[RFC wayland-protocols 1/1] unstable: add color management protocol
Sebastian Wick
sebastian at sebastianwick.net
Thu Feb 14 06:43:05 UTC 2019
On 2019-02-14 05:55, Sharma, Shashank via wayland-devel wrote:
> Hello Sebastian,
> My comments inline.
>
> Regards
> Shashank
>> -----Original Message-----
>> From: wayland-devel
>> [mailto:wayland-devel-bounces at lists.freedesktop.org] On
>> Behalf Of Sebastian Wick
>> Sent: Thursday, February 14, 2019 8:16 AM
>> To: wayland-devel at lists.freedesktop.org
>> Cc: Sebastian Wick <sebastian at sebastianwick.net>
>> Subject: [RFC wayland-protocols 1/1] unstable: add color management
>> protocol
>>
>> This protocol allows clients to attach a color space and rendering
>> intent to a
>> wl_surface. It further allows the client to be informed which color
>> spaces a wl_surface
>> was converted to on the last presented.
>>
>> Signed-off-by: Sebastian Wick <sebastian at sebastianwick.net>
>> ---
>> Makefile.am | 1 +
>> unstable/color-management/README | 4 +
>> .../color-management-unstable-v1.xml | 183
>> ++++++++++++++++++
>> 3 files changed, 188 insertions(+)
>> create mode 100644 unstable/color-management/README create mode
>> 100644
>> unstable/color-management/color-management-unstable-v1.xml
>>
>> diff --git a/Makefile.am b/Makefile.am
>> index 345ae6a..80abc1d 100644
>> --- a/Makefile.am
>> +++ b/Makefile.am
>> @@ -23,6 +23,7 @@ unstable_protocols =
>> \
>> unstable/xdg-decoration/xdg-decoration-unstable-v1.xml \
>>
>> unstable/linux-explicit-synchronization/linux-explicit-synchronization-
>> unstable-v1.xml \
>> unstable/primary-selection/primary-selection-unstable-v1.xml \
>> + unstable/color-management/color-management-unstable-v1.xml \
>> $(NULL)
>>
>> stable_protocols = \
>> diff --git a/unstable/color-management/README b/unstable/color-
>> management/README
>> new file mode 100644
>> index 0000000..140f1e9
>> --- /dev/null
>> +++ b/unstable/color-management/README
>> @@ -0,0 +1,4 @@
>> +Color management protocol
>> +
>> +Maintainers:
>> +Sebastian Wick <sebastian at sebastianwick.net>
>> diff --git
>> a/unstable/color-management/color-management-unstable-v1.xml
>> b/unstable/color-management/color-management-unstable-v1.xml
>> new file mode 100644
>> index 0000000..1615fe6
>> --- /dev/null
>> +++ b/unstable/color-management/color-management-unstable-v1.xml
>> @@ -0,0 +1,183 @@
>> +<?xml version="1.0" encoding="UTF-8"?>
>> +<protocol name="color_management_unstable_v1">
>> +
>> + <copyright>
>> + Copyright © 2019 Sebastian Wick
>> + Copyright © 2019 Erwin Burema
>> +
>> + Permission is hereby granted, free of charge, to any person
>> obtaining a
>> + copy of this software and associated documentation files (the
>> "Software"),
>> + to deal in the Software without restriction, including without
>> limitation
>> + the rights to use, copy, modify, merge, publish, distribute,
>> sublicense,
>> + and/or sell copies of the Software, and to permit persons to whom
>> the
>> + Software is furnished to do so, subject to the following
>> conditions:
>> +
>> + The above copyright notice and this permission notice (including
>> the next
>> + paragraph) shall be included in all copies or substantial
>> portions of the
>> + Software.
>> +
>> + THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
>> EXPRESS OR
>> + IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
>> MERCHANTABILITY,
>> + FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO
>> EVENT
>> SHALL
>> + THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES
>> OR OTHER
>> + LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
>> ARISING
>> + FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
>> OTHER
>> + DEALINGS IN THE SOFTWARE.
>> + </copyright>
>> +
>> + <description summary="color management protocol">
>> + This protocol provides the ability to specify the color space
>> + of a wl_surface. If further enumerates the color spaces currently
>> + in the system and allows to query feedback about which color
>> spaces
>> + a wl_surface was converted to on the last present.
>> + The idea behind the feedback system is to allow the client to do
>> color
>> + conversion to a color space which will likely be used to show the
>> surface
>> + which allows the compositor to skip the color conversion step.
>> + </description>
>> +
>> + <interface name="zwp_color_manager_v1" version="1">
>> + <description summary="color manager">
>> + The color manager is a singleton global object that provides
>> access
>> + to outputs' color spaces, let's you create new color spaces and
>> attach
>> + a color space to surfaces.
>> + </description>
>> +
>> + <enum name="render_intent">
>> + <description summary="render intent">
>> + render intents
>> + </description>
>> + <entry name="absolute" value="1"/>
>> + <entry name="relative" value="2"/>
>> + <entry name="perceptual" value="3"/>
>> + <entry name="saturation" value="4" />
> Would you elaborate a bit more on what do we want to achieve with
> render-intent ? How can the compositor utilize this filed during color
> conversion ?
It's about how to handle colors which are outside of the destionation
gamut
https://en.wikipedia.org/wiki/Color_management#Rendering_intent
>> + </enum>
>> +
>> + <enum name="well_known_color_space">
>> + <description summary="well-known color spaces">
> Its slightly difficult to define what is the definition of a
> "well-known" color space, as different use cases may have their own
> favorite. May be we can call it "general-purpose" or "standard" or
> something in those line.
>> + Well-known color spaces
>> + </description>
>> + <entry name="none" value="0" summary="no known color space" />
>> + <entry name="sRGB" value="1" summary="sRGB color space" />
>> + <entry name="adobeRGB" value="2" summary="Adobe RGB" />
>> + <entry name="DCI-3P" value="3" summary="DCI-3P" />
> I guess we mean DCI-P3 here
>> + <entry name="rec2020" value="4" summary="rec2020" />
>> + <entry name="rec2020-pq" value="5" summary="rec2020 with pq
>> curve (HDR)"
>> />
>> + <entry name="rec2020-hlg" value="6" summary="rec2020 with HLG
>> curve
>> (HDR)" />
> Well, this is slightly off the chart for me. If I understood this
> correctly, the cover letter talks about the intent is to focus on
> Colorspace only, but here we are differentiating enum values based on
> their EOTF HDR curve, which is the only difference between these three
> colorspaces. Then why not just mind the colorspace regardless of the
> HDR curve.
A color space consits of three primaries, the white point and the
EOTF/OETF.
Just because two color spaces share the same color gamut doesn't mean
that
they're the same. You're right though that a conversion between those
color
spaces would not require a render intent but since we don't know which
color
spaces we end up having to convert to we need a render intent always.
The point of the enum is to give some ICCs a recognizable name so
clients have
the ability to identify those color spaces without having to poke the
ICC
profile.
>
> On the other hand, we can accommodate the HDR specific information
> also, like the EOTF curve type, and the HDR metadata type and content
> for the frame, if we can add some additional filed in this same
> protocol. So the colorspace information will remain the same, and the
> presence of HDR metadata type / content filed can indicate if the
> current frame is HDR BT2020 frame or SDR BT2020 frame.
This is where my lack of knowledge about HDR standards might become a
problem. My understanding right now is that the HDR standards define a
color
space (which includes EOTF) and then has (potentially dynamic) metadata
which
specifies the luminance.
If my understanding is correct a request which sets the HDR metadata
would be
sufficient. Another request to set a custom tone curve would be highly
useful,
too.
>
> Ankit from my team, has something similar being implemented to address
> the color management and HDR requirements in the same protocol. May be
> you guys can merge your solutions and come-up with something even more
> exiting and usable.
I'd love to hear more about that. Feel free to contact me on IRC
(swick), mail
or just on this list.
>> + </enum>
>> +
>> + <enum name="error">
>> + <description summary="fatal color manager errors">
>> + These fatal protocol errors may be emitted in response to
>> + illegal color management requests.
>> + </description>
>> + <entry name="invalid_icc_profile" value="0" summary="invalid
>> ICC profile"/>
>> + </enum>
>> +
>> + <request name="set_color_space">
>> + <description summary="set the color space of a surface">
>> + Set the color space of a surface. The color space is double
>> buffered,
>> + and will be applied at the time wl_surface.commit of the
>> corresponding
>> + wl_surface is called.
>> +
>> + The render intent gives the compositor a hint what to
>> optimize for
>> + in color space conversions.
>> +
>> + If a surface has no color space set, sRGB and an arbitrary
>> render
>> + intent will be assumed.
>> + </description>
>> + <arg name="surface" type="object" interface="wl_surface"/>
>> + <arg name="color_space" type="object"
>> interface="zwp_color_space_v1"/>
>> + <arg name="render_intent" type="uint" enum="render_intent"/>
>> + </request>
>> +
>> + <request name="color_space_feedback">
>> + <description summary="color space feedback">
>> + Request color space feedback for the current content
>> submission
>> + on the given surface. This creates a new color_space_feedback
>> + object, which will deliver the feedback information once. If
>> + multiple color_space_feedback objects are created for the
>> same
>> + submission, they will all deliver the same information.
>> +
>> + For details on what information is returned, see the
>> + color_space_feedback interface.
>> + </description>
>> + <arg name="surface" type="object" interface="wl_surface"/>
>> + <arg name="feedback" type="new_id"
>> interface="zwp_color_space_feedback_v1"/>
>> + </request>
>> +
>> + <request name="color_space_from_icc">
>> + <description summary="create a color space from an ICC
>> profile">
>> + Create a color space from an ICC profile. Only three channel
>> + profiles are allowed.
>> + </description>
>> + <arg name="color_space" type="new_id"
>> interface="zwp_color_space_v1"/>
>> + <arg name="icc" type="fd"/>
>> + </request>
>> +
>> + <event name="color_space_added">
>> + <description summary="color space added">
>> + Send after binding or when a new color space is added to the
>> system.
>> + </description>
>> + <arg name="color_space" type="new_id"
>> interface="zwp_color_space_v1"/>
>> + </event>
>> + </interface>
>> +
>> + <interface name="zwp_color_space_v1" version="1">
>> + <description summary="a color space">
>> + A color space can be attached to a wl_surface and sends
>> + information like the ICC profile to let the client perform
>> color
>> + correction.
>> + </description>
>> +
>> + <event name="information">
>> + <description summary="color space information">
>> + Information describing the color space.
>> +
>> + The icc argument contains a fd to the corresponding ICC
>> profile.
>> + well-known can be used to easily identify a color-space.
>> +
>> + A color space can be associated with a wl_output.
>> + </description>
>> + <arg name="icc" type="fd"/>
>> + <arg name="well-known" type="uint"
>> enum="well_known_color_space"/>
>> + <arg name="output" type="object" interface="wl_output"
>> allow-null="true"/>
>> + </event>
>> +
>> + <event name="removed">
>> + <description summary="color space removed">
>> + This event is sent when the color space is removed from the
>> system.
>> + When this event is received, the client must destroy the
>> object.
>> + </description>
>> + </event>
>> +
>> + <request name="destroy" type="destructor">
>> + <description summary="destroy the color space object">
>> + Destroy the color_space object.
>> + </description>
>> + </request>
>> + </interface>
>> +
>> + <interface name="zwp_color_space_feedback_v1" version="1">
>> + <description summary="color space feedback">
>> + The surface content was converted to a number of color spaces
>> + on the last content update. They get listed in decreasing order
>> + of importance by the converted event.
>> + Once a color_space_feedback object has delivered a done
>> + event it is automatically destroyed.
>> + </description>
>> +
>> + <event name="converted">
>> + <description summary="color space conversion">
>> + The surface was converted to a color space.
>> + </description>
>> + <arg name="color_space" type="object"
>> interface="zwp_color_space_v1"/>
>> + </event>
>> +
>> + <event name="done">
>> + <description summary="done listing color space conversions">
>> + All color space conversion have been listed.
>> + </description>
>> + </event>
>> + </interface>
>> +
>> +</protocol>
>> --
>> 2.20.1
>>
>> _______________________________________________
>> wayland-devel mailing list
>> wayland-devel at lists.freedesktop.org
>> https://lists.freedesktop.org/mailman/listinfo/wayland-devel
> _______________________________________________
> wayland-devel mailing list
> wayland-devel at lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/wayland-devel
More information about the wayland-devel
mailing list