03:26 zmike: dcbaker: \o/
03:27 dcbaker: robclark: I thing in the next release of meson we’ll actually tell you which option :D
03:29 robclark: yeah, I figured it out in the end.. but the constant bool -> enabled/disabled/auto churn is a bit annoying.. I kinda wish we had some aliases true -> enabled / false -> disabled ;-)
03:32 dcbaker: robclark: Can in interest you in https://github.com/mesonbuild/meson/pull/11279
03:33 robclark: yes please
03:41 daniels: cwabbott: the a618 are ours and are known to be physically stable - lmk if you need deeper access to bottom out specific issues
11:03 eric_engestrom: robclark, dcbaker: https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21485
11:57 hakzsam: anyone from venus can help for CTS upgrade https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21390 ?
11:57 hakzsam: other drivers are stable now
11:58 hakzsam: daniels: or should I disable it? :D
13:14 DavidHeidelberg[m]: hakzsam: :D
13:41 DavidHeidelberg[m]: hakzsam: try to ping @gerddie at gitlab
13:41 hakzsam: Daniel did already
13:41 DavidHeidelberg[m]: cool! :)\
16:15 DavidHeidelberg[m]: anholt_: Not sure if the message got lost in the history, but for sure: any chance for re-assign CI caches from the 8888 to the port 80 and assign DNS name `caching-proxy`? (since we could share the default traces.yml with both LAVA and Google without altering proxy)
17:05 anholt_: DavidHeidelberg[m]: 8888 to 80, easy. caching-proxy, would need to look into the dnsmasq setup and how to do that. but, more importantly, should probably only do this maintenance (which might involve restarting NM) when I might have physical access to the system if necessary.
17:05 anholt_: which would be a few days
17:06 idr: "In my day, we had to walk up hill both ways in the snow to do system maintenance." Lol
17:07 DavidHeidelberg[m]: anholt_: no need to rush, if it's possible that's great enough ;-)
18:30 DavidHeidelberg[m]: robclark: the flaky godot on a618 documented: https://gitlab.freedesktop.org/mesa/mesa/-/issues/8337
18:30 DavidHeidelberg[m]: ratio of flake approx 1:11
18:33 robclark: thx
18:34 robclark: idr: when the sidewalk is a sheet of ice, walking downhill is also not great :-P
18:41 robclark: DavidHeidelberg[m]: tbh, even zoomed in to 200% I can't make out the difference when flipping btwn the images... does png have an alpha channel which might be different, or something like that?
18:45 DavidHeidelberg[m]: robclark: I think it's like rgb(255,0,0) to rgb(254,0,0) type of changes
18:46 robclark: hmm, ok
20:52 mareko: spambot in gitlab: https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21476#note_1790001
20:53 mareko: how do we handle it?
20:53 psykose: you wouldn't.. watch a movie..
20:55 jenatali: "Report abuse to administrator"
20:55 jenatali: In the "..." menu
22:55 DavidHeidelberg[m]: robclark: I tried zoom it and loop it on https://okias.pages.freedesktop.org/-/mesa/-/jobs/36850630/artifacts/results/summary/results/trace@freedreno-a618@godot@godot-tps-gles3-high.trace.html and "the flake" looks cleaner than original
22:57 robclark: I still can't spot the diff ;-)
22:58 DavidHeidelberg[m]:uploaded an image: (45KiB) < https://matrix.org/_matrix/media/v3/download/matrix.org/MGsLXEHXcgTroTzbaabdXIbP/godot_flake.png >
22:58 robclark: I guess we should use something less than exact diff
22:58 DavidHeidelberg[m]:uploaded an image: (45KiB) < https://matrix.org/_matrix/media/v3/download/matrix.org/BkEOaYcGKaYhTOOeDCSbeGBn/godot_orig.png >
22:58 DavidHeidelberg[m]: in the center
22:59 robclark: yeah, ok, I guess I just need to open the img in it's own tab so I can zoom
22:59 DavidHeidelberg[m]: the flake (first) look a bit smoother
22:59 robclark: yeah
22:59 DavidHeidelberg[m]: and the original is a bit more like steps
23:05 DavidHeidelberg[m]: it looks like to me, that most often produced output is the imperfect one