drm: Fixup locking WARNINGs in drm_mode_config_reset
diff mbox

Message ID 1438151563-8369-1-git-send-email-daniel.vetter@ffwll.ch
State New
Headers show

Commit Message

Daniel Vetter July 29, 2015, 6:32 a.m. UTC
With

commit 7a3f3d6667f5f9ffd1517f6b21d64bbf5312042c
Author: Daniel Vetter <daniel.vetter@ffwll.ch>
Date:   Thu Jul 9 23:44:28 2015 +0200

    drm: Check locking in drm_for_each_connector

we started checking the locking in drm_for_each_connector but somehow
I totally missed drm_mode_config_reset. There's no problem there since
this function should only be called in single-threaded contexts
(driver load or resume), so just wrap the loop with the right lock.

v2: Drink coffee and all that ...

Cc: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reported-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
---
 drivers/gpu/drm/drm_crtc.c | 2 ++
 1 file changed, 2 insertions(+)

Comments

Daniel Vetter July 29, 2015, 8:19 a.m. UTC | #1
On Wed, Jul 29, 2015 at 08:32:43AM +0200, Daniel Vetter wrote:
> With
> 
> commit 7a3f3d6667f5f9ffd1517f6b21d64bbf5312042c
> Author: Daniel Vetter <daniel.vetter@ffwll.ch>
> Date:   Thu Jul 9 23:44:28 2015 +0200
> 
>     drm: Check locking in drm_for_each_connector
> 
> we started checking the locking in drm_for_each_connector but somehow
> I totally missed drm_mode_config_reset. There's no problem there since
> this function should only be called in single-threaded contexts
> (driver load or resume), so just wrap the loop with the right lock.
> 
> v2: Drink coffee and all that ...
> 
> Cc: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
> Reported-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
> Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>

Tested-by: Jianwei Wang <jianwei.wang.chn@gmail.com>

> ---
>  drivers/gpu/drm/drm_crtc.c | 2 ++
>  1 file changed, 2 insertions(+)
> 
> diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c
> index c91c18b2b1d4..10c1a0f6680c 100644
> --- a/drivers/gpu/drm/drm_crtc.c
> +++ b/drivers/gpu/drm/drm_crtc.c
> @@ -5273,9 +5273,11 @@ void drm_mode_config_reset(struct drm_device *dev)
>  		if (encoder->funcs->reset)
>  			encoder->funcs->reset(encoder);
>  
> +	mutex_lock(&dev->mode_config.mutex);
>  	drm_for_each_connector(connector, dev)
>  		if (connector->funcs->reset)
>  			connector->funcs->reset(connector);
> +	mutex_unlock(&dev->mode_config.mutex);
>  }
>  EXPORT_SYMBOL(drm_mode_config_reset);
>  
> -- 
> 2.1.4
>
Laurent Pinchart July 29, 2015, 8:15 p.m. UTC | #2
Hi Daniel,

Thank you for the patch.

On Wednesday 29 July 2015 08:32:43 Daniel Vetter wrote:
> With
> 
> commit 7a3f3d6667f5f9ffd1517f6b21d64bbf5312042c
> Author: Daniel Vetter <daniel.vetter@ffwll.ch>
> Date:   Thu Jul 9 23:44:28 2015 +0200
> 
>     drm: Check locking in drm_for_each_connector
> 
> we started checking the locking in drm_for_each_connector but somehow
> I totally missed drm_mode_config_reset. There's no problem there since
> this function should only be called in single-threaded contexts
> (driver load or resume), so just wrap the loop with the right lock.
> 
> v2: Drink coffee and all that ...
> 
> Cc: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
> Reported-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
> Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>

This gets rid of the warning at driver load time with the rcar-du-drm driver.

Tested-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>

> ---
>  drivers/gpu/drm/drm_crtc.c | 2 ++
>  1 file changed, 2 insertions(+)
> 
> diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c
> index c91c18b2b1d4..10c1a0f6680c 100644
> --- a/drivers/gpu/drm/drm_crtc.c
> +++ b/drivers/gpu/drm/drm_crtc.c
> @@ -5273,9 +5273,11 @@ void drm_mode_config_reset(struct drm_device *dev)
>  		if (encoder->funcs->reset)
>  			encoder->funcs->reset(encoder);
> 
> +	mutex_lock(&dev->mode_config.mutex);
>  	drm_for_each_connector(connector, dev)
>  		if (connector->funcs->reset)
>  			connector->funcs->reset(connector);
> +	mutex_unlock(&dev->mode_config.mutex);

Wouldn't it make sense to protect the whole function with the mode_config 
mutex, given that it's documented as "protecting the KMS related lists and 
structures" ?

>  }
>  EXPORT_SYMBOL(drm_mode_config_reset);
Daniel Vetter July 29, 2015, 10:24 p.m. UTC | #3
On Wed, Jul 29, 2015 at 11:15:23PM +0300, Laurent Pinchart wrote:
> Hi Daniel,
> 
> Thank you for the patch.
> 
> On Wednesday 29 July 2015 08:32:43 Daniel Vetter wrote:
> > With
> > 
> > commit 7a3f3d6667f5f9ffd1517f6b21d64bbf5312042c
> > Author: Daniel Vetter <daniel.vetter@ffwll.ch>
> > Date:   Thu Jul 9 23:44:28 2015 +0200
> > 
> >     drm: Check locking in drm_for_each_connector
> > 
> > we started checking the locking in drm_for_each_connector but somehow
> > I totally missed drm_mode_config_reset. There's no problem there since
> > this function should only be called in single-threaded contexts
> > (driver load or resume), so just wrap the loop with the right lock.
> > 
> > v2: Drink coffee and all that ...
> > 
> > Cc: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
> > Reported-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
> > Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
> 
> This gets rid of the warning at driver load time with the rcar-du-drm driver.
> 
> Tested-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>

Thanks for testing, applied patch to drm-misc.
> 
> > ---
> >  drivers/gpu/drm/drm_crtc.c | 2 ++
> >  1 file changed, 2 insertions(+)
> > 
> > diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c
> > index c91c18b2b1d4..10c1a0f6680c 100644
> > --- a/drivers/gpu/drm/drm_crtc.c
> > +++ b/drivers/gpu/drm/drm_crtc.c
> > @@ -5273,9 +5273,11 @@ void drm_mode_config_reset(struct drm_device *dev)
> >  		if (encoder->funcs->reset)
> >  			encoder->funcs->reset(encoder);
> > 
> > +	mutex_lock(&dev->mode_config.mutex);
> >  	drm_for_each_connector(connector, dev)
> >  		if (connector->funcs->reset)
> >  			connector->funcs->reset(connector);
> > +	mutex_unlock(&dev->mode_config.mutex);
> 
> Wouldn't it make sense to protect the whole function with the mode_config 
> mutex, given that it's documented as "protecting the KMS related lists and 
> structures" ?

Well that was correct ages ago I guess but since then we've split out tons
of other locks. It doesn't protect anything else really since those lists
are all static after driver load, and even here it's just to shut up the
WARNING (which is right for runtime in general and not harmful to grab the
lock really either).

I do have a plan to fix up kerneldoc, but I want to get the various
improvements collabora is working on in first.
-Daniel

> 
> >  }
> >  EXPORT_SYMBOL(drm_mode_config_reset);
> 
> -- 
> Regards,
> 
> Laurent Pinchart
>

Patch
diff mbox

diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c
index c91c18b2b1d4..10c1a0f6680c 100644
--- a/drivers/gpu/drm/drm_crtc.c
+++ b/drivers/gpu/drm/drm_crtc.c
@@ -5273,9 +5273,11 @@  void drm_mode_config_reset(struct drm_device *dev)
 		if (encoder->funcs->reset)
 			encoder->funcs->reset(encoder);
 
+	mutex_lock(&dev->mode_config.mutex);
 	drm_for_each_connector(connector, dev)
 		if (connector->funcs->reset)
 			connector->funcs->reset(connector);
+	mutex_unlock(&dev->mode_config.mutex);
 }
 EXPORT_SYMBOL(drm_mode_config_reset);