-
Notifications
You must be signed in to change notification settings - Fork 13.2k
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
[WIP] Try using call-site aware IR PGO for LLVM #97153
Conversation
@bors try @rust-timer queue |
Awaiting bors try build completion. @rustbot label: +S-waiting-on-perf |
⌛ Trying commit 15da36b487fb82d429209e0a4e08ad57981be4c1 with merge 8851e36fefe79fb0e7ceac2986b07bcf5e534835... |
💔 Test failed - checks-actions |
This comment has been minimized.
This comment has been minimized.
@bors try @rust-timer queue |
Awaiting bors try build completion. @rustbot label: +S-waiting-on-perf |
⌛ Trying commit d183f765a21b7928379d815cea11e481c20050b2 with merge 87302328f42a41865f79dd5d01857803704966db... |
☀️ Try build successful - checks-actions |
Queued 87302328f42a41865f79dd5d01857803704966db with parent c067287, future comparison URL. |
Finished benchmarking commit (87302328f42a41865f79dd5d01857803704966db): comparison url. Instruction count
Max RSS (memory usage)Results
CyclesResults
If you disagree with this performance assessment, please file an issue in rust-lang/rustc-perf. Benchmarking this pull request likely means that it is perf-sensitive, so we're automatically marking it as not fit for rolling up. While you can manually mark this PR as fit for rollup, we strongly recommend not doing so since this PR may lead to changes in compiler perf. Next Steps: If you can justify the regressions found in this try perf run, please indicate this with @bors rollup=never Footnotes |
@bors try @rust-timer queue |
Awaiting bors try build completion. @rustbot label: +S-waiting-on-perf |
⌛ Trying commit 3736e6b044b5e07c156ef77dc2e5f3dd8759e733 with merge cbed134b20d1fe82e1106f135cc4383f4021933b... |
☀️ Try build successful - checks-actions |
Queued cbed134b20d1fe82e1106f135cc4383f4021933b with parent b5caa5a, future comparison URL. |
Finished benchmarking commit (cbed134b20d1fe82e1106f135cc4383f4021933b): comparison url. Instruction countThis benchmark run did not return any relevant results for this metric. Max RSS (memory usage)Results
CyclesResults
If you disagree with this performance assessment, please file an issue in rust-lang/rustc-perf. Benchmarking this pull request likely means that it is perf-sensitive, so we're automatically marking it as not fit for rolling up. While you can manually mark this PR as fit for rollup, we strongly recommend not doing so since this PR may lead to changes in compiler perf. @bors rollup=never Footnotes |
@bors try @rust-timer queue |
Awaiting bors try build completion. @rustbot label: +S-waiting-on-perf |
⌛ Trying commit 93ff904 with merge d3b259f9622b4fb396bba698cc1df6dfdf5d3026... |
@bors try @rust-timer queue |
Awaiting bors try build completion. @rustbot label: +S-waiting-on-perf |
⌛ Trying commit 44ad613 with merge 731282f819829b43dccdc7437b204705b7aec048... |
☀️ Try build successful - checks-actions |
@bors try @rust-timer queue |
Awaiting bors try build completion. @rustbot label: +S-waiting-on-perf |
⌛ Trying commit 44ad613 with merge bcd41abefbf14dc1eec130e985171a28a4ccbcd2... |
☀️ Try build successful - checks-actions |
Queued bcd41abefbf14dc1eec130e985171a28a4ccbcd2 with parent e78e747, future comparison URL. |
Finished benchmarking commit (bcd41abefbf14dc1eec130e985171a28a4ccbcd2): comparison url. Instruction count
Max RSS (memory usage)Results
CyclesResults
If you disagree with this performance assessment, please file an issue in rust-lang/rustc-perf. Benchmarking this pull request likely means that it is perf-sensitive, so we're automatically marking it as not fit for rolling up. While you can manually mark this PR as fit for rollup, we strongly recommend not doing so since this PR may lead to changes in compiler perf. Next Steps: If you can justify the regressions found in this try perf run, please indicate this with @bors rollup=never Footnotes |
Seems like the results here are worse than not doing this; I'm going to go ahead and close but feel free to reopen if needed. |
Some of the previous runs had nice cycle wins, but they were gone in the last round, and this type of PGO made dist builds quite slower, so it was not worth it. |
LLVM has a second mode of PGO, called call-site aware PGO. Let's try if and how it works.
r? @ghost