Skip to content

Update cc for remaining lockfiles #2067

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

Closed
wants to merge 1 commit into from

Conversation

ognevny
Copy link
Contributor

@ognevny ognevny commented Apr 11, 2025

I forgot to update cc in compile-benchmarks crates. when I tried to do so, I found that not all packages got rquired version of cc (>=1.1.10), so I just blibfly ran cargo update for all packages to guarantee fresh version of cc...

@ognevny ognevny force-pushed the update-internal-lockfile branch from 489f035 to e47d49a Compare April 11, 2025 21:09
@ognevny
Copy link
Contributor Author

ognevny commented Apr 11, 2025

this is a very huge diff. I'll try to update crates for each package manually

@ognevny ognevny force-pushed the update-internal-lockfile branch from e47d49a to 65cd677 Compare April 11, 2025 21:21
@ognevny
Copy link
Contributor Author

ognevny commented Apr 11, 2025

@Kobzol sorry for disturbing with my attempts on making opt-dist work for specific host. I tried to minimize diff here, but I didn't manage to win cargo in fight with cc update for cargo-0.60.0 (it updates cc to >=v1.1.10 only with full lockfile update)

edit: done, minimized the diff :)

@ognevny ognevny changed the title Update all lockfiles in collector package Update some lockfiles in collector/compile-benchmarks package Apr 11, 2025
@ognevny ognevny force-pushed the update-internal-lockfile branch from 65cd677 to 791012b Compare April 12, 2025 04:44
@ognevny ognevny changed the title Update some lockfiles in collector/compile-benchmarks package Update cc for remaining lockfiles Apr 12, 2025
@Kobzol
Copy link
Contributor

Kobzol commented Apr 12, 2025

Hmm, I have much bigger issue with this than the previous PR :) We only update the compile benchmarks if they stop compiling on our supported targets (which is right pretty much only x64 Linux), to keep benchmark stability.

I don't want to do these global lockfile updates for the benchmarks in a single PR.

That being said, we are currently just before a big benchmark update that will update most of the benchmarks to newer versions, which should also bump cc. So after that update the situation might be better for your target. The update will take at least several weeks though (and someone actually has to start doing it..).

#2024

@ognevny
Copy link
Contributor Author

ognevny commented Apr 12, 2025

understood. I'll use my patch downstream then. waiting for updates

@ognevny ognevny force-pushed the update-internal-lockfile branch from 791012b to 307ecf2 Compare April 12, 2025 07:34
@ognevny
Copy link
Contributor Author

ognevny commented Apr 14, 2025

by the way, when you open a PR with compile-benchmark, please mork your PR with closes #2067 so I'll get a notification when your PR is merged :)

@Kobzol
Copy link
Contributor

Kobzol commented Apr 14, 2025

There will probably be ~50 of such PRs (we need on PR to add a new benchmark and then later to remove the old one). I created a benchmark-update label to track this.

@ognevny
Copy link
Contributor Author

ognevny commented May 15, 2025

@Kobzol it seems that some outdated versions are still there :(

@Kobzol
Copy link
Contributor

Kobzol commented May 15, 2025

Could you please post in which benchmarks is the outdated cc version?

@ognevny
Copy link
Contributor Author

ognevny commented May 15, 2025

@ognevny ognevny closed this May 15, 2025
@ognevny ognevny force-pushed the update-internal-lockfile branch from 307ecf2 to f265cc6 Compare May 15, 2025 18:55
@ognevny
Copy link
Contributor Author

ognevny commented May 15, 2025

wait, how I closed it...

edit: now I can't change this branch anymore

@Kobzol
Copy link
Contributor

Kobzol commented May 20, 2025

These two are "stable" benchmarks that we are keeping ideally unchanged for a long time, they are only used for benchmarking stable/beta releases and should not be used for PR or local benchmarks. Why do you need them to work?

@Kobzol Kobzol reopened this May 20, 2025
@ognevny
Copy link
Contributor Author

ognevny commented May 20, 2025

These two are "stable" benchmarks that we are keeping ideally unchanged for a long time, they are only used for benchmarking stable/beta releases and should not be used for PR or local benchmarks. Why do you need them to work?

IIUC these benchmarks are not used with opt-dist? in this case I'll close PR

@Kobzol
Copy link
Contributor

Kobzol commented May 20, 2025

Ah, no, they shouldn't be used for opt-dist.

@ognevny
Copy link
Contributor Author

ognevny commented May 20, 2025

Ah, no, they shouldn't be used for opt-dist.

I don't have a run for current state, but syn seems to be used

@Kobzol
Copy link
Contributor

Kobzol commented May 20, 2025

https://github.com/rust-lang/rust/blob/98fdcaed50dfc5346263ed40e9eacb1fc5c124ec/src/build_helper/src/lib.rs#L12-L11 this is the list of benchmarks that opt-dist currently uses. Note that syn is different from syn-1.0.89.

@ognevny
Copy link
Contributor Author

ognevny commented May 20, 2025

https://github.com/rust-lang/rust/blob/98fdcaed50dfc5346263ed40e9eacb1fc5c124ec/src/build_helper/src/lib.rs#L12-L11 this is the list of benchmarks that opt-dist currently uses. Note that syn is different from syn-1.0.89.

so this list will be changed to include newer syn?

@Kobzol
Copy link
Contributor

Kobzol commented May 20, 2025

Yeah, it's in my TODO list :)

@ognevny
Copy link
Contributor Author

ognevny commented May 20, 2025

thanks for clarification!

@ognevny ognevny closed this May 20, 2025
@ognevny ognevny deleted the update-internal-lockfile branch May 20, 2025 08:40
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants