From: "Lisovskiy, Stanislav" <stanislav.lisovskiy@intel.com>
To: Ville Syrjala <ville.syrjala@linux.intel.com>
Cc: intel-gfx@lists.freedesktop.org
Subject: Re: [Intel-gfx] [PATCH v2 2/9] drm/i915: Split vlv_compute_pipe_wm() into two
Date: Wed, 21 Sep 2022 18:07:59 +0300 [thread overview]
Message-ID: <YysozyslzdzLvYSW@intel.com> (raw)
In-Reply-To: <20220622155452.32587-3-ville.syrjala@linux.intel.com>
On Wed, Jun 22, 2022 at 06:54:45PM +0300, Ville Syrjala wrote:
> From: Ville Syrjälä <ville.syrjala@linux.intel.com>
>
> Split vlv_compute_pipe_wm() into two halves. The first half computes
> the new raw watermarks, and the second half munges those up into real
> watermarks for the particular pipe.
>
> We can reuse the second half for watermark sanitation as well.
>
> Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Stanislav Lisovskiy <stanslav.lisovskiy@intel.com>
> ---
> drivers/gpu/drm/i915/intel_pm.c | 114 ++++++++++++++++++--------------
> 1 file changed, 64 insertions(+), 50 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
> index 395ed3c832d6..4ea43fa73075 100644
> --- a/drivers/gpu/drm/i915/intel_pm.c
> +++ b/drivers/gpu/drm/i915/intel_pm.c
> @@ -1904,64 +1904,17 @@ static bool vlv_raw_crtc_wm_is_valid(const struct intel_crtc_state *crtc_state,
> vlv_raw_plane_wm_is_valid(crtc_state, PLANE_CURSOR, level);
> }
>
> -static int vlv_compute_pipe_wm(struct intel_atomic_state *state,
> - struct intel_crtc *crtc)
> +static int _vlv_compute_pipe_wm(struct intel_crtc_state *crtc_state)
> {
> + struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
> struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
> - struct intel_crtc_state *crtc_state =
> - intel_atomic_get_new_crtc_state(state, crtc);
> struct vlv_wm_state *wm_state = &crtc_state->wm.vlv.optimal;
> const struct vlv_fifo_state *fifo_state =
> &crtc_state->wm.vlv.fifo_state;
> u8 active_planes = crtc_state->active_planes & ~BIT(PLANE_CURSOR);
> int num_active_planes = hweight8(active_planes);
> - bool needs_modeset = drm_atomic_crtc_needs_modeset(&crtc_state->uapi);
> - const struct intel_plane_state *old_plane_state;
> - const struct intel_plane_state *new_plane_state;
> - struct intel_plane *plane;
> enum plane_id plane_id;
> - int level, ret, i;
> - unsigned int dirty = 0;
> -
> - for_each_oldnew_intel_plane_in_state(state, plane,
> - old_plane_state,
> - new_plane_state, i) {
> - if (new_plane_state->hw.crtc != &crtc->base &&
> - old_plane_state->hw.crtc != &crtc->base)
> - continue;
> -
> - if (vlv_raw_plane_wm_compute(crtc_state, new_plane_state))
> - dirty |= BIT(plane->id);
> - }
> -
> - /*
> - * DSPARB registers may have been reset due to the
> - * power well being turned off. Make sure we restore
> - * them to a consistent state even if no primary/sprite
> - * planes are initially active.
> - */
> - if (needs_modeset)
> - crtc_state->fifo_changed = true;
> -
> - if (!dirty)
> - return 0;
> -
> - /* cursor changes don't warrant a FIFO recompute */
> - if (dirty & ~BIT(PLANE_CURSOR)) {
> - const struct intel_crtc_state *old_crtc_state =
> - intel_atomic_get_old_crtc_state(state, crtc);
> - const struct vlv_fifo_state *old_fifo_state =
> - &old_crtc_state->wm.vlv.fifo_state;
> -
> - ret = vlv_compute_fifo(crtc_state);
> - if (ret)
> - return ret;
> -
> - if (needs_modeset ||
> - memcmp(old_fifo_state, fifo_state,
> - sizeof(*fifo_state)) != 0)
> - crtc_state->fifo_changed = true;
> - }
> + int level;
>
> /* initially allow all levels */
> wm_state->num_levels = intel_wm_num_levels(dev_priv);
> @@ -2008,6 +1961,67 @@ static int vlv_compute_pipe_wm(struct intel_atomic_state *state,
> return 0;
> }
>
> +static int vlv_compute_pipe_wm(struct intel_atomic_state *state,
> + struct intel_crtc *crtc)
> +{
> + struct intel_crtc_state *crtc_state =
> + intel_atomic_get_new_crtc_state(state, crtc);
> + bool needs_modeset = drm_atomic_crtc_needs_modeset(&crtc_state->uapi);
> + const struct intel_plane_state *old_plane_state;
> + const struct intel_plane_state *new_plane_state;
> + struct intel_plane *plane;
> + unsigned int dirty = 0;
> + int i;
> +
> + for_each_oldnew_intel_plane_in_state(state, plane,
> + old_plane_state,
> + new_plane_state, i) {
> + if (new_plane_state->hw.crtc != &crtc->base &&
> + old_plane_state->hw.crtc != &crtc->base)
> + continue;
> +
> + if (vlv_raw_plane_wm_compute(crtc_state, new_plane_state))
> + dirty |= BIT(plane->id);
> + }
> +
> + /*
> + * DSPARB registers may have been reset due to the
> + * power well being turned off. Make sure we restore
> + * them to a consistent state even if no primary/sprite
> + * planes are initially active. We also force a FIFO
> + * recomputation so that we are sure to sanitize the
> + * FIFO setting we took over from the BIOS even if there
> + * are no active planes on the crtc.
> + */
> + if (needs_modeset)
> + dirty = ~0;
> +
> + if (!dirty)
> + return 0;
> +
> + /* cursor changes don't warrant a FIFO recompute */
> + if (dirty & ~BIT(PLANE_CURSOR)) {
> + const struct intel_crtc_state *old_crtc_state =
> + intel_atomic_get_old_crtc_state(state, crtc);
> + const struct vlv_fifo_state *old_fifo_state =
> + &old_crtc_state->wm.vlv.fifo_state;
> + const struct vlv_fifo_state *new_fifo_state =
> + &crtc_state->wm.vlv.fifo_state;
> + int ret;
> +
> + ret = vlv_compute_fifo(crtc_state);
> + if (ret)
> + return ret;
> +
> + if (needs_modeset ||
> + memcmp(old_fifo_state, new_fifo_state,
> + sizeof(*new_fifo_state)) != 0)
> + crtc_state->fifo_changed = true;
> + }
> +
> + return _vlv_compute_pipe_wm(crtc_state);
> +}
> +
> #define VLV_FIFO(plane, value) \
> (((value) << DSPARB_ ## plane ## _SHIFT_VLV) & DSPARB_ ## plane ## _MASK_VLV)
>
> --
> 2.35.1
>
next prev parent reply other threads:[~2022-09-21 15:07 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-06-22 15:54 [Intel-gfx] [PATCH v2 0/9] drm/i915: g4x/vlv/chv CxSR/wm fixes/cleanups Ville Syrjala
2022-06-22 15:54 ` [Intel-gfx] [PATCH v2 1/9] drm/i915: Split g4x_compute_pipe_wm() into two Ville Syrjala
2022-06-22 15:54 ` [Intel-gfx] [PATCH v2 2/9] drm/i915: Split vlv_compute_pipe_wm() " Ville Syrjala
2022-09-21 15:07 ` Lisovskiy, Stanislav [this message]
2022-06-22 15:54 ` [Intel-gfx] [PATCH v2 3/9] drm/i915: Simplify up g4x watermark sanitation Ville Syrjala
2022-09-21 15:10 ` Lisovskiy, Stanislav
2022-06-22 15:54 ` [Intel-gfx] [PATCH v2 4/9] drm/i915: Simplify up vlv " Ville Syrjala
2022-09-21 15:13 ` Lisovskiy, Stanislav
2022-06-22 15:54 ` [Intel-gfx] [PATCH v2 5/9] drm/i915: Add missing invalidate to g4x wm readout Ville Syrjala
2022-09-21 15:14 ` Lisovskiy, Stanislav
2022-06-22 15:54 ` [Intel-gfx] [PATCH v2 6/9] drm/i915: Fix g4x/vlv/chv CxSR vs. format/tiling/rotation changes Ville Syrjala
2022-10-07 5:57 ` Lisovskiy, Stanislav
2022-06-22 15:54 ` [Intel-gfx] [PATCH v2 7/9] drm/i915: Fix pipe gamma enable/disable vs. CxSR on gmch platforms Ville Syrjala
2022-10-07 6:01 ` Lisovskiy, Stanislav
2022-10-07 6:21 ` Ville Syrjälä
2022-06-22 15:54 ` [Intel-gfx] [PATCH v2 8/9] drm/i915: Write watermarks for disabled pipes " Ville Syrjala
2022-10-07 6:03 ` Lisovskiy, Stanislav
2022-06-22 15:54 ` [Intel-gfx] [PATCH v2 9/9] drm/i915: Enable atomic by default on ctg/elk Ville Syrjala
2022-06-23 18:34 ` Jani Nikula
2022-06-22 23:02 ` [Intel-gfx] ✓ Fi.CI.BAT: success for drm/i915: g4x/vlv/chv CxSR/wm fixes/cleanups (rev3) Patchwork
2022-06-27 11:01 ` [Intel-gfx] ✗ Fi.CI.IGT: failure " Patchwork
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=YysozyslzdzLvYSW@intel.com \
--to=stanislav.lisovskiy@intel.com \
--cc=intel-gfx@lists.freedesktop.org \
--cc=ville.syrjala@linux.intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox