- 8651777 [gpu] Add basic heuristics for better reduction occupancy (#15120) by Lei Zhang · 1 year, 6 months ago
- 9e9aff0 [PkgCI] Add llama2_7b_i4 recipe for correctness testing on cuda (#15113) by Kunwar Grover · 1 year, 6 months ago
- 58b5670 [rocm] Enable the ROCM compiler target backend by default. (#15111) by Stella Laurenzo · 1 year, 6 months ago
- c64b31f [PkgCI] Add tqdm bar while downloading artifacts (#15112) by Kunwar Grover · 1 year, 6 months ago
- 98d4f18 [PkgCI] Add llama2 recipe for NVIDIA A100 (#15093) by Kunwar Grover · 1 year, 6 months ago
- 480a4c6 [pjrt] Add ROCM and Vulkan backends. (#15106) by Stella Laurenzo · 1 year, 6 months ago
- 00cbed6 Bump torch-mlir to 6f81ad72938deb56c6d43bbc01388c1f8f1253c1 (#15110) by Quinn Dawkins · 1 year, 6 months ago
- 89f99bc [rocm] Add target dependency to ROCM bc file copies. (#15109) by Stella Laurenzo · 1 year, 6 months ago
- 3add457 Adding iree_io_file_handle_t placeholder. (#15101) by Ben Vanik · 1 year, 6 months ago
- 90a0225 [Torch] Assume strict symbolic shapes (#15107) by Quinn Dawkins · 1 year, 6 months ago
- 24d80e1 Preserve lowering config attribute during rematerialization. (#15103) by MaheshRavishankar · 1 year, 6 months ago
- d76a104 Restore PJRT to minimally functional after break with XLA. (#15105) by Stella Laurenzo · 1 year, 6 months ago
- 39b3b24 [python] Print error code in CompilerToolError (#15099) by Lei Zhang · 1 year, 6 months ago
- d413a7e [ROCM] Build rocm device bc files from source and installation setup. (#15096) by Stanley Winata · 1 year, 6 months ago
- bb51f6f [Utils] Fix bug in sub-byte check for byte alignment (#14942) by Kunwar Grover · 1 year, 7 months ago
- e50538c [LLVMGPU] Lower descriptor type to global address space (#15047) by Quinn Dawkins · 1 year, 7 months ago
- ad64ecc [experimental][ROCM] Add shared memory support on ROCM RT and Target. (#15097) by Stanley Winata · 1 year, 7 months ago
- b3e5a43 Add python binding for device profile flushing (#15095) by Quinn Dawkins · 1 year, 7 months ago
- 017c826 Re-enable RISC-V benchmark targets (#15092) by Jerry Wu · 1 year, 7 months ago
- 3389845 Update CNAME to redirect docs too (#15046) by Jacques Pienaar · 1 year, 7 months ago
- 9dc8013 Revert "[Flow] Allow CollapseDimensions pass to fold reduction dimensions as well (#14656) (#15090) by MaheshRavishankar · 1 year, 7 months ago
- 868b7fb [Reducer] Improve simplification for reducing strategies (#15080) by Kunwar Grover · 1 year, 7 months ago
- 18c7427 Log details when dumping no executable benchmarks. (#15068) by Scott Todd · 1 year, 7 months ago
- 8c45e71 Avoid using ccache for 'runtime' builds. (#15086) by Scott Todd · 1 year, 7 months ago
- caaf7dd Update Mac build docs regarding thin archives (#15085) by bjacob · 1 year, 7 months ago
- ad1a82e [docs][NFC] Update doc for using ctest_all.sh (#15071) by tyb0807 · 1 year, 7 months ago
- be3cd1d suppress warnings on XCode 15 (#15066) by bjacob · 1 year, 7 months ago
- fee7f6a [LLVMExternalProjects] Drop the IREE LinalgTransformDialect (#15067) by Quinn Dawkins · 1 year, 7 months ago
- ada5473 [Reducer] Add delta pass to reduce optimization barriers (#15081) by Kunwar Grover · 1 year, 7 months ago
- 5d479ea [Reducer] Verify module before running the interesting script (#15082) by Kunwar Grover · 1 year, 7 months ago
- 4e6d841 Integrate LLVM 2023-09-27 (#15048) by Stella Laurenzo · 1 year, 7 months ago
- 83df8c4 [ConstEval] Teach global hoisting to build a dot graph of its analysis (#15075) by Kunwar Grover · 1 year, 7 months ago
- c753168 [ConstEval] Do not hoist if the hoisted value increases size significantly (#15065) by Kunwar Grover · 1 year, 7 months ago
- 0af63ad Support scalable tile sizes in #iree_codegen.lowering_config (#15033) by Benjamin Maxwell · 1 year, 7 months ago
- 113f9d2 [LLVMGPU] Convert maximumf/minimumf to max/min for ROCM (#15069) by nirvedhmeshram · 1 year, 7 months ago
- 14ce232 [CPU] Improve computation of vector sizes (#14935) by Diego Caballero · 1 year, 7 months ago
- a53153a [LinalgExt][Bufferization] Clean-up bufferization pass for LinalgExt (#15040) by Abhishek Varma · 1 year, 7 months ago
- 1ba5e37 [LLVMGPU] Enable WarpReduction on ROCM + Let matvec use Warp Reduce. (#15034) by Stanley Winata · 1 year, 7 months ago
- 750784d Fix torch-mlir commit (#15062) by Kunwar Grover · 1 year, 7 months ago
- 60a2566 Add iree-reduce tool (#14892) by Kunwar Grover · 1 year, 7 months ago
- 1a63564 Refactor IREECodegenAttrs to use typed array parameters (#15032) by Benjamin Maxwell · 1 year, 7 months ago
- 15ace7e [Flow] Allow CollapseDimensions pass to fold reduction dimensions as well (#14656) by MaheshRavishankar · 1 year, 7 months ago
- 699b34c [vulkan] Add e2e coop matrix f16 matmul test (#15058) by Jakub Kuderski · 1 year, 7 months ago
- 1b5eec9 [vulkan] Enable coop matrix features when available (#15056) by Jakub Kuderski · 1 year, 7 months ago
- fdb863d NFC: Reserve larger ranges for different targets' pipelines (#15057) by Lei Zhang · 1 year, 7 months ago
- 1166f12 [vulkan][spirv] Update submodules and deps (#15052) by Jakub Kuderski · 1 year, 7 months ago
- f8c7e49 Correct input type typo in error message (#15051) by Julian Walker · 1 year, 7 months ago
- ffd5ad4 [Flow] Fix FuseDequantMatmul pass for multiple uses of dequant (#15045) by Kunwar Grover · 1 year, 7 months ago
- ef280a4 Integrate llvm 20230926 (#15043) by Stella Laurenzo · 1 year, 7 months ago
- 10d626c bump torch-mlir to b83d9f2cd95119725ae0c55802fcf312ea82256e (#15039) by Daniel Garvey · 1 year, 7 months ago
- f993d7c [spirv][vulkan] Switch from NV to KHR coop matrix extension (#15026) by Jakub Kuderski · 1 year, 7 months ago
- 2e3ce2e Fix test assuming deterministic constant ordering. (#15044) by Scott Todd · 1 year, 7 months ago
- 3d4c1ac Remove notice about missing Windows Python packages. (#14580) by Scott Todd · 1 year, 7 months ago
- 85319c8 Fix cuda fetch script (#15011) by Jakub Kuderski · 1 year, 7 months ago
- de51b22 Integrate llvm 20230922 (#15020) by Stella Laurenzo · 1 year, 7 months ago
- e77c78e Add a dedicated 'torch' input pipeline. (#15037) by Stella Laurenzo · 1 year, 7 months ago
- 3725acd [Flow] Fixed dropped dim computations to handle some ambiguous cases. (#15035) by MaheshRavishankar · 1 year, 7 months ago
- 465a214 Merge branch 'main' of github.com:openxla/iree into merge_pjrt_plugin by Stella Laurenzo · 1 year, 7 months ago
- a0c5673 [SPIRV] Handle sub-byte loads with EmulateNarrowTypes (#15029) by Quinn Dawkins · 1 year, 7 months ago
- 944879f [LLVMCPU] Revert KernelDispatch changes from #15018 (#15024) by Quinn Dawkins · 1 year, 7 months ago
- 4ca3b08 Subtree merge openxla-pjrt-plugin into integrations/pjrt. by Stella Laurenzo · 1 year, 7 months ago
- 63ae50c Set root dir for in-tree location. by Stella Laurenzo · 1 year, 7 months ago
- 85fd4f0 Remove items only needed in a standalone repo. by Stella Laurenzo · 1 year, 7 months ago
- a56cc00 Drop convperf and mmperf Dockerfiles and workflows. (#14903) by Scott Todd · 1 year, 7 months ago
- f1a4e74 Update build instructions. by Stella Laurenzo · 1 year, 7 months ago
- d8c197b Port CPU setup changes to CUDA. by Stella Laurenzo · 1 year, 7 months ago
- d79f88e Modularize setup.py. by Stella Laurenzo · 1 year, 7 months ago
- 3feab59 Adapt CPU plugin packaging to new structure. by Stella Laurenzo · 1 year, 7 months ago
- cf26c22 Note branch point for PJRT headers. by Stella Laurenzo · 1 year, 7 months ago
- 069bcda Remove elements needed only as a standalone project. by Stella Laurenzo · 1 year, 7 months ago
- be1e86d Remove features dependent on unsupportable dependencies and implement CMake build. by Stella Laurenzo · 1 year, 7 months ago
- 1d21e9a Initial re-organization in preparation for splitting. by Stella Laurenzo · 1 year, 7 months ago
- 2571cf1 Update nightly dependencies (#315) by iree-github-actions-bot · 1 year, 7 months ago
- 2b9ade5 [spirv] Fix transfer op crash when vectorizing memref (#15025) by Lei Zhang · 1 year, 7 months ago
- fc3292f Fix checking for pending PRs by Jacques Pienaar · 1 year, 7 months ago
- 3bf20db Fixing hal_inline file IO for variable initialization. (#15022) by Ben Vanik · 1 year, 7 months ago
- 875fde3 [GlobalOptimization] Fold unit extent dims before fusing dequant with matmul (#15018) by Quinn Dawkins · 1 year, 7 months ago
- 4cb46cb Add Tensor Tracing to inline-dynamic lowerings (#14983) by Rob Suderman · 1 year, 7 months ago
- fc1ff49 [ConstEval] Block hoisting of child uses inside dispatch. (#15014) by Stanley Winata · 1 year, 7 months ago
- 588748e [metal] Disable crashing reduce_window tests for now (#15013) by Lei Zhang · 1 year, 7 months ago
- 25a0b6c [SPIRV] Break SPIRVVectorize pass into GenericVectorization and SPIRVVectorLowering passes. (#15009) by Quinn Dawkins · 1 year, 7 months ago
- 3cb5f1a Revert commit https://github.com/openxla/iree/pull/15001/commits/1686c138b2262fccb56c7643546712b8cd06ac19 that made it in the integrate. (#15010) by MaheshRavishankar · 1 year, 7 months ago
- 44722fb Create CNAME (#15008) by Jacques Pienaar · 1 year, 7 months ago
- 8bb3a84 Integrate llvm 20230920 (#15001) by Stella Laurenzo · 1 year, 7 months ago
- a79da4f [Flow] Fix trailing constant accesses in slice raising patterns (#15002) by Quinn Dawkins · 1 year, 7 months ago
- 9be0a5c Define some more unimplemented instances by Jacques Pienaar · 1 year, 7 months ago
- bd05a55 Move SCFToCFG earlier in pipeline (#15006) by Rob Suderman · 1 year, 7 months ago
- 23e0ac3 Set correct tile sizes on root ops with pack/unpack ops (#15005) by Jerry Wu · 1 year, 7 months ago
- 8ac3a1a Optimized narrow mmt4d tile functions (for small M0) (#14995) by bjacob · 1 year, 7 months ago
- 5017492 Update Github runner to v2.309.0 (#15004) by Jerry Wu · 1 year, 7 months ago
- 6710ff6 add decompose complex ops pass to torch_to_iree (#14992) by Daniel Garvey · 1 year, 7 months ago
- cc7d6e9 Fixing cpuinfo Mac build break from #14969. by Ben Vanik · 1 year, 7 months ago
- 1ef2958 Integrate llvm-project to 5cf714bb2f75552b10e1eb62fd07aec4b6033881. (#15000) by Stella Laurenzo · 1 year, 7 months ago
- fb9e1b6 Adding `--iree-vulkan-experimental-indirect-bindings=true` flag. (#14977) by Ben Vanik · 1 year, 7 months ago
- 0f4dd73 Adding `--task_topology_cpu_ids=` flag. (#14969) by Ben Vanik · 1 year, 7 months ago
- 04259d0 Integrate llvm-project at f66cd9e9556a53142a26a5c21a72e21f1579217c (#14980) by Stella Laurenzo · 1 year, 7 months ago
- d24d024 Changing IREE_CPU_DATA* enums to free floating constants. (#14991) by Ben Vanik · 1 year, 7 months ago
- 8910e11 [CPU] Add mask cleanup patterns to vectorizer pass (#14988) by Diego Caballero · 1 year, 7 months ago
- 2f72249 [stablehlo] Add matcher for ApproxTopK custom call (#14899) by Rob Suderman · 1 year, 7 months ago
- ea72e7d Pin to old docker for build temporarily (#314) by Jacques Pienaar · 1 year, 7 months ago