Skip unnecessary components in x64 try builds #142963

Kobzol · 2025-06-24T13:57:39Z

We unnecessarily rebuild wasm-component-ld, llvm-bitcode-linker and Cranelift during the intermediate PGO builds several times times, which is unnecessarily and increases the duration of try builds. This PR also disables some unnecessary dist components.

r? @jieyouxu

rustbot · 2025-06-24T13:57:43Z

jieyouxu is not on the review rotation at the moment.
They may take a while to respond.

Kobzol · 2025-06-24T13:57:44Z

@bors2 try

rustbot · 2025-06-24T13:57:45Z

Some changes occurred in src/tools/opt-dist

cc @Kobzol

rust-bors · 2025-06-24T13:57:49Z

⌛ Trying commit ced1523 with merge e1cdefe…

To cancel the try build, run the command @bors2 try cancel.

Skip unnecessary components in x64 try builds We unnecessarily rebuild `wasm-component-ld`, `llvm-bitcode-linker` and Cranelift during the intermediate PGO builds several times times, which is unnecessarily and increases the duration of try builds. This PR also disables some unnecessary dist components. r? `@jieyouxu`

Kobzol · 2025-06-24T13:58:57Z

Recent try builds were ~2.5h.

Kobzol · 2025-06-24T14:05:05Z

@bors2 try

rust-bors · 2025-06-24T14:05:09Z

⌛ Trying commit 76a8080 with merge 936a203…

(The previously running try build was automatically cancelled.)

To cancel the try build, run the command @bors2 try cancel.

Skip unnecessary components in x64 try builds We unnecessarily rebuild `wasm-component-ld`, `llvm-bitcode-linker` and Cranelift during the intermediate PGO builds several times times, which is unnecessarily and increases the duration of try builds. This PR also disables some unnecessary dist components. r? `@jieyouxu`

jieyouxu · 2025-06-24T14:38:02Z

src/tools/opt-dist/src/main.rs

        for target in [
            "rust-docs",
            "rustc-docs",
+            "rustc-dev",


Question: to double-check, the "fast try build" is where the user never specified an actual job matcher, right?

Yeah, the try builds most often used for perf. runs. The are so slow now :( A few years ago after my optimizations they were barely 1 hour, but now it's 2.5h :( Just BOLT alone takes >1h, it writes 100 GiB of profiles to disk, lol.

jieyouxu

Thanks. Feel free to r=me once the try job comes back okay.

Kobzol · 2025-06-24T15:25:59Z

@rust-timer queue

rust-bors · 2025-06-24T17:05:58Z

☀️ Try build successful (CI)
Build commit: 936a203 (936a203ffd25c86bc604222fb47a7d3f0e429bd5, parent: 36b21637e93b038453924d3c66821089e71d8baa)

Kobzol · 2025-06-24T17:07:54Z

(Starting another build to time it with Docker cache primed).

@bors2 try

rust-bors · 2025-06-24T17:07:58Z

⌛ Trying commit 76a8080 with merge d732d09…

To cancel the try build, run the command @bors2 try cancel.

Skip unnecessary components in x64 try builds We unnecessarily rebuild `wasm-component-ld`, `llvm-bitcode-linker` and Cranelift during the intermediate PGO builds several times times, which is unnecessarily and increases the duration of try builds. This PR also disables some unnecessary dist components. r? `@jieyouxu`

rust-bors · 2025-06-24T19:13:32Z

☀️ Try build successful (CI)
Build commit: d732d09 (d732d09fce8a79b588d90ae46965cc2435c617f5, parent: 36b21637e93b038453924d3c66821089e71d8baa)

Kobzol · 2025-06-24T19:40:02Z

Ok, 2h. That's still quite slow, but much better than 2.5 hours before.

Kobzol · 2025-06-25T14:06:09Z

Dang, I skipped too much and it broke perf. We still need to build Cargo.

@bors2 try

rust-timer · 2025-06-26T12:24:48Z

Finished benchmarking commit (bdcbdb5): comparison URL.

Overall result: no relevant changes - no action needed

Benchmarking this pull request means it may be perf-sensitive – we'll automatically label it not fit for rolling up. You can override this, but we strongly advise not to, due to possible changes in compiler perf.

@bors rollup=never
@rustbot label: -S-waiting-on-perf -perf-regression

Instruction count

This benchmark run did not return any relevant results for this metric.

Max RSS (memory usage)

Results (primary 1.8%, secondary 4.2%)

A less reliable metric. May be of interest, but not used to determine the overall result above.

	mean	range	count
Regressions ❌ (primary)	1.8%	[1.8%, 1.8%]	1
Regressions ❌ (secondary)	4.2%	[2.7%, 5.8%]	2
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	1.8%	[1.8%, 1.8%]	1

Cycles

This benchmark run did not return any relevant results for this metric.

Binary size

This benchmark run did not return any relevant results for this metric.

Bootstrap: 690.11s -> 688.094s (-0.29%)
Artifact size: 372.15 MiB -> 372.11 MiB (-0.01%)

workingjubilee · 2025-06-27T05:12:27Z

Everyone takes another turn

@bors p=1

compiler-errors · 2025-06-27T16:59:23Z

This doesn't affect perf and @Kobzol marked it as rollup, let's just roll it up with the rest of the PRs.

@bors rollup=always p=0

Skip unnecessary components in x64 try builds We unnecessarily rebuild `wasm-component-ld`, `llvm-bitcode-linker` and Cranelift during the intermediate PGO builds several times times, which is unnecessarily and increases the duration of try builds. This PR also disables some unnecessary dist components. r? `@jieyouxu`

…rors Rollup of 8 pull requests Successful merges: - #139858 (New const traits syntax) - #140809 (Reduce special casing for the panic runtime) - #142963 (Skip unnecessary components in x64 try builds) - #142974 (Update stage0 to 1.89.0-beta.1) - #142987 (rustdoc: show attributes on enum variants) - #143002 (tests: Require `run-fail` ui tests to have an exit code (`SIGABRT` not ok)) - #143092 (const checks for lifetime-extended temporaries: avoid 'top-level scope' terminology) - #143096 (tag_for_variant: properly pass TypingEnv) r? `@ghost` `@rustbot` modify labels: rollup

Skip unnecessary components in x64 try builds We unnecessarily rebuild `wasm-component-ld`, `llvm-bitcode-linker` and Cranelift during the intermediate PGO builds several times times, which is unnecessarily and increases the duration of try builds. This PR also disables some unnecessary dist components. r? `@jieyouxu`

Skip unnecessary components in x64 try builds We unnecessarily rebuild `wasm-component-ld`, `llvm-bitcode-linker` and Cranelift during the intermediate PGO builds several times times, which is unnecessarily and increases the duration of try builds. This PR also disables some unnecessary dist components. r? ``@jieyouxu``

Skip unnecessary components in x64 try builds We unnecessarily rebuild `wasm-component-ld`, `llvm-bitcode-linker` and Cranelift during the intermediate PGO builds several times times, which is unnecessarily and increases the duration of try builds. This PR also disables some unnecessary dist components. r? ```@jieyouxu```

Rollup of 10 pull requests Successful merges: - #123476 (std::net: adding `unix_socket_exclbind` feature for solaris/illumos.) - #142708 (Do not include NUL-terminator in computed length) - #142963 (Skip unnecessary components in x64 try builds) - #142987 (rustdoc: show attributes on enum variants) - #143031 (Add windows-gnullvm hosts to the manifest) - #143082 (update internal `send_signal` comment) - #143110 (Use tidy to sort `sym::*` items) - #143111 (BTreeSet: remove duplicated code by reusing `from_sorted_iter`) - #143114 (Minor Documentation Improvements) - #143137 (Miri subtree update) r? `@ghost` `@rustbot` modify labels: rollup

Skip unnecessary components in x64 try builds We unnecessarily rebuild `wasm-component-ld`, `llvm-bitcode-linker` and Cranelift during the intermediate PGO builds several times times, which is unnecessarily and increases the duration of try builds. This PR also disables some unnecessary dist components. r? ````@jieyouxu````

Rollup of 10 pull requests Successful merges: - #123476 (std::net: adding `unix_socket_exclbind` feature for solaris/illumos.) - #142708 (Do not include NUL-terminator in computed length) - #142963 (Skip unnecessary components in x64 try builds) - #142974 (Update stage0 to 1.89.0-beta.1) - #142987 (rustdoc: show attributes on enum variants) - #143031 (Add windows-gnullvm hosts to the manifest) - #143082 (update internal `send_signal` comment) - #143110 (Use tidy to sort `sym::*` items) - #143111 (BTreeSet: remove duplicated code by reusing `from_sorted_iter`) - #143114 (Minor Documentation Improvements) r? `@ghost` `@rustbot` modify labels: rollup

Rollup of 9 pull requests Successful merges: - #123476 (std::net: adding `unix_socket_exclbind` feature for solaris/illumos.) - #142708 (Do not include NUL-terminator in computed length) - #142963 (Skip unnecessary components in x64 try builds) - #142987 (rustdoc: show attributes on enum variants) - #143031 (Add windows-gnullvm hosts to the manifest) - #143082 (update internal `send_signal` comment) - #143110 (Use tidy to sort `sym::*` items) - #143111 (BTreeSet: remove duplicated code by reusing `from_sorted_iter`) - #143114 (Minor Documentation Improvements) r? `@ghost` `@rustbot` modify labels: rollup try-job: dist-i586-gnu-i586-i686-musl

Rollup merge of #142963 - Kobzol:try-build-skip, r=jieyouxu Skip unnecessary components in x64 try builds We unnecessarily rebuild `wasm-component-ld`, `llvm-bitcode-linker` and Cranelift during the intermediate PGO builds several times times, which is unnecessarily and increases the duration of try builds. This PR also disables some unnecessary dist components. r? `````@jieyouxu`````

Rollup of 9 pull requests Successful merges: - rust-lang/rust#123476 (std::net: adding `unix_socket_exclbind` feature for solaris/illumos.) - rust-lang/rust#142708 (Do not include NUL-terminator in computed length) - rust-lang/rust#142963 (Skip unnecessary components in x64 try builds) - rust-lang/rust#142987 (rustdoc: show attributes on enum variants) - rust-lang/rust#143031 (Add windows-gnullvm hosts to the manifest) - rust-lang/rust#143082 (update internal `send_signal` comment) - rust-lang/rust#143110 (Use tidy to sort `sym::*` items) - rust-lang/rust#143111 (BTreeSet: remove duplicated code by reusing `from_sorted_iter`) - rust-lang/rust#143114 (Minor Documentation Improvements) r? `@ghost` `@rustbot` modify labels: rollup try-job: dist-i586-gnu-i586-i686-musl

Ship LLVM (`rust-dev`) in fast try builds again rust-lang#142963 stopped shipping `rust-dev` in fast try builds, which was not actually correct, because if a PR modifies the LLVM submodule, we should provide the prebuilt LLVM for rustc-perf even in a (fast) try build. So why didn't we find out about this earlier? Well, because soon before that PR landed, we started using new bors for try builds, which actually broke bootstrap's git change detection logic, because new bors used a different e-mail address for merge commits. So the two bugs kind of masked each other out. Recently, we fixed the e-mail address and git change detection with new bors, since it's now also used for auto builds, but that in turn broke fast try builds that modify LLVM (rust-lang#150722 (comment)), because rustc-perf saw that LLVM was modified in the try build, and it (correctly!) tried to download `rust-dev` for the given try build commit, but that was (incorrectly!) missing on CI, due to being skipped. This PR restored building the `rust-dev` component in fast try builds. In theory, we could use the same git detection logic that bootstrap uses and only do this if LLVM is actually modified in the given commit, but I'd rather do the correct thing here, than introduce additional opportunities for the git detection to desync.

Rollup merge of #151196 - try-builds-llvm, r=jieyouxu Ship LLVM (`rust-dev`) in fast try builds again #142963 stopped shipping `rust-dev` in fast try builds, which was not actually correct, because if a PR modifies the LLVM submodule, we should provide the prebuilt LLVM for rustc-perf even in a (fast) try build. So why didn't we find out about this earlier? Well, because soon before that PR landed, we started using new bors for try builds, which actually broke bootstrap's git change detection logic, because new bors used a different e-mail address for merge commits. So the two bugs kind of masked each other out. Recently, we fixed the e-mail address and git change detection with new bors, since it's now also used for auto builds, but that in turn broke fast try builds that modify LLVM (#150722 (comment)), because rustc-perf saw that LLVM was modified in the try build, and it (correctly!) tried to download `rust-dev` for the given try build commit, but that was (incorrectly!) missing on CI, due to being skipped. This PR restored building the `rust-dev` component in fast try builds. In theory, we could use the same git detection logic that bootstrap uses and only do this if LLVM is actually modified in the given commit, but I'd rather do the correct thing here, than introduce additional opportunities for the git detection to desync.

rustbot assigned jieyouxu Jun 24, 2025

rustbot added the S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. label Jun 24, 2025

rustbot added A-testsuite Area: The testsuite used to check the correctness of rustc T-infra Relevant to the infrastructure team, which will review and decide on the PR/issue. labels Jun 24, 2025

jieyouxu reviewed Jun 24, 2025

View reviewed changes

jieyouxu approved these changes Jun 24, 2025

View reviewed changes

This comment has been minimized.

Sign in to view

rustbot added the S-waiting-on-perf Status: Waiting on a perf run to be completed. label Jun 24, 2025

This comment has been minimized.

Sign in to view

Kobzol added 3 commits June 25, 2025 16:05

Skip unnecessary components in x64 try builds

e0f8e86

Skip more dist components

25dee4e

Do not build GCC in fast try builds

8034154

Kobzol force-pushed the try-build-skip branch from 76a8080 to 8034154 Compare June 25, 2025 14:05

bors added the S-waiting-on-bors Status: Waiting on bors to run and complete tests. Bors will change the label on completion. label Jun 26, 2025

rustbot removed the S-waiting-on-perf Status: Waiting on a perf run to be completed. label Jun 26, 2025

compiler-errors mentioned this pull request Jun 27, 2025

Rollup of 8 pull requests #143105

Closed

matthiaskrgr mentioned this pull request Jun 28, 2025

Rollup of 10 pull requests #143142

Closed

matthiaskrgr mentioned this pull request Jun 28, 2025

Rollup of 9 pull requests #143147

Closed

matthiaskrgr mentioned this pull request Jun 28, 2025

Rollup of 10 pull requests #143148

Closed

matthiaskrgr mentioned this pull request Jun 28, 2025

Rollup of 10 pull requests #143153

Closed

matthiaskrgr mentioned this pull request Jun 28, 2025

Rollup of 9 pull requests #143157

Merged

bors merged commit 5e725a6 into rust-lang:master Jun 29, 2025
10 checks passed

rustbot added this to the 1.90.0 milestone Jun 29, 2025

Kobzol deleted the try-build-skip branch June 29, 2025 06:11

This was referenced Jan 16, 2026

Update to LLVM 22 #150722

Open

Ship LLVM (rust-dev) in fast try builds again #151196

Merged

Uh oh!

Skip unnecessary components in x64 try builds #142963

Skip unnecessary components in x64 try builds #142963

Conversation

Kobzol commented Jun 24, 2025

Uh oh!

rustbot commented Jun 24, 2025

Uh oh!

Kobzol commented Jun 24, 2025

Uh oh!

rustbot commented Jun 24, 2025

Uh oh!

rust-bors bot commented Jun 24, 2025

Uh oh!

Kobzol commented Jun 24, 2025

Uh oh!

Kobzol commented Jun 24, 2025

Uh oh!

rust-bors bot commented Jun 24, 2025

Uh oh!

jieyouxu Jun 24, 2025

Choose a reason for hiding this comment

Uh oh!

Kobzol Jun 24, 2025

Choose a reason for hiding this comment

Uh oh!

jieyouxu left a comment

Choose a reason for hiding this comment

Uh oh!

Kobzol commented Jun 24, 2025

Uh oh!

This comment has been minimized.

rust-bors bot commented Jun 24, 2025

Uh oh!

This comment has been minimized.

Kobzol commented Jun 24, 2025

Uh oh!

rust-bors bot commented Jun 24, 2025

Uh oh!

rust-bors bot commented Jun 24, 2025

Uh oh!

Kobzol commented Jun 24, 2025

Uh oh!

Kobzol commented Jun 25, 2025

Uh oh!

rust-timer commented Jun 26, 2025

Overall result: no relevant changes - no action needed

Instruction count

Max RSS (memory usage)

Cycles

Binary size

Uh oh!

workingjubilee commented Jun 27, 2025

Uh oh!

compiler-errors commented Jun 27, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

7 participants

compiler-errors commented Jun 27, 2025 •

edited

Loading