[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: ‘core-updates’ rebased: testing needed!
From: |
Christopher Baines |
Subject: |
Re: ‘core-updates’ rebased: testing needed! |
Date: |
Thu, 27 Jun 2024 13:32:35 +0100 |
User-agent: |
mu4e 1.12.4; emacs 29.3 |
Ludovic Courtès <ludo@gnu.org> writes:
> 宋文武 <iyzsong@envs.net> skribis:
>
>> Hello, there are 399 build failed which only on core-updates if I count
>> it correctly (1 at the end of each line come from ci status, I think 2
>> are for dependencies failed), the 4xxxxxx are build number, which can be
>> used as "https://ci.guix.gnu.org/build/4602867/details".
>
> Thanks for all the work you’ve put last week into fixing these issues!
> We’re at 94% of substitute availability on x86_64-linux right now:
>
> https://qa.guix.gnu.org/branch/core-updates
>
> ci.guix is lagging behind on aarch64-linux (mostly because part of the
> builders are out-of-order right now¹) and also on powerpc64le-linux
> (that would need more investigation), but fortunately bordeaux.guix has
> relatively good figures on these platforms.
>
> On i686-linux, there are two important packages failing: gd (1 test
> failure) and libfaketime. Fixing these can make an important
> difference.
>
> Let’s keep this going and we should be able to merge real soon, maybe as
> soon as we get above 70% on aarch64 and powerpc64le? Is that a
> reasonable goal?
I'm not sure about powerpc64le, I don't think I'll be able to run the
one machine hooked up to the bordeaux build farm in the next few weeks.
signature.asc
Description: PGP signature