- a96dd9e New mmt4d builtin with general code and some assembly kernels (i8mm, dotprod) (#10339) by bjacob · 2 years, 7 months ago
- 6fd42a4 fix `minimum_duration_ns` in `iree_benchmark_register` (#10424) by bjacob · 2 years, 7 months ago
- 1c18688 Add workload flag for dispatch region formation (#10422) by Matthias Springer · 2 years, 7 months ago
- eaab515 Create DispatchWorkgroupsOps via DispatchRegionOps (#10400) by Matthias Springer · 2 years, 7 months ago
- 3a267bb [NFC] Move getLoopRanges to RegionOpUtils (#10420) by Matthias Springer · 2 years, 7 months ago
- 315d99c [NFC] Move ConvertToFlowPass into separate file (#10419) by Matthias Springer · 2 years, 7 months ago
- 21bcf35 [NFC] Move orderOperations to RegionOpUtils (#10417) by Matthias Springer · 2 years, 7 months ago
- 5c55fc4 Drop revert of "Remove LinalgStrategyEnablePass". (#10416) by MaheshRavishankar · 2 years, 7 months ago
- f29c86a Copy from upstream and drop reliance on upstream CodegenStrategy. (#10411) by Nicolas Vasilache · 2 years, 7 months ago
- 3300167 Make sure GCS dir changes with run attempt (#10410) by Geoffrey Martin-Noble · 2 years, 7 months ago
- 57262e6 Limit unrolling for transpose operations on RISC-V (#9913) by Diego Caballero · 2 years, 7 months ago
- 6502ccd Integrate llvm-project at https://github.com/llvm/llvm-project/commit/8ae9cf550b60 (#10387) by MaheshRavishankar · 2 years, 7 months ago
- 4c13cae Update Colab notebooks to use --output-format=mlir-ir. (#10409) by Scott Todd · 2 years, 7 months ago
- 3eb488f Add missing build_tf_integrations job to summary dependencies (#10395) by Geoffrey Martin-Noble · 2 years, 7 months ago
- 9782061 Drop PDL usage where unnecessary as well as tests that have been upst… (#10404) by Nicolas Vasilache · 2 years, 7 months ago
- d4f5305 Add a fused softmax strategy on GPU (#10383) by Nicolas Vasilache · 2 years, 7 months ago
- 3b45fe3 [Flow][NFC] Move cloning/moving helpers to RegionOpUtils (#10402) by Matthias Springer · 2 years, 7 months ago
- 0c9b299 [Flow] ConvertRegionToWorkgroups: Add workload builder callback (#10403) by Matthias Springer · 2 years, 7 months ago
- 7e1c336 Updates to IREE vs TFLite benchmarks (#10381) by mariecwhite · 2 years, 7 months ago
- a963df0 [NFC] Use OpBuilder instead of Rewriter for workload computation (#10401) by Matthias Springer · 2 years, 7 months ago
- d63bbf7 Split clone_preceding_op_into_dispatch transform op (#10396) by Matthias Springer · 2 years, 7 months ago
- c4ad40c Disable flaky test. (#10399) by MaheshRavishankar · 2 years, 7 months ago
- 17b33fc Use `--mount` instead of `--volume` in `docker run` (#10392) by Geoffrey Martin-Noble · 2 years, 7 months ago
- 9db55e7 Script the creation of a new VM image (#10390) by Geoffrey Martin-Noble · 2 years, 7 months ago
- 01ac12a Add a script and lint check for maximum path lengths. (#10367) by Scott Todd · 2 years, 7 months ago
- f483a1f Generalize TransposeSharedMem Pipeline to Fusion with Other ElementWise (#10360) by Kojo Acquah · 2 years, 7 months ago
- e995fd3 Copy patterns to iree-dialects and drop reliance on their existence in MLIR core(#10385) by Nicolas Vasilache · 2 years, 7 months ago
- 6f97d06 Fix the TF import rule generator (#10306) by Jerry Wu · 2 years, 7 months ago
- 8c511b3 Update TF code from ValueOrDie() to value(). (#10384) by Scott Todd · 2 years, 7 months ago
- 2185211 Adding subspan offset/length to hal.buffer_view.create. (#10378) by Ben Vanik · 2 years, 7 months ago
- d377acf Starting support for resource attribute serialization. (#10377) by Ben Vanik · 2 years, 7 months ago
- 66036b4 [spirv] Add fast math support for WebGPU (#10352) by Lei Zhang · 2 years, 7 months ago
- c17fc52 VMVX e2e matmul tests with target CPU features variants for ARM (#10366) by bjacob · 2 years, 7 months ago
- b86ef09 Add RISC-V 64-bit Linux unit tests (#10346) by CindyLiu · 2 years, 7 months ago
- b04e36e Update import tooling to write to bytecode-ir format (#10375) by Rob Suderman · 2 years, 7 months ago
- 9ef8abf Hiding semaphores from the compiler. (#10374) by Ben Vanik · 2 years, 7 months ago
- 1abccdf Adding stream.async.load/store folders for bitcast ops. (#10373) by Ben Vanik · 2 years, 7 months ago
- 128be3f Generate CMake rules to compile IREE modules (#10160) by Jerry Wu · 2 years, 7 months ago
- d1e7956 [spirv] Cherry-pick LLVM/MLIR commits for fast math mode (#10372) by Lei Zhang · 2 years, 7 months ago
- 7a435f0 Move iree/compiler/Dialect/Modules/ to iree/compiler/Modules/. (#10370) by Scott Todd · 2 years, 7 months ago
- 8a55b02 Remove old kokoro crosscompile scripts (#10355) by CindyLiu · 2 years, 7 months ago
- a348117 Rework GeneralDot lowerings for mul/general_dot conversion (#10262) by Rob Suderman · 2 years, 7 months ago
- 9ccf0d5 Fix typos in VectorizeConv.cpp comments. (#10365) by lorenzo chelini · 2 years, 7 months ago
- 3186975 [NFC] Switch to use upstream bitEnumContains methods. (#10361) by Han-Chung Wang · 2 years, 7 months ago
- 9121254 Add support for non-1 batch cases in img2col convolution pass (#10343) by Lei Zhang · 2 years, 7 months ago
- 8919460 [NFC] Move tensor.pad related passes from SPIRV/ to Common/ (#10357) by Han-Chung Wang · 2 years, 7 months ago
- 625e345 Integrate llvm-project at 55e1441f7b5d and bump dependencies (#10351) by Han-Chung Wang · 2 years, 7 months ago
- 83b0103 NFC: Rename `DispatchDefaultWorkgroupCountOp` to `DispatchWorkgroupCountFromDagRootOp` (#10344) by MaheshRavishankar · 2 years, 7 months ago
- e64f95c [spirv] Relax batch matmul recognization for more cases (#10349) by Lei Zhang · 2 years, 7 months ago
- 5765e4c Fix and enable broken tests (#10347) by Matthias Springer · 2 years, 7 months ago
- 7cb032c NFC: various clean ups to ConvertConv2DToImg2ColPass (#10340) by Lei Zhang · 2 years, 7 months ago
- f010f78 Refactor mhlo.scatter canonicalization to simplify linalg_ext lowering (#10328) by Rob Suderman · 2 years, 7 months ago
- 0c4885c Upgrade GitHub runner to v2.296.2 (#10338) by Geoffrey Martin-Noble · 2 years, 7 months ago
- 342b098 Replace MiniLM benchmark with seqlen=128 version. (#10335) by Jerry Wu · 2 years, 7 months ago
- 2ef36af Create a separate workflow for benchmark pipeline (#10333) by Jerry Wu · 2 years, 7 months ago
- 90bed7e Integrate llvm-project at 4d700ffe67be and bump dependencies. (#10336) by Han-Chung Wang · 2 years, 7 months ago
- 134523a [spirv] Recognize (batch) matmul-like linalg.generic (#10213) by Lei Zhang · 2 years, 7 months ago
- d3d4ed0 Add RISC-V cross compile/run support for iree_check_test (#10330) by CindyLiu · 2 years, 7 months ago
- f8803db Add dry-run and expanded filtering options to update script (#10321) by Geoffrey Martin-Noble · 2 years, 7 months ago
- 5aff56a [spirv] Add pipeline for performing subgroup reduction (#10286) by Lei Zhang · 2 years, 7 months ago
- c469de5 Place Docker workdir in consistent location (#10327) by Geoffrey Martin-Noble · 2 years, 7 months ago
- fbfafb4 [spirv] Add pipeline for performing warp reduction by Lei Zhang · 2 years, 8 months ago
- a4a07bc NFC: move LLVMGPUVectorization pass to Common/ by Lei Zhang · 2 years, 8 months ago
- 5d86f1c NFC: move VectorReduceToGPU pass to Common/ by Lei Zhang · 2 years, 8 months ago
- 1fb572d Add Transpose Pipeline Test for LLVMGPULowerExecutableTargetPass (#10277) by Kojo Acquah · 2 years, 7 months ago
- ed28683 Add target and result to transform.iree.bufferize (#10332) by Nicolas Vasilache · 2 years, 7 months ago
- d750bad Integrate llvm-project at 72142fbac496 and bump dependencies. (#10322) by Han-Chung Wang · 2 years, 7 months ago
- ec97010 Delete Kokoro directory (#10325) by Geoffrey Martin-Noble · 2 years, 7 months ago
- 63e89c1 Update reduction.mlir (#10324) by Nicolas Vasilache · 2 years, 7 months ago
- 08b5fbb Exclude more tests from ASan test configuration. (#10281) by Scott Todd · 2 years, 7 months ago
- 64536ac rename libdevice header name (#10320) by Thomas · 2 years, 7 months ago
- da96ec9 Install GitHub runner on startup in a tmpfs (#10318) by Geoffrey Martin-Noble · 2 years, 7 months ago
- 2825165 Integrate llvm-project at 2d52c6bfae80 and bump dependencies. (#10304) by Han-Chung Wang · 2 years, 7 months ago
- e83c8f2 Adapt distribution tiling params of mmt-4D to use default heuristics. (#10264) by Han-Chung Wang · 2 years, 7 months ago
- e912478 Create Python script for updating instance groups (#10310) by Geoffrey Martin-Noble · 2 years, 7 months ago
- 5bb4c59 Specify runner environment that jobs should target (#10302) by Geoffrey Martin-Noble · 2 years, 7 months ago
- eb855e8 Remove CUDA target Bazel file as build doesn't work with Bazel (#10317) by Thomas · 2 years, 7 months ago
- 82a75dc Add dcaballe to LLVM CPU CODEOWNERS (#10316) by Diego Caballero · 2 years, 7 months ago
- a8e4c38 Add softmax transform dialect spec and e2e test (#10274) by Nicolas Vasilache · 2 years, 7 months ago
- 2316736 Disable nounroll workaround for CUDA 11.3 (#10307) by Thomas · 2 years, 7 months ago
- 27ded29 Remove tooling/cpu_features, switch to base/internal/cpu. (#10303) by bjacob · 2 years, 7 months ago
- 82518af Add labels identifying runner release and environment (#10299) by Geoffrey Martin-Noble · 2 years, 7 months ago
- c71a9fe Fix shell highlights for sample iree-compile commands. (#10301) by Scott Todd · 2 years, 7 months ago
- 1012f88 Add `--iree-input-type=` flag to sample commands in docs. (#10295) by Scott Todd · 2 years, 7 months ago
- 223d02f [spirv] Cherry-pick more upstream commits for vector reduction (#10285) by Lei Zhang · 2 years, 8 months ago
- 915977b [vulkan] Update various targets to match current status (#10283) by Lei Zhang · 2 years, 8 months ago
- ea7fdab [spirv] Cherry-pick llvm/llvm-project@1b5d97dd (#10284) by Lei Zhang · 2 years, 8 months ago
- 55fc029 [LLVMGPU] Add shared memory swizzle transformation (#9943) by Thomas · 2 years, 8 months ago
- 5c5f678 Reworking CPU data queries and access for vmvx/builtins/compiler-produced code. (#10270) by Ben Vanik · 2 years, 8 months ago
- 63552ae Integrate llvm-project and bump dependencies (#10282) by Okwan Kwon · 2 years, 8 months ago
- 853dd6e Adding cpu_data.h and CPU data queries into iree/base/internal/cpu.h. by Ben Vanik · 2 years, 8 months ago
- 9336348 Initial mmt4d builtins - reference code move out of vmvx/module.c (#10272) by bjacob · 2 years, 8 months ago
- 18d2fb9 Refine web test runner style (text wrapping, console colors, etc.). (#10273) by Scott Todd · 2 years, 8 months ago
- db6b687 Bump llvm 20220901 (#10271) by Okwan Kwon · 2 years, 8 months ago
- cbfb816 Limit vector level tiling sizes aggressively for elementwise dispatches. (#10260) by Han-Chung Wang · 2 years, 8 months ago
- 73ac0be Add experimental path to use transformer dialect only in the codegen (#10168) by Thomas · 2 years, 8 months ago
- 2816df4 Specialize tiling sizes for N=2 and M=2 cases in matmul_to_mmt4d transform. (#10215) by Han-Chung Wang · 2 years, 8 months ago
- 4b4b854 Updating VMVX to use per-worker contexts and support workgroup state. by Ben Vanik · 2 years, 8 months ago
- 229d831 Fix cmake `iree_*_module` condition check to support cross-compile (#10266) by CindyLiu · 2 years, 8 months ago
- 2b22644 Add CI job to build benchmark tools (#10233) by Jerry Wu · 2 years, 8 months ago