[clr-interp] Improve the performance of interpreter calls#120067
Merged
davidwrighton merged 6 commits intodotnet:mainfrom Sep 26, 2025
Merged
[clr-interp] Improve the performance of interpreter calls#120067davidwrighton merged 6 commits intodotnet:mainfrom
davidwrighton merged 6 commits intodotnet:mainfrom
Conversation
- Instead of setting the target in the CallStubHeader at each invocation, store the target in the CallStubHeader immediately. - Avoid invoking the prestub unnecessarily. Instead, detect that we've already run the prestub to completion at least once before poisoning the m_interpreterCode field on MethodDesc. This may be enough to make some of our tests which are currently timing out pass.
Contributor
|
Tagging subscribers to this area: @mangod9 |
Member
Author
|
/azp run runtime-interpreter |
|
Azure Pipelines successfully started running 1 pipeline(s). |
… were not quite ready to execute.
…_InterpreterMethodCallPerf
Member
Author
|
/azp run runtime-interpreter |
|
Azure Pipelines successfully started running 1 pipeline(s). |
kg
reviewed
Sep 25, 2025
kg
approved these changes
Sep 25, 2025
davidwrighton
requested review from
Copilot
and removed request for
BrzVlad and
janvorli
Contributor
There was a problem hiding this comment.
Pull Request Overview
This PR optimizes interpreter call performance by reducing redundant work and avoiding unnecessary prestub invocations. The changes focus on storing call targets in CallStubHeader immediately rather than setting them at each invocation, and implementing a poisoning mechanism to prevent repeated failed attempts to get interpreter code.
- Store call targets in CallStubHeader at creation time to avoid repeated target setting
- Add poisoning mechanism to prevent redundant prestub calls for methods that cannot be interpreted
- Optimize call paths by detecting when prestub has already completed successfully
Reviewed Changes
Copilot reviewed 4 out of 4 changed files in this pull request and generated 3 comments.
| File | Description |
|---|---|
| src/coreclr/vm/wasm/helpers.cpp | Updates InvokeManagedMethod to handle NULL target by getting multi-callable address |
| src/coreclr/vm/method.hpp | Adds interpreter code poisoning mechanism with new methods and constants |
| src/coreclr/vm/interpexec.cpp | Major optimization changes including target storage in CallStubHeader and poisoning logic |
| src/coreclr/vm/callstubgenerator.h | Adds GetTarget() method to CallStubHeader for retrieving stored targets |
This was referenced Sep 26, 2025
Open
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
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
2 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.
This may be enough to make some of our tests which are currently timing out pass.