<i509vcb>
I'm not sure why push constants seem to not be lowered in one of the tests I am running
ourdumbfuture has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
<i509vcb>
Hmm deref_var &pushConstants (push_const PushConstants) seems to be getting lowered after I do sysvals?
ourdumbfuture has joined #asahi-gpu
<alyssa>
one of the lower_explicit_io's IIRC
<alyssa>
your pass order might need some tweaks
cr1901_ is now known as cr1901
<i509vcb>
alyssa: lower sysvals is done. Everything lives in the sysvals buffer now
kujeger has quit [Ping timeout: 480 seconds]
ourdumbfuture has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
<alyssa>
Woohoo!
<alyssa>
i509vcb: needs a rebase, the meta_begin stuff landed already (I merged ella's commit)
aratuk has joined #asahi-gpu
ourdumbfuture has joined #asahi-gpu
aratuk has quit []
novafacing9926 has joined #asahi-gpu
ourdumbfuture has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
novafacing992 has quit [Ping timeout: 480 seconds]
novafacing9926 is now known as novafacing992
aratuk has joined #asahi-gpu
JTL has joined #asahi-gpu
aratuk has quit []
kujeger has joined #asahi-gpu
ourdumbfuture has joined #asahi-gpu
hightower4 has joined #asahi-gpu
hightower3 has quit [Ping timeout: 480 seconds]
ourdumbfuture has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
<i509vcb>
alyssa: my branch is up to date, I just ended up filling in a few missing meta_begins I guess while testing things
jeisom has quit [Ping timeout: 480 seconds]
<alyssa>
i509vcb: split out the patch then :)
<alyssa>
thx
<i509vcb>
ok
<alyssa>
ella-0: please review i509vcb's MR when you get a chance, thank you!
possiblemeatball has quit [Quit: Quit]
cylm has quit [Ping timeout: 480 seconds]
tertu2 has quit [Quit: so long...]
crabbedhaloablut has joined #asahi-gpu
<lina>
alyssa: New kernel debug_flags bit 0x2000 to verbosely explain UAPI usage errors (you probably want that on always by default)
tertu has joined #asahi-gpu
tertu has quit [Ping timeout: 480 seconds]
tertu has joined #asahi-gpu
zhdi has joined #asahi-gpu
Major_Biscuit has joined #asahi-gpu
Major_Biscuit has quit []
MajorBiscuit has joined #asahi-gpu
zhdi has quit [Quit: Connection closed for inactivity]
chadmed has quit [Ping timeout: 480 seconds]
bisko has joined #asahi-gpu
kujeger has quit [Ping timeout: 480 seconds]
<lina>
alyssa: When you're around, there's a new Dolphin issue seemingly related to MSAA+Depth that goes away when I flush after every draw, so I think it's shader related ^^
bisko has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
jeisom has joined #asahi-gpu
chadmed has joined #asahi-gpu
chadmed has quit [Remote host closed the connection]
chadmed has joined #asahi-gpu
cr1901_ has joined #asahi-gpu
cr1901 has quit [Ping timeout: 480 seconds]
ourdumbfuture has joined #asahi-gpu
duckworld has quit [Ping timeout: 480 seconds]
ourdumbfuture has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
minecrell has quit [Quit: Ping timeout (120 seconds)]
minecrell has joined #asahi-gpu
ourdumbfuture has joined #asahi-gpu
<alyssa>
alright
alyssa has quit [Quit: alyssa]
alyssa has joined #asahi-gpu
alyssa has quit []
cr1901__ has joined #asahi-gpu
cr1901_ has quit [Ping timeout: 480 seconds]
alyssa has joined #asahi-gpu
kujeger has joined #asahi-gpu
nightquest has quit [Ping timeout: 480 seconds]
nightquest has joined #asahi-gpu
ourdumbfuture has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
ourdumbfuture has joined #asahi-gpu
bisko has joined #asahi-gpu
alyssa has quit [Quit: alyssa]
bisko has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
bisko has joined #asahi-gpu
Major_Biscuit has joined #asahi-gpu
MajorBiscuit has quit [Ping timeout: 480 seconds]
cylm has joined #asahi-gpu
cr1901_ has joined #asahi-gpu
cr1901__ has quit [Ping timeout: 480 seconds]
cr1901_ is now known as cr1901
bisko has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
ourdumbfuture has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
Major_Biscuit has quit [Ping timeout: 480 seconds]
alyssa has joined #asahi-gpu
iaguis has joined #asahi-gpu
ourdumbfuture has joined #asahi-gpu
<i509vcb>
alyssa: For indirect compute, I probably use CDM_INDIRECT instead of CDM_GLOBAL_SIZE? (Since that needs the group count at recording time)
<alyssa>
i509vcb: yes
duckworld has joined #asahi-gpu
<alyssa>
see the corresponding gallium driver implementation
rhysmdnz has quit [Quit: Bridge terminating on SIGTERM]
Guest942 has quit [Quit: Bridge terminating on SIGTERM]
rhysmdnz has joined #asahi-gpu
Guest1229 has joined #asahi-gpu
Major_Biscuit has joined #asahi-gpu
Major_Biscuit has quit []
iaguis has quit [Quit: leaving]
<i509vcb>
If vkCmdDispatch gets 0 in any groupCount dimension. is there really any reason to record the command in the first place?
ave36309 has quit [Ping timeout: 480 seconds]
iaguis has joined #asahi-gpu
<i509vcb>
Valid usage doesn't forbid 0
<alyssa>
no reason
<i509vcb>
dEQP-VK.compute.basic.empty_workgroup_all seem to be fine with me just immediately returning, so I guess thats valid from the cts
jeisom has quit [Ping timeout: 480 seconds]
ave36309 has joined #asahi-gpu
<i509vcb>
agx seems to not like negative height, hopefully I don't need to do lowering in shaders for this...