Commits · da362150174d5c2d535ee045413ed31357f8ca16 · Roger Ferrer / llvm-epi · GitLab

Dec 30, 2016
- [MemDep] Handle gep with zeros for invariant.group · da362150
  Piotr Padlewski authored Dec 30, 2016
  da362150
- Fix test change in r290736: restore index generation · e2770c0b
  Mehdi Amini authored Dec 30, 2016
  e2770c0b
- [LICM] When promoting scalars, allow inserting stores to thread-local allocas. · 76e06c88
  Michael Kuperstein authored Dec 30, 2016
  76e06c88
- Use continuous boosting factor for complete unroll. · cc76344e
  Dehao Chen authored Dec 30, 2016
  cc76344e
- Replace test from using llvm-lto to use llvm-link (NFC) · 30a9b6bb
  Mehdi Amini authored Dec 30, 2016
  30a9b6bb
- [AVR] Optimize 16-bit ORs with '0' · 453d0429
  Dylan McKay authored Dec 30, 2016
  453d0429
Dec 29, 2016
- NewGVN: Fix PR 31491 by ensuring that we touch the right instructions. Change... · e0bd37e7
  Daniel Berlin authored Dec 29, 2016
  e0bd37e7
- Revert "[COFF] Use 32-bit jump table entries in .rdata for Win64" · cd46c1df
  Reid Kleckner authored Dec 29, 2016
  cd46c1df
- [TBAAVerifier] Be stricter around verifying scalar nodes · 00d76a57
  Sanjoy Das authored Dec 29, 2016
  00d76a57
- [AMDGPU][mc] Enable absolute expressions in .hsa_code_object_isa directive · 25478d82
  Artem Tamazov authored Dec 29, 2016
  25478d82
- Introduce element-wise atomic memcpy intrinsic · 4f31e52f
  Igor Laevsky authored Dec 29, 2016
  4f31e52f
- [InstCombine] Fix some of the AVX-512 scalar arithmetic test cases to do a... · b57a84da
  Craig Topper authored Dec 29, 2016
  b57a84da
- [COFF] Use 32-bit jump table entries in .rdata for Win64 · c9e0a153
  Reid Kleckner authored Dec 29, 2016
  c9e0a153
Dec 28, 2016
- Add an index for Module Metadata record in the bitcode · e98f9258
  Mehdi Amini authored Dec 28, 2016
  e98f9258
- Revert "Add an index for Module Metadata record in the bitcode" · 2b59eca1
  Saleem Abdulrasool authored Dec 28, 2016
  2b59eca1
- Add an index for Module Metadata record in the bitcode · 32ca1481
  Mehdi Amini authored Dec 28, 2016
  32ca1481
- [WinEH] Don't assume endFunction is called while in .text · 92647369
  Reid Kleckner authored Dec 28, 2016
  92647369
- [PM] Introduce a devirtualization iteration layer for the new PM. · 05ca5acc
  Chandler Carruth authored Dec 28, 2016
  05ca5acc
- [PM] Teach the CGSCC's CG update utility to more carefully invalidate · 443e57e0
  Chandler Carruth authored Dec 28, 2016
  443e57e0
- This is a large patch for X86 AVX-512 of an optimization for reducing code... · 19c4fc5e
  Gadi Haber authored Dec 28, 2016
  19c4fc5e
- [PM] Teach the inliner's call graph update to handle inserting new edges · 9900d18b
  Chandler Carruth authored Dec 28, 2016
  9900d18b
- [PM] Actually commit the test update that was supposed to accompany · 69c5cc69
  Chandler Carruth authored Dec 28, 2016
  69c5cc69
- [InstCombine] Canonicalize insert splat sequences into an insert + shuffle · cd7ad713
  Michael Kuperstein authored Dec 28, 2016
  cd7ad713
Dec 27, 2016
- [sanitizer-coverage] sort the switch cases · f24e52c0
  Kostya Serebryany authored Dec 27, 2016
  f24e52c0
- [PM] Teach MemDep to invalidate its result object when its cached · e14524ca
  Chandler Carruth authored Dec 27, 2016
  e14524ca
- test: modernise ARM CodeGen tests · 0ce0dc25
  Saleem Abdulrasool authored Dec 27, 2016
  0ce0dc25
- Attempt to make the Windows bots green after r290609. · b1ea99a9
  Chad Rosier authored Dec 27, 2016
  b1ea99a9
- [PM] Add more dedicated testing to cover the invalidation logic added to · 7a73eabf
  Chandler Carruth authored Dec 27, 2016
  7a73eabf
- [MemCpyOpt] Don't sink LoadInst below possible clobber. · 7cb74462
  Bryant Wong authored Dec 27, 2016
  7cb74462
- [ThinLTO] Fix "||" vs "|" mixup. · e0ee5cf7
  Teresa Johnson authored Dec 27, 2016
  e0ee5cf7
- [AArch64][AsmParser] Add support for parsing shift/extend operands with symbols. · 2ff37b86
  Chad Rosier authored Dec 27, 2016
  2ff37b86
- [AMDGPU][llvm-mc] Predefined symbols to access register counts (.kernel.{v|s}gpr_count) · a01cce88
  Artem Tamazov authored Dec 27, 2016
  a01cce88
- [PM] Teach BasicAA how to invalidate its result object. · aa351675
  Chandler Carruth authored Dec 27, 2016
  aa351675
- [PM] Disable more of the loop passes -- LCSSA and LoopSimplify are also · 81c8edaf
  Chandler Carruth authored Dec 27, 2016
  81c8edaf
- [AMDGPU] Assembler: support SDWA and DPP for VOP2b instructions · e66365e0
  Sam Kolton authored Dec 27, 2016
  e66365e0
- RuntimeDyldELF: add R_AARCH64_ADD_ABS_LO12_NC reloc · 92090835
  Eugene Leviant authored Dec 27, 2016
  92090835
- [PM] Teach the AAManager and AAResults layer (the worst offender for · 17c630a0
  Chandler Carruth authored Dec 27, 2016
  17c630a0
- [PM] Turn on the new PM's inliner in addition to the current one for · 625038d5
  Chandler Carruth authored Dec 27, 2016
  625038d5
- [AVX-512] Add all forms of VPALIGNR, VALIGND, and VALIGNQ to the load folding tables. · e77e9011
  Craig Topper authored Dec 27, 2016
  e77e9011
- [PM] Add one of the features left out of the initial inliner patch: · 141bf5d1
  Chandler Carruth authored Dec 27, 2016
  141bf5d1