Re: [PATCH RESEND v2 12/32] drm/vkms: Introduce configfs for plane color encoding

[Louis Chauvet <louis.chauvet@bootlin.com>]

On 12/18/25 18:59, Luca Ceresoli wrote:
> On Wed Oct 29, 2025 at 3:36 PM CET, Louis Chauvet wrote:
>> To allows the userspace to test many hardware configuration, introduce a
>> new interface to configure the available color encoding per planes. VKMS
>> supports multiple color encoding, so the userspace can choose any
>> combination.
>>
>> The supported color encoding are configured by writing a color encoding
>> bitmask to the file `supported_color_encoding` and the default color
>> encoding is chosen by writing a color encoding bitmask to
>> `default_color_encoding`.
>>
>> Signed-off-by: Louis Chauvet <louis.chauvet@bootlin.com>
>> ---
>>   Documentation/gpu/vkms.rst           |  7 ++-
>>   drivers/gpu/drm/vkms/vkms_configfs.c | 98 ++++++++++++++++++++++++++++++++++++
>>   2 files changed, 104 insertions(+), 1 deletion(-)
>>
>> diff --git a/Documentation/gpu/vkms.rst b/Documentation/gpu/vkms.rst
>> index eac1a942d6c4..dab6811687a2 100644
>> --- a/Documentation/gpu/vkms.rst
>> +++ b/Documentation/gpu/vkms.rst
>> @@ -87,7 +87,7 @@ Start by creating one or more planes::
>>
>>     sudo mkdir /config/vkms/my-vkms/planes/plane0
>>
>> -Planes have 4 configurable attributes:
>> +Planes have 6 configurable attributes:
>>
>>   - type: Plane type: 0 overlay, 1 primary, 2 cursor (same values as those
>>     exposed by the "type" property of a plane)
>> @@ -97,6 +97,11 @@ Planes have 4 configurable attributes:
>>     (same values as those exposed by the "rotation" property of a plane)
>>   - default_rotation: Default rotation presented to the userspace, same values as
>>     possible_rotations.
>> +- supported_color_encoding: Available encoding for a plane, as a bitmask:
>                                           encodings
> 
>> +  0x01 YCBCR_BT601, 0x02: YCBCR_BT709, 0x04 YCBCR_BT2020 (same values as those exposed
>                             ^
> 
> Unintended colon? While I think it's nice to have, there is none elsewhere,
> even in previous patches, and I'd say we can live happy without.
> 
>> +  by the COLOR_ENCODING property of a plane)
>> +- default_color_encoding: Default color encoding presented to the userspace, same
>> +  values as supported_color_encoding
>>
>>   Continue by creating one or more CRTCs::
>>
>> diff --git a/drivers/gpu/drm/vkms/vkms_configfs.c b/drivers/gpu/drm/vkms/vkms_configfs.c
>> index 7cc8ba315ef0..ee2e8d141f9e 100644
>> --- a/drivers/gpu/drm/vkms/vkms_configfs.c
>> +++ b/drivers/gpu/drm/vkms/vkms_configfs.c
>> @@ -439,16 +439,114 @@ static ssize_t plane_default_rotation_store(struct config_item *item,
>>   	return count;
>>   }
>>
>> +static ssize_t plane_supported_color_encodings_show(struct config_item *item, char *page)
>> +{
>> +	struct vkms_configfs_plane *plane;
>> +	unsigned int supported_color_encoding;
>                       supported_color_encodings
> 
>> +
>> +	plane = plane_item_to_vkms_configfs_plane(item);
> 
> As for patch 9, for consistency:
> 
>    struct vkms_configfs_plane *plane = plane_item_to_vkms_configfs_plane(item);
> 
>> +
>> +	scoped_guard(mutex, &plane->dev->lock) {
>> +		supported_color_encoding = vkms_config_plane_get_supported_color_encodings(plane->config);
>> +	}
>> +
>> +	return sprintf(page, "%u", supported_color_encoding);
>> +}
>> +
>> +static ssize_t plane_supported_color_encodings_store(struct config_item *item,
>> +						     const char *page, size_t count)
>> +{
>> +	struct vkms_configfs_plane *plane = plane_item_to_vkms_configfs_plane(item);
>> +	int ret, val = 0;
>> +
>> +	ret = kstrtouint(page, 10, &val);
>> +	if (ret)
>> +		return ret;
>> +
>> +	/* Should be a supported value */
>> +	if (val & ~(BIT(DRM_COLOR_YCBCR_BT601) |
>> +		    BIT(DRM_COLOR_YCBCR_BT709) |
>> +		    BIT(DRM_COLOR_YCBCR_BT2020)))
>> +		return -EINVAL;
>> +	/* Should at least provide one color range */
>> +	if ((val & (BIT(DRM_COLOR_YCBCR_BT601) |
>> +		    BIT(DRM_COLOR_YCBCR_BT709) |
>> +		    BIT(DRM_COLOR_YCBCR_BT2020))) == 0)
> 
> I'm mentioning here as it comes to mind, but it's valid for other similar
> patches in this series: why not adding a
> 
>    #define DRM_COLOR_ENCODINGS_SUPPORTED ( \
>                     BIT(DRM_COLOR_YCBCR_BT601) |
>                     BIT(DRM_COLOR_YCBCR_BT709) |
>                     BIT(DRM_COLOR_YCBCR_BT2020))
> 
> and use it in place of the various bitwise-or sequences?
> 
> This would simplify work later on if adding a new color encoding (or color
> range, or...).
> 
> Somewhat like DRM_MODE_*_MASK.
> 
>> +		return -EINVAL;
>> +
>> +	scoped_guard(mutex, &plane->dev->lock) {
>> +		/* Ensures that the default rotation is included in supported rotation */
>> +		if (plane->dev->enabled)
>> +			return -EINVAL;
> 
> And here the comment is definitely wrong. :-)
> 
>> +
>> +		vkms_config_plane_set_supported_color_encodings(plane->config, val);
>> +	}
>> +
>> +	return count;
>> +}
>> +
>> +/* Plane default_color_encoding : vkms/<device>/planes/<plane>/default_color_encoding */
> 
> There's no such comment in other places, so for consistency remove it (or
> add it everywhere?!? ... no, just kidding).
> 
>> +
>> +static ssize_t plane_default_color_encoding_show(struct config_item *item, char *page)
>> +{
>> +	struct vkms_configfs_plane *plane;
>> +	unsigned int default_color_encoding;
>> +
>> +	plane = plane_item_to_vkms_configfs_plane(item);
>> +
>> +	scoped_guard(mutex, &plane->dev->lock) {
>> +		default_color_encoding = vkms_config_plane_get_default_color_encoding(plane->config);
>> +	}
>> +
>> +	return sprintf(page, "%u", default_color_encoding);
>> +}
>> +
>> +static ssize_t plane_default_color_encoding_store(struct config_item *item,
>> +						  const char *page, size_t count)
>> +{
>> +	struct vkms_configfs_plane *plane = plane_item_to_vkms_configfs_plane(item);
>> +	int ret, val = 0;
>> +
>> +	ret = kstrtouint(page, 10, &val);
>> +	if (ret)
>> +		return ret;
>> +
>> +	/* Should be a supported value */
>> +	if (val & ~(BIT(DRM_COLOR_YCBCR_BT601) |
>> +		    BIT(DRM_COLOR_YCBCR_BT709) |
>> +		    BIT(DRM_COLOR_YCBCR_BT2020)))
>> +		return -EINVAL;
>> +	/* Should at least provide one color range */
>> +	if ((val & (BIT(DRM_COLOR_YCBCR_BT601) |
>> +		    BIT(DRM_COLOR_YCBCR_BT709) |
>> +		    BIT(DRM_COLOR_YCBCR_BT2020))) == 0)
>> +		return -EINVAL;
> 
> Shouldn't you check that exactly one bit is set? As in patch 9.

Because this code is wrong... the default rotation should be 
DRM_COLOR_YCBCR_BT601 / DRM_COLOR_YCBCR_BT709 / DRM_COLOR_YCBCR_BT2020
not a bitfield...

>> +
>> +	scoped_guard(mutex, &plane->dev->lock) {
>> +		/* Ensures that the default rotation is included in supported rotation */
>> +		if (plane->dev->enabled)
>> +			return -EINVAL;
> 
> As before, wrong comment.
> 
> Luca
> 
> --
> Luca Ceresoli, Bootlin
> Embedded Linux and Kernel engineering
> https://bootlin.com