2026-04-07 16:52:29 <mps> can someone check build of !100321
2026-04-07 16:53:05 <mps> on loongarch64
2026-04-07 17:44:33 <mps> nvm, i managed to set loongarch64 build env and tested, it pass
2026-04-08 01:50:45 <huajingyun> Thanks:)
2026-04-08 06:48:59 <mps> huajingyun: you are welcome :)
2026-04-10 16:25:04 <omni> qt6-qtwebengine just got disabled on loongarch64 (armv7 too) with the Qt6 6.11.0 upgrade in 76488b70403c7a049a36938de6eeaf0fe6c47694
2026-04-11 01:31:50 <huajingyun> omni: Ah, thanks for the reminder. It seems the patches is no longer applicable to 6.11.0,I'll re-enable Longarch64 when I have time.
2026-04-12 07:36:39 <ncopa> linux-lts 6.12.81 fails to build on loongarch64 (3.22-stable branch)
2026-04-12 07:36:53 <ncopa> https://build.alpinelinux.org/buildlogs/build-3-22-loongarch64/main/linux-lts/linux-lts-6.12.81-r0.log
2026-04-12 07:37:27 <ncopa> .../arch/loongarch/mm/tlb.c:292:33: error: 'exception_handlers' undeclared (first use in this function); did you mean 'exception_table'?
2026-04-12 07:37:34 <ncopa> seems like an upstream problem
2026-04-12 07:38:58 <ncopa> I would appreciate help with finding an upstream patch to solve that. It is blocking the security releases we are planning for tomorrow (Monday)
2026-04-12 10:11:12 <mps> yesterday while upgrading kernel i made my loongarch64 unbootable
2026-04-12 10:12:26 <mps> also linux-lts don't boot, probably i forgot to add nvme to mkinitfs.conf
2026-04-12 15:02:29 <ncopa> commit 055c7e75190e0be43037bd663a3f6aced194416e
2026-04-12 15:02:29 <ncopa> Author: Tiezhu Yang <yangtiezhu@loongson.cn>
2026-04-12 15:02:29 <ncopa>     LoongArch: Handle percpu handler address for ORC unwinder
2026-04-12 15:02:29 <ncopa> Date:   Tue Feb 10 19:31:13 2026 +0800
2026-04-12 15:02:48 <ncopa> was likely not cherry-picked to linux-6.12.y
2026-04-12 15:03:16 <ncopa> but this was:
2026-04-12 15:03:18 <ncopa> commit 0e6f596d6ac635e80bb265d587b2287ef8fa1cd6
2026-04-12 15:03:19 <ncopa> Date:   Tue Feb 10 19:31:14 2026 +0800
2026-04-12 15:03:19 <ncopa> Author: Tiezhu Yang <yangtiezhu@loongson.cn>
2026-04-12 15:03:19 <ncopa>     LoongArch: Remove some extern variables in source files
2026-04-12 15:03:56 <ncopa> so we have to either cherry-pick 055c7e75190e0be43037bd663a3f6aced194416e or revert 0e6f596d6ac635e80bb265d587b2287ef8fa1cd6
2026-04-12 15:04:06 <ncopa> I don't know which loongarch people prefer
2026-04-12 15:06:08 <ncopa> linux-6.12.y has a    Revert "LoongArch: Remove unnecessary checks for ORC unwinder"
2026-04-12 15:06:45 <ncopa> So I assume we need to also revert 0e6f596d6ac635e80bb265d587b2287ef8fa1cd6
2026-04-12 15:08:35 <ncopa> which ends up with a conflict. Ugh....
2026-04-12 18:29:24 <mps> my local kernel linux-edge-dev and alpine linux-stable-dev have wrong /usr/src/linux-headers-*/arch/longarch dir
2026-04-12 18:30:19 <mps> should be /usr/src/linux-headers-*/arch/longarch64
2026-04-12 18:31:49 <mps> probably linux-lts-dev
2026-04-12 18:32:30 <mps> did anyone tried some akms modules to build
2026-04-13 03:21:57 <huajingyun> Thanks ncopa. It seems you've already used the upstream commit 0e6f596d6ac635e80bb265d587b2287ef8fa1cd6 to resolve the error
2026-04-13 03:22:23 <huajingyun> I just spoke with the relevant personnel, and today they submitted two commits for upstream Linux 6.12 as fixes
2026-04-13 03:22:29 <huajingyun> https://lore.kernel.org/loongarch/20260413023627.1363488-1-chenhuacai@loongson.cn/T/#u
2026-04-13 03:22:30 <huajingyun> https://lore.kernel.org/loongarch/20260413023647.1368051-1-chenhuacai@loongson.cn/T/#u
2026-04-13 03:31:00 <huajingyun> since they haven't been merged yet, could we apply these two patches to alpine first?
2026-04-13 07:20:01 <huajingyun> based on past patterns, is the 3.24 builder soon to be online?
2026-04-13 07:56:17 <achill> soon, yes. just waiting for ncopa to set it up
2026-04-13 08:05:01 <huajingyun> achill:  thank you, good to know:)
2026-04-13 08:15:55 <ncopa> hi
2026-04-13 08:16:01 <ncopa> yes im late with the 3.24 builders
2026-04-13 08:16:40 <ncopa> I did this for loongarch64: https://git.alpinelinux.org/aports/commit/?h=3.22-stable
2026-04-13 08:19:41 <ncopa> The commit 4cd641a79e69 ("LoongArch: Remove unnecessary checks for ORC unwinder") was reverted in linux-6.12.y:
2026-04-13 08:19:52 <ncopa> https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/log/arch/loongarch?h=v6.12.81
2026-04-13 09:10:30 <huajingyun> ncopa: okay I see, thank you for these links
2026-04-13 09:22:15 <ncopa> can you check what I did in 3.22-stable and 3.21-stable is ok?
2026-04-13 09:56:00 <DWwanghao> ncopa: I'm building the 3.22-stable and 3.21-stable ISO respectively to check if they boot properly. I'll report back once they're done
2026-04-13 10:01:12 <ncopa> thanks!
2026-04-13 10:01:35 <ncopa> I think you can just test boot the built packages from v3.22 and v3.21
2026-04-13 10:01:44 <ncopa> you dont need to rebuild yrouself
2026-04-13 10:01:55 <ncopa> i fixed the build yesterday
2026-04-13 10:03:56 <huajingyun> Yes, we are creating iso
2026-04-13 10:10:36 <huajingyun> ncopa: by the way, I also discussed https://git.alpinelinux.org/aports/commit/?h=3.22-stable with Tiezhu Yang, the submitter of the commit ("LoongArch: Remove unnecessary checks for ORC unwinder"),He also said it was ok
2026-04-13 10:46:48 <huajingyun> ncopa: It has been verified that 3.22-stable and 3.21-stable, based on Linux-lts 6.12.81, both work correctly
2026-04-13 10:59:31 <ncopa> very good. Thanks!
2026-04-13 12:15:27 <ncopa> tree-sitter fails to build on loongarch64 for some reason. I'm not too familiar with rust so I have no clue how to solve it. https://gitlab.alpinelinux.org/alpine/aports/-/jobs/2303970
2026-04-13 12:59:15 <omni> it may be fixed in a later version of the rustix crate
2026-04-13 13:18:02 <omni> I left comments on the MR, take copies of the Cargo.{toml,lock} files, to diff patches from, unless you're working in the tree-sitter git repository
2026-04-13 13:20:39 <omni> we need to do this all the time for exotic architectures that aren't x86_64 or aarch64, rust is very portable like that
2026-04-14 01:57:28 <huajingyun> ncopa: you are welcome :)
2026-04-14 13:14:28 <mps> huh, lost video out on upgrade kernel to 7.0, loongarch64
2026-04-14 17:27:58 <mps> here is dmesg out of video problem with 6.19 kernel and up https://tpaste.us/1yrk
2026-04-14 17:28:33 <mps> on 6.18 and previous it works fine
2026-04-14 19:39:03 <mps> found fix, build amd gpu as module and not in-kernel
2026-04-14 20:20:30 <mps> 7.0 work nice on loongarch64 \o/
2026-04-17 07:43:02 <huajingyun> fossdd: hi, I just created !426710, which is a fix from linux-tools for loongarch64, could you please take a look?  thanks
2026-04-17 07:43:10 <huajingyun> It is now blocking edge builder
2026-04-17 08:29:01 <ncopa> i merged it. Thank you!
2026-04-17 08:29:51 <ncopa> I really appreciate that you  keep an eye on the build status, and fix issues promptly
2026-04-17 08:39:14 <huajingyun> ncopa: you're welcome,thank you for the quick merge!  glad to help:)
2026-04-17 08:39:32 <huajingyun> I see that 3.24 builders is already online, but besides longarch64, there should also be ppc64le
2026-04-17 08:53:20 <ncopa> yes, still working no those
2026-04-17 08:54:03 <ncopa> and s390x
2026-04-17 08:55:42 <ncopa> build-3-24-loongarch64 is online now
2026-04-17 08:55:50 <ncopa> only s390x left
2026-04-17 08:59:05 <huajingyun> That's great! thank you for your work
2026-04-17 09:22:55 <huajingyun> we will also continue to keep an eye on the builder status, hoping to help fix some blockers in time (if some aports break the build)
2026-04-17 09:23:05 <huajingyun> looking forward to seeing how the build progress of the 3.24 builders will be after this weekend