Commits · 9e922e7d246ccd887f62966711b59a2d45227c07 · Raul Torres / llvm-target-spread · GitLab

Sep 29, 2014
- Use a loop to simplify the runtime unrolling prologue. · fc02e3c3
  Kevin Qin authored Sep 29, 2014
  fc02e3c3
- [Thumb2] ldrexd and strexd are not defined on v7M · a4eba5ad
  Oliver Stannard authored Sep 29, 2014
  a4eba5ad
- [x86] Make the new vector shuffle lowering lower blends as VSELECT · 6cbf4316
  Chandler Carruth authored Sep 29, 2014
  6cbf4316
- Remove dead code from DIBuilder · b76b57fe
  Jyoti Allur authored Sep 29, 2014
  b76b57fe
- [x86] Delete a bunch of really bad and totally unnecessary code in the · b1cc7a85
  Chandler Carruth authored Sep 29, 2014
  b1cc7a85
- [x86] Refactor all of the VSELECT-as-blend lowering code to avoid domain · d639c7a8
  Chandler Carruth authored Sep 29, 2014
  d639c7a8
- [x86] Improve naming and comments for VSELECT lowering. · 2f9e56e5
  Chandler Carruth authored Sep 29, 2014
  2f9e56e5
- [x86] Add the dispatch skeleton to the new vector shuffle lowering for · c7129276
  Chandler Carruth authored Sep 29, 2014
  c7129276
- [x86] Make the split-and-lower routine fully generic by relaxing the · 32a3ebda
  Chandler Carruth authored Sep 29, 2014
  32a3ebda
- [x86] Teach the new vector shuffle lowering to fall back on AVX-512 · 24e3b69c
  Chandler Carruth authored Sep 28, 2014
  24e3b69c
- [x86] Fix the new vector shuffle lowering's use of VSELECT for AVX2 · abe742e8
  Chandler Carruth authored Sep 28, 2014
  abe742e8
Sep 28, 2014
- Add MachineOperand::ChangeToFPImmediate and setFPImm · 93ffe58f
  Matt Arsenault authored Sep 28, 2014
  93ffe58f
- [x86] Fix a really silly bug that I introduced fixing another bug in the · 6578f920
  Chandler Carruth authored Sep 28, 2014
  6578f920
- [x86] Fix yet another bug in the new vector shuffle lowering's handling · b10c6b8e
  Chandler Carruth authored Sep 28, 2014
  b10c6b8e
- WinCOFFObjectWriter.cpp: make write_uint32_le more efficient · ba80b5d4
  Hans Wennborg authored Sep 28, 2014
  ba80b5d4
Sep 27, 2014
- [AArch64] Redundant store instructions should be removed as dead code · 463db9a7
  James Molloy authored Sep 27, 2014
  463db9a7
- Fix llvm::huge_valf multiple initializations with Visual C++. · 7b4133ac
  Yaron Keren authored Sep 27, 2014
  7b4133ac
- [x86] Fix yet another issue with widening vector shuffle elements. · f4b9e6b9
  Chandler Carruth authored Sep 27, 2014
  f4b9e6b9
- [x86] Fix terrible bugs everywhere in the new vector shuffle lowering · 4d03be17
  Chandler Carruth authored Sep 27, 2014
  4d03be17
- [x86] Flip the sentinel values used in the target shuffle mask decoding · 81e6b29f
  Chandler Carruth authored Sep 27, 2014
  81e6b29f
- Refactor reciprocal and reciprocal square root estimate into target-independent functions (part 2). · bdf1e388
  Sanjay Patel authored Sep 26, 2014
  bdf1e388
- Object: BSS/virtual sections don't have contents · dac39857
  David Majnemer authored Sep 26, 2014
  dac39857
- clang-format of ChangeStdinToBinary & ChangeStdoutToBinary. · abce3c4e
  Yaron Keren authored Sep 26, 2014
  abce3c4e
Sep 26, 2014
- [x86] Fix a moderately terrifying bug in the new 128-bit shuffle logic · f572f3b2
  Chandler Carruth authored Sep 26, 2014
  f572f3b2
- [IndVar] Don't widen loop compare unless IV user is sign extended. · 7b974b73
  Chad Rosier authored Sep 26, 2014
  7b974b73
- R600/SI: Use break instead of continue · 2dd3129b
  Matt Arsenault authored Sep 26, 2014
  2dd3129b
- R600/SI: Add a note about the order of the operands to div_scale · a276c3e0
  Matt Arsenault authored Sep 26, 2014
  a276c3e0
- R600/SI: Move finding SGPR operand to move to separate function · ee522bf2
  Matt Arsenault authored Sep 26, 2014
  ee522bf2
- R600/SI Allow same SGPR to be used for multiple operands · 6a0919fb
  Matt Arsenault authored Sep 26, 2014
  6a0919fb
- R600/SI: Partially move operand legalization to post-isel hook. · cb0ac3d1
  Matt Arsenault authored Sep 26, 2014
  cb0ac3d1
- R600/SI: Implement findCommutedOpIndices · 92befe79
  Matt Arsenault authored Sep 26, 2014
  92befe79
- R600/SI: Don't move operands that are required to be SGPRs · 5885bef6
  Matt Arsenault authored Sep 26, 2014
  5885bef6
- R600/SI: Don't assert on exotic operand types · 0bea8d83
  Matt Arsenault authored Sep 26, 2014
  0bea8d83
- R600/SI: Fix using wrong operand indices when commuting · aff65fbc
  Matt Arsenault authored Sep 26, 2014
  aff65fbc
- R600/SI: Remove apparently dead code in legalizeOperands · e50c1c4a
  Matt Arsenault authored Sep 26, 2014
  e50c1c4a
- Ignore annotation function calls in cost computation · 472b05b3
  David Peixotto authored Sep 26, 2014
  472b05b3
- [x86] The mnemonic is SHUFPS not SHUPFS. =[ I'm very bad at spelling · acd19064
  Chandler Carruth authored Sep 26, 2014
  acd19064
- [x86] In the new vector shuffle lowering, when trying to do another · 0c9ee10d
  Chandler Carruth authored Sep 26, 2014
  0c9ee10d
- [x86] Fix a large collection of bugs that crept in as I fleshed out the · 5afd4c26
  Chandler Carruth authored Sep 26, 2014
  5afd4c26
- Elide repeated register operand in Thumb1 instructions · 36c626e3
  Renato Golin authored Sep 26, 2014
  36c626e3