Re: [PATCH 3/8] drm/ingenic: Use standard drm_atomic_helper_commit_tail

From: Christophe JAILLET
Date: Sun Aug 08 2021 - 16:15:54 EST


Le 08/08/2021 à 22:09, Paul Cercueil a écrit :
Hi Christophe,

Le dim., août 8 2021 at 21:50:04 +0200, Christophe JAILLET <christophe.jaillet@xxxxxxxxxx> a écrit :
Le 08/08/2021 à 15:45, Paul Cercueil a écrit :
By making the CRTC's .vblank_enable() function return an error when it
is known that the hardware won't deliver a VBLANK, we can drop the
ingenic_drm_atomic_helper_commit_tail() function and use the standard
drm_atomic_helper_commit_tail() function instead.

Signed-off-by: Paul Cercueil <paul@xxxxxxxxxxxxxxx>
---
  drivers/gpu/drm/ingenic/ingenic-drm-drv.c | 28 ++++-------------------
  1 file changed, 4 insertions(+), 24 deletions(-)

diff --git a/drivers/gpu/drm/ingenic/ingenic-drm-drv.c b/drivers/gpu/drm/ingenic/ingenic-drm-drv.c
index bc71ba44ccf4..3ed7c27a8dde 100644
--- a/drivers/gpu/drm/ingenic/ingenic-drm-drv.c
+++ b/drivers/gpu/drm/ingenic/ingenic-drm-drv.c
@@ -706,29 +706,6 @@ static int ingenic_drm_encoder_atomic_check(struct drm_encoder *encoder,
      }
  }
  -static void ingenic_drm_atomic_helper_commit_tail(struct drm_atomic_state *old_state)
-{
-    /*
-     * Just your regular drm_atomic_helper_commit_tail(), but only calls
-     * drm_atomic_helper_wait_for_vblanks() if priv->no_vblank.
-     */
-    struct drm_device *dev = old_state->dev;
-    struct ingenic_drm *priv = drm_device_get_priv(dev);
-
-    drm_atomic_helper_commit_modeset_disables(dev, old_state);
-
-    drm_atomic_helper_commit_planes(dev, old_state, 0);
-
-    drm_atomic_helper_commit_modeset_enables(dev, old_state);
-
-    drm_atomic_helper_commit_hw_done(old_state);
-
-    if (!priv->no_vblank)
-        drm_atomic_helper_wait_for_vblanks(dev, old_state);
-
-    drm_atomic_helper_cleanup_planes(dev, old_state);
-}


Hi,
if this function is removed, shouldn't:
  static struct drm_mode_config_helper_funcs ingenic_drm_mode_config_helpers = {
      .atomic_commit_tail = ingenic_drm_atomic_helper_commit_tail,
  };
be updated as well?

I've not seen it in the serie.

It is there though :) At the bottom of this very patch.


My email client played me some tricks, apparently!
Sorry for the noise.

CJ

Just my 2v.
CJ

Cheers,
-Paul