Commits · e5c6a321637ee6b2a7b52e2791a34279ba26d8b7 · Roger Ferrer / llvm-epi · GitLab

Jan 31, 2017
- [mips] Addition of the immediate cases for the instructions [d]div, [d]divu · 12850eea
  Simon Dardis authored Jan 31, 2017
  12850eea
- [X86] Add AVX and SSE2 version of MOVSDmr to execution domain fixing table.... · 797e32dd
  Craig Topper authored Jan 31, 2017
  797e32dd
- [AVX-512] Fix copy and paste bug in execution domain fixing tables so that we... · 779e4c5b
  Craig Topper authored Jan 31, 2017
  779e4c5b
- [NVPTX] Implement NVPTXTargetLowering::getSqrtEstimate. · 1c9692a4
  Justin Lebar authored Jan 31, 2017
  1c9692a4
- [X86] Update the broadcast fallback patterns to use shuffle instructions from... · 06e038c6
  Craig Topper authored Jan 31, 2017
  06e038c6
- [X86] Add test cases for AVX1 broadcast fallback patterns when load can't be folded. · 88b0a473
  Craig Topper authored Jan 31, 2017
  88b0a473
- AMDGPU: Generalize matching of v_med3_f32 · f84e5d9a
  Matt Arsenault authored Jan 31, 2017
  f84e5d9a
- InferAddressSpaces: Handle icmp · 72f259b8
  Matt Arsenault authored Jan 31, 2017
  72f259b8
- InferAddressSpaces: Support memory intrinsics · 6d5a8d48
  Matt Arsenault authored Jan 31, 2017
  6d5a8d48
- InferAddressSpaces: Support atomics · 6c907a9b
  Matt Arsenault authored Jan 31, 2017
  6c907a9b
- InferAddressSpaces: Don't replace volatile users · d89a6e11
  Matt Arsenault authored Jan 31, 2017
  d89a6e11
- AMDGPU: Implement hook for InferAddressSpaces · b6491cc8
  Matt Arsenault authored Jan 31, 2017
  b6491cc8
- NVPTX: Move InferAddressSpaces to generic code · 850657a4
  Matt Arsenault authored Jan 31, 2017
  850657a4
- [SCEV] Simplify/generalize howFarToZero solving. · 10d1ff64
  Eli Friedman authored Jan 31, 2017
  10d1ff64
- [ExecutionDepsFix] Improve clearance calculation for loops · 578cf7aa
  Keno Fischer authored Jan 30, 2017
  578cf7aa
- [InstCombine] enable (X <<nsw C1) >>s C2 --> X <<nsw (C1 - C2) for vectors with splat constants · 8c5f2361
  Sanjay Patel authored Jan 30, 2017
  8c5f2361
- [WebAssembly] Add wasm support for llvm-readobj · 6d76b7b4
  Derek Schuff authored Jan 30, 2017
  6d76b7b4
- [InstCombine] add vector test for (X <<nsw C1) >>s C2 --> X <<nsw (C1 - C2); NFC · abbb118a
  Sanjay Patel authored Jan 30, 2017
  abbb118a
- NVPTX: Refactor NVPTXInferAddressSpaces to check TTI · 42b64783
  Matt Arsenault authored Jan 30, 2017
  42b64783
- [InstCombine] enable more lshr(shl X, C1), C2 folds for vectors with splat constants · 0c39d56a
  Sanjay Patel authored Jan 30, 2017
  0c39d56a
Jan 30, 2017
- Revert r292979 which causes compile time failure. · 6217fa44
  Dehao Chen authored Jan 30, 2017
  6217fa44
- [InstCombine] add tests for more shift-shift patterns; NFC · 98cc8414
  Sanjay Patel authored Jan 30, 2017
  98cc8414
- Fix line endings. · 23457332
  Eli Friedman authored Jan 30, 2017
  23457332
- Reapply [AMDGPU][mc][tests][NFC] Add coverage/smoke tests for Gfx7 and Gfx8. · 61eb79d7
  Artem Tamazov authored Jan 30, 2017
  61eb79d7
- Re-commit AMDGPU/GlobalISel: Add support for simple shaders · ca16621b
  Tom Stellard authored Jan 30, 2017
  ca16621b
- GlobalISel: correctly translate invoke when callee is a register. · 2bf8c9d3
  Tim Northover authored Jan 30, 2017
  2bf8c9d3
- [AMDGPU] Internalize non-kernel symbols · a3b72798
  Stanislav Mekhanoshin authored Jan 30, 2017
  a3b72798
- Change the llvm-obdump(1) behavior with the -macho flag and inappropriate file types. · 98898f2b
  Kevin Enderby authored Jan 30, 2017
  98898f2b
- GlobalISel: account for differing exception selector sizes. · c9449704
  Tim Northover authored Jan 30, 2017
  c9449704
- LSR: Don't drop address space when type doesn't match · 1f2ca663
  Matt Arsenault authored Jan 30, 2017
  1f2ca663
- GlobalISel: translate memset & memmove. · 79f43f19
  Tim Northover authored Jan 30, 2017
  79f43f19
- AMDGPU: Undo sub x, c -> add x, -c canonicalization · af635240
  Matt Arsenault authored Jan 30, 2017
  af635240
- GlobalISel: permit unused vregs without a register-class after ISel. · 480609d0
  Tim Northover authored Jan 30, 2017
  480609d0
- [X86][XOP] Fix test name · 3ddc94d3
  Simon Pilgrim authored Jan 30, 2017
  3ddc94d3
- [InstCombine] enable (X >>?exact C1) << C2 --> X >>?exact (C1-C2) for vectors with splat constants · 373db5ba
  Sanjay Patel authored Jan 30, 2017
  373db5ba
- SDAG: Update ChainNodesMatched during UpdateChains if a node is replaced · 8f520a73
  Justin Bogner authored Jan 30, 2017
  8f520a73
- [InstCombine] add vector splat tests for (X >>?exact C1) << C2 --> X >>?exact (C1-C2); NFC · 1a86607d
  Sanjay Patel authored Jan 30, 2017
  1a86607d
- NewGVN: Instead of changeToUnreachable, insert an instruction SimplifyCFG will... · a53a7224
  Daniel Berlin authored Jan 30, 2017
  a53a7224
- AMDGPU: Make i32 uaddo/usubo legal · ee3f0acf
  Matt Arsenault authored Jan 30, 2017
  ee3f0acf
- DAG: Fold fneg into compare with constant into the constant · 32e6bfa2
  Matt Arsenault authored Jan 30, 2017
  32e6bfa2