Commits · e63b574ff6a827ac64128490f117c94a54859992 · Roger Ferrer / llvm-epi · GitLab

Sep 14, 2022
- [mlir][LLVMIR] Add lifetime start and end marker instrinsics · e63b574f
  Jeff Niu authored Sep 14, 2022
  e63b574f
- [mlir][linalg] Vectorization for conv_1d_ncw_fcw · 026fac2a
  Stanley Winata authored Sep 14, 2022
  026fac2a
- [mlir][vector] Clean up and generalize lowering of warp_execute to scf · 4abb9e5d
  Thomas Raoux authored Sep 14, 2022
  4abb9e5d
- [mlir][sparse] Make sparse compiler more admissible. · 55a1d50f
  Peiming Liu authored Sep 12, 2022
  55a1d50f
- [mlir][Linalg] Add return type filter to the transform dialect · 593c14d4
  Nicolas Vasilache authored Sep 14, 2022
  593c14d4
- [MLIR][math] Use approximate matches for folded ops · 3108249d
  Jeff Niu authored Sep 13, 2022
  3108249d
- [MLIR] Fix toy lit substitutions · 5cff32b9
  Christian Sigg authored Sep 14, 2022
  5cff32b9
- [mlir][vector] Fold scalar vector.extract of non-splat n-D constants · 8c2ea144
  Jakub Kuderski authored Sep 13, 2022
  8c2ea144
Sep 13, 2022
- [mlir][sparse] Add sparse_tensor.select operation · 07150fec
  Jim Kitchen authored Sep 13, 2022
  07150fec
- [mlir] Add bytecode encodings for the builtin ElementsAttr attributes · 5fb1bbe6
  River Riddle authored Sep 12, 2022
  5fb1bbe6
- [mlir] Add fallback support for parsing/printing unknown external resources · 34300ee3
  River Riddle authored Sep 13, 2022
  34300ee3
- [mlir:Bytecode] Add support for encoding resources · 6ab2bcff
  River Riddle authored Sep 06, 2022
  6ab2bcff
- [MLIR] Remove unused lit test replacements. · c5197517
  Christian Sigg authored Sep 07, 2022
  c5197517
- [mlir][arith] Canonicalization patterns for subi. · d8a57c77
  Slava Zakharin authored Sep 09, 2022
  d8a57c77
- [mlir][Vector] Support broadcast vector type in distribution of vector.warp_execute_on_lane_0. · 845dc178
  Nicolas Vasilache authored Sep 13, 2022
  845dc178
- [mlir] improve the error message in expensive transform checks · 59a640a3
  Alex Zinenko authored Sep 13, 2022
  59a640a3
- [MLIR] Only substitute `mlir-spirv-cpu-runner` tool if enabled · 5c6d8a74
  Christian Sigg authored Sep 13, 2022
  5c6d8a74
- [mlir][NFC] Fix polynomial Approx test. · d67603a6
  Jianjian GUAN authored Sep 09, 2022
  d67603a6
- [mlir][tosa] Added tosa.reverse folder · 3e49a3e8
  Rob Suderman authored Sep 12, 2022
  3e49a3e8
- Apply clang-tidy fixes for performance-unnecessary-value-param in TestTilingInterface.cpp (NFC) · 4bd66792
  Mehdi Amini authored Aug 29, 2022
  
  4bd66792
Sep 12, 2022
- [mlir][sparse] add memSizes array to sparse storage format · 6607fdf7
  Aart Bik authored Sep 12, 2022
  6607fdf7
- Apply clang-tidy fixes for modernize-use-equals-default in TestPatterns.cpp (NFC) · cde2736c
  Mehdi Amini authored Aug 29, 2022
  
  cde2736c
- Apply clang-tidy fixes for modernize-use-equals-default in TestLinalgDecomposeOps.cpp (NFC) · 6e5c654e
  Mehdi Amini authored Aug 29, 2022
  
  6e5c654e
Sep 10, 2022
- [mlir][sparse] fix some asan detected leaks in integration tests · 0078f723
  Aart Bik authored Sep 09, 2022
  0078f723
- [mlir][linalg] Add decomposition from conv_2d_nchw · 8e484b52
  Stanley Winata authored Sep 09, 2022
  8e484b52
Sep 09, 2022
- [mlir][sparse] Add new option (enable-runtime-library) to sparse compiler pipeline · eb65327f
  Peiming Liu authored Sep 09, 2022
  eb65327f
- [mlir][arith] Support wide integer addition emulation · c7f64616
  Jakub Kuderski authored Sep 09, 2022
  c7f64616
- [mlir][spirv] Add support for fast math mode · 6329c738
  Lei Zhang authored Sep 09, 2022
  6329c738
- [mlir][CallGraph] Add special call graph node for representing unknown callees · 82f86b86
  Markus Böck authored Sep 09, 2022
  82f86b86
- [mlir][vector] Fix bug in transfer op flattening · 9e7c97d8
  Thomas Raoux authored Sep 09, 2022
  9e7c97d8
- [mlir][linalg] Fix tiling interface implementation ordering of parallel_insert_slice · 61a4b228
  Guray Ozen authored Sep 09, 2022
  61a4b228
- [mlir][linalg] Relax tiling constraint when there are multiple destination operands · a367c571
  Guray Ozen authored Sep 09, 2022
  a367c571
- [mlir][vector] Extend WarpExecutionOnLane0 pattern support to allow... · 20df17fd
  Nicolas Vasilache authored Sep 09, 2022
  20df17fd
- [mlir][vector] NFC - Clean up vector patterns and propagate benefit through populate functions · 27cc31b6
  Nicolas Vasilache authored Sep 09, 2022
  27cc31b6
- [mlir][vector] Don't duplicate transfer_read during vector distribution · 06413618
  Thomas Raoux authored Sep 08, 2022
  06413618
- [mlir][arith] Support wide integer constant emulation · 864236d1
  Jakub Kuderski authored Sep 08, 2022
  864236d1
- [mlir][Tensor] Add rewrites to extract slices through `tensor.collape_shape` · f4a478cd
  Christopher Bate authored Sep 08, 2022
  f4a478cd
- Reland "[mlir][arith] Add wide integer emulation pass" · 7fa1d743
  Jakub Kuderski authored Sep 08, 2022
  7fa1d743
- [mlir][Math] Add TruncOp. · 9d0b90e9
  Jianjian GUAN authored Sep 06, 2022
  9d0b90e9
- [mlir][math] Canonicalization for math.floor op · 5bb62105
  Kai Sasaki authored Sep 09, 2022
  5bb62105