[core] allow using initializer allocator for prepack #26617
+2
−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
Contributor
There was a problem hiding this comment.
Pull Request Overview
This PR enables the use of read-only initializer allocators for prepack operations, improving memory usage when an execution provider registers a dedicated initializer allocator (e.g., WebGPU EP). The change maintains backward compatibility by falling back to the standard allocator when no initializer-specific allocator is available.
- Replaces
GetAllocatorwithGetInitializerAllocatorin the non-shared prepack path - Maintains backward compatibility for EPs without initializer allocators
- Leverages existing
GetInitializerAllocatorinfrastructure
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
yuslepukhin
reviewed
Nov 20, 2025
| // we store the newly minted pre-packed data. | ||
|
|
||
| AllocatorPtr session_cpu_alloc = GetAllocator(kernel->Info().GetDevice(OrtMemType::OrtMemTypeDefault)); | ||
| AllocatorPtr session_initializer_alloc = GetInitializerAllocator(kernel->Info().GetDevice(OrtMemType::OrtMemTypeDefault)); |
Member
Contributor
Author
There was a problem hiding this comment.
The original variable name is confusing. It actually returns the registered allocator by looking up the map using the OrtDevice as key. for a node on webgpu EP, this returns a webgpu allocator instead of a CPU one.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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
This PR makes ORT to prefer initializer allocator when calling
OpKernel::PrePack.If an EP does not register an initializer allocator (currently only WebGPU does this), the behavior is kept unchanged.
Motivation and Context
Helps to improve the memory usage when doing prepack.