[bug#70178] Request for merging "mesa-updates" branch

[bug#70178] Request for merging "mesa-updates" branch

[John Kehayias via Guix-patches via]

Hi Guix,

I submitted a bunch of patches to mesa-updates a few days ago. From what I can tell things have gone well on x86_64, with most of the "new" failures listed are compared to previous mesa-updates and are ones broken on master too. A few failed from some time issue on hydra 128. And a few were legitimate failures which I have fixed. I'll go through again and make sure I didn't miss anything. I've been using this branch locally for a few things without issues so far.

So...I'm guessing the main thing is to wait for non-x86 coverage. Is there a way to see that on Bordeaux? As always, please report any issues, but hopefully this will be good to go once builds are done.

John

[bug#70178] Request for merging "mesa-updates" branch

[Christopher Baines]

[-- Attachment #1: Type: text/plain, Size: 954 bytes --]

John Kehayias via Guix-patches via <guix-patches@gnu.org> writes:

> I submitted a bunch of patches to mesa-updates a few days ago. From
> what I can tell things have gone well on x86_64, with most of the
> "new" failures listed are compared to previous mesa-updates and are
> ones broken on master too. A few failed from some time issue on hydra
> 128. And a few were legitimate failures which I have fixed. I'll go
> through again and make sure I didn't miss anything. I've been using
> this branch locally for a few things without issues so far.
>
> So...I'm guessing the main thing is to wait for non-x86 coverage. Is
> there a way to see that on Bordeaux? As always, please report any
> issues, but hopefully this will be good to go once builds are done.

The page on QA should show substitute availability [1]. It's probably
going to take a week, probably closer to two to get to a reasonable
level.

1: https://qa.guix.gnu.org/branch/mesa-updates

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 987 bytes --]

[bug#70178] Request for merging "mesa-updates" branch

[Christopher Baines]

[-- Attachment #1: Type: text/plain, Size: 1204 bytes --]

Christopher Baines <mail@cbaines.net> writes:

> John Kehayias via Guix-patches via <guix-patches@gnu.org> writes:
>
>> I submitted a bunch of patches to mesa-updates a few days ago. From
>> what I can tell things have gone well on x86_64, with most of the
>> "new" failures listed are compared to previous mesa-updates and are
>> ones broken on master too. A few failed from some time issue on hydra
>> 128. And a few were legitimate failures which I have fixed. I'll go
>> through again and make sure I didn't miss anything. I've been using
>> this branch locally for a few things without issues so far.
>>
>> So...I'm guessing the main thing is to wait for non-x86 coverage. Is
>> there a way to see that on Bordeaux? As always, please report any
>> issues, but hopefully this will be good to go once builds are done.
>
> The page on QA should show substitute availability [1]. It's probably
> going to take a week, probably closer to two to get to a reasonable
> level.

Substitute availability is getting there, it's just i686-linux which is
lagging behind (which is expected with the limited compute resources
available).

I've just pushed a merge commit to keep the branch up to date with
master.

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 987 bytes --]

bug#70178: Request for merging "mesa-updates" branch

[John Kehayias via Guix-patches via]

Hello,

On Mon, Apr 15, 2024 at 11:32 AM, Christopher Baines wrote:

> Christopher Baines <mail@cbaines.net> writes:
>
>> John Kehayias via Guix-patches via <guix-patches@gnu.org> writes:
>>
>>> I submitted a bunch of patches to mesa-updates a few days ago. From
>>> what I can tell things have gone well on x86_64, with most of the
>>> "new" failures listed are compared to previous mesa-updates and are
>>> ones broken on master too. A few failed from some time issue on hydra
>>> 128. And a few were legitimate failures which I have fixed. I'll go
>>> through again and make sure I didn't miss anything. I've been using
>>> this branch locally for a few things without issues so far.
>>>
>>> So...I'm guessing the main thing is to wait for non-x86 coverage. Is
>>> there a way to see that on Bordeaux? As always, please report any
>>> issues, but hopefully this will be good to go once builds are done.
>>
>> The page on QA should show substitute availability [1]. It's probably
>> going to take a week, probably closer to two to get to a reasonable
>> level.
>
> Substitute availability is getting there, it's just i686-linux which is
> lagging behind (which is expected with the limited compute resources
> available).
>
> I've just pushed a merge commit to keep the branch up to date with
> master.

Thanks for helping out. I merged with
2d5736cc3e869fadd2592cc13a8d332fac63b144. i686 is still a bit lower but
seemed to really slow down or pause, even after I gave it another day.
Anyways, hopefully is fine!

Also, thanks Simon for testing and reporting, apologies for not
responding earlier to your message.

Thanks,
John

[bug#70178] Request for merging "mesa-updates" branch

[Christopher Baines]

[-- Attachment #1: Type: text/plain, Size: 1795 bytes --]

John Kehayias <john.kehayias@protonmail.com> writes:

> Hello,
>
> On Mon, Apr 15, 2024 at 11:32 AM, Christopher Baines wrote:
>
>> Christopher Baines <mail@cbaines.net> writes:
>>
>>> John Kehayias via Guix-patches via <guix-patches@gnu.org> writes:
>>>
>>>> I submitted a bunch of patches to mesa-updates a few days ago. From
>>>> what I can tell things have gone well on x86_64, with most of the
>>>> "new" failures listed are compared to previous mesa-updates and are
>>>> ones broken on master too. A few failed from some time issue on hydra
>>>> 128. And a few were legitimate failures which I have fixed. I'll go
>>>> through again and make sure I didn't miss anything. I've been using
>>>> this branch locally for a few things without issues so far.
>>>>
>>>> So...I'm guessing the main thing is to wait for non-x86 coverage. Is
>>>> there a way to see that on Bordeaux? As always, please report any
>>>> issues, but hopefully this will be good to go once builds are done.
>>>
>>> The page on QA should show substitute availability [1]. It's probably
>>> going to take a week, probably closer to two to get to a reasonable
>>> level.
>>
>> Substitute availability is getting there, it's just i686-linux which is
>> lagging behind (which is expected with the limited compute resources
>> available).
>>
>> I've just pushed a merge commit to keep the branch up to date with
>> master.
>
> Thanks for helping out. I merged with
> 2d5736cc3e869fadd2592cc13a8d332fac63b144. i686 is still a bit lower but
> seemed to really slow down or pause, even after I gave it another day.
> Anyways, hopefully is fine!

Great. As for the i686 builds, this was probably slowed down by the
x86_64 builds taking priority after pushing the master merge, since that
triggered builds for both i686 and x86_64.

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 987 bytes --]

[bug#70178] Request for merging "mesa-updates" branch

[Simon Streit]

Hello John,

John Kehayias via Guix-patches via <guix-patches@gnu.org> writes:

> I submitted a bunch of patches to mesa-updates a few days ago. From
> what I can tell things have gone well on x86_64, with most of the
> "new" failures listed are compared to previous mesa-updates and are
> ones broken on master too. A few failed from some time issue on hydra
> 128. And a few were legitimate failures which I have fixed. I'll go
> through again and make sure I didn't miss anything. I've been using
> this branch locally for a few things without issues so far.

I've rolled out the branch on an old machine and it is looking all good
here.  The current mesa update also fixes an issue where rendering is
broken with GTK4 applications in Gnome on master.


Kind regards,

-- 
Simon