Allow choosing memory type for CudaPinned allocator #26892
+38
−2
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
2) Added missing override stub function in tests
Contributor
|
/azp run Linux QNN CI Pipeline, Win_TRT_Minimal_CUDA_Test_CI, Windows ARM64 QNN CI Pipeline, Windows GPU Doc Gen CI Pipeline |
|
Azure Pipelines successfully started running 4 pipeline(s). |
Contributor
There was a problem hiding this comment.
Pull request overview
This PR enables the CudaPinned allocator to use configurable memory types (DeviceAllocator or ArenaAllocator) instead of being hardcoded to DeviceAllocator, addressing issue #26887.
Key Changes:
- Added a new
CreateCudaPinnedAllocatormethod across the CUDA provider interface hierarchy with arena configuration parameters - Updated the allocator creation logic in
environment.ccto pass arena configuration to the new method - Implemented the actual allocator creation logic in
CUDAExecutionProvider
Reviewed changes
Copilot reviewed 6 out of 6 changed files in this pull request and generated 2 comments.
Show a summary per file
| File | Description |
|---|---|
| onnxruntime/core/providers/cuda/cuda_provider_factory.h | Adds interface declaration for CreateCudaPinnedAllocator with arena configuration parameters |
| onnxruntime/core/providers/cuda/cuda_provider_factory.cc | Implements the factory method to delegate to CUDAExecutionProvider::CreateCudaPinnedAllocator |
| onnxruntime/core/providers/cuda/cuda_execution_provider.h | Declares static method CreateCudaPinnedAllocator in the execution provider |
| onnxruntime/core/providers/cuda/cuda_execution_provider.cc | Implements the allocator creation with arena configuration support |
| onnxruntime/core/session/environment.cc | Updates call site to use new method name and pass arena configuration |
| onnxruntime/test/providers/cuda/test_cases/cuda_test_provider.cc | Adds stub implementation for test provider interface |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
Co-authored-by: Copilot <[email protected]>
…ateCudaPinnedAllocator
Contributor
|
/azp run Linux QNN CI Pipeline, Win_TRT_Minimal_CUDA_Test_CI, Windows ARM64 QNN CI Pipeline, Windows GPU Doc Gen CI Pipeline |
|
Azure Pipelines successfully started running 4 pipeline(s). |
tianleiwu
approved these changes
Jan 8, 2026
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Description
CudaPinned allocator uses hardcoded to DeviceAllocator memory type.
The PR comes to allow choosing memory type for CudaPinned allocator between DeviceAllocator and ArenaAllocator.
Motivation and Context
Fixed issue #26887