-
Notifications
You must be signed in to change notification settings - Fork 3k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[MigraphX EP] [ROCm EP] Upstream ROCm changes for bugfixes and features #23249
base: main
Are you sure you want to change the base?
Conversation
…t#23204) <!-- Describe your changes. --> Changed all support tensor type from ir 9 to ir 10. <!-- - Why is this change required? What problem does it solve? - If it fixes an open issue, please link to the issue here. --> - See issue microsoft#23205 Co-authored-by: Yueqing Zhang <[email protected]>
This works around having too many gfx targets which otherwise fails linking.
* create package for migraphx ep * add migrahx to the gpu providers for benchmark.py * remove rocm from migraphx perfs tests
* Add support for MIGraphX Exhaustive tune flag in MIGraphX EP (#46) * Add support for MIGraphX Exhaustive tune flag in MIGraphX EP Enable exhaustive tune by either python interface of environment env in bash * Apply lintrunner pass * Fix compile errors. * Lintrunner pass * handle review comments --------- Co-authored-by: Ted Themistokleous <[email protected]>
* Force MIGraphXEP when MIGraphX is chosen * Fix lint
#53) * Ensure we support all inputs for MatMulInteger and ConvInteger. Limit to int8 for now Allow for models with biases/full input and only check for int8 support in EP * Add support for uint8 types --------- Co-authored-by: Ted Themistokleous <[email protected]>
…ripts (#58) (#59) Co-authored-by: Ted Themistokleous <[email protected]>
This fixes SWDEV-486455
* Force hipify to use copied version from rocm-6.3.0-14776 build * Update hipify-perl location and permissions * Update hipify path to remove absolute path
GCC 13 introduces -W(no-)dangling-reference but AMDClang does not support this yet.
Raise the tolerance of CudaKernelTest.SoftmaxGrad tests. This is a temporary workaround of https://ontrack-internal.amd.com/browse/SWDEV-477109 to make CI happy.
* Fix SWDEV-483388: mistakenly invoke hipMemcpy when ORT_ENABLE_STREAM=True * Fix another hipMemcpy * Remove MIGRAPHX_STREAM_SYNC guard which does not exists in current version.
* Added maximum gridDim.y overflow heck before calling transposeNoOverlap kernel so that TransposeBigMLFloat16 test passes * Fix formatting
ping @tianleiwu this effects both ROCm and MIGraphX EPs as this adds various features that weren't upstreamed yet |
/azp run Windows ARM64 QNN CI Pipeline,Windows x64 QNN CI Pipeline,Windows CPU CI Pipeline,Windows GPU CUDA CI Pipeline,Windows GPU DML CI Pipeline,Windows GPU Doc Gen CI Pipeline,Windows GPU TensorRT CI Pipeline,ONNX Runtime Web CI Pipeline,Linux CPU CI Pipeline,Linux CPU Minimal Build E2E CI Pipeline |
/azp run Linux GPU CI Pipeline,Linux GPU TensorRT CI Pipeline,Linux OpenVINO CI Pipeline,Linux QNN CI Pipeline,MacOS CI Pipeline,orttraining-linux-ci-pipeline,orttraining-linux-gpu-ci-pipeline,onnxruntime-binary-size-checks-ci-pipeline,Big Models,Linux Android Emulator QNN CI Pipeline |
/azp run Android CI Pipeline,iOS CI Pipeline,ONNX Runtime React Native CI Pipeline,CoreML CI Pipeline,Linux DNNL CI Pipeline,Linux MIGraphX CI Pipeline,Linux ROCm CI Pipeline |
Azure Pipelines successfully started running 7 pipeline(s). |
Azure Pipelines successfully started running 8 pipeline(s). |
Azure Pipelines successfully started running 10 pipeline(s). |
Description
Add support to mainline Onnxruntime of changes from the ROCm Team's changes
Motivation and Context
Various bugfixes, and changes added between ROCm 6.2 and 6.3 that haven't been upstreamed yet to mainline