[NFC] Avoid quadratic time when precomputing blocks #6862
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
tlively
approved these changes
Aug 21, 2024
Member
tlively
left a comment
tlively
left a comment
There was a problem hiding this comment.
LGTM % clarifying the example in the comment.
src/passes/Precompute.cpp
Outdated
Comment on lines
405
to
407
| // (br $out) | ||
| // ) | ||
| // (undo that side effect exactly, and nothing more) |
Member
There was a problem hiding this comment.
This br would skip over the undo of the side effects, so I don't think this example makes sense.
tlively
approved these changes
Aug 21, 2024
Closed
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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Outdated suggestions cannot be applied.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.
When precomputing fails on a child block of a parent block, there is no point
to precompute the parent, as that will fail as well.
At least, I can't think of any instruction sequence in wasm that could cause that
(see detailed comment in the code). I also fuzzed this (see second commit,
which is later reverted), and failed to find anything after a few thousand
iterations. But in theory this is not NFC if there is something that could be
optimized.
This makes
--precomputeon Emscripten'stest_biggerswitchgo from 1.44seconds to 0.02 seconds (not a typo, that is 72x faster). The absolute number
is not that big, but we do run this pass more than once, so it saves a noticeable
chunk of time.