Revert create_output memoization
#120
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
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Outdated suggestions cannot be applied.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.
This PR reverts the changes made to
create_outputfrom #115. The change in question was to use a hashmap to memoize which output paths had been written whencreate_outputwas called (the goal was to avoid writing to the same path multiple times)It turns out this change could cause severe slowdowns in some cases. I started digging into why building the eza package locally was taking so long-- around 3-4 minutes-- even though the actual Rust build was pretty fast (about 30 seconds). I tested again with a version of Brioche before #115, and the build was much faster: about 1m18s
I did lots of experiments and rebuilds, and finally narrowed the problem down to
create_output. Undoing thecreate_outputchanges from #115 while leaving everything else in place made the build finish in 1m10s, even faster than before!