Df/caching termlike hashes #3338
Merged
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
4 tasks
|
This re-creates the one |
|
treeowl
force-pushed
the
df/caching-termlike-hashes
branch
2 times, most recently
from
2e495a1 to
7316801
Compare
|
treeowl
force-pushed
the
df/caching-termlike-hashes
branch
2 times, most recently
from
699ae15 to
82e7c64
Compare
|
|
/KEVM-perf-run |
Cache the hash of the `TermLikeF` in the `TermLike`, and extract it in the `Hashable` instance instead of recursing into the `TermLikeF`. The precomputed hash in produced using the derived `Hashable` instance of `TermLikeF`, which will call `hash` on the respective sub-structure (like `And`, `Or`, ...), wrapped into one of the `TermLikeF` constructors. For all such sub-structures, the derived `Hashable` instance will then recurse into the contents (containing a `TermLike` again for the case of `TermLike`), thereby using the cached hashes directly instead of performing a deep recursion. Partially addresses #3113
Write a specialized
from :: TermLike Concrete -> TermLike variable
I was hoping this would remove a number of the new `Hashable`
constraints; unfortunately it was not so successful in that regard. It
did, however remove quite a lot of `InternalVariable` constraints, so I
guess that's a win.
`fourmolu` support for CPP is experimental, limited, and basically buggy. It's apparently okay to make whole top-level declarations conditional, so let's do it that way.
treeowl
force-pushed
the
df/caching-termlike-hashes
branch
from
82e7c64 to
1e6ae6c
Compare
|
/KEVM-perf-run |
|
The KEVM test suggests I lack adequate permissions. |
|
That's a gotcha in the github settings - you need to be publicly a member of the RV organosation, i.e. make your membership publicly visible. |
|
/KEVM-perf-run |
|
Running KEVM test with 1e6ae6c and K version as https://github.com/runtimeverification/haskell-backend/actions/runs/3466091969 |
|
Running test with 1e6ae6c and kevm-bug-storagevar03-with-conditions.tar.gz as https://github.com/runtimeverification/haskell-backend/actions/runs/3466278670 |
|
KEVM test finished: https://github.com/runtimeverification/haskell-backend/actions/runs/3466091969 EDIT: pasting results here 🎉 looks like a consistent 10% speedup 🎉
|
|
Looks promising! It would be good to see what happens with the stricter versions. |
|
indeed...worth a try. These are the top consumers now (from the
|
goodlyrottenapple
approved these changes
Nov 22, 2022
|
@treeowl ^^^ we have an |
jberthold
approved these changes
Nov 22, 2022
ana-pantilie
approved these changes
Nov 22, 2022
|
|
Aha, okay. I'll use that in the future. I was waiting for @ana-pantilie before pulling the lever on this one. |
This was referenced Nov 30, 2022
Merged
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.
Cache the hash of the TermLikeF in the TermLike, and extract it in the Hashable instance instead of recursing into the TermLikeF.
The precomputed hash in produced using the derived Hashable instance of TermLikeF, which will call hash on the respective sub-structure (like And, Or, ...), wrapped into one of the TermLikeF constructors. For all such sub-structures, the derived Hashable instance will then recurse into the contents (containing a TermLike again for the case of TermLike), thereby using the cached hashes directly instead of performing a deep recursion.
Partially addresses #3113
Fixes #nnnn
Scope:
Estimate:
Review checklist
The author performs the actions on the checklist. The reviewer evaluates the work and checks the boxes as they are completed.