inference: fixes cache lookup with extended lattice elements #53953
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
|
@nanosoldier |
|
Your benchmark job has completed - possible performance regressions were detected. A full report can be found here. |
aviatesk
force-pushed
the
avi/cache-lookup-extended-lattice
branch
from
c65542c
to
5975795
Compare
|
@nanosoldier |
|
Your benchmark job has completed - possible performance regressions were detected. A full report can be found here. |
aviatesk
force-pushed
the
avi/cache-lookup-extended-lattice
branch
from
5975795
to
f36db1b
Compare
|
@nanosoldier |
|
Your benchmark job has completed - possible performance regressions were detected. A full report can be found here. |
aviatesk
force-pushed
the
avi/cache-lookup-extended-lattice
branch
from
f36db1b
to
6e0c259
Compare
|
@nanosoldier |
aviatesk
force-pushed
the
avi/cache-lookup-extended-lattice
branch
from
6e0c259
to
13e854f
Compare
|
Your benchmark job has completed - possible performance regressions were detected. A full report can be found here. |
aviatesk
force-pushed
the
avi/cache-lookup-extended-lattice
branch
from
13e854f
to
435cef5
Compare
The previous local cache lookup system wasn't working well for caches with extended lattice elements that are transformed from the caller context to the callee context by `matching_cache_argtypes`. To address this, the current commit moves the call to `matching_cache_argtypes` right before `cache_lookup`, instead of within the `InferenceResult` constructor. Note that this adjustment leads to `given_argtypes` being allocated even when there's a cache hit, potentially affecting performance negatively. I'm looking to further optimize `matching_cache_argtypes` in an upcoming commit.
aviatesk
force-pushed
the
avi/cache-lookup-extended-lattice
branch
from
435cef5
to
a5356a4
Compare
aviatesk
added a commit
to JuliaDiff/Diffractor.jl
that referenced
this pull request
Apr 9, 2024
aviatesk
added a commit
that referenced
this pull request
Apr 9, 2024
External abstract interpreters (mainly Diffractor) may want to ir-interpret top-level code chunk.
aviatesk
added a commit
to JuliaDiff/Diffractor.jl
that referenced
this pull request
Apr 9, 2024
aviatesk
added a commit
to JuliaDiff/Diffractor.jl
that referenced
this pull request
Apr 9, 2024
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.
The previous local cache lookup system wasn't working well for caches with extended lattice elements that are transformed from the caller context to the callee context by
matching_cache_argtypes. To address this, the current commit moves the call tomatching_cache_argtypesright beforecache_lookup, instead of within theInferenceResultconstructor.Note that this adjustment leads to
given_argtypesbeing allocated even when there's a cache hit, potentially affecting performance negatively. I'm looking to further optimizematching_cache_argtypesin an upcoming commit.