On Mon Dec 22, 2025 at 11:11 AM CET, Louis Chauvet wrote:
> To allows the userspace to test many hardware configuration, introduce a
> new interface to configure the available color ranges per planes. VKMS
> supports multiple color ranges, so the userspace can choose any
> combination.
>
> The supported color ranges are configured by writing a color range bitmask
> to the file `supported_color_ranges` and the default color range is
> chosen by writing a color encoding bitmask to `default_color_range`.
>
> Signed-off-by: Louis Chauvet <[email protected]>
> --- a/Documentation/ABI/testing/configfs-vkms
> +++ b/Documentation/ABI/testing/configfs-vkms
> @@ -138,6 +138,21 @@ Description:
> Default color encoding presented to userspace, same
> values as supported_color_encoding.
>
> +What:
> /sys/kernel/config/vkms/<device>/planes/<plane>/supported_color_ranges
> +Date: Nov 2025
This should be Jan 2026 I guess. Same for the previous patches in the
series which I already reviewed, sorry I didn't notice before.
BTW I wonder whether it is really important to have a date here. The time
before a patch is applied can make it quite wrong, but mostly I don't see
an obvious usefulness.
> --- a/drivers/gpu/drm/vkms/vkms_configfs.c
> +++ b/drivers/gpu/drm/vkms/vkms_configfs.c
> +static ssize_t plane_default_color_range_show(struct config_item *item, char
> *page)
> +{
> + struct vkms_configfs_plane *plane =
> plane_item_to_vkms_configfs_plane(item);;
Double semicolon.
> +static ssize_t plane_default_color_range_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 & ~VKMS_SUPPORTED_COLOR_RANGES)
> + return -EINVAL;
> + /* Should at least provide one color range */
> + if ((val & VKMS_SUPPORTED_COLOR_RANGES) == 0)
> + return -EINVAL;
As for patch 13, these 3 lines are redundant, the is_power_of_2() below is
enough.
> +
> + if (!is_power_of_2(val))
> + return -EINVAL;
> +
> + /* Convert bit position to the proper enum value */
> + val = __ffs(val) + DRM_COLOR_YCBCR_LIMITED_RANGE;
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
I wonder whether this should just be '+ 1'. After all it's just the __ffs
semantics counting from 1 as opposed to the BIT() semantics counting from
0. Any pair of BIT() to read and __ffs() to write will need a '+ 1',
regardless of the meaning of the bits.
Same in patch 13, but realized just now.
Luca
--
Luca Ceresoli, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com
