Re: [PATCH 1/4] pinctrl: sh-pfc: r8a7796: Add VIN4, VIN5 pins, groups and functions

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Hi Geert,

On 2018-03-09 23:24:37 +0100, Geert Uytterhoeven wrote:
> Hi Niklas,
> 
> On Fri, Mar 9, 2018 at 10:58 PM, Niklas Söderlund
> <niklas.soderlund@xxxxxxxxxxxx> wrote:
> > On 2018-03-09 13:33:03 +0100, Geert Uytterhoeven wrote:
> >> P.S. Apparently R-Car Gen2 and Gen3 also support 8-bit YCbCr input data
> >>      on the DATA8-15 pins, for which we don't have pin groups yet.
> >>      Niklas: is this mode supported by the VIN driver?
> >
> > * Gen2
> > I can't find DATA12-15 in the datasheet I have, where did you find them?
> > I'm looking at Tables 26.3, 26.4 and 26.5 on v1.0 of the Gen2 datasheet.
> > But yes on Gen2 the VIN driver supports capturing from these data pins.
> 
> Sorry, on Gen2 (all but V2H, to make matters more complicated), the DATA pins
> are not numbered from 0 to 23, but split in 3 blocks of 8 pins, matching R, G,
> and B blocks. But apart from the numbering, the formats are mostly the same (4
> bit width is the exception, and supported on Gen2 only).
> 
> What I meant is the third mode in e.g. Table 26.5, "ITU-R BT.601/BT.709/BT.656
> 8-bit YCbCr-422 (VnDMR2/YDS = 1)", which uses VI0_G[7:0] instead of VIO_B[7:0]
> for transfering 8-bit YCbCr data. Apparently the PFC driver doesn't have a pin
> group for that combo.
> 
> (I have v2.00 of the datasheet, but the table looks identical in v1.0).
> 
> > * Gen3
> > Currently CSI-2 are the only supported input method for the Gen3
> > patches. It would be possible with a small hack to run the Gen2 driver
> > on Gen3 and have it try and use the DATA pins, but this is not tested as
> > prior to V3M we had no device to test this on as the DATA pins where all
> > routed to EXIO connectors.
> >
> > There have been some talks about adding support for this to the driver,
> > I know Jacopo posted a patch-set a while ago for this but I have not
> > tested it. Looking at the Gen3 Table 26.8.1 it sure looks like all
> > DATA lines DATA0-23 could be used on some SoCs to capture 24 bit RGB and
> > YCbCr.
> 
> My question was about the second mode in the table, which is the same one
> as the third mode on Gen2.

Ahh I see my bad. The current VIN driver for Gen2 which is the only one 
where this is supported can use the following input formats:

ITU-R BT.656 8-Bit
ITU-R BT.656 10-Bit
ITU-R BT.601 8-Bit
ITU-R BT.601 10-Bit
ITU-R BT.601/BT.709/BT.1358 16-Bit
ITU-R BT.601/BT.709 (RGB888) 24-Bit

But I do not have the hardware to be able to verify that all modes 
actually work. All modes are carried over from the driver conversion 
from soc-camera.

Looking at the Koelsch the modes that are tested are 'ITU-R 
BT.601/BT.709 (RGB888) 24-Bit' from the ADV7612 and 'ITU-R BT.656 8-Bit' 
from the ADV7180.

> 
> Gr{oetje,eeting}s,
> 
>                         Geert
> 
> --
> Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@xxxxxxxxxxxxxx
> 
> In personal conversations with technical people, I call myself a hacker. But
> when I'm talking to journalists I just say "programmer" or something like that.
>                                 -- Linus Torvalds

-- 
Regards,
Niklas Söderlund



[Index of Archives]     [Linux Samsung SOC]     [Linux Wireless]     [Linux Kernel]     [ATH6KL]     [Linux Bluetooth]     [Linux Netdev]     [Kernel Newbies]     [IDE]     [Security]     [Git]     [Netfilter]     [Bugtraq]     [Yosemite News]     [MIPS Linux]     [ARM Linux]     [Linux Security]     [Linux RAID]     [Linux ATA RAID]     [Samba]     [Device Mapper]

  Powered by Linux