ChanServ changed the topic of #panfrost to: Panfrost - FLOSS Mali Midgard & Bifrost - Logs https://oftc.irclog.whitequark.org/panfrost - <macc24> i have been here before it was popular
WoC is now known as Guest1262
WoC has joined #panfrost
Guest1262 has quit [Ping timeout: 480 seconds]
vstehle1 has quit [Ping timeout: 480 seconds]
camus1 has quit [Remote host closed the connection]
camus has joined #panfrost
camus has quit [Remote host closed the connection]
camus has joined #panfrost
jambalaya has quit [Ping timeout: 480 seconds]
jambalaya has joined #panfrost
camus has quit [Remote host closed the connection]
camus has joined #panfrost
anarsoul has quit []
anarsoul has joined #panfrost
vstehle1 has joined #panfrost
camus1 has joined #panfrost
camus has quit [Read error: Connection reset by peer]
camus1 has quit [Remote host closed the connection]
camus has joined #panfrost
camus1 has joined #panfrost
camus has quit [Ping timeout: 480 seconds]
fahien24 has quit []
rasterman has joined #panfrost
fahien2 has quit []
italove has quit []
ndufresne has quit [Quit: The Lounge - https://thelounge.chat]
tomeu has quit [Quit: The Lounge - https://thelounge.chat]
fahien2 has joined #panfrost
fahien2 has quit [Remote host closed the connection]
italove has joined #panfrost
ndufresne has joined #panfrost
italove has quit [Remote host closed the connection]
ndufresne has quit [Remote host closed the connection]
italove has joined #panfrost
tomeu has joined #panfrost
abas has joined #panfrost
abas has quit [Remote host closed the connection]
camus has joined #panfrost
camus1 has quit [Ping timeout: 480 seconds]
fahien2 has joined #panfrost
fahien2 is now known as fahien
floof58_ has quit [Ping timeout: 480 seconds]
chewitt has quit [Ping timeout: 480 seconds]
floof58_ has joined #panfrost
floof58_ has quit [Ping timeout: 480 seconds]
camus1 has joined #panfrost
floof58_ has joined #panfrost
camus has quit [Ping timeout: 480 seconds]
Rathann has joined #panfrost
camus1 has quit [Remote host closed the connection]
camus has joined #panfrost
floof58_ has quit [Ping timeout: 480 seconds]
floof58_ has joined #panfrost
floof58_ has quit [Ping timeout: 480 seconds]
floof58 has joined #panfrost
camus1 has joined #panfrost
camus has quit [Ping timeout: 480 seconds]
Danct12 has joined #panfrost
nlhowell has joined #panfrost
floof58_ has joined #panfrost
floof58 has quit [Ping timeout: 480 seconds]
camus1 has quit [Ping timeout: 480 seconds]
camus has joined #panfrost
tomeu has quit []
ndufresne1 has joined #panfrost
ndufresne1 is now known as ndufresne
nlhowell is now known as Guest1315
nlhowell has joined #panfrost
<HdkR> Today I just learned about this forward pixel kill feature of Mali. I wonder how that works
<alyssa> HdkR: I think that blog explains it, no?
<alyssa> "Calculations already in flight can be terminated at any time if we spot that a later thread will write opaque data to the same pixel location... . By adding a simple FIFO buffer to the start of the pipeline, we can extend the forward pixel kill zone, making it more likely to spot overdraw, and at the same time giving the pipeline the chance to kill threads before they are even started."
<HdkR> What does it mean to "spot that a later thread will write opaque data" though?
<HdkR> Is the fifo just a store queue to the tile and it always takes the latest result at some indeterminate commit step?
Guest1315 has quit [Ping timeout: 480 seconds]
<HdkR> Which would mean it is completely a hardware construct and nothing to do on the compiler side?
nlhowell has quit [Ping timeout: 480 seconds]
<alyssa> 100% hardware
<alyssa> Software just gets (on Bifrost) a "can kill?" and "can be killed?" bits (like early-z some bits)
<alyssa> (e.g. a fragment shader that has SSBO writes cannot be killed, but it can still kill)
nlhowell has joined #panfrost
<HdkR> That's pretty cute
indy has quit []
indy has joined #panfrost
JulianGro has quit [Remote host closed the connection]
JulianGro has joined #panfrost
jrsharp has quit [Remote host closed the connection]
nlhowell has quit [Ping timeout: 480 seconds]
<cphealy> Would it be possible to have a method for disabling this like we have for disabling AFBC? The use case would be for various testing purposes.
camus has quit []
nlhowell has joined #panfrost
warpme_ has joined #panfrost
JulianGro has quit [Remote host closed the connection]
warpme_ has quit [resistance.oftc.net weber.oftc.net]
cphealy has quit [resistance.oftc.net weber.oftc.net]
bluebugs has quit [resistance.oftc.net weber.oftc.net]
ids1024 has quit [resistance.oftc.net weber.oftc.net]
remexre has quit [resistance.oftc.net weber.oftc.net]
WoC has quit [resistance.oftc.net weber.oftc.net]
tlwoerner has quit [resistance.oftc.net weber.oftc.net]
taowa has quit [resistance.oftc.net weber.oftc.net]
robink has quit [resistance.oftc.net weber.oftc.net]
philpax has quit [resistance.oftc.net weber.oftc.net]
kenzie has quit [resistance.oftc.net weber.oftc.net]
austriancoder has quit [resistance.oftc.net weber.oftc.net]
ezequielg has quit [resistance.oftc.net weber.oftc.net]
daniels has quit [resistance.oftc.net weber.oftc.net]
alyssa has quit [resistance.oftc.net weber.oftc.net]
suihkulokki has quit [resistance.oftc.net weber.oftc.net]
narmstrong has quit [resistance.oftc.net weber.oftc.net]
jolan has quit [resistance.oftc.net weber.oftc.net]
cphealy has joined #panfrost
bluebugs has joined #panfrost
austriancoder has joined #panfrost
remexre has joined #panfrost
ezequielg has joined #panfrost
robink has joined #panfrost
philpax has joined #panfrost
narmstrong has joined #panfrost
suihkulokki has joined #panfrost
alyssa has joined #panfrost
taowa has joined #panfrost
kenzie has joined #panfrost
daniels has joined #panfrost
jolan has joined #panfrost
ids1024 has joined #panfrost
tlwoerner has joined #panfrost
warpme_ has joined #panfrost
WoC has joined #panfrost
nlhowell has quit [Ping timeout: 480 seconds]
<alyssa> cphealy: Yes on Bifrost.
<alyssa> Unclear on midgard.
rasterman has quit [Quit: Gettin' stinky!]
<alyssa> bbrezillon: Rebasing the XML MR now
<alyssa> I think this is really close to landable, just needs to be passed over the finish line. I'm on it :)
<alyssa> The C changes look good anyway. So just some bug in the XML I guess
<alyssa> Hm
<alyssa> note to self: cleanup Texture Descriptor Is 64b
<alyssa> ok, that explains v5 being horribly broken
<alyssa> I guess really I should just audit all 4 XMLs
<alyssa> instead of trying to play cute games with the diffs