GNU bug report logs -
#74780
Request for merging "mesa-updates" branch
Previous Next
To add a comment to this bug, you must first unarchive it, by sending
a message to control AT debbugs.gnu.org, with unarchive 74780 in the body.
You can then email your comments to 74780 AT debbugs.gnu.org in the normal way.
Toggle the display of automated, internal messages from the tracker.
Report forwarded
to
guix-patches <at> gnu.org
:
bug#74780
; Package
guix-patches
.
(Wed, 11 Dec 2024 06:34:02 GMT)
Full text and
rfc822 format available.
Acknowledgement sent
to
John Kehayias <john.kehayias <at> protonmail.com>
:
New bug report received and forwarded. Copy sent to
guix-patches <at> gnu.org
.
(Wed, 11 Dec 2024 06:34:02 GMT)
Full text and
rfc822 format available.
Message #5 received at submit <at> debbugs.gnu.org (full text, mbox):
Hi Guix,
Currently the mesa-updates branch just has a few updates which seem to have built fine (on x86_64 at least), see <https://issues.guix.gnu.org/74779>.
I don't know of other pending changes other than a libva update I'll see about. Perhaps we can do some ungrafting too?
John
Information forwarded
to
guix-patches <at> gnu.org
:
bug#74780
; Package
guix-patches
.
(Fri, 20 Dec 2024 20:33:02 GMT)
Full text and
rfc822 format available.
Message #8 received at 74780 <at> debbugs.gnu.org (full text, mbox):
On Wed, Dec 11, 2024 at 06:30 AM, John Kehayias wrote:
> Hi Guix,
>
> Currently the mesa-updates branch just has a few updates which seem to
> have built fine (on x86_64 at least), see
> <https://issues.guix.gnu.org/74779>.
>
> I don't know of other pending changes other than a libva update I'll
> see about. Perhaps we can do some ungrafting too?
>
> John
(CC-ing Efraim after the patch to build the asahi vulkan driver on
aarch64-linux)
I just rebased on master after further updating wayland-protocols and
mesa to their very freshest versions.
Not much else going on here, so maybe just let this build and then merge
to keep things current. Unfortunately the QA page hasn't shown a status
for this branch but I'll check later.
There are discussions about libglvnd support in Mesa I need to chime in
on, so we could tackle that, but not until the new year for me. That
will require a decent number of changes and lots more testing, so my
preference would be to merge this branch when substitutes are available
and then immediately work on libglvnd or whatever else is pending.
John
Information forwarded
to
guix-patches <at> gnu.org
:
bug#74780
; Package
guix-patches
.
(Sat, 21 Dec 2024 16:34:02 GMT)
Full text and
rfc822 format available.
Message #11 received at 74780 <at> debbugs.gnu.org (full text, mbox):
[Message part 1 (text/plain, inline)]
On Fri, Dec 20, 2024 at 08:31:57PM +0000, John Kehayias wrote:
> On Wed, Dec 11, 2024 at 06:30 AM, John Kehayias wrote:
>
> > Hi Guix,
> >
> > Currently the mesa-updates branch just has a few updates which seem to
> > have built fine (on x86_64 at least), see
> > <https://issues.guix.gnu.org/74779>.
> >
> > I don't know of other pending changes other than a libva update I'll
> > see about. Perhaps we can do some ungrafting too?
> >
> > John
>
> (CC-ing Efraim after the patch to build the asahi vulkan driver on
> aarch64-linux)
I didn't test it (I don't have the hardware), but it built just fine and
I assume the Asahi people know what they're doing.
> I just rebased on master after further updating wayland-protocols and
> mesa to their very freshest versions.
I'll rebuild again on a couple of architectures but I don't expect any
issues.
> Not much else going on here, so maybe just let this build and then merge
> to keep things current. Unfortunately the QA page hasn't shown a status
> for this branch but I'll check later.
>
> There are discussions about libglvnd support in Mesa I need to chime in
> on, so we could tackle that, but not until the new year for me. That
> will require a decent number of changes and lots more testing, so my
> preference would be to merge this branch when substitutes are available
> and then immediately work on libglvnd or whatever else is pending.
That makes sense. Regular smallish updates are one of the reasons for
the team/topic -based branches.
--
Efraim Flashner <efraim <at> flashner.co.il> אפרים פלשנר
GPG key = A28B F40C 3E55 1372 662D 14F7 41AA E7DC CA3D 8351
Confidentiality cannot be guaranteed on emails sent or received unencrypted
[signature.asc (application/pgp-signature, inline)]
Information forwarded
to
guix-patches <at> gnu.org
:
bug#74780
; Package
guix-patches
.
(Sun, 22 Dec 2024 09:21:02 GMT)
Full text and
rfc822 format available.
Message #14 received at 74780 <at> debbugs.gnu.org (full text, mbox):
[Message part 1 (text/plain, inline)]
On Fri, Dec 20, 2024 at 08:31:57PM +0000, John Kehayias wrote:
> On Wed, Dec 11, 2024 at 06:30 AM, John Kehayias wrote:
>
> > Hi Guix,
> >
> > Currently the mesa-updates branch just has a few updates which seem to
> > have built fine (on x86_64 at least), see
> > <https://issues.guix.gnu.org/74779>.
> >
> > I don't know of other pending changes other than a libva update I'll
> > see about. Perhaps we can do some ungrafting too?
> >
> > John
>
> (CC-ing Efraim after the patch to build the asahi vulkan driver on
> aarch64-linux)
>
> I just rebased on master after further updating wayland-protocols and
> mesa to their very freshest versions.
>
> Not much else going on here, so maybe just let this build and then merge
> to keep things current. Unfortunately the QA page hasn't shown a status
> for this branch but I'll check later.
>
> There are discussions about libglvnd support in Mesa I need to chime in
> on, so we could tackle that, but not until the new year for me. That
> will require a decent number of changes and lots more testing, so my
> preference would be to merge this branch when substitutes are available
> and then immediately work on libglvnd or whatever else is pending.
I built mesa for aarch64, armhf, riscv64 and powerpc64le. They all
built without any problems.
--
Efraim Flashner <efraim <at> flashner.co.il> אפרים פלשנר
GPG key = A28B F40C 3E55 1372 662D 14F7 41AA E7DC CA3D 8351
Confidentiality cannot be guaranteed on emails sent or received unencrypted
[signature.asc (application/pgp-signature, inline)]
Information forwarded
to
guix-patches <at> gnu.org
:
bug#74780
; Package
guix-patches
.
(Sun, 29 Dec 2024 14:59:02 GMT)
Full text and
rfc822 format available.
Message #17 received at 74780 <at> debbugs.gnu.org (full text, mbox):
Hey John and all,
It’s the turn of the “mesa-updates” branch to be merged:
https://qa.guix.gnu.org/
:-)
Ludo’.
Information forwarded
to
guix-patches <at> gnu.org
:
bug#74780
; Package
guix-patches
.
(Mon, 30 Dec 2024 02:48:01 GMT)
Full text and
rfc822 format available.
Message #20 received at 74780 <at> debbugs.gnu.org (full text, mbox):
Hi all,
On Sun, Dec 29, 2024 at 03:57 PM, Ludovic Courtès wrote:
> Hey John and all,
>
> It’s the turn of the “mesa-updates” branch to be merged:
>
> <https://qa.guix.gnu.org/>
>
> :-)
>
> Ludo’.
Thanks for checking in.
I just rebased and pushed. I'll keep an eye on it but didn't notice
any big failures (plenty on earlier evaluations but mostly due to the
python merge from what I can tell).
The question is what the substitute coverage will be like on
non-x86_64-linux. It had already been about 9-10 days from the last
update and didn't seem like much progress. And with QA back up but
very behind, not sure if we expect that to tell us anything soon.
Should I give it a day or so to build the current rebase and then just
merge? From Efraim's local builds and given the nature of the commits,
I don't expect any big surprises.
(Side note: does i686-linux have priority on Berlin or something like
that? The last few weeks I notice many more active i686 builds after a
push with many pending builds. The current evaluation, for instance:
the vast majority of workers are idle, and then the vast majority of
active are i686 builds. Or is that from just generally being behind on
i686 so there's just more of that to do?)
Thanks!
John
Information forwarded
to
guix-patches <at> gnu.org
:
bug#74780
; Package
guix-patches
.
(Sat, 04 Jan 2025 05:35:02 GMT)
Full text and
rfc822 format available.
Message #23 received at 74780 <at> debbugs.gnu.org (full text, mbox):
On Sun, Dec 29, 2024 at 09:46 PM, John Kehayias wrote:
> Hi all,
>
> On Sun, Dec 29, 2024 at 03:57 PM, Ludovic Courtès wrote:
>
>> Hey John and all,
>>
>> It’s the turn of the “mesa-updates” branch to be merged:
>>
>> <https://qa.guix.gnu.org/>
>>
>> :-)
>>
>> Ludo’.
>
> Thanks for checking in.
>
> I just rebased and pushed. I'll keep an eye on it but didn't notice
> any big failures (plenty on earlier evaluations but mostly due to the
> python merge from what I can tell).
>
> The question is what the substitute coverage will be like on
> non-x86_64-linux. It had already been about 9-10 days from the last
> update and didn't seem like much progress. And with QA back up but
> very behind, not sure if we expect that to tell us anything soon.
>
> Should I give it a day or so to build the current rebase and then just
> merge? From Efraim's local builds and given the nature of the commits,
> I don't expect any big surprises.
>
> (Side note: does i686-linux have priority on Berlin or something like
> that? The last few weeks I notice many more active i686 builds after a
> push with many pending builds. The current evaluation, for instance:
> the vast majority of workers are idle, and then the vast majority of
> active are i686 builds. Or is that from just generally being behind on
> i686 so there's just more of that to do?)
>
> Thanks!
> John
QA page still hasn't updated and it has been another 5 days from the
last message. Unless there are any objections I'll do a rebase and
merge later this weekend.
John
Information forwarded
to
guix-patches <at> gnu.org
:
bug#74780
; Package
guix-patches
.
(Sat, 04 Jan 2025 17:39:02 GMT)
Full text and
rfc822 format available.
Message #26 received at 74780 <at> debbugs.gnu.org (full text, mbox):
Hi John,
John Kehayias <john.kehayias <at> protonmail.com> skribis:
> QA page still hasn't updated and it has been another 5 days from the
> last message. Unless there are any objections I'll do a rebase and
> merge later this weekend.
Sure. Please do check the situation at
<https://ci.guix.gnu.org/jobset/mesa-updates> and how it compares to
‘master’.
You can check that direct dependents of packages that were upgraded
still build fine. For example, “guix build -P1 mesa --no-grafts -n -v1”
will indicate whether substitutes are available for all of the direct
dependents of ‘mesa’, meaning that they were successfully built.
Thanks for working on this!
Ludo’.
Reply sent
to
John Kehayias <john.kehayias <at> protonmail.com>
:
You have taken responsibility.
(Mon, 06 Jan 2025 02:12:02 GMT)
Full text and
rfc822 format available.
Notification sent
to
John Kehayias <john.kehayias <at> protonmail.com>
:
bug acknowledged by developer.
(Mon, 06 Jan 2025 02:12:03 GMT)
Full text and
rfc822 format available.
Message #31 received at 74780-done <at> debbugs.gnu.org (full text, mbox):
Hi Ludo’!
On Sat, Jan 04, 2025 at 06:38 PM, Ludovic Courtès wrote:
> Sure. Please do check the situation at
> <https://ci.guix.gnu.org/jobset/mesa-updates> and how it compares to
> ‘master’.
>
Yes, I've been keeping an eye on that and by overall percentage (from
the Cuirass homepage), the mesa-updates branch has been about equal to
master for a while. Though I know the missing stuff is for non-x86
which we get more of from Bordeaux, which is what I usually try to
wait for, unfortunately harder to tell without the handy QA page
comparison.
> You can check that direct dependents of packages that were upgraded
> still build fine. For example, “guix build -P1 mesa --no-grafts -n -v1”
> will indicate whether substitutes are available for all of the direct
> dependents of ‘mesa’, meaning that they were successfully built.
>
Oh, the new dependents option for guix build, very nice! I hadn't
tried it before but this is great.
I didn't see something like this built-in (maybe I missed it?) but
seems pretty easy to do a comparison then to another branch. I just
did the same command on my 'master' worktree and compared, to see just
a few minor differences (so I restarted the builds since they didn't
have a clear error).
A little bash-fu could easily have a nice little comparison between
branches using substitute coverage, or maybe we want to add that as an
option directly? Maybe specifying two branches or commits to see what
package names differ (in hash or substitutes)?
> Thanks for working on this!
>
> Ludo’.
Most welcome!
Merged just now: 23231c296fb2c2af8c1c0a6ead1dd6f0833f7c45 to
a65ebe5fad6921dddb165f417761886fc114ad29.
Thanks,
John
bug archived.
Request was from
Debbugs Internal Request <help-debbugs <at> gnu.org>
to
internal_control <at> debbugs.gnu.org
.
(Mon, 03 Feb 2025 12:24:15 GMT)
Full text and
rfc822 format available.
This bug report was last modified 134 days ago.
Previous Next
GNU bug tracking system
Copyright (C) 1999 Darren O. Benham,
1997,2003 nCipher Corporation Ltd,
1994-97 Ian Jackson.