<jmcneill>
I have set OVL_UI_ATTR_CTL(0) to 0x401 (XRGB_8888, enabled) and OVL_UI_ATTR_CTL(1) to 0x1 (ARGB_8888, enabled)
<jmcneill>
A cursor pixel value of 0x00000000 is rendered black instead of transparent
megi has quit [Ping timeout: 250 seconds]
<catphish>
jmcneill: i am quite familiar with it
<catphish>
jmcneill: as far as i know there's no blending between layers in the same overlay :(
<jmcneill>
ah nuts :/
<jmcneill>
the datasheet makes it seem like it might be possible
<catphish>
you have to use 2 overlays to make use of blending, layers in the same overlay just overwrite each other
<catphish>
the example in the datasheet only shows them overlapping, not blending :(
<jmcneill>
I was hoping for a solution that would work with both mixer0 and mixer1, that's why I went down this path in the first place
<catphish>
can't you use UI1 over VIDEO?
<catphish>
on both?
<jmcneill>
The diagram you mean? I saw that.. it was all of the alpha fields in OVL_UI_ATTR_CTL that made me hopeful (because they are not present in OVL_V_ATTCTL)
<jmcneill>
oh that's not a bad idea
<catphish>
i'm not 100% certain, but my tests yield the same results as yours
<catphish>
layers in the same overlay just overwrite each other, not blend
<catphish>
but you can use the video layer in XRGB mode
<jmcneill>
yeah, trying to think if there are any drawbacks
<catphish>
then UI1 in ARGB mode, then the blender should work
<jmcneill>
is VI scaler going to cause issues with the fb?
<catphish>
i don't know, i've never run linux on it
<catphish>
well, then the v scaler will scale your fb :)
<jmcneill>
yeah, this is a problem :/
<catphish>
you want to blend 2 layers on top of your existing fb?
<catphish>
if so, this will work on mixer0 but you're out of luck on mixer1 :(
<catphish>
unless you render everything on mixer0 then use writeback to output it through mixer1
<catphish>
this is hard but might help
<jmcneill>
the goal is to use a separate layer to render the pointer
<jmcneill>
but I'm starting to think maybe this is too much effort for what it's worth :)
<catphish>
not sure i understand why you need the scaler
<jmcneill>
video plane
<catphish>
you can render your pointer on UI1 and blend it over your fb on the video layer
<catphish>
or do you need video, and ui, and a third layer for the pointer?
<catphish>
if so, it's probably not going to happen :(
<jmcneill>
that was what i was going for
<jmcneill>
although I'm not sure if in practice that will ever happen
<jmcneill>
my original expectation was that the window server (Xorg) was going to use the video layer but it doesn't
<catphish>
i went through all this, ended up just rendering it all in software
<catphish>
it's cheap to do binary transparency in software
<catphish>
it makes sense in a way to leave the video layer for actual video, as it has better scaling features
<jmcneill>
I think what I'll do is use CH2 only if it is present
<jmcneill>
and fallback to software rendering otherwise
<catphish>
i'm afraid i don't have any other ideas, jernej is the real expert on this, he might have advice if you stick around long enough
<catphish>
that sounds like a sensble idea
LargePrime has joined #linux-sunxi
lurchi__ is now known as lurchi_
<catphish>
good luck anyway :)
<jmcneill>
geez, I ended up with a rectangle again!
<jmcneill>
do I need to do anything special beyond specifying ARGB to get blending to work?
pmpp_ has joined #linux-sunxi
pmpp has quit [Disconnected by services]
lurchi_ is now known as lurchi__
<anarsoul>
mripard: wens: so regarding strict clock rate check
<anarsoul>
I'm thinking about introducing dt property for sun4i
<anarsoul>
that will disable strict check on platforms that don't need it
<anarsoul>
is it something acceptable?
pmpp_ has quit [Ping timeout: 250 seconds]
pmpp has joined #linux-sunxi
return0e has quit [Ping timeout: 240 seconds]
<wens>
seeing as I was the one who originally came up with it, go for it :)
return0e has joined #linux-sunxi
<lvrp16>
all 4 h5 boards have been running at 768mhz ddr for more than 3 days without a single memory error. is there a multiplier issue preventing 792mhz (24x33)? i've been testing h3 and it's crashed at 768 and 744 after 24 hours. 720 is still running. seems like h5 memory controller is much improved from h3.
<wens>
lvrp16: there's two multiples, one 5 bit wide and one 2 bit wide
<wens>
* multipliers
<wens>
lvrp16: so the highest you can go with just one multiplier is 32, beyond that you get multiples of 2 or 3
catphish has quit [Quit: Leaving]
<lvrp16>
oh so i should try 816 and 840?
<lvrp16>
will give it a go now, thanks
<lvrp16>
wens: 816 did not tune. 840 did tune but stopped shortly
victhor has quit [Remote host closed the connection]
sunshavi has quit [Ping timeout: 246 seconds]
tlwoerner has quit [Quit: Leaving]
tlwoerner has joined #linux-sunxi
sunshavi has joined #linux-sunxi
shfil has quit [Quit: Connection closed for inactivity]
leviathanch has joined #linux-sunxi
pmpp has quit [Ping timeout: 245 seconds]
pmpp has joined #linux-sunxi
_whitelogger has joined #linux-sunxi
TheSeven has quit [Ping timeout: 240 seconds]
TheSeven has joined #linux-sunxi
dddddd has quit [Remote host closed the connection]
Andy-D has quit [Ping timeout: 250 seconds]
Andy-D has joined #linux-sunxi
lurchi_ has joined #linux-sunxi
akaWolf has quit [Remote host closed the connection]
pmpp_ has joined #linux-sunxi
pmpp has quit [Disconnected by services]
lurchi__ has quit [Ping timeout: 246 seconds]
merlin1991 has quit [Remote host closed the connection]
kelvan has quit [Remote host closed the connection]
merlin1991 has joined #linux-sunxi
kelvan has joined #linux-sunxi
_whitelogger has joined #linux-sunxi
TheSeven has quit [Ping timeout: 250 seconds]
TheSeven has joined #linux-sunxi
IgorPec has joined #linux-sunxi
selfbg has joined #linux-sunxi
<jernej>
jmcneill: HW cursor is not implemented for DE2 on Linux. One UI plane could be wasted on mixer0, but mixer1 doesn't have enough resources.
<jernej>
well, at least with current implementation.
<jernej>
regarding alpha blending, make sure that alpha mode and alpha are set to 0 in UI layer settings
<jernej>
then, you have to set right blending mode in global registers, that is just hardcoded to 0x03010301
pmpp_ has quit [Ping timeout: 272 seconds]
<jernej>
you also have to properly set Z order in route register
pmpp has joined #linux-sunxi
<jernej>
after that, there are some things you have to consider
<jernej>
VI layer doesn't support alpha blending, alpha number is just ignored, so if you use it, it has to be at the bottom, otherwise it will just cover any other plane
<jernej>
last but not least, depending on your setting, bottom layer has to cover whole screen, otherwise you have to set background color
<jernej>
Linux DRM driver always set it to opaque black (0xff000000)
reinforce has joined #linux-sunxi
pmpp has quit [Ping timeout: 268 seconds]
lkcl has quit [Ping timeout: 240 seconds]
lkcl has joined #linux-sunxi
f0xx has joined #linux-sunxi
nuuuciano has quit [Ping timeout: 244 seconds]
yann has quit [Ping timeout: 272 seconds]
tjader has quit [Ping timeout: 268 seconds]
tjader has joined #linux-sunxi
clemens3 has joined #linux-sunxi
pmpp has joined #linux-sunxi
clemens3 has quit [Ping timeout: 240 seconds]
pmpp has quit [Ping timeout: 250 seconds]
clemens3 has joined #linux-sunxi
ninolein has joined #linux-sunxi
loswillios has quit [Ping timeout: 246 seconds]
loswillios has joined #linux-sunxi
akaWolf has joined #linux-sunxi
AneoX has joined #linux-sunxi
yann has joined #linux-sunxi
tnovotny has joined #linux-sunxi
msimpson has joined #linux-sunxi
megi has joined #linux-sunxi
aalm has quit [Quit: xyz 2.3]
maz has joined #linux-sunxi
BenG83 has joined #linux-sunxi
LargePrime has quit [Ping timeout: 272 seconds]
tllim has quit [Read error: Connection reset by peer]
LargePrime has joined #linux-sunxi
msimpson_ has joined #linux-sunxi
msimpson has quit [Read error: Connection reset by peer]
swiftgeek has quit [Ping timeout: 240 seconds]
victhor has joined #linux-sunxi
Gerwin_J has joined #linux-sunxi
<fALSO>
jernej, from what i investigated yesterday
<fALSO>
moeicenowy, sent a big patch to linux-sunxi
<fALSO>
about PINE64
<fALSO>
that is also an H6
swiftgeek has joined #linux-sunxi
<jmcneill>
jernej: I ended up burning a UI plane for it. My blending issue was around strange routing and uninitialized BLD_CTL.
<jmcneill>
I have UI (CH1) going to pipe 0, VI (CH0) going to pipe 1, and UI (CH2) giong to pipe 2. Seems I have to set BLD_CTL for all 3 pipes -- I was only setting it for 0 and 2.
<jmcneill>
All good now (for RT-Mixer0 at least)
<jmcneill>
Fallback to software cursor for RT-Mixer1
selfbg has quit [Read error: Connection reset by peer]
selfbg has joined #linux-sunxi
<mripard>
anarsoul: if that's our only option, I'd rather go for a parameter
<fALSO>
jernej, im trying to figure out a way on the DTS to save the MAC address of the ethernet device
<fALSO>
before submiting the patch
maz has quit [Ping timeout: 250 seconds]
mzki has joined #linux-sunxi
nashpa_ has joined #linux-sunxi
shfil has joined #linux-sunxi
s_frit has quit [Remote host closed the connection]
s_frit has joined #linux-sunxi
\\Mr_C\\ has quit [Quit: (Read error: Connection reset by beer)]
maz has joined #linux-sunxi
yann has quit [Ping timeout: 250 seconds]
\\Mr_C\\ has joined #linux-sunxi
yann has joined #linux-sunxi
nashpa_ is now known as nashpa
aalm has joined #linux-sunxi
dddddd has joined #linux-sunxi
parazyd has quit [Ping timeout: 240 seconds]
LargePrime has quit [Ping timeout: 240 seconds]
LargePrime has joined #linux-sunxi
afaerber has joined #linux-sunxi
Gerwin_J has quit [Quit: Gerwin_J]
aalm has quit [Ping timeout: 245 seconds]
maz has quit [Ping timeout: 240 seconds]
leviathanch has quit [Remote host closed the connection]
\x has quit [K-Lined]
\x has joined #linux-sunxi
\x has quit [Changing host]
\x has joined #linux-sunxi
\x has quit [K-Lined]
fkluknav has quit [Remote host closed the connection]
fkluknav has joined #linux-sunxi
_whitelogger has joined #linux-sunxi
<fALSO>
moeicenowy, willing to help me out with the DTS of the orange pi one plus ?
nick- has joined #linux-sunxi
nuuuciano has joined #linux-sunxi
maz has joined #linux-sunxi
clemens3 has quit [Remote host closed the connection]
lurchi_ is now known as lurchi__
freemangordon has joined #linux-sunxi
maz has quit [Ping timeout: 240 seconds]
Andy-D has quit [Remote host closed the connection]
selfbg has quit [Remote host closed the connection]
<fALSO>
going to try the uboot patch later in the day, ill report to you later
yann has joined #linux-sunxi
yann has quit [Ping timeout: 268 seconds]
clemens3 has joined #linux-sunxi
pmpp is now known as pmp-p
yann has joined #linux-sunxi
jbrown has quit [Ping timeout: 252 seconds]
tnovotny has quit [Quit: Leaving]
parazyd has joined #linux-sunxi
msimpson_ has quit [Read error: Connection reset by peer]
reinforce has joined #linux-sunxi
wasutton3_mobile has quit [Ping timeout: 245 seconds]
BenG83 has joined #linux-sunxi
wasutton3_mobile has joined #linux-sunxi
<unkraut>
hi, i am trying to get the lcd panel sold as s070wv20-ct16 to work on a banana pi r1 (A20) with a recent kernel arch linux, i already managed to get it to work in u-boot (2019-01 mainline checkout from sunday)
<unkraut>
does anyone have a working kernel config for a similar setup ?
yann has quit [Ping timeout: 240 seconds]
<unkraut>
the device tree should be mostly correct as it does work in u-boot
clemens3 has quit [Ping timeout: 250 seconds]
BenG83 has quit [Remote host closed the connection]
<unkraut>
LCD0_CS goes from high to low immediatly after kernel is loaded
<unkraut>
LCD_DE and BL_EN go low later
<fALSO>
moeicenowy, just tested with that patch for uboot, and the alias on the dtb
<fALSO>
and the macaddress is now fixed :-)
<fALSO>
awesome
shfil has quit [Quit: Connection closed for inactivity]