ChanServ changed the topic of #asahi-gpu to: Asahi Linux GPU development (no user support, NO binary reversing) | Keep things on topic | GitHub: https://alx.sh/g | Wiki: https://alx.sh/w | Logs: https://alx.sh/l/asahi-gpu
dapselt^ has quit [Remote host closed the connection]
dapsayemolt^ has joined #asahi-gpu
pb17 has quit [Ping timeout: 480 seconds]
m5zs7k_ has quit []
m5zs7k has joined #asahi-gpu
pb17 has joined #asahi-gpu
chrisl has quit [Ping timeout: 480 seconds]
pb17 has quit [Ping timeout: 480 seconds]
opticron has quit [Read error: Connection reset by peer]
opticron has joined #asahi-gpu
pb17 has joined #asahi-gpu
cow123 has quit [Ping timeout: 480 seconds]
chrisl has joined #asahi-gpu
chrisl has quit [Ping timeout: 480 seconds]
chrisl has joined #asahi-gpu
chrisl has quit [Ping timeout: 480 seconds]
pb17 has quit [Ping timeout: 480 seconds]
marvin24 has joined #asahi-gpu
marvin24_ has quit [Ping timeout: 480 seconds]
pb17 has joined #asahi-gpu
chrisl has joined #asahi-gpu
chrisl has quit [Ping timeout: 480 seconds]
r0ni has quit [Read error: Connection reset by peer]
chrisl has joined #asahi-gpu
pb17 has quit [Ping timeout: 480 seconds]
chrisl has quit [Ping timeout: 480 seconds]
pb17 has joined #asahi-gpu
<chadmed> alyssa: is this the DCP fb format we will need for planar scanout?
john-cabaj has quit [Ping timeout: 480 seconds]
chrisl has joined #asahi-gpu
chrisl has quit [Ping timeout: 480 seconds]
ece964311 has joined #asahi-gpu
ece96431 has quit [Ping timeout: 480 seconds]
pb17 has quit [Ping timeout: 480 seconds]
pb17 has joined #asahi-gpu
chrisl has joined #asahi-gpu
dapsayemolt^ has quit [Ping timeout: 480 seconds]
chrisl has quit [Ping timeout: 480 seconds]
jmax has joined #asahi-gpu
pb17 has quit [Ping timeout: 480 seconds]
pb17 has joined #asahi-gpu
chrisl has joined #asahi-gpu
chrisl has quit [Ping timeout: 480 seconds]
chadmed has quit [Quit: Konversation terminated!]
chadmed has joined #asahi-gpu
pb17 has quit [Ping timeout: 480 seconds]
chrisl has joined #asahi-gpu
chrisl has quit [Ping timeout: 480 seconds]
pb17 has joined #asahi-gpu
chrisl has joined #asahi-gpu
dapsayemolt^ has joined #asahi-gpu
chrisl has quit [Ping timeout: 480 seconds]
bass_exe has quit [Ping timeout: 480 seconds]
chrisl has joined #asahi-gpu
bass_exe has joined #asahi-gpu
chrisl has quit [Ping timeout: 480 seconds]
pb17 has quit [Ping timeout: 480 seconds]
pb17 has joined #asahi-gpu
chrisl has joined #asahi-gpu
chrisl has quit [Ping timeout: 480 seconds]
john-cabaj has joined #asahi-gpu
r0ni has joined #asahi-gpu
pb17 has quit [Ping timeout: 480 seconds]
chrisl has joined #asahi-gpu
pb17 has joined #asahi-gpu
chrisl has quit [Ping timeout: 480 seconds]
<alyssa> chadmed: no, that one is infinitely more messed up
<alyssa> Apple calls that one interchange-tiled
<alyssa> and I cannot figure out any way to use it on Linux without, like, rewriting kwin
<alyssa> so... "will" is "will not" ;)
cow123 has joined #asahi-gpu
chrisl has joined #asahi-gpu
chrisl has quit [Ping timeout: 480 seconds]
pb17 has quit [Ping timeout: 480 seconds]
<alyssa> ZLS *does* support twiddled, the bits are where you'd expect them to be
<alyssa> so when apple says sparse z/s starts with m2 they mean that specifically their weird version of sparse does
<alyssa> we probably could do sparse z/s on m1 with some shenanigans if we were so motivated
<alyssa> ..yeah actually we definitely could
<alyssa> ooh dEQP-GLES3.functional.texture.filtering.3d.sizes.3x7x5_linear_mipmap_linear: failing
pb17 has joined #asahi-gpu
chrisl has joined #asahi-gpu
chrisl has quit [Ping timeout: 480 seconds]
<alyssa> gathering data for this layout is... annoying
<alyssa> since apple only seems to use it with sparse, and apple's idea of sparse is crazy.
<alyssa> but juryrigged something
<alyssa> for 3x7x5, mipmapped 3D
<alyssa> according to the sparse map that spans 5 pages, each specified at 0x200 offsets
<alyssa> in units of 4
<alyssa> which is what we expect
<alyssa> PBE descriptors looks almost like we expect. only weird thing is page-aligned layers isn't set?!
<alyssa> even though obviously they are
<alyssa> ditto for the texture descriptor
<alyssa> first,last level = 0..2
<alyssa> unk mipmappe dis set
chrisl has joined #asahi-gpu
<alyssa> oh dEQP-GLES3.functional.texture.filtering.2d.sizes.3x7_linear_mipmap_linear fails too
chrisl has quit [Ping timeout: 480 seconds]
<alyssa> oh, sure
<alyssa> mipmapping first divides, then rounds
<alyssa> instead of rounding first then dividng
<alyssa> so the hw is a bit more efficient than my original guess.
<alyssa> cool
<alyssa> nice
dapsayemolt^ has quit [Remote host closed the connection]
<alyssa> there we go
<alyssa> ...except for compressed
<alyssa> so it shares the same weird special case as gpu-tiled
<alyssa> that makes sense, I guess
<alyssa> wait what
<alyssa> then why did we break dEQP-GLES31.functional.copy_image.non_compressed.viewclass_128_bits.rgba32f_rgba32ui.texture2d_to_cubemap
<alyssa> serious wat
<alyssa> found a fault for a 65x65x6 cubemap case with r32f .. ok
<alyssa> so why is 65x65 different than 3x7?
<alyssa> er, 65x63
<alyssa> i guess maybe major vs minor axis is the actual issue here
<alyssa> ohhhh
<alyssa> I *think* it does round-then-divide to figure out where levels start
<alyssa> but divide-then-round to figure out the actual tile sizes in the levels
<alyssa> this is better than always doing round-then-divide, because at least within a level stuff is nice and packed so you access fewer cache lines --> faster perf
<alyssa> (perf is the same as divide-then-round, the "correct" approach)
<alyssa> but worse then divide-then-round for RAM usage since there's extra padding sometimes
<alyssa> but that doesn't matter for perf since levels are cache line padded anyway
<alyssa> ...except for compressed?
<alyssa> OK yes here we go
<alyssa> cts likes this now
<alyssa> /* AGX: Even the exceptions have exceptions */
<alyssa> ok, gl cts passing now with twiddled images
<alyssa> vk, you're up next
pb17 has quit [Ping timeout: 480 seconds]
pb17 has joined #asahi-gpu
WindowPain6 has joined #asahi-gpu
WindowPain has quit [Ping timeout: 480 seconds]
WindowPain6 is now known as WindowPain
chrisl has joined #asahi-gpu
chrisl has quit [Ping timeout: 480 seconds]
<jannau> agx driver compiles again after replacing a ton stale rust dependencies with upstream changes or patches under active discussion
chrisl has joined #asahi-gpu
chrisl has quit [Ping timeout: 480 seconds]
master91122 has quit [Quit: Connection closed for inactivity]
chrisl has joined #asahi-gpu
chrisl has quit [Ping timeout: 480 seconds]
chrisl has joined #asahi-gpu
pb17 has quit [Ping timeout: 480 seconds]
chrisl has quit [Ping timeout: 480 seconds]
<alyssa> yay!
<alyssa> ok, after that... significant diversion
<alyssa> I'm now confident in my understanding of the different tilings so https://lore.kernel.org/asahi/20250225-apple-twiddled-modifiers-v2-1-cf69729e87f6@rosenzweig.io/T/#u sent out
pb17 has joined #asahi-gpu
<alyssa> jannau: amazing work :)
<jannau> doesn't work yet because reserved CPU memory shouldn't be handled in the same way as iomem
<sven> jannau: \o/
<alyssa> jannau: hey progress is progress =D
<jannau> in principle this should have all timed used of_reservedmem but that looks annoying to wrap for rust
<jannau> I suppose this should be used by all other asc drivers as well instead of the the automatic mapping I implemented for dart
cow123 has quit [Read error: Connection reset by peer]
<jannau> and working with adding more hacks
<alyssa> Yay!
cow123 has joined #asahi-gpu
chrisl has joined #asahi-gpu
chrisl has quit [Ping timeout: 480 seconds]
Calandracas_ has joined #asahi-gpu
pb17 has quit [Ping timeout: 480 seconds]
<jannau> less than 2.5k added rust abstractions left (excluding drm): https://paste.centos.org/view/325a8533
<jannau> this is after abstactions already under discussion on rust-for-linux
pjakobsson_ has quit [Ping timeout: 480 seconds]
Calandracas has quit [Ping timeout: 480 seconds]
<jannau> I still have to replace the xarray abstractions
pb17 has joined #asahi-gpu
chadmed has quit [Quit: Konversation terminated!]
chadmed has joined #asahi-gpu
ypwn_ has joined #asahi-gpu
ypwn has quit [Read error: Connection reset by peer]
pjakobsson has joined #asahi-gpu