#
✕ nick:chadmed
#asahi-dev · 2024-05-14 13:05
chadmedbut breaking all gtk and firefox is probably too much breakage
#asahi-dev · 2024-05-14 13:05
chadmedim of the opinion that we should just ship this as-is to provide impetus to fix this stuff
#asahi-dev · 2024-05-14 13:01
chadmedfirefox is gtk3...
#asahi-dev · 2024-05-14 12:56
chadmedgimp also misbehaves
#asahi-dev · 2024-05-14 12:55
chadmedfirefox and sysprof both misbehave
#asahi-dev · 2024-05-14 12:54
chadmedleio: no rush or anything, like i said judging by how gtk apps behave i have little hope that mutter does the right thing currently :p
#asahi-dev · 2024-05-14 11:30
chadmedgtk hangs the entire commit chain
#asahi-dev · 2024-05-14 11:30
chadmedugh yeah this is a mess, weston just crashes
#asahi-dev · 2024-05-14 11:29
chadmedcmon....
#asahi-dev · 2024-05-14 11:29
chadmed"let's do a new plane type but also tell all of userspace that plane types are now irrelevant" ???
#asahi-dev · 2024-05-14 11:29
chadmedyeah just read it, that discussion was not very helpful
#asahi-dev · 2024-05-14 11:24
chadmedim gonna guess off the bat that mutter does not behave sensibly here....
#asahi-dev · 2024-05-14 11:24
chadmedugh firefox...
#asahi-dev · 2024-05-14 11:21
chadmeds/reject/rejected
#asahi-dev · 2024-05-14 11:21
chadmedi probably could have saved ~3 days if i had just tried that initially...
#asahi-dev · 2024-05-14 11:21
chadmedkwin does the right thing and if a cursor swap is reject will immediately fall back to software cursors
#asahi-dev · 2024-05-14 11:19
chadmedleio: could i possibly get you to test https://github.com/chadmed/linux/tree/dcp-overlay-planes on gnome
#asahi-dev · 2024-05-12 20:54
chadmedtheres no point even enumerating a type anymore if we're just going to ignore them
#asahi-dev · 2024-05-12 20:54
chadmedif the driver has not exposed a cursor plane, it should be assumed that the hardware does not support hardware cursors. its that simple
#asahi-dev · 2024-05-12 20:53
chadmedand we are proof of that
#asahi-dev · 2024-05-12 20:53
chadmedin most hardware they behave mostly the same, but cursors are special and they shouldnt be treated the same as content for an overlay
#asahi-dev · 2024-05-12 20:52
chadmedthere is primary, overlay, and cursor
#asahi-dev · 2024-05-12 20:52
chadmedthe problem with consuming an overlay with the cursor is that most hardware either doesnt support them at all or only has two or three planes
#asahi-dev · 2024-05-12 20:51
chadmedi mean overlay planes in general, the point is for that use case
#asahi-dev · 2024-05-12 20:49
chadmedi think "universal" really is a misnomer here
#asahi-dev · 2024-05-12 20:49
chadmedleio: that is the point yeah. i dont think it is reasonable for compositors to use generic overlays for the cursor. planes have been universal since 3.x and neither xorg nor any wayland compositor has yet to use overlay planes for the cursor
#asahi-dev · 2024-05-12 12:36
chadmedso we're still "good" :)
#asahi-dev · 2024-05-12 12:36
chadmedit doesnt mean that all plane types will be treated the same
#asahi-dev · 2024-05-12 12:36
chadmedoh i think we misunderstood the docs. plane types havent/arent going away, userspace just gets to see all of them rather than kms tying the primary and cursor to the crtc internally and using unique ioctls for them
#asahi-dev · 2024-05-12 12:18
chadmedi am slightly worried about universal planes though...
#asahi-dev · 2024-05-12 12:17
chadmedhw cursor specifically is no big deal right now but fixing as much as we can for multiple planes in general will make it less painful to wire them up for video playback and such when we want that eventually
#asahi-dev · 2024-05-12 12:16
chadmedno (obvious) colour space issues observed here, but obv the pros have different screens. will test tomorrow on j314.
#asahi-dev · 2024-05-12 11:39
chadmedi was just about ready to give up too!
#asahi-dev · 2024-05-12 11:38
chadmedbut at least it doesnt just crash now :)
#asahi-dev · 2024-05-12 11:37
chadmedanyway the issue is that dcp does _not_ like framebuffers smaller than 32x32, and will fail to swap. kwin is squishing the cursor down past 8x64
#asahi-dev · 2024-05-12 11:36
chadmedtheres no point until at least kwin supports universal planes, as the old api that distinguishes between overlays and cursors is dead
#asahi-dev · 2024-05-12 11:36
chadmedyou wont see it happen, we havent wired up hardware planes yet :p
#asahi-dev · 2024-05-12 11:35
chadmedits very odd and im sure its not _supposed_ to do that (we are probably doing something odd)
#asahi-dev · 2024-05-12 11:35
chadmedLarstiQ: when you move to the right of the screen, rather than clip out past the bounds of the primary plane as youd expect, the cursor just gets compressed along the x axis such that the rightmost edge of its framebuffer never clips outside the rightmost edge of the primary below it
#asahi-dev · 2024-05-12 11:20
chadmedalso surface 0 works on m2 on 13.5. it's getting late now but i will build this patch on j314 and see if it works there now too with 13.5 fw
#asahi-dev · 2024-05-12 11:18
chadmedonly two issues left: stupid behaviour near the edge of the screen (why tf do compositors try to squish the cursor? just noclip out of the bounds of the primary plane), and the fact that alpha doesn't seem to work as it should (the cursor plane is filled with black instead of transparency)
#asahi-dev · 2024-05-12 11:16
chadmedof any size btw
#asahi-dev · 2024-05-12 11:16
chadmedthis fixes dcp crashing when using hardware cursors
#asahi-dev · 2024-05-12 11:16
chadmedhttps://github.com/chadmed/linux/commit/36a275ae2a4a4dd599360fd413b8c1813c886321
#asahi-dev · 2024-05-12 09:55
chadmedbut at least it doesnt crash
#asahi-dev · 2024-05-12 09:54
chadmednew issue: alpha doesn't alpha and the cursor plane is filled with black
#asahi-dev · 2024-05-12 09:53
chadmeduserspace will always want the cursor over the primary plane, so just feed the planes into req->surf[] in the order userspace tells us!
#asahi-dev · 2024-05-12 09:53
chadmedfor_each_oldnew_plane_in_state seems to not have a set order, and dcp seems to expect to receive surfaces in order
#asahi-dev · 2024-05-12 09:52
chadmedvery funny and silly fix: l = new_state->zpos
#asahi-dev · 2024-05-12 09:52
chadmedit even works with small cursors

plus ›