ChanServ changed the topic of #asahi-gpu to: Asahi Linux GPU development (no user support, NO binary reversing) | Keep things on topic | GitHub: | Wiki: | Logs:
omioolddbotm^ has joined #asahi-gpu
possiblemeatball has quit [Ping timeout: 480 seconds]
omioolddbotm^ has quit [Remote host closed the connection]
possiblemeatball has joined #asahi-gpu
possiblemeatball has quit [Ping timeout: 480 seconds]
possiblemeatball has joined #asahi-gpu
hightower4 has joined #asahi-gpu
hightower3 has quit [Ping timeout: 480 seconds]
Hibyehello has quit [Read error: Connection reset by peer]
Hibyehello has joined #asahi-gpu
possiblemeatball has quit [Quit: Quit]
chip_x has quit [Ping timeout: 480 seconds]
yamii has joined #asahi-gpu
yamii_ has quit [Remote host closed the connection]
balrog has quit [Remote host closed the connection]
balrog has joined #asahi-gpu
possiblemeatball has joined #asahi-gpu
nyilas has joined #asahi-gpu
akemin_dayo has joined #asahi-gpu
Leftas7 has joined #asahi-gpu
<alyssa> lina: uh oh
<alyssa> I think I'm figuring out what the 02345 buffer is...
nyilas has quit []
yamii has quit []
jn has quit []
ChaosPrincess has quit []
yuka_ has quit []
linuxgemini1 has quit []
Misthios has quit []
mkurz has quit []
yrlf has quit []
Mrmaxmeier has quit []
Retr0id has quit []
cy8aer has quit []
Mary6 has quit []
leftas has quit []
jole has quit []
manawyrm has quit []
kit_ty_kate3 has quit []
mairacanal has quit []
Yamakaja has quit []
akemin-dayo has quit []
V_ has quit []
m42uko_ has quit []
VinDuv has quit []
tired- has quit []
deflax has quit []
flokli has quit []
povik has quit []
marcan has quit []
asie has quit []
rappet has quit []
kaprests has quit []
CME has quit []
paddatrapper_ has quit []
mjolnir[m] has quit []
matrix_ds has quit []
Retr0id has joined #asahi-gpu
rappet has joined #asahi-gpu
nyilas has joined #asahi-gpu
yamii has joined #asahi-gpu
jn has joined #asahi-gpu
ChaosPrincess has joined #asahi-gpu
linuxgemini1 has joined #asahi-gpu
yuka_ has joined #asahi-gpu
Misthios has joined #asahi-gpu
mkurz has joined #asahi-gpu
cy8aer has joined #asahi-gpu
Mrmaxmeier has joined #asahi-gpu
yrlf has joined #asahi-gpu
Mary6 has joined #asahi-gpu
VinDuv has joined #asahi-gpu
jole has joined #asahi-gpu
manawyrm has joined #asahi-gpu
kit_ty_kate3 has joined #asahi-gpu
mairacanal has joined #asahi-gpu
V_ has joined #asahi-gpu
Yamakaja has joined #asahi-gpu
m42uko_ has joined #asahi-gpu
tired- has joined #asahi-gpu
deflax has joined #asahi-gpu
flokli has joined #asahi-gpu
povik has joined #asahi-gpu
paddatrapper_ has joined #asahi-gpu
matrix_ds has joined #asahi-gpu
mjolnir[m] has joined #asahi-gpu
CME has joined #asahi-gpu
kaprests has joined #asahi-gpu
marcan has joined #asahi-gpu
asie has joined #asahi-gpu
matrix_ds has quit [Max SendQ exceeded]
matrix_ds has joined #asahi-gpu
<alyssa> OK. I think I have a more or less complete picture of bindless texture handles then
<alyssa> still mystified by bindless samplers.
<alyssa> namely
<alyssa> add the register (if present) to the 64-bit uniform (must be present) and that's the address of the texture descriptor in memory, bam, done.
<alyssa> Metal's texture handles are 32-bit offsets into the heap or something like that
<alyssa> byte offsets
<alyssa> multiples of 24
<alyssa> as for the samplers though
<alyssa> the sampler register is a 16-bit index, not a byte offset (!)
<alyssa> but it isn't at all clear to me where the sampler descriptors themselves live. presumably this is indexing into a global heap but I can't figure out where.
<alyssa> i feel like i'm getting warmer but this isn't clicking yet
<alyssa> either my Metal code is wrong or something is horribly broken in my decoder ..
<alyssa> I'm clearly missing something dumb
<alyssa> but, if changing the sampler descriptor is not changing memory as far as I can see, IDK what to make of that other than not mapping memory hard enough
<alyssa> IOGPU Compute word 38: # of bindless samplers + 1
<alyssa> IOGPU Compute word 39: # of bindless samplers
<alyssa> ??
<alyssa> (seeing as they each bump up when I create a bindless sampler even if it's not actually used)
<alyssa> and corresponding index in the 02345 buffer gets the second word replaced by 0x780 or 0x781
<alyssa> but that, doesn't tell me where the actual samplers live!
<alyssa> 02345 is the only interesting thing here but, I'm printing it and it's not telling me what I need
<alyssa> 02345 is clearly a table of 8-byte records of some kind
<alyssa> relating to the bindless samplers
<alyssa> but .. I'm not seeing where the cntents go
<alyssa> there'sjust, not enough entropy here
<alyssa> unless.
<alyssa> no this would be pathological even by AGX standards..
<alyssa> no it's not that either.
<alyssa> at a loss here, where is this missing memory
<alyssa> managed to provoke 0x783
balrog has quit [Read error: No route to host]
balrog has joined #asahi-gpu
balrog has quit [Quit: Bye]
balrog has joined #asahi-gpu
nopeslide13 has joined #asahi-gpu
<eric_engestrom> marcan (lina, alyssa): for the manjaro/alarm snafu with enabling asahi in the can't-work-upstream package, this reminds me that I was surprised by the choice to rename the mesa package in the asahi repo instead of naming it `mesa` so that it would shadow the one from alarm; perhaps this should still be done now to avoid this kind of issues in the future? (although I can't think of something else they could do)
nopeslide1 has quit [Ping timeout: 480 seconds]
bcrumb has joined #asahi-gpu
bcrumb has quit [Ping timeout: 480 seconds]
maximbaz has quit [Quit: bye]
maximbaz has joined #asahi-gpu
darkapex2 has joined #asahi-gpu
darkapex1 has quit [Ping timeout: 480 seconds]
bcrumb has joined #asahi-gpu
bcrumb has quit [Quit: WeeChat 3.8]
balrog has quit [Quit: Bye]
balrog has joined #asahi-gpu
nopeslide13 has quit []
nopeslide13 has joined #asahi-gpu
nyilas has quit [Remote host closed the connection]
yrlf has quit [Quit: The Lounge -]
yrlf has joined #asahi-gpu
pthariensflame has joined #asahi-gpu
pthariensflame has quit []
<jannau> eric_engestrom: I suppose the idea was to make it optional and easy to revert to the upstream version in the case we screw up