Conversation
nixpkgs-ci
bot
added
10.rebuild-linux: 11-100
|
Seems a bit odd but given we're already doing it for Darwin I guess it doesn't hurt to do it for musl too… |
|
A quick web search does suggest the problem would be fixed with a more recent gnulib, though. Would you be interested in seeing whether there's a gnulib patch we can backport instead? |
|
I did not find any discussions related to GNULIBHEADERS_OVERRIDE_WINT_T and musl. Can you share the results of your search? |
|
I found https://lists.gnu.org/archive/html/help-libidn/2021-07/msg00006.html which mentions a similar problem on Cygwin. Generally I find that musl problems tend to actually be not-Glibc problems, so I'd expect it to have probably been the same fix. |
nixpkgs-ci
bot
added
10.rebuild-linux: 0
|
That discussion is from 2021, and our gnulib is much more recent than that. So I doubt it's the same issue. |
|
But do we use our gnulib here? Usually packages vendor gnulib sources. |
|
You're right, gss uses a vendored gnulib copy. However this copy is also dated August 2022, so after the discussion. It would be nice if we could try building with a more recent gnulib, but it does not seem easy to reproduce the steps done to create the gss release tarball, which determines the gnulib version used. Swapping out the gnulib src is also not easy (at least with the paths I tried). |
Should let us drop the Darwin hack, and not have to apply it on other platfomrs, like musl, too. I used Nixpkgs' gnulib source because it works, but if it stops working in future we could use fetchgit to get the gnulib submodule. Fixes: 90805a4 ("gss: use autoreconfHook") Closes: NixOS#436189
|
Had a go at a proper fix: #444996
|
2 participants
Depends on #436187
Things done
passthru.tests.nixpkgs-reviewon this PR. See nixpkgs-review usage../result/bin/.Add a 👍 reaction to pull requests you find important.