate on vectors of bytes · cf624b23
  Roman Lebedev authored Nov 23, 2022
  cf624b23
Nov 13, 2022
- [clang][test] Avoid UB in overload.cl · e564f515
  Arthur Eubanks authored Nov 13, 2022
  
  e564f515
Nov 11, 2022
- The handling of 'funsafe-math-optimizations' doesn't update the 'MathErrno' · 91628f06
  Zahira Ammarguellat authored Nov 07, 2022
  91628f06
Nov 08, 2022
- [clang][test] Use opt -passes=<name> instead of opt -name · 5f9a8268
  Bjorn Pettersson authored Nov 06, 2022
  5f9a8268
Nov 04, 2022
- [IR] Switch everything to use memory attribute · 304f1d59
  Nikita Popov authored Nov 02, 2022
  304f1d59
Oct 14, 2022
- AMDGPU: Add __builtin_amdgcn_permlane64 · f59f116b
  Matt Arsenault authored Sep 28, 2022
  
  f59f116b
Sep 25, 2022
- [AMDGPU] Pattern for flat atomic fadd f64 intrinsic with local addr · dcc756d0
  Petar Avramovic authored Sep 25, 2022
  dcc756d0
Sep 23, 2022
- [clang][AMDGPU] Temporarily disable clang atomic fadd test for gfx90a · a3becb33
  Petar Avramovic authored Sep 23, 2022
  a3becb33
Sep 16, 2022
- [AMDGPU] Added __builtin_amdgcn_ds_bvh_stack_rtn · e5409659
  Stanislav Mekhanoshin authored Sep 15, 2022
  e5409659
Sep 02, 2022
- [test] Change cc1 -fvisibility to -fvisibility= · 74742147
  Fangrui Song authored Sep 02, 2022
  
  74742147
- Revert "[InstCombine] Treat passing undef to noundef params as UB" · 18de7c6a
  Muhammad Omair Javaid authored Sep 02, 2022
  18de7c6a
- [InstCombine] Treat passing undef to noundef params as UB · c911befa
  Arthur Eubanks authored Aug 31, 2022
  c911befa
Sep 01, 2022
- [OpenCL][SPIR-V] Test extern functions with a pointer arg. · 6b1a0452
  Anastasia Stulova authored Sep 01, 2022
  6b1a0452
Aug 23, 2022
- [AMDGPU] Add builtin s_sendmsg_rtn · 9f6cb3e9
  Yaxun (Sam) Liu authored Aug 18, 2022
  9f6cb3e9
Aug 20, 2022
- [AMDGPU] Add iglp_opt builtin and MFMA GEMM Opt strategy · b0f4678b
  Austin Kerbow authored Aug 17, 2022
  b0f4678b
Jul 28, 2022
- [AMDGPU] Add amdgcn_sched_group_barrier builtin · f5b21680
  Austin Kerbow authored Jun 13, 2022
  f5b21680
Jul 22, 2022
- Strengthen -Wint-conversion to default to an error · 7068aa98
  Aaron Ballman authored Jul 22, 2022
  7068aa98
Jul 18, 2022
- [AMDGPU] Support for gfx940 fp8 smfmac · 523a99c0
  Stanislav Mekhanoshin authored Jul 15, 2022
  523a99c0
- [AMDGPU] Support for gfx940 fp8 mfma · 2695f0a6
  Stanislav Mekhanoshin authored Jul 15, 2022
  2695f0a6
- [AMDGPU] Support for gfx940 fp8 conversions · 9fa5a6b7
  Stanislav Mekhanoshin authored Jul 15, 2022
  9fa5a6b7
Jul 01, 2022
- [AMDGPU] Add WMMA clang builtins · 4a782252
  Piotr Sobczak authored Jul 01, 2022
  4a782252
Jun 16, 2022
- [AMDGPU] gfx11 new dot instruction codegen support · 2d43de13
  Joe Nash authored Jun 15, 2022
  2d43de13
May 11, 2022
- [AMDGPU] Add llvm.amdgcn.sched.barrier intrinsic · 2db70021
  Austin Kerbow authored Mar 24, 2022
  2db70021