Fix regression in jiterpreter trace generation #99083
Merged
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
… have much larger back branch dispatch tables than necessary
|
Tagging subscribers to 'arch-wasm': @lewing Issue DetailsShould fix some or all of dotnet/perf-autofiling-issues#29881 When I migrated the jiterpreter to a new branch target scanning pass I failed to filter the branch target tables it generates properly. This combined with an existing bug in the cfg code meant that traces in large methods ended up having big branch dispatch tables at the top where most of the branch targets were never hit, eating up a valuable chunk of the trace's 4KB limit. This PR adds appropriate filtering in both the scanning pass and the cfg so that the dispatch table is as small as possible.
|
ghost
assigned
vargaz
approved these changes
Feb 28, 2024
This was referenced Feb 29, 2024
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
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Outdated suggestions cannot be applied.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.
Should fix some or all of dotnet/perf-autofiling-issues#29881
When I migrated the jiterpreter to a new branch target scanning pass I failed to filter the branch target tables it generates properly. This combined with an existing bug in the cfg code meant that traces in large methods ended up having big branch dispatch tables at the top where most of the branch targets were never hit, eating up a valuable chunk of the trace's 4KB limit.
This PR adds appropriate filtering in both the scanning pass and the cfg so that the dispatch table is as small as possible.