07:33pixelcluster: why is it again that we can't just have marge-bot run clang-format in projects that want it (I'd like if clang-format were automatic in mesa, for example)?
07:37pixelcluster: can't blame contributors getting upset by having to run the tool whose entire purpose it is to automate tedious code formatting multiple times, or having to go out of their way to install a different version because their current version is slightly incompatible...
07:37pixelcluster: so it'd be really cool if marge could just do that step as part of the rebase or something like that
08:30daniels: pinchartl: yeah for sure
08:30daniels: pixelcluster: iirc it was discussed but no-one worked on Marge to do it
08:37MrCooper: maybe if clang-format didn't make things worse in some cases
08:43pixelcluster: MrCooper: this is strictly about the parts of the code that already enforce clang-format
08:44pixelcluster: I definitely don't intend to just clang-format everything whether it's wanted or not, but for the parts where clang-format is enforced, you already can't merge without your code being formatted by clang-format
08:44MrCooper: yep, it means humans couldn't clean up after clang-format messing up
08:44MrCooper: this has been attempted in Mesa
08:45pixelcluster: when? any links to discussions about this?
08:48MrCooper: well, not exactly what you're suggesting but running clang-format in CI and failing if it complains
08:49MrCooper: anyway, technical solutions to social issues and all that
08:50pixelcluster: running clang-format in CI and failing if it complains is what we do
08:50pixelcluster: it's good to enforce clang-format and there haven't really been issues with it
08:50pixelcluster: we don't need to re-hash that
09:09pq: I *thought* I had no problems with gmail subscribed to mailing lists, until I realized I hadn't been receiving any email from some lists at all for years. Mostly related to X.org... like the members list. Due to the recent worsening, I have migrated all my gitlab and mailing list subscriptions away from gmail.
09:09pq: btw. does lei work for fd.o lists?
09:09pq: or x.org lists?
09:32daniels: not yet, but it will
10:39eric_engestrom: spam, but I don't have the rights to flag it as such: https://gitlab.freedesktop.org/freedesktop/freedesktop-wiki-templates/-/issues/6
11:01eric_engestrom: pixelcluster: regarding getting marge to run clang-format, I wrote a proposal (https://gitlab.com/marge-org/marge-bot/-/issues/391) but didn't get around to implement it yet
11:04pixelcluster: huh, why so complicated?
11:05pixelcluster: fwiw I have a branch at https://gitlab.com/pixelcluster2/marge-bot/-/commits/clang-format which just uses another filter-branch to run clang-format
11:06pixelcluster: I suppose you could extend that to rustfmt pretty trivially by providing a --formatter option which would include both git clang-format and rustfmt
11:16pixelcluster: it works fine (see here, https://gitlab.freedesktop.org/pixelcluster/mesa/-/merge_requests/2 if you click "compare previous version" on marge's push you see the clang-format changes)
11:49daniels: eric_engestrom: you should be able to now
11:50eric_engestrom: daniels: indeed, thanks!
11:51eric_engestrom: although I guess I also need to register damspam for it ^^
11:53eric_engestrom: daniels: I can't register: "This request can only be filed by a Maintainer (or above)"
11:54eric_engestrom: could you register it? `damspam request-webhook freedesktop/freedesktop-wiki-templates`
13:40gallo[m]: krastevm: about the LAVA question. Could you share your job definition? btw, the LAVA upstream repo is located in gitlab: https://gitlab.com/lava/lava
15:10daniels: bentiss: if you've got a moment please, can you check out fdo-equinix-arm-11? I can't get to its recovery console via SSH and I also can't access the web console since it turns out you can't use your 2FA recovery codes if you don't have the TOTP device anymore ...
15:11daniels: (wonder if it's a good or a bad thing that I haven't logged in to the web console since Sep at the latest)
15:11bentiss: daniels: OK, let me see if it works...
15:11daniels: thanks!
15:15bentiss: hmm... arm-11 is not responding, indeed, but arm-12 is still running podman, I thought I switched everything back to docker
15:16bentiss: looks like ml-29 is also running podman :)
15:18bentiss: daniels: started the reboot process of arm-11
15:19bentiss: daniels: and FWIW, regarding the totp, I personnaly store them on 2 different yubikeys when I create them. This way they are not linked to any phone, but only on your keys, and you just have to keep one safe at home
15:20bentiss: (and you can use the yubikey authenticator app directly on your desktop, to not have to pull up your phone)
15:26bentiss: daniels: arm-11 is probably dead. It stops at "Welcome to GRUB!"
15:29daniels: great ...
15:30bentiss: daniels: we should probably just tell Ed about that an spin up a new one
15:30daniels: bentiss: I only have Solo keys rather than Yubis; not sure how I get TOTP working on those
15:30daniels: bentiss: yeah
15:30bentiss: i.e. not release it immediately
15:30bentiss: daniels: on solo v1 not sure you can. On solo v2 it's possible, but the firmware is not there yet
15:31daniels: yeah, I'm v2 and indeed last I saw it was 'just' waiting for firmware
15:32bentiss: heh... "solo2 app oath list" -> thread 'main' panicked at 'explicit panic', src/transport/ctap.rs:90:18
15:33daniels: :D
15:35bentiss: it's really too bad they promised a lot and did not delivered, and/or just managed to take the nitrokey changes which now support gpg/totp/oauth/piv :(
15:38bentiss: sigh... I also haven't worked on integrating the new cloud-init generation with the runners
17:28bentiss: daniels: sigh: Error: Could not create Device: POST https://api.equinix.com/metal/v1/projects/xxxxx/devices: 422 There aren't available servers in that metro
17:28bentiss: I guess this will have to wait until Monday then
17:28daniels: fff
17:28daniels: no problem, thanks
17:28daniels: I did mange to get my account unblocked itmt, but I also have to leave ... shit, actually 3 minutes ago
17:28daniels: o/
17:28bentiss: \o
18:22hch12907: hey, just a heads up - lists.freedesktop.org currently times out on ipv6
18:36BroNeedsHelp: ay
18:36BroNeedsHelp: anyone