[PATCH v2 0/1] drm/sti: hdmi: call drm_edid_connector_update when edid is NULL

[PATCH v2 0/1] drm/sti: hdmi: call drm_edid_connector_update when edid is NULL

[Chu Guangqing]

call drm_edid_connector_update to reset the information when edid is NULL

v2:
 - add hdmi: to the commit subject
 - remove DRM_ERROR after reset reset edid
v1:
  https://lore.kernel.org/all/20251014065406.4017-1-chuguangqing@inspur.com/

Chu Guangqing (1):
  drm/sti: hdmi: call drm_edid_connector_update when edid is NULL

 drivers/gpu/drm/sti/sti_hdmi.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

-- 
2.43.7

[PATCH v2 1/1] drm/sti: hdmi: call drm_edid_connector_update when edid is NULL

[Chu Guangqing]

call drm_edid_connector_update to reset the information when edid is NULL.
We can see the following comments in drm_edid.c
If EDID is NULL, reset the information.

Signed-off-by: Chu Guangqing <chuguangqing@inspur.com>
---
 drivers/gpu/drm/sti/sti_hdmi.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/sti/sti_hdmi.c b/drivers/gpu/drm/sti/sti_hdmi.c
index 4e7c3d78b2b9..e0be1be8bcdd 100644
--- a/drivers/gpu/drm/sti/sti_hdmi.c
+++ b/drivers/gpu/drm/sti/sti_hdmi.c
@@ -1008,7 +1008,7 @@ static int sti_hdmi_connector_get_modes(struct drm_connector *connector)
 	return count;
 
 fail:
-	DRM_ERROR("Can't read HDMI EDID\n");
+	drm_edid_connector_update(connector, NULL);
 	return 0;
 }
 
-- 
2.43.7

Re: [PATCH v2 1/1] drm/sti: hdmi: call drm_edid_connector_update when edid is NULL

[Raphaël Gallais-Pou]

Le Mon, Oct 20, 2025 at 09:30:39AM +0800, Chu Guangqing a écrit :
> call drm_edid_connector_update to reset the information when edid is NULL.
> We can see the following comments in drm_edid.c
> If EDID is NULL, reset the information.
> 
> Signed-off-by: Chu Guangqing <chuguangqing@inspur.com>
> ---

Hi,

Acked-by: Raphael Gallais-Pou <rgallaispou@gmail.com>

Best regards,
Raphaël

Re: [PATCH v2 1/1] drm/sti: hdmi: call drm_edid_connector_update when edid is NULL

[Jani Nikula]

On Mon, 20 Oct 2025, Chu Guangqing <chuguangqing@inspur.com> wrote:
> call drm_edid_connector_update to reset the information when edid is NULL.
> We can see the following comments in drm_edid.c
> If EDID is NULL, reset the information.
>
> Signed-off-by: Chu Guangqing <chuguangqing@inspur.com>
> ---
>  drivers/gpu/drm/sti/sti_hdmi.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/drm/sti/sti_hdmi.c b/drivers/gpu/drm/sti/sti_hdmi.c
> index 4e7c3d78b2b9..e0be1be8bcdd 100644
> --- a/drivers/gpu/drm/sti/sti_hdmi.c
> +++ b/drivers/gpu/drm/sti/sti_hdmi.c
> @@ -1008,7 +1008,7 @@ static int sti_hdmi_connector_get_modes(struct drm_connector *connector)
>  	return count;
>  
>  fail:
> -	DRM_ERROR("Can't read HDMI EDID\n");
> +	drm_edid_connector_update(connector, NULL);

The context above has:

	drm_edid = drm_edid_read(connector);

	drm_edid_connector_update(connector, drm_edid);

	cec_notifier_set_phys_addr(hdmi->notifier,
				   connector->display_info.source_physical_address);

	if (!drm_edid)
		goto fail;

i.e. drm_edid_connector_update() already gets called regardless of
whether drm_edid is NULL or not. Precisely as intended in commit
f7945d9fa8b7 ("drm/sti/sti_hdmi: convert to struct drm_edid").


BR,
Jani.


>  	return 0;
>  }

-- 
Jani Nikula, Intel

Re: [PATCH v2 1/1] drm/sti: hdmi: call drm_edid_connector_update when edid is NULL

[Raphaël Gallais-Pou]

Le Tue, Oct 21, 2025 at 11:55:01AM +0300, Jani Nikula a écrit :
> On Mon, 20 Oct 2025, Chu Guangqing <chuguangqing@inspur.com> wrote:
> > call drm_edid_connector_update to reset the information when edid is NULL.
> > We can see the following comments in drm_edid.c
> > If EDID is NULL, reset the information.
> >
> > Signed-off-by: Chu Guangqing <chuguangqing@inspur.com>
> > ---
> >  drivers/gpu/drm/sti/sti_hdmi.c | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > diff --git a/drivers/gpu/drm/sti/sti_hdmi.c b/drivers/gpu/drm/sti/sti_hdmi.c
> > index 4e7c3d78b2b9..e0be1be8bcdd 100644
> > --- a/drivers/gpu/drm/sti/sti_hdmi.c
> > +++ b/drivers/gpu/drm/sti/sti_hdmi.c
> > @@ -1008,7 +1008,7 @@ static int sti_hdmi_connector_get_modes(struct drm_connector *connector)
> >  	return count;
> >  
> >  fail:
> > -	DRM_ERROR("Can't read HDMI EDID\n");
> > +	drm_edid_connector_update(connector, NULL);
> 
> The context above has:
> 
> 	drm_edid = drm_edid_read(connector);
> 
> 	drm_edid_connector_update(connector, drm_edid);
> 
> 	cec_notifier_set_phys_addr(hdmi->notifier,
> 				   connector->display_info.source_physical_address);
> 
> 	if (!drm_edid)
> 		goto fail;
> 
> i.e. drm_edid_connector_update() already gets called regardless of
> whether drm_edid is NULL or not. Precisely as intended in commit
> f7945d9fa8b7 ("drm/sti/sti_hdmi: convert to struct drm_edid").

Woops indeed you are right.  Looks like I was distracted during the
review.

Sorry but looking at it again this patch has no mean to be.  Regarding
the noisy message, I'll send one removing it, but it is not pressing
IMHO.  It only appears if the driver probe with connector plugged.

Thanks for your reply Jani,

Best regards,
Raphaël
> 
> 
> BR,
> Jani.
> 
> 
> >  	return 0;
> >  }
> 
> -- 
> Jani Nikula, Intel