carefully cache freshly-inferred edge for call-site inlining #51975
Conversation
This file contains 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
|
Your benchmark job has completed - possible performance regressions were detected. A full report can be found here. |
aviatesk
force-pushed
the
avi/callsite-inline-fresh-edge
branch
3 times, most recently
from
c033de1 to
f480b9a
Compare
vtjnash
approved these changes
Nov 2, 2023
Currently call-site inlining fails on freshly-inferred edge if its source isn't inlineable. This happens because such sources are exclusively cached globally. For successful call-site inlining, it's necessary to cache these locally also, ensuring the inliner can access them later. To this end, the type of the `cache_mode` field of `InferenceState` has been switched to `UInt8` from `Symbol`. This change allows it to represent multiple caching strategies. A new caching mode can be introduced, e.g. `VOLATILE_CACHE_MODE`, in the future.
aviatesk
force-pushed
the
avi/callsite-inline-fresh-edge
branch
from
f480b9a to
8415b88
Compare
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Outdated suggestions cannot be applied.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.
Currently call-site inlining fails on freshly-inferred edge if its source isn't inlineable. This happens because such sources are exclusively cached globally. For successful call-site inlining, it's necessary to cache these locally also, ensuring the inliner can access them later.
To this end, the type of the
cache_modefield ofInferenceStatehas been switched toUInt8fromSymbol. This change allows it to encode multiple caching strategies. A new caching mode can be introduced, e.g.VOLATILE_CACHE_MODE, in the future.@nanosoldier
runbenchmarks("inference", vs=":master")