MoeIcenowy has quit [Quit: ZNC 1.6.5+deb1+deb9u1 - http://znc.in]
MoeIcenowy has joined #panfrost
raster has joined #panfrost
MoeIcenowy has quit [Quit: ZNC 1.6.5+deb1+deb9u1 - http://znc.in]
MoeIcenowy has joined #panfrost
MoeIcenowy has quit [Quit: ZNC 1.6.5+deb1+deb9u1 - http://znc.in]
MoeIcenowy has joined #panfrost
NeuroScr has quit [Quit: NeuroScr]
stikonas has joined #panfrost
narmstrong has quit [Ping timeout: 268 seconds]
narmstrong has joined #panfrost
jernej has quit [Read error: Connection reset by peer]
stikonas has quit [Ping timeout: 258 seconds]
milloni_ is now known as milloni
<tomeu>
alyssa: sent the CI patch, but it's stuck in mesa-dev moderation, due to the size of the file with the expectations
<tomeu>
tell me if you would prefer a MR, or how do you want to do the review
<alyssa>
tomeu: No, I definitely prefer email for review (gitlab fails to render diffs this big on my hardware thanks to the shear amount of JavaScript, aparently..)
<alyssa>
Although I can't promise I'll get to it today, I did get the email and will look through soon! :)
<daniels>
alyssa: MR rendering performance has improved a lot recently
<alyssa>
daniels: That's good to hear, but I am still using an Arm Chromebook with no GPU accel ;)
afaerber has quit [Quit: Leaving]
herbmillerjr has joined #panfrost
<daniels>
by 'rendering performance', I mean the time you spend in frontend JavaScript
afaerber has joined #panfrost
raster has quit [Ping timeout: 250 seconds]
raster has joined #panfrost
jernej has joined #panfrost
<alyssa>
Oh, I'll have to try it again then :)
pH5 has joined #panfrost
raster has quit [Remote host closed the connection]
herbmilleriw has joined #panfrost
belgin has joined #panfrost
narmstrong has quit []
narmstrong has joined #panfrost
belgin has quit [Read error: Connection reset by peer]
belgin has joined #panfrost
belgin has quit [Client Quit]
stikonas has joined #panfrost
afaerber has quit [Quit: Leaving]
afaerber has joined #panfrost
herbmilleriw has quit [Remote host closed the connection]
herbmilleriw has joined #panfrost
stikonas has quit [Read error: Connection reset by peer]
stikonas_ has joined #panfrost
indy has quit [Read error: Connection reset by peer]
stikonas_ is now known as stikonas
herbmilleriw has quit [Client Quit]
indy has joined #panfrost
herbmilleriw has joined #panfrost
NeuroScr has joined #panfrost
NeuroScr has quit [Quit: NeuroScr]
NeuroScr has joined #panfrost
mateo` has quit [Quit: WeeChat 2.4]
mateo` has joined #panfrost
<anarsoul>
alyssa: ping
<anarsoul>
alyssa: does mpv work for you with mesa master?
<alyssa>
anarsoul: pong
* alyssa
tris
<anarsoul>
if yes - can you please dump nir fs shader for me? :)
<alyssa>
anarsoul: Apparently not.
<alyssa>
Though I probably regressed stuff recently
<anarsoul>
alyssa: out of curiosity, do you get 2 undefs in fs shader?
<alyssa>
anarsoul: Nope?
<anarsoul>
I wonder where they come from...
<anarsoul>
apparently some lowering pass is buggy
<alyssa>
Mine's broken due to overzealous copyprop, whoops
<alyssa>
Ugh, too many regressions
<anarsoul>
:(
<alyssa>
..Not going to lie, I'm excited for CI :)
* alyssa
bisects
<alyssa>
Pretty recent issue, have a good commit
<alyssa>
So many bugs :(
<anarsoul>
hehe
<anarsoul>
could you dump nir fs shader for me for mpv with good commit please?
<alyssa>
Don't be so sure it's a good commit yet
NeuroScr has quit [Quit: NeuroScr]
<alyssa>
"mask & (mask + 1)"
<alyssa>
The code I just wrote is too clever for me to understand it :(
<urjaman>
i thiiink that's a test for it not being an actual power of 2?
<urjaman>
that is, if it is 2^n - 1, that will result in a zero...
<alyssa>
urjaman: Test for it not being 2^n-1
<urjaman>
yeah
<alyssa>
urjaman: The context is wanting to test for a nontrivial mask
<alyssa>
Something other than x/xy/xyz/xyzw
* alyssa
rants
* HdkR
still wants popcount in AArch64
<anarsoul>
alyssa: looks like you have wallpapering disabled in panfrost -- but how do you use sway then? :)
<alyssa>
anarsoul: Disabling buffer age hackily
ezequielg has quit [Quit: leaving]
<anarsoul>
I see.
<anarsoul>
I'd better go fix wallpapering for lima (it works, but it skips reloading buffer for some cases)