Re: [PATCH 17/19] drm/i915: Use new atomic iterator macros in wm code

[Paulo Zanoni <paulo.r.zanoni@intel.com>]

Em Qui, 2016-11-03 às 18:49 +0200, Ville Syrjälä escreveu:
> On Mon, Oct 17, 2016 at 02:37:16PM +0200, Maarten Lankhorst wrote:
> > 
> > Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com
> > >
> > ---
> >  drivers/gpu/drm/i915/intel_pm.c | 12 ++++++------
> >  1 file changed, 6 insertions(+), 6 deletions(-)
> > 
> > diff --git a/drivers/gpu/drm/i915/intel_pm.c
> > b/drivers/gpu/drm/i915/intel_pm.c
> > index 2df06b703e3d..163b73b493bf 100644
> > --- a/drivers/gpu/drm/i915/intel_pm.c
> > +++ b/drivers/gpu/drm/i915/intel_pm.c
> > @@ -3227,7 +3227,7 @@ skl_get_total_relative_data_rate(struct
> > intel_crtc_state *intel_cstate)
> >  		return 0;
> >  
> >  	/* Calculate and cache data rate for each plane */
> > -	for_each_plane_in_state(state, plane, pstate, i) {
> > +	for_each_new_plane_in_state(state, plane, pstate, i) {
> 
> Looks like this patch needs a refresh due to my tardiness in
> reviewing
> it.
> 
> Also unrelated, but I recently noticed that
> skl_get_total_relative_data_rate() doesn't exclude the cursor plane
> from the calculation. I'm thinking it should. Adding some Cc:s...

It seems we're safe because skl_plane_relative_data_rate() returns 0
for cursor. But an explicitly check at the caller would probably be
much much better instead of hidden deep in the code. OTOH, Matt's plans
to exclude cursor case-handling would remove this need again.

> 
> > 
> >  		id = skl_wm_plane_id(to_intel_plane(plane));
> >  		intel_plane = to_intel_plane(plane);
> >  
> > @@ -3364,14 +3364,14 @@ skl_allocate_pipe_ddb(struct
> > intel_crtc_state *cstate,
> >  	alloc_size -= cursor_blocks;
> >  
> >  	/* 1. Allocate the mininum required blocks for each active
> > plane */
> > -	for_each_plane_in_state(state, plane, pstate, i) {
> > +	for_each_new_plane_in_state(state, plane, pstate, i) {
> >  		intel_plane = to_intel_plane(plane);
> >  		id = skl_wm_plane_id(intel_plane);
> >  
> >  		if (intel_plane->pipe != pipe)
> >  			continue;
> >  
> > -		if (!to_intel_plane_state(pstate)->base.visible) {
> > +		if (!pstate->visible) {
> >  			minimum[id] = 0;
> >  			y_minimum[id] = 0;
> >  			continue;
> > @@ -3933,7 +3933,7 @@ pipes_modified(struct drm_atomic_state
> > *state)
> >  	struct drm_crtc_state *cstate;
> >  	uint32_t i, ret = 0;
> >  
> > -	for_each_crtc_in_state(state, crtc, cstate, i)
> > +	for_each_new_crtc_in_state(state, crtc, cstate, i)
> >  		ret |= drm_crtc_mask(crtc);
> >  
> >  	return ret;
> > @@ -4048,7 +4048,7 @@ skl_compute_wm(struct drm_atomic_state
> > *state)
> >  	 * since any racing commits that want to update them would
> > need to
> >  	 * hold _all_ CRTC state mutexes.
> >  	 */
> > -	for_each_crtc_in_state(state, crtc, cstate, i)
> > +	for_each_new_crtc_in_state(state, crtc, cstate, i)
> >  		changed = true;
> >  	if (!changed)
> >  		return 0;
> > @@ -4070,7 +4070,7 @@ skl_compute_wm(struct drm_atomic_state
> > *state)
> >  	 * should allow skl_update_pipe_wm() to return failure in
> > cases where
> >  	 * no suitable watermark values can be found.
> >  	 */
> > -	for_each_crtc_in_state(state, crtc, cstate, i) {
> > +	for_each_new_crtc_in_state(state, crtc, cstate, i) {
> >  		struct intel_crtc *intel_crtc =
> > to_intel_crtc(crtc);
> >  		struct intel_crtc_state *intel_cstate =
> >  			to_intel_crtc_state(cstate);
> > -- 
> > 2.7.4
> > 
> > _______________________________________________
> > Intel-gfx mailing list
> > Intel-gfx@lists.freedesktop.org
> > https://lists.freedesktop.org/mailman/listinfo/intel-gfx
> 
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx