Commits · 1a5d623ded8e0845d42c9ea272425a11391bf412 · Roger Ferrer / llvm-epi-0.8 · GitLab

This project is archived. Its data is read-only.

May 18, 2019
- [NFC][AArch64] Autogenerate fcopysign.ll test · 1a5d623d
  Roman Lebedev authored May 18, 2019
  1a5d623d
- [NFC][AArch64] Autogenerate bitfield-insert.ll, selectcc-to-shiftand.ll tests · 13ac317e
  Roman Lebedev authored May 18, 2019
  13ac317e
- [NFC][AArch64] Add some ubfx tests with immediates · d1be3c44
  Roman Lebedev authored May 18, 2019
  d1be3c44
- UpdateTestChecks: fix AMDGPU handling · 98092f37
  Roman Lebedev authored May 18, 2019
  98092f37
- UpdateTestChecks: arm64-eabi handlind · 822b9c97
  Roman Lebedev authored May 18, 2019
  822b9c97
- AMDGPU/GlobalISel: Implement s64->s64 [SU]ITOFP · 2f29220d
  Matt Arsenault authored May 17, 2019
  2f29220d
- GlobalISel: Implement lower for S64->S32 [SU]ITOFP · 02b5ca8c
  Matt Arsenault authored May 17, 2019
  02b5ca8c
May 17, 2019
- GlobalISel: Define integer min/max instructions · f3cedf48
  Matt Arsenault authored May 17, 2019
  f3cedf48
- [X86][SSE] Fold movmsk(not(x)) -> not(movmsk) · 065431c8
  Simon Pilgrim authored May 17, 2019
  065431c8
- [X86][SSE] Match all-of bool scalar reductions into a bitcast/movmsk + cmp. · 2c2f8e74
  Simon Pilgrim authored May 17, 2019
  2c2f8e74
- [DAGCombiner] visitShiftByConstant(): drop bogus signbit check · 64c756b9
  Roman Lebedev authored May 17, 2019
  64c756b9
- [X86][AVX] isNOT - add extract_subvector(xor X, -1) -> extract_subvector(X) fold. · 62c7032c
  Simon Pilgrim authored May 17, 2019
  62c7032c
- AMDGPU/GlobalISel: Legalize G_FCEIL · a510b570
  Matt Arsenault authored May 17, 2019
  a510b570
- AMDGPU/GlobalISel: Legalize G_INTRINSIC_TRUNC · 6aebcd54
  Matt Arsenault authored May 17, 2019
  6aebcd54
- AMDGPU/GlobalISel: Legalize G_FRINT · 6aafc5e1
  Matt Arsenault authored May 17, 2019
  6aafc5e1
- AMDGPU/GlobalISel: Legalize G_FCOPYSIGN · 1448f568
  Matt Arsenault authored May 17, 2019
  1448f568
- AMDGPU/GlobalISel: RegBankSelect for llvm.amdgcn.s.buffer.load · 568f1938
  Matt Arsenault authored May 17, 2019
  568f1938
- AMDGPU/GlobalISel: Use subreg index instead of extra unmerge · a3b5a386
  Matt Arsenault authored May 17, 2019
  a3b5a386
- AMDGPU/GlobalISel: Use waterfall loop for buffer_load · b3dc7363
  Matt Arsenault authored May 17, 2019
  b3dc7363
- Re-land r360859: "[MergeICmps] Simplify the code." · 632dfdda
  Clement Courbet authored May 17, 2019
  632dfdda
- [AMDGPU] detect WaW hazards when moving/merging load/store instructions · c4bc61ba
  Rhys Perry authored May 17, 2019
  c4bc61ba
- [SystemZ] Bugfix in SystemZTargetLowering::combineIntDIVREM() · 9427961c
  Jonas Paulsson authored May 17, 2019
  9427961c
- Revert r360859: "Reland r360771 "[MergeICmps] Simplify the code."" · d764e7c6
  Nico Weber authored May 17, 2019
  d764e7c6
May 16, 2019
- [CodeGen] Fixed de-optimization of legalize subvector extract · e3cbdaf1
  Tim Renouf authored May 16, 2019
  e3cbdaf1
- [X86] Use 0x9 instead of 0x1 as the immediate in some masked floor pattern.... · f09b9d41
  Craig Topper authored May 16, 2019
  f09b9d41
- AMDGPU: Introduce TokenFactor for ABI register copies in call sequence · 99e6f4d1
  Matt Arsenault authored May 16, 2019
  99e6f4d1
- AMDGPU: Assume xnack is enabled by default · df24c92c
  Matt Arsenault authored May 16, 2019
  df24c92c
- [RISCV][NFC] Add nounwind attribute to functions missing it in test/CodeGen/RISCV · 3966b02c
  Alex Bradbury authored May 16, 2019
  3966b02c
- [AArch64] Handle ISD::LROUND and ISD::LLROUND · 2d28db6b
  Adhemerval Zanella authored May 16, 2019
  2d28db6b
- [CodeGen] Add lround/llround builtins · 73643b50
  Adhemerval Zanella authored May 16, 2019
  73643b50
- RegAllocFast: Improve hinting heuristic · 828b685e
  Matt Arsenault authored May 16, 2019
  828b685e
- [NFC] Fixup FileCheck option name in tests added in rL360881 · 62650cf4
  Roman Lebedev authored May 16, 2019
  62650cf4
- [NFC][CodeGen] Add some more tests for pulling binops through shifts · ec6608d5
  Roman Lebedev authored May 16, 2019
  ec6608d5
- AMDGPU/GlobalISel: Correct regbank for 1-bit and/or/xor · a8f88c38
  Matt Arsenault authored May 16, 2019
  a8f88c38
- Reland r360771 "[MergeICmps] Simplify the code." · c4fdd717
  Clement Courbet authored May 16, 2019
  c4fdd717
- GlobalISel: Add G_FCOPYSIGN · 55146d31
  Matt Arsenault authored May 16, 2019
  55146d31
May 15, 2019
- [X86] Delay creating index register negations during address matching until... · e43bdf14
  Craig Topper authored May 15, 2019
  e43bdf14
- [codeview] Fix SDNode representation of annotation labels · 48824903
  Reid Kleckner authored May 15, 2019
  48824903
- [AArch64] only indicate CFI on Windows if we emitted CFI · 814435fe
  Mandeep Singh Grang authored May 15, 2019
  814435fe
- RegAlloc: try to fail more gracefully when out of registers · 664ceeda
  Nicolai Haehnle authored May 15, 2019
  664ceeda