-
Notifications
You must be signed in to change notification settings - Fork 12.7k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Revert "Compare tagged/niche-filling layout and pick the best one" #74611
Conversation
We can re-land this PR at a later point if there's a way to recover the perf regression here, but given the pretty large variety of performance regressions over the last week it seems good to eagerly revert this to be better able to evaluate the new version. @bors rollup=never |
Oh wow that's definitely much larger than just computing both layouts was (then again I didn't use @bors r+ |
|
📌 Commit 1d68600 has been approved by |
@bors rollup=never Due to perf effects. |
⌛ Testing commit 1d68600 with merge f4c6964303040a7712afce36bfe89c3663a032ae... |
💔 Test failed - checks-azure |
@bors retry - error inflating zlib |
☀️ Test successful - checks-actions, checks-azure |
…ercote Reland rust-lang#74069 Investigation in rust-lang#74716 has concluded that this PR is indeed not a regression (and in fact the rollup itself is not either). This reverts the revert in rust-lang#74611. r? @nnethercote cc @eddyb
Reverts #74069. It caused a performance regression, see #74069 (comment). perf: https://perf.rust-lang.org/compare.html?start=d3df8512d2c2afc6d2e7d8b5b951dd7f2ad77b02&end=cfade73820883adf654fe34fd6b0b03a99458a51
r? @eddyb
cc @nnethercote