Commits · f96b4ab92d6a95a0720d18b1cd948a95d0cb54e7 · Roger Ferrer / llvm-epi · GitLab

May 14, 2017
- [X86][AVX2] Fix costs for v4i64 ashr by splat · f96b4ab9
  Simon Pilgrim authored May 14, 2017
  f96b4ab9
- [X86][AVX1] Account for cost of extract/insert of 256-bit shifts by splat · de4467b1
  Simon Pilgrim authored May 14, 2017
  de4467b1
- [X86] Add avx512vl command lines to the 128/256-bit vector-lzcnt tests so we... · c27dc9f7
  Craig Topper authored May 14, 2017
  c27dc9f7
- [X86] Cleanup some of the check-prefixes in the vector-lzcnt tests. · 87804dfe
  Craig Topper authored May 14, 2017
  87804dfe
- [X86] Remove unused value from IntrinsicType enum. NFC · ceea1a76
  Craig Topper authored May 14, 2017
  ceea1a76
- [X86][AVX1] Account for cost of extract/insert of 256-bit SDIV/UDIV by mul sequences · d3f0d03c
  Simon Pilgrim authored May 14, 2017
  d3f0d03c
- Revert r302978 and r302981. · c1c78617
  Akira Hatanaka authored May 14, 2017
  c1c78617
- Fix DynamicLibraryTest.cpp on FreeBSD and NetBSD · 4043373e
  Dimitry Andric authored May 14, 2017
  4043373e
- [COFF] Gracefully handle empty .drectve sections · ee97c5f0
  Shoaib Meenai authored May 14, 2017
  ee97c5f0
- [X86][XOP] XOP's general v16i8 shifts will be used instead of v8i16 shift + mask. · 5bef9c62
  Simon Pilgrim authored May 14, 2017
  5bef9c62
- [X86][SSE] Account for cost of extract/insert of v32i8 vector shifts · aa8dffb6
  Simon Pilgrim authored May 14, 2017
  aa8dffb6
- [StringExtras] Add llvm::to_integer. · 7d186b01
  Zachary Turner authored May 14, 2017
  7d186b01
- [X86][XOP] Account for cost of extract/insert of 256-bit vector shifts · 4599eaa0
  Simon Pilgrim authored May 14, 2017
  4599eaa0
- [X86][AVX] Allow 32-bit targets to peek through subvectors to extract constant... · f3ee9c69
  Simon Pilgrim authored May 14, 2017
  f3ee9c69
- [X86][AVX] Add additional 32-bit target vector shift tests · f3e87ac5
  Simon Pilgrim authored May 14, 2017
  f3e87ac5
- [Simplify] Remove some leftover dead code · 497fdd7d
  Tobias Grosser authored May 14, 2017
  497fdd7d
- [Polly] Fix code generation of llvm.expect intrinsic · b693f42b
  Tobias Grosser authored May 14, 2017
  b693f42b
- update of the url · 2f10a671
  Sylvestre Ledru authored May 14, 2017
  2f10a671
- [InstSimplify] Add patterns for folding (A & B) | (~A ^ B) -> (~A ^ B) and its commuted variants. · 479daaf7
  Craig Topper authored May 14, 2017
  479daaf7
- foo · 982cc3b1
  Craig Topper authored May 14, 2017
  982cc3b1
- [BasicAA] Alphabetize includes. NFC · dfc8955e
  Craig Topper authored May 14, 2017
  dfc8955e
- [clang-tidy] TwineLocalCheck: add param # checking · 7696b71d
  Yan Wang authored May 14, 2017
  7696b71d
- Renable test that was disabled due to cost analysis · 90a9ef6c
  Xinliang David Li authored May 14, 2017
  90a9ef6c
- Fix test failure on windows -- do not return deleted func · 392e9756
  Xinliang David Li authored May 14, 2017
  392e9756
- [llvm-pdbdump] Add the option to sort functions and data. · 0683be2e
  Zachary Turner authored May 14, 2017
  0683be2e
- [SelectionDAG] Added support for EXTRACT_SUBVECTOR/CONCAT_VECTORS demandedelts... · 754c1618
  Simon Pilgrim authored May 13, 2017
  754c1618
- Add missing files · d891d89c
  Peter Collingbourne authored May 13, 2017
  d891d89c
- Move lib/LibDriver -> lib/ToolDrivers/llvm-lib. NFCI. · c6f07c42
  Peter Collingbourne authored May 13, 2017
  c6f07c42
May 13, 2017
- [X86][SSE] Test showing missing EXTRACT_SUBVECTOR/CONCAT_VECTORS demandedelts... · 78b0ce03
  Simon Pilgrim authored May 13, 2017
  78b0ce03
- [SelectionDAG] Add VECTOR_SHUFFLE support to ComputeNumSignBits · 7666afd0
  Simon Pilgrim authored May 13, 2017
  7666afd0
- [X86][SSE] Test showing inability of ComputeNumSignBits to resolve shuffles · ded23a7f
  Simon Pilgrim authored May 13, 2017
  ded23a7f
- [ValueTracking] Remove const_casts on several calls to computeKnownBits and ComputeSignBit. NFC · 9fe35797
  Craig Topper authored May 13, 2017
  9fe35797
- MSan: Mark MemorySanitizer tests that use x86 intrinsics as REQUIRES: x86 · 3a3e115e
  Justin Bogner authored May 13, 2017
  3a3e115e
- [x86, SSE] AVX1 PR28129 (256-bit all-ones rematerialization) · ef46c276
  Simon Pilgrim authored May 13, 2017
  ef46c276
- [LoopOptimizer][Fix]PR32859, PR24738 · 7d62e4b4
  Simon Pilgrim authored May 13, 2017
  7d62e4b4
- [Simplify] Remove identical write removal. NFC. · fa7be883
  Michael Kruse authored May 13, 2017
  fa7be883
- [Simplify] Remove writes that are overwritten. · f263610b
  Michael Kruse authored May 13, 2017
  f263610b
- This reverts r302984 · 1d12790f
  Vivek Pandya authored May 13, 2017
  1d12790f
- Simplify MIR Output used for Codegen Testing · d20de87f
  Vivek Pandya authored May 13, 2017
  d20de87f
- [APInt] Use Lo_32/Hi_32/Make_64 in a few more places in the divide code. NFCI · 2c9a7066
  Craig Topper authored May 13, 2017
  2c9a7066