<chewitt>
there are also midgard/bifrost drivers in the LE github
<chewitt>
but might not build against 5.9+ kernels as we switched to lima/panfrost
<chewitt>
prob. not much difference tho
<asdf28>
chewitt: yes, it's mali utgard, thanks for the link. i have already tried this driver
<asdf28>
but something is wrong when it comes to 3D performance using the newer drivers
<chewitt>
new ARM drivers, or lima?
<asdf28>
whereas the legacy kernel runs it silky smooth
<asdf28>
both
<asdf28>
i have tried the ARM wayland opennGL drivers and mesa openGL
<chewitt>
sounds like a rock > you < hard-place situation
<asdf28>
and both kernel drivers, mali and lima
<asdf28>
they all work, but slow
<asdf28>
i just wanted to test if it's the kernel maybe
<asdf28>
so it would be interesting to use the newer drivers with the old kernel
<asdf28>
and i saw that in the github repository linked above, someone backported the DRM driver to it, that would be all that's needed i think
<asdf28>
it's not relevant for LibreELEC though, because there are no 3D games in there
<asdf28>
but with a fast enough 3D driver, lakka could be ported to use a mainline kernel as well
yann has joined #linux-amlogic
yann has quit [Ping timeout: 260 seconds]
<chewitt>
the trick is to find a repeatable test with lima, and provide feedback to the devs on what fails
<chewitt>
then they can target something specific to fix
<asdf28>
it's weird because it does not seem to be related to raw CPU or GPU performance. i remember that i fiddled with CPU and GPU clock speeds and it had no noticeable effect
<asdf28>
there's also nothing that "fails", so to speak
<asdf28>
it's just slow
<chewitt>
ahh, I guess the current focus is conformance not performance
<chewitt>
performance usually comes later once you've implemented a complete(r) system
<asdf28>
i wouldn't expect it to be as fast as the proprietar driver though
<asdf28>
yes
<asdf28>
the openGL DRM driver from ARM is the one that i'm wondering about
<asdf28>
the one that's called "libmali" in the libreelec packages
<chewitt>
long-term I'd expect lima to be faster; less weird outdated stuff to be done in code
<asdf28>
because it's also very slow compared to the legacy drivers that the board comes with
<chewitt>
lima is still more of a community effort, panfrost has commercial interest behind it that drive focus
<chewitt>
i'm sure it will get there, but, things take time
Barada has quit [Read error: Connection reset by peer]
Barada has joined #linux-amlogic
kaspter has quit [Remote host closed the connection]
kaspter has joined #linux-amlogic
<asdf28>
one thing that i noticed is that the 3D performance takes a big hit once the driver is using DRM
<asdf28>
the older drivers all use the framebuffer device
<asdf28>
but i can't directly compare them because i can't get the FB drivers to work on newer kernels
<asdf28>
so i can't compare the drivers using the same kernel
<asdf28>
this is why i wanted to try the DRM drivers using the legacy kernel
<asdf28>
if it's slow there as well, this would rule out everything else
chewitt has quit [Ping timeout: 256 seconds]
Guest27673 has quit [Ping timeout: 272 seconds]
camus has joined #linux-amlogic
kaspter has quit [Ping timeout: 272 seconds]
camus is now known as kaspter
chewitt has joined #linux-amlogic
Barada has quit [Quit: Barada]
warpme_ has joined #linux-amlogic
Darkmatter66 has quit [Ping timeout: 260 seconds]
ldevulder_ has joined #linux-amlogic
ldevulder has quit [Ping timeout: 240 seconds]
kaspter has quit [Quit: kaspter]
ldevulder_ is now known as ldevulder
<narmstrong>
comparing Lima and the ARM Mali acceleration on fbdev is not fair
<narmstrong>
xdarklight: I'm back from vacation, we can discuss to solve your issues !
<narmstrong>
xdarklight: hmm, *our* issues :-)
<asdf28>
no, i was comparing the ARM fbdev driver with the ARM DRM driver
<asdf28>
the DRM OpenGL driver is a lot slower
<asdf28>
and i was wondering if the fbdev driver is faster by itself or because of something else, such as the legacy kernel
<narmstrong>
"fast" how ?
<narmstrong>
DRM will respect the display vsync, fbdev won't necessarely
<narmstrong>
the fbdec "stack" is allmost inexistant, thus you have maximum performance
yann has joined #linux-amlogic
<asdf28>
it's 20% faster in some 3D games, i'd say
<narmstrong>
in fbdev ? what kind of games did you manage to run on fbdev ? emulators ?
<narmstrong>
against what ? wayland ? or gbm ?
<asdf28>
yes only emulators
plntyk has joined #linux-amlogic
<asdf28>
i have also tried glmark, but the fbdev driver does not support some tests of it it seems
<asdf28>
i compared it to the ARM wayland driver running in KMS mode (wayland is not involved in that mode)
<asdf28>
i think that would be gbm then?
<narmstrong>
yes, wayland libmali can work in gbm mode without the wayland compositor
<asdf28>
if one would backport the meson DRM driver to the legacy kernel, you think it wouldn't outperform the old fbdev driver?
<narmstrong>
no, it would not be faster
<asdf28>
oh, that's interesting. i wasn't sure about that, but i wanted to try it
sputnik has joined #linux-amlogic
sputnik is now known as Guest4809
yann has quit [Ping timeout: 260 seconds]
buzzmarshall has joined #linux-amlogic
warpme_ has quit [Quit: Connection closed for inactivity]
<xdarklight>
hi narmstrong - welcome back!
<xdarklight>
narmstrong: so I basically have multiple topics: 1) my understanding is that some of the meson-dw-hdmi code should be organized differently. this would help me as the code to integrate with the VPU registers is almost identical on the 32-bit SoCs, which don't use the DW HDMI controller though
<xdarklight>
narmstrong: 2) common clock framework support for the video clocks and 3) any topic that comes up along the way ;)
<xdarklight>
narmstrong: for 1) and 2) I have patches to "make it work", but it's not in a state where I can upstream it. so now that I've proven that I can make it work I'm curious to find the next steps with you. I believe that some of these steps will also make the drm maintainers happy/happier, because AFAIU some decoupling - related to 1) for example - is also what they want
cmeerw has joined #linux-amlogic
yann has joined #linux-amlogic
yann has quit [Ping timeout: 260 seconds]
<narmstrong>
xdarklight: indeed, 1) would be useful for you hdmi controller and a base for the rework of the dsi integration
<narmstrong>
for 2) we would still need to control a lot of the clock muxes to avoid CCF to use a random clock path
<xdarklight>
narmstrong: when do you have time to work on a plan for all of this? also how do you like to discuss this (like: here on IRC, some kind of phone call, web meeting, ...)?
<narmstrong>
xdarklight: I have currently a lot of work on non amlogic stuff so it’s hard to answer
<narmstrong>
And usually I’m not very available in the evening, so maybe we could start sketching stuff by email then discuss via irc or slack ?
<xdarklight>
narmstrong: sure. I think it makes sense to start by discussing the reorganization part first (I think it's mostly independent of the CCF stuff)
Darkmatter66 has joined #linux-amlogic
Anessen97_0 has joined #linux-amlogic
Guest4809 has quit [Remote host closed the connection]
Guest4809 has joined #linux-amlogic
yann has joined #linux-amlogic
yann has quit [Ping timeout: 256 seconds]
TheAssassin has quit [Quit: No Ping reply in 180 seconds.]
TheAssassin has joined #linux-amlogic
warpme_ has joined #linux-amlogic
lvrp16 has quit [Ping timeout: 260 seconds]
lvrp16 has joined #linux-amlogic
yann has joined #linux-amlogic
yann has quit [Ping timeout: 256 seconds]
cmeerw has quit [Ping timeout: 258 seconds]
ldevulder has quit [Read error: Connection reset by peer]