Cache file data to reduce lock contention #116
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
jpenilla
added a commit
to PaperMC/patched-spigot-fernflower
that referenced
this pull request
Apr 7, 2022
Zip file instances lock on all reads, so we should try to load everything we need into memory all at once, rather than reading multiple times
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.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.
Zip file instances lock on all reads, so we should try to
load everything we need into memory all at once, rather than
reading multiple times
This does have some negative impact on base memory usage, but reduces allocations from file reads quite a bit, so it's probably about a wash on that front.
These changes knock about 10s off the runtime to decompile 1.18.2 (~40s to ~30s) on my machine (Ryzen 7 3800x, win 11, 64gb ram, SSD), with no change to the output.
Further work on true lazy loading of classes is upcoming, but this was a change with way lower impact, and where my work started so I pulled it out on its own.
Some reports from JFR that led to these changes:
Original
After introducing the ZipFile cache (step 1)
After caching data in the LazyLoader link (step 2)
(yes, this abuses LazyLoader, but LazyLoader is dumb and I don't care)