Gather PGO profiles in parallel#159524
Conversation
|
@bors try jobs=dist-x86_64-linux |
This comment has been minimized.
This comment has been minimized.
Gather PGO profiles in parallel try-job: dist-x86_64-linux
70e86de to
6d17554
Compare
|
@bors try jobs=dist-x86_64-linux |
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
Gather PGO profiles in parallel try-job: dist-x86_64-linux
This comment has been minimized.
This comment has been minimized.
|
Finished benchmarking commit (e87c5cd): comparison URL. Overall result: no relevant changes - no action neededBenchmarking means the PR may be perf-sensitive. Consider adding rollup=never if this change is not fit for rolling up. @rustbot label: -S-waiting-on-perf -perf-regression Instruction countThis perf run didn't have relevant results for this metric. Max RSS (memory usage)This perf run didn't have relevant results for this metric. CyclesResults (secondary 25.1%)A less reliable metric. May be of interest, but not used to determine the overall result above.
Binary sizeThis perf run didn't have relevant results for this metric. Bootstrap: 485.656s -> 485.33s (-0.07%) |
|
r? @jieyouxu rustbot has assigned @jieyouxu. Use Why was this reviewer chosen?The reviewer was selected based on:
|
This comment has been minimized.
This comment has been minimized.
|
This makes the job faster by about 20 minutes 🎉 |
6d17554 to
c373810
Compare
|
@bors r+ |
…uwer Rollup of 6 pull requests Successful merges: - #159524 (Gather PGO profiles in parallel) - #158850 (Avoid underflow for splatted rust-call receivers) - #159215 ([compiletest] Differentiate between LLVM LLDB and Apple LLDB) - #159347 (Document rustc_public lib.rs and update README) - #159469 (Reorganize `tests/ui/issues` [27/N]) - #159512 (Add regression tests for fixed dead-code issues)
Rollup merge of #159524 - Kobzol:opt-dist-parallel, r=jieyouxu Gather PGO profiles in parallel This PR modifies `opt-dist` to: 1. Gather PGO profiles in parallel, with 4 benchmarks executed at once. To improve CPU utilization. 2. Optimize both LLVM and rustc with BOLT in parallel, because this takes ~4 minutes and is mostly sequential.
This PR modifies
opt-distto: