[Intel-gfx] [PATCH 06/39] drm/i915: move wm_disp funcs to display.funcs

[Murthy, Arun R]

> -----Original Message-----
> From: Nikula, Jani <jani.nikula at intel.com>
> Sent: Friday, August 12, 2022 12:25 PM
> To: Murthy, Arun R <arun.r.murthy at intel.com>; intel-
> gfx at lists.freedesktop.org
> Cc: De Marchi, Lucas <lucas.demarchi at intel.com>
> Subject: RE: [Intel-gfx] [PATCH 06/39] drm/i915: move wm_disp funcs to
> display.funcs
> 
> On Fri, 12 Aug 2022, "Murthy, Arun R" <arun.r.murthy at intel.com> wrote:
> >> diff --git a/drivers/gpu/drm/i915/display/intel_display_core.h
> >> b/drivers/gpu/drm/i915/display/intel_display_core.h
> >> index 98c6ccdc9100..a6843ebcca5a 100644
> >> --- a/drivers/gpu/drm/i915/display/intel_display_core.h
> >> +++ b/drivers/gpu/drm/i915/display/intel_display_core.h
> >> @@ -8,6 +8,7 @@
> >>
> >>  #include <linux/types.h>
> >>
> >> +struct drm_i915_private;
> >>  struct intel_atomic_state;
> >>  struct intel_cdclk_funcs;
> >>  struct intel_clock_gating_funcs;
> >> @@ -31,6 +32,23 @@ struct intel_display_funcs {
> >>       void (*commit_modeset_enables)(struct intel_atomic_state
> >> *state); };
> >>
> >> +/* functions used for watermark calcs for display. */ struct
> >> +intel_wm_funcs {
> >> +     /* update_wm is for legacy wm management */
> >> +     void (*update_wm)(struct drm_i915_private *dev_priv);
> >> +     int (*compute_pipe_wm)(struct intel_atomic_state *state,
> >> +                            struct intel_crtc *crtc);
> >> +     int (*compute_intermediate_wm)(struct intel_atomic_state *state,
> >> +                                    struct intel_crtc *crtc);
> >> +     void (*initial_watermarks)(struct intel_atomic_state *state,
> >> +                                struct intel_crtc *crtc);
> >> +     void (*atomic_update_watermarks)(struct intel_atomic_state *state,
> >> +                                      struct intel_crtc *crtc);
> >> +     void (*optimize_watermarks)(struct intel_atomic_state *state,
> >> +                                 struct intel_crtc *crtc);
> >> +     int (*compute_global_watermarks)(struct intel_atomic_state
> >> +*state);
> >> };
> >> +
> >>  struct intel_display {
> >>       /* Display functions */
> >>       struct {
> >> @@ -48,6 +66,9 @@ struct intel_display {
> >>
> >>               /* pm private clock gating functions */
> >>               const struct intel_clock_gating_funcs *clock_gating;
> >> +
> >> +             /* pm display functions */
> >> +             const struct intel_wm_funcs *wm;
> >>       } funcs;
> >
> > Can the wm, dbuf, clock related move to a struct intel_pm ? which
> > makes it more meaningful else again we end up creating a struct
> > intel_display a long one like i915_private.
> 
> Well, I think the display wm/dbuf stuff also needs to be split from intel_pm.c
> so it's not so clear cut. And for now, I'm keeping the functions together.

Ok, lets take it up in the later. But adding a TODO would make it clear that some more clean up in this regards is in queue.

Reviewed-by: Arun R Murthy <arun.r.murthy at intel.com>

Thanks and Regards,
Arun R Murthy
--------------------