ChanServ changed the topic of #lima to: Development channel for open source lima driver for ARM Mali4** GPUs - Kernel driver has landed in mainline, userspace driver is part of mesa - Logs at https://oftc.irclog.whitequark.org/lima/
camus has joined #lima
camus1 has quit [Ping timeout: 480 seconds]
mripard has quit [Ping timeout: 480 seconds]
mripard has joined #lima
camus1 has joined #lima
camus has quit [Ping timeout: 480 seconds]
<rellla> anarsoul: feel free to pick it up
<enunes> anarsoul: yeah I tried to implement that at some point back in https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/1357 , that was tricky
<enunes> eventually I gave up on my implementation that was too complicated and not greatly optimized, to just enable the nir lowering to enable more applications and tests
<enunes> I think one of the problems was that the varying separation happened pretty early in the mesa shader pipeline and it was not very easy to merge it back at the ppir stage
<enunes> but maybe I missed something back then
<enunes> hmm yeah looks like the other problem may be that the perspective division not necessarily comes from a varying
<enunes> but it might be good enough to just apply the "optimization" in case we detect that it does come from a varying
camus has joined #lima
camus1 has quit [Read error: Connection reset by peer]
anarsoul has quit [Read error: Connection reset by peer]
anarsoul has joined #lima
camus1 has joined #lima
camus has quit [Ping timeout: 480 seconds]
camus has joined #lima
camus1 has quit [Ping timeout: 480 seconds]
Net147 has quit [Quit: Quit]
Net147 has joined #lima
adjtm_ has quit []
adjtm has joined #lima
camus1 has joined #lima
camus has quit [Ping timeout: 480 seconds]
camus has joined #lima
camus1 has quit [Ping timeout: 480 seconds]
seraglios has joined #lima
camus1 has joined #lima
camus has quit [Ping timeout: 480 seconds]
seraglios has quit [autokilled: This host has violated network policy. Mail support@oftc.net if you feel this is in error. (2021-09-23 15:35:09)]
camus1 has quit [Read error: Connection reset by peer]
camus has joined #lima
drod has joined #lima
<anarsoul> enunes: exactly my idea, we don't need to make it perfect, just detect if it comes from the same varying and apply optimization
<anarsoul> enunes: btw, do you know why projector is split into separate source in nir?
camus1 has joined #lima
camus has quit [Read error: Connection reset by peer]
<enunes> anarsoul: I dont remember where it is split, if I recall correctly when I tried to trace it, it seemed to be even before it reaches nir, maybe in glsl ir
<enunes> if I didnt miss anything and that is indeed the case, I suppose it is just that other drivers never cared about keeping it in the same varying
<anarsoul> enunes: well, it doesn't depend on frontend whether it's in the same varying :)
<anarsoul> it's up to the shader
drod has quit [Remote host closed the connection]
camus has joined #lima
camus1 has quit [Read error: Connection reset by peer]