| Message ID | 20181011163942.28267-4-nicholas.kazlauskas@amd.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | A DRM API for adaptive sync and variable refresh rate support | expand |
On Thu, 11 Oct 2018 12:39:41 -0400 Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> wrote: > These include the drm_connector 'vrr_capable' and the drm_crtc > 'vrr_enabled' properties. > > Signed-off-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> > --- > Documentation/gpu/drm-kms.rst | 7 +++++++ > drivers/gpu/drm/drm_connector.c | 22 ++++++++++++++++++++++ > 2 files changed, 29 insertions(+) > > diff --git a/Documentation/gpu/drm-kms.rst b/Documentation/gpu/drm-kms.rst > index 4b1501b4835b..8da2a178cf85 100644 > --- a/Documentation/gpu/drm-kms.rst > +++ b/Documentation/gpu/drm-kms.rst > @@ -575,6 +575,13 @@ Explicit Fencing Properties > .. kernel-doc:: drivers/gpu/drm/drm_atomic_uapi.c > :doc: explicit fencing properties > > + > +Variable Refresh Properties > +--------------------------- > + > +.. kernel-doc:: drivers/gpu/drm/drm_connector.c > + :doc: Variable refresh properties > + > Existing KMS Properties > ----------------------- > > diff --git a/drivers/gpu/drm/drm_connector.c b/drivers/gpu/drm/drm_connector.c > index f0deeb7298d0..2a12853ca917 100644 > --- a/drivers/gpu/drm/drm_connector.c > +++ b/drivers/gpu/drm/drm_connector.c > @@ -1254,6 +1254,28 @@ int drm_mode_create_scaling_mode_property(struct drm_device *dev) > } > EXPORT_SYMBOL(drm_mode_create_scaling_mode_property); > > +/** > + * DOC: Variable refresh properties > + * > + * Variable refresh rate control is supported via properties on the > + * &drm_connector and &drm_crtc objects. > + * > + * "vrr_capable": > + * Optional &drm_connector boolean property that drivers should attach > + * with drm_connector_attach_vrr_capable_property() on connectors that > + * could support variable refresh rates. Drivers should update the > + * property value by calling drm_connector_set_vrr_capable_property(). > + * > + * Absence of the property should indicate absence of support. > + * > + * "vrr_enabled": > + * Default &drm_crtc boolean property that notifies the driver that the > + * variable refresh rate adjustment should be enabled for the CRTC. Hi, where is the documentation that explains how drivers must implement "variable refresh rate adjustment"? What should and could userspace expect to get if it flicks this switch? I also think the kernel documentation should include a description of what VRR actually is and how it conceptually works as far as userspace is concerned. That is, the kernel documentation should describe what this thing does, so that we avoid every driver implementing a different thing. For example, one driver could prevent the luminance flickering itself by tuning the timings while another driver might not do that, which means that an application tested on the former driver will look just fine while it is unbearable to watch on the latter. Thanks, pq > + * > + * Support for variable refresh rate will depend on the "vrr_capable" > + * property exposed on the &drm_connector object. > + */ > + > /** > * drm_connector_attach_vrr_capable_property - creates the > * vrr_capable property
On 10/26/18 7:37 AM, Pekka Paalanen wrote: > On Thu, 11 Oct 2018 12:39:41 -0400 > Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> wrote: > >> These include the drm_connector 'vrr_capable' and the drm_crtc >> 'vrr_enabled' properties. >> >> Signed-off-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> >> --- >> Documentation/gpu/drm-kms.rst | 7 +++++++ >> drivers/gpu/drm/drm_connector.c | 22 ++++++++++++++++++++++ >> 2 files changed, 29 insertions(+) >> >> diff --git a/Documentation/gpu/drm-kms.rst b/Documentation/gpu/drm-kms.rst >> index 4b1501b4835b..8da2a178cf85 100644 >> --- a/Documentation/gpu/drm-kms.rst >> +++ b/Documentation/gpu/drm-kms.rst >> @@ -575,6 +575,13 @@ Explicit Fencing Properties >> .. kernel-doc:: drivers/gpu/drm/drm_atomic_uapi.c >> :doc: explicit fencing properties >> >> + >> +Variable Refresh Properties >> +--------------------------- >> + >> +.. kernel-doc:: drivers/gpu/drm/drm_connector.c >> + :doc: Variable refresh properties >> + >> Existing KMS Properties >> ----------------------- >> >> diff --git a/drivers/gpu/drm/drm_connector.c b/drivers/gpu/drm/drm_connector.c >> index f0deeb7298d0..2a12853ca917 100644 >> --- a/drivers/gpu/drm/drm_connector.c >> +++ b/drivers/gpu/drm/drm_connector.c >> @@ -1254,6 +1254,28 @@ int drm_mode_create_scaling_mode_property(struct drm_device *dev) >> } >> EXPORT_SYMBOL(drm_mode_create_scaling_mode_property); >> >> +/** >> + * DOC: Variable refresh properties >> + * >> + * Variable refresh rate control is supported via properties on the >> + * &drm_connector and &drm_crtc objects. >> + * >> + * "vrr_capable": >> + * Optional &drm_connector boolean property that drivers should attach >> + * with drm_connector_attach_vrr_capable_property() on connectors that >> + * could support variable refresh rates. Drivers should update the >> + * property value by calling drm_connector_set_vrr_capable_property(). >> + * >> + * Absence of the property should indicate absence of support. >> + * >> + * "vrr_enabled": >> + * Default &drm_crtc boolean property that notifies the driver that the >> + * variable refresh rate adjustment should be enabled for the CRTC. > > Hi, > > where is the documentation that explains how drivers must implement > "variable refresh rate adjustment"? > > What should and could userspace expect to get if it flicks this switch? > > I also think the kernel documentation should include a description of > what VRR actually is and how it conceptually works as far as userspace > is concerned. > > That is, the kernel documentation should describe what this thing does, > so that we avoid every driver implementing a different thing. For > example, one driver could prevent the luminance flickering itself by > tuning the timings while another driver might not do that, which means > that an application tested on the former driver will look just fine > while it is unbearable to watch on the latter. > > > Thanks, > pq I felt it was best to leave this more on the vague side to not impose restrictions yet on what a driver must do. If you think it's worth defining what the "baseline" expectation is for userspace, I would probably describe it as "utilizing the monitor's variable refresh rate to reduce stuttering or tearing that can occur during flipping". This is currently what the amdgpu driver has enabled for support. The implementation also isn't much more complex than just passing the variable refresh range to the hardware. I wouldn't want every driver to be forced to implement some sort of luminance flickering by default. It's not noticeable on many panels and any tuning would inherently add latency to the output. It would probably be better left as a new property or a driver specific feature. In general I would imagine that most future VRR features would end up as new properties. Anything that's purely software could be implemented as a drm helper that every driver can use. I think the target presentation timestamp feature is a good example for that. > >> + * >> + * Support for variable refresh rate will depend on the "vrr_capable" >> + * property exposed on the &drm_connector object. >> + */ >> + >> /** >> * drm_connector_attach_vrr_capable_property - creates the >> * vrr_capable property > Nicholas Kazlauskas
On Fri, Oct 26, 2018 at 02:49:31PM +0000, Kazlauskas, Nicholas wrote: > On 10/26/18 7:37 AM, Pekka Paalanen wrote: > > On Thu, 11 Oct 2018 12:39:41 -0400 > > Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> wrote: > > > >> These include the drm_connector 'vrr_capable' and the drm_crtc > >> 'vrr_enabled' properties. > >> > >> Signed-off-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> > >> --- > >> Documentation/gpu/drm-kms.rst | 7 +++++++ > >> drivers/gpu/drm/drm_connector.c | 22 ++++++++++++++++++++++ > >> 2 files changed, 29 insertions(+) > >> > >> diff --git a/Documentation/gpu/drm-kms.rst b/Documentation/gpu/drm-kms.rst > >> index 4b1501b4835b..8da2a178cf85 100644 > >> --- a/Documentation/gpu/drm-kms.rst > >> +++ b/Documentation/gpu/drm-kms.rst > >> @@ -575,6 +575,13 @@ Explicit Fencing Properties > >> .. kernel-doc:: drivers/gpu/drm/drm_atomic_uapi.c > >> :doc: explicit fencing properties > >> > >> + > >> +Variable Refresh Properties > >> +--------------------------- > >> + > >> +.. kernel-doc:: drivers/gpu/drm/drm_connector.c > >> + :doc: Variable refresh properties > >> + > >> Existing KMS Properties > >> ----------------------- > >> > >> diff --git a/drivers/gpu/drm/drm_connector.c b/drivers/gpu/drm/drm_connector.c > >> index f0deeb7298d0..2a12853ca917 100644 > >> --- a/drivers/gpu/drm/drm_connector.c > >> +++ b/drivers/gpu/drm/drm_connector.c > >> @@ -1254,6 +1254,28 @@ int drm_mode_create_scaling_mode_property(struct drm_device *dev) > >> } > >> EXPORT_SYMBOL(drm_mode_create_scaling_mode_property); > >> > >> +/** > >> + * DOC: Variable refresh properties > >> + * > >> + * Variable refresh rate control is supported via properties on the > >> + * &drm_connector and &drm_crtc objects. > >> + * > >> + * "vrr_capable": > >> + * Optional &drm_connector boolean property that drivers should attach > >> + * with drm_connector_attach_vrr_capable_property() on connectors that > >> + * could support variable refresh rates. Drivers should update the > >> + * property value by calling drm_connector_set_vrr_capable_property(). > >> + * > >> + * Absence of the property should indicate absence of support. > >> + * > >> + * "vrr_enabled": > >> + * Default &drm_crtc boolean property that notifies the driver that the > >> + * variable refresh rate adjustment should be enabled for the CRTC. > > > > Hi, > > > > where is the documentation that explains how drivers must implement > > "variable refresh rate adjustment"? > > > > What should and could userspace expect to get if it flicks this switch? > > > > I also think the kernel documentation should include a description of > > what VRR actually is and how it conceptually works as far as userspace > > is concerned. > > > > That is, the kernel documentation should describe what this thing does, > > so that we avoid every driver implementing a different thing. For > > example, one driver could prevent the luminance flickering itself by > > tuning the timings while another driver might not do that, which means > > that an application tested on the former driver will look just fine > > while it is unbearable to watch on the latter. > > > > > > Thanks, > > pq > > I felt it was best to leave this more on the vague side to not impose > restrictions yet on what a driver must do. > > If you think it's worth defining what the "baseline" expectation is for > userspace, I would probably describe it as "utilizing the monitor's > variable refresh rate to reduce stuttering or tearing that can occur > during flipping". This is currently what the amdgpu driver has enabled > for support. The implementation also isn't much more complex than just > passing the variable refresh range to the hardware. > > I wouldn't want every driver to be forced to implement some sort of > luminance flickering by default. It's not noticeable on many panels and > any tuning would inherently add latency to the output. It would probably > be better left as a new property or a driver specific feature. > > In general I would imagine that most future VRR features would end up as > new properties. Anything that's purely software could be implemented as > a drm helper that every driver can use. I think the target presentation > timestamp feature is a good example for that. Speaking of timestamps. What is the expected behaviour of vblank timestamps when vrr is enabled?
On 10/26/18 10:53 AM, Ville Syrjälä wrote: > On Fri, Oct 26, 2018 at 02:49:31PM +0000, Kazlauskas, Nicholas wrote: >> On 10/26/18 7:37 AM, Pekka Paalanen wrote: >>> On Thu, 11 Oct 2018 12:39:41 -0400 >>> Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> wrote: >>> >>>> These include the drm_connector 'vrr_capable' and the drm_crtc >>>> 'vrr_enabled' properties. >>>> >>>> Signed-off-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> >>>> --- >>>> Documentation/gpu/drm-kms.rst | 7 +++++++ >>>> drivers/gpu/drm/drm_connector.c | 22 ++++++++++++++++++++++ >>>> 2 files changed, 29 insertions(+) >>>> >>>> diff --git a/Documentation/gpu/drm-kms.rst b/Documentation/gpu/drm-kms.rst >>>> index 4b1501b4835b..8da2a178cf85 100644 >>>> --- a/Documentation/gpu/drm-kms.rst >>>> +++ b/Documentation/gpu/drm-kms.rst >>>> @@ -575,6 +575,13 @@ Explicit Fencing Properties >>>> .. kernel-doc:: drivers/gpu/drm/drm_atomic_uapi.c >>>> :doc: explicit fencing properties >>>> >>>> + >>>> +Variable Refresh Properties >>>> +--------------------------- >>>> + >>>> +.. kernel-doc:: drivers/gpu/drm/drm_connector.c >>>> + :doc: Variable refresh properties >>>> + >>>> Existing KMS Properties >>>> ----------------------- >>>> >>>> diff --git a/drivers/gpu/drm/drm_connector.c b/drivers/gpu/drm/drm_connector.c >>>> index f0deeb7298d0..2a12853ca917 100644 >>>> --- a/drivers/gpu/drm/drm_connector.c >>>> +++ b/drivers/gpu/drm/drm_connector.c >>>> @@ -1254,6 +1254,28 @@ int drm_mode_create_scaling_mode_property(struct drm_device *dev) >>>> } >>>> EXPORT_SYMBOL(drm_mode_create_scaling_mode_property); >>>> >>>> +/** >>>> + * DOC: Variable refresh properties >>>> + * >>>> + * Variable refresh rate control is supported via properties on the >>>> + * &drm_connector and &drm_crtc objects. >>>> + * >>>> + * "vrr_capable": >>>> + * Optional &drm_connector boolean property that drivers should attach >>>> + * with drm_connector_attach_vrr_capable_property() on connectors that >>>> + * could support variable refresh rates. Drivers should update the >>>> + * property value by calling drm_connector_set_vrr_capable_property(). >>>> + * >>>> + * Absence of the property should indicate absence of support. >>>> + * >>>> + * "vrr_enabled": >>>> + * Default &drm_crtc boolean property that notifies the driver that the >>>> + * variable refresh rate adjustment should be enabled for the CRTC. >>> >>> Hi, >>> >>> where is the documentation that explains how drivers must implement >>> "variable refresh rate adjustment"? >>> >>> What should and could userspace expect to get if it flicks this switch? >>> >>> I also think the kernel documentation should include a description of >>> what VRR actually is and how it conceptually works as far as userspace >>> is concerned. >>> >>> That is, the kernel documentation should describe what this thing does, >>> so that we avoid every driver implementing a different thing. For >>> example, one driver could prevent the luminance flickering itself by >>> tuning the timings while another driver might not do that, which means >>> that an application tested on the former driver will look just fine >>> while it is unbearable to watch on the latter. >>> >>> >>> Thanks, >>> pq >> >> I felt it was best to leave this more on the vague side to not impose >> restrictions yet on what a driver must do. >> >> If you think it's worth defining what the "baseline" expectation is for >> userspace, I would probably describe it as "utilizing the monitor's >> variable refresh rate to reduce stuttering or tearing that can occur >> during flipping". This is currently what the amdgpu driver has enabled >> for support. The implementation also isn't much more complex than just >> passing the variable refresh range to the hardware. >> >> I wouldn't want every driver to be forced to implement some sort of >> luminance flickering by default. It's not noticeable on many panels and >> any tuning would inherently add latency to the output. It would probably >> be better left as a new property or a driver specific feature. >> >> In general I would imagine that most future VRR features would end up as >> new properties. Anything that's purely software could be implemented as >> a drm helper that every driver can use. I think the target presentation >> timestamp feature is a good example for that. > > Speaking of timestamps. What is the expected behaviour of vblank > timestamps when vrr is enabled? > When vrr is enabled the duration of the vertical front porch will be extended until flip or timeout occurs. The vblank timestamp will vary based on duration of the vertical front porch. The min/max duration for the front porch can be specified by the driver via the min/max range. No changes to vblank timestamping handling should be necessary to accommodate variable refresh rate. I think it's probably best to update the documentation for vrr_enable with some of the specifics I described above. That should help clarify userspace expectations as well. Nicholas Kazlauskas
On Fri, Oct 26, 2018 at 05:34:15PM +0000, Kazlauskas, Nicholas wrote: > On 10/26/18 10:53 AM, Ville Syrjälä wrote: > > On Fri, Oct 26, 2018 at 02:49:31PM +0000, Kazlauskas, Nicholas wrote: > >> On 10/26/18 7:37 AM, Pekka Paalanen wrote: > >>> On Thu, 11 Oct 2018 12:39:41 -0400 > >>> Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> wrote: > >>> > >>>> These include the drm_connector 'vrr_capable' and the drm_crtc > >>>> 'vrr_enabled' properties. > >>>> > >>>> Signed-off-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> > >>>> --- > >>>> Documentation/gpu/drm-kms.rst | 7 +++++++ > >>>> drivers/gpu/drm/drm_connector.c | 22 ++++++++++++++++++++++ > >>>> 2 files changed, 29 insertions(+) > >>>> > >>>> diff --git a/Documentation/gpu/drm-kms.rst b/Documentation/gpu/drm-kms.rst > >>>> index 4b1501b4835b..8da2a178cf85 100644 > >>>> --- a/Documentation/gpu/drm-kms.rst > >>>> +++ b/Documentation/gpu/drm-kms.rst > >>>> @@ -575,6 +575,13 @@ Explicit Fencing Properties > >>>> .. kernel-doc:: drivers/gpu/drm/drm_atomic_uapi.c > >>>> :doc: explicit fencing properties > >>>> > >>>> + > >>>> +Variable Refresh Properties > >>>> +--------------------------- > >>>> + > >>>> +.. kernel-doc:: drivers/gpu/drm/drm_connector.c > >>>> + :doc: Variable refresh properties > >>>> + > >>>> Existing KMS Properties > >>>> ----------------------- > >>>> > >>>> diff --git a/drivers/gpu/drm/drm_connector.c b/drivers/gpu/drm/drm_connector.c > >>>> index f0deeb7298d0..2a12853ca917 100644 > >>>> --- a/drivers/gpu/drm/drm_connector.c > >>>> +++ b/drivers/gpu/drm/drm_connector.c > >>>> @@ -1254,6 +1254,28 @@ int drm_mode_create_scaling_mode_property(struct drm_device *dev) > >>>> } > >>>> EXPORT_SYMBOL(drm_mode_create_scaling_mode_property); > >>>> > >>>> +/** > >>>> + * DOC: Variable refresh properties > >>>> + * > >>>> + * Variable refresh rate control is supported via properties on the > >>>> + * &drm_connector and &drm_crtc objects. > >>>> + * > >>>> + * "vrr_capable": > >>>> + * Optional &drm_connector boolean property that drivers should attach > >>>> + * with drm_connector_attach_vrr_capable_property() on connectors that > >>>> + * could support variable refresh rates. Drivers should update the > >>>> + * property value by calling drm_connector_set_vrr_capable_property(). > >>>> + * > >>>> + * Absence of the property should indicate absence of support. > >>>> + * > >>>> + * "vrr_enabled": > >>>> + * Default &drm_crtc boolean property that notifies the driver that the > >>>> + * variable refresh rate adjustment should be enabled for the CRTC. > >>> > >>> Hi, > >>> > >>> where is the documentation that explains how drivers must implement > >>> "variable refresh rate adjustment"? > >>> > >>> What should and could userspace expect to get if it flicks this switch? > >>> > >>> I also think the kernel documentation should include a description of > >>> what VRR actually is and how it conceptually works as far as userspace > >>> is concerned. > >>> > >>> That is, the kernel documentation should describe what this thing does, > >>> so that we avoid every driver implementing a different thing. For > >>> example, one driver could prevent the luminance flickering itself by > >>> tuning the timings while another driver might not do that, which means > >>> that an application tested on the former driver will look just fine > >>> while it is unbearable to watch on the latter. > >>> > >>> > >>> Thanks, > >>> pq > >> > >> I felt it was best to leave this more on the vague side to not impose > >> restrictions yet on what a driver must do. > >> > >> If you think it's worth defining what the "baseline" expectation is for > >> userspace, I would probably describe it as "utilizing the monitor's > >> variable refresh rate to reduce stuttering or tearing that can occur > >> during flipping". This is currently what the amdgpu driver has enabled > >> for support. The implementation also isn't much more complex than just > >> passing the variable refresh range to the hardware. > >> > >> I wouldn't want every driver to be forced to implement some sort of > >> luminance flickering by default. It's not noticeable on many panels and > >> any tuning would inherently add latency to the output. It would probably > >> be better left as a new property or a driver specific feature. > >> > >> In general I would imagine that most future VRR features would end up as > >> new properties. Anything that's purely software could be implemented as > >> a drm helper that every driver can use. I think the target presentation > >> timestamp feature is a good example for that. > > > > Speaking of timestamps. What is the expected behaviour of vblank > > timestamps when vrr is enabled? > > > > When vrr is enabled the duration of the vertical front porch will be > extended until flip or timeout occurs. The vblank timestamp will vary > based on duration of the vertical front porch. The min/max duration for > the front porch can be specified by the driver via the min/max range. > > No changes to vblank timestamping handling should be necessary to > accommodate variable refresh rate. The problem is that the timestamp is supposed to correspond to the first active pixel. And since we don't know how long the front porch will be we can't realistically report the true value. So I guess just assuming min front porch length is as good as anything else? > > I think it's probably best to update the documentation for vrr_enable > with some of the specifics I described above. That should help clarify > userspace expectations as well. > > Nicholas Kazlauskas
On Fri, Oct 26, 2018 at 05:34:15PM +0000, Kazlauskas, Nicholas wrote: > On 10/26/18 10:53 AM, Ville Syrjälä wrote: > > On Fri, Oct 26, 2018 at 02:49:31PM +0000, Kazlauskas, Nicholas wrote: > >> On 10/26/18 7:37 AM, Pekka Paalanen wrote: > >>> On Thu, 11 Oct 2018 12:39:41 -0400 > >>> Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> wrote: > >>> > >>>> These include the drm_connector 'vrr_capable' and the drm_crtc > >>>> 'vrr_enabled' properties. > >>>> > >>>> Signed-off-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> > >>>> --- > >>>> Documentation/gpu/drm-kms.rst | 7 +++++++ > >>>> drivers/gpu/drm/drm_connector.c | 22 ++++++++++++++++++++++ > >>>> 2 files changed, 29 insertions(+) > >>>> > >>>> diff --git a/Documentation/gpu/drm-kms.rst b/Documentation/gpu/drm-kms.rst > >>>> index 4b1501b4835b..8da2a178cf85 100644 > >>>> --- a/Documentation/gpu/drm-kms.rst > >>>> +++ b/Documentation/gpu/drm-kms.rst > >>>> @@ -575,6 +575,13 @@ Explicit Fencing Properties > >>>> .. kernel-doc:: drivers/gpu/drm/drm_atomic_uapi.c > >>>> :doc: explicit fencing properties > >>>> > >>>> + > >>>> +Variable Refresh Properties > >>>> +--------------------------- > >>>> + > >>>> +.. kernel-doc:: drivers/gpu/drm/drm_connector.c > >>>> + :doc: Variable refresh properties > >>>> + > >>>> Existing KMS Properties > >>>> ----------------------- > >>>> > >>>> diff --git a/drivers/gpu/drm/drm_connector.c b/drivers/gpu/drm/drm_connector.c > >>>> index f0deeb7298d0..2a12853ca917 100644 > >>>> --- a/drivers/gpu/drm/drm_connector.c > >>>> +++ b/drivers/gpu/drm/drm_connector.c > >>>> @@ -1254,6 +1254,28 @@ int drm_mode_create_scaling_mode_property(struct drm_device *dev) > >>>> } > >>>> EXPORT_SYMBOL(drm_mode_create_scaling_mode_property); > >>>> > >>>> +/** > >>>> + * DOC: Variable refresh properties > >>>> + * > >>>> + * Variable refresh rate control is supported via properties on the > >>>> + * &drm_connector and &drm_crtc objects. > >>>> + * > >>>> + * "vrr_capable": > >>>> + * Optional &drm_connector boolean property that drivers should attach > >>>> + * with drm_connector_attach_vrr_capable_property() on connectors that > >>>> + * could support variable refresh rates. Drivers should update the > >>>> + * property value by calling drm_connector_set_vrr_capable_property(). > >>>> + * > >>>> + * Absence of the property should indicate absence of support. > >>>> + * > >>>> + * "vrr_enabled": > >>>> + * Default &drm_crtc boolean property that notifies the driver that the > >>>> + * variable refresh rate adjustment should be enabled for the CRTC. > >>> > >>> Hi, > >>> > >>> where is the documentation that explains how drivers must implement > >>> "variable refresh rate adjustment"? > >>> > >>> What should and could userspace expect to get if it flicks this switch? > >>> > >>> I also think the kernel documentation should include a description of > >>> what VRR actually is and how it conceptually works as far as userspace > >>> is concerned. > >>> > >>> That is, the kernel documentation should describe what this thing does, > >>> so that we avoid every driver implementing a different thing. For > >>> example, one driver could prevent the luminance flickering itself by > >>> tuning the timings while another driver might not do that, which means > >>> that an application tested on the former driver will look just fine > >>> while it is unbearable to watch on the latter. > >>> > >>> > >>> Thanks, > >>> pq > >> > >> I felt it was best to leave this more on the vague side to not impose > >> restrictions yet on what a driver must do. > >> > >> If you think it's worth defining what the "baseline" expectation is for > >> userspace, I would probably describe it as "utilizing the monitor's > >> variable refresh rate to reduce stuttering or tearing that can occur > >> during flipping". This is currently what the amdgpu driver has enabled I would also mention that without VRR, the display engine processes the flips independently of the rendering speed which might result into stuttering or tearing. Might also be worth giving a quick example with numbers in the documentation. Something like: For Eg if the rendering speed is 40Hz, without VRR, display engine will flip at constant 60Hz, which means that same frame will be displayed twice which will be observed as stutter/repetition. With VRR enabled, the vertical front porch will be extended and the flip will be processed when its ready or at max blanking time resulting a smooth transition without repetition. > >> for support. The implementation also isn't much more complex than just > >> passing the variable refresh range to the hardware. > >> > >> I wouldn't want every driver to be forced to implement some sort of > >> luminance flickering by default. It's not noticeable on many panels and > >> any tuning would inherently add latency to the output. It would probably > >> be better left as a new property or a driver specific feature. > >> > >> In general I would imagine that most future VRR features would end up as > >> new properties. Anything that's purely software could be implemented as > >> a drm helper that every driver can use. I think the target presentation > >> timestamp feature is a good example for that. > > > > Speaking of timestamps. What is the expected behaviour of vblank > > timestamps when vrr is enabled? > > > > When vrr is enabled the duration of the vertical front porch will be > extended until flip or timeout occurs. The vblank timestamp will vary > based on duration of the vertical front porch. The min/max duration for > the front porch can be specified by the driver via the min/max range. This min max range that is read from monitor descriptor is only used to program the HW registers right? Its not exposed to the userspace, correct? Manasi > > No changes to vblank timestamping handling should be necessary to > accommodate variable refresh rate. > > I think it's probably best to update the documentation for vrr_enable > with some of the specifics I described above. That should help clarify > userspace expectations as well. > > Nicholas Kazlauskas
On 10/26/18 3:13 PM, Manasi Navare wrote: > On Fri, Oct 26, 2018 at 05:34:15PM +0000, Kazlauskas, Nicholas wrote: >> On 10/26/18 10:53 AM, Ville Syrjälä wrote: >>> On Fri, Oct 26, 2018 at 02:49:31PM +0000, Kazlauskas, Nicholas wrote: >>>> On 10/26/18 7:37 AM, Pekka Paalanen wrote: >>>>> On Thu, 11 Oct 2018 12:39:41 -0400 >>>>> Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> wrote: >>>>> >>>>>> These include the drm_connector 'vrr_capable' and the drm_crtc >>>>>> 'vrr_enabled' properties. >>>>>> >>>>>> Signed-off-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> >>>>>> --- >>>>>> Documentation/gpu/drm-kms.rst | 7 +++++++ >>>>>> drivers/gpu/drm/drm_connector.c | 22 ++++++++++++++++++++++ >>>>>> 2 files changed, 29 insertions(+) >>>>>> >>>>>> diff --git a/Documentation/gpu/drm-kms.rst b/Documentation/gpu/drm-kms.rst >>>>>> index 4b1501b4835b..8da2a178cf85 100644 >>>>>> --- a/Documentation/gpu/drm-kms.rst >>>>>> +++ b/Documentation/gpu/drm-kms.rst >>>>>> @@ -575,6 +575,13 @@ Explicit Fencing Properties >>>>>> .. kernel-doc:: drivers/gpu/drm/drm_atomic_uapi.c >>>>>> :doc: explicit fencing properties >>>>>> >>>>>> + >>>>>> +Variable Refresh Properties >>>>>> +--------------------------- >>>>>> + >>>>>> +.. kernel-doc:: drivers/gpu/drm/drm_connector.c >>>>>> + :doc: Variable refresh properties >>>>>> + >>>>>> Existing KMS Properties >>>>>> ----------------------- >>>>>> >>>>>> diff --git a/drivers/gpu/drm/drm_connector.c b/drivers/gpu/drm/drm_connector.c >>>>>> index f0deeb7298d0..2a12853ca917 100644 >>>>>> --- a/drivers/gpu/drm/drm_connector.c >>>>>> +++ b/drivers/gpu/drm/drm_connector.c >>>>>> @@ -1254,6 +1254,28 @@ int drm_mode_create_scaling_mode_property(struct drm_device *dev) >>>>>> } >>>>>> EXPORT_SYMBOL(drm_mode_create_scaling_mode_property); >>>>>> >>>>>> +/** >>>>>> + * DOC: Variable refresh properties >>>>>> + * >>>>>> + * Variable refresh rate control is supported via properties on the >>>>>> + * &drm_connector and &drm_crtc objects. >>>>>> + * >>>>>> + * "vrr_capable": >>>>>> + * Optional &drm_connector boolean property that drivers should attach >>>>>> + * with drm_connector_attach_vrr_capable_property() on connectors that >>>>>> + * could support variable refresh rates. Drivers should update the >>>>>> + * property value by calling drm_connector_set_vrr_capable_property(). >>>>>> + * >>>>>> + * Absence of the property should indicate absence of support. >>>>>> + * >>>>>> + * "vrr_enabled": >>>>>> + * Default &drm_crtc boolean property that notifies the driver that the >>>>>> + * variable refresh rate adjustment should be enabled for the CRTC. >>>>> >>>>> Hi, >>>>> >>>>> where is the documentation that explains how drivers must implement >>>>> "variable refresh rate adjustment"? >>>>> >>>>> What should and could userspace expect to get if it flicks this switch? >>>>> >>>>> I also think the kernel documentation should include a description of >>>>> what VRR actually is and how it conceptually works as far as userspace >>>>> is concerned. >>>>> >>>>> That is, the kernel documentation should describe what this thing does, >>>>> so that we avoid every driver implementing a different thing. For >>>>> example, one driver could prevent the luminance flickering itself by >>>>> tuning the timings while another driver might not do that, which means >>>>> that an application tested on the former driver will look just fine >>>>> while it is unbearable to watch on the latter. >>>>> >>>>> >>>>> Thanks, >>>>> pq >>>> >>>> I felt it was best to leave this more on the vague side to not impose >>>> restrictions yet on what a driver must do. >>>> >>>> If you think it's worth defining what the "baseline" expectation is for >>>> userspace, I would probably describe it as "utilizing the monitor's >>>> variable refresh rate to reduce stuttering or tearing that can occur >>>> during flipping". This is currently what the amdgpu driver has enabled > > I would also mention that without VRR, the display engine processes the flips > independently of the rendering speed which might result into stuttering or tearing. > > Might also be worth giving a quick example with numbers in the documentation. > Something like: For Eg if the rendering speed is 40Hz, without VRR, display engine > will flip at constant 60Hz, which means that same frame will be displayed twice which > will be observed as stutter/repetition. > With VRR enabled, the vertical front porch will be extended and the flip will > be processed when its ready or at max blanking time resulting a smooth transition without repetition. Good points about mentioning the problems it solves in the documentation. > >>>> for support. The implementation also isn't much more complex than just >>>> passing the variable refresh range to the hardware. >>>> >>>> I wouldn't want every driver to be forced to implement some sort of >>>> luminance flickering by default. It's not noticeable on many panels and >>>> any tuning would inherently add latency to the output. It would probably >>>> be better left as a new property or a driver specific feature. >>>> >>>> In general I would imagine that most future VRR features would end up as >>>> new properties. Anything that's purely software could be implemented as >>>> a drm helper that every driver can use. I think the target presentation >>>> timestamp feature is a good example for that. >>> >>> Speaking of timestamps. What is the expected behaviour of vblank >>> timestamps when vrr is enabled? >>> >> >> When vrr is enabled the duration of the vertical front porch will be >> extended until flip or timeout occurs. The vblank timestamp will vary >> based on duration of the vertical front porch. The min/max duration for >> the front porch can be specified by the driver via the min/max range. > > This min max range that is read from monitor descriptor is only used to program > the HW registers right? Its not exposed to the userspace, correct? > > Manasi Currently the range isn't exposed to userspace. I think I wouldn't mind having them for testing purposes but that can just be done via debugfs. They might make more sense as properties but I don't have any integration patches to utilize them in userspace. > >> >> No changes to vblank timestamping handling should be necessary to >> accommodate variable refresh rate. >> >> I think it's probably best to update the documentation for vrr_enable >> with some of the specifics I described above. That should help clarify >> userspace expectations as well. >> >> Nicholas Kazlauskas Nicholas Kazlauskas
On Fri, Oct 26, 2018 at 08:06:11PM +0000, Kazlauskas, Nicholas wrote: > On 10/26/18 3:13 PM, Manasi Navare wrote: > > On Fri, Oct 26, 2018 at 05:34:15PM +0000, Kazlauskas, Nicholas wrote: > >> On 10/26/18 10:53 AM, Ville Syrjälä wrote: > >>> On Fri, Oct 26, 2018 at 02:49:31PM +0000, Kazlauskas, Nicholas wrote: > >>>> On 10/26/18 7:37 AM, Pekka Paalanen wrote: > >>>>> On Thu, 11 Oct 2018 12:39:41 -0400 > >>>>> Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> wrote: > >>>>> > >>>>>> These include the drm_connector 'vrr_capable' and the drm_crtc > >>>>>> 'vrr_enabled' properties. > >>>>>> > >>>>>> Signed-off-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> > >>>>>> --- > >>>>>> Documentation/gpu/drm-kms.rst | 7 +++++++ > >>>>>> drivers/gpu/drm/drm_connector.c | 22 ++++++++++++++++++++++ > >>>>>> 2 files changed, 29 insertions(+) > >>>>>> > >>>>>> diff --git a/Documentation/gpu/drm-kms.rst b/Documentation/gpu/drm-kms.rst > >>>>>> index 4b1501b4835b..8da2a178cf85 100644 > >>>>>> --- a/Documentation/gpu/drm-kms.rst > >>>>>> +++ b/Documentation/gpu/drm-kms.rst > >>>>>> @@ -575,6 +575,13 @@ Explicit Fencing Properties > >>>>>> .. kernel-doc:: drivers/gpu/drm/drm_atomic_uapi.c > >>>>>> :doc: explicit fencing properties > >>>>>> > >>>>>> + > >>>>>> +Variable Refresh Properties > >>>>>> +--------------------------- > >>>>>> + > >>>>>> +.. kernel-doc:: drivers/gpu/drm/drm_connector.c > >>>>>> + :doc: Variable refresh properties > >>>>>> + > >>>>>> Existing KMS Properties > >>>>>> ----------------------- > >>>>>> > >>>>>> diff --git a/drivers/gpu/drm/drm_connector.c b/drivers/gpu/drm/drm_connector.c > >>>>>> index f0deeb7298d0..2a12853ca917 100644 > >>>>>> --- a/drivers/gpu/drm/drm_connector.c > >>>>>> +++ b/drivers/gpu/drm/drm_connector.c > >>>>>> @@ -1254,6 +1254,28 @@ int drm_mode_create_scaling_mode_property(struct drm_device *dev) > >>>>>> } > >>>>>> EXPORT_SYMBOL(drm_mode_create_scaling_mode_property); > >>>>>> > >>>>>> +/** > >>>>>> + * DOC: Variable refresh properties > >>>>>> + * > >>>>>> + * Variable refresh rate control is supported via properties on the > >>>>>> + * &drm_connector and &drm_crtc objects. > >>>>>> + * > >>>>>> + * "vrr_capable": > >>>>>> + * Optional &drm_connector boolean property that drivers should attach > >>>>>> + * with drm_connector_attach_vrr_capable_property() on connectors that > >>>>>> + * could support variable refresh rates. Drivers should update the > >>>>>> + * property value by calling drm_connector_set_vrr_capable_property(). > >>>>>> + * > >>>>>> + * Absence of the property should indicate absence of support. > >>>>>> + * > >>>>>> + * "vrr_enabled": > >>>>>> + * Default &drm_crtc boolean property that notifies the driver that the > >>>>>> + * variable refresh rate adjustment should be enabled for the CRTC. > >>>>> > >>>>> Hi, > >>>>> > >>>>> where is the documentation that explains how drivers must implement > >>>>> "variable refresh rate adjustment"? > >>>>> > >>>>> What should and could userspace expect to get if it flicks this switch? > >>>>> > >>>>> I also think the kernel documentation should include a description of > >>>>> what VRR actually is and how it conceptually works as far as userspace > >>>>> is concerned. > >>>>> > >>>>> That is, the kernel documentation should describe what this thing does, > >>>>> so that we avoid every driver implementing a different thing. For > >>>>> example, one driver could prevent the luminance flickering itself by > >>>>> tuning the timings while another driver might not do that, which means > >>>>> that an application tested on the former driver will look just fine > >>>>> while it is unbearable to watch on the latter. > >>>>> > >>>>> > >>>>> Thanks, > >>>>> pq > >>>> > >>>> I felt it was best to leave this more on the vague side to not impose > >>>> restrictions yet on what a driver must do. > >>>> > >>>> If you think it's worth defining what the "baseline" expectation is for > >>>> userspace, I would probably describe it as "utilizing the monitor's > >>>> variable refresh rate to reduce stuttering or tearing that can occur > >>>> during flipping". This is currently what the amdgpu driver has enabled > > > > I would also mention that without VRR, the display engine processes the flips > > independently of the rendering speed which might result into stuttering or tearing. > > > > Might also be worth giving a quick example with numbers in the documentation. > > Something like: For Eg if the rendering speed is 40Hz, without VRR, display engine > > will flip at constant 60Hz, which means that same frame will be displayed twice which > > will be observed as stutter/repetition. > > With VRR enabled, the vertical front porch will be extended and the flip will > > be processed when its ready or at max blanking time resulting a smooth transition without repetition. > > Good points about mentioning the problems it solves in the documentation. > > > > >>>> for support. The implementation also isn't much more complex than just > >>>> passing the variable refresh range to the hardware. > >>>> > >>>> I wouldn't want every driver to be forced to implement some sort of > >>>> luminance flickering by default. It's not noticeable on many panels and > >>>> any tuning would inherently add latency to the output. It would probably > >>>> be better left as a new property or a driver specific feature. > >>>> > >>>> In general I would imagine that most future VRR features would end up as > >>>> new properties. Anything that's purely software could be implemented as > >>>> a drm helper that every driver can use. I think the target presentation > >>>> timestamp feature is a good example for that. > >>> > >>> Speaking of timestamps. What is the expected behaviour of vblank > >>> timestamps when vrr is enabled? > >>> > >> > >> When vrr is enabled the duration of the vertical front porch will be > >> extended until flip or timeout occurs. The vblank timestamp will vary > >> based on duration of the vertical front porch. The min/max duration for > >> the front porch can be specified by the driver via the min/max range. > > > > This min max range that is read from monitor descriptor is only used to program > > the HW registers right? Its not exposed to the userspace, correct? > > > > Manasi > > Currently the range isn't exposed to userspace. > > I think I wouldn't mind having them for testing purposes but that can > just be done via debugfs. They might make more sense as properties but I > don't have any integration patches to utilize them in userspace. > > > > >> > >> No changes to vblank timestamping handling should be necessary to > >> accommodate variable refresh rate. > >> > >> I think it's probably best to update the documentation for vrr_enable > >> with some of the specifics I described above. That should help clarify > >> userspace expectations as well. > >> > >> Nicholas Kazlauskas Yes debugfs for min, max Vrefresh range will be good. Manasi > > Nicholas Kazlauskas
On Fri, 26 Oct 2018 17:34:15 +0000 "Kazlauskas, Nicholas" <Nicholas.Kazlauskas@amd.com> wrote: > On 10/26/18 10:53 AM, Ville Syrjälä wrote: > > On Fri, Oct 26, 2018 at 02:49:31PM +0000, Kazlauskas, Nicholas wrote: > >> On 10/26/18 7:37 AM, Pekka Paalanen wrote: > >>> Hi, > >>> > >>> where is the documentation that explains how drivers must implement > >>> "variable refresh rate adjustment"? > >>> > >>> What should and could userspace expect to get if it flicks this switch? > >>> > >>> I also think the kernel documentation should include a description of > >>> what VRR actually is and how it conceptually works as far as userspace > >>> is concerned. > >>> > >>> That is, the kernel documentation should describe what this thing does, > >>> so that we avoid every driver implementing a different thing. For > >>> example, one driver could prevent the luminance flickering itself by > >>> tuning the timings while another driver might not do that, which means > >>> that an application tested on the former driver will look just fine > >>> while it is unbearable to watch on the latter. > >>> > >>> > >>> Thanks, > >>> pq > >> > >> I felt it was best to leave this more on the vague side to not impose > >> restrictions yet on what a driver must do. > >> > >> If you think it's worth defining what the "baseline" expectation is for > >> userspace, I would probably describe it as "utilizing the monitor's > >> variable refresh rate to reduce stuttering or tearing that can occur > >> during flipping". This is currently what the amdgpu driver has enabled > >> for support. The implementation also isn't much more complex than just > >> passing the variable refresh range to the hardware. Hi, sorry, but that says nothing. Also, tearing has nothing to do here. VRR reduces stuttering if userspace is already tear-free. If userspace was driving the display in a tearing fashion, VRR will not reduce or remove tearing, it just makes it happen at different points and times. Tearing happens because framebuffer flips are executed regardless of the refresh cycle, so adjusting the refresh timings won't help. However... > >> > >> I wouldn't want every driver to be forced to implement some sort of > >> luminance flickering by default. It's not noticeable on many panels and > >> any tuning would inherently add latency to the output. It would probably > >> be better left as a new property or a driver specific feature. The point is to give userspace guaranteed expectations. If the expectation is that some displays might actually emit bad luminance flickering, then userspace must always assume the worst and implement the needed slewing algorithms to avoid it, even if it would be unnecessary. Userspace is farther away from the hardware than the drivers are, and if userspace is required to implement luminance flickering avoidance, that implementation must be done in all display servers and KMS apps that might enable VRR. That would also call for a userspace hardware database, so that people can set up quirks to enable/disable/adjust the algorithms for specific hardware with the hopes that other users could then have a good out-of-the-box experience. Instead, if possible, I would like to see some guarantees from the kernel drivers that userspace does not need to worry about luminance flickering. Unless you would deem all hardware that can exhibit luminance flickering as faulty and unsupported? We need a baseline default expectation. You can modify that expectation later with new properties, but I believe something needs to be defined as the default. Even if the definition is really just "hardware takes care of not flickering". > >> > >> In general I would imagine that most future VRR features would end up as > >> new properties. Anything that's purely software could be implemented as > >> a drm helper that every driver can use. I think the target presentation > >> timestamp feature is a good example for that. > > > > Speaking of timestamps. What is the expected behaviour of vblank > > timestamps when vrr is enabled? > > > > When vrr is enabled the duration of the vertical front porch will be > extended until flip or timeout occurs. The vblank timestamp will vary > based on duration of the vertical front porch. The min/max duration for > the front porch can be specified by the driver via the min/max range. ...This is actually useful information, it explains things. Do all VRR hardware implementations fundamentally work like this? With that definition, there is only more parameter to be exposed to userspace in the future: what is the length of the timeout? No need to expose maximum-refresh-freq because that information is already present in the programmed video mode. Btw. even this definition does not give any hint about problems like the luminance flickering. If possible flickering is an issue that cannot be simply ignored, then something should be documented about it. Do you know of any other "hidden" issues that could require userspace or drivers to be careful in how it will drive VRR? Thanks, pq > No changes to vblank timestamping handling should be necessary to > accommodate variable refresh rate. > > I think it's probably best to update the documentation for vrr_enable > with some of the specifics I described above. That should help clarify > userspace expectations as well. > > Nicholas Kazlauskas
On 10/29/18 4:36 AM, Pekka Paalanen wrote: > On Fri, 26 Oct 2018 17:34:15 +0000 > "Kazlauskas, Nicholas" <Nicholas.Kazlauskas@amd.com> wrote: > >> On 10/26/18 10:53 AM, Ville Syrjälä wrote: >>> On Fri, Oct 26, 2018 at 02:49:31PM +0000, Kazlauskas, Nicholas wrote: >>>> On 10/26/18 7:37 AM, Pekka Paalanen wrote: > >>>>> Hi, >>>>> >>>>> where is the documentation that explains how drivers must implement >>>>> "variable refresh rate adjustment"? >>>>> >>>>> What should and could userspace expect to get if it flicks this switch? >>>>> >>>>> I also think the kernel documentation should include a description of >>>>> what VRR actually is and how it conceptually works as far as userspace >>>>> is concerned. >>>>> >>>>> That is, the kernel documentation should describe what this thing does, >>>>> so that we avoid every driver implementing a different thing. For >>>>> example, one driver could prevent the luminance flickering itself by >>>>> tuning the timings while another driver might not do that, which means >>>>> that an application tested on the former driver will look just fine >>>>> while it is unbearable to watch on the latter. >>>>> >>>>> >>>>> Thanks, >>>>> pq >>>> >>>> I felt it was best to leave this more on the vague side to not impose >>>> restrictions yet on what a driver must do. >>>> >>>> If you think it's worth defining what the "baseline" expectation is for >>>> userspace, I would probably describe it as "utilizing the monitor's >>>> variable refresh rate to reduce stuttering or tearing that can occur >>>> during flipping". This is currently what the amdgpu driver has enabled >>>> for support. The implementation also isn't much more complex than just >>>> passing the variable refresh range to the hardware. > > Hi, > > sorry, but that says nothing. > > Also, tearing has nothing to do here. VRR reduces stuttering if > userspace is already tear-free. If userspace was driving the display in > a tearing fashion, VRR will not reduce or remove tearing, it just makes > it happen at different points and times. Tearing happens because > framebuffer flips are executed regardless of the refresh cycle, so > adjusting the refresh timings won't help. > > However... > >>>> >>>> I wouldn't want every driver to be forced to implement some sort of >>>> luminance flickering by default. It's not noticeable on many panels and >>>> any tuning would inherently add latency to the output. It would probably >>>> be better left as a new property or a driver specific feature. > > The point is to give userspace guaranteed expectations. If the > expectation is that some displays might actually emit bad luminance > flickering, then userspace must always assume the worst and implement > the needed slewing algorithms to avoid it, even if it would be > unnecessary. > > Userspace is farther away from the hardware than the drivers are, and > if userspace is required to implement luminance flickering avoidance, > that implementation must be done in all display servers and KMS apps > that might enable VRR. That would also call for a userspace hardware > database, so that people can set up quirks to enable/disable/adjust the > algorithms for specific hardware with the hopes that other users could > then have a good out-of-the-box experience. Instead, if possible, I > would like to see some guarantees from the kernel drivers that > userspace does not need to worry about luminance flickering. > > Unless you would deem all hardware that can exhibit luminance > flickering as faulty and unsupported? > > We need a baseline default expectation. You can modify that expectation > later with new properties, but I believe something needs to be defined > as the default. Even if the definition is really just "hardware takes > care of not flickering". > >>>> >>>> In general I would imagine that most future VRR features would end up as >>>> new properties. Anything that's purely software could be implemented as >>>> a drm helper that every driver can use. I think the target presentation >>>> timestamp feature is a good example for that. >>> >>> Speaking of timestamps. What is the expected behaviour of vblank >>> timestamps when vrr is enabled? >>> >> >> When vrr is enabled the duration of the vertical front porch will be >> extended until flip or timeout occurs. The vblank timestamp will vary >> based on duration of the vertical front porch. The min/max duration for >> the front porch can be specified by the driver via the min/max range. > > ...This is actually useful information, it explains things. Do all VRR > hardware implementations fundamentally work like this? > > With that definition, there is only more parameter to be exposed to > userspace in the future: what is the length of the timeout? No need to > expose maximum-refresh-freq because that information is already present > in the programmed video mode. > > Btw. even this definition does not give any hint about problems like > the luminance flickering. If possible flickering is an issue that > cannot be simply ignored, then something should be documented about it. > > Do you know of any other "hidden" issues that could require userspace > or drivers to be careful in how it will drive VRR? > > > Thanks, > pq There are two issues that can manifest when enabling variable refresh rate - "stuttering" and luminance flickering. Luminance flickering is dependent on the panel, but stuttering will occur on anything (and will be worse on panels with a wider range). Both of these issues can occur for content that flips on demand or at random. The large and frequent fluctuations in flip times remove any sort of consistency for user input and rendering. The stuttering is actually the primary reason the mesa patches exist (with luminance flickering removal being a bonus). Whether variable refresh rate should be enabled or not will depend on whether the userspace content is appropriate or not. A skewing algorithm would address both issues but would still make the user experience worse for those applications. Using a fixed rate for consistency is likely still preferable for those applications. This is why I'd prefer to define a baseline like: "When the vrr_enable property is set to true for vrr_capable hardware the vertical front porch will be extended until either flip or timeout occurs. The vertical front porch timing and timeout duration are determined by the driver based on the panel's reported minimum and maximum variable refresh range." The patches don't presently expose the minimum/maximum range but this does guarantee bounds. The inner bounds would then be determined by the driver to allow flexibility in implementing algorithms that can improve the user experience. An example of such an algorithm would be low framerate compensation for flip rates below the range. The amdgpu driver has an implementation for this that can do frame doubling and tripling for monitors with an appropriate range. You could also implement some sort of skewing for luminance flickering reduction for panels that really need it on top of this. Future properties that utilize variable refresh rate would be compatible with this baseline - everything is based on the determination of the inner bounds on minimum and maximum refresh. > >> No changes to vblank timestamping handling should be necessary to >> accommodate variable refresh rate. >> >> I think it's probably best to update the documentation for vrr_enable >> with some of the specifics I described above. That should help clarify >> userspace expectations as well. >> >> Nicholas Kazlauskas > Nicholas Kazlauskas
On 2018-10-26 7:59 p.m., Ville Syrjälä wrote: > On Fri, Oct 26, 2018 at 05:34:15PM +0000, Kazlauskas, Nicholas wrote: >> On 10/26/18 10:53 AM, Ville Syrjälä wrote: >>> >>> Speaking of timestamps. What is the expected behaviour of vblank >>> timestamps when vrr is enabled? >> >> When vrr is enabled the duration of the vertical front porch will be >> extended until flip or timeout occurs. The vblank timestamp will vary >> based on duration of the vertical front porch. The min/max duration for >> the front porch can be specified by the driver via the min/max range. >> >> No changes to vblank timestamping handling should be necessary to >> accommodate variable refresh rate. > > The problem is that the timestamp is supposed to correspond to the first > active pixel. And since we don't know how long the front porch will be > we can't realistically report the true value. So I guess just assuming > min front porch length is as good as anything else? That (and documenting that the timestamp corresponds to the earliest possible first active pixel, not necessarily the actual one, with VRR) might be good enough for the actual vblank event timestamps. However, I'm not so sure about the timestamps of page flip completion events. Those could be very misleading if the flip completes towards the timeout, which could result in bad behaviour of applications which use them for animation timing. Maybe the timestamp could be updated appropriately (yes, I'm hand-waving :) in drm_crtc_send_vblank_event?
On Mon, Oct 29, 2018 at 05:37:49PM +0100, Michel Dänzer wrote: > On 2018-10-26 7:59 p.m., Ville Syrjälä wrote: > > On Fri, Oct 26, 2018 at 05:34:15PM +0000, Kazlauskas, Nicholas wrote: > >> On 10/26/18 10:53 AM, Ville Syrjälä wrote: > >>> > >>> Speaking of timestamps. What is the expected behaviour of vblank > >>> timestamps when vrr is enabled? > >> > >> When vrr is enabled the duration of the vertical front porch will be > >> extended until flip or timeout occurs. The vblank timestamp will vary > >> based on duration of the vertical front porch. The min/max duration for > >> the front porch can be specified by the driver via the min/max range. > >> > >> No changes to vblank timestamping handling should be necessary to > >> accommodate variable refresh rate. > > > > The problem is that the timestamp is supposed to correspond to the first > > active pixel. And since we don't know how long the front porch will be > > we can't realistically report the true value. So I guess just assuming > > min front porch length is as good as anything else? > > That (and documenting that the timestamp corresponds to the earliest > possible first active pixel, not necessarily the actual one, with VRR) > might be good enough for the actual vblank event timestamps. > > > However, I'm not so sure about the timestamps of page flip completion > events. Those could be very misleading if the flip completes towards the > timeout, which could result in bad behaviour of applications which use > them for animation timing. > > Maybe the timestamp could be updated appropriately (yes, I'm hand-waving > :) in drm_crtc_send_vblank_event? Hmm. Updated how? Whether it's a page flip event or vblank even we won't know when the first active pixel will come. Although I suppose if there is some kind of vrr slew rate limit we could at least account for that to report a more correct "this is the earliest you migth be able to see your frame" timestamp. Oh, or are you actually saying that shceduling a new flip before the timeout is actually going to latch that flip immediately? I figured that the flip would get latched on the next start of vblank regardless, and the act of scheduling a flip will just kick the hardware to start scanning the previously latched frame earlier. scanout A | ... vblank | scanout A | ... vblank | scanout B | ... vblank ^ ^ ^ ^ | | flip C latch C flip B latch B vs. scanout A | ... vblank | scanout B | ... vblank | scanout C | ... vblank ^ ^ ^ ^ | latch B | latch C flip B flip C The latter would seem more like a tearing flip without the tearing.
On 10/29/18 2:03 PM, Ville Syrjälä wrote: > On Mon, Oct 29, 2018 at 05:37:49PM +0100, Michel Dänzer wrote: >> On 2018-10-26 7:59 p.m., Ville Syrjälä wrote: >>> On Fri, Oct 26, 2018 at 05:34:15PM +0000, Kazlauskas, Nicholas wrote: >>>> On 10/26/18 10:53 AM, Ville Syrjälä wrote: >>>>> >>>>> Speaking of timestamps. What is the expected behaviour of vblank >>>>> timestamps when vrr is enabled? >>>> >>>> When vrr is enabled the duration of the vertical front porch will be >>>> extended until flip or timeout occurs. The vblank timestamp will vary >>>> based on duration of the vertical front porch. The min/max duration for >>>> the front porch can be specified by the driver via the min/max range. >>>> >>>> No changes to vblank timestamping handling should be necessary to >>>> accommodate variable refresh rate. >>> >>> The problem is that the timestamp is supposed to correspond to the first >>> active pixel. And since we don't know how long the front porch will be >>> we can't realistically report the true value. So I guess just assuming >>> min front porch length is as good as anything else? >> >> That (and documenting that the timestamp corresponds to the earliest >> possible first active pixel, not necessarily the actual one, with VRR) >> might be good enough for the actual vblank event timestamps. >> >> >> However, I'm not so sure about the timestamps of page flip completion >> events. Those could be very misleading if the flip completes towards the >> timeout, which could result in bad behaviour of applications which use >> them for animation timing. >> >> Maybe the timestamp could be updated appropriately (yes, I'm hand-waving >> :) in drm_crtc_send_vblank_event? > > Hmm. Updated how? Whether it's a page flip event or vblank even we won't > know when the first active pixel will come. Although I suppose if > there is some kind of vrr slew rate limit we could at least account > for that to report a more correct "this is the earliest you migth be > able to see your frame" timestamp. > > Oh, or are you actually saying that shceduling a new flip before the > timeout is actually going to latch that flip immediately? I figured > that the flip would get latched on the next start of vblank regardless, > and the act of scheduling a flip will just kick the hardware to start > scanning the previously latched frame earlier. > > scanout A | ... vblank | scanout A | ... vblank | scanout B | ... vblank > ^ ^ ^ ^ > | | flip C latch C > flip B latch B > > vs. > > scanout A | ... vblank | scanout B | ... vblank | scanout C | ... vblank > ^ ^ ^ ^ > | latch B | latch C > flip B flip C > > The latter would seem more like a tearing flip without the tearing. > I'm not sure any of this is necessary. The vblank timestamp is determined (at least if you're using the helpers) by using the scanout position. The crtc_vtotal won't change when variable refresh rate is enabled. What does change is the behavior of the scanout position. The vpos will continue going will beyond the end of crtc_vtotal up until whatever the vtotal would be for the minimum refresh rate. There's no clamping performed for these calculations so it ends up working as expected. Nicholas Kazlauskas
On 2018-10-29 8:11 p.m., Kazlauskas, Nicholas wrote: > On 10/29/18 2:03 PM, Ville Syrjälä wrote: >> On Mon, Oct 29, 2018 at 05:37:49PM +0100, Michel Dänzer wrote: >>> On 2018-10-26 7:59 p.m., Ville Syrjälä wrote: >>>> On Fri, Oct 26, 2018 at 05:34:15PM +0000, Kazlauskas, Nicholas wrote: >>>>> On 10/26/18 10:53 AM, Ville Syrjälä wrote: >>>>>> >>>>>> Speaking of timestamps. What is the expected behaviour of vblank >>>>>> timestamps when vrr is enabled? >>>>> >>>>> When vrr is enabled the duration of the vertical front porch will be >>>>> extended until flip or timeout occurs. The vblank timestamp will vary >>>>> based on duration of the vertical front porch. The min/max duration for >>>>> the front porch can be specified by the driver via the min/max range. >>>>> >>>>> No changes to vblank timestamping handling should be necessary to >>>>> accommodate variable refresh rate. >>>> >>>> The problem is that the timestamp is supposed to correspond to the first >>>> active pixel. And since we don't know how long the front porch will be >>>> we can't realistically report the true value. So I guess just assuming >>>> min front porch length is as good as anything else? >>> >>> That (and documenting that the timestamp corresponds to the earliest >>> possible first active pixel, not necessarily the actual one, with VRR) >>> might be good enough for the actual vblank event timestamps. >>> >>> >>> However, I'm not so sure about the timestamps of page flip completion >>> events. Those could be very misleading if the flip completes towards the >>> timeout, which could result in bad behaviour of applications which use >>> them for animation timing. >>> >>> Maybe the timestamp could be updated appropriately (yes, I'm hand-waving >>> :) in drm_crtc_send_vblank_event? >> >> [...] > > I'm not sure any of this is necessary. > > The vblank timestamp is determined (at least if you're using the > helpers) by using the scanout position. > > The crtc_vtotal won't change when variable refresh rate is enabled. What > does change is the behavior of the scanout position. > > The vpos will continue going will beyond the end of crtc_vtotal up until > whatever the vtotal would be for the minimum refresh rate. There's no > clamping performed for these calculations so it ends up working as expected. I'm afraid not. amdgpu_display_get_crtc_scanoutpos returns how many lines after "the end of vblank" (which is calculated based on crtc_vtotal) the current scanout position is. drm_calc_vbltimestamp_from_scanoutpos converts that to a time delta from "the end of vblank", and subtracts that from the current time (when the scanout position was sampled). The result is the timestamp corresponding to "the end of vblank" based on crtc_vtotal.
On 2018-10-29 7:03 p.m., Ville Syrjälä wrote: > On Mon, Oct 29, 2018 at 05:37:49PM +0100, Michel Dänzer wrote: >> On 2018-10-26 7:59 p.m., Ville Syrjälä wrote: >>> On Fri, Oct 26, 2018 at 05:34:15PM +0000, Kazlauskas, Nicholas wrote: >>>> On 10/26/18 10:53 AM, Ville Syrjälä wrote: >>>>> >>>>> Speaking of timestamps. What is the expected behaviour of vblank >>>>> timestamps when vrr is enabled? >>>> >>>> When vrr is enabled the duration of the vertical front porch will be >>>> extended until flip or timeout occurs. The vblank timestamp will vary >>>> based on duration of the vertical front porch. The min/max duration for >>>> the front porch can be specified by the driver via the min/max range. >>>> >>>> No changes to vblank timestamping handling should be necessary to >>>> accommodate variable refresh rate. >>> >>> The problem is that the timestamp is supposed to correspond to the first >>> active pixel. And since we don't know how long the front porch will be >>> we can't realistically report the true value. So I guess just assuming >>> min front porch length is as good as anything else? >> >> That (and documenting that the timestamp corresponds to the earliest >> possible first active pixel, not necessarily the actual one, with VRR) >> might be good enough for the actual vblank event timestamps. >> >> >> However, I'm not so sure about the timestamps of page flip completion >> events. Those could be very misleading if the flip completes towards the >> timeout, which could result in bad behaviour of applications which use >> them for animation timing. >> >> Maybe the timestamp could be updated appropriately (yes, I'm hand-waving >> :) in drm_crtc_send_vblank_event? > > Hmm. Updated how? Whether it's a page flip event or vblank even we won't > know when the first active pixel will come. Although I suppose if > there is some kind of vrr slew rate limit we could at least account > for that to report a more correct "this is the earliest you migth be > able to see your frame" timestamp. > > Oh, or are you actually saying that shceduling a new flip before the > timeout is actually going to latch that flip immediately? I figured > that the flip would get latched on the next start of vblank regardless, > and the act of scheduling a flip will just kick the hardware to start > scanning the previously latched frame earlier. > > scanout A | ... vblank | scanout A | ... vblank | scanout B | ... vblank > ^ ^ ^ ^ > | | flip C latch C > flip B latch B This would kind of defeat the point of VRR, wouldn't it? If a flip was scheduled after the start of vblank, the vblank would always time out, resulting in the minimum refresh rate. > scanout A | ... vblank | scanout B | ... vblank | scanout C | ... vblank > ^ ^ ^ ^ > | latch B | latch C > flip B flip C So this is what happens. Regardless, when the flip is latched, AMD hardware generates a "pflip" interrupt, and its handler calls drm_crtc_send_vblank_event (and in the case of DC drm_crtc_accurate_vblank_count before that). So the time when drm_crtc_send_vblank_event is called might be a good approximation of when scanout of the next frame starts. Another possibility might be to wait for the hardware vline counter to wrap around to 0 before calling drm_crtc_accurate_vblank_count, then the calculations should be based on 0 instead of crtc_vtotal.
On 10/30/18 5:29 AM, Michel Dänzer wrote: > On 2018-10-29 7:03 p.m., Ville Syrjälä wrote: >> On Mon, Oct 29, 2018 at 05:37:49PM +0100, Michel Dänzer wrote: >>> On 2018-10-26 7:59 p.m., Ville Syrjälä wrote: >>>> On Fri, Oct 26, 2018 at 05:34:15PM +0000, Kazlauskas, Nicholas wrote: >>>>> On 10/26/18 10:53 AM, Ville Syrjälä wrote: >>>>>> >>>>>> Speaking of timestamps. What is the expected behaviour of vblank >>>>>> timestamps when vrr is enabled? >>>>> >>>>> When vrr is enabled the duration of the vertical front porch will be >>>>> extended until flip or timeout occurs. The vblank timestamp will vary >>>>> based on duration of the vertical front porch. The min/max duration for >>>>> the front porch can be specified by the driver via the min/max range. >>>>> >>>>> No changes to vblank timestamping handling should be necessary to >>>>> accommodate variable refresh rate. >>>> >>>> The problem is that the timestamp is supposed to correspond to the first >>>> active pixel. And since we don't know how long the front porch will be >>>> we can't realistically report the true value. So I guess just assuming >>>> min front porch length is as good as anything else? >>> >>> That (and documenting that the timestamp corresponds to the earliest >>> possible first active pixel, not necessarily the actual one, with VRR) >>> might be good enough for the actual vblank event timestamps. >>> >>> >>> However, I'm not so sure about the timestamps of page flip completion >>> events. Those could be very misleading if the flip completes towards the >>> timeout, which could result in bad behaviour of applications which use >>> them for animation timing. >>> >>> Maybe the timestamp could be updated appropriately (yes, I'm hand-waving >>> :) in drm_crtc_send_vblank_event? >> >> Hmm. Updated how? Whether it's a page flip event or vblank even we won't >> know when the first active pixel will come. Although I suppose if >> there is some kind of vrr slew rate limit we could at least account >> for that to report a more correct "this is the earliest you migth be >> able to see your frame" timestamp. >> >> Oh, or are you actually saying that shceduling a new flip before the >> timeout is actually going to latch that flip immediately? I figured >> that the flip would get latched on the next start of vblank regardless, >> and the act of scheduling a flip will just kick the hardware to start >> scanning the previously latched frame earlier. >> >> scanout A | ... vblank | scanout A | ... vblank | scanout B | ... vblank >> ^ ^ ^ ^ >> | | flip C latch C >> flip B latch B > > This would kind of defeat the point of VRR, wouldn't it? If a flip was > scheduled after the start of vblank, the vblank would always time out, > resulting in the minimum refresh rate. > > >> scanout A | ... vblank | scanout B | ... vblank | scanout C | ... vblank >> ^ ^ ^ ^ >> | latch B | latch C >> flip B flip C > > So this is what happens. > > Regardless, when the flip is latched, AMD hardware generates a "pflip" > interrupt, and its handler calls drm_crtc_send_vblank_event (and in the > case of DC drm_crtc_accurate_vblank_count before that). So the time when > drm_crtc_send_vblank_event is called might be a good approximation of > when scanout of the next frame starts. > > Another possibility might be to wait for the hardware vline counter to > wrap around to 0 before calling drm_crtc_accurate_vblank_count, then the > calculations should be based on 0 instead of crtc_vtotal. > > I understand the issue you're describing now. The timestamp is supposed to signify the end of the current vblank. The call to get scanout position is supposed to return the number of lines until scanout (a negative value) or the number of lines since the next scanout began (a positive value). The amdgpu driver calculates the number of lines based on a hardware register status position which returns an increasing value from 0 that indicates the current vpos/hpos for the display. For any vpos below vbl_start we know the value is correct since the next vblank hasn't begun yet. But for anythign about vbl_start it's probably wrong since it applies a corrective offset based on the fixed value of crtc_vtotal. It's even worse when the value is above crtc_vtotal since it'll be calculating the number of lines since the last scanout since it'll be a positive value. So the issue becomes determining when the vfront porch will end. When the flip address gets written the vfront porch will end at the start of the next line leaving only the back porch plus part of the line. But we don't know when the flip will occur, if at all. It hasn't occurred yet in this case. Waiting for the wrap around to 0 might be the best choice here since there's no guarantee the flip will occur. Nicholas Kazlauskas
On 10/30/18 11:34 AM, Kazlauskas, Nicholas wrote: > On 10/30/18 5:29 AM, Michel Dänzer wrote: >> On 2018-10-29 7:03 p.m., Ville Syrjälä wrote: >>> On Mon, Oct 29, 2018 at 05:37:49PM +0100, Michel Dänzer wrote: >>>> On 2018-10-26 7:59 p.m., Ville Syrjälä wrote: >>>>> On Fri, Oct 26, 2018 at 05:34:15PM +0000, Kazlauskas, Nicholas wrote: >>>>>> On 10/26/18 10:53 AM, Ville Syrjälä wrote: >>>>>>> >>>>>>> Speaking of timestamps. What is the expected behaviour of vblank >>>>>>> timestamps when vrr is enabled? >>>>>> >>>>>> When vrr is enabled the duration of the vertical front porch will be >>>>>> extended until flip or timeout occurs. The vblank timestamp will vary >>>>>> based on duration of the vertical front porch. The min/max duration for >>>>>> the front porch can be specified by the driver via the min/max range. >>>>>> >>>>>> No changes to vblank timestamping handling should be necessary to >>>>>> accommodate variable refresh rate. >>>>> >>>>> The problem is that the timestamp is supposed to correspond to the first >>>>> active pixel. And since we don't know how long the front porch will be >>>>> we can't realistically report the true value. So I guess just assuming >>>>> min front porch length is as good as anything else? >>>> >>>> That (and documenting that the timestamp corresponds to the earliest >>>> possible first active pixel, not necessarily the actual one, with VRR) >>>> might be good enough for the actual vblank event timestamps. >>>> >>>> >>>> However, I'm not so sure about the timestamps of page flip completion >>>> events. Those could be very misleading if the flip completes towards the >>>> timeout, which could result in bad behaviour of applications which use >>>> them for animation timing. >>>> >>>> Maybe the timestamp could be updated appropriately (yes, I'm hand-waving >>>> :) in drm_crtc_send_vblank_event? >>> >>> Hmm. Updated how? Whether it's a page flip event or vblank even we won't >>> know when the first active pixel will come. Although I suppose if >>> there is some kind of vrr slew rate limit we could at least account >>> for that to report a more correct "this is the earliest you migth be >>> able to see your frame" timestamp. >>> >>> Oh, or are you actually saying that shceduling a new flip before the >>> timeout is actually going to latch that flip immediately? I figured >>> that the flip would get latched on the next start of vblank regardless, >>> and the act of scheduling a flip will just kick the hardware to start >>> scanning the previously latched frame earlier. >>> >>> scanout A | ... vblank | scanout A | ... vblank | scanout B | ... vblank >>> ^ ^ ^ ^ >>> | | flip C latch C >>> flip B latch B >> >> This would kind of defeat the point of VRR, wouldn't it? If a flip was >> scheduled after the start of vblank, the vblank would always time out, >> resulting in the minimum refresh rate. >> >> >>> scanout A | ... vblank | scanout B | ... vblank | scanout C | ... vblank >>> ^ ^ ^ ^ >>> | latch B | latch C >>> flip B flip C >> >> So this is what happens. >> >> Regardless, when the flip is latched, AMD hardware generates a "pflip" >> interrupt, and its handler calls drm_crtc_send_vblank_event (and in the >> case of DC drm_crtc_accurate_vblank_count before that). So the time when >> drm_crtc_send_vblank_event is called might be a good approximation of >> when scanout of the next frame starts. >> >> Another possibility might be to wait for the hardware vline counter to >> wrap around to 0 before calling drm_crtc_accurate_vblank_count, then the >> calculations should be based on 0 instead of crtc_vtotal. >> >> > > > I understand the issue you're describing now. The timestamp is supposed > to signify the end of the current vblank. The call to get scanout > position is supposed to return the number of lines until scanout (a > negative value) or the number of lines since the next scanout began (a > positive value). > > The amdgpu driver calculates the number of lines based on a hardware > register status position which returns an increasing value from 0 that > indicates the current vpos/hpos for the display. > > For any vpos below vbl_start we know the value is correct since the next > vblank hasn't begun yet. But for anythign about vbl_start it's probably > wrong since it applies a corrective offset based on the fixed value of > crtc_vtotal. It's even worse when the value is above crtc_vtotal since > it'll be calculating the number of lines since the last scanout since > it'll be a positive value. > > So the issue becomes determining when the vfront porch will end. > > When the flip address gets written the vfront porch will end at the > start of the next line leaving only the back porch plus part of the > line. But we don't know when the flip will occur, if at all. It hasn't > occurred yet in this case. > > Waiting for the wrap around to 0 might be the best choice here since > there's no guarantee the flip will occur. > > Nicholas Kazlauskas > _______________________________________________ > dri-devel mailing list > dri-devel@lists.freedesktop.org > https://lists.freedesktop.org/mailman/listinfo/dri-devel > I put some more thought into this and I don't think this is as bad as I had originally thought. I think the vblank timestamp is supposed to be for the first active pixel of the next scanout. The usage of which is for clients to time their content/animation/etc. The client likely doesn't care when they issue their flip, just that their content is matched for that vblank timestamp. The fixed refresh model works really well for that kind of application. Utilizing variable refresh rate would be a mistake in that case since the client would then have to time based on when they flip which is a lot harder to "predict" precisely. I did some more investigation into when amdgpu gets the scanout position and what values we get back out of it. The timestamp is updated shortly after the crtc irq vblank which is typically within a few lines of vbl_start. This makes sense, since we can provide the prediction timestamp early. Waiting for the vblank to wrap around to 0 doesn't really make sense here since that would mean we already hit timeout or the flip occurred - making the timestamp useless and probably burning a ton of CPU to do it. What this means is that we're going to have to make some sort of prediction based on a fixed duration to have a reasonable result. I think assuming the min front porch duration (max refresh) is the logical approach here - in most cases this will be crtc_vtotal since it's the monitor's fastest refresh. For clients that flip early this will end up being the correct timestamp. No need to explicitly check for variable refresh enabled in the get scanout pos handler either. There is one little caveat with get scanout pos with that approach, however. For usages other than the timestamp the number of lines until scanout will still be incorrect when going above crtc_vtotal. We can either assume the worst in this case (the max front porch duration, the min refresh/timeout) or clamp and just return the size of the back porch instead The vpos will look stalled, but the flip will likely happen sooner than the timeout in most cases so this will likely work out better. Nicholas Kazlauskas
On 2018-10-31 2:38 p.m., Kazlauskas, Nicholas wrote: > On 10/30/18 11:34 AM, Kazlauskas, Nicholas wrote: >> On 10/30/18 5:29 AM, Michel Dänzer wrote: >>> On 2018-10-29 7:03 p.m., Ville Syrjälä wrote: >>>> On Mon, Oct 29, 2018 at 05:37:49PM +0100, Michel Dänzer wrote: >>>>> On 2018-10-26 7:59 p.m., Ville Syrjälä wrote: >>>>>> On Fri, Oct 26, 2018 at 05:34:15PM +0000, Kazlauskas, Nicholas wrote: >>>>>>> On 10/26/18 10:53 AM, Ville Syrjälä wrote: >>>>>>>> >>>>>>>> Speaking of timestamps. What is the expected behaviour of vblank >>>>>>>> timestamps when vrr is enabled? >>>>>>> >>>>>>> When vrr is enabled the duration of the vertical front porch will be >>>>>>> extended until flip or timeout occurs. The vblank timestamp will vary >>>>>>> based on duration of the vertical front porch. The min/max duration for >>>>>>> the front porch can be specified by the driver via the min/max range. >>>>>>> >>>>>>> No changes to vblank timestamping handling should be necessary to >>>>>>> accommodate variable refresh rate. >>>>>> >>>>>> The problem is that the timestamp is supposed to correspond to the first >>>>>> active pixel. And since we don't know how long the front porch will be >>>>>> we can't realistically report the true value. So I guess just assuming >>>>>> min front porch length is as good as anything else? >>>>> >>>>> That (and documenting that the timestamp corresponds to the earliest >>>>> possible first active pixel, not necessarily the actual one, with VRR) >>>>> might be good enough for the actual vblank event timestamps. >>>>> >>>>> >>>>> However, I'm not so sure about the timestamps of page flip completion >>>>> events. Those could be very misleading if the flip completes towards the >>>>> timeout, which could result in bad behaviour of applications which use >>>>> them for animation timing. >>>>> >>>>> Maybe the timestamp could be updated appropriately (yes, I'm hand-waving >>>>> :) in drm_crtc_send_vblank_event? >>>> >>>> Hmm. Updated how? Whether it's a page flip event or vblank even we won't >>>> know when the first active pixel will come. Although I suppose if >>>> there is some kind of vrr slew rate limit we could at least account >>>> for that to report a more correct "this is the earliest you migth be >>>> able to see your frame" timestamp. >>>> >>>> Oh, or are you actually saying that shceduling a new flip before the >>>> timeout is actually going to latch that flip immediately? I figured >>>> that the flip would get latched on the next start of vblank regardless, >>>> and the act of scheduling a flip will just kick the hardware to start >>>> scanning the previously latched frame earlier. >>>> >>>> scanout A | ... vblank | scanout A | ... vblank | scanout B | ... vblank >>>> ^ ^ ^ ^ >>>> | | flip C latch C >>>> flip B latch B >>> >>> This would kind of defeat the point of VRR, wouldn't it? If a flip was >>> scheduled after the start of vblank, the vblank would always time out, >>> resulting in the minimum refresh rate. >>> >>> >>>> scanout A | ... vblank | scanout B | ... vblank | scanout C | ... vblank >>>> ^ ^ ^ ^ >>>> | latch B | latch C >>>> flip B flip C >>> >>> So this is what happens. >>> >>> Regardless, when the flip is latched, AMD hardware generates a "pflip" >>> interrupt, and its handler calls drm_crtc_send_vblank_event (and in the >>> case of DC drm_crtc_accurate_vblank_count before that). So the time when >>> drm_crtc_send_vblank_event is called might be a good approximation of >>> when scanout of the next frame starts. >>> >>> Another possibility might be to wait for the hardware vline counter to >>> wrap around to 0 before calling drm_crtc_accurate_vblank_count, then the >>> calculations should be based on 0 instead of crtc_vtotal. >> >> >> I understand the issue you're describing now. The timestamp is supposed >> to signify the end of the current vblank. The call to get scanout >> position is supposed to return the number of lines until scanout (a >> negative value) or the number of lines since the next scanout began (a >> positive value). >> >> The amdgpu driver calculates the number of lines based on a hardware >> register status position which returns an increasing value from 0 that >> indicates the current vpos/hpos for the display. >> >> For any vpos below vbl_start we know the value is correct since the next >> vblank hasn't begun yet. But for anythign about vbl_start it's probably >> wrong since it applies a corrective offset based on the fixed value of >> crtc_vtotal. It's even worse when the value is above crtc_vtotal since >> it'll be calculating the number of lines since the last scanout since >> it'll be a positive value. >> >> So the issue becomes determining when the vfront porch will end. >> >> When the flip address gets written the vfront porch will end at the >> start of the next line leaving only the back porch plus part of the >> line. But we don't know when the flip will occur, if at all. It hasn't >> occurred yet in this case. >> >> Waiting for the wrap around to 0 might be the best choice here since >> there's no guarantee the flip will occur. > > I put some more thought into this and I don't think this is as bad as I > had originally thought. > > I think the vblank timestamp is supposed to be for the first active > pixel of the next scanout. The usage of which is for clients to time > their content/animation/etc. > > The client likely doesn't care when they issue their flip, just that > their content is matched for that vblank timestamp. The fixed refresh > model works really well for that kind of application. > > Utilizing variable refresh rate would be a mistake in that case since > the client would then have to time based on when they flip which is a > lot harder to "predict" precisely. It's only a "mistake" as long as the timestamps are misleading. :) As discussed before, accurate presentation timestamps are one requirement for achieving perfectly smooth animation. > I did some more investigation into when amdgpu gets the scanout position > and what values we get back out of it. The timestamp is updated shortly > after the crtc irq vblank which is typically within a few lines of > vbl_start. This makes sense, since we can provide the prediction > timestamp early. Waiting for the vblank to wrap around to 0 doesn't > really make sense here since that would mean we already hit timeout or > the flip occurred Sounds like you're mixing up the two cases of "actual" vblank events (triggered by the "vblank" interrupt => drm_(crtc_)handle_vblank) and flip completion events (triggered by the PFLIP interrupt with our hardware => drm_crtc_send_vblank_event). Actual vblank events need to be delivered to userspace at the start of vblank, so we indeed can't wait until the timestamp is accurate for them. We just need to document the exact semantics of their timestamp with VRR. For page flip completion events though, the timestamp needs to be accurate and correspond to when the flipped frame starts being scanned out, otherwise we'll surely break at least some userspace relying on this information.
On 10/31/18 10:12 AM, Michel Dänzer wrote: > On 2018-10-31 2:38 p.m., Kazlauskas, Nicholas wrote: >> On 10/30/18 11:34 AM, Kazlauskas, Nicholas wrote: >>> On 10/30/18 5:29 AM, Michel Dänzer wrote: >>>> On 2018-10-29 7:03 p.m., Ville Syrjälä wrote: >>>>> On Mon, Oct 29, 2018 at 05:37:49PM +0100, Michel Dänzer wrote: >>>>>> On 2018-10-26 7:59 p.m., Ville Syrjälä wrote: >>>>>>> On Fri, Oct 26, 2018 at 05:34:15PM +0000, Kazlauskas, Nicholas wrote: >>>>>>>> On 10/26/18 10:53 AM, Ville Syrjälä wrote: >>>>>>>>> >>>>>>>>> Speaking of timestamps. What is the expected behaviour of vblank >>>>>>>>> timestamps when vrr is enabled? >>>>>>>> >>>>>>>> When vrr is enabled the duration of the vertical front porch will be >>>>>>>> extended until flip or timeout occurs. The vblank timestamp will vary >>>>>>>> based on duration of the vertical front porch. The min/max duration for >>>>>>>> the front porch can be specified by the driver via the min/max range. >>>>>>>> >>>>>>>> No changes to vblank timestamping handling should be necessary to >>>>>>>> accommodate variable refresh rate. >>>>>>> >>>>>>> The problem is that the timestamp is supposed to correspond to the first >>>>>>> active pixel. And since we don't know how long the front porch will be >>>>>>> we can't realistically report the true value. So I guess just assuming >>>>>>> min front porch length is as good as anything else? >>>>>> >>>>>> That (and documenting that the timestamp corresponds to the earliest >>>>>> possible first active pixel, not necessarily the actual one, with VRR) >>>>>> might be good enough for the actual vblank event timestamps. >>>>>> >>>>>> >>>>>> However, I'm not so sure about the timestamps of page flip completion >>>>>> events. Those could be very misleading if the flip completes towards the >>>>>> timeout, which could result in bad behaviour of applications which use >>>>>> them for animation timing. >>>>>> >>>>>> Maybe the timestamp could be updated appropriately (yes, I'm hand-waving >>>>>> :) in drm_crtc_send_vblank_event? >>>>> >>>>> Hmm. Updated how? Whether it's a page flip event or vblank even we won't >>>>> know when the first active pixel will come. Although I suppose if >>>>> there is some kind of vrr slew rate limit we could at least account >>>>> for that to report a more correct "this is the earliest you migth be >>>>> able to see your frame" timestamp. >>>>> >>>>> Oh, or are you actually saying that shceduling a new flip before the >>>>> timeout is actually going to latch that flip immediately? I figured >>>>> that the flip would get latched on the next start of vblank regardless, >>>>> and the act of scheduling a flip will just kick the hardware to start >>>>> scanning the previously latched frame earlier. >>>>> >>>>> scanout A | ... vblank | scanout A | ... vblank | scanout B | ... vblank >>>>> ^ ^ ^ ^ >>>>> | | flip C latch C >>>>> flip B latch B >>>> >>>> This would kind of defeat the point of VRR, wouldn't it? If a flip was >>>> scheduled after the start of vblank, the vblank would always time out, >>>> resulting in the minimum refresh rate. >>>> >>>> >>>>> scanout A | ... vblank | scanout B | ... vblank | scanout C | ... vblank >>>>> ^ ^ ^ ^ >>>>> | latch B | latch C >>>>> flip B flip C >>>> >>>> So this is what happens. >>>> >>>> Regardless, when the flip is latched, AMD hardware generates a "pflip" >>>> interrupt, and its handler calls drm_crtc_send_vblank_event (and in the >>>> case of DC drm_crtc_accurate_vblank_count before that). So the time when >>>> drm_crtc_send_vblank_event is called might be a good approximation of >>>> when scanout of the next frame starts. >>>> >>>> Another possibility might be to wait for the hardware vline counter to >>>> wrap around to 0 before calling drm_crtc_accurate_vblank_count, then the >>>> calculations should be based on 0 instead of crtc_vtotal. >>> >>> >>> I understand the issue you're describing now. The timestamp is supposed >>> to signify the end of the current vblank. The call to get scanout >>> position is supposed to return the number of lines until scanout (a >>> negative value) or the number of lines since the next scanout began (a >>> positive value). >>> >>> The amdgpu driver calculates the number of lines based on a hardware >>> register status position which returns an increasing value from 0 that >>> indicates the current vpos/hpos for the display. >>> >>> For any vpos below vbl_start we know the value is correct since the next >>> vblank hasn't begun yet. But for anythign about vbl_start it's probably >>> wrong since it applies a corrective offset based on the fixed value of >>> crtc_vtotal. It's even worse when the value is above crtc_vtotal since >>> it'll be calculating the number of lines since the last scanout since >>> it'll be a positive value. >>> >>> So the issue becomes determining when the vfront porch will end. >>> >>> When the flip address gets written the vfront porch will end at the >>> start of the next line leaving only the back porch plus part of the >>> line. But we don't know when the flip will occur, if at all. It hasn't >>> occurred yet in this case. >>> >>> Waiting for the wrap around to 0 might be the best choice here since >>> there's no guarantee the flip will occur. >> >> I put some more thought into this and I don't think this is as bad as I >> had originally thought. >> >> I think the vblank timestamp is supposed to be for the first active >> pixel of the next scanout. The usage of which is for clients to time >> their content/animation/etc. >> >> The client likely doesn't care when they issue their flip, just that >> their content is matched for that vblank timestamp. The fixed refresh >> model works really well for that kind of application. >> >> Utilizing variable refresh rate would be a mistake in that case since >> the client would then have to time based on when they flip which is a >> lot harder to "predict" precisely. > > It's only a "mistake" as long as the timestamps are misleading. :) As > discussed before, accurate presentation timestamps are one requirement > for achieving perfectly smooth animation. For most 3D games the game world/rendering can advance by an arbitrary timestep - and faster will be smoother, which is what the native mode would give you. For fixed interval content/animation where you can't do that variable refresh rate could vastly improve the output. But like discussed before that would need a way to specify the interval/presentation timestamp to control that front porch duration. The timestamp will be misleading in any other case. > > >> I did some more investigation into when amdgpu gets the scanout position >> and what values we get back out of it. The timestamp is updated shortly >> after the crtc irq vblank which is typically within a few lines of >> vbl_start. This makes sense, since we can provide the prediction >> timestamp early. Waiting for the vblank to wrap around to 0 doesn't >> really make sense here since that would mean we already hit timeout or >> the flip occurred > > Sounds like you're mixing up the two cases of "actual" vblank events > (triggered by the "vblank" interrupt => drm_(crtc_)handle_vblank) and > flip completion events (triggered by the PFLIP interrupt with our > hardware => drm_crtc_send_vblank_event). > > Actual vblank events need to be delivered to userspace at the start of > vblank, so we indeed can't wait until the timestamp is accurate for > them. We just need to document the exact semantics of their timestamp > with VRR. > > For page flip completion events though, the timestamp needs to be > accurate and correspond to when the flipped frame starts being scanned > out, otherwise we'll surely break at least some userspace relying on > this information. > > Yeah, I was. I guess what's sent is the estimated vblank timestamp calculated at the start of the interrupt. And since that's just a guess rather than what's actually going to happen it's going to be wrong in a lot of cases. I could see the wrap-around method working if the vblank timestamp was somehow updated in amdgpu or in drm_crtc_send_vblank_event. This would be a relatively simple fix but would break anything in userspace that relied on the timestamp for vblank interrupt and the flip completion being the same value. But I guess that's the case for any fix for this potential problem. Nicholas Kazlauskas
On 2018-10-31 3:41 p.m., Kazlauskas, Nicholas wrote: > On 10/31/18 10:12 AM, Michel Dänzer wrote: >> On 2018-10-31 2:38 p.m., Kazlauskas, Nicholas wrote: >>> On 10/30/18 11:34 AM, Kazlauskas, Nicholas wrote: >>>> >>>> I understand the issue you're describing now. The timestamp is supposed >>>> to signify the end of the current vblank. The call to get scanout >>>> position is supposed to return the number of lines until scanout (a >>>> negative value) or the number of lines since the next scanout began (a >>>> positive value). >>>> >>>> The amdgpu driver calculates the number of lines based on a hardware >>>> register status position which returns an increasing value from 0 that >>>> indicates the current vpos/hpos for the display. >>>> >>>> For any vpos below vbl_start we know the value is correct since the next >>>> vblank hasn't begun yet. But for anythign about vbl_start it's probably >>>> wrong since it applies a corrective offset based on the fixed value of >>>> crtc_vtotal. It's even worse when the value is above crtc_vtotal since >>>> it'll be calculating the number of lines since the last scanout since >>>> it'll be a positive value. >>>> >>>> So the issue becomes determining when the vfront porch will end. >>>> >>>> When the flip address gets written the vfront porch will end at the >>>> start of the next line leaving only the back porch plus part of the >>>> line. But we don't know when the flip will occur, if at all. It hasn't >>>> occurred yet in this case. >>>> >>>> Waiting for the wrap around to 0 might be the best choice here since >>>> there's no guarantee the flip will occur. >>> >>> I put some more thought into this and I don't think this is as bad as I >>> had originally thought. >>> >>> I think the vblank timestamp is supposed to be for the first active >>> pixel of the next scanout. The usage of which is for clients to time >>> their content/animation/etc. >>> >>> The client likely doesn't care when they issue their flip, just that >>> their content is matched for that vblank timestamp. The fixed refresh >>> model works really well for that kind of application. >>> >>> Utilizing variable refresh rate would be a mistake in that case since >>> the client would then have to time based on when they flip which is a >>> lot harder to "predict" precisely. >> >> It's only a "mistake" as long as the timestamps are misleading. :) As >> discussed before, accurate presentation timestamps are one requirement >> for achieving perfectly smooth animation. > > For most 3D games the game world/rendering can advance by an arbitrary > timestep - and faster will be smoother, which is what the native mode > would give you. > > For fixed interval content/animation where you can't do that variable > refresh rate could vastly improve the output. But like discussed before > that would need a way to specify the interval/presentation timestamp to > control that front porch duration. The timestamp will be misleading in > any other case. I don't agree that an accurate timestamp can ever be more "misleading" than an inaccurate one. But yeah, accurate timestamps and time-based presentation are two sides of the same coin which can buy the holy grail of perfectly smooth animation. :) >>> I did some more investigation into when amdgpu gets the scanout position >>> and what values we get back out of it. The timestamp is updated shortly >>> after the crtc irq vblank which is typically within a few lines of >>> vbl_start. This makes sense, since we can provide the prediction >>> timestamp early. Waiting for the vblank to wrap around to 0 doesn't >>> really make sense here since that would mean we already hit timeout or >>> the flip occurred >> >> Sounds like you're mixing up the two cases of "actual" vblank events >> (triggered by the "vblank" interrupt => drm_(crtc_)handle_vblank) and >> flip completion events (triggered by the PFLIP interrupt with our >> hardware => drm_crtc_send_vblank_event). >> >> Actual vblank events need to be delivered to userspace at the start of >> vblank, so we indeed can't wait until the timestamp is accurate for >> them. We just need to document the exact semantics of their timestamp >> with VRR. >> >> For page flip completion events though, the timestamp needs to be >> accurate and correspond to when the flipped frame starts being scanned >> out, otherwise we'll surely break at least some userspace relying on >> this information. >> > Yeah, I was. I guess what's sent is the estimated vblank timestamp > calculated at the start of the interrupt. s/interrupt/vblank/, yeah. > And since that's just a guess rather than what's actually going to > happen it's going to be wrong in a lot of cases. > > I could see the wrap-around method working if the vblank timestamp was > somehow updated in amdgpu or in drm_crtc_send_vblank_event. DC already calls drm_crtc_accurate_vblank_count before drm_crtc_send_vblank_event, we "just" need to make sure that results in an accurate timestamp. > This would be a relatively simple fix but would break anything in > userspace that relied on the timestamp for vblank interrupt and the > flip completion being the same value. Hmm, that's a good point. So while VRR is enabled, maybe it is safer to defer delivery of vblank events until the accurate timestamp is known as well, at least by default. If there is userspace which needs the events earlier even with VRR but can live with the guessed timestamp, a flag or something could be added for that.
On 10/31/18 12:20 PM, Michel Dänzer wrote: > On 2018-10-31 3:41 p.m., Kazlauskas, Nicholas wrote: >> On 10/31/18 10:12 AM, Michel Dänzer wrote: >>> On 2018-10-31 2:38 p.m., Kazlauskas, Nicholas wrote: >>>> On 10/30/18 11:34 AM, Kazlauskas, Nicholas wrote: >>>>> >>>>> I understand the issue you're describing now. The timestamp is supposed >>>>> to signify the end of the current vblank. The call to get scanout >>>>> position is supposed to return the number of lines until scanout (a >>>>> negative value) or the number of lines since the next scanout began (a >>>>> positive value). >>>>> >>>>> The amdgpu driver calculates the number of lines based on a hardware >>>>> register status position which returns an increasing value from 0 that >>>>> indicates the current vpos/hpos for the display. >>>>> >>>>> For any vpos below vbl_start we know the value is correct since the next >>>>> vblank hasn't begun yet. But for anythign about vbl_start it's probably >>>>> wrong since it applies a corrective offset based on the fixed value of >>>>> crtc_vtotal. It's even worse when the value is above crtc_vtotal since >>>>> it'll be calculating the number of lines since the last scanout since >>>>> it'll be a positive value. >>>>> >>>>> So the issue becomes determining when the vfront porch will end. >>>>> >>>>> When the flip address gets written the vfront porch will end at the >>>>> start of the next line leaving only the back porch plus part of the >>>>> line. But we don't know when the flip will occur, if at all. It hasn't >>>>> occurred yet in this case. >>>>> >>>>> Waiting for the wrap around to 0 might be the best choice here since >>>>> there's no guarantee the flip will occur. >>>> >>>> I put some more thought into this and I don't think this is as bad as I >>>> had originally thought. >>>> >>>> I think the vblank timestamp is supposed to be for the first active >>>> pixel of the next scanout. The usage of which is for clients to time >>>> their content/animation/etc. >>>> >>>> The client likely doesn't care when they issue their flip, just that >>>> their content is matched for that vblank timestamp. The fixed refresh >>>> model works really well for that kind of application. >>>> >>>> Utilizing variable refresh rate would be a mistake in that case since >>>> the client would then have to time based on when they flip which is a >>>> lot harder to "predict" precisely. >>> >>> It's only a "mistake" as long as the timestamps are misleading. :) As >>> discussed before, accurate presentation timestamps are one requirement >>> for achieving perfectly smooth animation. >> >> For most 3D games the game world/rendering can advance by an arbitrary >> timestep - and faster will be smoother, which is what the native mode >> would give you. >> >> For fixed interval content/animation where you can't do that variable >> refresh rate could vastly improve the output. But like discussed before >> that would need a way to specify the interval/presentation timestamp to >> control that front porch duration. The timestamp will be misleading in >> any other case. > > I don't agree that an accurate timestamp can ever be more "misleading" > than an inaccurate one. But yeah, accurate timestamps and time-based > presentation are two sides of the same coin which can buy the holy grail > of perfectly smooth animation. :) > > >>>> I did some more investigation into when amdgpu gets the scanout position >>>> and what values we get back out of it. The timestamp is updated shortly >>>> after the crtc irq vblank which is typically within a few lines of >>>> vbl_start. This makes sense, since we can provide the prediction >>>> timestamp early. Waiting for the vblank to wrap around to 0 doesn't >>>> really make sense here since that would mean we already hit timeout or >>>> the flip occurred >>> >>> Sounds like you're mixing up the two cases of "actual" vblank events >>> (triggered by the "vblank" interrupt => drm_(crtc_)handle_vblank) and >>> flip completion events (triggered by the PFLIP interrupt with our >>> hardware => drm_crtc_send_vblank_event). >>> >>> Actual vblank events need to be delivered to userspace at the start of >>> vblank, so we indeed can't wait until the timestamp is accurate for >>> them. We just need to document the exact semantics of their timestamp >>> with VRR. >>> >>> For page flip completion events though, the timestamp needs to be >>> accurate and correspond to when the flipped frame starts being scanned >>> out, otherwise we'll surely break at least some userspace relying on >>> this information. >>> >> Yeah, I was. I guess what's sent is the estimated vblank timestamp >> calculated at the start of the interrupt. > > s/interrupt/vblank/, yeah. > > >> And since that's just a guess rather than what's actually going to >> happen it's going to be wrong in a lot of cases. >> >> I could see the wrap-around method working if the vblank timestamp was >> somehow updated in amdgpu or in drm_crtc_send_vblank_event. > > DC already calls drm_crtc_accurate_vblank_count before > drm_crtc_send_vblank_event, we "just" need to make sure that results in > an accurate timestamp. > > >> This would be a relatively simple fix but would break anything in >> userspace that relied on the timestamp for vblank interrupt and the >> flip completion being the same value. > > Hmm, that's a good point. So while VRR is enabled, maybe it is safer to > defer delivery of vblank events until the accurate timestamp is known as > well, at least by default. If there is userspace which needs the events > earlier even with VRR but can live with the guessed timestamp, a flag or > something could be added for that. > > I was under the impression that the vblank timestamp was reused but it's already going to differ since we call drm_crtc_accurate_vblank_count before drm_crtc_send_vblank_event. Thanks for pointing that out. Since that works for updating timestamp what's left is making sure that it waits for the wrap around if it's above crtc_vtotal. It might make sense to add a new flag for this that's only used within amdgpu_get_crtc_scanout_position so the other call sites aren't affected. There isn't a way to get an accurate timestamp with VRR enabled until after the flip happens. So deferring it kind of defeats the purpose of a client using it to make predictions before the flip for displaying their content. Nicholas Kazlauskas
On Wed, 31 Oct 2018 17:54:34 +0000 "Kazlauskas, Nicholas" <Nicholas.Kazlauskas@amd.com> wrote: > On 10/31/18 12:20 PM, Michel Dänzer wrote: > > On 2018-10-31 3:41 p.m., Kazlauskas, Nicholas wrote: > >> On 10/31/18 10:12 AM, Michel Dänzer wrote: > >>> On 2018-10-31 2:38 p.m., Kazlauskas, Nicholas wrote: > >>>> On 10/30/18 11:34 AM, Kazlauskas, Nicholas wrote: > >>>>> > >>>>> I understand the issue you're describing now. The timestamp is supposed > >>>>> to signify the end of the current vblank. The call to get scanout > >>>>> position is supposed to return the number of lines until scanout (a > >>>>> negative value) or the number of lines since the next scanout began (a > >>>>> positive value). > >>>>> > >>>>> The amdgpu driver calculates the number of lines based on a hardware > >>>>> register status position which returns an increasing value from 0 that > >>>>> indicates the current vpos/hpos for the display. > >>>>> > >>>>> For any vpos below vbl_start we know the value is correct since the next > >>>>> vblank hasn't begun yet. But for anythign about vbl_start it's probably > >>>>> wrong since it applies a corrective offset based on the fixed value of > >>>>> crtc_vtotal. It's even worse when the value is above crtc_vtotal since > >>>>> it'll be calculating the number of lines since the last scanout since > >>>>> it'll be a positive value. > >>>>> > >>>>> So the issue becomes determining when the vfront porch will end. > >>>>> > >>>>> When the flip address gets written the vfront porch will end at the > >>>>> start of the next line leaving only the back porch plus part of the > >>>>> line. But we don't know when the flip will occur, if at all. It hasn't > >>>>> occurred yet in this case. > >>>>> > >>>>> Waiting for the wrap around to 0 might be the best choice here since > >>>>> there's no guarantee the flip will occur. > >>>> > >>>> I put some more thought into this and I don't think this is as bad as I > >>>> had originally thought. > >>>> > >>>> I think the vblank timestamp is supposed to be for the first active > >>>> pixel of the next scanout. The usage of which is for clients to time > >>>> their content/animation/etc. > >>>> > >>>> The client likely doesn't care when they issue their flip, just that > >>>> their content is matched for that vblank timestamp. The fixed refresh > >>>> model works really well for that kind of application. > >>>> > >>>> Utilizing variable refresh rate would be a mistake in that case since > >>>> the client would then have to time based on when they flip which is a > >>>> lot harder to "predict" precisely. > >>> > >>> It's only a "mistake" as long as the timestamps are misleading. :) As > >>> discussed before, accurate presentation timestamps are one requirement > >>> for achieving perfectly smooth animation. > >> > >> For most 3D games the game world/rendering can advance by an arbitrary > >> timestep - and faster will be smoother, which is what the native mode > >> would give you. > >> > >> For fixed interval content/animation where you can't do that variable > >> refresh rate could vastly improve the output. But like discussed before > >> that would need a way to specify the interval/presentation timestamp to > >> control that front porch duration. The timestamp will be misleading in > >> any other case. > > > > I don't agree that an accurate timestamp can ever be more "misleading" > > than an inaccurate one. But yeah, accurate timestamps and time-based > > presentation are two sides of the same coin which can buy the holy grail > > of perfectly smooth animation. :) > > > > > >>>> I did some more investigation into when amdgpu gets the scanout position > >>>> and what values we get back out of it. The timestamp is updated shortly > >>>> after the crtc irq vblank which is typically within a few lines of > >>>> vbl_start. This makes sense, since we can provide the prediction > >>>> timestamp early. Waiting for the vblank to wrap around to 0 doesn't > >>>> really make sense here since that would mean we already hit timeout or > >>>> the flip occurred > >>> > >>> Sounds like you're mixing up the two cases of "actual" vblank events > >>> (triggered by the "vblank" interrupt => drm_(crtc_)handle_vblank) and > >>> flip completion events (triggered by the PFLIP interrupt with our > >>> hardware => drm_crtc_send_vblank_event). > >>> > >>> Actual vblank events need to be delivered to userspace at the start of > >>> vblank, so we indeed can't wait until the timestamp is accurate for > >>> them. We just need to document the exact semantics of their timestamp > >>> with VRR. > >>> > >>> For page flip completion events though, the timestamp needs to be > >>> accurate and correspond to when the flipped frame starts being scanned > >>> out, otherwise we'll surely break at least some userspace relying on > >>> this information. > >>> > >> Yeah, I was. I guess what's sent is the estimated vblank timestamp > >> calculated at the start of the interrupt. > > > > s/interrupt/vblank/, yeah. > > > > > >> And since that's just a guess rather than what's actually going to > >> happen it's going to be wrong in a lot of cases. > >> > >> I could see the wrap-around method working if the vblank timestamp was > >> somehow updated in amdgpu or in drm_crtc_send_vblank_event. > > > > DC already calls drm_crtc_accurate_vblank_count before > > drm_crtc_send_vblank_event, we "just" need to make sure that results in > > an accurate timestamp. > > > > > >> This would be a relatively simple fix but would break anything in > >> userspace that relied on the timestamp for vblank interrupt and the > >> flip completion being the same value. > > > > Hmm, that's a good point. So while VRR is enabled, maybe it is safer to > > defer delivery of vblank events until the accurate timestamp is known as > > well, at least by default. If there is userspace which needs the events > > earlier even with VRR but can live with the guessed timestamp, a flag or > > something could be added for that. > > > > > > I was under the impression that the vblank timestamp was reused but it's > already going to differ since we call drm_crtc_accurate_vblank_count > before drm_crtc_send_vblank_event. Thanks for pointing that out. > > Since that works for updating timestamp what's left is making sure that > it waits for the wrap around if it's above crtc_vtotal. It might make > sense to add a new flag for this that's only used within > amdgpu_get_crtc_scanout_position so the other call sites aren't affected. > > There isn't a way to get an accurate timestamp with VRR enabled until > after the flip happens. So deferring it kind of defeats the purpose of a > client using it to make predictions before the flip for displaying their > content. Hi, I don't think I understood half of the discussion above, but I'd like to offer how Weston uses vblank timestamps and pageflip timestamps today. When Weston starts rendering "from idle", which means it has lost track of the past flip timestamps since it has not been flipping, it uses drmWaitVBlank() to query the latest flip time non-blocking: drmVBlank vbl = { .request.type = DRM_VBLANK_RELATIVE, .request.sequence = 0, .request.signal = 0, }; If that returns a valid answer, it is taken as the base timestamp. (If it doesn't result in a valid timestamp, Weston makes a flip to the current FB and uses the realized pageflip timestamp of that as the base timestamp.) The base timestamp is used to predict the next possible flip time by adding one refresh period to it. This obviously depends heavily on fixed refresh rate. However, if userspace can know the VRR flip timeout length, I think it should be possible to compute the next possible flip window being between: base timestamp + minimum refresh period + { 0, timeout }. Of course, the prediction of the next possible flip window is only an estimate, so some inaccuracy is allowed. We still rely on the after-the-fact feedback in the form of pageflip timestamps to be accurate, so that we can potentially adjust the userspace timings to eventually hit what we predict. Implementing support for flip target timestamps would be merely an addition to all the above, that should improve the sub-refresh-period timing accuracy even further. But, to compute target timestamps, userspace would still need to predict the flip window to avoid being half a refresh period or more off on average. Therefore, if at all possible, I would like you to set the vblank and pageflip timestamp semantics such that this prediction is not fundamentally broken. Thanks, pq
On 2018-10-31 6:54 p.m., Kazlauskas, Nicholas wrote: > On 10/31/18 12:20 PM, Michel Dänzer wrote: >> On 2018-10-31 3:41 p.m., Kazlauskas, Nicholas wrote: >>> On 10/31/18 10:12 AM, Michel Dänzer wrote: >>>> On 2018-10-31 2:38 p.m., Kazlauskas, Nicholas wrote: >>>>> On 10/30/18 11:34 AM, Kazlauskas, Nicholas wrote: >>>>>> >>>>>> I understand the issue you're describing now. The timestamp is supposed >>>>>> to signify the end of the current vblank. The call to get scanout >>>>>> position is supposed to return the number of lines until scanout (a >>>>>> negative value) or the number of lines since the next scanout began (a >>>>>> positive value). >>>>>> >>>>>> The amdgpu driver calculates the number of lines based on a hardware >>>>>> register status position which returns an increasing value from 0 that >>>>>> indicates the current vpos/hpos for the display. >>>>>> >>>>>> For any vpos below vbl_start we know the value is correct since the next >>>>>> vblank hasn't begun yet. But for anythign about vbl_start it's probably >>>>>> wrong since it applies a corrective offset based on the fixed value of >>>>>> crtc_vtotal. It's even worse when the value is above crtc_vtotal since >>>>>> it'll be calculating the number of lines since the last scanout since >>>>>> it'll be a positive value. >>>>>> >>>>>> So the issue becomes determining when the vfront porch will end. >>>>>> >>>>>> When the flip address gets written the vfront porch will end at the >>>>>> start of the next line leaving only the back porch plus part of the >>>>>> line. But we don't know when the flip will occur, if at all. It hasn't >>>>>> occurred yet in this case. >>>>>> >>>>>> Waiting for the wrap around to 0 might be the best choice here since >>>>>> there's no guarantee the flip will occur. >>>>> >>>>> I put some more thought into this and I don't think this is as bad as I >>>>> had originally thought. >>>>> >>>>> I think the vblank timestamp is supposed to be for the first active >>>>> pixel of the next scanout. The usage of which is for clients to time >>>>> their content/animation/etc. >>>>> >>>>> The client likely doesn't care when they issue their flip, just that >>>>> their content is matched for that vblank timestamp. The fixed refresh >>>>> model works really well for that kind of application. >>>>> >>>>> Utilizing variable refresh rate would be a mistake in that case since >>>>> the client would then have to time based on when they flip which is a >>>>> lot harder to "predict" precisely. >>>> >>>> It's only a "mistake" as long as the timestamps are misleading. :) As >>>> discussed before, accurate presentation timestamps are one requirement >>>> for achieving perfectly smooth animation. >>> >>> For most 3D games the game world/rendering can advance by an arbitrary >>> timestep - and faster will be smoother, which is what the native mode >>> would give you. >>> >>> For fixed interval content/animation where you can't do that variable >>> refresh rate could vastly improve the output. But like discussed before >>> that would need a way to specify the interval/presentation timestamp to >>> control that front porch duration. The timestamp will be misleading in >>> any other case. >> >> I don't agree that an accurate timestamp can ever be more "misleading" >> than an inaccurate one. But yeah, accurate timestamps and time-based >> presentation are two sides of the same coin which can buy the holy grail >> of perfectly smooth animation. :) >> >> >>>>> I did some more investigation into when amdgpu gets the scanout position >>>>> and what values we get back out of it. The timestamp is updated shortly >>>>> after the crtc irq vblank which is typically within a few lines of >>>>> vbl_start. This makes sense, since we can provide the prediction >>>>> timestamp early. Waiting for the vblank to wrap around to 0 doesn't >>>>> really make sense here since that would mean we already hit timeout or >>>>> the flip occurred >>>> >>>> Sounds like you're mixing up the two cases of "actual" vblank events >>>> (triggered by the "vblank" interrupt => drm_(crtc_)handle_vblank) and >>>> flip completion events (triggered by the PFLIP interrupt with our >>>> hardware => drm_crtc_send_vblank_event). >>>> >>>> Actual vblank events need to be delivered to userspace at the start of >>>> vblank, so we indeed can't wait until the timestamp is accurate for >>>> them. We just need to document the exact semantics of their timestamp >>>> with VRR. >>>> >>>> For page flip completion events though, the timestamp needs to be >>>> accurate and correspond to when the flipped frame starts being scanned >>>> out, otherwise we'll surely break at least some userspace relying on >>>> this information. >>>> >>> Yeah, I was. I guess what's sent is the estimated vblank timestamp >>> calculated at the start of the interrupt. >> >> s/interrupt/vblank/, yeah. >> >> >>> And since that's just a guess rather than what's actually going to >>> happen it's going to be wrong in a lot of cases. >>> >>> I could see the wrap-around method working if the vblank timestamp was >>> somehow updated in amdgpu or in drm_crtc_send_vblank_event. >> >> DC already calls drm_crtc_accurate_vblank_count before >> drm_crtc_send_vblank_event, we "just" need to make sure that results in >> an accurate timestamp. >> >> >>> This would be a relatively simple fix but would break anything in >>> userspace that relied on the timestamp for vblank interrupt and the >>> flip completion being the same value. >> >> Hmm, that's a good point. So while VRR is enabled, maybe it is safer to >> defer delivery of vblank events until the accurate timestamp is known as >> well, at least by default. If there is userspace which needs the events >> earlier even with VRR but can live with the guessed timestamp, a flag or >> something could be added for that. > > I was under the impression that the vblank timestamp was reused but it's > already going to differ since we call drm_crtc_accurate_vblank_count > before drm_crtc_send_vblank_event. Thanks for pointing that out. > > Since that works for updating timestamp what's left is making sure that > it waits for the wrap around if it's above crtc_vtotal. It might make > sense to add a new flag for this that's only used within > amdgpu_get_crtc_scanout_position so the other call sites aren't affected. > > There isn't a way to get an accurate timestamp with VRR enabled until > after the flip happens. So deferring it kind of defeats the purpose of a > client using it to make predictions before the flip for displaying their > content. That's generally not a valid use-case. The KMS API is defined such that if userspace receives the vblank event for vblank period N and then submits a page flip, the page flip cannot (normally[0]) take effect before vblank period N+1. There are mainly two valid use-cases for waiting for vblank: 1. Wait for vblank N, then submit page flip for vblank N+1. 2. Wait for vblank N, then try to do something before vblank N ends, e.g. draw directly to the front buffer, as a poor man's way of avoiding tearing or similar artifacts. Use-case 1 is used by Xorg (both for Present and DRI2) and at least Weston, but presumably most if not all Wayland compositors. From Pekka's description of how Weston uses it (thanks!), it's clear that the timestamp of vblank events does need to accurately correspond to the end of the vblank period. Use-case 2 is also used by Xorg, but only when not flipping, which rules out VRR. Also, I'd argue that blocking waits for vblank are more suitable for this use-case than vblank events. Therefore, in summary I propose this is how it could work: * Without VRR, timestamps continue to work the same way they do now. * With VRR, delivery of both vblank and page flip completion events must be deferred until the timestamp accurately corresponds to the actual end of vblank. For blocking vblank waits, I'm not sure, but I'm leaning towards continuing to return to userspace at the beginning of vblank by default, even if the timestamp is inaccurate. Deferring delivery of events should avoid busy-loops. Bonus points for minimizing the number of interrupts needed. :) Note that both blocking vblank waits and vblank events are currently triggered from drm_(crtc_)handle_vblank, so some rework would be required for this. It's possible that there's other userspace which doesn't care about the accuracy of the timestamps, but cares about getting events as early as possible. In that case, flags might need to be added. [0] Our drivers can complete the flip in the same vblank period under some circumstances if userspace specifies it as the target using the DRM_MODE_PAGE_FLIP_TARGET_ABSOLUTE/RELATIVE flag, but there's no guarantee for it to work. And no other driver supports this yet. Without those flags, the flip cannot complete in the same vblank period, or it would break existing userspace using use-case 1 above.
On 11/1/18 6:58 AM, Michel Dänzer wrote: > On 2018-10-31 6:54 p.m., Kazlauskas, Nicholas wrote: >> On 10/31/18 12:20 PM, Michel Dänzer wrote: >>> On 2018-10-31 3:41 p.m., Kazlauskas, Nicholas wrote: >>>> On 10/31/18 10:12 AM, Michel Dänzer wrote: >>>>> On 2018-10-31 2:38 p.m., Kazlauskas, Nicholas wrote: >>>>>> On 10/30/18 11:34 AM, Kazlauskas, Nicholas wrote: >>>>>>> >>>>>>> I understand the issue you're describing now. The timestamp is supposed >>>>>>> to signify the end of the current vblank. The call to get scanout >>>>>>> position is supposed to return the number of lines until scanout (a >>>>>>> negative value) or the number of lines since the next scanout began (a >>>>>>> positive value). >>>>>>> >>>>>>> The amdgpu driver calculates the number of lines based on a hardware >>>>>>> register status position which returns an increasing value from 0 that >>>>>>> indicates the current vpos/hpos for the display. >>>>>>> >>>>>>> For any vpos below vbl_start we know the value is correct since the next >>>>>>> vblank hasn't begun yet. But for anythign about vbl_start it's probably >>>>>>> wrong since it applies a corrective offset based on the fixed value of >>>>>>> crtc_vtotal. It's even worse when the value is above crtc_vtotal since >>>>>>> it'll be calculating the number of lines since the last scanout since >>>>>>> it'll be a positive value. >>>>>>> >>>>>>> So the issue becomes determining when the vfront porch will end. >>>>>>> >>>>>>> When the flip address gets written the vfront porch will end at the >>>>>>> start of the next line leaving only the back porch plus part of the >>>>>>> line. But we don't know when the flip will occur, if at all. It hasn't >>>>>>> occurred yet in this case. >>>>>>> >>>>>>> Waiting for the wrap around to 0 might be the best choice here since >>>>>>> there's no guarantee the flip will occur. >>>>>> >>>>>> I put some more thought into this and I don't think this is as bad as I >>>>>> had originally thought. >>>>>> >>>>>> I think the vblank timestamp is supposed to be for the first active >>>>>> pixel of the next scanout. The usage of which is for clients to time >>>>>> their content/animation/etc. >>>>>> >>>>>> The client likely doesn't care when they issue their flip, just that >>>>>> their content is matched for that vblank timestamp. The fixed refresh >>>>>> model works really well for that kind of application. >>>>>> >>>>>> Utilizing variable refresh rate would be a mistake in that case since >>>>>> the client would then have to time based on when they flip which is a >>>>>> lot harder to "predict" precisely. >>>>> >>>>> It's only a "mistake" as long as the timestamps are misleading. :) As >>>>> discussed before, accurate presentation timestamps are one requirement >>>>> for achieving perfectly smooth animation. >>>> >>>> For most 3D games the game world/rendering can advance by an arbitrary >>>> timestep - and faster will be smoother, which is what the native mode >>>> would give you. >>>> >>>> For fixed interval content/animation where you can't do that variable >>>> refresh rate could vastly improve the output. But like discussed before >>>> that would need a way to specify the interval/presentation timestamp to >>>> control that front porch duration. The timestamp will be misleading in >>>> any other case. >>> >>> I don't agree that an accurate timestamp can ever be more "misleading" >>> than an inaccurate one. But yeah, accurate timestamps and time-based >>> presentation are two sides of the same coin which can buy the holy grail >>> of perfectly smooth animation. :) >>> >>> >>>>>> I did some more investigation into when amdgpu gets the scanout position >>>>>> and what values we get back out of it. The timestamp is updated shortly >>>>>> after the crtc irq vblank which is typically within a few lines of >>>>>> vbl_start. This makes sense, since we can provide the prediction >>>>>> timestamp early. Waiting for the vblank to wrap around to 0 doesn't >>>>>> really make sense here since that would mean we already hit timeout or >>>>>> the flip occurred >>>>> >>>>> Sounds like you're mixing up the two cases of "actual" vblank events >>>>> (triggered by the "vblank" interrupt => drm_(crtc_)handle_vblank) and >>>>> flip completion events (triggered by the PFLIP interrupt with our >>>>> hardware => drm_crtc_send_vblank_event). >>>>> >>>>> Actual vblank events need to be delivered to userspace at the start of >>>>> vblank, so we indeed can't wait until the timestamp is accurate for >>>>> them. We just need to document the exact semantics of their timestamp >>>>> with VRR. >>>>> >>>>> For page flip completion events though, the timestamp needs to be >>>>> accurate and correspond to when the flipped frame starts being scanned >>>>> out, otherwise we'll surely break at least some userspace relying on >>>>> this information. >>>>> >>>> Yeah, I was. I guess what's sent is the estimated vblank timestamp >>>> calculated at the start of the interrupt. >>> >>> s/interrupt/vblank/, yeah. >>> >>> >>>> And since that's just a guess rather than what's actually going to >>>> happen it's going to be wrong in a lot of cases. >>>> >>>> I could see the wrap-around method working if the vblank timestamp was >>>> somehow updated in amdgpu or in drm_crtc_send_vblank_event. >>> >>> DC already calls drm_crtc_accurate_vblank_count before >>> drm_crtc_send_vblank_event, we "just" need to make sure that results in >>> an accurate timestamp. >>> >>> >>>> This would be a relatively simple fix but would break anything in >>>> userspace that relied on the timestamp for vblank interrupt and the >>>> flip completion being the same value. >>> >>> Hmm, that's a good point. So while VRR is enabled, maybe it is safer to >>> defer delivery of vblank events until the accurate timestamp is known as >>> well, at least by default. If there is userspace which needs the events >>> earlier even with VRR but can live with the guessed timestamp, a flag or >>> something could be added for that. >> >> I was under the impression that the vblank timestamp was reused but it's >> already going to differ since we call drm_crtc_accurate_vblank_count >> before drm_crtc_send_vblank_event. Thanks for pointing that out. >> >> Since that works for updating timestamp what's left is making sure that >> it waits for the wrap around if it's above crtc_vtotal. It might make >> sense to add a new flag for this that's only used within >> amdgpu_get_crtc_scanout_position so the other call sites aren't affected. >> >> There isn't a way to get an accurate timestamp with VRR enabled until >> after the flip happens. So deferring it kind of defeats the purpose of a >> client using it to make predictions before the flip for displaying their >> content. > > That's generally not a valid use-case. The KMS API is defined such that > if userspace receives the vblank event for vblank period N and then > submits a page flip, the page flip cannot (normally[0]) take effect > before vblank period N+1. > > > There are mainly two valid use-cases for waiting for vblank: > > 1. Wait for vblank N, then submit page flip for vblank N+1. > > 2. Wait for vblank N, then try to do something before vblank N ends, > e.g. draw directly to the front buffer, as a poor man's way of avoiding > tearing or similar artifacts. > > > Use-case 1 is used by Xorg (both for Present and DRI2) and at least > Weston, but presumably most if not all Wayland compositors. From Pekka's > description of how Weston uses it (thanks!), it's clear that the > timestamp of vblank events does need to accurately correspond to the end > of the vblank period. > > Use-case 2 is also used by Xorg, but only when not flipping, which rules > out VRR. Also, I'd argue that blocking waits for vblank are more > suitable for this use-case than vblank events. > > Therefore, in summary I propose this is how it could work: > > * Without VRR, timestamps continue to work the same way they do now. > > * With VRR, delivery of both vblank and page flip completion events must > be deferred until the timestamp accurately corresponds to the actual end > of vblank. For blocking vblank waits, I'm not sure, but I'm leaning > towards continuing to return to userspace at the beginning of vblank by > default, even if the timestamp is inaccurate. > > Deferring delivery of events should avoid busy-loops. Bonus points for > minimizing the number of interrupts needed. :) > > Note that both blocking vblank waits and vblank events are currently > triggered from drm_(crtc_)handle_vblank, so some rework would be > required for this. > > It's possible that there's other userspace which doesn't care about the > accuracy of the timestamps, but cares about getting events as early as > possible. In that case, flags might need to be added. > > > [0] Our drivers can complete the flip in the same vblank period under > some circumstances if userspace specifies it as the target using the > DRM_MODE_PAGE_FLIP_TARGET_ABSOLUTE/RELATIVE flag, but there's no > guarantee for it to work. And no other driver supports this yet. Without > those flags, the flip cannot complete in the same vblank period, or it > would break existing userspace using use-case 1 above. > Deferring the pageflip timestamp is something we can do today in the driver with no impact on existing behavior/semantics. Since hpos is used in the calculation as well I think we don't need even need to wait for the roll-over necessarily. Deferring the event that comes from drm_handle_vblank worries me though. The implementation for that would be sending that event on the roll-over at timeout or grouping it together on the pageflip. Those are pretty long delays for both. To me it still makes more sense to just define the vblank timestamp coming out of drm_crtc_handle_vblank as an estimate when VRR is enabled. An estimation which ends up being the same value as what userspace gets today, the estimation for the current mode. I think my opinion would be different if the timestamps were supposed to be the same value but that's incredibly unlikely given that they're offset from the ktime_get in most cases. Nicholas Kazlauskas
On 2018-11-01 3:58 p.m., Kazlauskas, Nicholas wrote: > On 11/1/18 6:58 AM, Michel Dänzer wrote: >> On 2018-10-31 6:54 p.m., Kazlauskas, Nicholas wrote: >>> On 10/31/18 12:20 PM, Michel Dänzer wrote: >>>> On 2018-10-31 3:41 p.m., Kazlauskas, Nicholas wrote: >>>>> On 10/31/18 10:12 AM, Michel Dänzer wrote: >>>>>> On 2018-10-31 2:38 p.m., Kazlauskas, Nicholas wrote: >>>>>>> >>>>>>> I did some more investigation into when amdgpu gets the scanout position >>>>>>> and what values we get back out of it. The timestamp is updated shortly >>>>>>> after the crtc irq vblank which is typically within a few lines of >>>>>>> vbl_start. This makes sense, since we can provide the prediction >>>>>>> timestamp early. Waiting for the vblank to wrap around to 0 doesn't >>>>>>> really make sense here since that would mean we already hit timeout or >>>>>>> the flip occurred >>>>>> >>>>>> Sounds like you're mixing up the two cases of "actual" vblank events >>>>>> (triggered by the "vblank" interrupt => drm_(crtc_)handle_vblank) and >>>>>> flip completion events (triggered by the PFLIP interrupt with our >>>>>> hardware => drm_crtc_send_vblank_event). >>>>>> >>>>>> Actual vblank events need to be delivered to userspace at the start of >>>>>> vblank, so we indeed can't wait until the timestamp is accurate for >>>>>> them. We just need to document the exact semantics of their timestamp >>>>>> with VRR. >>>>>> >>>>>> For page flip completion events though, the timestamp needs to be >>>>>> accurate and correspond to when the flipped frame starts being scanned >>>>>> out, otherwise we'll surely break at least some userspace relying on >>>>>> this information. >>>>>> >>>>> Yeah, I was. I guess what's sent is the estimated vblank timestamp >>>>> calculated at the start of the interrupt. >>>> >>>> s/interrupt/vblank/, yeah. >>>> >>>> >>>>> And since that's just a guess rather than what's actually going to >>>>> happen it's going to be wrong in a lot of cases. >>>>> >>>>> I could see the wrap-around method working if the vblank timestamp was >>>>> somehow updated in amdgpu or in drm_crtc_send_vblank_event. >>>> >>>> DC already calls drm_crtc_accurate_vblank_count before >>>> drm_crtc_send_vblank_event, we "just" need to make sure that results in >>>> an accurate timestamp. >>>> >>>> >>>>> This would be a relatively simple fix but would break anything in >>>>> userspace that relied on the timestamp for vblank interrupt and the >>>>> flip completion being the same value. >>>> >>>> Hmm, that's a good point. So while VRR is enabled, maybe it is safer to >>>> defer delivery of vblank events until the accurate timestamp is known as >>>> well, at least by default. If there is userspace which needs the events >>>> earlier even with VRR but can live with the guessed timestamp, a flag or >>>> something could be added for that. >>> >>> I was under the impression that the vblank timestamp was reused but it's >>> already going to differ since we call drm_crtc_accurate_vblank_count >>> before drm_crtc_send_vblank_event. Thanks for pointing that out. >>> >>> Since that works for updating timestamp what's left is making sure that >>> it waits for the wrap around if it's above crtc_vtotal. It might make >>> sense to add a new flag for this that's only used within >>> amdgpu_get_crtc_scanout_position so the other call sites aren't affected. >>> >>> There isn't a way to get an accurate timestamp with VRR enabled until >>> after the flip happens. So deferring it kind of defeats the purpose of a >>> client using it to make predictions before the flip for displaying their >>> content. >> >> That's generally not a valid use-case. The KMS API is defined such that >> if userspace receives the vblank event for vblank period N and then >> submits a page flip, the page flip cannot (normally[0]) take effect >> before vblank period N+1. >> >> >> There are mainly two valid use-cases for waiting for vblank: >> >> 1. Wait for vblank N, then submit page flip for vblank N+1. >> >> 2. Wait for vblank N, then try to do something before vblank N ends, >> e.g. draw directly to the front buffer, as a poor man's way of avoiding >> tearing or similar artifacts. >> >> >> Use-case 1 is used by Xorg (both for Present and DRI2) and at least >> Weston, but presumably most if not all Wayland compositors. From Pekka's >> description of how Weston uses it (thanks!), it's clear that the >> timestamp of vblank events does need to accurately correspond to the end >> of the vblank period. >> >> Use-case 2 is also used by Xorg, but only when not flipping, which rules >> out VRR. Also, I'd argue that blocking waits for vblank are more >> suitable for this use-case than vblank events. >> >> Therefore, in summary I propose this is how it could work: >> >> * Without VRR, timestamps continue to work the same way they do now. >> >> * With VRR, delivery of both vblank and page flip completion events must >> be deferred until the timestamp accurately corresponds to the actual end >> of vblank. For blocking vblank waits, I'm not sure, but I'm leaning >> towards continuing to return to userspace at the beginning of vblank by >> default, even if the timestamp is inaccurate. >> >> Deferring delivery of events should avoid busy-loops. Bonus points for >> minimizing the number of interrupts needed. :) >> >> Note that both blocking vblank waits and vblank events are currently >> triggered from drm_(crtc_)handle_vblank, so some rework would be >> required for this. >> >> It's possible that there's other userspace which doesn't care about the >> accuracy of the timestamps, but cares about getting events as early as >> possible. In that case, flags might need to be added. >> >> >> [0] Our drivers can complete the flip in the same vblank period under >> some circumstances if userspace specifies it as the target using the >> DRM_MODE_PAGE_FLIP_TARGET_ABSOLUTE/RELATIVE flag, but there's no >> guarantee for it to work. And no other driver supports this yet. Without >> those flags, the flip cannot complete in the same vblank period, or it >> would break existing userspace using use-case 1 above. >> > Deferring the pageflip timestamp is something we can do today in the > driver with no impact on existing behavior/semantics. Since hpos is used > in the calculation as well I think we don't need even need to wait for > the roll-over necessarily. Not sure offhand how hpos helps for this, but I'll trust your word for it. :) > Deferring the event that comes from drm_handle_vblank worries me though. > The implementation for that would be sending that event on the roll-over > at timeout or grouping it together on the pageflip. Those are pretty > long delays for both. > > To me it still makes more sense to just define the vblank timestamp > coming out of drm_crtc_handle_vblank as an estimate when VRR is enabled. > An estimation which ends up being the same value as what userspace gets > today, the estimation for the current mode. > > I think my opinion would be different if the timestamps were supposed to > be the same value but that's incredibly unlikely given that they're > offset from the ktime_get in most cases. Right now, without VRR, they're supposed to be the same within measuring accuracy. If you're seeing bigger differences than that, can you provide some examples? You might have discovered a bug. Anyway, I re-read Pekka's description, and it sounds like Weston only uses the vblank timestamp if no flip completed in the last vblank period, which normally shouldn't happen with VRR. Otherwise it uses the timestamp from the flip completion event. So, it sounds like having different timestamps between the two cases should be fine for Weston after all. Sorry for the confusion above. So, I'm okay with trying to change only the timestamps of the flip completion events. That'll certainly make things easier. :)
On 11/1/18 1:05 PM, Michel Dänzer wrote: > On 2018-11-01 3:58 p.m., Kazlauskas, Nicholas wrote: >> On 11/1/18 6:58 AM, Michel Dänzer wrote: >>> On 2018-10-31 6:54 p.m., Kazlauskas, Nicholas wrote: >>>> On 10/31/18 12:20 PM, Michel Dänzer wrote: >>>>> On 2018-10-31 3:41 p.m., Kazlauskas, Nicholas wrote: >>>>>> On 10/31/18 10:12 AM, Michel Dänzer wrote: >>>>>>> On 2018-10-31 2:38 p.m., Kazlauskas, Nicholas wrote: >>>>>>>> >>>>>>>> I did some more investigation into when amdgpu gets the scanout position >>>>>>>> and what values we get back out of it. The timestamp is updated shortly >>>>>>>> after the crtc irq vblank which is typically within a few lines of >>>>>>>> vbl_start. This makes sense, since we can provide the prediction >>>>>>>> timestamp early. Waiting for the vblank to wrap around to 0 doesn't >>>>>>>> really make sense here since that would mean we already hit timeout or >>>>>>>> the flip occurred >>>>>>> >>>>>>> Sounds like you're mixing up the two cases of "actual" vblank events >>>>>>> (triggered by the "vblank" interrupt => drm_(crtc_)handle_vblank) and >>>>>>> flip completion events (triggered by the PFLIP interrupt with our >>>>>>> hardware => drm_crtc_send_vblank_event). >>>>>>> >>>>>>> Actual vblank events need to be delivered to userspace at the start of >>>>>>> vblank, so we indeed can't wait until the timestamp is accurate for >>>>>>> them. We just need to document the exact semantics of their timestamp >>>>>>> with VRR. >>>>>>> >>>>>>> For page flip completion events though, the timestamp needs to be >>>>>>> accurate and correspond to when the flipped frame starts being scanned >>>>>>> out, otherwise we'll surely break at least some userspace relying on >>>>>>> this information. >>>>>>> >>>>>> Yeah, I was. I guess what's sent is the estimated vblank timestamp >>>>>> calculated at the start of the interrupt. >>>>> >>>>> s/interrupt/vblank/, yeah. >>>>> >>>>> >>>>>> And since that's just a guess rather than what's actually going to >>>>>> happen it's going to be wrong in a lot of cases. >>>>>> >>>>>> I could see the wrap-around method working if the vblank timestamp was >>>>>> somehow updated in amdgpu or in drm_crtc_send_vblank_event. >>>>> >>>>> DC already calls drm_crtc_accurate_vblank_count before >>>>> drm_crtc_send_vblank_event, we "just" need to make sure that results in >>>>> an accurate timestamp. >>>>> >>>>> >>>>>> This would be a relatively simple fix but would break anything in >>>>>> userspace that relied on the timestamp for vblank interrupt and the >>>>>> flip completion being the same value. >>>>> >>>>> Hmm, that's a good point. So while VRR is enabled, maybe it is safer to >>>>> defer delivery of vblank events until the accurate timestamp is known as >>>>> well, at least by default. If there is userspace which needs the events >>>>> earlier even with VRR but can live with the guessed timestamp, a flag or >>>>> something could be added for that. >>>> >>>> I was under the impression that the vblank timestamp was reused but it's >>>> already going to differ since we call drm_crtc_accurate_vblank_count >>>> before drm_crtc_send_vblank_event. Thanks for pointing that out. >>>> >>>> Since that works for updating timestamp what's left is making sure that >>>> it waits for the wrap around if it's above crtc_vtotal. It might make >>>> sense to add a new flag for this that's only used within >>>> amdgpu_get_crtc_scanout_position so the other call sites aren't affected. >>>> >>>> There isn't a way to get an accurate timestamp with VRR enabled until >>>> after the flip happens. So deferring it kind of defeats the purpose of a >>>> client using it to make predictions before the flip for displaying their >>>> content. >>> >>> That's generally not a valid use-case. The KMS API is defined such that >>> if userspace receives the vblank event for vblank period N and then >>> submits a page flip, the page flip cannot (normally[0]) take effect >>> before vblank period N+1. >>> >>> >>> There are mainly two valid use-cases for waiting for vblank: >>> >>> 1. Wait for vblank N, then submit page flip for vblank N+1. >>> >>> 2. Wait for vblank N, then try to do something before vblank N ends, >>> e.g. draw directly to the front buffer, as a poor man's way of avoiding >>> tearing or similar artifacts. >>> >>> >>> Use-case 1 is used by Xorg (both for Present and DRI2) and at least >>> Weston, but presumably most if not all Wayland compositors. From Pekka's >>> description of how Weston uses it (thanks!), it's clear that the >>> timestamp of vblank events does need to accurately correspond to the end >>> of the vblank period. >>> >>> Use-case 2 is also used by Xorg, but only when not flipping, which rules >>> out VRR. Also, I'd argue that blocking waits for vblank are more >>> suitable for this use-case than vblank events. >>> >>> Therefore, in summary I propose this is how it could work: >>> >>> * Without VRR, timestamps continue to work the same way they do now. >>> >>> * With VRR, delivery of both vblank and page flip completion events must >>> be deferred until the timestamp accurately corresponds to the actual end >>> of vblank. For blocking vblank waits, I'm not sure, but I'm leaning >>> towards continuing to return to userspace at the beginning of vblank by >>> default, even if the timestamp is inaccurate. >>> >>> Deferring delivery of events should avoid busy-loops. Bonus points for >>> minimizing the number of interrupts needed. :) >>> >>> Note that both blocking vblank waits and vblank events are currently >>> triggered from drm_(crtc_)handle_vblank, so some rework would be >>> required for this. >>> >>> It's possible that there's other userspace which doesn't care about the >>> accuracy of the timestamps, but cares about getting events as early as >>> possible. In that case, flags might need to be added. >>> >>> >>> [0] Our drivers can complete the flip in the same vblank period under >>> some circumstances if userspace specifies it as the target using the >>> DRM_MODE_PAGE_FLIP_TARGET_ABSOLUTE/RELATIVE flag, but there's no >>> guarantee for it to work. And no other driver supports this yet. Without >>> those flags, the flip cannot complete in the same vblank period, or it >>> would break existing userspace using use-case 1 above. >>> >> Deferring the pageflip timestamp is something we can do today in the >> driver with no impact on existing behavior/semantics. Since hpos is used >> in the calculation as well I think we don't need even need to wait for >> the roll-over necessarily. > > Not sure offhand how hpos helps for this, but I'll trust your word for > it. :) If we're still in the vfront porch when on the pageflip irq timestamp query then it should be going over to the back porch on the next line. So there's still the full back porch to go through, plus a little bit of the current line horizontally. That horizontal bit is included in the timestamp calculation already so it should be fine. > > >> Deferring the event that comes from drm_handle_vblank worries me though. >> The implementation for that would be sending that event on the roll-over >> at timeout or grouping it together on the pageflip. Those are pretty >> long delays for both. >> >> To me it still makes more sense to just define the vblank timestamp >> coming out of drm_crtc_handle_vblank as an estimate when VRR is enabled. >> An estimation which ends up being the same value as what userspace gets >> today, the estimation for the current mode. >> >> I think my opinion would be different if the timestamps were supposed to >> be the same value but that's incredibly unlikely given that they're >> offset from the ktime_get in most cases. > > Right now, without VRR, they're supposed to be the same within measuring > accuracy. If you're seeing bigger differences than that, can you provide > some examples? You might have discovered a bug. There's a little bit of uncertainty since it's two separate calls to get_crtc_scanoutpos and two different calls to ktime_get for the end time. The numbers differ but not by a significant amount. I wouldn't consider it a bug at least. > > > Anyway, I re-read Pekka's description, and it sounds like Weston only > uses the vblank timestamp if no flip completed in the last vblank > period, which normally shouldn't happen with VRR. Otherwise it uses the > timestamp from the flip completion event. > > So, it sounds like having different timestamps between the two cases > should be fine for Weston after all. Sorry for the confusion above. > > So, I'm okay with trying to change only the timestamps of the flip > completion events. That'll certainly make things easier. :) > > Sounds good. The only change needed to handle this in the pageflip is a small bugfix for the case where vpos >= crtc_vtotal on the pageflip query. I'll include it as part of the next patch series with the updated documentation. Nicholas Kazlauskas
diff --git a/Documentation/gpu/drm-kms.rst b/Documentation/gpu/drm-kms.rst index 4b1501b4835b..8da2a178cf85 100644 --- a/Documentation/gpu/drm-kms.rst +++ b/Documentation/gpu/drm-kms.rst @@ -575,6 +575,13 @@ Explicit Fencing Properties .. kernel-doc:: drivers/gpu/drm/drm_atomic_uapi.c :doc: explicit fencing properties + +Variable Refresh Properties +--------------------------- + +.. kernel-doc:: drivers/gpu/drm/drm_connector.c + :doc: Variable refresh properties + Existing KMS Properties ----------------------- diff --git a/drivers/gpu/drm/drm_connector.c b/drivers/gpu/drm/drm_connector.c index f0deeb7298d0..2a12853ca917 100644 --- a/drivers/gpu/drm/drm_connector.c +++ b/drivers/gpu/drm/drm_connector.c @@ -1254,6 +1254,28 @@ int drm_mode_create_scaling_mode_property(struct drm_device *dev) } EXPORT_SYMBOL(drm_mode_create_scaling_mode_property); +/** + * DOC: Variable refresh properties + * + * Variable refresh rate control is supported via properties on the + * &drm_connector and &drm_crtc objects. + * + * "vrr_capable": + * Optional &drm_connector boolean property that drivers should attach + * with drm_connector_attach_vrr_capable_property() on connectors that + * could support variable refresh rates. Drivers should update the + * property value by calling drm_connector_set_vrr_capable_property(). + * + * Absence of the property should indicate absence of support. + * + * "vrr_enabled": + * Default &drm_crtc boolean property that notifies the driver that the + * variable refresh rate adjustment should be enabled for the CRTC. + * + * Support for variable refresh rate will depend on the "vrr_capable" + * property exposed on the &drm_connector object. + */ + /** * drm_connector_attach_vrr_capable_property - creates the * vrr_capable property
These include the drm_connector 'vrr_capable' and the drm_crtc 'vrr_enabled' properties. Signed-off-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> --- Documentation/gpu/drm-kms.rst | 7 +++++++ drivers/gpu/drm/drm_connector.c | 22 ++++++++++++++++++++++ 2 files changed, 29 insertions(+)