Dne nedelja, 22. maj 2022 ob 12:48:27 CEST je benlypan@gmail.com napisal(a):
Hi Jernej!
Which SDK? All SDKs that I have or found on internet don't mention YUV nor
RGB
scanline limit. That doesn't mean there is none, I'm just unable to verify your claim. Did you test this by yourself? Also, please make YUV scanline change separate patch with fixes tag.
Here is the code about scanline config for V3s: https://github.com/Tina-Linux/tina-v3s-linux-4.9/blob/master/drivers/video/
fbdev/sunxi/disp2/disp/de/lowlevel_sun8iw8/de_feat.c#L55
https://github.com/Tina-Linux/tina-v3s-linux-4.9/blob/master/drivers/video/
fbdev/sunxi/disp2/disp/de/lowlevel_sun8iw8/de_rtmx.c#L1212
And here is for D1: https://gitlab.com/weidongshan/tina-d1-h/-/blob/main/lichee/linux-5.4/
drivers/video/fbdev/sunxi/disp2/disp/de/lowlevel_v2x/de_feat.c#L182
https://gitlab.com/weidongshan/tina-d1-h/-/blob/main/lichee/linux-5.4/
drivers/video/fbdev/sunxi/disp2/disp/de/lowlevel_v2x/de_rtmx.c#L1588
Thanks for the links!
I was tested it on V3s. Without this patch, the plane will not display correctly if src_w large than 1024, for both YUV and RGB. I don't have a device with D1, so D1 is not tested. But according to the SDK code of D1, the scanline value of mixer1 is 1024.
I'm new to submitting patches, So may I ask a question? If there is no problem with this patch, what I need to do is to send new patches as you suggested?
https://www.kernel.org/doc/html/latest/process/submitting-patches.html
I already requested changes, so you need to send v2. But wait until discussion is finished.
Best regards, Jernej
Best regards, Genfu Pan