akaizen has quit [Remote host closed the connection]
akaizen has joined #linux-rockchip
akaizen has quit [Read error: Connection reset by peer]
akaizen has joined #linux-rockchip
linuxium has quit []
mcbrick has quit [Remote host closed the connection]
massi has joined #linux-rockchip
<_brain>
morning
rellla has joined #linux-rockchip
rellla2 has joined #linux-rockchip
rellla has quit [Ping timeout: 264 seconds]
AstralixNB has joined #linux-rockchip
<_brain>
AstralixNB: taking a look at HDMI and FB for the RK3066, should I port the driver from Android, or write from scratch? :>
rellla2 is now known as rellla
<AstralixNB>
_brain, I have no idea as I didn't check that. But I is very likely that these drivers have just to be adapted to an existing lcdc.
eebrah|FTS has quit [Ping timeout: 240 seconds]
mmind00 has joined #linux-rockchip
hramrach has quit [Ping timeout: 240 seconds]
<_brain>
AstralixNB: so it would seem, looking at the stuff now. gonna take a loot at adapting some video-related stuff tonight
<AstralixNB>
The new mainline ARM strategy is to cut things into common pieces and reuse them as often as it can be.
<AstralixNB>
As ARM SOCs are assembled from IPs it makes more sense to write a single driver for this IP then to write a new driver per SOC like it was done in the past.
<AstralixNB>
So major work is to fiind the most matching existing driver and put in as little additional code, options, quirks, as possible to adopt the own inpllementation of the IP.
eebrah|noPC has joined #linux-rockchip
<_brain>
and make it as configurable through the dts as possible, instead?
<_brain>
obviously, :)
<AstralixNB>
Yes, right
<mmind00>
also for video stuff, looking at the (upcoming) "Common Display Framework" is most likely the way to go
<AstralixNB>
Isn't LCDC and Common Display Framework the same? I thought LCDC is the predecessor of it
<AstralixNB>
Ah, ok. It was cause the LCDC made some steps to the direction of auto detection and unification...
<AstralixNB>
Do we have to go the CDF way from the start or can we put an intermediate LCDC? I am not sure if we do something good if the interface we intergate is too futuristic and we have no one able to connect ontop of us
<mmind00>
current gpu support always includes private APIs between the components involved, this is what the CDF is meant to solve ... but the CDF itself is in the works for quite some time - including changing APIs