| Message ID | 20230208211016.7034-1-ville.syrjala@linux.intel.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | [v2,1/2] drm: Introduce plane SIZE_HINTS property | expand |
On Wed, 8 Feb 2023 23:10:16 +0200 Ville Syrjala <ville.syrjala@linux.intel.com> wrote: > From: Ville Syrjälä <ville.syrjala@linux.intel.com> > > Add a new immutable plane property by which a plane can advertise > a handful of recommended plane sizes. This would be mostly exposed > by cursor planes as a slightly more capable replacement for > the DRM_CAP_CURSOR_WIDTH/HEIGHT caps, which can only declare > a one size fits all limit for the whole device. > > Currently eg. amdgpu/i915/nouveau just advertize the max cursor > size via the cursor size caps. But always using the max sized > cursor can waste a surprising amount of power, so a better > stragey is desirable. > > Most other drivers don't specify any cursor size at all, in > which case the ioctl code just claims that 64x64 is a great > choice. Whether that is actually true is debatable. > > A poll of various compositor developers informs us that > blindly probing with setcursor/atomic ioctl to determine > suitable cursor sizes is not acceptable, thus the > introduction of the new property to supplant the cursor > size caps. The compositor will now be free to select a > more optimal cursor size from the short list of options. > > Note that the reported sizes (either via the property or the > caps) make no claims about things such as plane scaling. So > these things should only really be consulted for simple > "cursor like" use cases. > > v2: Try to add some docs > > Cc: Simon Ser <contact@emersion.fr> > Cc: Jonas Ådahl <jadahl@redhat.com> > Cc: Daniel Stone <daniel@fooishbar.org> > Cc: Pekka Paalanen <pekka.paalanen@collabora.com> > Acked-by: Harry Wentland <harry.wentland@amd.com> > Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> > --- > drivers/gpu/drm/drm_mode_config.c | 7 +++++ > drivers/gpu/drm/drm_plane.c | 48 +++++++++++++++++++++++++++++++ > include/drm/drm_mode_config.h | 5 ++++ > include/drm/drm_plane.h | 4 +++ > include/uapi/drm/drm_mode.h | 11 +++++++ > 5 files changed, 75 insertions(+) > > diff --git a/drivers/gpu/drm/drm_mode_config.c b/drivers/gpu/drm/drm_mode_config.c > index 87eb591fe9b5..21860f94a18c 100644 > --- a/drivers/gpu/drm/drm_mode_config.c > +++ b/drivers/gpu/drm/drm_mode_config.c > @@ -374,6 +374,13 @@ static int drm_mode_create_standard_properties(struct drm_device *dev) > return -ENOMEM; > dev->mode_config.modifiers_property = prop; > > + prop = drm_property_create(dev, > + DRM_MODE_PROP_IMMUTABLE | DRM_MODE_PROP_BLOB, > + "SIZE_HINTS", 0); > + if (!prop) > + return -ENOMEM; > + dev->mode_config.size_hints_property = prop; > + > return 0; > } > > diff --git a/drivers/gpu/drm/drm_plane.c b/drivers/gpu/drm/drm_plane.c > index 24e7998d1731..ae51b1f83755 100644 > --- a/drivers/gpu/drm/drm_plane.c > +++ b/drivers/gpu/drm/drm_plane.c > @@ -140,6 +140,21 @@ > * DRM_FORMAT_MOD_LINEAR. Before linux kernel release v5.1 there have been > * various bugs in this area with inconsistencies between the capability > * flag and per-plane properties. > + * > + * SIZE_HINTS: > + * Blob property which contains the set of recommended plane size > + * which can used for simple "cursor like" use cases (eg. no scaling). > + * Using these hints frees userspace from extensive probing of > + * supported plane sizes through atomic/setcursor ioctls. > + * > + * For optimal usage userspace should pick the smallest size > + * that satisfies its own requirements. > + * > + * The blob contains an array of struct drm_plane_size_hint. > + * > + * Drivers should only attach this property to planes that > + * support a very limited set of sizes (eg. cursor planes > + * on typical hardware). Hi Ville, sounds good. Maybe a minor nit about "typical hardware". Would e.g. "legacy PC hardware" be more accurate? Acked-by: Pekka Paalanen <pekka.paalanen@collabora.com> but let's see if that other option (allocate cap size, make FB with image size, guarantee zero padding) from my other email would be viable too. Thanks, pq > */ > > static unsigned int drm_num_planes(struct drm_device *dev) > @@ -1582,3 +1597,36 @@ int drm_plane_create_scaling_filter_property(struct drm_plane *plane, > return 0; > } > EXPORT_SYMBOL(drm_plane_create_scaling_filter_property); > + > +/** > + * drm_plane_add_size_hint_property - create a size hint property > + * > + * @plane: drm plane > + * @hints: size hints > + * @num_hints: number of size hints > + * > + * Create a size hints property for the plane. > + * > + * RETURNS: > + * Zero for success or -errno > + */ > +int drm_plane_add_size_hints_property(struct drm_plane *plane, > + const struct drm_plane_size_hint *hints, > + int num_hints) > +{ > + struct drm_device *dev = plane->dev; > + struct drm_mode_config *config = &dev->mode_config; > + struct drm_property_blob *blob; > + > + blob = drm_property_create_blob(dev, > + array_size(sizeof(hints[0]), num_hints), > + hints); > + if (IS_ERR(blob)) > + return PTR_ERR(blob); > + > + drm_object_attach_property(&plane->base, config->size_hints_property, > + blob->base.id); > + > + return 0; > +} > +EXPORT_SYMBOL(drm_plane_add_size_hints_property); > diff --git a/include/drm/drm_mode_config.h b/include/drm/drm_mode_config.h > index e5b053001d22..5bc8aed9b445 100644 > --- a/include/drm/drm_mode_config.h > +++ b/include/drm/drm_mode_config.h > @@ -949,6 +949,11 @@ struct drm_mode_config { > */ > struct drm_property *modifiers_property; > > + /** > + * @size_hints_propertty: Plane SIZE_HINTS property. > + */ > + struct drm_property *size_hints_property; > + > /* cursor size */ > uint32_t cursor_width, cursor_height; > > diff --git a/include/drm/drm_plane.h b/include/drm/drm_plane.h > index 51291983ea44..1997d7d64b69 100644 > --- a/include/drm/drm_plane.h > +++ b/include/drm/drm_plane.h > @@ -32,6 +32,7 @@ > #include <drm/drm_util.h> > > struct drm_crtc; > +struct drm_plane_size_hint; > struct drm_printer; > struct drm_modeset_acquire_ctx; > > @@ -945,5 +946,8 @@ drm_plane_get_damage_clips(const struct drm_plane_state *state); > > int drm_plane_create_scaling_filter_property(struct drm_plane *plane, > unsigned int supported_filters); > +int drm_plane_add_size_hints_property(struct drm_plane *plane, > + const struct drm_plane_size_hint *hints, > + int num_hints); > > #endif > diff --git a/include/uapi/drm/drm_mode.h b/include/uapi/drm/drm_mode.h > index 46becedf5b2f..9d7c5967264f 100644 > --- a/include/uapi/drm/drm_mode.h > +++ b/include/uapi/drm/drm_mode.h > @@ -849,6 +849,17 @@ struct drm_color_lut { > __u16 reserved; > }; > > +/** > + * struct drm_plane_size_hint - Plane size hints > + * > + * The plane SIZE_HINTS property blob contains an > + * array of struct drm_plane_size_hint. > + */ > +struct drm_plane_size_hint { > + __u16 width; > + __u16 height; > +}; > + > /** > * struct hdr_metadata_infoframe - HDR Metadata Infoframe Data. > *
On Thu, Feb 09, 2023 at 01:58:55PM +0200, Pekka Paalanen wrote: > On Wed, 8 Feb 2023 23:10:16 +0200 > Ville Syrjala <ville.syrjala@linux.intel.com> wrote: > > > From: Ville Syrjälä <ville.syrjala@linux.intel.com> > > > > Add a new immutable plane property by which a plane can advertise > > a handful of recommended plane sizes. This would be mostly exposed > > by cursor planes as a slightly more capable replacement for > > the DRM_CAP_CURSOR_WIDTH/HEIGHT caps, which can only declare > > a one size fits all limit for the whole device. > > > > Currently eg. amdgpu/i915/nouveau just advertize the max cursor > > size via the cursor size caps. But always using the max sized > > cursor can waste a surprising amount of power, so a better > > stragey is desirable. > > > > Most other drivers don't specify any cursor size at all, in > > which case the ioctl code just claims that 64x64 is a great > > choice. Whether that is actually true is debatable. > > > > A poll of various compositor developers informs us that > > blindly probing with setcursor/atomic ioctl to determine > > suitable cursor sizes is not acceptable, thus the > > introduction of the new property to supplant the cursor > > size caps. The compositor will now be free to select a > > more optimal cursor size from the short list of options. > > > > Note that the reported sizes (either via the property or the > > caps) make no claims about things such as plane scaling. So > > these things should only really be consulted for simple > > "cursor like" use cases. > > > > v2: Try to add some docs > > > > Cc: Simon Ser <contact@emersion.fr> > > Cc: Jonas Ådahl <jadahl@redhat.com> > > Cc: Daniel Stone <daniel@fooishbar.org> > > Cc: Pekka Paalanen <pekka.paalanen@collabora.com> > > Acked-by: Harry Wentland <harry.wentland@amd.com> > > Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> > > --- > > drivers/gpu/drm/drm_mode_config.c | 7 +++++ > > drivers/gpu/drm/drm_plane.c | 48 +++++++++++++++++++++++++++++++ > > include/drm/drm_mode_config.h | 5 ++++ > > include/drm/drm_plane.h | 4 +++ > > include/uapi/drm/drm_mode.h | 11 +++++++ > > 5 files changed, 75 insertions(+) > > > > diff --git a/drivers/gpu/drm/drm_mode_config.c b/drivers/gpu/drm/drm_mode_config.c > > index 87eb591fe9b5..21860f94a18c 100644 > > --- a/drivers/gpu/drm/drm_mode_config.c > > +++ b/drivers/gpu/drm/drm_mode_config.c > > @@ -374,6 +374,13 @@ static int drm_mode_create_standard_properties(struct drm_device *dev) > > return -ENOMEM; > > dev->mode_config.modifiers_property = prop; > > > > + prop = drm_property_create(dev, > > + DRM_MODE_PROP_IMMUTABLE | DRM_MODE_PROP_BLOB, > > + "SIZE_HINTS", 0); > > + if (!prop) > > + return -ENOMEM; > > + dev->mode_config.size_hints_property = prop; > > + > > return 0; > > } > > > > diff --git a/drivers/gpu/drm/drm_plane.c b/drivers/gpu/drm/drm_plane.c > > index 24e7998d1731..ae51b1f83755 100644 > > --- a/drivers/gpu/drm/drm_plane.c > > +++ b/drivers/gpu/drm/drm_plane.c > > @@ -140,6 +140,21 @@ > > * DRM_FORMAT_MOD_LINEAR. Before linux kernel release v5.1 there have been > > * various bugs in this area with inconsistencies between the capability > > * flag and per-plane properties. > > + * > > + * SIZE_HINTS: > > + * Blob property which contains the set of recommended plane size > > + * which can used for simple "cursor like" use cases (eg. no scaling). > > + * Using these hints frees userspace from extensive probing of > > + * supported plane sizes through atomic/setcursor ioctls. > > + * > > + * For optimal usage userspace should pick the smallest size > > + * that satisfies its own requirements. > > + * > > + * The blob contains an array of struct drm_plane_size_hint. > > + * > > + * Drivers should only attach this property to planes that > > + * support a very limited set of sizes (eg. cursor planes > > + * on typical hardware). > > Hi Ville, > > sounds good. Maybe a minor nit about "typical hardware". Would e.g. > "legacy PC hardware" be more accurate? "legacy" doesn't feel quite right for current and upcoming hardware.
On Wed, Feb 08, 2023 at 11:10:16PM +0200, Ville Syrjala wrote: > From: Ville Syrjälä <ville.syrjala@linux.intel.com> > > Add a new immutable plane property by which a plane can advertise > a handful of recommended plane sizes. This would be mostly exposed > by cursor planes as a slightly more capable replacement for > the DRM_CAP_CURSOR_WIDTH/HEIGHT caps, which can only declare > a one size fits all limit for the whole device. > > Currently eg. amdgpu/i915/nouveau just advertize the max cursor > size via the cursor size caps. But always using the max sized > cursor can waste a surprising amount of power, so a better > stragey is desirable. > > Most other drivers don't specify any cursor size at all, in > which case the ioctl code just claims that 64x64 is a great > choice. Whether that is actually true is debatable. > > A poll of various compositor developers informs us that > blindly probing with setcursor/atomic ioctl to determine > suitable cursor sizes is not acceptable, thus the > introduction of the new property to supplant the cursor > size caps. The compositor will now be free to select a > more optimal cursor size from the short list of options. > > Note that the reported sizes (either via the property or the > caps) make no claims about things such as plane scaling. So > these things should only really be consulted for simple > "cursor like" use cases. > > v2: Try to add some docs > > Cc: Simon Ser <contact@emersion.fr> > Cc: Jonas Ådahl <jadahl@redhat.com> > Cc: Daniel Stone <daniel@fooishbar.org> > Cc: Pekka Paalanen <pekka.paalanen@collabora.com> > Acked-by: Harry Wentland <harry.wentland@amd.com> > Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> > --- > drivers/gpu/drm/drm_mode_config.c | 7 +++++ > drivers/gpu/drm/drm_plane.c | 48 +++++++++++++++++++++++++++++++ > include/drm/drm_mode_config.h | 5 ++++ > include/drm/drm_plane.h | 4 +++ > include/uapi/drm/drm_mode.h | 11 +++++++ > 5 files changed, 75 insertions(+) > > diff --git a/drivers/gpu/drm/drm_mode_config.c b/drivers/gpu/drm/drm_mode_config.c > index 87eb591fe9b5..21860f94a18c 100644 > --- a/drivers/gpu/drm/drm_mode_config.c > +++ b/drivers/gpu/drm/drm_mode_config.c > @@ -374,6 +374,13 @@ static int drm_mode_create_standard_properties(struct drm_device *dev) > return -ENOMEM; > dev->mode_config.modifiers_property = prop; > > + prop = drm_property_create(dev, > + DRM_MODE_PROP_IMMUTABLE | DRM_MODE_PROP_BLOB, > + "SIZE_HINTS", 0); > + if (!prop) > + return -ENOMEM; > + dev->mode_config.size_hints_property = prop; > + > return 0; > } > > diff --git a/drivers/gpu/drm/drm_plane.c b/drivers/gpu/drm/drm_plane.c > index 24e7998d1731..ae51b1f83755 100644 > --- a/drivers/gpu/drm/drm_plane.c > +++ b/drivers/gpu/drm/drm_plane.c > @@ -140,6 +140,21 @@ > * DRM_FORMAT_MOD_LINEAR. Before linux kernel release v5.1 there have been > * various bugs in this area with inconsistencies between the capability > * flag and per-plane properties. > + * > + * SIZE_HINTS: > + * Blob property which contains the set of recommended plane size > + * which can used for simple "cursor like" use cases (eg. no scaling). > + * Using these hints frees userspace from extensive probing of > + * supported plane sizes through atomic/setcursor ioctls. > + * > + * For optimal usage userspace should pick the smallest size > + * that satisfies its own requirements. > + * > + * The blob contains an array of struct drm_plane_size_hint. > + * > + * Drivers should only attach this property to planes that > + * support a very limited set of sizes (eg. cursor planes > + * on typical hardware). This is a bit awkward since problematic drivers would only expose this property if they are new enough. A way to avoid this is for all new drivers expose this property, but special case the size 0x0 as "everything below the max limit goes". Then userspace can do (ignoring the missing cap fallback). if (has(SIZE_HINTS)) size = figure_out_size(SIZE_HINTS, DRM_CAP_CURSOR_WIDTH, DRM_CAP_CURSOR_HEIGHT, preferred_size); else size = DRM_CAP_CURSOR_WIDTH x DRM_CAP_CURSOR_WIDTH; With `figure_out_size()` knowing how to deal with 0x0 in the size hints to use `preferred_size` directly. Jonas > */ > > static unsigned int drm_num_planes(struct drm_device *dev) > @@ -1582,3 +1597,36 @@ int drm_plane_create_scaling_filter_property(struct drm_plane *plane, > return 0; > } > EXPORT_SYMBOL(drm_plane_create_scaling_filter_property); > + > +/** > + * drm_plane_add_size_hint_property - create a size hint property > + * > + * @plane: drm plane > + * @hints: size hints > + * @num_hints: number of size hints > + * > + * Create a size hints property for the plane. > + * > + * RETURNS: > + * Zero for success or -errno > + */ > +int drm_plane_add_size_hints_property(struct drm_plane *plane, > + const struct drm_plane_size_hint *hints, > + int num_hints) > +{ > + struct drm_device *dev = plane->dev; > + struct drm_mode_config *config = &dev->mode_config; > + struct drm_property_blob *blob; > + > + blob = drm_property_create_blob(dev, > + array_size(sizeof(hints[0]), num_hints), > + hints); > + if (IS_ERR(blob)) > + return PTR_ERR(blob); > + > + drm_object_attach_property(&plane->base, config->size_hints_property, > + blob->base.id); > + > + return 0; > +} > +EXPORT_SYMBOL(drm_plane_add_size_hints_property); > diff --git a/include/drm/drm_mode_config.h b/include/drm/drm_mode_config.h > index e5b053001d22..5bc8aed9b445 100644 > --- a/include/drm/drm_mode_config.h > +++ b/include/drm/drm_mode_config.h > @@ -949,6 +949,11 @@ struct drm_mode_config { > */ > struct drm_property *modifiers_property; > > + /** > + * @size_hints_propertty: Plane SIZE_HINTS property. > + */ > + struct drm_property *size_hints_property; > + > /* cursor size */ > uint32_t cursor_width, cursor_height; > > diff --git a/include/drm/drm_plane.h b/include/drm/drm_plane.h > index 51291983ea44..1997d7d64b69 100644 > --- a/include/drm/drm_plane.h > +++ b/include/drm/drm_plane.h > @@ -32,6 +32,7 @@ > #include <drm/drm_util.h> > > struct drm_crtc; > +struct drm_plane_size_hint; > struct drm_printer; > struct drm_modeset_acquire_ctx; > > @@ -945,5 +946,8 @@ drm_plane_get_damage_clips(const struct drm_plane_state *state); > > int drm_plane_create_scaling_filter_property(struct drm_plane *plane, > unsigned int supported_filters); > +int drm_plane_add_size_hints_property(struct drm_plane *plane, > + const struct drm_plane_size_hint *hints, > + int num_hints); > > #endif > diff --git a/include/uapi/drm/drm_mode.h b/include/uapi/drm/drm_mode.h > index 46becedf5b2f..9d7c5967264f 100644 > --- a/include/uapi/drm/drm_mode.h > +++ b/include/uapi/drm/drm_mode.h > @@ -849,6 +849,17 @@ struct drm_color_lut { > __u16 reserved; > }; > > +/** > + * struct drm_plane_size_hint - Plane size hints > + * > + * The plane SIZE_HINTS property blob contains an > + * array of struct drm_plane_size_hint. > + */ > +struct drm_plane_size_hint { > + __u16 width; > + __u16 height; > +}; > + > /** > * struct hdr_metadata_infoframe - HDR Metadata Infoframe Data. > * > -- > 2.39.1 >
On Thu, 9 Feb 2023 15:10:38 +0200 Ville Syrjälä <ville.syrjala@linux.intel.com> wrote: > On Thu, Feb 09, 2023 at 01:58:55PM +0200, Pekka Paalanen wrote: > > On Wed, 8 Feb 2023 23:10:16 +0200 > > Ville Syrjala <ville.syrjala@linux.intel.com> wrote: > > > > > From: Ville Syrjälä <ville.syrjala@linux.intel.com> > > > > > > Add a new immutable plane property by which a plane can advertise > > > a handful of recommended plane sizes. This would be mostly exposed > > > by cursor planes as a slightly more capable replacement for > > > the DRM_CAP_CURSOR_WIDTH/HEIGHT caps, which can only declare > > > a one size fits all limit for the whole device. > > > > > > Currently eg. amdgpu/i915/nouveau just advertize the max cursor > > > size via the cursor size caps. But always using the max sized > > > cursor can waste a surprising amount of power, so a better > > > stragey is desirable. > > > > > > Most other drivers don't specify any cursor size at all, in > > > which case the ioctl code just claims that 64x64 is a great > > > choice. Whether that is actually true is debatable. > > > > > > A poll of various compositor developers informs us that > > > blindly probing with setcursor/atomic ioctl to determine > > > suitable cursor sizes is not acceptable, thus the > > > introduction of the new property to supplant the cursor > > > size caps. The compositor will now be free to select a > > > more optimal cursor size from the short list of options. > > > > > > Note that the reported sizes (either via the property or the > > > caps) make no claims about things such as plane scaling. So > > > these things should only really be consulted for simple > > > "cursor like" use cases. > > > > > > v2: Try to add some docs > > > > > > Cc: Simon Ser <contact@emersion.fr> > > > Cc: Jonas Ådahl <jadahl@redhat.com> > > > Cc: Daniel Stone <daniel@fooishbar.org> > > > Cc: Pekka Paalanen <pekka.paalanen@collabora.com> > > > Acked-by: Harry Wentland <harry.wentland@amd.com> > > > Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> > > > --- > > > drivers/gpu/drm/drm_mode_config.c | 7 +++++ > > > drivers/gpu/drm/drm_plane.c | 48 +++++++++++++++++++++++++++++++ > > > include/drm/drm_mode_config.h | 5 ++++ > > > include/drm/drm_plane.h | 4 +++ > > > include/uapi/drm/drm_mode.h | 11 +++++++ > > > 5 files changed, 75 insertions(+) > > > > > > diff --git a/drivers/gpu/drm/drm_mode_config.c b/drivers/gpu/drm/drm_mode_config.c > > > index 87eb591fe9b5..21860f94a18c 100644 > > > --- a/drivers/gpu/drm/drm_mode_config.c > > > +++ b/drivers/gpu/drm/drm_mode_config.c > > > @@ -374,6 +374,13 @@ static int drm_mode_create_standard_properties(struct drm_device *dev) > > > return -ENOMEM; > > > dev->mode_config.modifiers_property = prop; > > > > > > + prop = drm_property_create(dev, > > > + DRM_MODE_PROP_IMMUTABLE | DRM_MODE_PROP_BLOB, > > > + "SIZE_HINTS", 0); > > > + if (!prop) > > > + return -ENOMEM; > > > + dev->mode_config.size_hints_property = prop; > > > + > > > return 0; > > > } > > > > > > diff --git a/drivers/gpu/drm/drm_plane.c b/drivers/gpu/drm/drm_plane.c > > > index 24e7998d1731..ae51b1f83755 100644 > > > --- a/drivers/gpu/drm/drm_plane.c > > > +++ b/drivers/gpu/drm/drm_plane.c > > > @@ -140,6 +140,21 @@ > > > * DRM_FORMAT_MOD_LINEAR. Before linux kernel release v5.1 there have been > > > * various bugs in this area with inconsistencies between the capability > > > * flag and per-plane properties. > > > + * > > > + * SIZE_HINTS: > > > + * Blob property which contains the set of recommended plane size > > > + * which can used for simple "cursor like" use cases (eg. no scaling). > > > + * Using these hints frees userspace from extensive probing of > > > + * supported plane sizes through atomic/setcursor ioctls. > > > + * > > > + * For optimal usage userspace should pick the smallest size > > > + * that satisfies its own requirements. > > > + * > > > + * The blob contains an array of struct drm_plane_size_hint. > > > + * > > > + * Drivers should only attach this property to planes that > > > + * support a very limited set of sizes (eg. cursor planes > > > + * on typical hardware). > > > > Hi Ville, > > > > sounds good. Maybe a minor nit about "typical hardware". Would e.g. > > "legacy PC hardware" be more accurate? > > "legacy" doesn't feel quite right for current and upcoming hardware. It's an example, not everything. Although, I didn't expect current and upcoming hardware to keep such limitations either but to move towards universal rather than specialized planes. Maybe just drop the whole "(eg. ...)"? Thanks, pq
On Thu, Feb 09, 2023 at 03:16:23PM +0100, Jonas Ådahl wrote: > On Wed, Feb 08, 2023 at 11:10:16PM +0200, Ville Syrjala wrote: > > From: Ville Syrjälä <ville.syrjala@linux.intel.com> > > > > Add a new immutable plane property by which a plane can advertise > > a handful of recommended plane sizes. This would be mostly exposed > > by cursor planes as a slightly more capable replacement for > > the DRM_CAP_CURSOR_WIDTH/HEIGHT caps, which can only declare > > a one size fits all limit for the whole device. > > > > Currently eg. amdgpu/i915/nouveau just advertize the max cursor > > size via the cursor size caps. But always using the max sized > > cursor can waste a surprising amount of power, so a better > > stragey is desirable. > > > > Most other drivers don't specify any cursor size at all, in > > which case the ioctl code just claims that 64x64 is a great > > choice. Whether that is actually true is debatable. > > > > A poll of various compositor developers informs us that > > blindly probing with setcursor/atomic ioctl to determine > > suitable cursor sizes is not acceptable, thus the > > introduction of the new property to supplant the cursor > > size caps. The compositor will now be free to select a > > more optimal cursor size from the short list of options. > > > > Note that the reported sizes (either via the property or the > > caps) make no claims about things such as plane scaling. So > > these things should only really be consulted for simple > > "cursor like" use cases. > > > > v2: Try to add some docs > > > > Cc: Simon Ser <contact@emersion.fr> > > Cc: Jonas Ådahl <jadahl@redhat.com> > > Cc: Daniel Stone <daniel@fooishbar.org> > > Cc: Pekka Paalanen <pekka.paalanen@collabora.com> > > Acked-by: Harry Wentland <harry.wentland@amd.com> > > Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> > > --- > > drivers/gpu/drm/drm_mode_config.c | 7 +++++ > > drivers/gpu/drm/drm_plane.c | 48 +++++++++++++++++++++++++++++++ > > include/drm/drm_mode_config.h | 5 ++++ > > include/drm/drm_plane.h | 4 +++ > > include/uapi/drm/drm_mode.h | 11 +++++++ > > 5 files changed, 75 insertions(+) > > > > diff --git a/drivers/gpu/drm/drm_mode_config.c b/drivers/gpu/drm/drm_mode_config.c > > index 87eb591fe9b5..21860f94a18c 100644 > > --- a/drivers/gpu/drm/drm_mode_config.c > > +++ b/drivers/gpu/drm/drm_mode_config.c > > @@ -374,6 +374,13 @@ static int drm_mode_create_standard_properties(struct drm_device *dev) > > return -ENOMEM; > > dev->mode_config.modifiers_property = prop; > > > > + prop = drm_property_create(dev, > > + DRM_MODE_PROP_IMMUTABLE | DRM_MODE_PROP_BLOB, > > + "SIZE_HINTS", 0); > > + if (!prop) > > + return -ENOMEM; > > + dev->mode_config.size_hints_property = prop; > > + > > return 0; > > } > > > > diff --git a/drivers/gpu/drm/drm_plane.c b/drivers/gpu/drm/drm_plane.c > > index 24e7998d1731..ae51b1f83755 100644 > > --- a/drivers/gpu/drm/drm_plane.c > > +++ b/drivers/gpu/drm/drm_plane.c > > @@ -140,6 +140,21 @@ > > * DRM_FORMAT_MOD_LINEAR. Before linux kernel release v5.1 there have been > > * various bugs in this area with inconsistencies between the capability > > * flag and per-plane properties. > > + * > > + * SIZE_HINTS: > > + * Blob property which contains the set of recommended plane size > > + * which can used for simple "cursor like" use cases (eg. no scaling). > > + * Using these hints frees userspace from extensive probing of > > + * supported plane sizes through atomic/setcursor ioctls. > > + * > > + * For optimal usage userspace should pick the smallest size > > + * that satisfies its own requirements. > > + * > > + * The blob contains an array of struct drm_plane_size_hint. > > + * > > + * Drivers should only attach this property to planes that > > + * support a very limited set of sizes (eg. cursor planes > > + * on typical hardware). > > This is a bit awkward since problematic drivers would only expose > this property if they are new enough. > > A way to avoid this is for all new drivers expose this property, but > special case the size 0x0 as "everything below the max limit goes". Then > userspace can do (ignoring the missing cap fallback). I don't think there are any drivers that need that. So I'm thinking we can just ignore that for now. > > if (has(SIZE_HINTS)) > size = figure_out_size(SIZE_HINTS, > DRM_CAP_CURSOR_WIDTH, > DRM_CAP_CURSOR_HEIGHT, > preferred_size); > else > size = DRM_CAP_CURSOR_WIDTH x DRM_CAP_CURSOR_WIDTH; > > With `figure_out_size()` knowing how to deal with 0x0 in the size hints > to use `preferred_size` directly. > > > Jonas > > > */ > > > > static unsigned int drm_num_planes(struct drm_device *dev) > > @@ -1582,3 +1597,36 @@ int drm_plane_create_scaling_filter_property(struct drm_plane *plane, > > return 0; > > } > > EXPORT_SYMBOL(drm_plane_create_scaling_filter_property); > > + > > +/** > > + * drm_plane_add_size_hint_property - create a size hint property > > + * > > + * @plane: drm plane > > + * @hints: size hints > > + * @num_hints: number of size hints > > + * > > + * Create a size hints property for the plane. > > + * > > + * RETURNS: > > + * Zero for success or -errno > > + */ > > +int drm_plane_add_size_hints_property(struct drm_plane *plane, > > + const struct drm_plane_size_hint *hints, > > + int num_hints) > > +{ > > + struct drm_device *dev = plane->dev; > > + struct drm_mode_config *config = &dev->mode_config; > > + struct drm_property_blob *blob; > > + > > + blob = drm_property_create_blob(dev, > > + array_size(sizeof(hints[0]), num_hints), > > + hints); > > + if (IS_ERR(blob)) > > + return PTR_ERR(blob); > > + > > + drm_object_attach_property(&plane->base, config->size_hints_property, > > + blob->base.id); > > + > > + return 0; > > +} > > +EXPORT_SYMBOL(drm_plane_add_size_hints_property); > > diff --git a/include/drm/drm_mode_config.h b/include/drm/drm_mode_config.h > > index e5b053001d22..5bc8aed9b445 100644 > > --- a/include/drm/drm_mode_config.h > > +++ b/include/drm/drm_mode_config.h > > @@ -949,6 +949,11 @@ struct drm_mode_config { > > */ > > struct drm_property *modifiers_property; > > > > + /** > > + * @size_hints_propertty: Plane SIZE_HINTS property. > > + */ > > + struct drm_property *size_hints_property; > > + > > /* cursor size */ > > uint32_t cursor_width, cursor_height; > > > > diff --git a/include/drm/drm_plane.h b/include/drm/drm_plane.h > > index 51291983ea44..1997d7d64b69 100644 > > --- a/include/drm/drm_plane.h > > +++ b/include/drm/drm_plane.h > > @@ -32,6 +32,7 @@ > > #include <drm/drm_util.h> > > > > struct drm_crtc; > > +struct drm_plane_size_hint; > > struct drm_printer; > > struct drm_modeset_acquire_ctx; > > > > @@ -945,5 +946,8 @@ drm_plane_get_damage_clips(const struct drm_plane_state *state); > > > > int drm_plane_create_scaling_filter_property(struct drm_plane *plane, > > unsigned int supported_filters); > > +int drm_plane_add_size_hints_property(struct drm_plane *plane, > > + const struct drm_plane_size_hint *hints, > > + int num_hints); > > > > #endif > > diff --git a/include/uapi/drm/drm_mode.h b/include/uapi/drm/drm_mode.h > > index 46becedf5b2f..9d7c5967264f 100644 > > --- a/include/uapi/drm/drm_mode.h > > +++ b/include/uapi/drm/drm_mode.h > > @@ -849,6 +849,17 @@ struct drm_color_lut { > > __u16 reserved; > > }; > > > > +/** > > + * struct drm_plane_size_hint - Plane size hints > > + * > > + * The plane SIZE_HINTS property blob contains an > > + * array of struct drm_plane_size_hint. > > + */ > > +struct drm_plane_size_hint { > > + __u16 width; > > + __u16 height; > > +}; > > + > > /** > > * struct hdr_metadata_infoframe - HDR Metadata Infoframe Data. > > * > > -- > > 2.39.1 > >
On Tue, Feb 14, 2023 at 11:25:56AM +0200, Ville Syrjälä wrote: > On Thu, Feb 09, 2023 at 03:16:23PM +0100, Jonas Ådahl wrote: > > On Wed, Feb 08, 2023 at 11:10:16PM +0200, Ville Syrjala wrote: > > > From: Ville Syrjälä <ville.syrjala@linux.intel.com> > > > > > > Add a new immutable plane property by which a plane can advertise > > > a handful of recommended plane sizes. This would be mostly exposed > > > by cursor planes as a slightly more capable replacement for > > > the DRM_CAP_CURSOR_WIDTH/HEIGHT caps, which can only declare > > > a one size fits all limit for the whole device. > > > > > > Currently eg. amdgpu/i915/nouveau just advertize the max cursor > > > size via the cursor size caps. But always using the max sized > > > cursor can waste a surprising amount of power, so a better > > > stragey is desirable. > > > > > > Most other drivers don't specify any cursor size at all, in > > > which case the ioctl code just claims that 64x64 is a great > > > choice. Whether that is actually true is debatable. > > > > > > A poll of various compositor developers informs us that > > > blindly probing with setcursor/atomic ioctl to determine > > > suitable cursor sizes is not acceptable, thus the > > > introduction of the new property to supplant the cursor > > > size caps. The compositor will now be free to select a > > > more optimal cursor size from the short list of options. > > > > > > Note that the reported sizes (either via the property or the > > > caps) make no claims about things such as plane scaling. So > > > these things should only really be consulted for simple > > > "cursor like" use cases. > > > > > > v2: Try to add some docs > > > > > > Cc: Simon Ser <contact@emersion.fr> > > > Cc: Jonas Ådahl <jadahl@redhat.com> > > > Cc: Daniel Stone <daniel@fooishbar.org> > > > Cc: Pekka Paalanen <pekka.paalanen@collabora.com> > > > Acked-by: Harry Wentland <harry.wentland@amd.com> > > > Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> > > > --- > > > drivers/gpu/drm/drm_mode_config.c | 7 +++++ > > > drivers/gpu/drm/drm_plane.c | 48 +++++++++++++++++++++++++++++++ > > > include/drm/drm_mode_config.h | 5 ++++ > > > include/drm/drm_plane.h | 4 +++ > > > include/uapi/drm/drm_mode.h | 11 +++++++ > > > 5 files changed, 75 insertions(+) > > > > > > diff --git a/drivers/gpu/drm/drm_mode_config.c b/drivers/gpu/drm/drm_mode_config.c > > > index 87eb591fe9b5..21860f94a18c 100644 > > > --- a/drivers/gpu/drm/drm_mode_config.c > > > +++ b/drivers/gpu/drm/drm_mode_config.c > > > @@ -374,6 +374,13 @@ static int drm_mode_create_standard_properties(struct drm_device *dev) > > > return -ENOMEM; > > > dev->mode_config.modifiers_property = prop; > > > > > > + prop = drm_property_create(dev, > > > + DRM_MODE_PROP_IMMUTABLE | DRM_MODE_PROP_BLOB, > > > + "SIZE_HINTS", 0); > > > + if (!prop) > > > + return -ENOMEM; > > > + dev->mode_config.size_hints_property = prop; > > > + > > > return 0; > > > } > > > > > > diff --git a/drivers/gpu/drm/drm_plane.c b/drivers/gpu/drm/drm_plane.c > > > index 24e7998d1731..ae51b1f83755 100644 > > > --- a/drivers/gpu/drm/drm_plane.c > > > +++ b/drivers/gpu/drm/drm_plane.c > > > @@ -140,6 +140,21 @@ > > > * DRM_FORMAT_MOD_LINEAR. Before linux kernel release v5.1 there have been > > > * various bugs in this area with inconsistencies between the capability > > > * flag and per-plane properties. > > > + * > > > + * SIZE_HINTS: > > > + * Blob property which contains the set of recommended plane size > > > + * which can used for simple "cursor like" use cases (eg. no scaling). > > > + * Using these hints frees userspace from extensive probing of > > > + * supported plane sizes through atomic/setcursor ioctls. > > > + * > > > + * For optimal usage userspace should pick the smallest size > > > + * that satisfies its own requirements. > > > + * > > > + * The blob contains an array of struct drm_plane_size_hint. > > > + * > > > + * Drivers should only attach this property to planes that > > > + * support a very limited set of sizes (eg. cursor planes > > > + * on typical hardware). > > > > This is a bit awkward since problematic drivers would only expose > > this property if they are new enough. > > > > A way to avoid this is for all new drivers expose this property, but > > special case the size 0x0 as "everything below the max limit goes". Then > > userspace can do (ignoring the missing cap fallback). > > I don't think there are any drivers that need that. > So I'm thinking we can just ignore that for now. None the less, userspace not seeing SIZE_HINTS will be required to indefinitely use the existing "old" behavior using the size cap as the buffer size with a fallback, and drivers without any size limitations that, i.e. details that are hard to express with a set of accepted sizes, will still use the inoptimal buffer sizes. If I read [1] correctly, AMD has no particular size limitations other than a size limit, but without a SIZE_HINTS, userspace still needs to use the maximum size. [1] https://gitlab.freedesktop.org/drm/intel/-/issues/7687#note_1760865 Jonas > > > > > if (has(SIZE_HINTS)) > > size = figure_out_size(SIZE_HINTS, > > DRM_CAP_CURSOR_WIDTH, > > DRM_CAP_CURSOR_HEIGHT, > > preferred_size); > > else > > size = DRM_CAP_CURSOR_WIDTH x DRM_CAP_CURSOR_WIDTH; > > > > With `figure_out_size()` knowing how to deal with 0x0 in the size hints > > to use `preferred_size` directly. > > > > > > Jonas > > > > > */ > > > > > > static unsigned int drm_num_planes(struct drm_device *dev) > > > @@ -1582,3 +1597,36 @@ int drm_plane_create_scaling_filter_property(struct drm_plane *plane, > > > return 0; > > > } > > > EXPORT_SYMBOL(drm_plane_create_scaling_filter_property); > > > + > > > +/** > > > + * drm_plane_add_size_hint_property - create a size hint property > > > + * > > > + * @plane: drm plane > > > + * @hints: size hints > > > + * @num_hints: number of size hints > > > + * > > > + * Create a size hints property for the plane. > > > + * > > > + * RETURNS: > > > + * Zero for success or -errno > > > + */ > > > +int drm_plane_add_size_hints_property(struct drm_plane *plane, > > > + const struct drm_plane_size_hint *hints, > > > + int num_hints) > > > +{ > > > + struct drm_device *dev = plane->dev; > > > + struct drm_mode_config *config = &dev->mode_config; > > > + struct drm_property_blob *blob; > > > + > > > + blob = drm_property_create_blob(dev, > > > + array_size(sizeof(hints[0]), num_hints), > > > + hints); > > > + if (IS_ERR(blob)) > > > + return PTR_ERR(blob); > > > + > > > + drm_object_attach_property(&plane->base, config->size_hints_property, > > > + blob->base.id); > > > + > > > + return 0; > > > +} > > > +EXPORT_SYMBOL(drm_plane_add_size_hints_property); > > > diff --git a/include/drm/drm_mode_config.h b/include/drm/drm_mode_config.h > > > index e5b053001d22..5bc8aed9b445 100644 > > > --- a/include/drm/drm_mode_config.h > > > +++ b/include/drm/drm_mode_config.h > > > @@ -949,6 +949,11 @@ struct drm_mode_config { > > > */ > > > struct drm_property *modifiers_property; > > > > > > + /** > > > + * @size_hints_propertty: Plane SIZE_HINTS property. > > > + */ > > > + struct drm_property *size_hints_property; > > > + > > > /* cursor size */ > > > uint32_t cursor_width, cursor_height; > > > > > > diff --git a/include/drm/drm_plane.h b/include/drm/drm_plane.h > > > index 51291983ea44..1997d7d64b69 100644 > > > --- a/include/drm/drm_plane.h > > > +++ b/include/drm/drm_plane.h > > > @@ -32,6 +32,7 @@ > > > #include <drm/drm_util.h> > > > > > > struct drm_crtc; > > > +struct drm_plane_size_hint; > > > struct drm_printer; > > > struct drm_modeset_acquire_ctx; > > > > > > @@ -945,5 +946,8 @@ drm_plane_get_damage_clips(const struct drm_plane_state *state); > > > > > > int drm_plane_create_scaling_filter_property(struct drm_plane *plane, > > > unsigned int supported_filters); > > > +int drm_plane_add_size_hints_property(struct drm_plane *plane, > > > + const struct drm_plane_size_hint *hints, > > > + int num_hints); > > > > > > #endif > > > diff --git a/include/uapi/drm/drm_mode.h b/include/uapi/drm/drm_mode.h > > > index 46becedf5b2f..9d7c5967264f 100644 > > > --- a/include/uapi/drm/drm_mode.h > > > +++ b/include/uapi/drm/drm_mode.h > > > @@ -849,6 +849,17 @@ struct drm_color_lut { > > > __u16 reserved; > > > }; > > > > > > +/** > > > + * struct drm_plane_size_hint - Plane size hints > > > + * > > > + * The plane SIZE_HINTS property blob contains an > > > + * array of struct drm_plane_size_hint. > > > + */ > > > +struct drm_plane_size_hint { > > > + __u16 width; > > > + __u16 height; > > > +}; > > > + > > > /** > > > * struct hdr_metadata_infoframe - HDR Metadata Infoframe Data. > > > * > > > -- > > > 2.39.1 > > > > > -- > Ville Syrjälä > Intel >
On Tue, Feb 14, 2023 at 10:54:27AM +0100, Jonas Ådahl wrote: > On Tue, Feb 14, 2023 at 11:25:56AM +0200, Ville Syrjälä wrote: > > On Thu, Feb 09, 2023 at 03:16:23PM +0100, Jonas Ådahl wrote: > > > On Wed, Feb 08, 2023 at 11:10:16PM +0200, Ville Syrjala wrote: > > > > From: Ville Syrjälä <ville.syrjala@linux.intel.com> > > > > > > > > Add a new immutable plane property by which a plane can advertise > > > > a handful of recommended plane sizes. This would be mostly exposed > > > > by cursor planes as a slightly more capable replacement for > > > > the DRM_CAP_CURSOR_WIDTH/HEIGHT caps, which can only declare > > > > a one size fits all limit for the whole device. > > > > > > > > Currently eg. amdgpu/i915/nouveau just advertize the max cursor > > > > size via the cursor size caps. But always using the max sized > > > > cursor can waste a surprising amount of power, so a better > > > > stragey is desirable. > > > > > > > > Most other drivers don't specify any cursor size at all, in > > > > which case the ioctl code just claims that 64x64 is a great > > > > choice. Whether that is actually true is debatable. > > > > > > > > A poll of various compositor developers informs us that > > > > blindly probing with setcursor/atomic ioctl to determine > > > > suitable cursor sizes is not acceptable, thus the > > > > introduction of the new property to supplant the cursor > > > > size caps. The compositor will now be free to select a > > > > more optimal cursor size from the short list of options. > > > > > > > > Note that the reported sizes (either via the property or the > > > > caps) make no claims about things such as plane scaling. So > > > > these things should only really be consulted for simple > > > > "cursor like" use cases. > > > > > > > > v2: Try to add some docs > > > > > > > > Cc: Simon Ser <contact@emersion.fr> > > > > Cc: Jonas Ådahl <jadahl@redhat.com> > > > > Cc: Daniel Stone <daniel@fooishbar.org> > > > > Cc: Pekka Paalanen <pekka.paalanen@collabora.com> > > > > Acked-by: Harry Wentland <harry.wentland@amd.com> > > > > Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> > > > > --- > > > > drivers/gpu/drm/drm_mode_config.c | 7 +++++ > > > > drivers/gpu/drm/drm_plane.c | 48 +++++++++++++++++++++++++++++++ > > > > include/drm/drm_mode_config.h | 5 ++++ > > > > include/drm/drm_plane.h | 4 +++ > > > > include/uapi/drm/drm_mode.h | 11 +++++++ > > > > 5 files changed, 75 insertions(+) > > > > > > > > diff --git a/drivers/gpu/drm/drm_mode_config.c b/drivers/gpu/drm/drm_mode_config.c > > > > index 87eb591fe9b5..21860f94a18c 100644 > > > > --- a/drivers/gpu/drm/drm_mode_config.c > > > > +++ b/drivers/gpu/drm/drm_mode_config.c > > > > @@ -374,6 +374,13 @@ static int drm_mode_create_standard_properties(struct drm_device *dev) > > > > return -ENOMEM; > > > > dev->mode_config.modifiers_property = prop; > > > > > > > > + prop = drm_property_create(dev, > > > > + DRM_MODE_PROP_IMMUTABLE | DRM_MODE_PROP_BLOB, > > > > + "SIZE_HINTS", 0); > > > > + if (!prop) > > > > + return -ENOMEM; > > > > + dev->mode_config.size_hints_property = prop; > > > > + > > > > return 0; > > > > } > > > > > > > > diff --git a/drivers/gpu/drm/drm_plane.c b/drivers/gpu/drm/drm_plane.c > > > > index 24e7998d1731..ae51b1f83755 100644 > > > > --- a/drivers/gpu/drm/drm_plane.c > > > > +++ b/drivers/gpu/drm/drm_plane.c > > > > @@ -140,6 +140,21 @@ > > > > * DRM_FORMAT_MOD_LINEAR. Before linux kernel release v5.1 there have been > > > > * various bugs in this area with inconsistencies between the capability > > > > * flag and per-plane properties. > > > > + * > > > > + * SIZE_HINTS: > > > > + * Blob property which contains the set of recommended plane size > > > > + * which can used for simple "cursor like" use cases (eg. no scaling). > > > > + * Using these hints frees userspace from extensive probing of > > > > + * supported plane sizes through atomic/setcursor ioctls. > > > > + * > > > > + * For optimal usage userspace should pick the smallest size > > > > + * that satisfies its own requirements. > > > > + * > > > > + * The blob contains an array of struct drm_plane_size_hint. > > > > + * > > > > + * Drivers should only attach this property to planes that > > > > + * support a very limited set of sizes (eg. cursor planes > > > > + * on typical hardware). > > > > > > This is a bit awkward since problematic drivers would only expose > > > this property if they are new enough. > > > > > > A way to avoid this is for all new drivers expose this property, but > > > special case the size 0x0 as "everything below the max limit goes". Then > > > userspace can do (ignoring the missing cap fallback). > > > > I don't think there are any drivers that need that. > > So I'm thinking we can just ignore that for now. > > None the less, userspace not seeing SIZE_HINTS will be required to > indefinitely use the existing "old" behavior using the size cap as the > buffer size with a fallback, and drivers without any size limitations > that, i.e. details that are hard to express with a set of accepted > sizes, will still use the inoptimal buffer sizes. > > If I read [1] correctly, AMD has no particular size limitations other > than a size limit, but without a SIZE_HINTS, userspace still needs to > use the maximum size. Simon pointed out they have pretty much the same exact limits as i915. Ie. only a few power of two sizes, and stride must match width. > > [1] https://gitlab.freedesktop.org/drm/intel/-/issues/7687#note_1760865 > > > Jonas > > > > > > > > > if (has(SIZE_HINTS)) > > > size = figure_out_size(SIZE_HINTS, > > > DRM_CAP_CURSOR_WIDTH, > > > DRM_CAP_CURSOR_HEIGHT, > > > preferred_size); > > > else > > > size = DRM_CAP_CURSOR_WIDTH x DRM_CAP_CURSOR_WIDTH; > > > > > > With `figure_out_size()` knowing how to deal with 0x0 in the size hints > > > to use `preferred_size` directly. > > > > > > > > > Jonas > > > > > > > */ > > > > > > > > static unsigned int drm_num_planes(struct drm_device *dev) > > > > @@ -1582,3 +1597,36 @@ int drm_plane_create_scaling_filter_property(struct drm_plane *plane, > > > > return 0; > > > > } > > > > EXPORT_SYMBOL(drm_plane_create_scaling_filter_property); > > > > + > > > > +/** > > > > + * drm_plane_add_size_hint_property - create a size hint property > > > > + * > > > > + * @plane: drm plane > > > > + * @hints: size hints > > > > + * @num_hints: number of size hints > > > > + * > > > > + * Create a size hints property for the plane. > > > > + * > > > > + * RETURNS: > > > > + * Zero for success or -errno > > > > + */ > > > > +int drm_plane_add_size_hints_property(struct drm_plane *plane, > > > > + const struct drm_plane_size_hint *hints, > > > > + int num_hints) > > > > +{ > > > > + struct drm_device *dev = plane->dev; > > > > + struct drm_mode_config *config = &dev->mode_config; > > > > + struct drm_property_blob *blob; > > > > + > > > > + blob = drm_property_create_blob(dev, > > > > + array_size(sizeof(hints[0]), num_hints), > > > > + hints); > > > > + if (IS_ERR(blob)) > > > > + return PTR_ERR(blob); > > > > + > > > > + drm_object_attach_property(&plane->base, config->size_hints_property, > > > > + blob->base.id); > > > > + > > > > + return 0; > > > > +} > > > > +EXPORT_SYMBOL(drm_plane_add_size_hints_property); > > > > diff --git a/include/drm/drm_mode_config.h b/include/drm/drm_mode_config.h > > > > index e5b053001d22..5bc8aed9b445 100644 > > > > --- a/include/drm/drm_mode_config.h > > > > +++ b/include/drm/drm_mode_config.h > > > > @@ -949,6 +949,11 @@ struct drm_mode_config { > > > > */ > > > > struct drm_property *modifiers_property; > > > > > > > > + /** > > > > + * @size_hints_propertty: Plane SIZE_HINTS property. > > > > + */ > > > > + struct drm_property *size_hints_property; > > > > + > > > > /* cursor size */ > > > > uint32_t cursor_width, cursor_height; > > > > > > > > diff --git a/include/drm/drm_plane.h b/include/drm/drm_plane.h > > > > index 51291983ea44..1997d7d64b69 100644 > > > > --- a/include/drm/drm_plane.h > > > > +++ b/include/drm/drm_plane.h > > > > @@ -32,6 +32,7 @@ > > > > #include <drm/drm_util.h> > > > > > > > > struct drm_crtc; > > > > +struct drm_plane_size_hint; > > > > struct drm_printer; > > > > struct drm_modeset_acquire_ctx; > > > > > > > > @@ -945,5 +946,8 @@ drm_plane_get_damage_clips(const struct drm_plane_state *state); > > > > > > > > int drm_plane_create_scaling_filter_property(struct drm_plane *plane, > > > > unsigned int supported_filters); > > > > +int drm_plane_add_size_hints_property(struct drm_plane *plane, > > > > + const struct drm_plane_size_hint *hints, > > > > + int num_hints); > > > > > > > > #endif > > > > diff --git a/include/uapi/drm/drm_mode.h b/include/uapi/drm/drm_mode.h > > > > index 46becedf5b2f..9d7c5967264f 100644 > > > > --- a/include/uapi/drm/drm_mode.h > > > > +++ b/include/uapi/drm/drm_mode.h > > > > @@ -849,6 +849,17 @@ struct drm_color_lut { > > > > __u16 reserved; > > > > }; > > > > > > > > +/** > > > > + * struct drm_plane_size_hint - Plane size hints > > > > + * > > > > + * The plane SIZE_HINTS property blob contains an > > > > + * array of struct drm_plane_size_hint. > > > > + */ > > > > +struct drm_plane_size_hint { > > > > + __u16 width; > > > > + __u16 height; > > > > +}; > > > > + > > > > /** > > > > * struct hdr_metadata_infoframe - HDR Metadata Infoframe Data. > > > > * > > > > -- > > > > 2.39.1 > > > > > > > > -- > > Ville Syrjälä > > Intel > >
On Tue, Feb 14, 2023 at 12:28:44PM +0200, Ville Syrjälä wrote: > On Tue, Feb 14, 2023 at 10:54:27AM +0100, Jonas Ådahl wrote: > > On Tue, Feb 14, 2023 at 11:25:56AM +0200, Ville Syrjälä wrote: > > > On Thu, Feb 09, 2023 at 03:16:23PM +0100, Jonas Ådahl wrote: > > > > On Wed, Feb 08, 2023 at 11:10:16PM +0200, Ville Syrjala wrote: > > > > > From: Ville Syrjälä <ville.syrjala@linux.intel.com> > > > > > > > > > > Add a new immutable plane property by which a plane can advertise > > > > > a handful of recommended plane sizes. This would be mostly exposed > > > > > by cursor planes as a slightly more capable replacement for > > > > > the DRM_CAP_CURSOR_WIDTH/HEIGHT caps, which can only declare > > > > > a one size fits all limit for the whole device. > > > > > > > > > > Currently eg. amdgpu/i915/nouveau just advertize the max cursor > > > > > size via the cursor size caps. But always using the max sized > > > > > cursor can waste a surprising amount of power, so a better > > > > > stragey is desirable. > > > > > > > > > > Most other drivers don't specify any cursor size at all, in > > > > > which case the ioctl code just claims that 64x64 is a great > > > > > choice. Whether that is actually true is debatable. > > > > > > > > > > A poll of various compositor developers informs us that > > > > > blindly probing with setcursor/atomic ioctl to determine > > > > > suitable cursor sizes is not acceptable, thus the > > > > > introduction of the new property to supplant the cursor > > > > > size caps. The compositor will now be free to select a > > > > > more optimal cursor size from the short list of options. > > > > > > > > > > Note that the reported sizes (either via the property or the > > > > > caps) make no claims about things such as plane scaling. So > > > > > these things should only really be consulted for simple > > > > > "cursor like" use cases. > > > > > > > > > > v2: Try to add some docs > > > > > > > > > > Cc: Simon Ser <contact@emersion.fr> > > > > > Cc: Jonas Ådahl <jadahl@redhat.com> > > > > > Cc: Daniel Stone <daniel@fooishbar.org> > > > > > Cc: Pekka Paalanen <pekka.paalanen@collabora.com> > > > > > Acked-by: Harry Wentland <harry.wentland@amd.com> > > > > > Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> > > > > > --- > > > > > drivers/gpu/drm/drm_mode_config.c | 7 +++++ > > > > > drivers/gpu/drm/drm_plane.c | 48 +++++++++++++++++++++++++++++++ > > > > > include/drm/drm_mode_config.h | 5 ++++ > > > > > include/drm/drm_plane.h | 4 +++ > > > > > include/uapi/drm/drm_mode.h | 11 +++++++ > > > > > 5 files changed, 75 insertions(+) > > > > > > > > > > diff --git a/drivers/gpu/drm/drm_mode_config.c b/drivers/gpu/drm/drm_mode_config.c > > > > > index 87eb591fe9b5..21860f94a18c 100644 > > > > > --- a/drivers/gpu/drm/drm_mode_config.c > > > > > +++ b/drivers/gpu/drm/drm_mode_config.c > > > > > @@ -374,6 +374,13 @@ static int drm_mode_create_standard_properties(struct drm_device *dev) > > > > > return -ENOMEM; > > > > > dev->mode_config.modifiers_property = prop; > > > > > > > > > > + prop = drm_property_create(dev, > > > > > + DRM_MODE_PROP_IMMUTABLE | DRM_MODE_PROP_BLOB, > > > > > + "SIZE_HINTS", 0); > > > > > + if (!prop) > > > > > + return -ENOMEM; > > > > > + dev->mode_config.size_hints_property = prop; > > > > > + > > > > > return 0; > > > > > } > > > > > > > > > > diff --git a/drivers/gpu/drm/drm_plane.c b/drivers/gpu/drm/drm_plane.c > > > > > index 24e7998d1731..ae51b1f83755 100644 > > > > > --- a/drivers/gpu/drm/drm_plane.c > > > > > +++ b/drivers/gpu/drm/drm_plane.c > > > > > @@ -140,6 +140,21 @@ > > > > > * DRM_FORMAT_MOD_LINEAR. Before linux kernel release v5.1 there have been > > > > > * various bugs in this area with inconsistencies between the capability > > > > > * flag and per-plane properties. > > > > > + * > > > > > + * SIZE_HINTS: > > > > > + * Blob property which contains the set of recommended plane size > > > > > + * which can used for simple "cursor like" use cases (eg. no scaling). > > > > > + * Using these hints frees userspace from extensive probing of > > > > > + * supported plane sizes through atomic/setcursor ioctls. > > > > > + * > > > > > + * For optimal usage userspace should pick the smallest size > > > > > + * that satisfies its own requirements. > > > > > + * > > > > > + * The blob contains an array of struct drm_plane_size_hint. > > > > > + * > > > > > + * Drivers should only attach this property to planes that > > > > > + * support a very limited set of sizes (eg. cursor planes > > > > > + * on typical hardware). > > > > > > > > This is a bit awkward since problematic drivers would only expose > > > > this property if they are new enough. > > > > > > > > A way to avoid this is for all new drivers expose this property, but > > > > special case the size 0x0 as "everything below the max limit goes". Then > > > > userspace can do (ignoring the missing cap fallback). > > > > > > I don't think there are any drivers that need that. > > > So I'm thinking we can just ignore that for now. > > > > None the less, userspace not seeing SIZE_HINTS will be required to > > indefinitely use the existing "old" behavior using the size cap as the > > buffer size with a fallback, and drivers without any size limitations > > that, i.e. details that are hard to express with a set of accepted > > sizes, will still use the inoptimal buffer sizes. > > > > If I read [1] correctly, AMD has no particular size limitations other > > than a size limit, but without a SIZE_HINTS, userspace still needs to > > use the maximum size. > > Simon pointed out they have pretty much the same exact limits as i915. > Ie. only a few power of two sizes, and stride must match width. How about various ARM drivers, where the cursor plane is a regular overlay plane with an artificial 'cursor' stamp on it? Either way, the documentation creates an impossible expectation - drivers, existing of future, that does not "support for a very limited set of sizes" but actually any size below a limit, can't communicate to userspace that it can handle cursor buffers with an arbitrary sizes, without userspace breaking on todays kernels. Jonas > > > > > [1] https://gitlab.freedesktop.org/drm/intel/-/issues/7687#note_1760865 > > > > > > Jonas > > > > > > > > > > > > > if (has(SIZE_HINTS)) > > > > size = figure_out_size(SIZE_HINTS, > > > > DRM_CAP_CURSOR_WIDTH, > > > > DRM_CAP_CURSOR_HEIGHT, > > > > preferred_size); > > > > else > > > > size = DRM_CAP_CURSOR_WIDTH x DRM_CAP_CURSOR_WIDTH; > > > > > > > > With `figure_out_size()` knowing how to deal with 0x0 in the size hints > > > > to use `preferred_size` directly. > > > > > > > > > > > > Jonas > > > > > > > > > */ > > > > > > > > > > static unsigned int drm_num_planes(struct drm_device *dev) > > > > > @@ -1582,3 +1597,36 @@ int drm_plane_create_scaling_filter_property(struct drm_plane *plane, > > > > > return 0; > > > > > } > > > > > EXPORT_SYMBOL(drm_plane_create_scaling_filter_property); > > > > > + > > > > > +/** > > > > > + * drm_plane_add_size_hint_property - create a size hint property > > > > > + * > > > > > + * @plane: drm plane > > > > > + * @hints: size hints > > > > > + * @num_hints: number of size hints > > > > > + * > > > > > + * Create a size hints property for the plane. > > > > > + * > > > > > + * RETURNS: > > > > > + * Zero for success or -errno > > > > > + */ > > > > > +int drm_plane_add_size_hints_property(struct drm_plane *plane, > > > > > + const struct drm_plane_size_hint *hints, > > > > > + int num_hints) > > > > > +{ > > > > > + struct drm_device *dev = plane->dev; > > > > > + struct drm_mode_config *config = &dev->mode_config; > > > > > + struct drm_property_blob *blob; > > > > > + > > > > > + blob = drm_property_create_blob(dev, > > > > > + array_size(sizeof(hints[0]), num_hints), > > > > > + hints); > > > > > + if (IS_ERR(blob)) > > > > > + return PTR_ERR(blob); > > > > > + > > > > > + drm_object_attach_property(&plane->base, config->size_hints_property, > > > > > + blob->base.id); > > > > > + > > > > > + return 0; > > > > > +} > > > > > +EXPORT_SYMBOL(drm_plane_add_size_hints_property); > > > > > diff --git a/include/drm/drm_mode_config.h b/include/drm/drm_mode_config.h > > > > > index e5b053001d22..5bc8aed9b445 100644 > > > > > --- a/include/drm/drm_mode_config.h > > > > > +++ b/include/drm/drm_mode_config.h > > > > > @@ -949,6 +949,11 @@ struct drm_mode_config { > > > > > */ > > > > > struct drm_property *modifiers_property; > > > > > > > > > > + /** > > > > > + * @size_hints_propertty: Plane SIZE_HINTS property. > > > > > + */ > > > > > + struct drm_property *size_hints_property; > > > > > + > > > > > /* cursor size */ > > > > > uint32_t cursor_width, cursor_height; > > > > > > > > > > diff --git a/include/drm/drm_plane.h b/include/drm/drm_plane.h > > > > > index 51291983ea44..1997d7d64b69 100644 > > > > > --- a/include/drm/drm_plane.h > > > > > +++ b/include/drm/drm_plane.h > > > > > @@ -32,6 +32,7 @@ > > > > > #include <drm/drm_util.h> > > > > > > > > > > struct drm_crtc; > > > > > +struct drm_plane_size_hint; > > > > > struct drm_printer; > > > > > struct drm_modeset_acquire_ctx; > > > > > > > > > > @@ -945,5 +946,8 @@ drm_plane_get_damage_clips(const struct drm_plane_state *state); > > > > > > > > > > int drm_plane_create_scaling_filter_property(struct drm_plane *plane, > > > > > unsigned int supported_filters); > > > > > +int drm_plane_add_size_hints_property(struct drm_plane *plane, > > > > > + const struct drm_plane_size_hint *hints, > > > > > + int num_hints); > > > > > > > > > > #endif > > > > > diff --git a/include/uapi/drm/drm_mode.h b/include/uapi/drm/drm_mode.h > > > > > index 46becedf5b2f..9d7c5967264f 100644 > > > > > --- a/include/uapi/drm/drm_mode.h > > > > > +++ b/include/uapi/drm/drm_mode.h > > > > > @@ -849,6 +849,17 @@ struct drm_color_lut { > > > > > __u16 reserved; > > > > > }; > > > > > > > > > > +/** > > > > > + * struct drm_plane_size_hint - Plane size hints > > > > > + * > > > > > + * The plane SIZE_HINTS property blob contains an > > > > > + * array of struct drm_plane_size_hint. > > > > > + */ > > > > > +struct drm_plane_size_hint { > > > > > + __u16 width; > > > > > + __u16 height; > > > > > +}; > > > > > + > > > > > /** > > > > > * struct hdr_metadata_infoframe - HDR Metadata Infoframe Data. > > > > > * > > > > > -- > > > > > 2.39.1 > > > > > > > > > > > -- > > > Ville Syrjälä > > > Intel > > > > > -- > Ville Syrjälä > Intel >
On Tue, Feb 14, 2023 at 12:01:49PM +0100, Jonas Ådahl wrote: > On Tue, Feb 14, 2023 at 12:28:44PM +0200, Ville Syrjälä wrote: > > On Tue, Feb 14, 2023 at 10:54:27AM +0100, Jonas Ådahl wrote: > > > On Tue, Feb 14, 2023 at 11:25:56AM +0200, Ville Syrjälä wrote: > > > > On Thu, Feb 09, 2023 at 03:16:23PM +0100, Jonas Ådahl wrote: > > > > > On Wed, Feb 08, 2023 at 11:10:16PM +0200, Ville Syrjala wrote: > > > > > > From: Ville Syrjälä <ville.syrjala@linux.intel.com> > > > > > > > > > > > > Add a new immutable plane property by which a plane can advertise > > > > > > a handful of recommended plane sizes. This would be mostly exposed > > > > > > by cursor planes as a slightly more capable replacement for > > > > > > the DRM_CAP_CURSOR_WIDTH/HEIGHT caps, which can only declare > > > > > > a one size fits all limit for the whole device. > > > > > > > > > > > > Currently eg. amdgpu/i915/nouveau just advertize the max cursor > > > > > > size via the cursor size caps. But always using the max sized > > > > > > cursor can waste a surprising amount of power, so a better > > > > > > stragey is desirable. > > > > > > > > > > > > Most other drivers don't specify any cursor size at all, in > > > > > > which case the ioctl code just claims that 64x64 is a great > > > > > > choice. Whether that is actually true is debatable. > > > > > > > > > > > > A poll of various compositor developers informs us that > > > > > > blindly probing with setcursor/atomic ioctl to determine > > > > > > suitable cursor sizes is not acceptable, thus the > > > > > > introduction of the new property to supplant the cursor > > > > > > size caps. The compositor will now be free to select a > > > > > > more optimal cursor size from the short list of options. > > > > > > > > > > > > Note that the reported sizes (either via the property or the > > > > > > caps) make no claims about things such as plane scaling. So > > > > > > these things should only really be consulted for simple > > > > > > "cursor like" use cases. > > > > > > > > > > > > v2: Try to add some docs > > > > > > > > > > > > Cc: Simon Ser <contact@emersion.fr> > > > > > > Cc: Jonas Ådahl <jadahl@redhat.com> > > > > > > Cc: Daniel Stone <daniel@fooishbar.org> > > > > > > Cc: Pekka Paalanen <pekka.paalanen@collabora.com> > > > > > > Acked-by: Harry Wentland <harry.wentland@amd.com> > > > > > > Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> > > > > > > --- > > > > > > drivers/gpu/drm/drm_mode_config.c | 7 +++++ > > > > > > drivers/gpu/drm/drm_plane.c | 48 +++++++++++++++++++++++++++++++ > > > > > > include/drm/drm_mode_config.h | 5 ++++ > > > > > > include/drm/drm_plane.h | 4 +++ > > > > > > include/uapi/drm/drm_mode.h | 11 +++++++ > > > > > > 5 files changed, 75 insertions(+) > > > > > > > > > > > > diff --git a/drivers/gpu/drm/drm_mode_config.c b/drivers/gpu/drm/drm_mode_config.c > > > > > > index 87eb591fe9b5..21860f94a18c 100644 > > > > > > --- a/drivers/gpu/drm/drm_mode_config.c > > > > > > +++ b/drivers/gpu/drm/drm_mode_config.c > > > > > > @@ -374,6 +374,13 @@ static int drm_mode_create_standard_properties(struct drm_device *dev) > > > > > > return -ENOMEM; > > > > > > dev->mode_config.modifiers_property = prop; > > > > > > > > > > > > + prop = drm_property_create(dev, > > > > > > + DRM_MODE_PROP_IMMUTABLE | DRM_MODE_PROP_BLOB, > > > > > > + "SIZE_HINTS", 0); > > > > > > + if (!prop) > > > > > > + return -ENOMEM; > > > > > > + dev->mode_config.size_hints_property = prop; > > > > > > + > > > > > > return 0; > > > > > > } > > > > > > > > > > > > diff --git a/drivers/gpu/drm/drm_plane.c b/drivers/gpu/drm/drm_plane.c > > > > > > index 24e7998d1731..ae51b1f83755 100644 > > > > > > --- a/drivers/gpu/drm/drm_plane.c > > > > > > +++ b/drivers/gpu/drm/drm_plane.c > > > > > > @@ -140,6 +140,21 @@ > > > > > > * DRM_FORMAT_MOD_LINEAR. Before linux kernel release v5.1 there have been > > > > > > * various bugs in this area with inconsistencies between the capability > > > > > > * flag and per-plane properties. > > > > > > + * > > > > > > + * SIZE_HINTS: > > > > > > + * Blob property which contains the set of recommended plane size > > > > > > + * which can used for simple "cursor like" use cases (eg. no scaling). > > > > > > + * Using these hints frees userspace from extensive probing of > > > > > > + * supported plane sizes through atomic/setcursor ioctls. > > > > > > + * > > > > > > + * For optimal usage userspace should pick the smallest size > > > > > > + * that satisfies its own requirements. > > > > > > + * > > > > > > + * The blob contains an array of struct drm_plane_size_hint. > > > > > > + * > > > > > > + * Drivers should only attach this property to planes that > > > > > > + * support a very limited set of sizes (eg. cursor planes > > > > > > + * on typical hardware). > > > > > > > > > > This is a bit awkward since problematic drivers would only expose > > > > > this property if they are new enough. > > > > > > > > > > A way to avoid this is for all new drivers expose this property, but > > > > > special case the size 0x0 as "everything below the max limit goes". Then > > > > > userspace can do (ignoring the missing cap fallback). > > > > > > > > I don't think there are any drivers that need that. > > > > So I'm thinking we can just ignore that for now. > > > > > > None the less, userspace not seeing SIZE_HINTS will be required to > > > indefinitely use the existing "old" behavior using the size cap as the > > > buffer size with a fallback, and drivers without any size limitations > > > that, i.e. details that are hard to express with a set of accepted > > > sizes, will still use the inoptimal buffer sizes. > > > > > > If I read [1] correctly, AMD has no particular size limitations other > > > than a size limit, but without a SIZE_HINTS, userspace still needs to > > > use the maximum size. > > > > Simon pointed out they have pretty much the same exact limits as i915. > > Ie. only a few power of two sizes, and stride must match width. > > How about various ARM drivers, where the cursor plane is a regular > overlay plane with an artificial 'cursor' stamp on it? They don't even bother with the size cap currently. So the generic ioctl code currently just decides that 64x64 is good enough for them. > > Either way, the documentation creates an impossible expectation - > drivers, existing of future, that does not "support for a very limited > set of sizes" but actually any size below a limit, can't communicate to > userspace that it can handle cursor buffers with an arbitrary sizes, > without userspace breaking on todays kernels. I don't see how anything would break. You just won't get a 100% optimal result potentially if you don't declare any smaller sizes. And I think we can always specify that magic 0x0 value later (or a new cap/etc) should an actual user for it appear. I guess to play it safe we could specify 0x0 as a value that might become valid in the future, so userspace should check for it and treat it the same as not having the prop at all.
On 2/14/23 05:28, Ville Syrjälä wrote: > On Tue, Feb 14, 2023 at 10:54:27AM +0100, Jonas Ådahl wrote: >> On Tue, Feb 14, 2023 at 11:25:56AM +0200, Ville Syrjälä wrote: >>> On Thu, Feb 09, 2023 at 03:16:23PM +0100, Jonas Ådahl wrote: >>>> On Wed, Feb 08, 2023 at 11:10:16PM +0200, Ville Syrjala wrote: >>>>> From: Ville Syrjälä <ville.syrjala@linux.intel.com> >>>>> >>>>> Add a new immutable plane property by which a plane can advertise >>>>> a handful of recommended plane sizes. This would be mostly exposed >>>>> by cursor planes as a slightly more capable replacement for >>>>> the DRM_CAP_CURSOR_WIDTH/HEIGHT caps, which can only declare >>>>> a one size fits all limit for the whole device. >>>>> >>>>> Currently eg. amdgpu/i915/nouveau just advertize the max cursor >>>>> size via the cursor size caps. But always using the max sized >>>>> cursor can waste a surprising amount of power, so a better >>>>> stragey is desirable. >>>>> >>>>> Most other drivers don't specify any cursor size at all, in >>>>> which case the ioctl code just claims that 64x64 is a great >>>>> choice. Whether that is actually true is debatable. >>>>> >>>>> A poll of various compositor developers informs us that >>>>> blindly probing with setcursor/atomic ioctl to determine >>>>> suitable cursor sizes is not acceptable, thus the >>>>> introduction of the new property to supplant the cursor >>>>> size caps. The compositor will now be free to select a >>>>> more optimal cursor size from the short list of options. >>>>> >>>>> Note that the reported sizes (either via the property or the >>>>> caps) make no claims about things such as plane scaling. So >>>>> these things should only really be consulted for simple >>>>> "cursor like" use cases. >>>>> >>>>> v2: Try to add some docs >>>>> >>>>> Cc: Simon Ser <contact@emersion.fr> >>>>> Cc: Jonas Ådahl <jadahl@redhat.com> >>>>> Cc: Daniel Stone <daniel@fooishbar.org> >>>>> Cc: Pekka Paalanen <pekka.paalanen@collabora.com> >>>>> Acked-by: Harry Wentland <harry.wentland@amd.com> >>>>> Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> >>>>> --- >>>>> drivers/gpu/drm/drm_mode_config.c | 7 +++++ >>>>> drivers/gpu/drm/drm_plane.c | 48 +++++++++++++++++++++++++++++++ >>>>> include/drm/drm_mode_config.h | 5 ++++ >>>>> include/drm/drm_plane.h | 4 +++ >>>>> include/uapi/drm/drm_mode.h | 11 +++++++ >>>>> 5 files changed, 75 insertions(+) >>>>> >>>>> diff --git a/drivers/gpu/drm/drm_mode_config.c b/drivers/gpu/drm/drm_mode_config.c >>>>> index 87eb591fe9b5..21860f94a18c 100644 >>>>> --- a/drivers/gpu/drm/drm_mode_config.c >>>>> +++ b/drivers/gpu/drm/drm_mode_config.c >>>>> @@ -374,6 +374,13 @@ static int drm_mode_create_standard_properties(struct drm_device *dev) >>>>> return -ENOMEM; >>>>> dev->mode_config.modifiers_property = prop; >>>>> >>>>> + prop = drm_property_create(dev, >>>>> + DRM_MODE_PROP_IMMUTABLE | DRM_MODE_PROP_BLOB, >>>>> + "SIZE_HINTS", 0); >>>>> + if (!prop) >>>>> + return -ENOMEM; >>>>> + dev->mode_config.size_hints_property = prop; >>>>> + >>>>> return 0; >>>>> } >>>>> >>>>> diff --git a/drivers/gpu/drm/drm_plane.c b/drivers/gpu/drm/drm_plane.c >>>>> index 24e7998d1731..ae51b1f83755 100644 >>>>> --- a/drivers/gpu/drm/drm_plane.c >>>>> +++ b/drivers/gpu/drm/drm_plane.c >>>>> @@ -140,6 +140,21 @@ >>>>> * DRM_FORMAT_MOD_LINEAR. Before linux kernel release v5.1 there have been >>>>> * various bugs in this area with inconsistencies between the capability >>>>> * flag and per-plane properties. >>>>> + * >>>>> + * SIZE_HINTS: >>>>> + * Blob property which contains the set of recommended plane size >>>>> + * which can used for simple "cursor like" use cases (eg. no scaling). >>>>> + * Using these hints frees userspace from extensive probing of >>>>> + * supported plane sizes through atomic/setcursor ioctls. >>>>> + * >>>>> + * For optimal usage userspace should pick the smallest size >>>>> + * that satisfies its own requirements. >>>>> + * >>>>> + * The blob contains an array of struct drm_plane_size_hint. >>>>> + * >>>>> + * Drivers should only attach this property to planes that >>>>> + * support a very limited set of sizes (eg. cursor planes >>>>> + * on typical hardware). >>>> >>>> This is a bit awkward since problematic drivers would only expose >>>> this property if they are new enough. >>>> >>>> A way to avoid this is for all new drivers expose this property, but >>>> special case the size 0x0 as "everything below the max limit goes". Then >>>> userspace can do (ignoring the missing cap fallback). >>> >>> I don't think there are any drivers that need that. >>> So I'm thinking we can just ignore that for now. >> >> None the less, userspace not seeing SIZE_HINTS will be required to >> indefinitely use the existing "old" behavior using the size cap as the >> buffer size with a fallback, and drivers without any size limitations >> that, i.e. details that are hard to express with a set of accepted >> sizes, will still use the inoptimal buffer sizes. >> >> If I read [1] correctly, AMD has no particular size limitations other >> than a size limit, but without a SIZE_HINTS, userspace still needs to >> use the maximum size. > > Simon pointed out they have pretty much the same exact limits as i915. > Ie. only a few power of two sizes, and stride must match width. > That's an artificial limitation in the driver. The HW has no such limitation and it would be nice to drop that from our driver as well. Harry >> >> [1] https://gitlab.freedesktop.org/drm/intel/-/issues/7687#note_1760865 >> >> >> Jonas >> >>> >>>> >>>> if (has(SIZE_HINTS)) >>>> size = figure_out_size(SIZE_HINTS, >>>> DRM_CAP_CURSOR_WIDTH, >>>> DRM_CAP_CURSOR_HEIGHT, >>>> preferred_size); >>>> else >>>> size = DRM_CAP_CURSOR_WIDTH x DRM_CAP_CURSOR_WIDTH; >>>> >>>> With `figure_out_size()` knowing how to deal with 0x0 in the size hints >>>> to use `preferred_size` directly. >>>> >>>> >>>> Jonas >>>> >>>>> */ >>>>> >>>>> static unsigned int drm_num_planes(struct drm_device *dev) >>>>> @@ -1582,3 +1597,36 @@ int drm_plane_create_scaling_filter_property(struct drm_plane *plane, >>>>> return 0; >>>>> } >>>>> EXPORT_SYMBOL(drm_plane_create_scaling_filter_property); >>>>> + >>>>> +/** >>>>> + * drm_plane_add_size_hint_property - create a size hint property >>>>> + * >>>>> + * @plane: drm plane >>>>> + * @hints: size hints >>>>> + * @num_hints: number of size hints >>>>> + * >>>>> + * Create a size hints property for the plane. >>>>> + * >>>>> + * RETURNS: >>>>> + * Zero for success or -errno >>>>> + */ >>>>> +int drm_plane_add_size_hints_property(struct drm_plane *plane, >>>>> + const struct drm_plane_size_hint *hints, >>>>> + int num_hints) >>>>> +{ >>>>> + struct drm_device *dev = plane->dev; >>>>> + struct drm_mode_config *config = &dev->mode_config; >>>>> + struct drm_property_blob *blob; >>>>> + >>>>> + blob = drm_property_create_blob(dev, >>>>> + array_size(sizeof(hints[0]), num_hints), >>>>> + hints); >>>>> + if (IS_ERR(blob)) >>>>> + return PTR_ERR(blob); >>>>> + >>>>> + drm_object_attach_property(&plane->base, config->size_hints_property, >>>>> + blob->base.id); >>>>> + >>>>> + return 0; >>>>> +} >>>>> +EXPORT_SYMBOL(drm_plane_add_size_hints_property); >>>>> diff --git a/include/drm/drm_mode_config.h b/include/drm/drm_mode_config.h >>>>> index e5b053001d22..5bc8aed9b445 100644 >>>>> --- a/include/drm/drm_mode_config.h >>>>> +++ b/include/drm/drm_mode_config.h >>>>> @@ -949,6 +949,11 @@ struct drm_mode_config { >>>>> */ >>>>> struct drm_property *modifiers_property; >>>>> >>>>> + /** >>>>> + * @size_hints_propertty: Plane SIZE_HINTS property. >>>>> + */ >>>>> + struct drm_property *size_hints_property; >>>>> + >>>>> /* cursor size */ >>>>> uint32_t cursor_width, cursor_height; >>>>> >>>>> diff --git a/include/drm/drm_plane.h b/include/drm/drm_plane.h >>>>> index 51291983ea44..1997d7d64b69 100644 >>>>> --- a/include/drm/drm_plane.h >>>>> +++ b/include/drm/drm_plane.h >>>>> @@ -32,6 +32,7 @@ >>>>> #include <drm/drm_util.h> >>>>> >>>>> struct drm_crtc; >>>>> +struct drm_plane_size_hint; >>>>> struct drm_printer; >>>>> struct drm_modeset_acquire_ctx; >>>>> >>>>> @@ -945,5 +946,8 @@ drm_plane_get_damage_clips(const struct drm_plane_state *state); >>>>> >>>>> int drm_plane_create_scaling_filter_property(struct drm_plane *plane, >>>>> unsigned int supported_filters); >>>>> +int drm_plane_add_size_hints_property(struct drm_plane *plane, >>>>> + const struct drm_plane_size_hint *hints, >>>>> + int num_hints); >>>>> >>>>> #endif >>>>> diff --git a/include/uapi/drm/drm_mode.h b/include/uapi/drm/drm_mode.h >>>>> index 46becedf5b2f..9d7c5967264f 100644 >>>>> --- a/include/uapi/drm/drm_mode.h >>>>> +++ b/include/uapi/drm/drm_mode.h >>>>> @@ -849,6 +849,17 @@ struct drm_color_lut { >>>>> __u16 reserved; >>>>> }; >>>>> >>>>> +/** >>>>> + * struct drm_plane_size_hint - Plane size hints >>>>> + * >>>>> + * The plane SIZE_HINTS property blob contains an >>>>> + * array of struct drm_plane_size_hint. >>>>> + */ >>>>> +struct drm_plane_size_hint { >>>>> + __u16 width; >>>>> + __u16 height; >>>>> +}; >>>>> + >>>>> /** >>>>> * struct hdr_metadata_infoframe - HDR Metadata Infoframe Data. >>>>> * >>>>> -- >>>>> 2.39.1 >>>>> >>> >>> -- >>> Ville Syrjälä >>> Intel >>> >
On Tue, Feb 14, 2023 at 02:27:19PM -0500, Harry Wentland wrote: > > > On 2/14/23 05:28, Ville Syrjälä wrote: > > On Tue, Feb 14, 2023 at 10:54:27AM +0100, Jonas Ådahl wrote: > >> On Tue, Feb 14, 2023 at 11:25:56AM +0200, Ville Syrjälä wrote: > >>> On Thu, Feb 09, 2023 at 03:16:23PM +0100, Jonas Ådahl wrote: > >>>> On Wed, Feb 08, 2023 at 11:10:16PM +0200, Ville Syrjala wrote: > >>>>> From: Ville Syrjälä <ville.syrjala@linux.intel.com> > >>>>> > >>>>> Add a new immutable plane property by which a plane can advertise > >>>>> a handful of recommended plane sizes. This would be mostly exposed > >>>>> by cursor planes as a slightly more capable replacement for > >>>>> the DRM_CAP_CURSOR_WIDTH/HEIGHT caps, which can only declare > >>>>> a one size fits all limit for the whole device. > >>>>> > >>>>> Currently eg. amdgpu/i915/nouveau just advertize the max cursor > >>>>> size via the cursor size caps. But always using the max sized > >>>>> cursor can waste a surprising amount of power, so a better > >>>>> stragey is desirable. > >>>>> > >>>>> Most other drivers don't specify any cursor size at all, in > >>>>> which case the ioctl code just claims that 64x64 is a great > >>>>> choice. Whether that is actually true is debatable. > >>>>> > >>>>> A poll of various compositor developers informs us that > >>>>> blindly probing with setcursor/atomic ioctl to determine > >>>>> suitable cursor sizes is not acceptable, thus the > >>>>> introduction of the new property to supplant the cursor > >>>>> size caps. The compositor will now be free to select a > >>>>> more optimal cursor size from the short list of options. > >>>>> > >>>>> Note that the reported sizes (either via the property or the > >>>>> caps) make no claims about things such as plane scaling. So > >>>>> these things should only really be consulted for simple > >>>>> "cursor like" use cases. > >>>>> > >>>>> v2: Try to add some docs > >>>>> > >>>>> Cc: Simon Ser <contact@emersion.fr> > >>>>> Cc: Jonas Ådahl <jadahl@redhat.com> > >>>>> Cc: Daniel Stone <daniel@fooishbar.org> > >>>>> Cc: Pekka Paalanen <pekka.paalanen@collabora.com> > >>>>> Acked-by: Harry Wentland <harry.wentland@amd.com> > >>>>> Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> > >>>>> --- > >>>>> drivers/gpu/drm/drm_mode_config.c | 7 +++++ > >>>>> drivers/gpu/drm/drm_plane.c | 48 +++++++++++++++++++++++++++++++ > >>>>> include/drm/drm_mode_config.h | 5 ++++ > >>>>> include/drm/drm_plane.h | 4 +++ > >>>>> include/uapi/drm/drm_mode.h | 11 +++++++ > >>>>> 5 files changed, 75 insertions(+) > >>>>> > >>>>> diff --git a/drivers/gpu/drm/drm_mode_config.c b/drivers/gpu/drm/drm_mode_config.c > >>>>> index 87eb591fe9b5..21860f94a18c 100644 > >>>>> --- a/drivers/gpu/drm/drm_mode_config.c > >>>>> +++ b/drivers/gpu/drm/drm_mode_config.c > >>>>> @@ -374,6 +374,13 @@ static int drm_mode_create_standard_properties(struct drm_device *dev) > >>>>> return -ENOMEM; > >>>>> dev->mode_config.modifiers_property = prop; > >>>>> > >>>>> + prop = drm_property_create(dev, > >>>>> + DRM_MODE_PROP_IMMUTABLE | DRM_MODE_PROP_BLOB, > >>>>> + "SIZE_HINTS", 0); > >>>>> + if (!prop) > >>>>> + return -ENOMEM; > >>>>> + dev->mode_config.size_hints_property = prop; > >>>>> + > >>>>> return 0; > >>>>> } > >>>>> > >>>>> diff --git a/drivers/gpu/drm/drm_plane.c b/drivers/gpu/drm/drm_plane.c > >>>>> index 24e7998d1731..ae51b1f83755 100644 > >>>>> --- a/drivers/gpu/drm/drm_plane.c > >>>>> +++ b/drivers/gpu/drm/drm_plane.c > >>>>> @@ -140,6 +140,21 @@ > >>>>> * DRM_FORMAT_MOD_LINEAR. Before linux kernel release v5.1 there have been > >>>>> * various bugs in this area with inconsistencies between the capability > >>>>> * flag and per-plane properties. > >>>>> + * > >>>>> + * SIZE_HINTS: > >>>>> + * Blob property which contains the set of recommended plane size > >>>>> + * which can used for simple "cursor like" use cases (eg. no scaling). > >>>>> + * Using these hints frees userspace from extensive probing of > >>>>> + * supported plane sizes through atomic/setcursor ioctls. > >>>>> + * > >>>>> + * For optimal usage userspace should pick the smallest size > >>>>> + * that satisfies its own requirements. > >>>>> + * > >>>>> + * The blob contains an array of struct drm_plane_size_hint. > >>>>> + * > >>>>> + * Drivers should only attach this property to planes that > >>>>> + * support a very limited set of sizes (eg. cursor planes > >>>>> + * on typical hardware). > >>>> > >>>> This is a bit awkward since problematic drivers would only expose > >>>> this property if they are new enough. > >>>> > >>>> A way to avoid this is for all new drivers expose this property, but > >>>> special case the size 0x0 as "everything below the max limit goes". Then > >>>> userspace can do (ignoring the missing cap fallback). > >>> > >>> I don't think there are any drivers that need that. > >>> So I'm thinking we can just ignore that for now. > >> > >> None the less, userspace not seeing SIZE_HINTS will be required to > >> indefinitely use the existing "old" behavior using the size cap as the > >> buffer size with a fallback, and drivers without any size limitations > >> that, i.e. details that are hard to express with a set of accepted > >> sizes, will still use the inoptimal buffer sizes. > >> > >> If I read [1] correctly, AMD has no particular size limitations other > >> than a size limit, but without a SIZE_HINTS, userspace still needs to > >> use the maximum size. > > > > Simon pointed out they have pretty much the same exact limits as i915. > > Ie. only a few power of two sizes, and stride must match width. > > > > That's an artificial limitation in the driver. The HW has no such > limitation and it would be nice to drop that from our driver as > well. OK. Then I guess we might get a driver that wants such a magic 0x0 size hint (or something). Can we expect patches for that soonish? Or should we, for the time being, just mark the 0x0 value as potentially reserved for that? Someone should also look at the igt coverage for funny cursor sizes. Currently I think we just check POT sizes + on i915 a few non-square variants of those with reduced non-POT height (which is what some Intel hw supports).
On Tue, Feb 14, 2023 at 01:19:51PM +0200, Ville Syrjälä wrote: > On Tue, Feb 14, 2023 at 12:01:49PM +0100, Jonas Ådahl wrote: > > On Tue, Feb 14, 2023 at 12:28:44PM +0200, Ville Syrjälä wrote: > > > On Tue, Feb 14, 2023 at 10:54:27AM +0100, Jonas Ådahl wrote: > > > > On Tue, Feb 14, 2023 at 11:25:56AM +0200, Ville Syrjälä wrote: > > > > > On Thu, Feb 09, 2023 at 03:16:23PM +0100, Jonas Ådahl wrote: > > > > > > On Wed, Feb 08, 2023 at 11:10:16PM +0200, Ville Syrjala wrote: > > > > > > > From: Ville Syrjälä <ville.syrjala@linux.intel.com> > > > > > > > > > > > > > > Add a new immutable plane property by which a plane can advertise > > > > > > > a handful of recommended plane sizes. This would be mostly exposed > > > > > > > by cursor planes as a slightly more capable replacement for > > > > > > > the DRM_CAP_CURSOR_WIDTH/HEIGHT caps, which can only declare > > > > > > > a one size fits all limit for the whole device. > > > > > > > > > > > > > > Currently eg. amdgpu/i915/nouveau just advertize the max cursor > > > > > > > size via the cursor size caps. But always using the max sized > > > > > > > cursor can waste a surprising amount of power, so a better > > > > > > > stragey is desirable. > > > > > > > > > > > > > > Most other drivers don't specify any cursor size at all, in > > > > > > > which case the ioctl code just claims that 64x64 is a great > > > > > > > choice. Whether that is actually true is debatable. > > > > > > > > > > > > > > A poll of various compositor developers informs us that > > > > > > > blindly probing with setcursor/atomic ioctl to determine > > > > > > > suitable cursor sizes is not acceptable, thus the > > > > > > > introduction of the new property to supplant the cursor > > > > > > > size caps. The compositor will now be free to select a > > > > > > > more optimal cursor size from the short list of options. > > > > > > > > > > > > > > Note that the reported sizes (either via the property or the > > > > > > > caps) make no claims about things such as plane scaling. So > > > > > > > these things should only really be consulted for simple > > > > > > > "cursor like" use cases. > > > > > > > > > > > > > > v2: Try to add some docs > > > > > > > > > > > > > > Cc: Simon Ser <contact@emersion.fr> > > > > > > > Cc: Jonas Ådahl <jadahl@redhat.com> > > > > > > > Cc: Daniel Stone <daniel@fooishbar.org> > > > > > > > Cc: Pekka Paalanen <pekka.paalanen@collabora.com> > > > > > > > Acked-by: Harry Wentland <harry.wentland@amd.com> > > > > > > > Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> > > > > > > > --- > > > > > > > drivers/gpu/drm/drm_mode_config.c | 7 +++++ > > > > > > > drivers/gpu/drm/drm_plane.c | 48 +++++++++++++++++++++++++++++++ > > > > > > > include/drm/drm_mode_config.h | 5 ++++ > > > > > > > include/drm/drm_plane.h | 4 +++ > > > > > > > include/uapi/drm/drm_mode.h | 11 +++++++ > > > > > > > 5 files changed, 75 insertions(+) > > > > > > > > > > > > > > diff --git a/drivers/gpu/drm/drm_mode_config.c b/drivers/gpu/drm/drm_mode_config.c > > > > > > > index 87eb591fe9b5..21860f94a18c 100644 > > > > > > > --- a/drivers/gpu/drm/drm_mode_config.c > > > > > > > +++ b/drivers/gpu/drm/drm_mode_config.c > > > > > > > @@ -374,6 +374,13 @@ static int drm_mode_create_standard_properties(struct drm_device *dev) > > > > > > > return -ENOMEM; > > > > > > > dev->mode_config.modifiers_property = prop; > > > > > > > > > > > > > > + prop = drm_property_create(dev, > > > > > > > + DRM_MODE_PROP_IMMUTABLE | DRM_MODE_PROP_BLOB, > > > > > > > + "SIZE_HINTS", 0); > > > > > > > + if (!prop) > > > > > > > + return -ENOMEM; > > > > > > > + dev->mode_config.size_hints_property = prop; > > > > > > > + > > > > > > > return 0; > > > > > > > } > > > > > > > > > > > > > > diff --git a/drivers/gpu/drm/drm_plane.c b/drivers/gpu/drm/drm_plane.c > > > > > > > index 24e7998d1731..ae51b1f83755 100644 > > > > > > > --- a/drivers/gpu/drm/drm_plane.c > > > > > > > +++ b/drivers/gpu/drm/drm_plane.c > > > > > > > @@ -140,6 +140,21 @@ > > > > > > > * DRM_FORMAT_MOD_LINEAR. Before linux kernel release v5.1 there have been > > > > > > > * various bugs in this area with inconsistencies between the capability > > > > > > > * flag and per-plane properties. > > > > > > > + * > > > > > > > + * SIZE_HINTS: > > > > > > > + * Blob property which contains the set of recommended plane size > > > > > > > + * which can used for simple "cursor like" use cases (eg. no scaling). > > > > > > > + * Using these hints frees userspace from extensive probing of > > > > > > > + * supported plane sizes through atomic/setcursor ioctls. > > > > > > > + * > > > > > > > + * For optimal usage userspace should pick the smallest size > > > > > > > + * that satisfies its own requirements. > > > > > > > + * > > > > > > > + * The blob contains an array of struct drm_plane_size_hint. > > > > > > > + * > > > > > > > + * Drivers should only attach this property to planes that > > > > > > > + * support a very limited set of sizes (eg. cursor planes > > > > > > > + * on typical hardware). > > > > > > > > > > > > This is a bit awkward since problematic drivers would only expose > > > > > > this property if they are new enough. > > > > > > > > > > > > A way to avoid this is for all new drivers expose this property, but > > > > > > special case the size 0x0 as "everything below the max limit goes". Then > > > > > > userspace can do (ignoring the missing cap fallback). > > > > > > > > > > I don't think there are any drivers that need that. > > > > > So I'm thinking we can just ignore that for now. > > > > > > > > None the less, userspace not seeing SIZE_HINTS will be required to > > > > indefinitely use the existing "old" behavior using the size cap as the > > > > buffer size with a fallback, and drivers without any size limitations > > > > that, i.e. details that are hard to express with a set of accepted > > > > sizes, will still use the inoptimal buffer sizes. > > > > > > > > If I read [1] correctly, AMD has no particular size limitations other > > > > than a size limit, but without a SIZE_HINTS, userspace still needs to > > > > use the maximum size. > > > > > > Simon pointed out they have pretty much the same exact limits as i915. > > > Ie. only a few power of two sizes, and stride must match width. > > > > How about various ARM drivers, where the cursor plane is a regular > > overlay plane with an artificial 'cursor' stamp on it? > > They don't even bother with the size cap currently. So the > generic ioctl code currently just decides that 64x64 is good > enough for them. > > > > > Either way, the documentation creates an impossible expectation - > > drivers, existing of future, that does not "support for a very limited > > set of sizes" but actually any size below a limit, can't communicate to > > userspace that it can handle cursor buffers with an arbitrary sizes, > > without userspace breaking on todays kernels. > > I don't see how anything would break. You just won't get a 100% > optimal result potentially if you don't declare any smaller > sizes. And I think we can always specify that magic 0x0 value > later (or a new cap/etc) should an actual user for it appear. Or maybe better than the 0x0 magic value inside the blob would be to just have the property with value 0 (ie. no blob at all)?
diff --git a/drivers/gpu/drm/drm_mode_config.c b/drivers/gpu/drm/drm_mode_config.c index 87eb591fe9b5..21860f94a18c 100644 --- a/drivers/gpu/drm/drm_mode_config.c +++ b/drivers/gpu/drm/drm_mode_config.c @@ -374,6 +374,13 @@ static int drm_mode_create_standard_properties(struct drm_device *dev) return -ENOMEM; dev->mode_config.modifiers_property = prop; + prop = drm_property_create(dev, + DRM_MODE_PROP_IMMUTABLE | DRM_MODE_PROP_BLOB, + "SIZE_HINTS", 0); + if (!prop) + return -ENOMEM; + dev->mode_config.size_hints_property = prop; + return 0; } diff --git a/drivers/gpu/drm/drm_plane.c b/drivers/gpu/drm/drm_plane.c index 24e7998d1731..ae51b1f83755 100644 --- a/drivers/gpu/drm/drm_plane.c +++ b/drivers/gpu/drm/drm_plane.c @@ -140,6 +140,21 @@ * DRM_FORMAT_MOD_LINEAR. Before linux kernel release v5.1 there have been * various bugs in this area with inconsistencies between the capability * flag and per-plane properties. + * + * SIZE_HINTS: + * Blob property which contains the set of recommended plane size + * which can used for simple "cursor like" use cases (eg. no scaling). + * Using these hints frees userspace from extensive probing of + * supported plane sizes through atomic/setcursor ioctls. + * + * For optimal usage userspace should pick the smallest size + * that satisfies its own requirements. + * + * The blob contains an array of struct drm_plane_size_hint. + * + * Drivers should only attach this property to planes that + * support a very limited set of sizes (eg. cursor planes + * on typical hardware). */ static unsigned int drm_num_planes(struct drm_device *dev) @@ -1582,3 +1597,36 @@ int drm_plane_create_scaling_filter_property(struct drm_plane *plane, return 0; } EXPORT_SYMBOL(drm_plane_create_scaling_filter_property); + +/** + * drm_plane_add_size_hint_property - create a size hint property + * + * @plane: drm plane + * @hints: size hints + * @num_hints: number of size hints + * + * Create a size hints property for the plane. + * + * RETURNS: + * Zero for success or -errno + */ +int drm_plane_add_size_hints_property(struct drm_plane *plane, + const struct drm_plane_size_hint *hints, + int num_hints) +{ + struct drm_device *dev = plane->dev; + struct drm_mode_config *config = &dev->mode_config; + struct drm_property_blob *blob; + + blob = drm_property_create_blob(dev, + array_size(sizeof(hints[0]), num_hints), + hints); + if (IS_ERR(blob)) + return PTR_ERR(blob); + + drm_object_attach_property(&plane->base, config->size_hints_property, + blob->base.id); + + return 0; +} +EXPORT_SYMBOL(drm_plane_add_size_hints_property); diff --git a/include/drm/drm_mode_config.h b/include/drm/drm_mode_config.h index e5b053001d22..5bc8aed9b445 100644 --- a/include/drm/drm_mode_config.h +++ b/include/drm/drm_mode_config.h @@ -949,6 +949,11 @@ struct drm_mode_config { */ struct drm_property *modifiers_property; + /** + * @size_hints_propertty: Plane SIZE_HINTS property. + */ + struct drm_property *size_hints_property; + /* cursor size */ uint32_t cursor_width, cursor_height; diff --git a/include/drm/drm_plane.h b/include/drm/drm_plane.h index 51291983ea44..1997d7d64b69 100644 --- a/include/drm/drm_plane.h +++ b/include/drm/drm_plane.h @@ -32,6 +32,7 @@ #include <drm/drm_util.h> struct drm_crtc; +struct drm_plane_size_hint; struct drm_printer; struct drm_modeset_acquire_ctx; @@ -945,5 +946,8 @@ drm_plane_get_damage_clips(const struct drm_plane_state *state); int drm_plane_create_scaling_filter_property(struct drm_plane *plane, unsigned int supported_filters); +int drm_plane_add_size_hints_property(struct drm_plane *plane, + const struct drm_plane_size_hint *hints, + int num_hints); #endif diff --git a/include/uapi/drm/drm_mode.h b/include/uapi/drm/drm_mode.h index 46becedf5b2f..9d7c5967264f 100644 --- a/include/uapi/drm/drm_mode.h +++ b/include/uapi/drm/drm_mode.h @@ -849,6 +849,17 @@ struct drm_color_lut { __u16 reserved; }; +/** + * struct drm_plane_size_hint - Plane size hints + * + * The plane SIZE_HINTS property blob contains an + * array of struct drm_plane_size_hint. + */ +struct drm_plane_size_hint { + __u16 width; + __u16 height; +}; + /** * struct hdr_metadata_infoframe - HDR Metadata Infoframe Data. *