Re: [PATCH 05/16] drm/i915: Move some computations out from hsw_compute_wm_parameters()

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



2013/10/11 Ville Syrjälä <ville.syrjala@xxxxxxxxxxxxxxx>:
> On Thu, Oct 10, 2013 at 07:34:38PM -0300, Paulo Zanoni wrote:
>> 2013/10/9  <ville.syrjala@xxxxxxxxxxxxxxx>:
>> > From: Ville Syrjälä <ville.syrjala@xxxxxxxxxxxxxxx>
>> >
>> > Move the watermark max computations into haswell_update_wm(). This
>> > allows keeping the 1/2 vs. 5/6 split code in one place, and avoid having
>> > to pass around so many things. We also save a bit of stack space by only
>> > requiring one copy of struct hsw_wm_maximums.
>> >
>> > Also move the intel_wm_config out from hsw_compute_wm_parameters() and
>> > pass it it. We'll have some need for it in haswell_update_wm() later.
>> >
>> > Signed-off-by: Ville Syrjälä <ville.syrjala@xxxxxxxxxxxxxxx>
>> > ---
>> >  drivers/gpu/drm/i915/intel_pm.c | 33 ++++++++++++++-------------------
>> >  1 file changed, 14 insertions(+), 19 deletions(-)
>> >
>> > diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
>> > index b09715f..0fe6c36 100644
>> > --- a/drivers/gpu/drm/i915/intel_pm.c
>> > +++ b/drivers/gpu/drm/i915/intel_pm.c
>> > @@ -2586,13 +2586,11 @@ static void intel_setup_wm_latency(struct drm_device *dev)
>> >
>> >  static void hsw_compute_wm_parameters(struct drm_crtc *crtc,
>> >                                       struct hsw_pipe_wm_parameters *p,
>> > -                                     struct hsw_wm_maximums *lp_max_1_2,
>> > -                                     struct hsw_wm_maximums *lp_max_5_6)
>> > +                                     struct intel_wm_config *config)
>> >  {
>> >         struct drm_device *dev = crtc->dev;
>> >         struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
>> >         enum pipe pipe = intel_crtc->pipe;
>> > -       struct intel_wm_config config = {};
>> >         struct drm_plane *plane;
>> >
>> >         p->active = intel_crtc_active(crtc);
>> > @@ -2609,7 +2607,7 @@ static void hsw_compute_wm_parameters(struct drm_crtc *crtc,
>> >         }
>> >
>> >         list_for_each_entry(crtc, &dev->mode_config.crtc_list, head)
>> > -               config.num_pipes_active += intel_crtc_active(crtc);
>> > +               config->num_pipes_active += intel_crtc_active(crtc);
>> >
>> >         list_for_each_entry(plane, &dev->mode_config.plane_list, head) {
>> >                 struct intel_plane *intel_plane = to_intel_plane(plane);
>> > @@ -2617,17 +2615,9 @@ static void hsw_compute_wm_parameters(struct drm_crtc *crtc,
>> >                 if (intel_plane->pipe == pipe)
>> >                         p->spr = intel_plane->wm;
>> >
>> > -               config.sprites_enabled |= intel_plane->wm.enabled;
>> > -               config.sprites_scaled |= intel_plane->wm.scaled;
>> > +               config->sprites_enabled |= intel_plane->wm.enabled;
>> > +               config->sprites_scaled |= intel_plane->wm.scaled;
>> >         }
>> > -
>> > -       ilk_wm_max(dev, 1, &config, INTEL_DDB_PART_1_2, lp_max_1_2);
>> > -
>> > -       /* 5/6 split only in single pipe config on IVB+ */
>> > -       if (INTEL_INFO(dev)->gen >= 7 && config.num_pipes_active <= 1)
>> > -               ilk_wm_max(dev, 1, &config, INTEL_DDB_PART_5_6, lp_max_5_6);
>> > -       else
>> > -               *lp_max_5_6 = *lp_max_1_2;
>> >  }
>> >
>> >  /* Compute new watermarks for the pipe */
>> > @@ -2889,14 +2879,15 @@ static void haswell_update_wm(struct drm_crtc *crtc)
>> >         struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
>> >         struct drm_device *dev = crtc->dev;
>> >         struct drm_i915_private *dev_priv = dev->dev_private;
>> > -       struct hsw_wm_maximums lp_max_1_2, lp_max_5_6;
>> > +       struct hsw_wm_maximums max;
>> >         struct hsw_pipe_wm_parameters params = {};
>> >         struct hsw_wm_values results = {};
>> >         enum intel_ddb_partitioning partitioning;
>> >         struct intel_pipe_wm pipe_wm = {};
>> >         struct intel_pipe_wm lp_wm_1_2 = {}, lp_wm_5_6 = {}, *best_lp_wm;
>> > +       struct intel_wm_config config = {};
>> >
>> > -       hsw_compute_wm_parameters(crtc, &params, &lp_max_1_2, &lp_max_5_6);
>> > +       hsw_compute_wm_parameters(crtc, &params, &config);
>> >
>> >         intel_compute_pipe_wm(crtc, &params, &pipe_wm);
>> >
>> > @@ -2905,10 +2896,14 @@ static void haswell_update_wm(struct drm_crtc *crtc)
>> >
>> >         intel_crtc->wm.active = pipe_wm;
>> >
>> > -       ilk_wm_merge(dev, &lp_max_1_2, &lp_wm_1_2);
>> > -       ilk_wm_merge(dev, &lp_max_5_6, &lp_wm_5_6);
>> > +       ilk_wm_max(dev, 1, &config, INTEL_DDB_PART_1_2, &max);
>> > +       ilk_wm_merge(dev, &max, &lp_wm_1_2);
>> > +
>> > +       /* 5/6 split only in single pipe config on IVB+ */
>> > +       if (INTEL_INFO(dev)->gen >= 7 && config.num_pipes_active <= 1) {
>>
>> No need to calculate 5_6 on zero pipes, I guess.
>>
>>
>> > +               ilk_wm_max(dev, 1, &config, INTEL_DDB_PART_5_6, &max);
>> > +               ilk_wm_merge(dev, &max, &lp_wm_5_6);
>>
>> Oh, so now you've moved ilk_wm_merge to the "if" statement, as I
>> requested on the review to a previous patch :)
>>
>> >
>> > -       if (lp_max_1_2.pri != lp_max_5_6.pri) {
>>
>> By removing this check, you're now also calculating 5_6 watermarks for
>> the case where we just have 1 pipe but the sprites are disabled.
>
> Hmm, right. I guess we should just add a more explicit
> config.sprites_enabled check here to avoid that overhead.

Since you've addressed this issue on patch 17/16, and it's just an
optimization, not a real bug:

Reviewed-by: Paulo Zanoni <paulo.r.zanoni@xxxxxxxxx>

>
>>
>>
>> >                 best_lp_wm = hsw_find_best_result(dev, &lp_wm_1_2, &lp_wm_5_6);
>> >         } else {
>> >                 best_lp_wm = &lp_wm_1_2;
>> > --
>> > 1.8.1.5
>> >
>> > _______________________________________________
>> > Intel-gfx mailing list
>> > Intel-gfx@xxxxxxxxxxxxxxxxxxxxx
>> > http://lists.freedesktop.org/mailman/listinfo/intel-gfx
>>
>>
>>
>> --
>> Paulo Zanoni
>
> --
> Ville Syrjälä
> Intel OTC



-- 
Paulo Zanoni
_______________________________________________
Intel-gfx mailing list
Intel-gfx@xxxxxxxxxxxxxxxxxxxxx
http://lists.freedesktop.org/mailman/listinfo/intel-gfx





[Index of Archives]     [Linux USB Devel]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]
  Powered by Linux