On 16/04/2020 19:10, Steve Longerbeam wrote:
> Hi Hans,
>
> On 4/16/20 3:52 AM, Hans Verkuil wrote:
>> Hi Steve,
>>
>> On 25/03/2020 18:34, Steve Longerbeam wrote:
>>> The function init_mbus_colorimetry() is used to initialize the imx
>>> subdevice mbus colorimetry to some sane defaults when the subdevice is
>>> registered. Currently it guesses at a colorspace based on the passed
>>> mbus pixel format. If the format is RGB, it chooses colorspace
>>> V4L2_COLORSPACE_SRGB, and if the format is YUV, it chooses
>>> V4L2_COLORSPACE_SMPTE170M.
>>>
>>> While that might be a good guess, it's not necessarily true that a RGB
>>> pixel format encoding uses a SRGB colorspace, or that a YUV encoding
>>> uses a SMPTE170M colorspace. Instead of making this dubious guess,
>>> just default the colorspace to SRGB.
>>>
>>> Reported-by: Hans Verkuil <hverkuil@xxxxxxxxx>
>>> Signed-off-by: Steve Longerbeam <slongerbeam@xxxxxxxxx>
>>> ---
>>> drivers/staging/media/imx/imx-media-utils.c | 3 +--
>>> 1 file changed, 1 insertion(+), 2 deletions(-)
>>>
>>> diff --git a/drivers/staging/media/imx/imx-media-utils.c b/drivers/staging/media/imx/imx-media-utils.c
>>> index fae981698c49..8344675bfd05 100644
>>> --- a/drivers/staging/media/imx/imx-media-utils.c
>>> +++ b/drivers/staging/media/imx/imx-media-utils.c
>>> @@ -236,8 +236,7 @@ static const struct imx_media_pixfmt ipu_rgb_formats[] = {
>>> static void init_mbus_colorimetry(struct v4l2_mbus_framefmt *mbus,
>>> const struct imx_media_pixfmt *fmt)
>>> {
>>> - mbus->colorspace = (fmt->cs == IPUV3_COLORSPACE_RGB) ?
>>> - V4L2_COLORSPACE_SRGB : V4L2_COLORSPACE_SMPTE170M;
>>> + mbus->colorspace = V4L2_COLORSPACE_SRGB;
>>> mbus->xfer_func = V4L2_MAP_XFER_FUNC_DEFAULT(mbus->colorspace);
>>> mbus->ycbcr_enc = V4L2_MAP_YCBCR_ENC_DEFAULT(mbus->colorspace);
>>> mbus->quantization =
>> The quantization is probably wrong as well since it checks fmt->cs.
>> The first argument should just be 'true'.
>
> I don't think so. The default quantization depends in part on whether
> the pixel format is RGB or YUV, which is the reason for passing boolean
> (fmt->cs == IPUV3_COLORSPACE_RGB) as the first argument.
Ah yes, that's true. Sorry for the noise.
Regards,
Hans
>
> I realize "fmt->cs" is a misnomer, but it is borrowing from earlier
> misnomers, i.e. the 'enum ipu_color_space' enumerations. Fixing those
> names would touch lots of imx driver code.
>
>
>>
>> In any case, this patch no longer applies after the imx utils patch series.
>
> Ok, I'll re-submit after the utils patch series is merged.
>
> Steve
>
