<forkbomb>
mszyprow_: originally the issue was that after setting up the clocks for the first time, the wm1811 never stops using FLL1 as SYSCLK, and so it can't change the rate of FLL1
<forkbomb>
that meant that the first audio session would work, but any subsequent ones wouldn't, unless they used the same rates as the first one
<forkbomb>
my "hack" was to always switch to FLL2 before configuring sysclk
<forkbomb>
with that, audio works normally once you `aplay` something
<forkbomb>
but yeah, i will check the clock drivers. thanks for the pointer
<memeka>
mszyprow_: ping again :D
<mszyprow_>
memeka: pong
<memeka>
mszyprow_: GScaler (via internal path) integration to Exynos DRM / FIMD / DECON as video plane (Exynos 542x/5433)
<memeka>
is this the same as what i was asking a while ago?
<memeka>
memeka> is there any plans on implementing multiple planes and supporting NV12/NV21 on exynos54xx DRM (in order to use DRMPRIME)?
<mszyprow_>
memeka: yes
<memeka>
and is it close to the top of the queue or somewhere deep? :D
<mszyprow_>
memeka: rather close to the top, we are checking how this works in old reference kernels
<memeka>
mszyprow_: cool let me know if you need tests :P
<mszyprow_>
memeka: this was used only in very old kernels, probably only on exynos5250
<mszyprow_>
memeka: when we will have anything to test, I will let you know
<memeka>
thanks
<mszyprow_>
forkbomb: if you need any help related to codecs/i2s/clocks, feel free to ask snawrocki
<forkbomb>
ok
<mszyprow_>
forkbomb: asoc is the one of few kernel subsystems which I have completely no idea how it works
<forkbomb>
me neither, to be honest :P
<forkbomb>
btw, i don't know if you have access to the 5420 TRM, but i think the dsim PLLTMR register is wrong in mainline
<forkbomb>
the vendor kernel i have says it's 0x50
<forkbomb>
(it's pretty annoying that samsung doesn't release datasheets, would be much easier to check it myself...)
<mszyprow_>
forkbomb: let me check...
<mszyprow_>
forkbomb: afair there is no exynos54xx board in mainline, which use dsi display
<forkbomb>
no, there's not
<forkbomb>
i'll add bringing up my 5420 board to the TODO list, i guess
<mszyprow_>
forkbomb: according to TRM, it should be 0x58
<forkbomb>
ah, yeah... i was looking in regs-dsim.h instead of regs-mipidsim.h
<mszyprow_>
forkbomb: 0x4C is PLLCTRL, 0x50 is PLLCTRL1 and 0x54 is PLLCTRL2
<mszyprow_>
well, includes in vendor kernels are ugly mess
<forkbomb>
indeed
<forkbomb>
thanks for checking!
<mszyprow_>
especially if they are not preprocessed with something like unifdef tool
<forkbomb>
mszyprow_: that looks like a very useful tool
<mszyprow_>
yes, I use it often when I try to extract something from vendor code
<mszyprow_>
helps to keep myself sane when parsing multilevel conditional code
<forkbomb>
heh
<mszyprow_>
forkbomb: which 5420based board do you want to play with?
<forkbomb>
mszyprow_: it's called "ha3g". Galaxy Note 3 3G
<mszyprow_>
okay, I never seen that one
<forkbomb>
it seems to be a pretty niche device
<forkbomb>
but it uses the same pmic as peach, so that's nice
<mszyprow_>
memeka: btw, there are already 2 rgb planes in exynos drm mixer crtc device
<mszyprow_>
memeka: although the second is labelled as 'cursor' plane, it is a full featured one - its size can be up to full-hd and supports crop/overlay
<memeka>
mszyprow_: last time i checked, i could not get far because there was no support for NV12/NV21 on drm...
<mszyprow_>
memeka: right, the buffer must be rgb16/24/32 not nv12
<mszyprow_>
but the second plane and support for it is already there ;P
<memeka>
mszyprow_: right, so then it's just adding the gscaler in the mix, to seamlessly convert nv12
<mszyprow_>
memeka: "just"
<memeka>
:)
<memeka>
i remember i could play video to DRM using gscaler in the pipeline, but ideally it should work w/o that
kxkamil has joined #linux-exynos
<memeka>
mszyprow_: also i remember that gstreamer was showing nv12/21 as accepted by the DRM, but then it was failing
<mszyprow_>
memeka: some older exynos drm lacked proper check for available hw blocks
<mszyprow_>
memeka: and always reported nv12 plane regardless the hw support
<memeka>
mszyprow_: this was with kernel 4.14 :P
<mszyprow_>
strange
<memeka>
it's weird, since checking with a tool from libdrm, nv12 was now shown in the list of formats
<mszyprow_>
maybe you had exynos drm vidi enabled in the config
<memeka>
but gst-inspect was showing it
<mszyprow_>
it reports both rgb and nv12 as supported
<mszyprow_>
but this is just a 'driver' for virtual device
<mszyprow_>
frankly it should never reach mainline in the current state
<mszyprow_>
as it is not really useful for anything
<memeka>
don't think i had, but not sure
<memeka>
like i said, modetest from libdrm was showing the correct stuff
<memeka>
maybe it was a gstreamer thing
<mszyprow_>
otherwise I see no way to get nv12 as supported on exynos 542x drm
<mszyprow_>
maybe gstreamer has something hardcoded...
<memeka>
yup, maybe gst
<mszyprow_>
as nv12 plane is indeed supported on exynos4 socs
<memeka>
anw, hope to test gscaler integration soon :P :P
nighty- has quit [Quit: Disappears in a puff of smoke]
Turl has quit [Ping timeout: 256 seconds]
_whitelogger_ has joined #linux-exynos
ssvb has joined #linux-exynos
willmore has quit [Ping timeout: 260 seconds]
libv has quit [Ping timeout: 260 seconds]
libv has joined #linux-exynos
nashpa_ has joined #linux-exynos
mszyprow_ has quit [Ping timeout: 260 seconds]
_whitelogger has quit [Remote host closed the connection]
Wizzup has joined #linux-exynos
nashpa_ has quit [Quit: Going away]
nashpa has joined #linux-exynos
Turl has joined #linux-exynos
willmore_ is now known as willore
willore is now known as willmore
nighty- has joined #linux-exynos
genii has joined #linux-exynos
afaerber has joined #linux-exynos
kxkamil has quit [Quit: Leaving]
afaerber has quit [Ping timeout: 252 seconds]
aalm has quit [Ping timeout: 240 seconds]
afaerber has joined #linux-exynos
afaerber has quit [Ping timeout: 260 seconds]
Ultrasauce has quit [Remote host closed the connection]
afaerber has joined #linux-exynos
Vasco_O is now known as Vasco
LiquidAcid has joined #linux-exynos
afaerber has quit [Quit: Leaving]
Ultrasauce has joined #linux-exynos
Ultrasauce has quit [Read error: Connection reset by peer]
Ultrasauce has joined #linux-exynos
aalm has joined #linux-exynos
LiquidAcid has quit [Quit: Leaving]
Vasco is now known as Vasco_O
libv has quit [Ping timeout: 260 seconds]
libv has joined #linux-exynos
kloczek has quit [Remote host closed the connection]
kloczek has joined #linux-exynos
kloczek has quit [Remote host closed the connection]