Den 22.09.2022 16.25, skrev Maxime Ripard:
> The TV mode property has been around for a while now to select and get the
> current TV mode output on an analog TV connector.
>
> Despite that property name being generic, its content isn't and has been
> driver-specific which makes it hard to build any generic behaviour on top
> of it, both in kernel and user-space.
>
> Let's create a new enum tv norm property, that can contain any of the
> analog TV standards currently supported by kernel drivers. Each driver can
> then pass in a bitmask of the modes it supports, and the property
> creation function will filter out the modes not supported.
>
> We'll then be able to phase out the older tv mode property.
>
> Signed-off-by: Maxime Ripard <[email protected]>
>
Please can you add per patch changelogs, it's hard to review when I have
to recall what might have happened with each patch. If you do it drm
style and put in the commit message it should be easy enough to do.
> diff --git a/drivers/gpu/drm/drm_connector.c b/drivers/gpu/drm/drm_connector.c
> index 4e4fbc9e0049..e7aa8de08f5b 100644
> --- a/drivers/gpu/drm/drm_connector.c
> +++ b/drivers/gpu/drm/drm_connector.c
> @@ -980,6 +980,18 @@ static const struct drm_prop_enum_list drm_dvi_i_subconnector_enum_list[] = {
> DRM_ENUM_NAME_FN(drm_get_dvi_i_subconnector_name,
> drm_dvi_i_subconnector_enum_list)
>
> +static const struct drm_prop_enum_list drm_tv_mode_enum_list[] = {
> + { DRM_MODE_TV_MODE_NONE, "None" },
> + { DRM_MODE_TV_MODE_NTSC, "NTSC" },
I think going back to plain NTSC and PAL is a good choice for the common
variants.
> + { DRM_MODE_TV_MODE_NTSC_443, "NTSC-443" },
> + { DRM_MODE_TV_MODE_NTSC_J, "NTSC-J" },
> + { DRM_MODE_TV_MODE_PAL, "PAL" },
> + { DRM_MODE_TV_MODE_PAL_M, "PAL-M" },
> + { DRM_MODE_TV_MODE_PAL_N, "PAL-N" },
> + { DRM_MODE_TV_MODE_SECAM, "SECAM" },
> +};
> +DRM_ENUM_NAME_FN(drm_get_tv_mode_name, drm_tv_mode_enum_list)
> +
> static const struct drm_prop_enum_list drm_tv_select_enum_list[] = {
> { DRM_MODE_SUBCONNECTOR_Automatic, "Automatic" }, /* DVI-I and TV-out */
> { DRM_MODE_SUBCONNECTOR_Composite, "Composite" }, /* TV-out */
> @@ -1645,6 +1657,10 @@ EXPORT_SYMBOL(drm_mode_create_tv_margin_properties);
> * responsible for allocating a list of format names and passing them to
> * this routine.
> *
> + * NOTE: This functions registers the deprecated "mode" connector
> + * property to select the analog TV mode (ie, NTSC, PAL, etc.). New
> + * drivers must use drm_mode_create_tv_properties() instead.
> + *
> * Returns:
> * 0 on success or a negative error code on failure.
> */
> @@ -1686,7 +1702,6 @@ int drm_mode_create_tv_properties_legacy(struct drm_device *dev,
> if (drm_mode_create_tv_margin_properties(dev))
> goto nomem;
>
> -
> if (num_modes) {
> dev->mode_config.legacy_tv_mode_property =
> drm_property_create(dev, DRM_MODE_PROP_ENUM,
> @@ -1735,6 +1750,49 @@ int drm_mode_create_tv_properties_legacy(struct drm_device *dev,
> }
> EXPORT_SYMBOL(drm_mode_create_tv_properties_legacy);
>
> +/**
> + * drm_mode_create_tv_properties - create TV specific connector properties
> + * @dev: DRM device
> + * @supported_tv_modes: Bitmask of TV modes supported (See DRM_MODE_TV_MODE_*)
> +
> + * Called by a driver's TV initialization routine, this function creates
> + * the TV specific connector properties for a given device. Caller is
> + * responsible for allocating a list of format names and passing them to
> + * this routine.
Copy-paste error, there are no format names in this version.
> + *
> + * Returns:
> + * 0 on success or a negative error code on failure.
> + */
> +int drm_mode_create_tv_properties(struct drm_device *dev,
> + unsigned int supported_tv_modes)
> +{
> + struct drm_prop_enum_list tv_mode_list[DRM_MODE_TV_MODE_MAX];
> + struct drm_property *tv_mode;
> + unsigned int i, len = 0;
> +
> + if (dev->mode_config.tv_mode_property)
> + return 0;
> +
> + for (i = 0; i < DRM_MODE_TV_MODE_MAX; i++) {
> + if (!(supported_tv_modes & BIT(i)))
> + continue;
> +
> + tv_mode_list[len].type = i;
> + tv_mode_list[len].name = drm_get_tv_mode_name(i);
> + len++;
> + }
> +
> + tv_mode = drm_property_create_enum(dev, 0, "TV mode",
> + tv_mode_list, len);
> + if (!tv_mode)
> + return -ENOMEM;
> +
> + dev->mode_config.tv_mode_property = tv_mode;
> +
> + return drm_mode_create_tv_properties_legacy(dev, 0, NULL);
> +}
> +EXPORT_SYMBOL(drm_mode_create_tv_properties);
> +
> /**
> * drm_mode_create_scaling_mode_property - create scaling mode property
> * @dev: DRM device
> diff --git a/include/drm/drm_connector.h b/include/drm/drm_connector.h
> index d566b4a4709c..fffacbfd0a45 100644
> --- a/include/drm/drm_connector.h
> +++ b/include/drm/drm_connector.h
> @@ -143,6 +143,71 @@ enum subpixel_order {
>
> };
>
> +/**
> + * enum drm_connector_tv_mode - Analog TV output mode
> + *
> + * This enum is used to indicate the TV output mode used on an analog TV
> + * connector.
> + *
> + * WARNING: The values of this enum is uABI since they're exposed in the
> + * "TV mode" connector property.
> + */
> +enum drm_connector_tv_mode {
> + /**
> + * @DRM_MODE_TV_MODE_NONE: Placeholder to not default on one
> + * variant or the other when nothing is set.
> + */
> + DRM_MODE_TV_MODE_NONE = 0,
How is this supposed to be used?
Noralf.
> +
> + /**
> + * @DRM_MODE_TV_MODE_NTSC: CCIR System M (aka 525-lines)
> + * together with the NTSC Color Encoding.
> + */
> + DRM_MODE_TV_MODE_NTSC,
> +
> + /**
> + * @DRM_MODE_TV_MODE_NTSC_443: Variant of
> + * @DRM_MODE_TV_MODE_NTSC. Uses a color subcarrier frequency
> + * of 4.43 MHz.
> + */
> + DRM_MODE_TV_MODE_NTSC_443,
> +
> + /**
> + * @DRM_MODE_TV_MODE_NTSC_J: Variant of @DRM_MODE_TV_MODE_NTSC
> + * used in Japan. Uses a black level equals to the blanking
> + * level.
> + */
> + DRM_MODE_TV_MODE_NTSC_J,
> +
> + /**
> + * @DRM_MODE_TV_MODE_PAL: CCIR System B together with the PAL
> + * color system.
> + */
> + DRM_MODE_TV_MODE_PAL,
> +
> + /**
> + * @DRM_MODE_TV_MODE_PAL_M: CCIR System M (aka 525-lines)
> + * together with the PAL color encoding
> + */
> + DRM_MODE_TV_MODE_PAL_M,
> +
> + /**
> + * @DRM_MODE_TV_MODE_PAL_N: CCIR System N together with the PAL
> + * color encoding. It uses 625 lines, but has a color subcarrier
> + * frequency of 3.58MHz, the SECAM color space, and narrower
> + * channels compared to most of the other PAL variants.
> + */
> + DRM_MODE_TV_MODE_PAL_N,
> +
> + /**
> + * @DRM_MODE_TV_MODE_SECAM: CCIR System B together with the
> + * SECAM color system.
> + */
> + DRM_MODE_TV_MODE_SECAM,
> +
> + DRM_MODE_TV_MODE_MAX,
> +};
> +
> /**
> * struct drm_scrambling: sink's scrambling support.
> */
> @@ -696,6 +761,7 @@ struct drm_connector_tv_margins {
> * @subconnector: detected subconnector
> * @margins: TV margins
> * @legacy_mode: Legacy TV mode, driver specific value
> + * @mode: TV mode
> * @brightness: brightness in percent
> * @contrast: contrast in percent
> * @flicker_reduction: flicker reduction in percent
> @@ -708,6 +774,7 @@ struct drm_tv_connector_state {
> enum drm_mode_subconnector subconnector;
> struct drm_connector_tv_margins margins;
> unsigned int legacy_mode;
> + unsigned int mode;
> unsigned int brightness;
> unsigned int contrast;
> unsigned int flicker_reduction;
> @@ -1789,6 +1856,7 @@ const char *drm_get_subpixel_order_name(enum subpixel_order order);
> const char *drm_get_dpms_name(int val);
> const char *drm_get_dvi_i_subconnector_name(int val);
> const char *drm_get_dvi_i_select_name(int val);
> +const char *drm_get_tv_mode_name(int val);
> const char *drm_get_tv_subconnector_name(int val);
> const char *drm_get_tv_select_name(int val);
> const char *drm_get_dp_subconnector_name(int val);
> @@ -1802,6 +1870,8 @@ int drm_mode_create_tv_margin_properties(struct drm_device *dev);
> int drm_mode_create_tv_properties_legacy(struct drm_device *dev,
> unsigned int num_modes,
> const char * const modes[]);
> +int drm_mode_create_tv_properties(struct drm_device *dev,
> + unsigned int supported_tv_modes);
> void drm_connector_attach_tv_margin_properties(struct drm_connector *conn);
> int drm_mode_create_scaling_mode_property(struct drm_device *dev);
> int drm_connector_attach_content_type_property(struct drm_connector *dev);
> diff --git a/include/drm/drm_mode_config.h b/include/drm/drm_mode_config.h
> index 35a827175c24..10a6f7d1df0d 100644
> --- a/include/drm/drm_mode_config.h
> +++ b/include/drm/drm_mode_config.h
> @@ -718,9 +718,17 @@ struct drm_mode_config {
> /**
> * @legacy_tv_mode_property: Optional TV property to select
> * the output TV mode.
> + *
> + * Superseeded by @tv_mode_property
> */
> struct drm_property *legacy_tv_mode_property;
>
> + /**
> + * @tv_mode_property: Optional TV property to select the TV
> + * standard output on the connector.
> + */
> + struct drm_property *tv_mode_property;
> +
> /**
> * @tv_left_margin_property: Optional TV property to set the left
> * margin (expressed in pixels).
>
On Sat, Sep 24, 2022 at 05:52:29PM +0200, Noralf Tr?nnes wrote:
> Den 22.09.2022 16.25, skrev Maxime Ripard:
> > The TV mode property has been around for a while now to select and get the
> > current TV mode output on an analog TV connector.
> >
> > Despite that property name being generic, its content isn't and has been
> > driver-specific which makes it hard to build any generic behaviour on top
> > of it, both in kernel and user-space.
> >
> > Let's create a new enum tv norm property, that can contain any of the
> > analog TV standards currently supported by kernel drivers. Each driver can
> > then pass in a bitmask of the modes it supports, and the property
> > creation function will filter out the modes not supported.
> >
> > We'll then be able to phase out the older tv mode property.
> >
> > Signed-off-by: Maxime Ripard <[email protected]>
>
> Please can you add per patch changelogs, it's hard to review when I have
> to recall what might have happened with each patch. If you do it drm
> style and put in the commit message it should be easy enough to do.
I certainly don't want to start that discussion, but I'm really not a
fan of that format either. I'll do it for that series if you prefer.
> > +/**
> > + * enum drm_connector_tv_mode - Analog TV output mode
> > + *
> > + * This enum is used to indicate the TV output mode used on an analog TV
> > + * connector.
> > + *
> > + * WARNING: The values of this enum is uABI since they're exposed in the
> > + * "TV mode" connector property.
> > + */
> > +enum drm_connector_tv_mode {
> > + /**
> > + * @DRM_MODE_TV_MODE_NONE: Placeholder to not default on one
> > + * variant or the other when nothing is set.
> > + */
> > + DRM_MODE_TV_MODE_NONE = 0,
>
> How is this supposed to be used?
It's not supposed to be used. It was a suggestion from Mateusz to avoid
to default to any standard when we don't initialize something. I don't
have any strong feeling about it, so I can drop it if you prefer.
Maxime
Den 26.09.2022 12.01, skrev Maxime Ripard:
> On Sat, Sep 24, 2022 at 05:52:29PM +0200, Noralf Trønnes wrote:
>> Den 22.09.2022 16.25, skrev Maxime Ripard:
>>> The TV mode property has been around for a while now to select and get the
>>> current TV mode output on an analog TV connector.
>>>
>>> Despite that property name being generic, its content isn't and has been
>>> driver-specific which makes it hard to build any generic behaviour on top
>>> of it, both in kernel and user-space.
>>>
>>> Let's create a new enum tv norm property, that can contain any of the
>>> analog TV standards currently supported by kernel drivers. Each driver can
>>> then pass in a bitmask of the modes it supports, and the property
>>> creation function will filter out the modes not supported.
>>>
>>> We'll then be able to phase out the older tv mode property.
>>>
>>> Signed-off-by: Maxime Ripard <[email protected]>
>>
>> Please can you add per patch changelogs, it's hard to review when I have
>> to recall what might have happened with each patch. If you do it drm
>> style and put in the commit message it should be easy enough to do.
>
> I certainly don't want to start that discussion, but I'm really not a
> fan of that format either. I'll do it for that series if you prefer.
>
The format isn't important, but especially a big series like this and
being weeks between each iteration it's difficult to follow and see
which review comments that you have chosen to implement and how. It's
almost a full review each time. Even if I see that I have acked/rewieved
a patch, if I don't remember, I have to go back to the previous version
and see if I had any comments and if you followed up on that.
>>> +/**
>>> + * enum drm_connector_tv_mode - Analog TV output mode
>>> + *
>>> + * This enum is used to indicate the TV output mode used on an analog TV
>>> + * connector.
>>> + *
>>> + * WARNING: The values of this enum is uABI since they're exposed in the
>>> + * "TV mode" connector property.
>>> + */
>>> +enum drm_connector_tv_mode {
>>> + /**
>>> + * @DRM_MODE_TV_MODE_NONE: Placeholder to not default on one
>>> + * variant or the other when nothing is set.
>>> + */
>>> + DRM_MODE_TV_MODE_NONE = 0,
>>
>> How is this supposed to be used?
>
> It's not supposed to be used. It was a suggestion from Mateusz to avoid
> to default to any standard when we don't initialize something. I don't
> have any strong feeling about it, so I can drop it if you prefer.
>
The confusing thing to me is that "None" is part of the property enum
list, so the idea is that it can end up in userspace if there's a driver
error? Hmm, that won't work since TV_MODE_NONE won't be part of the
bitmask that the driver sets. So userspace reading the property ends up
with a value for which there's no enum name to match.
So usespace should be trained to know that zero for this property is a
driver error? No, not a good idea.
I think to catch a bug like this drm_atomic_connector_get_property()
should check the tv.mode value and see if it's a legal enum value and if
not it has to just pick a legal one and print an error. But I'm not sure
it's worth it to catch a bug like this. And I don't see any other enum
properties being checked for validity either before being returned to
userspace.
Based on this reasoning I think you should drop the NONE value.
Noralf.