- 8e68e98 Add full int8 ViT benchmark by mariecwhite · 1 year, 6 months ago
- 6bbdb72 [cuda] Mark event related APIs as unimplemented (#15382) by Lei Zhang · 1 year, 6 months ago
- 8d7dc80 Stop trying to use builtin `_Float16` (#15388) by bjacob · 1 year, 6 months ago
- 6b93f11 bump torch-mlir (#15389) by Daniel Garvey · 1 year, 6 months ago
- 668c020 Cast tensor.empty type to TypeConverter's type during materialization. (#15375) by Han-Chung Wang · 1 year, 6 months ago
- 77a8c55 [NFC] Move CPU materialize_encoding tests to Common/CPU/test (#15376) by Han-Chung Wang · 1 year, 6 months ago
- d1d63c3 Add riscv vector extension in cpu feature using hwcap (#15306) by Yun Hsiang · 1 year, 6 months ago
- 85f4006 [shlo] Misc fixes exposed by jax test. (#15379) by Jacques Pienaar · 1 year, 6 months ago
- fd9cd2f Fix some minspec/optional feature bitrot. (#15378) by Stella Laurenzo · 1 year, 6 months ago
- 332ac35 Drop AMDGPU in-tree build of device libraries. (#15374) by Stella Laurenzo · 1 year, 6 months ago
- fcdddcb Bump ARM64 runner image (#15366) by Jerry Wu · 1 year, 6 months ago
- 7c58c58 Use `c2-standard-16` VM to run x86_64 e2e benchmark tests (#15361) by Jerry Wu · 1 year, 6 months ago
- 4a20b91 Drop vulkan-spirv test cases from modules/check/test/*. (#15356) by Scott Todd · 1 year, 6 months ago
- 3d1d8c8 ukernels: sub-8-bit support, types `s16 * u4 -> s32` and `s16 * s16 -> s32` (#15343) by bjacob · 1 year, 6 months ago
- c0525ad Update the usage of the transform dialect interpreter (#15340) by Nicolas Vasilache · 1 year, 6 months ago
- 3112576 Adding gcloud CLI support to arm64 runner (#15308) by Fredrik Knutsson · 1 year, 6 months ago
- 4d06d20 [pjrt] Towards more mechanical stub generation. (#15363) by Jacques Pienaar · 1 year, 6 months ago
- 18b3dd7 LLVM integrate integrate-llvm-20231030 (#15351) by Stella Laurenzo · 1 year, 6 months ago
- bd72855 use getTypeBitWidth() to get the element type's bit width (#15360) by Okwan Kwon · 1 year, 6 months ago
- 03d655a Fix size calculation in the tensor.empty materialization pattern. (#15359) by Han-Chung Wang · 1 year, 6 months ago
- 9f7d6d4 Add falcon benchmarks by mariecwhite · 1 year, 6 months ago
- af171c5 Exclude executable files in root .gitignore. (#15266) by Scott Todd · 1 year, 6 months ago
- 256fe4f Add "torch" as an `InputType` in `iree/compiler/tools/core.py`. (#15358) by Scott Todd · 1 year, 6 months ago
- 5223596 [cuda] Support building node DAG in graph command buffer (#14857) by Eugene Zhulenev · 1 year, 6 months ago
- 05928c5 Enable the 'clang' project when building the ROCm target. (#15346) by Scott Todd · 1 year, 6 months ago
- 5c9556c Add a pass to materialize encodings into nop. (#15312) by Han-Chung Wang · 1 year, 6 months ago
- 092a74d [pjrt] Add primitive jax2tf tests (#15341) by Jacques Pienaar · 1 year, 6 months ago
- 2706526 [ROCM] add device path and use it to setup device (#15234) by nirvedhmeshram · 1 year, 6 months ago
- 7b92a6d [cuda] Avoid sorting when composing kernel arguments (#15325) by Lei Zhang · 1 year, 6 months ago
- 546e372 Add fallback for undo-ing encodings. (#15302) by Han-Chung Wang · 1 year, 6 months ago
- ada35b3 Disable folding casting ops into contraction ops by default. (#15342) by Han-Chung Wang · 1 year, 6 months ago
- 5350a37 Add CI configuration recipes with screenshots to contributing.md. (#15331) by Scott Todd · 1 year, 6 months ago
- f210990 Fix unsigned integers for collapse/expand (#15317) by Rob Suderman · 1 year, 6 months ago
- 5f86cd7 Implement transposition / broadcast on host transfers (#15300) by Rob Suderman · 1 year, 6 months ago
- 41e7289 ukernels: stop abusing signless as signed (#15338) by bjacob · 1 year, 6 months ago
- aeb38b0 Fix Artifact Saver for long path names (#15316) by Rob Suderman · 1 year, 6 months ago
- 8223344 Update RISCV-QEMU emulator to 8.2.1 (#15320) by CindyLiu · 1 year, 6 months ago
- cfa2370 Support intput/expected output in benchmark definitions (#15327) by Jerry Wu · 1 year, 6 months ago
- 7845e69 [Vulkan] Make target triples/envs options lists (#15309) by Quinn Dawkins · 1 year, 6 months ago
- d0c0278 [Codegen] Improve For induction variable packing pattern (#15333) by Quinn Dawkins · 1 year, 6 months ago
- 6b8954b [NFC-ish] Split out kernel config into <BACKEND>SelectLoweringStrategy (#15321) by Quinn Dawkins · 1 year, 6 months ago
- 960e8cb ExpandVectors: support the new `linalg.batch_vecmat` (#15310) by bjacob · 1 year, 6 months ago
- 11debb4 bump LLVM to acc6f3e9c1af6c7445aae6f10d4b016ac84112d3 (#15296) by bjacob · 1 year, 6 months ago
- e329737 Recommend `-G Ninja` in all in-tree documentation. (#15328) by Scott Todd · 1 year, 6 months ago
- 9cc729f [AArch64][SVE] Add e2e tests for small and large matmuls (#15292) by Benjamin Maxwell · 1 year, 6 months ago
- 39afb04 [Winograd] Adapt winograd for tiling + implement tiling/decompose (#15250) by Abhishek Varma · 1 year, 6 months ago
- 44c9346 Patch some Windows C code that breaks under gcc/clang. (#15326) by Scott Todd · 1 year, 6 months ago
- f29895e Small bugfix in EmplaceAllocations (#15261) by Daniel Garvey · 1 year, 6 months ago
- 222bcf2 [Flow] Add patterns to lower tensor.reshape to flow.tensor.reshape (#15226) by saienduri · 1 year, 6 months ago
- c878412 Fix benchmark dispatch repetition for nested calls (#15318) by Quinn Dawkins · 1 year, 6 months ago
- b65ff2f Bump gtest version (#15301) by Jacques Pienaar · 1 year, 6 months ago
- 41c2b6d Delete event during deferred callback (#15303) by Rob Suderman · 1 year, 6 months ago
- 5b43ed0 Add more icons to website headings and content tabs. (#15311) by Scott Todd · 1 year, 6 months ago
- 8103349 Using configured LLVMTarget for linking. (#15304) by Han-Chung Wang · 1 year, 6 months ago
- 5610d8c [VM] Fix lowering of case values in vm.br_table (#15305) by Quinn Dawkins · 1 year, 6 months ago
- e9f4b1e [Flow] Add TensorBitCastOp (#15260) by Quinn Dawkins · 1 year, 6 months ago
- c4be76f Tweak configure_ci.py regex patterns. (#15298) by Scott Todd · 1 year, 6 months ago
- 117e82d Add pass to expand all vectors in vecmat/matvec operations into matrices to enable tiling (#15273) by NatashaKnk · 1 year, 6 months ago
- b207d4d Ensure lowering configs match tiled ops (#15178) by Jerry Wu · 1 year, 6 months ago
- d4748be Add PyTorch "advanced AOT API" documentation. (#15297) by Scott Todd · 1 year, 6 months ago
- dbc5112 Add output verification to linux benchmark tool (#15281) by Jerry Wu · 1 year, 6 months ago
- 1206e60 Refactor HostToDevice Transfer to seperate 0-dim and splat cases (#15285) by Rob Suderman · 1 year, 6 months ago
- 85efc86 [CPU][SVE] Enforce > 1 pow2 sizes when materializing scalable matmul lowering_configs (#15276) by Benjamin Maxwell · 1 year, 6 months ago
- 1b177e95 Making flow.dispatch/stream.async.dispatch take multiple symbols. (#15295) by Ben Vanik · 1 year, 6 months ago
- a3a64d2 Adding selection condition to hal.executable.variant. (#15284) by Ben Vanik · 1 year, 6 months ago
- 466e812 [Codegen] Add ability to specify transform dialect libraries (#14788) by Quinn Dawkins · 1 year, 6 months ago
- 97a1394 Adding `vm.br_table` op. (#15286) by Ben Vanik · 1 year, 6 months ago
- 446f96e [metal] Enable end-to-end TOSA / StableHLO op tests (#15289) by Lei Zhang · 1 year, 6 months ago
- 1567c80 [vm] Fix vm ref trace function definition (#15247) by Lei Zhang · 1 year, 6 months ago
- 205c88d [metal] Retain device until command buffer completion (#15288) by Lei Zhang · 1 year, 6 months ago
- 28732a2 Start LLVM integrate integrate-llvm-20231018_6 (#15223) by Kunwar Grover · 1 year, 6 months ago
- e62db18 Update stale reference to ancient manylinux image. by Stella Laurenzo · 1 year, 6 months ago
- 6a850cb [scf] Pushed SCF support through rest of the Flow/Stream/HAL (#15063) by Rob Suderman · 1 year, 6 months ago
- 8b1e2ac Fixed VariantOp crash in LLVMCPULowerExecutableTarget (#15277) by Abdul Raheem · 1 year, 6 months ago
- e161480 Fix PJRT Build for Mac OS X (#15280) by Rob Suderman · 1 year, 6 months ago
- 2446351 [scf] Fix Resource processing for `scf.for` (#15055) by Rob Suderman · 1 year, 6 months ago
- 2bfc636 PJRT C API v0.35 (#15269) by Rob Suderman · 1 year, 6 months ago
- bb1efe8 Update Github runner to 2.311.0 (#15274) by Jerry Wu · 1 year, 6 months ago
- 8c34b97 Use custom iree.dev domain in links to documentation site. (#15036) by Scott Todd · 1 year, 6 months ago
- 094d9ba [Python] Add Fence VMRef Binding to enable async-exec on py. (#15263) by Stanley Winata · 1 year, 6 months ago
- e79fb2b [metal] Fix staging buffer alignment calculation (#15272) by Lei Zhang · 1 year, 6 months ago
- 0bdf338 [metal] Fix imported vm buffer ref counting in allocator (#15271) by Lei Zhang · 1 year, 6 months ago
- bd97cc5 Bump TF version in samples Dockerfile for Colab. (#15270) by Scott Todd · 1 year, 6 months ago
- e74287f Make TopK work with arbitrary rank (#15268) by Rob Suderman · 1 year, 6 months ago
- 99dc6bc Update CNAME by Jacques Pienaar · 1 year, 6 months ago
- 7d650ae [CPU] Fix check for scalable tile sizes (#15264) by Benjamin Maxwell · 1 year, 6 months ago
- aa5602d Improvements to e2e matmul tests (take 2) (#15259) by bjacob · 1 year, 6 months ago
- 1b8802a Re-land [EmitC] Add common baseclass for conversions and simplify patterns. (#15267) by Scott Todd · 1 year, 6 months ago
- 284c210 Revert "[EmitC][NFC] Introduce common baseclass for conversions and simplify pattern insertion" (#15265) by Scott Todd · 1 year, 6 months ago
- 69067bc [EmitC][NFC] Introduce common baseclass for conversions and simplify pattern insertion (#14717) by Simon Camphausen · 1 year, 6 months ago
- 66d6968 NFC: Remove dead hal.dispatch.extern builder code (#15262) by Quinn Dawkins · 1 year, 6 months ago
- 78e9dbc [FlashAttention] Adapt attention for tiling + implement tiling/decompose (#15217) by Abhishek Varma · 1 year, 6 months ago
- 3323519 Port samples/dynamic_shapes/ to PyTorch using SHARK-Turbine. (#15255) by Scott Todd · 1 year, 6 months ago
- 20e2112 Cleaning up hal.executable.variant syntax. (#15254) by Ben Vanik · 1 year, 6 months ago
- a95a28a Move expected output URL into the parameter of `iree_run_module_test` (#15248) by Jerry Wu · 1 year, 6 months ago
- eb9b8b6 Revert "Improvements to e2e matmul tests" (#15252) by bjacob · 1 year, 6 months ago
- 71c22da Improvements to e2e matmul tests (#15243) by bjacob · 1 year, 6 months ago
- 5a20dce bf16: select appropriate tile sizes on x86 and Arm, and enable in x86 bitcode build (#15244) by bjacob · 1 year, 6 months ago
- 02e34b0 Optimize `moveOp[Up,Down]InBlock` functions in `SimplifyGlobalAccesses`. (#15245) by Scott Todd · 1 year, 6 months ago
- df00df9 Disable flaky metal `e2e/*_ops` tests. (#15240) by Scott Todd · 1 year, 6 months ago