Commits · f4572249d78d13b16cb287500a3bcd226ececb31 · Roger Ferrer / llvm-epi · GitLab

Jul 10, 2019
- Move three folds for FADD, FSUB and FMUL in the DAG combiner away from Unsafe... · f4572249
  Michael Berg authored Jul 10, 2019
  f4572249
- [X86] Regenerate tests. NFCI. · 5a6d40be
  Simon Pilgrim authored Jul 10, 2019
  5a6d40be
- [X86] Add tests for an alternative sequence for _mm_storel_pi/_mm_storeh_pi intrinsics. NFC · ab5a30ac
  Craig Topper authored Jul 10, 2019
  ab5a30ac
- [X86][SSE] EltsFromConsecutiveLoads - add basic dereferenceable support · 6a585839
  Simon Pilgrim authored Jul 10, 2019
  6a585839
Jul 09, 2019
- [X86][AMDGPU][DAGCombiner] Move call to allowsMemoryAccess into... · 84a1f073
  Craig Topper authored Jul 09, 2019
  84a1f073
- [LegalizeTypes] Fix saturation bug for smul.fix.sat · 59029017
  Bjorn Pettersson authored Jul 09, 2019
  59029017
- Fixing @llvm.memcpy not honoring volatile. · 336f3e16
  Guillaume Chatelet authored Jul 09, 2019
  336f3e16
Jul 08, 2019
- Standardize on MSVC behavior for triples with no environment · 2f07c2e9
  Reid Kleckner authored Jul 08, 2019
  2f07c2e9
- Add, and infer, a nofree function attribute · b4b21d80
  Brian Homerding authored Jul 08, 2019
  b4b21d80
- [X86] Allow execution domain fixing to turn SHUFPD into SHUFPS. · 1deca50a
  Craig Topper authored Jul 08, 2019
  1deca50a
- [X86] Make movsd commutable to shufpd with a 0x02 immediate on pre-SSE4.1 targets. · d8261f02
  Craig Topper authored Jul 08, 2019
  d8261f02
- [X86] Add MOVSDrr->MOVLPDrm entry to load folding table. Add custom handling... · 46f2b583
  Craig Topper authored Jul 08, 2019
  46f2b583
Jul 06, 2019
- [X86] Add PS<->PD domain changing support for MOVH/MOVL load instructions and... · e753247b
  Craig Topper authored Jul 06, 2019
  e753247b
- [X86] Remove patterns from MOVLPSmr and MOVHPSmr instructions. · 317d6093
  Craig Topper authored Jul 06, 2019
  317d6093
- [X86] Add patterns to select MOVLPDrm from MOVSD+load and MOVHPD from UNPCKL+load. · 913105ca
  Craig Topper authored Jul 06, 2019
  913105ca
- [X86] Copy some test cases from vector-shuffle-sse1.ll to... · 8c036bf7
  Craig Topper authored Jul 06, 2019
  8c036bf7
- RegUsageInfoCollector: Skip AMDGPU entry point functions · 705e46f4
  Matt Arsenault authored Jul 05, 2019
  705e46f4
Jul 05, 2019
- This reverts r365061 and r365062 (test update) · 9dcfbbae
  Robert Lougher authored Jul 05, 2019
  9dcfbbae
- Revert r365198 as this accidentally commited something that · 2478b620
  Robert Lougher authored Jul 05, 2019
  2478b620
- This reverts r365061 and r365062 (test update) · 3bea2b15
  Robert Lougher authored Jul 05, 2019
  3bea2b15
- [X86][SSE] LowerINSERT_VECTOR_ELT - early out for out of range indices · 8b25d9bf
  Simon Pilgrim authored Jul 05, 2019
  8b25d9bf
- [X86] Add custom isel to select ADD/SUB/OR/XOR/AND to their non-immediate... · 171732ae
  Craig Topper authored Jul 04, 2019
  171732ae
Jul 04, 2019
- [DAGCombiner] Don't combine (addcarry (uaddo X, Y), 0, Carry) -> (addcarry X,... · e9aed963
  Craig Topper authored Jul 04, 2019
  e9aed963
- [X86][SSE] Add partial dereferenceable vector load test inspired by PR21780 · e602f70d
  Simon Pilgrim authored Jul 04, 2019
  e602f70d
- [X86][SSE] Add some partial dereferenceable vector load tests inspired by PR16739 · 146f1f2e
  Simon Pilgrim authored Jul 04, 2019
  146f1f2e
- [X86] Regenerate load fold peephole test. · 8351c327
  Simon Pilgrim authored Jul 04, 2019
  8351c327
- [X86][AVX1] Combine concat_vectors(pshufd(x,c),pshufd(y,c)) -> vpermilps(concat_vectors(x,y),c) · fde766de
  Simon Pilgrim authored Jul 04, 2019
  fde766de
Jul 03, 2019
- [X86] Update test; NFC · 11953acb
  Robert Lougher authored Jul 03, 2019
  11953acb
- [X86] Avoid SFB - Skip meta instructions · 720baf04
  Robert Lougher authored Jul 03, 2019
  720baf04
- [CodeGen] Make branch funnels pass the machine verifier · 83bbe2f4
  Francis Visoiu Mistrih authored Jul 03, 2019
  83bbe2f4
- [X86] ComputeNumSignBitsForTargetNode - add target shuffle support. · 26812c76
  Simon Pilgrim authored Jul 03, 2019
  26812c76
- [DAGCombine] More diamong carry pattern optimization. · bddb8c35
  Amaury Sechet authored Jul 03, 2019
  bddb8c35
- [X86][AVX] combineX86ShufflesRecursively - peek through extract_subvector · 783dbe40
  Simon Pilgrim authored Jul 03, 2019
  783dbe40
- [X86][AVX] Combine vpermi(bitcast(x)) -> bitcast(vpermi(x)) · 868d0b7f
  Simon Pilgrim authored Jul 03, 2019
  868d0b7f
- [X86][AVX] combineX86ShuffleChainWithExtract - add number of non-zero... · 0c230209
  Simon Pilgrim authored Jul 03, 2019
  0c230209
- [Codegen][X86][AArch64][ARM][PowerPC] Inc-of-add vs sub-of-not (PR42457) · c4b83a60
  Roman Lebedev authored Jul 03, 2019
  c4b83a60
- [triple] Use 'macabi' environment name for the Mac Catalyst triples · 3dbdbbec
  Alex Lorenz authored Jul 03, 2019
  3dbdbbec
- Add support for the 'macCatalyst' MachO platform · da1dfecd
  Alex Lorenz authored Jul 02, 2019
  da1dfecd
Jul 02, 2019
- [X86] Copy test cases from vector-zext.ll to vector-zext-widen.ll. Same for vector-sext.ll. NFC · fa4e825a
  Craig Topper authored Jul 02, 2019
  fa4e825a
- [X86] getTargetConstantBitsFromNode - remove unnecessary getZExtValue() (PR42486) · 56138749
  Simon Pilgrim authored Jul 02, 2019
  56138749