Commits · ee2d77f6d693b88a335f34d559d7f82fb6998859 · Roger Ferrer / llvm-epi · GitLab

Feb 23, 2017
- Disable TLS for stack protector on Android API<17. · ee2d77f6
  Evgeniy Stepanov authored Feb 23, 2017
  ee2d77f6
- [GlobalISel] Emit opt remarks on isel fallbacks. · ae9dadec
  Ahmed Bougacha authored Feb 23, 2017
  ae9dadec
- Correct register pressure calculation in presence of subregs · ce3ddd2d
  Stanislav Mekhanoshin authored Feb 23, 2017
  ce3ddd2d
- [Hexagon] Avoid IMPLICIT_DEFs as new-value producers · 2cfc7a48
  Krzysztof Parzyszek authored Feb 23, 2017
  2cfc7a48
- AMDGPU/SI: Fix trunc i16 pattern · 70293a04
  Jan Vesely authored Feb 23, 2017
  70293a04
- [Hexagon] Patterns for CTPOP, BSWAP and BITREVERSE · af5ff65d
  Krzysztof Parzyszek authored Feb 23, 2017
  af5ff65d
- [ARM] GlobalISel: Lower call returns · a8cb0cd8
  Diana Picus authored Feb 23, 2017
  a8cb0cd8
- [ARM] GlobalISel: Lower call parameters in regs · a606713c
  Diana Picus authored Feb 23, 2017
  a606713c
- [X86][AVX] Disable VCVTSS2SD & VCVTSD2SS memory folding and fix the register... · 4b2c968c
  Ayman Musa authored Feb 23, 2017
  4b2c968c
- Fix assertion failure in ARMConstantIslandPass. · 5ac6adbb
  Kristof Beyls authored Feb 23, 2017
  5ac6adbb
- [X86][AVX512] Change VCVTSS2SD and VCVTSD2SS node types to keep consistency... · 6e670cf4
  Ayman Musa authored Feb 23, 2017
  6e670cf4
- AMDGPU: Add another BFE pattern · a9e16e65
  Matt Arsenault authored Feb 23, 2017
  a9e16e65
- AMDGPU: Use clamp with f64 · 79a45db7
  Matt Arsenault authored Feb 22, 2017
  79a45db7
- AMDGPU: Fold FP clamp as modifier bit · d5c6515b
  Matt Arsenault authored Feb 22, 2017
  d5c6515b
- AMDGPU : Update TrapCode based on Trap Handler ABI. · f2cce02e
  Wei Ding authored Feb 22, 2017
  f2cce02e
- AMDGPU: Add replacement bfe intrinsics · f5262256
  Matt Arsenault authored Feb 22, 2017
  f5262256
Feb 22, 2017
- [AVR] Disable integrated assembler for a few tests · 19d95334
  Dylan McKay authored Feb 22, 2017
  19d95334
- [Hexagon] Implement @llvm.readcyclecounter() · ab57c2ba
  Krzysztof Parzyszek authored Feb 22, 2017
  ab57c2ba
- AMDGPU: Don't add emergency stack slot if all spills are SGPR->VGPR · 7b6c5d28
  Matt Arsenault authored Feb 22, 2017
  7b6c5d28
- [Hexagon] Add intrinsics for masked vector stores · 65971d97
  Krzysztof Parzyszek authored Feb 22, 2017
  65971d97
- AMDGPU: Don't look at chain users when adjusting writemask · 93e65ea7
  Matt Arsenault authored Feb 22, 2017
  93e65ea7
- AMDGPU: Always allocate emergency stack slot at offset 0 · 707780b4
  Matt Arsenault authored Feb 22, 2017
  707780b4
- AMDGPU: Change exp with compr bit printing · 61ec6a03
  Matt Arsenault authored Feb 22, 2017
  61ec6a03
- Revert "AMDGPU : Update TrapCode based on Trap Handler ABI." · 6ade56e0
  Wei Ding authored Feb 22, 2017
  6ade56e0
- AMDGPU : Update TrapCode based on Trap Handler ABI. · 4991d357
  Wei Ding authored Feb 22, 2017
  4991d357
- Bring back 2>&1 redirection for this test · e8a0f5ef
  Matthias Braun authored Feb 22, 2017
  e8a0f5ef
- [AArch64] Extend AArch64RedundantCopyElimination to do simple copy propagation. · 6bb79157
  Geoff Berry authored Feb 22, 2017
  6bb79157
- MIRTests: Remove unnecessary 2>&1 redirection · f1141285
  Matthias Braun authored Feb 22, 2017
  f1141285
- [WebAssembly] Configure codegen to legalize f16 values. · a63e8eb1
  Dan Gohman authored Feb 22, 2017
  a63e8eb1
- [DAGCombiner] revert r295336 · 8e48f416
  Bill Seurer authored Feb 22, 2017
  8e48f416
- [X86][GlobalISel] Initial implementation , select G_ADD gpr, gpr · f7359d89
  Igor Breger authored Feb 22, 2017
  f7359d89
- [X86] Regenerate CSE test with codegen instead of just the instruction count · 07056a06
  Simon Pilgrim authored Feb 22, 2017
  07056a06
- [ARM] Fix constant islands pass. · 56db97d4
  Roger Ferrer Ibanez authored Feb 22, 2017
  56db97d4
- [ARM] Classification Improvements to ARM Sched-Models. NFCI. · b6727228
  Javed Absar authored Feb 22, 2017
  b6727228
- [AVX-512] Allow legacy scalar min/max intrinsics to select EVEX instructions when available · 56d40229
  Craig Topper authored Feb 22, 2017
  56d40229
- AMDGPU: Add cvt.pkrtz intrinsic · 1f17c668
  Matt Arsenault authored Feb 22, 2017
  1f17c668
- AMDGPU: Remove some uses of llvm.SI.export in tests · 3ea06336
  Matt Arsenault authored Feb 22, 2017
  3ea06336
- AMDGPU: Remove llvm.AMDGPU.clamp intrinsic · 9417505f
  Matt Arsenault authored Feb 21, 2017
  9417505f
- AMDGPU: Redefine clamp node as clamp 0.0-1.0 · 2fdf2a1a
  Matt Arsenault authored Feb 21, 2017
  2fdf2a1a
Feb 21, 2017
- [NVPTX] Unify vectorization of load/stores of aggregate arguments and return values. · 29bbdc1c
  Artem Belevich authored Feb 21, 2017
  29bbdc1c