04:59 <anholt_> eric_engestrom: huh, seemed to be up here. restarted it, anyway.

04:59 <anholt_> I'll be out of town for the next week, so turn off my farm if it fails

07:27 <hyung_song> Hello. I'm trying to fork a repository to make a merge request, but fork button is disabled saying "You have reached your project limit". Account created last Friday and project list is empty. Also unable to create blank project. Any guide to fix this?

07:29 <daniels> hyung_song: this link should be available in a banner at the top of the page https://gitlab.freedesktop.org/freedesktop/freedesktop/-/wikis/home

07:30 <hyung_song> daniels: Thank you very much!

07:33 <daniels> np

07:35 <bentiss> FWIW, just enabled https://docs.gitlab.com/ee/user/project/repository/forking_workflow.html#from-the-ui

07:35 <bentiss> Ristovski: see above :)

09:02 <mupuf> Nice, bentiss!

09:33 <Ristovski> bentiss: very nice! thanks a lot

09:35 <bentiss> Ristovski: hopefully it'll work without too much quirks. It is enabled by default in 16.0, but we can turn it on in our current 15.11.x installation.

09:49 <kode54> this job looks like it crashed early or something: https://gitlab.freedesktop.org/mesa/mesa/-/jobs/42255702

11:20 <eric_engestrom> anholt_: I'll disable your farm since it's been blocking MRs since yesterday, and you can have a look when you're back :)

11:20 <eric_engestrom> anholt_: thanks, but it's still failing with "curl: (52) Empty reply from server"

11:38 <eric_engestrom> pushed 520c5091d7de4290f87b directly to main to bypass marge's queue so that the rest of it can go through normally

12:46 <tlwoerner> can we please get https://grillo-0.github.io/blog/feed.xml (a gsoc contributor) added to the planet?

12:52 <daniels> tlwoerner: can you please raise an MR on https://gitlab.freedesktop.org/freedesktop/planet.freedesktop.org ?

13:03 <tlwoerner> daniels: ok https://gitlab.freedesktop.org/freedesktop/planet.freedesktop.org/-/merge_requests/42

13:03 <daniels> tlwoerner: thanks!

14:15 <kxkamil> I see container error on gitlab, any way to cope with it ?

14:16 <kxkamil> container error 500 (Internal Server Error)

14:28 <daniels> kxkamil: can you please send a link to the job?

14:29 <kxkamil> there are two, first: https://gitlab.freedesktop.org/drm/igt-gpu-tools/-/jobs/42272865

14:30 <kxkamil> or more, https://gitlab.freedesktop.org/drm/igt-gpu-tools/-/jobs/42273349

14:30 <daniels> yeah, honestly I think it's 'hope it doesn't keep mhappening'

14:32 <kxkamil> hmm, last one has another one: https://gitlab.freedesktop.org/drm/igt-gpu-tools/-/jobs/42273353

14:32 <kxkamil> no disk space ?

14:33 <kxkamil> should I try re-run in hope it will work ?

14:50 <kxkamil> "mountpoint for cgroup not found"

14:50 <daniels> time="2023-05-23T13:03:59Z" level=error msg="container_linux.go:346: starting container process caused \"process_linux.go:297: applying cgroup configuration for process caused \\\"mountpoint for cgroup not found\\\"\"\n"

14:50 <daniels> yeah

14:51 <daniels> I think that means that igt is broken and needs to be updated to use at least podman but ideally ci-templates

14:56 <bentiss> daniels: looks like podman-free-space is not happy on ml-13... and we only have 140GB of free space

14:57 <bentiss> I won't have the time to debug this today, and I won't have time to switch x86 runners to fedora-coreos either... so maybe we should just respin it?

14:59 <bentiss> actually, all three x86 runners are showing issues with podman-free-space

15:10 <daniels> err

15:10 <daniels> wtf is /var/log a 0-byte file

15:15 <daniels> bentiss: deployed a fix to -13 and -17

15:15 <daniels> is -20 dead?

15:16 <daniels> will push the fixes to git when I figure out where I put my keys :)

15:35 <bentiss> daniels: I think we only have -13, -17 and -21

16:14 <daniels> bentiss: all three are happily freeing space now btw and I've pushed the changes

16:15 <kxkamil> thx, can I respin igt pipeline ?

16:19 <daniels> kxkamil: that's not the problem though

16:20 <daniels> the problem is that igt needs to change the way it does CI to use ci-templates like everyone else, or at least use podman directly

16:21 <kxkamil> could you point to projects which do it the correct way ?

16:21 <kxkamil> is igt the only one ?

16:22 <daniels> igt is the only one I know of with this issue, yeah

16:22 <daniels> https://freedesktop.pages.freedesktop.org/ci-templates/ has all the documentation for it

16:22 <kxkamil> thx a lot

16:22 <daniels> you can look at almost any other project on fd.o for examples - mesa, wayland, weston, etc

16:25 <kxkamil> checked .gitlab-ci.yml - there is podman

16:27 <kxkamil> ah, there is, i will do it, now write it into my todo

16:27 <kxkamil> thx once again

16:47 <bentiss> daniels: thanks a bunch!

16:48 <daniels> bentiss: np!

16:48 <daniels> kxkamil: np

20:54 <republic83> So some immediate snippets, first say that we start with hashing a add, we have 1024 indexes and 1 offset and inversions, we add 1023 times 512 and 6, provided that was the index, next up, there is mul, again 1024 indexes, but this time we find that 2 mul 6, this has one nesting, first 1022-8=1014 1018-2=1016 1014+1016=2030-1024=1006+12=1018 so 1024-1018 is 6, that determines a max of the stuff, now you have two nestings

20:54 <republic83> indexes should be taken off from 6*6 i.e 36

20:54 <republic83> Removing those you get answer 12

20:55 <republic83> Pretty much some version of universal gate stuff