Re: [Intel-gfx] [PATCH v2 rebased 06/11] drm/i915/display: Share intel_connector_needs_modeset()

["Ville Syrjälä" <ville.syrjala@linux.intel.com>]

On Fri, Dec 13, 2019 at 04:14:55PM -0800, Lucas De Marchi wrote:
> On Thu, Dec 12, 2019 at 05:52:49PM +0200, Ville Syrjälä wrote:
> >On Wed, Dec 11, 2019 at 10:45:21AM -0800, José Roberto de Souza wrote:
> >> intel_connector_needs_modeset() will be used outside of
> >> intel_display.c in a future patch so it would only be necessary to
> >> remove the state and add the prototype to the header file.
> >>
> >> But while at it, I simplified the arguments and changed to intel
> >> types and moved it to a better place intel_atomic.c.
> >>
> >> That allowed us to convert the whole
> >> intel_encoders_update_prepare/complete to intel type too.
> >>
> >> No behavior changes intended here.
> >>
> >> Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
> >> Signed-off-by: José Roberto de Souza <jose.souza@intel.com>
> >> ---
> >>  drivers/gpu/drm/i915/display/intel_atomic.c  | 32 ++++++++++++
> >>  drivers/gpu/drm/i915/display/intel_atomic.h  |  3 ++
> >>  drivers/gpu/drm/i915/display/intel_display.c | 53 ++++++--------------
> >>  3 files changed, 51 insertions(+), 37 deletions(-)
> >>
> >> diff --git a/drivers/gpu/drm/i915/display/intel_atomic.c b/drivers/gpu/drm/i915/display/intel_atomic.c
> >> index fd0026fc3618..6e93a39a6fec 100644
> >> --- a/drivers/gpu/drm/i915/display/intel_atomic.c
> >> +++ b/drivers/gpu/drm/i915/display/intel_atomic.c
> >> @@ -174,6 +174,38 @@ intel_digital_connector_duplicate_state(struct drm_connector *connector)
> >>  	return &state->base;
> >>  }
> >>
> >> +/**
> >> + * intel_digital_connector_needs_modeset - check if connector needs a modeset
> >> + */
> >> +bool
> >> +intel_digital_connector_needs_modeset(struct intel_atomic_state *state,
> >
> >Why "digital"? Oh because intel_atomic_get_old_connector_state() return
> >a ditgital_connector_state. A bit surprising.
> >
> >I suggest using just drm_connector_state here to keep this function
> >totally generic.
> >
> >> +				      struct intel_connector *connector)
> >> +{
> >> +	struct intel_digital_connector_state *old_connector_state, *new_connector_state;
> >> +	struct intel_crtc *old_crtc, *new_crtc;
> >> +	struct intel_crtc_state *new_crtc_state;
> >> +
> >> +	old_connector_state = intel_atomic_get_old_connector_state(state,
> >> +								   connector);
> >
> >Could be done when declaring the variable. Dunno which is prettier
> >though.
> >
> >> +	if (old_connector_state->base.crtc)
> >> +		old_crtc = to_intel_crtc(old_connector_state->base.crtc);
> >> +	else
> >> +		old_crtc = NULL;
> >
> >Simple
> >old_crtc = to_intel_crtc(old_connector_state->base.crtc);
> >will do. Can be done when declaring the variable as well.
> >
> >> +
> >> +	new_connector_state = intel_atomic_get_new_connector_state(state,
> >> +								   connector);
> >> +	if (new_connector_state->base.crtc) {
> >> +		new_crtc = to_intel_crtc(new_connector_state->base.crtc);
> >
> >ditto.
> >
> >> +		new_crtc_state = intel_atomic_get_new_crtc_state(state, new_crtc);
> >
> >Then this just becomes
> >if (new_crtc)
> >	new_crtc_state = ...;
> >
> >Or maybe
> >new_crtc_state = new_crtc ? get : NULL;
> >but that could be a bit ugly.
> >
> >
> >> +	} else {
> >> +		new_crtc_state = NULL;
> >> +		new_crtc = NULL;
> >> +	}
> >> +
> >> +	return new_crtc != old_crtc ||
> >> +	       (new_crtc && drm_atomic_crtc_needs_modeset(&new_crtc_state->uapi));
> >
> >Hmm. In fact this function could be one of those special cases where we
> >might even want to use all drm_ types internally since we don't actually
> >need anything else.
> 
> so... do you mean to bring intel_connector_needs_modeset() as is?

Maybe... Yeah, looks more useful as is.

> >> -static bool
> >> -intel_connector_needs_modeset(struct intel_atomic_state *state,
> >> -			      const struct drm_connector_state *old_conn_state,
> >> -			      const struct drm_connector_state *new_conn_state)
> >> -{
> >> -	struct intel_crtc *old_crtc = old_conn_state->crtc ?
> >> -				      to_intel_crtc(old_conn_state->crtc) : NULL;
> >> -	struct intel_crtc *new_crtc = new_conn_state->crtc ?
> >> -				      to_intel_crtc(new_conn_state->crtc) : NULL;

You could toss out those ternary operators while at it. They're not needed.

> >> -
> >> -	return new_crtc != old_crtc ||
> >> -	       (new_crtc &&
> >> -		needs_modeset(intel_atomic_get_new_crtc_state(state, new_crtc)));
> >> -}
> >> -

-- 
Ville Syrjälä
Intel
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx