09:56daniels: lumag: wondering if you had opinions about https://lore.kernel.org/dri-devel/20250627131751.2004-1-marius.vlad@collabora.com/T/#m3d7281f5e27d4e0864ffaf1644614ed2ce96924e ?
10:08Ermine: Hello, is displayport spec open?
10:09K900: Depends on what you mean by "open"
10:09K900: There's no patent fees
10:10K900: But you need to be a VESA member to actually access the latest specs
10:17Ermine: I meant actual access, so answer is "no"...
10:20K900: You can find leaked PDFs if you're into that kind of thing
10:20K900: But also why do you need it?
10:27MrCooper: FWIW, AFAIK X.org foundation members can get access to VESA specs
10:28Ermine: I want to try to resolve a dp-related bug in amdgpu (and also try to accomplish a related TODO item)
10:30K900: That's unlikely to be the kind of thing you can do by reading the specs
10:30K900: All the protocol level stuff is handled in firmware
10:31K900: That you're not going to get to unless you work at AMD
10:34Ermine: i know that specs are not enough, but i'd like to know some theory to understand stuff
12:23demarchi: mripard: did you have a chance to look at the failures on dim push we have with the additional checks?
12:23demarchi: mripard: haaving quite a few people tripping on that and not able to push
12:25mripard: demarchi: I've been pretty sick all week, so I started but I'm not at full speed
12:25mripard: I'll give it a try this afternoon
12:31demarchi: mripard: just thought about a quick thing to avoid the issue.... use --diff-filter=a to exclude added files
12:32demarchi: and we could even pass the added files separately through get_maintainer by using --diff-filter=A as a second call
12:32demarchi: other than that, I think we will need to change get_maintainer, but I can't grok what it's doing
12:44demarchi: https://gitlab.freedesktop.org/drm/maintainer-tools/-/merge_requests/83
12:49mripard: demarchi: looks good to me, thanks for looking into it
13:06mripard: demarchi: we have some other issue that could use a similar solution, but I'm not sure I see how
13:07demarchi: mripard: what other issue?
13:07mripard: MAINTAINERS alone will return torvalds' repo
13:07mripard: and if it's in a patch with other changes, torvalds' repo might appear first
13:08mripard: it looks from the man page that there's nothing in git show to ignore a file, which would be the best solution I guess
13:09demarchi: humn... then we probably need a path filter... `-- :(exclude)MAINTAINERS`
13:10mripard: does that work with show too?
13:16mripard: it does
13:17mripard: c8dc57916973 is in such a case if you want to give it a try
13:28demarchi: mripard: thanks... I will task switch now and try that later today
15:40alyssa: jenatali: um https://gitlab.freedesktop.org/mesa/mesa/-/jobs/81108560
15:40alyssa: ../src/gallium/drivers/d3d12/d3d12_context_graphics.cpp(1322): error C3861: '__typeof__': identifier not found
15:40alyssa: it worked.. yesterday..
15:40jenatali: C++ hooray
15:40alyssa: yeah i'm just dropping the dozen commit
15:40jenatali: You need a C++ version which does decltype() instead, or just convert caller to std::swap
15:41alyssa: oh sure std::swap
15:41jenatali: That's not dozen?
15:41alyssa: ...oh, right
15:41alyssa: dozen is C and d3d12 is C++?
15:41jenatali: Yeah
15:41alyssa: charming
15:41jenatali: Yeah
15:41jenatali: Except that there's a couple files in d3d12 that are C because nir is more ergonomic in C but D3D is more ergonomic in C++
15:42alyssa: vulkan: ergonomic in neither
15:43alyssa: q
15:49zmike: I just looked at https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests and it seems like there's a lot of open MRs
15:50zmike: should we just start merging stuff?
15:50mareko: yes
15:51Mary: totally forgot that I opened an MR on piglit for realtime priority oops
15:53zmike: there's a lot of them that even have rbs
15:53zmike: and nobody clicked merge
15:54zmike: yolooooooooooooo
15:55Hazematman: yomo - you only merge once
15:58zmike: mareko: https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/1025
15:58zmike: this should address your recent findings
15:58anholt: I disagree with "just start merging stuff". If there's a review problem, then get reviewing. If attention is missing on something of yours, ask someone.
15:59anholt:still reviews piglit tests
15:59zmike: I've assigned a bunch of MRs which already had rbs
15:59anholt: great!
15:59zmike: and hopefully my mail will draw attention to other MRs
17:30eric_engestrom: mesa devs, quick reminder to let me know on https://gitlab.freedesktop.org/mesa/mesa/-/issues/13155 if there is something that should be mentioned in the 25.2 announcement email :)
18:55zmike: how close are we