Commits · e57b49ee165d7d86cf5fa83be1190483d56e12a9 · Roger Ferrer / llvm-epi-0.8 · GitLab

This project is archived. Its data is read-only.

Apr 27, 2012
- Add mcpu to tests to prevent them from using AVX instructions on Sandy Bridge after r155618. · e57b49ee
  Craig Topper authored Apr 27, 2012
  e57b49ee
- Implement a bastardized ABI. · 1ec87ee0
  Evan Cheng authored Apr 27, 2012
  1ec87ee0
- - thumbv6 shouldn't imply +thumb2. Cortex-M0 doesn't suppport 32-bit Thumb2 · f52003de
  Evan Cheng authored Apr 27, 2012
  f52003de
- Use ConstantExpr::getExtractElement when constant-folding vectors · 90f3798f
  Dan Gohman authored Apr 27, 2012
  90f3798f
- Add instcombine patterns for the following transformations: · 7813dcee
  Chad Rosier authored Apr 26, 2012
  7813dcee
Apr 26, 2012
- Fix the SD scheduler to avoid gluing the same node twice. · 03fa574a
  Andrew Trick authored Apr 26, 2012
  03fa574a
- Use VLD1 in NEON extenting-load patterns instead of VLDR. · 3de97b7a
  Tim Northover authored Apr 26, 2012
  3de97b7a
- Teach the reassociate pass to fold chains of multiplies with repeated · 739ef80f
  Chandler Carruth authored Apr 26, 2012
  739ef80f
- Specify cpu to unbreak tests. · 8a8e9d1b
  Evan Cheng authored Apr 26, 2012
  8a8e9d1b
- If triple is armv7 / thumbv7 and a CPU is specified, do not automatically assume · 9f7ad310
  Evan Cheng authored Apr 26, 2012
  9f7ad310
Apr 25, 2012
- Try to fix llvm-arm-linux builder with -mcpu. · 6eeeb7e1
  Jakob Stoklund Olesen authored Apr 25, 2012
  6eeeb7e1
- Trivial change to make the test use -mcpu=generic so as to avoid · 82cac0ac
  Preston Gurd authored Apr 25, 2012
  82cac0ac
- Actually delete now-empty file. · eeb9e581
  Chandler Carruth authored Apr 25, 2012
  eeb9e581
- Reverting r155468. Chris and Chandler have convinced me that it's dangerous and · 2fd0c691
  Lang Hames authored Apr 25, 2012
  2fd0c691
- Do not use $gp as a dedicated global register if the target ABI is not O32. · 2020e27d
  Akira Hatanaka authored Apr 25, 2012
  2020e27d
- ARM: improved assembler diagnostics for missing CPU features. · 5117ef74
  Jim Grosbach authored Apr 24, 2012
  5117ef74
Apr 24, 2012
- ConstantFoldSelectInstruction swapped the operands of the select. · 450d69a5
  Nadav Rotem authored Apr 24, 2012
  450d69a5
- Fix the testcase. We do expect two vblendw on XMMs. · d50c3b2c
  Nadav Rotem authored Apr 24, 2012
  d50c3b2c
- Add a testcase for 155440 · edef7179
  Nadav Rotem authored Apr 24, 2012
  edef7179
- MachineBasicBlock::SplitCriticalEdge() should follow LLVM IR variant and... · 2d14d8ac
  Evan Cheng authored Apr 24, 2012
  2d14d8ac
- Add support for llvm.arm.neon.vmull* intrinsics to InstCombine. This fixes · 84531c2b
  Lang Hames authored Apr 24, 2012
  84531c2b
- Fix a crash on valid (if UB) bitcode that is produced for some global · aacb8a58
  Chandler Carruth authored Apr 24, 2012
  aacb8a58
- Add missing test cases for ARM VLD3 (single 3-element structure to all lanes) · 70be447e
  Kevin Enderby authored Apr 24, 2012
  70be447e
- Add missing test cases for ARM VLD4 (single 4-element structure to all lanes) · c8d223e4
  Kevin Enderby authored Apr 24, 2012
  c8d223e4
- AVX: We lower VECTOR_SHUFFLE and BUILD_VECTOR nodes into vbroadcast instructions · aa3ff8da
  Nadav Rotem authored Apr 24, 2012
  aa3ff8da
- FileCheck-ize tests. · 1981c0e5
  Bill Wendling authored Apr 24, 2012
  1981c0e5
- FileCheck-ize these tests. · 4cf911c0
  Bill Wendling authored Apr 24, 2012
  4cf911c0
- FileCheck-ize these tests. Harden some of them. · cd6df16c
  Bill Wendling authored Apr 24, 2012
  cd6df16c
Apr 23, 2012
- Optimize the vector UINT_TO_FP, SINT_TO_FP and FP_TO_SINT operations where the... · 3f8acfc3
  Nadav Rotem authored Apr 23, 2012
  3f8acfc3
- This patch fixes a problem which arose when using the Post-RA scheduler · 9a091475
  Preston Gurd authored Apr 23, 2012
  9a091475
- ARM: Add testcases for two-operand variants of VSRA/VRSRA/VSRI. · f6371b52
  Jim Grosbach authored Apr 23, 2012
  f6371b52
- Add ARM mode tests for the NEON vector shift-accumulate tests. · 76cdd136
  Jim Grosbach authored Apr 23, 2012
  76cdd136
- Tidy up. Reformat for ease of reading. · 5c7e9e5e
  Jim Grosbach authored Apr 23, 2012
  5c7e9e5e
- Revert r155365, r155366, and r155367. All three of these have regression · 3c3bb55a
  Chandler Carruth authored Apr 23, 2012
  3c3bb55a
- Hexagon V5 (floating point) support. · a3f8ba24
  Sirish Pande authored Apr 23, 2012
  a3f8ba24
- Support for Hexagon architectural feature, new value jump. · 2c7bf00f
  Sirish Pande authored Apr 23, 2012
  2c7bf00f
- Support for Hexagon VLIW Packetizer. · 6cd22515
  Sirish Pande authored Apr 23, 2012
  6cd22515
- Reapply r155136 after fixing PR12599. · 43bcb970
  Jakob Stoklund Olesen authored Apr 23, 2012
  43bcb970
Apr 22, 2012
- cleaned line endings in the newly added test file · 6c6cdec3
  Elena Demikhovsky authored Apr 22, 2012
  6c6cdec3
- Tidy up this test more: · 8ffa7c8a
  Chandler Carruth authored Apr 22, 2012
  8ffa7c8a