Skip to content

Add installation notes for Arch Linux users #3607

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

Merged
merged 2 commits into from
Mar 16, 2020
Merged

Add installation notes for Arch Linux users #3607

merged 2 commits into from
Mar 16, 2020

Conversation

murlakatamenka
Copy link
Contributor

Note: 2 of the AUR packages mentioned are maintained by me and are updated rather promptly.

@matklad
Copy link
Member

matklad commented Mar 16, 2020

bors r+

Thanks! And thanks for packaging in the first place :)

bors bot added a commit that referenced this pull request Mar 16, 2020
3607: Add installation notes for Arch Linux users r=matklad a=murlakatamenka

Note: 2 of the AUR packages mentioned are maintained by me and are updated rather promptly.

Co-authored-by: murlakatamenka <[email protected]>
@lnicola
Copy link
Member

lnicola commented Mar 16, 2020

Thanks for packaging it. It might be worth enabling jemalloc, see #3459 and #3531.

@matklad matklad merged commit 648df02 into rust-lang:master Mar 16, 2020
@murlakatamenka
Copy link
Contributor Author

@lnicola shouldn't it be unnecessary? Currently binary is built with just cargo build --release --locked and thus uses default toolchain and system's allocator. I assume people use stable/nightly-x86_64-unknown-linux-gnu by default and use musl target just to compile a generic *nix binary that would fit everyone.

@lnicola
Copy link
Member

lnicola commented Mar 16, 2020

Currently binary is built with just cargo build --release --locked and thus uses default toolchain and system's allocator. I assume people use stable/nightly-x86_64-unknown-linux-gnu by default

Yes, but enabling jemalloc (nb. that's a package feature, not a toolchain) makes it marginally faster (8.7% in my test) and enables memory usage reporting via a custom LSP / Code command. I don't know if that's worth the effort and build time hit, though. We don't try to ship the fastest binary anyway -- that would imply adding codegen-units = 1 and lto = on.

@murlakatamenka
Copy link
Contributor Author

I meant that using jemalloc is more beneficial for musl target (with slower allocator) and questionable for default unkown-linux-gnu one that will be used for building on users' Arch machines (with faster glibc allocator). But now I learned that it may still make sense to use jemalloc for non-musl targets.

Thanks, I'll test the tradeoff of longer compile time / faster binary and see whether using this feature is worth it.

@matklad
Copy link
Member

matklad commented Mar 17, 2020 via email

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.

3 participants