- b68b3d4 Reduce default max tile size for workgroup distribution for Generic ops. (#9478) by MaheshRavishankar · 2 years, 10 months ago
- 88be147 [vulkan] Disable clz.mlir tests due to Adreno GPU failure (#9509) by Lei Zhang · 2 years, 10 months ago
- f235749 Implement On Demand Downloading of CUDA Dependencies for non CI Builds (#9425) by Kojo Acquah · 2 years, 10 months ago
- bba260d Some updates to build external rocm (#9502) by nirvedhmeshram · 2 years, 10 months ago
- 0d19c32 Adding __NR_futex workaround for riscv32. (#9501) by Ben Vanik · 2 years, 10 months ago
- 23da926 [docs] Wrap shell command in shell codeblock by Geoffrey Martin-Noble · 2 years, 10 months ago
- a4ca33e Bring CanonicalizedSequenceOp in sync with SequenceOp (#9494) by Matthias Springer · 2 years, 10 months ago
- 0c1e81b Add e2e tests for large aligned matrix for cuda only (#9492) by Thomas · 2 years, 10 months ago
- a7f3357 Fix transform dialect op cloning. (#9496) by Nicolas Vasilache · 2 years, 10 months ago
- 5e523ee Outline workgroup count computing IR in a new DispatchWorkgroupsOp::workgroup_count region. (#9362) by Nicolas Vasilache · 2 years, 10 months ago
- 597b310 Merge pull request #9471 from google/benvanik-tracy-fibers-enable by Ben Vanik · 2 years, 10 months ago
- de97197 [LLVMGPU] Fix bug when trying to promote 4d contraction op (#9488) by Thomas · 2 years, 10 months ago
- f1d67b8 Topk Parallel Reduction Initial Implementation (#9364) by Kojo Acquah · 2 years, 10 months ago
- 75a02ab ThreadSanitizer continuous integration in Buildkite (#9371) by bjacob · 2 years, 10 months ago
- 9e57f53 Add a flag for controlling the max stack allocation limit allowed on CPU backend. (#9477) by MaheshRavishankar · 2 years, 10 months ago
- 4e3b095 Deal with slow e2e tests on VMVX (#9435) by bjacob · 2 years, 10 months ago
- a34c7ac Apply code upgrades for python >= 3.7 (#9401) by Ramiro Leal-Cavazos · 2 years, 10 months ago
- 47d7b45 Revert "Use clang-format-12 in presubmits" (#9480) by Geoffrey Martin-Noble · 2 years, 10 months ago
- 14f92e6 Bump pybind requirement to 2.8.0 (#9476) by Geoffrey Martin-Noble · 2 years, 10 months ago
- 997911a [spirv] Improve CodeGen flow for NVIDIA architectures (#9458) by Lei Zhang · 2 years, 10 months ago
- 40a49c6 Use clang-format-12 in presubmits (#9473) by Geoffrey Martin-Noble · 2 years, 10 months ago
- 9297a78 Documentation fix for vision_inference sample (#9472) by Matt Walsh · 2 years, 10 months ago
- d5d9161 Merge pull request #9466 from google/benvanik-optimal-buffers by Ben Vanik · 2 years, 10 months ago
- 35c2864 [spirv] Drop math.ctlz expansion given we have native conversion (#9470) by Lei Zhang · 2 years, 10 months ago
- e3f043b Bumping bytecode module version for HAL changes (pending #8014). by Ben Vanik · 2 years, 10 months ago
- 5844fd6 Reworking IREE_HAL_BUFFER_USAGE_* bits to carry more information. by Ben Vanik · 2 years, 10 months ago
- 7ec857d Cherry-pick a few SPIR-V related patches in LLVM/MLIR (#9465) by Lei Zhang · 2 years, 10 months ago
- 74ba582 Adding MSVC x86-32 ELF loading. (#9467) by Ben Vanik · 2 years, 10 months ago
- 6a4c9df Add Linux iree-dist artifact back to the release asset (#9464) by CindyLiu · 2 years, 10 months ago
- 6e4fa1c Adding IREE_HAL_MEMORY_TYPE_OPTIMAL* bits. by Ben Vanik · 3 years ago
- af0a780 Renaming to iree_hal_buffer_mapping_*_range for consistency. by Ben Vanik · 3 years ago
- 22fba1a Merge pull request #9461 from google/benvanik-npy by Ben Vanik · 2 years, 10 months ago
- 2a1925c Refactor `IREE_ENABLE_EMITC` -> `IREE_OUTPUT_FORMAT_C`. (#9455) by Scott Todd · 2 years, 10 months ago
- ade5b1b Integrate llvm-project at 25c8a061c573 and bump dependencies. (#9463) by Han-Chung Wang · 2 years, 10 months ago
- aecb725 Allowing `--function_input=@file.npy` in command line tools. by Ben Vanik · 2 years, 10 months ago
- 5338ea6 Adding numpy npy file load/save utility. by Ben Vanik · 2 years, 10 months ago
- fc11126 Cleanups to TransformDialectInterpreter and refactor (#9446) by Nicolas Vasilache · 2 years, 10 months ago
- a7fe4ae Merge pull request #9434 from matthias-springer/transform_py_bindings by Matthias Springer · 2 years, 10 months ago
- 00f60fb Use MLIR transform dialect Python bindings by Matthias Springer · 2 years, 10 months ago
- be4c7b4 Merge pull request #9457 from google/benvanik-vulkan-uuid by Ben Vanik · 2 years, 10 months ago
- 8c82983 Allowing deviceUUIDs to be used in Vulkan device paths. by Ben Vanik · 2 years, 10 months ago
- bf4a610 Adding iree_string_view_parse_hex_bytes utility. by Ben Vanik · 2 years, 10 months ago
- f8bbc1b Fixing default CUDA device selection. (#9454) by Ben Vanik · 2 years, 10 months ago
- 0f1f8eb Always add local_task and VMVX if building the compiler. (#9450) by Scott Todd · 2 years, 10 months ago
- f66ea1b Always pass `--output-format=vm-c` in iree_c_module. (#9451) by Scott Todd · 2 years, 10 months ago
- e029d0f Switching from cuDeviceGetUuid_v2 to cuDeviceGetUuid for older CUDA. (#9448) by Ben Vanik · 2 years, 10 months ago
- 795f6f9 Add BUILD file for vm/test/emitc and samples/emitc_modules (#9431) by Jacques Pienaar · 2 years, 10 months ago
- b3101c2 Integrate llvm-project and bump dependencies (#9441) by Okwan Kwon · 2 years, 10 months ago
- 2814631 Remove explicit `--iree-mlir-to-vm-bytecode-module` flag usage. (#9445) by Scott Todd · 2 years, 10 months ago
- 379c2e4 Merge pull request #9443 from google/benvanik-device-flags by Ben Vanik · 2 years, 10 months ago
- 33a7caa Rename "translate" to "compile" across build systems. (#9442) by Scott Todd · 2 years, 10 months ago
- bba52ae Changing uses of --driver= to --device= (that I can find). by Ben Vanik · 2 years, 10 months ago
- 46ec250 Replacing --driver with --device in tools. by Ben Vanik · 2 years, 10 months ago
- 6b65b53 Adding a --device flag to replace --driver. by Ben Vanik · 2 years, 10 months ago
- 7876a79 Adding --list_drivers, --list_devices, and --dump_devices flags. by Ben Vanik · 2 years, 11 months ago
- 243875c Enable 32-bit lowering of tosa.apply_scale for RISC-V (#9407) by Diego Caballero · 2 years, 10 months ago
- 8aa71c0 NFC: update Vulkan attributes to use AttrDef (#9439) by Lei Zhang · 2 years, 10 months ago
- 5aa452b Fix case where there are no test temp directories to create (#9438) by Geoffrey Martin-Noble · 2 years, 10 months ago
- 9cbbdbb NFC: move Vulkan attribute code to VulkanAttributes.cpp (#9437) by Lei Zhang · 2 years, 10 months ago
- 89e0274 Switch tests to runtime skipping if drivers are missing. (#9429) by Scott Todd · 2 years, 10 months ago
- 931b405 Drop LinalgExt::TileOp, LinalgExt::InParallelOp and LinalgExt HAL shim ops (#9424) by Nicolas Vasilache · 2 years, 10 months ago
- 6d41cd0 Add tool to collect compilation statistics (#9227) by Jerry Wu · 2 years, 10 months ago
- a7afce7 Add a pass to create a Flow::DispatchWorkgroupsOp using the transform dialect. (#9313) by Nicolas Vasilache · 2 years, 11 months ago
- 796d768 Adding an optional workgroup count region to flow.dispatch.workgroups. (#9428) by Ben Vanik · 2 years, 11 months ago
- fdabffc Bump llvm 20220609 2 (#9427) by Okwan Kwon · 2 years, 11 months ago
- 10f1822 Use CMake to create temporary test directories eliminating run_test.* (#9405) by Geoffrey Martin-Noble · 2 years, 11 months ago
- cd62cb5 Merge pull request #9422 from google/benvanik-flags-prefix by Ben Vanik · 2 years, 11 months ago
- 0748091 Update docs in samples/vision_inference. (#9420) by Scott Todd · 2 years, 11 months ago
- 4a917a9 Removing line number from --help flags header. by Ben Vanik · 2 years, 11 months ago
- c5f97a1 Using -fmacro-prefix-map to clean up __FILE__ macros. by Ben Vanik · 2 years, 11 months ago
- f697fbc Remove third_party/renderdoc_api (again). (#9419) by Scott Todd · 2 years, 11 months ago
- 70c2bb0 Making runtime APIs with sized arrays consistently (count, array). (#9416) by Ben Vanik · 2 years, 11 months ago
- 097d826 Renaming iree/tools/utils/ -> iree/tooling/. (#9417) by Ben Vanik · 2 years, 11 months ago
- 6d77ca0 Adding a device_util helper for tooling. (#9415) by Ben Vanik · 2 years, 11 months ago
- 432cb62 Integrate llvm-project and bump dependencies (#9414) by Okwan Kwon · 2 years, 11 months ago
- 3c0d84b Adding iree_hal_driver_create_device_by_path & co. (#9413) by Ben Vanik · 2 years, 11 months ago
- 7ca0f46 Replace samples/vision_inference with a Python script + README. (#9403) by Scott Todd · 2 years, 11 months ago
- f3362af Use upstream vectorize transform op (#9412) by Oleksandr "Alex" Zinenko · 2 years, 11 months ago
- d97e8bc Add `--no-tests=error` to ctest invocations in CI scripts. (#9376) by Scott Todd · 2 years, 11 months ago
- 754719f Use upstream transform op for fusion (#9411) by Oleksandr "Alex" Zinenko · 2 years, 11 months ago
- 5f548a6 [vulkan] Create command pools with CREATE_RESET_COMMAND_BUFFER_BIT (#9404) by Lei Zhang · 2 years, 11 months ago
- c16f33e Merge pull request #9409 from ftynse/xform-dialect by Oleksandr "Alex" Zinenko · 2 years, 11 months ago
- 452802a Use upstream MLIR SCF transform dialect extension by Alex Zinenko · 2 years, 11 months ago
- 4700cdc Use upstream decompose and generalize transform ops (#9408) by Oleksandr "Alex" Zinenko · 2 years, 11 months ago
- d83f4e4 Fixing race on semaphore timepoint cancellation. (#9400) by Ben Vanik · 2 years, 11 months ago
- d42e002 Create a clang-tidy option for building IREE (#9402) by Rob Suderman · 2 years, 11 months ago
- 74d741e Fix samples after driver refactoring. (#9391) by Scott Todd · 2 years, 11 months ago
- 718e04e Update documentation for local-task and local-sync drivers. (#9399) by Scott Todd · 2 years, 11 months ago
- 1c84fdf Removing iree_hal_driver_id_t in favor of just string names. (#9389) by Ben Vanik · 2 years, 11 months ago
- cc04405 check if the def op is null (#9396) by Okwan Kwon · 2 years, 11 months ago
- a54d662 Integrate llvm-project and bump dependencies (#9390) by Okwan Kwon · 2 years, 11 months ago
- cfe6913 Replace "bug report" issue template with issue forms. (#9352) by Scott Todd · 2 years, 11 months ago
- 5f1d195 Define module target in iree_bytecode_module (#9387) by Jerry Wu · 2 years, 11 months ago
- 259b4df Set 4 threads to run CTest on Android devices (match number of little cores) (#9372) by bjacob · 2 years, 11 months ago
- f9535d7 Renaming HAL function host allocator args to host_allocator. (#9388) by Ben Vanik · 2 years, 11 months ago
- fcd82d6 Adding vm.ctlz.i32/i64 ops. (#9382) by Ben Vanik · 2 years, 11 months ago
- 1c59c11 Renaming file_path to path and adding some basic URI utils. (#9384) by Ben Vanik · 2 years, 11 months ago
- 2474ecc Merge pull request #9380 from google/benvanik-external-hal-drivers by Ben Vanik · 2 years, 11 months ago
- f9b3708 dequantize softmax (#9337) by bjacob · 2 years, 11 months ago
- 9b86ad5 Dont' add `vulkan_headers` submodule by default (#9386) by Marius Brehler · 2 years, 11 months ago