07:57javierm: tzimmermann: sorry, I wasn't online. For which patch-set do you need review ?
07:59tzimmermann: javierm, hi! i just wanted to talk to you first. i have a series that reworks existing simpledrm/ofdrm and added drivers for efi and vesa. it's 18 patches. you'd be the optimal reviewer. shall i send it in one batch?
08:01javierm: tzimmermann: sure, one batch works for me
08:02tzimmermann: great. I've finally foudn some nice way of sharing code among these drivers, so the efi and vesa drivers aren't that large. and the series adds additional features to some of the driveres, such as edid and palette support
08:02javierm: tzimmermann: that's awesome
08:02tzimmermann: thanks for reveiwing. i think, i'll have it ready during next week
08:03javierm: tzimmermann: you are welcome
08:18tzimmermann: sima, lee jones asked for an ack by fbdev people on the backlight-cleanup series. since you looked thourhg it already, could you please send a reply with an a-b if you're OK with the changes?
08:23sima: tzimmermann, will do
08:23tzimmermann: thanks
08:23sima: tzimmermann, v3 still current one or do I need to poke lore?
08:24tzimmermann: sima, this one: https://lore.kernel.org/dri-devel/20250313165151.GE3645863@google.com/T/#t
08:25sima: tzimmermann, I thought lee jones was backlight maintainer?
08:25tzimmermann: same
08:25tzimmermann: but apparently not
08:25sima: why does he need review on backlight stuff ...
08:26tzimmermann: MAINTAINERS agrees
08:26tzimmermann: i have no idea
08:26sima: eh I'll whack an ack on the entire thing
08:26tzimmermann: thanks
08:27phasta: sima, airlied, can you check whether you're OK with this change to DRM's kconfig?
08:27phasta: https://lore.kernel.org/dri-devel/20250313133629.10418-2-tvrtko.ursulin@igalia.com/
08:27sima: phasta, not sure why this needs our ack, but you can have it
08:28phasta: I dunno, I never applied a change for "whole DRM" so far
08:28phasta: better safe than sorry
08:28sima: (narrator: this doesn't need a maintainer ack)
08:29sima: phasta, drm core changes are very much in-scope for drm-misc and the usual rule of "get appropriate amounts of review, minimally an ack" applies
08:29maazmombasawala: Any piglit maintainers here who can have a look at a merge request - https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/993 ? It's a fix for a crash in egl_khr_fence_sync I found when testing svga driver.
08:30phasta:is scared since airlied showed him the photos of the huge snakes in his garden at LPC
08:30sima: there's nothing special here
08:30sima: phasta, that's just 'straya
08:30sima: should probably have some myself, I imagine it's good for relaxing after looking at scary kernel code
08:31sima: but here in europe ancestors long ago killed all the true beasts
10:17phasta: does DRM also have the "policy" that changes must have been in drm-misc-next for 1-2 weeks to make it into the upcoming merge window?
10:17phasta: Or is everything that is in drm-misc-next on day X being sent to Linus?
10:26mripard: phasta: it has that policy
10:27mripard: it's why we stop sending drm-misc-next PR when rc6 is released
10:27mripard: and resume with rc1
10:27mripard: (and why drm-misc-next-fixes is a thing)
10:27phasta: you mean the drm-misc-next PR for drm-next
10:28mripard: yes
12:01S9: Help only LLVMPipe works. Zink give black window and the others gives glx: failed to create drisw screen Error: glXCreateContext failed
12:38S9: Why "kmsro: driver missing" is in `eglinfo` stdErr but "eglinfo: eglInitialize failed" in stdOut?
13:29S9: virpipe loose connection to rendering server, zink has some more extensions and report iGPU characteristics but Device platform: eglinfo: eglInitialize failed, softpipe report less extensions than llvmpipe but both fails like zink, and all other radeonsi r300 r600 nouveau freedreno vc4 v3d panfrost etnaviv tegra crocus iris i915 svga virgl lima swrast asahi d3d12 swr error i915g ilo r300g r600g vmwgfx glsoftware auto xswrast p
13:33alyssa: sima: Is drm in feature freeze? I am trying to understand the flow chart
13:33sima: alyssa, yup
13:34alyssa: ok
13:34sima: but the flow chart is only really relevant for bugfixes, not for anything else, that's always for drm-misc-next
13:34alyssa: i'm trying to figure out where to apply https://lore.kernel.org/asahi/14a845e8-54d0-45f8-b8b9-927609d92ede@stanley.mountain/
13:34alyssa: the adp driver seems to be in drm-next but not torvalds' tree
13:34sima: alyssa, at least I think so, but tzimmermann mripard mlankhorst would actually know
13:34sima: alyssa, yeah so that should go into drm-misc-next-fixes, but maybe it's not yet rolled forward
13:35sima: best to ping drm-misc maintainers, since they actually know what's up, unlike me :-P
13:35alyssa: sima: duly noted :p
13:35javierm: alyssa: https://drm.pages.freedesktop.org/maintainer-tools/committer/committer-drm-misc.html#where-do-i-apply-my-patch is what I always check FYI
13:35alyssa: javierm: yes that's what i'm reading
13:35alyssa: and I'm in doubt so asking my favourite maintainer on IRC
13:35sima: alyssa, but worst case you can push to drm-misc-next and then ask drm-misc maintainers to cherry-pick using dim (don't do the cherry-pick yourself)
13:35alyssa: per the instructions
13:36mripard: alyssa: so sima's your favorite maintainer. great.
13:36mripard: :)
13:36sima: yeah, not sure what I should do with the personality cult ...
13:36mripard: alyssa: but yeah, that one should be for drm-misc-next-fixes
13:37mripard: and I believe tzimmermann said it was open last week?
13:37javierm: yeah, he said:
13:37javierm: #dri-devel.03-07.log:13:08 < tzimmermann>| PSA: With -rc6 around the corner drm-misc-next-fixes is now open. Features still go into drm-misc-next. Fixes for stable still go into drm-misc-fixes. Fixes for v6.15 go into drm-misc-next-fixes. Patches in -fixes branches should be small and have a Fixes tag.
13:37alyssa: mripard: if you'd prefer I can annoy you with all my silly questions instead from now on ;)
13:38alyssa: javierm: ahhh i see
13:39javierm: alyssa: I also missed it btw, I just did a grep for drm-misc-next-fixes | grep tzimmermann in my recent irssi logs :D
13:39alyssa: javierm: ahahah ok
14:25karolherbst: quick review on https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34004 ? Fixes some SPIR-V 1.6 related crashes in rusticl, but needs fixes in vtn
15:04Lynne: even with -Damdgpu-virtio=false, mesa still tries to init a virtgpu context
15:28S9: ioctl(-1, DRM_IOCTL_I915_GEM_GET_APERTURE, 0x7fffc74014c0) = -1 EBADF (Bad file descriptor) DRM_IOCTL_I915_GEM_APERTURE failed: Bad file descriptor ioctl(-1, DRM_IOCTL_I915_GETPARAM, 0x7fffc74014d0) = -1 EBADF (Bad file descriptor) get chip id failed: -1 [9] ioctl(-1, DRM_IOCTL_I915_GETPARAM, 0x7fffc74014d0) = -1 EBADF (Bad file descriptor) i915 does not support EXECBUFER2 libva error: /run/current-system/profile/lib/dri/i965_
15:28S9: openDriver() returns -1 vaInitialize failed with error code -1 (unknown libva error),exit
15:29S9: I don't get it well, but maybe there is a difference between video/3D acceleration/drivers. Gnome WM/DE use which acceleration/driver?
15:33ccr: are you sure that the kernel driver loads correctly at all, and that you have the required firmware blobs for it
15:41S9: ccr: /var/log/debug:Mar 12 16:00:25 localhost vmunix: [ 6.673017] i915 0000:00:02.0: [drm:intel_dmc_init [i915]] Loading i915/adlp_dmc.bin /var/log/messages:Feb 23 09:38:36 localhost vmunix: [ 4.457407] i915 0000:00:02.0: [drm] Finished loading DMC firmware i915/adlp_dmc.bin (v2.20)
15:44S9: I even tested on that side: Mar 9 13:43:35 localhost vmunix: [ 3.869105] xe 0000:00:02.0: [drm] Finished loading DMC firmware i915/adlp_dmc.bin (v2.20)
15:53ccr: I'm not 100% sure but I think for Alder Lake there should also be some GuC/HuC firmware(s) being loaded, something like "[drm] GuC firmware i915/adlp_guc_70.bin" etc
15:56S9: ccr: Yeah but verbose debug erase beginning of logs, where it states about GuC/HuC
15:58S9: 123[ 4.031687] i915 0000:00:02.0: [drm] Finished loading DMC firmware i915/adlp_dmc.bin (v2.20) [ 4.063238] i915 0000:00:02.0: [drm] GT0: GuC firmware i915/adlp_guc_70.bin version 70.36.0 [ 4.063245] i915 0000:00:02.0: [drm] GT0: HuC firmware i915/tgl_huc.bin version 7.9.3
17:49S9: How to get hardware acceleration on WM/DE?
18:06zmike: landing these full-tree MRs today with zero flakes feels pretty damn good
18:15zmike: jenatali: how do I not -Werror in msvc for deprecated functions? e.g., https://gitlab.freedesktop.org/mesa/mesa/-/jobs/72957553
18:15jenatali: O.o
18:15jenatali: Lemme look it up, I assume there's some way to demote that specific warning back to just a warning
18:17zmike: on linux it's -Wno-error=deprecated-declarations
18:17jenatali: https://developercommunity.visualstudio.com/t/impossible-to-treat-warning-as-error-except-specif/473936
18:18jenatali: So... seems impossible :(
18:19jenatali: Options then would be, flip on /We for specific warnings to promote them to errors, fully suppress the deprecated function warning (either locally with pragmas or globally), or just don't use the [[deprecated]] tag for MSVC :(
18:19zmike: 🤕
18:19zmike: alright
19:33zmike: 252 files changed, 2079 insertions(+), 2302 deletions(-)
19:33zmike: it's a net win!
19:37jenatali: \o/
19:55alyssa: Yay!
19:55alyssa: depointerize?
19:56zmike: we are now pointer-free.
19:57HdkR: Pointers? Never touch the stuff. I pass everything by value.
19:57alyssa: rusty
19:57karolherbst: glad that I didn't touch pipe_surface up until now
20:00zmike: me too
20:05alyssa: i never understand surface_create and now i never have to
20:08alyssa: zmike: oh i think i know the panfrost issue
20:08alyssa: and i assume asahi is blowing up in the same way
20:08alyssa: let's check
20:08alyssa: will be a good break from glaring at ir3
20:16alyssa: hmm asahi CTS is clean
20:16alyssa: although it should have the same bug..
20:26alyssa: hmm no not the panfrost issue i expected