[PATCH v1] drm: Log recovery method with wedged event

[PATCH v1] drm: Log recovery method with wedged event

[Raag Jadav]

Currently, drm_dev_wedged_event() logs a basic message of device being
wedged, but doesn't print the recovery method used by the caller. Log it
for easier bug report analysis.

While at it, as the device wedging is a catastrophic event, raise its log
level to drm_warn().

Signed-off-by: Raag Jadav <raag.jadav@intel.com>
---
 drivers/gpu/drm/drm_drv.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/drm_drv.c b/drivers/gpu/drm/drm_drv.c
index cb53baa70995..1a1213f4235c 100644
--- a/drivers/gpu/drm/drm_drv.c
+++ b/drivers/gpu/drm/drm_drv.c
@@ -598,8 +598,8 @@ int drm_dev_wedged_event(struct drm_device *dev, unsigned long method,
 		/* Caller is unsure about recovery, do the best we can at this point. */
 		snprintf(event_string, sizeof(event_string), "%s", "WEDGED=unknown");
 
-	drm_info(dev, "device wedged, %s\n", method == DRM_WEDGE_RECOVERY_NONE ?
-		 "but no recovery needed" : "needs recovery");
+	drm_warn(dev, "device wedged with method %#lx, %s\n", method,
+		 method == DRM_WEDGE_RECOVERY_NONE ? "no recovery needed" : "needs recovery");
 
 	if (info && (info->comm[0] != '\0') && (info->pid >= 0)) {
 		snprintf(pid_string, sizeof(pid_string), "PID=%u", info->pid);
-- 
2.43.0

Re: [PATCH v1] drm: Log recovery method with wedged event

[Christian König]

On 10/19/26 04:58, Raag Jadav wrote:
> Currently, drm_dev_wedged_event() logs a basic message of device being
> wedged, but doesn't print the recovery method used by the caller. Log it
> for easier bug report analysis.

That sounds like a good idea to me.

> While at it, as the device wedging is a catastrophic event, raise its log
> level to drm_warn().

Device wedging is rarely something catastrophic but rather normal part of execution handling.

So I would rather object to that.

Regards,
Christian.

> 
> Signed-off-by: Raag Jadav <raag.jadav@intel.com>
> ---
>  drivers/gpu/drm/drm_drv.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/gpu/drm/drm_drv.c b/drivers/gpu/drm/drm_drv.c
> index cb53baa70995..1a1213f4235c 100644
> --- a/drivers/gpu/drm/drm_drv.c
> +++ b/drivers/gpu/drm/drm_drv.c
> @@ -598,8 +598,8 @@ int drm_dev_wedged_event(struct drm_device *dev, unsigned long method,
>  		/* Caller is unsure about recovery, do the best we can at this point. */
>  		snprintf(event_string, sizeof(event_string), "%s", "WEDGED=unknown");
>  
> -	drm_info(dev, "device wedged, %s\n", method == DRM_WEDGE_RECOVERY_NONE ?
> -		 "but no recovery needed" : "needs recovery");
> +	drm_warn(dev, "device wedged with method %#lx, %s\n", method,
> +		 method == DRM_WEDGE_RECOVERY_NONE ? "no recovery needed" : "needs recovery");
>  
>  	if (info && (info->comm[0] != '\0') && (info->pid >= 0)) {
>  		snprintf(pid_string, sizeof(pid_string), "PID=%u", info->pid);

Re: [PATCH v1] drm: Log recovery method with wedged event

[Rodrigo Vivi]

On Mon, Oct 19, 2026 at 04:29:35PM +0200, Christian König wrote:
> On 10/19/26 04:58, Raag Jadav wrote:
> > Currently, drm_dev_wedged_event() logs a basic message of device being
> > wedged, but doesn't print the recovery method used by the caller. Log it
> > for easier bug report analysis.
> 
> That sounds like a good idea to me.

+1

> 
> > While at it, as the device wedging is a catastrophic event, raise its log
> > level to drm_warn().
> 
> Device wedging is rarely something catastrophic but rather normal part of execution handling.
> 
> So I would rather object to that.

right, unless it has some option to differentiate the temporary wedge by the
cathastrofic one I'd also prefer to keep the info.

> 
> Regards,
> Christian.
> 
> > 
> > Signed-off-by: Raag Jadav <raag.jadav@intel.com>
> > ---
> >  drivers/gpu/drm/drm_drv.c | 4 ++--
> >  1 file changed, 2 insertions(+), 2 deletions(-)
> > 
> > diff --git a/drivers/gpu/drm/drm_drv.c b/drivers/gpu/drm/drm_drv.c
> > index cb53baa70995..1a1213f4235c 100644
> > --- a/drivers/gpu/drm/drm_drv.c
> > +++ b/drivers/gpu/drm/drm_drv.c
> > @@ -598,8 +598,8 @@ int drm_dev_wedged_event(struct drm_device *dev, unsigned long method,
> >  		/* Caller is unsure about recovery, do the best we can at this point. */
> >  		snprintf(event_string, sizeof(event_string), "%s", "WEDGED=unknown");
> >  
> > -	drm_info(dev, "device wedged, %s\n", method == DRM_WEDGE_RECOVERY_NONE ?
> > -		 "but no recovery needed" : "needs recovery");
> > +	drm_warn(dev, "device wedged with method %#lx, %s\n", method,
> > +		 method == DRM_WEDGE_RECOVERY_NONE ? "no recovery needed" : "needs recovery");
> >  
> >  	if (info && (info->comm[0] != '\0') && (info->pid >= 0)) {
> >  		snprintf(pid_string, sizeof(pid_string), "PID=%u", info->pid);
>