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
gnumake3: remove #37658
gnumake3: remove #37658
Conversation
Failure on aarch64-linux (full log) Attempted: coq2html, coq_8_3, lensfun, ngrep Partial log (click to expand)
|
Success on x86_64-linux (full log) Attempted: coq2html, coq_8_3, lensfun, ngrep Partial log (click to expand)
|
FYI the glibc update is now in master. |
@GrahamcOfBorg eval |
Failure on aarch64-linux (full log) Attempted: coq2html, coq_8_3, lensfun, ngrep Partial log (click to expand)
|
Success on x86_64-linux (full log) Attempted: coq2html, coq_8_3, lensfun, ngrep Partial log (click to expand)
|
Success on x86_64-linux (full log) Attempted: coq2html, coq_8_3, lensfun, ngrep Partial log (click to expand)
|
Failure on aarch64-linux (full log) Attempted: coq2html, coq_8_3, lensfun, ngrep Partial log (click to expand)
|
@@ -9,8 +9,8 @@ stdenv.mkDerivation rec { | |||
sha256 = "0cfk8jjhs9nbfjfdy98plrj9ayi59aph0nx6ppslgjhlcvacm2xf"; | |||
}; | |||
|
|||
nativeBuildInputs = [ pkgconfig ]; | |||
buildInputs = [ glib zlib libpng cmake gnumake3 ]; | |||
nativeBuildInputs = [ cmake pkgconfig ]; |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Could you add ninja, it is usually faster than make.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I'd think that in the context of Nix, make vs. ninja wouldn't make much difference. Ninja is certainly faster at determining that nothing needs to be done, but when you are building the project for the first time, and Ninja hasn't built up its databases yet, then I'd be surprised if the build time were changed significantly.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I measured it and both ninja and make back-ends seem to average slightly above five seconds on my machine. For larger codebases, though, ninja is significantly faster even when building from scratch according to my experiments. http://voices.canonical.com/jussi.pakkanen/2012/02/28/speed-bumps-hide-in-places-where-you-least-expect-them/ agrees with my observations.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I don't think that individual packages built with CMake should switch to Ninja without a good reason; however I'd mildly approve switching all CMake projects in Nixpkgs to Ninja by default. The upsides of Ninja are possibly faster build, more concise output, better debugability (especially with issues of parallel building) due to the ability to query the build graph, but an irksome downside is that console output (compiler warnings and errors) is not colorized.
There are still a few interesting patches in this PR: 3d637bb2eb9344665ca926fc60a25b7d4a3a6cfb, f82df4b90a0a4299ed24f01140e583b94a084f30, 161521544485292924b1fb0483e81d55e3b76b47 and b08af8c52b731107ffdccff2f294fd54f251c78f. @pbogdan could you please rebase this on master and remove the patches that are no longer relevant, so that we can test & merge? Thanks. |
cc @Zimmi48 coq_8_3 is still maintained right? |
|
Success on aarch64-linux (full log) Attempted: coq2html, lensfun, ngrep Partial log (click to expand)
|
Success on x86_64-linux (full log) Attempted: coq2html, lensfun, ngrep Partial log (click to expand)
|
Motivation for this change
It won't build any more after glibc 2.27 update - https://hydra.nixos.org/build/71662181 (the update is in staging only ATM). Rather than trying to patch make itself I tried building everything that depends on it with the default one and all of them succeeded (
coq_8_3
needed a small patch to itsconfigure
script, I also got rid of redundant parts of the existing patch).As it's more of a clean up that doesn't directly depend on anything in staging I'm opening this against master. Please let me know if it should go into staging instead.
/cc maintainers of affected packages @jwiegley @roconnor @vbgl @bjornfor
Things done
build-use-sandbox
innix.conf
on non-NixOS)nix-shell -p nox --run "nox-review wip"
./result/bin/
)EDIT: build got fixed in 492ca7b, leaving this open in case we still want to get rid of it.