Commits · 6a0bc9954218d9e0a6be0f99d062919099a1938e · Roger Ferrer / llvm-epi-0.8 · GitLab

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

May 23, 2014
- [Graph Writer] Limit the length of the graph name because Windows can't handle it. · 6a0bc995
  Michael J. Spencer authored May 22, 2014
  6a0bc995
- Make these bool bitfields. · 276d7a14
  Eric Christopher authored May 22, 2014
  276d7a14
- [RuntimeDyld] Teach RuntimeDyldMachO how to handle scattered VANILLA relocs on · 7f9fc2b3
  Lang Hames authored May 22, 2014
  7f9fc2b3
May 22, 2014
- Update some AliasAnalysis pass docs for getAdjustedAnalysisPointer. · d328db13
  Eric Christopher authored May 22, 2014
  d328db13
- R600: Add definition for flat address space ID. · 46b51b7f
  Matt Arsenault authored May 22, 2014
  46b51b7f
- R600: Try to convert BFE back to standard bit ops when possible. · 05e96f44
  Matt Arsenault authored May 22, 2014
  05e96f44
- R600: Add dag combine for BFE · 5565f65e
  Matt Arsenault authored May 22, 2014
  5565f65e
- R600: Implement ComputeNumSignBitsForTargetNode for BFE · bf8694d3
  Matt Arsenault authored May 22, 2014
  bf8694d3
- R600: Implement computeMaskedBitsForTargetNode for BFE · af6df9d9
  Matt Arsenault authored May 22, 2014
  af6df9d9
- R600: Expand mul24 for GPUs without it · 493c5f1b
  Matt Arsenault authored May 22, 2014
  493c5f1b
- R600: Expand mad24 for GPUs without it · f15a0562
  Matt Arsenault authored May 22, 2014
  f15a0562
- R600: Add intrinsics for mad24 · eb260206
  Matt Arsenault authored May 22, 2014
  eb260206
- Return false if we're not going to do anything. · 9eff5178
  Eric Christopher authored May 22, 2014
  9eff5178
- R600/SI: Move instruction pattern to instruction definition · f37abc71
  Matt Arsenault authored May 22, 2014
  f37abc71
- Remove LLVMContextImpl::optimizationRemarkEnabledFor. · 0b761a48
  Diego Novillo authored May 22, 2014
  0b761a48
- [X86] Improve the lowering of BITCAST from MVT::f64 to MVT::v4i16/MVT::v8i8. · c8dd1ad8
  Andrea Di Biagio authored May 22, 2014
  c8dd1ad8
- ARM64: remove '#' from annotation of add/sub immediate · b2a6fdb1
  Tim Northover authored May 22, 2014
  b2a6fdb1
- Add support for missed and analysis optimization remarks. · 7f8af8bf
  Diego Novillo authored May 22, 2014
  7f8af8bf
- Segmented stacks: omit __morestack call when there's no frame. · f9e798ba
  Tim Northover authored May 22, 2014
  f9e798ba
- ARM64: these work too · 2dce43c2
  Tim Northover authored May 22, 2014
  2dce43c2
- Yes they do · 5949b605
  Tim Northover authored May 22, 2014
  5949b605
- ARM64: model pre/post-indexed operations properly. · 4a3ab28a
  Tim Northover authored May 22, 2014
  4a3ab28a
- ARM64: separate load/store operands to simplify assembler · c350acfd
  Tim Northover authored May 22, 2014
  c350acfd
- [mips] Make unalignedload.ll test stricter and easier to modify for MIPS32r6/MIPS64r6 · 18b68e16
  Daniel Sanders authored May 22, 2014
  18b68e16
- [mips] Change lwl and lwr in inlineasm_constraint.ll to lw · e1f46e58
  Daniel Sanders authored May 22, 2014
  e1f46e58
- [mips] Use addiu in inline assembly tests since addi is not available in all ISA's · b2305958
  Daniel Sanders authored May 22, 2014
  b2305958
- Extend sys::getHostCPUFeatures to work on AArch64 platforms · 9288b218
  Bradley Smith authored May 22, 2014
  9288b218
- [mips][mips64r6] addi is not available on MIPS32r6/MIPS64r6 · 36ff7c2a
  Daniel Sanders authored May 22, 2014
  36ff7c2a
- [mips][mips64r6] Test that paired single instructions are invalid · a3566c63
  Daniel Sanders authored May 22, 2014
  a3566c63
- [mips][mips64r6] Add b[on]vc · 5c582b2f
  Daniel Sanders authored May 22, 2014
  5c582b2f
- ARM64: assert if we see i64 -> i64 extend in the DAG. · 0f627227
  Tim Northover authored May 22, 2014
  0f627227
- AArch64/ARM64: enable more AArch64 tests. · 71d04225
  Tim Northover authored May 22, 2014
  71d04225
- MC: initialise MCAsmParser variable · 9dd60cfb
  Saleem Abdulrasool authored May 22, 2014
  9dd60cfb
- Remove unused variable. · 65382d73
  Eric Christopher authored May 22, 2014
  65382d73
- ARM: introduce llvm.arm.undefined intrinsic · 2bd1262a
  Saleem Abdulrasool authored May 22, 2014
  2bd1262a
- R600/SI: Match fp_to_uint / uint_to_fp for f64 · c3a73c30
  Matt Arsenault authored May 22, 2014
  c3a73c30
- MC: formalise some assertions into proper errors · 6663f8f2
  Saleem Abdulrasool authored May 22, 2014
  6663f8f2
- Override runOnMachineFunction for ARMISelDAGToDAG so that we can · 0e6e7cf3
  Eric Christopher authored May 22, 2014
  0e6e7cf3
- Override runOnMachineFunction for X86ISelDAGToDAG so that we can · 4f09c592
  Eric Christopher authored May 22, 2014
  4f09c592
- Avoid using subtarget features when adding X86 specific passes to · 0d5c99eb
  Eric Christopher authored May 22, 2014
  0d5c99eb