<anarsoul>
yuq825: looks like there's some issue with reusing plb pp stream :\
<anarsoul>
I get weird misrendering in X11 with it enabled
<anarsoul>
but it's fixed if I call lima_generate_pp_stream() after retrieving pp_stream bo from cache
<anarsoul>
OK, fixed it
<anarsoul>
we also need to add fb params to pp stream key
<yuq825>
I get this a bit more thinking, we are trading CPU time building pp_stream for GPU time and memory bandwidth executing tasks, a bound box pp_stream cache trade some GPU time for CPU time back because the bound box is not accurate, the problem is the bound box accuracy (we may even use a pp_stream whose bound box can cover current task's bound box)
<yuq825>
that how to choose bound box
<yuq825>
how to choose existing pp_stream in the cache
<yuq825>
maybe a coverage rate of current task's damage area can be used to choose the pp_stream in the cache
<yuq825>
not exact bound box match
<yuq825>
when the rate is low, we may re-build pp_stream
<yuq825>
accurately
<anarsoul>
yuq825: it just gets too complicated, so I'm not sure we'll get much of out it
<yuq825>
it's OK we just have accurate bound box based pp_stream cache first, we need it anyway, we may optimize it by this idea in the future
<yuq825>
just remind this bound method does not only have memory trade off, also GPU time trade off
Net147 has joined #lima
<anarsoul>
yuq825: yeah, I guess it's impossible to find perfect solution here
<anarsoul>
I think bounding box should work well for desktop when usually it's a single window that gets updated
<anarsoul>
yuq825: btw please merge your multi-job MR
<anarsoul>
it seems to be stable and since it's quite massive and reworks a lot it would be nice to get it merged
<anarsoul>
otherwise there will be conflicts with other MRs
<yuq825>
ok, I'll merge your kernel fix first, then this MR
yuq825 has quit [Remote host closed the connection]
yuq825 has joined #lima
dddddd has quit [Ping timeout: 268 seconds]
megi has quit [Ping timeout: 268 seconds]
Barada has joined #lima
buzzmarshall has quit [Remote host closed the connection]
kaspter has quit [Remote host closed the connection]
champagneg has quit [Ping timeout: 240 seconds]
kaspter has joined #lima
champagneg has joined #lima
yann has quit [Ping timeout: 272 seconds]
champagneg has quit [Ping timeout: 265 seconds]
champagneg has joined #lima
buzzmarshall has joined #lima
monstr has quit [Remote host closed the connection]
Net147 has quit [Ping timeout: 265 seconds]
champagneg has quit [Ping timeout: 265 seconds]
champagneg has joined #lima
champagneg has quit [Quit: WeeChat 2.3]
<rellla>
anarsoul: i am one step further with dEQP-GLES2.functional.depth_stencil_clear.depth_stencil_masked :)
<rellla>
have to stop thinking now, but this is what i found out:
<rellla>
i logged stencil values. so let's say we have 0x67 as the stencil value after the last clear, which is the last action that writes stencil values
<rellla>
so the (stencil) value i get in the resulting image is always the ndx-1 one, but it should be the ndx value instead
<rellla>
i will think about that later, but maybe you have an idea in the meantime. maybe we use the wrong rsw for drawing or there is some off-by-one issue somewhere ...
yann has quit [Ping timeout: 268 seconds]
<rellla>
to complete the number example: if stencil value to test again is 0x67, we test 0x46 with LEQUAL which succeeds and is written to the image
<rellla>
the next is 0x55 which should also succeed, but that one and the following ones of course do not succeed or aren't written to the image as G value
anarsoul|2 has joined #lima
z3ntu_ has quit [Ping timeout: 246 seconds]
anarsoul has quit [Ping timeout: 240 seconds]
mariogrip has quit [Ping timeout: 248 seconds]
Viciouss has quit [Ping timeout: 240 seconds]
Viciouss has joined #lima
mariogrip has joined #lima
mariogrip has quit [Quit: killed]
anarsoul|2 is now known as anarsoul
jernej has quit [Remote host closed the connection]