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
cylm_ has quit [Quit: WeeChat 4.0.4]
amarioguy has joined #asahi-gpu
possiblemeatball has quit [Quit: Quit]
jeisom has quit [Ping timeout: 480 seconds]
crabbedhaloablut has joined #asahi-gpu
marshmallow has quit [Remote host closed the connection]
marshmallow has joined #asahi-gpu
_jannau__ is now known as _jannau_
elros has joined #asahi-gpu
eldriitch has joined #asahi-gpu
jeisom has joined #asahi-gpu
elros_ has joined #asahi-gpu
elros has quit [Ping timeout: 480 seconds]
cylm has joined #asahi-gpu
darkapex3 has quit []
darkapex has joined #asahi-gpu
possiblemeatball has joined #asahi-gpu
skippypaws has quit [Read error: Network is unreachable]
sa1 has quit [Read error: Network is unreachable]
sa1 has joined #asahi-gpu
skippypaws has joined #asahi-gpu
hayden_dev has quit [Read error: Network is unreachable]
azenla has quit [Read error: Network is unreachable]
hayden_dev has joined #asahi-gpu
daniels has quit [Read error: Network is unreachable]
azenla has joined #asahi-gpu
jonmasters has quit [Write error: connection closed]
robher has quit [Read error: Network is unreachable]
austriancoder has quit [Read error: Network is unreachable]
i509vcb has quit [Read error: Network is unreachable]
nightstrike has quit [Read error: Network is unreachable]
i509vcb has joined #asahi-gpu
austriancoder has joined #asahi-gpu
eric_engestrom has quit [Read error: Network is unreachable]
daniels has joined #asahi-gpu
Whistler_ has quit [Remote host closed the connection]
eric_engestrom has joined #asahi-gpu
joshtaylor has quit [Remote host closed the connection]
joshtaylor has joined #asahi-gpu
Whistler_ has joined #asahi-gpu
jonmasters has joined #asahi-gpu
robher has joined #asahi-gpu
nightstrike has joined #asahi-gpu
balrog has quit [Remote host closed the connection]
balrog has joined #asahi-gpu
alyssa has quit [Quit: alyssa]
c10l has quit [Quit: Bye o/]
c10l has joined #asahi-gpu
karolherbst has quit [Ping timeout: 480 seconds]
alyssa has joined #asahi-gpu
<alyssa> i509vcb: rebased agxv/main, that branch has layered rendering now
<alyssa> smoke tested with vkcube but did not CTS
<alyssa> good luck / have fun :)
<i509vcb> neat, I guess I'm wondering what mistake I was making during rebase as to make every commit be "click incoming changes" in editor
<i509vcb> I guess I'll try to see if the layered rendering stuff gives me working blits
karolherbst has joined #asahi-gpu
<alyssa> i509vcb: keep in mind i didn't do agxv integration, I just mean that the core compiler/common support is merged in by virtue of the rebase :)
alyssa has quit [Quit: alyssa]
<i509vcb> Yeah I suspected that
elros_ has quit [Quit: WeeChat 4.0.4]
<ella-0> layered rendering!!! cool. Currently in the process of moving to uni so not gonna be able to do much for a couple weeks. If no ones got to that by the time I'm settled I'll look into agxv integration.
<ella-0> Also anyone got some advice on how I should handle occlusion query availability? I'm assuming it'll be an extra compute pass that fills in an availability buffer but not sure.
Guest328 has quit [Quit: Bridge terminating on SIGTERM]
rhysmdnz has quit [Quit: Bridge terminating on SIGTERM]
Jamie has joined #asahi-gpu
Jamie is now known as Guest641
rhysmdnz has joined #asahi-gpu
flom84 has joined #asahi-gpu
flom84 has quit [Ping timeout: 480 seconds]
<i509vcb> ella-0: I think part of implementing CmdBlit with vk_meta requires layered rendering, so I guess I'll probably get a lot of that done
delroth has quit [Remote host closed the connection]
delroth has joined #asahi-gpu
<ella-0> Yeah it does
ourdumbfuture has joined #asahi-gpu
<i509vcb> Still patiently waiting for collabora to fix image copies in vk meta of course
<i509vcb> (alyssa said that was something still wip)
<Mary> Will queue another CTS run tonight and diff stuffs a bit compared to my last run
<i509vcb> I'm not 100% familar with layered rendering in vk, how would you typically indicate that to the pipeline?
<i509vcb> s/pipeline/shader
<i509vcb> CmdBeginRendering will know the number of layers because of binding the framebuffer
jeisom has quit [Ping timeout: 480 seconds]
darkapex has quit [Remote host closed the connection]
darkapex has joined #asahi-gpu
<i509vcb> descriptor set binding of VK_DESCRIPTOR_TYPE_SAMPLED_IMAGE with descriptorCount > 1?
rappet has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
rappet has joined #asahi-gpu
<i509vcb> this is going several layers deep in vulkan CreateInfo types...
stek has joined #asahi-gpu
chadmed has quit [Read error: Connection reset by peer]
chadmed has joined #asahi-gpu
jeisom has joined #asahi-gpu
stek has quit [Quit: Leaving]
ourdumbfuture has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
crabbedhaloablut has quit []
ourdumbfuture has joined #asahi-gpu
alyssa has joined #asahi-gpu
<alyssa> i509vcb: it doesn't look like you necesarily have that information but you dont really need it
<alyssa> when compiling frag shaders, just asssume the framebuffer is always layered
<alyssa> the only case where it would matter is for spilled render targets
<alyssa> but if you just always use a 2D Array PBE descriptor for that (with layers=1 if necessary) that should still work out fine, at a slight cost to instruction count in the spilled path
<alyssa> but if you're spilling attachments you're already screwed so that seems like a reasonable tradeoff to me lol
<alyssa> the layered bool matters a lot more for the background and EOT programs, but those are independent of pipeline and you should have that information available for them just fine, I think
<alyssa> I should probably simplify this in the gl driver but whatever, it's not really hurting anything there..
<alyssa> kapeesh? :)
<alyssa> 5
chadmed has quit [Remote host closed the connection]
chadmed has joined #asahi-gpu
ourdumbfuture has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
ourdumbfuture has joined #asahi-gpu
<i509vcb> Sure I guess that makes sense
<alyssa> :+1:
<i509vcb> Although I am definitely still confused as to why agx_nir_predicate_layer seems to never lower load_layer_id in the blit meta shader
<i509vcb> The meta shader uses an actual raw load_layer_id
<alyssa> i509vcb: run nir_lower_sysvals_to_varyings with the appropriate modification to support layer
<alyssa> before the predication pass etc
<alyssa> for us LAYER is a varying, not a sysval
<i509vcb> sysvals_to_varyings has no layer option?
<i509vcb> Or I guess I am adding new things
ourdumbfuture has quit [Quit: My Mac has gone to sleep. ZZZzzz…]