build-configs: rust: disable IBT/Retpoline/Rethunk for the moment

[ojeda]

Until we have the needed pieces in place in rustc and the kernel to support IBT/Retpoline/Rethunk (which will take at least a couple months); disable the relevant configs for the moment.

This avoids the 2000+ warnings from objtool that we currently generate in the logs.

Link: https://storage.kernelci.org/mainline/master/v6.1-11554-g785d21ba2f44/x86_64/x86_64_defconfig+rust/rustc-1.62/logs/build-warnings.log
Link: Rust-for-Linux/linux#945
Signed-off-by: Miguel Ojeda [email protected]

[ojeda]

I tested the fragment locally with scripts/kconfig/merge_config.sh, and I see in the config the kselftest fragment using # CONFIG_DUMMY is not set, so I hope this is the right way to disable them in the KernelCI YAML on top of the defconfig!

[ojeda]

@10ne1 Pinging in case you didn't see it. What should we do to proceed?

[10ne1]

Hi @ojeda Sorry for the delayed response, I'm away from computers / internet due to EoY holidays season, will test this ASAP when I'll be back on 3'rd January. Thanks!

[ojeda]

No rush, Alice may be taking a look too -- Happy New Year! :)

[aliceinwire]

yes, I asked @gctucker or @mgalka to enable rustc-1.66 docker image on kernelci for testing. still waiting for a replay.

[10ne1]

I've tested this locally and it works with both the mainline and rust-for-linux maintainer trees.

I also think this is independent of the docker image bump so is ready to land as is. Thanks!

Reviewed-by: Adrian Ratiu <[email protected]>

[aliceinwire]

Thanks

[aliceinwire]

@ojeda just as a suggestion. I just realized that the link file in the comment is on storage.kernelci.org. files on storage.kernelci.org get cleaned after around one month AFAIR, maybe is better to copy the content to gist.github.com and use that link next time.

[ojeda]

Yeah, avoiding ephemeral links sounds better. Thanks for pointing it out! Perhaps I should have just copied a sample inside the Git commit directly.

Should it be done only for the links in the commit messages, or also for the ones sometimes are put here in GitHub from staging? i.e. is it only for things that end up in the Git repository?

[aliceinwire]

@ojeda KernelCI doesn't have a rule about this. that's why it was only a suggestion. In my opinion commit messages as they are lasting more time, is maybe better if they link something a bit more permanent.

[ojeda]

Understood, thanks!