drm/drm_ioctl.c: Test client capability value early when setting.

Message ID 20180228152741.22507-1-Liviu.Dudau@arm.com
State New
Headers show
Series
  • drm/drm_ioctl.c: Test client capability value early when setting.
Related show

Commit Message

Liviu Dudau Feb. 28, 2018, 3:27 p.m.
The drm_setclientcap() function implementing the DRM_IOCTL_SET_CLIENT_CAP
ioctl expects that any capability set by the client will have a value of 1.
Make the check early so that we don't have to test the value for each
capability.

Signed-off-by: Liviu Dudau <liviu.dudau@arm.com>
---
 drivers/gpu/drm/drm_ioctl.c | 9 +++------
 1 file changed, 3 insertions(+), 6 deletions(-)

Comments

Ville Syrjälä Feb. 28, 2018, 3:40 p.m. | #1
On Wed, Feb 28, 2018 at 03:27:41PM +0000, Liviu Dudau wrote:
> The drm_setclientcap() function implementing the DRM_IOCTL_SET_CLIENT_CAP
> ioctl expects that any capability set by the client will have a value of 1.
> Make the check early so that we don't have to test the value for each
> capability.

What if we want a a non-boolean capability at some point?

> 
> Signed-off-by: Liviu Dudau <liviu.dudau@arm.com>
> ---
>  drivers/gpu/drm/drm_ioctl.c | 9 +++------
>  1 file changed, 3 insertions(+), 6 deletions(-)
> 
> diff --git a/drivers/gpu/drm/drm_ioctl.c b/drivers/gpu/drm/drm_ioctl.c
> index af782911c505..02ffa0e8d77b 100644
> --- a/drivers/gpu/drm/drm_ioctl.c
> +++ b/drivers/gpu/drm/drm_ioctl.c
> @@ -306,22 +306,19 @@ drm_setclientcap(struct drm_device *dev, void *data, struct drm_file *file_priv)
>  {
>  	struct drm_set_client_cap *req = data;
>  
> +	if (req->value > 1)
> +		return -EINVAL;
> +
>  	switch (req->capability) {
>  	case DRM_CLIENT_CAP_STEREO_3D:
> -		if (req->value > 1)
> -			return -EINVAL;
>  		file_priv->stereo_allowed = req->value;
>  		break;
>  	case DRM_CLIENT_CAP_UNIVERSAL_PLANES:
> -		if (req->value > 1)
> -			return -EINVAL;
>  		file_priv->universal_planes = req->value;
>  		break;
>  	case DRM_CLIENT_CAP_ATOMIC:
>  		if (!drm_core_check_feature(dev, DRIVER_ATOMIC))
>  			return -EINVAL;
> -		if (req->value > 1)
> -			return -EINVAL;
>  		file_priv->atomic = req->value;
>  		file_priv->universal_planes = req->value;
>  		break;
> -- 
> 2.16.2
> 
> _______________________________________________
> dri-devel mailing list
> dri-devel@lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/dri-devel
Liviu Dudau Feb. 28, 2018, 3:44 p.m. | #2
On Wed, Feb 28, 2018 at 05:40:41PM +0200, Ville Syrjälä wrote:
> On Wed, Feb 28, 2018 at 03:27:41PM +0000, Liviu Dudau wrote:
> > The drm_setclientcap() function implementing the DRM_IOCTL_SET_CLIENT_CAP
> > ioctl expects that any capability set by the client will have a value of 1.
> > Make the check early so that we don't have to test the value for each
> > capability.
> 
> What if we want a a non-boolean capability at some point?

Well, I'm adding another boolean capability soon, so you will be going
against the trend :)

I guess you will have 2 options: revert the patch or add a condition to
the test.

I don't have strong feelings, just felt like too much copying when
adding another capability so I thought to do some "cleanup".

Best regards,
Liviu

> 
> > 
> > Signed-off-by: Liviu Dudau <liviu.dudau@arm.com>
> > ---
> >  drivers/gpu/drm/drm_ioctl.c | 9 +++------
> >  1 file changed, 3 insertions(+), 6 deletions(-)
> > 
> > diff --git a/drivers/gpu/drm/drm_ioctl.c b/drivers/gpu/drm/drm_ioctl.c
> > index af782911c505..02ffa0e8d77b 100644
> > --- a/drivers/gpu/drm/drm_ioctl.c
> > +++ b/drivers/gpu/drm/drm_ioctl.c
> > @@ -306,22 +306,19 @@ drm_setclientcap(struct drm_device *dev, void *data, struct drm_file *file_priv)
> >  {
> >  	struct drm_set_client_cap *req = data;
> >  
> > +	if (req->value > 1)
> > +		return -EINVAL;
> > +
> >  	switch (req->capability) {
> >  	case DRM_CLIENT_CAP_STEREO_3D:
> > -		if (req->value > 1)
> > -			return -EINVAL;
> >  		file_priv->stereo_allowed = req->value;
> >  		break;
> >  	case DRM_CLIENT_CAP_UNIVERSAL_PLANES:
> > -		if (req->value > 1)
> > -			return -EINVAL;
> >  		file_priv->universal_planes = req->value;
> >  		break;
> >  	case DRM_CLIENT_CAP_ATOMIC:
> >  		if (!drm_core_check_feature(dev, DRIVER_ATOMIC))
> >  			return -EINVAL;
> > -		if (req->value > 1)
> > -			return -EINVAL;
> >  		file_priv->atomic = req->value;
> >  		file_priv->universal_planes = req->value;
> >  		break;
> > -- 
> > 2.16.2
> > 
> > _______________________________________________
> > dri-devel mailing list
> > dri-devel@lists.freedesktop.org
> > https://lists.freedesktop.org/mailman/listinfo/dri-devel
> 
> -- 
> Ville Syrjälä
> Intel OTC
Ville Syrjälä Feb. 28, 2018, 3:57 p.m. | #3
On Wed, Feb 28, 2018 at 03:44:44PM +0000, Liviu Dudau wrote:
> On Wed, Feb 28, 2018 at 05:40:41PM +0200, Ville Syrjälä wrote:
> > On Wed, Feb 28, 2018 at 03:27:41PM +0000, Liviu Dudau wrote:
> > > The drm_setclientcap() function implementing the DRM_IOCTL_SET_CLIENT_CAP
> > > ioctl expects that any capability set by the client will have a value of 1.
> > > Make the check early so that we don't have to test the value for each
> > > capability.
> > 
> > What if we want a a non-boolean capability at some point?
> 
> Well, I'm adding another boolean capability soon, so you will be going
> against the trend :)

Plenty of non-bools in driver specific counterparts I believe.

> I guess you will have 2 options: revert the patch or add a condition to
> the test.
> 
> I don't have strong feelings, just felt like too much copying when
> adding another capability so I thought to do some "cleanup".
> 
> Best regards,
> Liviu
> 
> > 
> > > 
> > > Signed-off-by: Liviu Dudau <liviu.dudau@arm.com>
> > > ---
> > >  drivers/gpu/drm/drm_ioctl.c | 9 +++------
> > >  1 file changed, 3 insertions(+), 6 deletions(-)
> > > 
> > > diff --git a/drivers/gpu/drm/drm_ioctl.c b/drivers/gpu/drm/drm_ioctl.c
> > > index af782911c505..02ffa0e8d77b 100644
> > > --- a/drivers/gpu/drm/drm_ioctl.c
> > > +++ b/drivers/gpu/drm/drm_ioctl.c
> > > @@ -306,22 +306,19 @@ drm_setclientcap(struct drm_device *dev, void *data, struct drm_file *file_priv)
> > >  {
> > >  	struct drm_set_client_cap *req = data;
> > >  
> > > +	if (req->value > 1)
> > > +		return -EINVAL;
> > > +
> > >  	switch (req->capability) {
> > >  	case DRM_CLIENT_CAP_STEREO_3D:
> > > -		if (req->value > 1)
> > > -			return -EINVAL;
> > >  		file_priv->stereo_allowed = req->value;
> > >  		break;
> > >  	case DRM_CLIENT_CAP_UNIVERSAL_PLANES:
> > > -		if (req->value > 1)
> > > -			return -EINVAL;
> > >  		file_priv->universal_planes = req->value;
> > >  		break;
> > >  	case DRM_CLIENT_CAP_ATOMIC:
> > >  		if (!drm_core_check_feature(dev, DRIVER_ATOMIC))
> > >  			return -EINVAL;
> > > -		if (req->value > 1)
> > > -			return -EINVAL;
> > >  		file_priv->atomic = req->value;
> > >  		file_priv->universal_planes = req->value;
> > >  		break;
> > > -- 
> > > 2.16.2
> > > 
> > > _______________________________________________
> > > dri-devel mailing list
> > > dri-devel@lists.freedesktop.org
> > > https://lists.freedesktop.org/mailman/listinfo/dri-devel
> > 
> > -- 
> > Ville Syrjälä
> > Intel OTC
> 
> -- 
> ====================
> | I would like to |
> | fix the world,  |
> | but they're not |
> | giving me the   |
>  \ source code!  /
>   ---------------
>     ¯\_(ツ)_/¯
Liviu Dudau Feb. 28, 2018, 4:34 p.m. | #4
On Wed, Feb 28, 2018 at 05:57:15PM +0200, Ville Syrjälä wrote:
> On Wed, Feb 28, 2018 at 03:44:44PM +0000, Liviu Dudau wrote:
> > On Wed, Feb 28, 2018 at 05:40:41PM +0200, Ville Syrjälä wrote:
> > > On Wed, Feb 28, 2018 at 03:27:41PM +0000, Liviu Dudau wrote:
> > > > The drm_setclientcap() function implementing the DRM_IOCTL_SET_CLIENT_CAP
> > > > ioctl expects that any capability set by the client will have a value of 1.
> > > > Make the check early so that we don't have to test the value for each
> > > > capability.
> > > 
> > > What if we want a a non-boolean capability at some point?
> > 
> > Well, I'm adding another boolean capability soon, so you will be going
> > against the trend :)
> 
> Plenty of non-bools in driver specific counterparts I believe.

So, is that a NACK? 


> 
> > I guess you will have 2 options: revert the patch or add a condition to
> > the test.
> > 
> > I don't have strong feelings, just felt like too much copying when
> > adding another capability so I thought to do some "cleanup".
> > 
> > Best regards,
> > Liviu
> > 
> > > 
> > > > 
> > > > Signed-off-by: Liviu Dudau <liviu.dudau@arm.com>
> > > > ---
> > > >  drivers/gpu/drm/drm_ioctl.c | 9 +++------
> > > >  1 file changed, 3 insertions(+), 6 deletions(-)
> > > > 
> > > > diff --git a/drivers/gpu/drm/drm_ioctl.c b/drivers/gpu/drm/drm_ioctl.c
> > > > index af782911c505..02ffa0e8d77b 100644
> > > > --- a/drivers/gpu/drm/drm_ioctl.c
> > > > +++ b/drivers/gpu/drm/drm_ioctl.c
> > > > @@ -306,22 +306,19 @@ drm_setclientcap(struct drm_device *dev, void *data, struct drm_file *file_priv)
> > > >  {
> > > >  	struct drm_set_client_cap *req = data;
> > > >  
> > > > +	if (req->value > 1)
> > > > +		return -EINVAL;
> > > > +
> > > >  	switch (req->capability) {
> > > >  	case DRM_CLIENT_CAP_STEREO_3D:
> > > > -		if (req->value > 1)
> > > > -			return -EINVAL;
> > > >  		file_priv->stereo_allowed = req->value;
> > > >  		break;
> > > >  	case DRM_CLIENT_CAP_UNIVERSAL_PLANES:
> > > > -		if (req->value > 1)
> > > > -			return -EINVAL;
> > > >  		file_priv->universal_planes = req->value;
> > > >  		break;
> > > >  	case DRM_CLIENT_CAP_ATOMIC:
> > > >  		if (!drm_core_check_feature(dev, DRIVER_ATOMIC))
> > > >  			return -EINVAL;
> > > > -		if (req->value > 1)
> > > > -			return -EINVAL;
> > > >  		file_priv->atomic = req->value;
> > > >  		file_priv->universal_planes = req->value;
> > > >  		break;
> > > > -- 
> > > > 2.16.2
> > > > 
> > > > _______________________________________________
> > > > dri-devel mailing list
> > > > dri-devel@lists.freedesktop.org
> > > > https://lists.freedesktop.org/mailman/listinfo/dri-devel
> > > 
> > > -- 
> > > Ville Syrjälä
> > > Intel OTC
> > 
> > -- 
> > ====================
> > | I would like to |
> > | fix the world,  |
> > | but they're not |
> > | giving me the   |
> >  \ source code!  /
> >   ---------------
> >     ¯\_(ツ)_/¯
> 
> -- 
> Ville Syrjälä
> Intel OTC
Daniel Vetter March 6, 2018, 10:13 a.m. | #5
On Wed, Feb 28, 2018 at 04:34:30PM +0000, Liviu Dudau wrote:
> On Wed, Feb 28, 2018 at 05:57:15PM +0200, Ville Syrjälä wrote:
> > On Wed, Feb 28, 2018 at 03:44:44PM +0000, Liviu Dudau wrote:
> > > On Wed, Feb 28, 2018 at 05:40:41PM +0200, Ville Syrjälä wrote:
> > > > On Wed, Feb 28, 2018 at 03:27:41PM +0000, Liviu Dudau wrote:
> > > > > The drm_setclientcap() function implementing the DRM_IOCTL_SET_CLIENT_CAP
> > > > > ioctl expects that any capability set by the client will have a value of 1.
> > > > > Make the check early so that we don't have to test the value for each
> > > > > capability.
> > > > 
> > > > What if we want a a non-boolean capability at some point?
> > > 
> > > Well, I'm adding another boolean capability soon, so you will be going
> > > against the trend :)
> > 
> > Plenty of non-bools in driver specific counterparts I believe.
> 
> So, is that a NACK? 

Yeah I think this is overoptimizing for not repeating code.
-Daniel

> 
> 
> > 
> > > I guess you will have 2 options: revert the patch or add a condition to
> > > the test.
> > > 
> > > I don't have strong feelings, just felt like too much copying when
> > > adding another capability so I thought to do some "cleanup".
> > > 
> > > Best regards,
> > > Liviu
> > > 
> > > > 
> > > > > 
> > > > > Signed-off-by: Liviu Dudau <liviu.dudau@arm.com>
> > > > > ---
> > > > >  drivers/gpu/drm/drm_ioctl.c | 9 +++------
> > > > >  1 file changed, 3 insertions(+), 6 deletions(-)
> > > > > 
> > > > > diff --git a/drivers/gpu/drm/drm_ioctl.c b/drivers/gpu/drm/drm_ioctl.c
> > > > > index af782911c505..02ffa0e8d77b 100644
> > > > > --- a/drivers/gpu/drm/drm_ioctl.c
> > > > > +++ b/drivers/gpu/drm/drm_ioctl.c
> > > > > @@ -306,22 +306,19 @@ drm_setclientcap(struct drm_device *dev, void *data, struct drm_file *file_priv)
> > > > >  {
> > > > >  	struct drm_set_client_cap *req = data;
> > > > >  
> > > > > +	if (req->value > 1)
> > > > > +		return -EINVAL;
> > > > > +
> > > > >  	switch (req->capability) {
> > > > >  	case DRM_CLIENT_CAP_STEREO_3D:
> > > > > -		if (req->value > 1)
> > > > > -			return -EINVAL;
> > > > >  		file_priv->stereo_allowed = req->value;
> > > > >  		break;
> > > > >  	case DRM_CLIENT_CAP_UNIVERSAL_PLANES:
> > > > > -		if (req->value > 1)
> > > > > -			return -EINVAL;
> > > > >  		file_priv->universal_planes = req->value;
> > > > >  		break;
> > > > >  	case DRM_CLIENT_CAP_ATOMIC:
> > > > >  		if (!drm_core_check_feature(dev, DRIVER_ATOMIC))
> > > > >  			return -EINVAL;
> > > > > -		if (req->value > 1)
> > > > > -			return -EINVAL;
> > > > >  		file_priv->atomic = req->value;
> > > > >  		file_priv->universal_planes = req->value;
> > > > >  		break;
> > > > > -- 
> > > > > 2.16.2
> > > > > 
> > > > > _______________________________________________
> > > > > dri-devel mailing list
> > > > > dri-devel@lists.freedesktop.org
> > > > > https://lists.freedesktop.org/mailman/listinfo/dri-devel
> > > > 
> > > > -- 
> > > > Ville Syrjälä
> > > > Intel OTC
> > > 
> > > -- 
> > > ====================
> > > | I would like to |
> > > | fix the world,  |
> > > | but they're not |
> > > | giving me the   |
> > >  \ source code!  /
> > >   ---------------
> > >     ¯\_(ツ)_/¯
> > 
> > -- 
> > Ville Syrjälä
> > Intel OTC
> 
> -- 
> ====================
> | I would like to |
> | fix the world,  |
> | but they're not |
> | giving me the   |
>  \ source code!  /
>   ---------------
>     ¯\_(ツ)_/¯

Patch

diff --git a/drivers/gpu/drm/drm_ioctl.c b/drivers/gpu/drm/drm_ioctl.c
index af782911c505..02ffa0e8d77b 100644
--- a/drivers/gpu/drm/drm_ioctl.c
+++ b/drivers/gpu/drm/drm_ioctl.c
@@ -306,22 +306,19 @@  drm_setclientcap(struct drm_device *dev, void *data, struct drm_file *file_priv)
 {
 	struct drm_set_client_cap *req = data;
 
+	if (req->value > 1)
+		return -EINVAL;
+
 	switch (req->capability) {
 	case DRM_CLIENT_CAP_STEREO_3D:
-		if (req->value > 1)
-			return -EINVAL;
 		file_priv->stereo_allowed = req->value;
 		break;
 	case DRM_CLIENT_CAP_UNIVERSAL_PLANES:
-		if (req->value > 1)
-			return -EINVAL;
 		file_priv->universal_planes = req->value;
 		break;
 	case DRM_CLIENT_CAP_ATOMIC:
 		if (!drm_core_check_feature(dev, DRIVER_ATOMIC))
 			return -EINVAL;
-		if (req->value > 1)
-			return -EINVAL;
 		file_priv->atomic = req->value;
 		file_priv->universal_planes = req->value;
 		break;