Google Git
Sign in
opensecura / 3p / openxla / iree / HEAD
  1. 612ad91 Upgrade all remaining code to free create functions. NFC. (#21902) by Jakub Kuderski · 13 hours ago latest-snapshot main
  2. fdad8dc Integrate LLVM at llvm-project/llvm@daf8f9fc1ccc6c5679bc89058fd66d8ea4da9d59 (#21893) by Rahul Kayaith · 13 hours ago
  3. bcd64b8 [Codegen] Upgrade iree dialects to free create functions. NFC. (#21898) by Jakub Kuderski · 14 hours ago
  4. b046d2e [LLVMCPU] Respect dominance when doing replacement of tile and fused values (#21901) by MaheshRavishankar · 15 hours ago
  5. 8d518fb [GPU] Remove MMAScheduleAttr (#21884) by Kunwar Grover · 17 hours ago
  6. 13b03d6 Upgrade IREE plugins to free create functions. NFC. (#21896) by Jakub Kuderski · 17 hours ago
  7. 5723e7e Reland "[VectorDistribute] Refactor layout configuration to a simpler logic" (#21895) by Kunwar Grover · 18 hours ago
  8. 0aded27 [ROCM] Update Ukernel infra to handle InnerTiledOp/Multi_MMA_MFMA (#21759) by Abhishek Varma · 19 hours ago
  9. f347ffa [Codegen] Upgrade Transforms and Utils to free create functions. NFC. (#21882) by Jakub Kuderski · 21 hours ago
  10. b61a4ce Upgrade GlobalOpt, InputConversion, ExternalInterfacess to free create function. NFC. (#21878) by Jakub Kuderski · 21 hours ago
  11. e6f54a2 [docs] Update the file config file for running ONNX operator tests on CPU. (#21892) by Han-Chung Wang · 21 hours ago
  12. 150be06 Bump version to 3.8.0 after 3.7.0 release. (#21852) by Sahil Faizal · 23 hours ago
  13. a523efe Add gfx950 ukernel patterns (#21856) by sebvince · 23 hours ago
  14. 60c1c1d [Codegen] Upgrade Dialect and Interfaces to free create functions. NFC. (#21881) by Jakub Kuderski · 23 hours ago
  15. f78d05f [Codegen] Upgrade LLVMCPU and LLVMGPU to free create functions. NFC. (#21880) by Jakub Kuderski · 24 hours ago
  16. cda3ce1 [Codegen] Upgrade Common, SPIRV, VMVX to free create functions. NFC. (#21879) by Jakub Kuderski · 24 hours ago
  17. e031c87 Upgrade Preprocessing and Modules to free create functions. NFC. (#21877) by Jakub Kuderski · 24 hours ago
  18. 4b10e33 [docs] Clarify compiler coding standards (#21886) by Jakub Kuderski · 25 hours ago
  19. bbc82b0 Revert "[VectorDistribute] Refactor layout configuration to a simpler logic" (#21887) by Kunwar Grover · 25 hours ago
  20. aa024b8 [StableHLO][CHLO]Refactor CHLO decompositions to follow upstream StableHLO (#21682) by Lekkala_Sravya-mcw · 25 hours ago
  21. dd1688b [VectorDistribute] Refactor layout configuration to a simpler logic (#21883) by Kunwar Grover · 36 hours ago
  22. f8d3f76 Avoid needles isa checks. NFC. (#21885) by Jakub Kuderski · 2 days ago
  23. 09647ef [NFC] Code Quality changes (#21876) by Muzammil · 3 days ago
  24. c56ee1f [Codegen][AMDGPU] Fix matmul miscompile on RDNA4 (#21873) by Jakub Kuderski · 3 days ago
  25. 124fb35 [GPU] Use Affine map for size calculations of alloca's in fission pass (#21870) by Nirvedh Meshram · 4 days ago
  26. 4a3c014 [CPU] Remove passing tests from expected_compile_failures list. (#21871) by Han-Chung Wang · 4 days ago
  27. fce488a [GPU] Remove reshape by expansion in workgroup scope of combine layout pass (#21869) by Nirvedh Meshram · 4 days ago
  28. 3354861 [Codegen][IGEMM] Do not pre-pad convs with CHW layout or small input channel size (#21839) by Vivian Zhang · 4 days ago
  29. ed30f30 [GPU] Add pattern to fold fill into pad ops (#21864) by Nirvedh Meshram · 4 days ago
  30. 963e2e9 [CodeGen] Do not fuse parallel ops if they directly write to destination. (#21837) by Han-Chung Wang · 4 days ago
  31. 1c54e4d [Test] Add onnx_ops test suites with O2/O3 optimization level. (#21838) by Han-Chung Wang · 4 days ago
  32. c2a1627 [Encoding] Support SetEncoding on scaled contraction ops (#21825) by Max191 · 4 days ago
  33. f807607 [Integrate] Drop llvm/llvm-project@b4c31dc revert. (#21851) by Han-Chung Wang · 4 days ago
  34. 5db83bf [Codegen][Tuner] retire the C/Python binding for querying mma intrinsic. NFC. (#21816) by Bangtian Liu · 5 days ago
  35. 0516edc [Codegen][Tuner]: improve python binding to query target info (#21812) by Bangtian Liu · 5 days ago
  36. 933f798 [DT] Fuse encoding ops more aggressively for multi-use, gather, and slices ops. (#21830) by Han-Chung Wang · 6 days ago
  37. b4da7b2 Integrate LLVM at llvm/llvm-project@9c7727c62af0 (#21835) by Fabian Mora · 6 days ago
  38. 83789af [iree-test-suites] Add data tiling tests for LLAMA 8B (#21832) by Abhishek Varma · 6 days ago
  39. a327b2d [Hoisting] Fix the double-free issue in `HoistIntoGlobalsPass::cleanupDeadOp`. (#21699) by Jerry Shih · 6 days ago
  40. 9303360 [Codegen][GPU] Use arithmetic intensity to guide gemm size categorization - step 3 (#21826) by Zhuoran Yin · 6 days ago
  41. 6633605 Integrate LLVM at llvm/llvm-project@74275a11038c (#21831) by Muzammil · 6 days ago
  42. 3212d89 Revert "[VectorDistribute] Correctly find new dimensions during reduction config" (#21810) by Kunwar Grover · 6 days ago
  43. b2ee8fa [codegen][rocdl] Remove ROCDLKernelConfig and ROCDLSelectLoweringStrategy (#21820) by Fabian Mora · 6 days ago
  44. 960809f [Codegen][LLVMGPU] Remove LLVMGPUWarpReduction pipeline (#21821) by James Newling · 6 days ago
  45. 95163e7 Revert "[codegen] more consumer fusion (#21521)" (#21819) by Praveen G · 6 days ago
  46. 9a76ffb [LinalgExt][NFC] Delete duplicated SingleBlockImplicitTerminator trait. (#21818) by Han-Chung Wang · 7 days ago
  47. d249161 [Codegen] Rewrite test so LLVMGPUWarpReduction is not used (#21770) by James Newling · 7 days ago
  48. f0e04ae Migrate ROCM ukernels from tuning spec to ukernel descriptor lowering (#21794) by Jorn Tuyls · 7 days ago
  49. 6cfd70e Move ROCM tests to fix dialect not registered error (#21811) by Jorn Tuyls · 7 days ago
  50. 4d91ffb [codegen] more consumer fusion (#21521) by Oleksandr "Alex" Zinenko · 8 days ago
  51. c37c680 [VectorDistribute] Do not handle bit extend during matmul configuration (#21798) by Kunwar Grover · 10 days ago
  52. 8c26dfc [VectorDistribute] Correctly find new dimensions during reduction config (#21797) by Kunwar Grover · 10 days ago
  53. 26f63c1 [GPU][DT] Fix LHS operand offset calculation for DataTiledMMAAttr (#21808) by Zhewen Yu · 11 days ago
  54. b7341d9 [ROCM] Add zero fill check to ukernel patterns (#21793) by Jorn Tuyls · 11 days ago
  55. 9fbb1fd [GPU] Add pattern to sink extract_slice through generic ops (#21796) by Nirvedh Meshram · 12 days ago
  56. ce92024 [Codegen][GPU] Adding new heuristics to take all dimensions into account when distributing tiles (#21803) by Zhuoran Yin · 12 days ago
  57. 31404c6 Integrate LLVM at llvm/llvm-project@f2e6ca805dbb (#21805) by Ian Wood · 12 days ago
  58. 1c0dfca Drop TensorCore/MMA pipelines. (#21741) by MaheshRavishankar · 12 days ago
  59. 3ea1e6c [Codegen][LLVMGPU] Give ops same config irrespective of generalized/specialized (#21769) by James Newling · 12 days ago
  60. dd684c4 [Dispatch][GlobalOpt] Improve transpose fusion for conv (#21778) by Ian Wood · 12 days ago
  61. 0c5ef6a [Codegen][GPU] Use arithmetic intensity to guide gemm size categorization - step 2 (#21691) by Zhuoran Yin · 12 days ago
  62. 5ab8a51 [Codegen] Remove WarpReduction from ROCDL pipeline (#21795) by James Newling · 12 days ago
  63. 7460fcd [Codegen][Tuner] expose python binding to query target info (#21782) by Bangtian Liu · 12 days ago
  64. 639c7cf Integrate LLVM at llvm/llvm-project@4b84223aad4f (#21791) by Ian Wood · 13 days ago
  65. 44b9780 [NFC] Change debug messages (#21768) by Muzammil · 13 days ago
  66. 1a13c77 [GPU][DT] Fix matmul narrow dim selection (#21764) by Zhewen Yu · 13 days ago
  67. 73c0d4f [Codegen] Add XOR-based Swizzle Attribute (#21562) by sebvince · 13 days ago
  68. f14e6b2 [ROCM] Update Ukernel infra to allow ROCM-specific bitcode ukernel lowering (#21681) by Abhishek Varma · 13 days ago
  69. 25d8239 [Codegen][IGEMM] Fix and preserve padding dim order for convs (#21772) by Vivian Zhang · 14 days ago
  70. 8ba9f68 [ROCM] Fix redefinition of symbol error for including tensor ukernels (#21780) by Jorn Tuyls · 14 days ago
  71. 33e2146 [Codegen] Add corner case for SwapExtractWithCollapsePattern (#21773) by Vivian Zhang · 14 days ago
  72. f1e9219 [DispatchCreation] Fix trailing unit dims case for collapse of expand folding (#21677) by Daniel Garvey · 14 days ago
  73. e6fb1e1 [Codegen] PV and QK matmul's must have same acc layout (#21729) by James Newling · 2 weeks ago
  74. 9bb1a2b [ROCM] Port mlir ukernels to ukernel descriptor lowering flow (#21683) by Jorn Tuyls · 2 weeks ago
  75. 46de78a [DT] Graduate data-tiling fusion from experimental flag to binding option. (#21745) by Han-Chung Wang · 3 weeks ago
  76. 80de240 Adding IREE_HAL_COMMAND_BUFFER_MODE_UNRETAINED flag. (#21755) by Ben Vanik · 3 weeks ago
  77. 657e2de [RISCV] Remove unused cmake variables. (#21746) by Han-Kuan Chen · 3 weeks ago
  78. 914868c Temporarily disable the circular buffer for parameter uploads. (#21758) by Andrew Woloszyn · 3 weeks ago
  79. b15a081 [NFC] Moving iree_hal_amdgpu_bitmap to iree/base/internal/. (#21666) by Ben Vanik · 3 weeks ago
  80. 2c378d0 [Codegen] Add matmul and batched matmul to list of ops to generalize (#21720) by James Newling · 3 weeks ago
  81. 1993c4f [Dispatch] CollapseDims for extract_slice and scf.forall (#21708) by Ian Wood · 3 weeks ago
  82. d7fc56e [ConstEval] Do not jit parameterized flow.tensor.constants (#21748) by Kunwar Grover · 3 weeks ago
  83. 3df650b Fixing flake-y host call CTS test. by Ben Vanik · 3 weeks ago
  84. 2758226 Fixing merge conflict from #21619 + #21653. (#21751) by Ben Vanik · 3 weeks ago
  85. 7755c30 Adding iree_hal_device_queue_host_call and emulation. (#21653) by Ben Vanik · 3 weeks ago
  86. 53daa95 Adding semaphore creation and wait flags for controlling behavior. (#21619) by Ben Vanik · 3 weeks ago
  87. b0895d6 Integrate LLVM at bfab8085af878dbcafaf5dfac4e34dc17a20971c (#21747) by Kunwar Grover · 3 weeks ago
  88. 9a9dfe8 Integrate LLVM at llvm/llvm-project@c65c0e87fc73 (#21744) by Han-Chung Wang · 3 weeks ago
  89. e5b0780 Apply UnsignedWhenEquivalent at the ModuleOp level. (#21743) by Erick Ochoa Lopez · 3 weeks ago
  90. 8d12c30 [CPU] Improve TileRootAndFuseProducerConsumer pass and deprecate TileAndFuse pass. (#21674) by Han-Chung Wang · 3 weeks ago
  91. 337c8aa [Codegen] Improve early bufferized padding codegen (#21694) by Max191 · 3 weeks ago
  92. 1f7762d Remove myself from samples/ CODEOWNERS. (#21726) by Scott Todd · 3 weeks ago
  93. ad1a8f0 Integrate llvm/llvm-project@6fc1deb8b749 (#21732) by Han-Chung Wang · 3 weeks ago
  94. c5dcac2 Bump sarisia/actions-status-discord from 1.15.3 to 1.15.4 in the github-actions group (#21730) by dependabot[bot] · 3 weeks ago
  95. 2aef563 Fix SmallVector conversion error with gcc (#21725) by Jorn Tuyls · 3 weeks ago
  96. 022a3c2 [ROCM] Readd SpecializeExports pass (#21727) by Quinn Dawkins · 3 weeks ago
  97. e2a75af [DT] Drop the data-tiling hint after encodings are set. (#21724) by Han-Chung Wang · 3 weeks ago
  98. 80af7ac Revert "Move windows builds to experimental to unblock release packages." (#21723) by Scott Todd · 3 weeks ago
  99. adf3eb9 Drop needless template parameters from patterns. NFC. (#21721) by Jakub Kuderski · 3 weeks ago
  100. 3df06b7 [Integrate] Drop LLVM revert of "Remove matmul_transpose variants" (#21344) by Han-Chung Wang · 3 weeks ago