Commits · 91c2a99740ae33b0c21f5cb3362499a44b9dd54e · Lorenzo Albano / LLVM bpEVL · GitLab

Jul 10, 2014
- [x86] Tweak the v16i8 single input special case lowering for shuffles · a34a8e23
  Chandler Carruth authored Jul 10, 2014
  a34a8e23
- Fix isDereferenceablePointer not to try to take the size of an unsized type. · 66e23f12
  Hal Finkel authored Jul 10, 2014
  66e23f12
- Allow isDereferenceablePointer to look through some bitcasts · 2e42c34d
  Hal Finkel authored Jul 10, 2014
  2e42c34d
- MC: modernise for loop · 1e76cbdf
  Saleem Abdulrasool authored Jul 10, 2014
  1e76cbdf
- MC: add and use an accessor for WinCFI · 427c08d4
  Saleem Abdulrasool authored Jul 10, 2014
  427c08d4
- Remove move assignment operator to appease older GCCs. · 8876c3fa
  Peter Collingbourne authored Jul 10, 2014
  8876c3fa
- [x86] Initial improvements to the new shuffle lowering for v16i8 · 7d2ffb54
  Chandler Carruth authored Jul 10, 2014
  7d2ffb54
- Explicitly define move constructor and move assignment operator to appease MSVC. · 05b9ebf2
  Peter Collingbourne authored Jul 10, 2014
  05b9ebf2
- SpecialCaseList: use std::unique_ptr. · d5feb7ba
  Peter Collingbourne authored Jul 10, 2014
  d5feb7ba
- [AArch64]Fix an assertion failure in DAG Combiner about concating 2 build_vector. · 71224b02
  Hao Liu authored Jul 10, 2014
  71224b02
- R600/SI: Add support for llvm.convert.{to|from}.fp16 · b0df9257
  Matt Arsenault authored Jul 10, 2014
  b0df9257
- [x86] Refactor some of the new code for lowering v16i8 shuffles to · b3840a55
  Chandler Carruth authored Jul 10, 2014
  b3840a55
- [dfsan] Handle bitcast aliases. · 2e28edf8
  Peter Collingbourne authored Jul 10, 2014
  2e28edf8
- [SDAG] Make the new zext-vector-inreg node default to expand so targets · d3561f6f
  Chandler Carruth authored Jul 09, 2014
  d3561f6f
Jul 09, 2014
- Recommit r212203: Don't try to construct debug LexicalScopes hierarchy for... · 029bd335
  David Blaikie authored Jul 09, 2014
  029bd335
- Decouple llvm::SpecialCaseList text representation and its LLVM IR semantics. · b7dd329f
  Alexey Samsonov authored Jul 09, 2014
  b7dd329f
- Add trunc (select c, a, b) -> select c (trunc a), (trunc b) combine. · 658c5576
  Matt Arsenault authored Jul 09, 2014
  658c5576
- AArch64: Better codegen for storing to __fp16. · 34cc92b4
  Jim Grosbach authored Jul 09, 2014
  34cc92b4
- TargetRegisterInfo: Remove function that fell out of use years ago. · c560a6ca
  Benjamin Kramer authored Jul 09, 2014
  c560a6ca
- [X86] AVX512: Enable it in the Loop Vectorizer · 2820a5b9
  Adam Nemet authored Jul 09, 2014
  2820a5b9
- Make AArch64FastISel::EmitIntExt explicitly check its source and destination types · 1ce0c37b
  Louis Gerbarg authored Jul 09, 2014
  1ce0c37b
- Fix for PR20059 (instcombine reorders shufflevector after instruction that may trap) · 58814445
  Sanjay Patel authored Jul 09, 2014
  58814445
- Add Imagination Technologies to the vendors in llvm::Triple · c5626f44
  Daniel Sanders authored Jul 09, 2014
  c5626f44
- Generic: add range-adapter for option parsing. · ac002d3e
  Tim Northover authored Jul 09, 2014
  ac002d3e
- [x86] Fix a bug in my new zext-vector-inreg DAG trickery where we were · 5865a73a
  Chandler Carruth authored Jul 09, 2014
  5865a73a
- Sink two variables only used in an assert into the assert itself. Should · 14cad41e
  Chandler Carruth authored Jul 09, 2014
  14cad41e
- X86: When lowering v8i32 himuls use the correct shuffle masks for AVX2. · d6f1733a
  Benjamin Kramer authored Jul 09, 2014
  d6f1733a
- [x86] Add a ZERO_EXTEND_VECTOR_INREG DAG node and use it when widening · afe4b250
  Chandler Carruth authored Jul 09, 2014
  afe4b250
- [mips][mips64r6] Correct select patterns that have the condition or true/false values backwards · e31155fd
  Daniel Sanders authored Jul 09, 2014
  e31155fd
- [mips][mips64r6] Correct cond names in the cmp.cond.[ds] instructions · dc06718e
  Daniel Sanders authored Jul 09, 2014
  dc06718e
- [x86] Initialize a pointer to null to fix a bug in r212602. · ef5dcf57
  Chandler Carruth authored Jul 09, 2014
  ef5dcf57
- [mips][mips64r6] Use JALR for indirect branches instead of JR (which is not... · f5a5fbd3
  Daniel Sanders authored Jul 09, 2014
  f5a5fbd3
- [mips][mips64r6] Use JALR for returns instead of JR (which is not available on MIPS32r6/MIPS64r6) · 338513b3
  Daniel Sanders authored Jul 09, 2014
  338513b3
- [x86] Re-apply a variant of the x86 side of r212324 now that the rest · 2ebc9426
  Chandler Carruth authored Jul 09, 2014
  2ebc9426
- [ASan/Win] Don't instrument COMDAT globals. Properly fixes PR20244. · e40fb373
  Timur Iskhodzhanov authored Jul 09, 2014
  e40fb373
- SourceMgr: consistently use 'unsigned' for the memory buffer ID type · a5b27a71
  Dmitri Gribenko authored Jul 09, 2014
  a5b27a71
- [SDAG] At the suggestion of Hal, switch to an output parameter that · f0a33b71
  Chandler Carruth authored Jul 09, 2014
  f0a33b71
- MipsTargetStreamer.h: Avoid "using" to appease msc17. · 843c4cb4
  NAKAMURA Takumi authored Jul 08, 2014
  843c4cb4
- AArch64: Better codegen for loading from __fp16. · 04691a53
  Jim Grosbach authored Jul 08, 2014
  04691a53
- Improve BasicAA CS-CS queries · 8ae0f8d6
  Hal Finkel authored Jul 08, 2014
  8ae0f8d6