[PATCH] gpu/drm/drm_framebuffer.c: Add judgement for return value of drm_get_format_info().

[PATCH] gpu/drm/drm_framebuffer.c: Add judgement for return value of drm_get_format_info().

[Peng Hao]

Since drm_get_format_info() may return NULL, so a judgement of return
value is needed to add.

Signed-off-by: Peng Hao <penghao@dingdao.com>
---
 drivers/gpu/drm/drm_framebuffer.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/gpu/drm/drm_framebuffer.c b/drivers/gpu/drm/drm_framebuffer.c
index aff3746dedfb..be7dd1998c04 100644
--- a/drivers/gpu/drm/drm_framebuffer.c
+++ b/drivers/gpu/drm/drm_framebuffer.c
@@ -194,6 +194,10 @@ static int framebuffer_check(struct drm_device *dev,
 
 	/* now let the driver pick its own format info */
 	info = drm_get_format_info(dev, r);
+	if (!info) {
+		drm_dbg_kms(dev, "no matched format info\n");
+		return -EFAULT;
+	}
 
 	for (i = 0; i < info->num_planes; i++) {
 		unsigned int width = fb_plane_width(r->width, info, i);
-- 
2.37.1

[PATCH] gpu/drm/drm_framebuffer.c: Add judgement for return value of drm_get_format_info().

[Peng Hao]

Since drm_get_format_info() may return NULL, so a judgement of return
value is needed to add.

Signed-off-by: Peng Hao <penghao@dingdao.com>
---
 drivers/gpu/drm/drm_framebuffer.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/gpu/drm/drm_framebuffer.c b/drivers/gpu/drm/drm_framebuffer.c
index aff3746dedfb..be7dd1998c04 100644
--- a/drivers/gpu/drm/drm_framebuffer.c
+++ b/drivers/gpu/drm/drm_framebuffer.c
@@ -194,6 +194,10 @@ static int framebuffer_check(struct drm_device *dev,
 
 	/* now let the driver pick its own format info */
 	info = drm_get_format_info(dev, r);
+	if (!info) {
+		drm_dbg_kms(dev, "no matched format info\n");
+		return -EFAULT;
+	}
 
 	for (i = 0; i < info->num_planes; i++) {
 		unsigned int width = fb_plane_width(r->width, info, i);
-- 
2.37.1

Re: [PATCH] gpu/drm/drm_framebuffer.c: Add judgement for return value of drm_get_format_info().

[Ville Syrjälä]

On Fri, Oct 27, 2023 at 05:19:12PM +0800, Peng Hao wrote:
> Since drm_get_format_info() may return NULL,

Not in this case since we already checked it earlier.

> so a judgement of return
> value is needed to add.
> 
> Signed-off-by: Peng Hao <penghao@dingdao.com>
> ---
>  drivers/gpu/drm/drm_framebuffer.c | 4 ++++
>  1 file changed, 4 insertions(+)
> 
> diff --git a/drivers/gpu/drm/drm_framebuffer.c b/drivers/gpu/drm/drm_framebuffer.c
> index aff3746dedfb..be7dd1998c04 100644
> --- a/drivers/gpu/drm/drm_framebuffer.c
> +++ b/drivers/gpu/drm/drm_framebuffer.c
> @@ -194,6 +194,10 @@ static int framebuffer_check(struct drm_device *dev,
>  
>  	/* now let the driver pick its own format info */
>  	info = drm_get_format_info(dev, r);
> +	if (!info) {
> +		drm_dbg_kms(dev, "no matched format info\n");
> +		return -EFAULT;
> +	}
>  
>  	for (i = 0; i < info->num_planes; i++) {
>  		unsigned int width = fb_plane_width(r->width, info, i);
> -- 
> 2.37.1

-- 
Ville Syrjälä
Intel

Re: [PATCH] gpu/drm/drm_framebuffer.c: Add judgement for return value of drm_get_format_info().

[彭昊]

[-- Attachment #1: Type: text/plain, Size: 2322 bytes --]

Hi, Ville & all
Thanks for quick reply!
According to your reply, the *info has been checked info earlier, and I have read the code of framebuffer_check() and related drm_get_format_info() which it calls, but unfortunatelly, I haven't seen  judgement(check) for it. What I only found related with check is the sentence below (in *drm_format_info() in drivers/gpu/drm/drm_fourcc.c)：
```
info = __drm_format_info(format);
WARN_ON(!info);
``` 
But to the best of my knowledge, "WARN_ON" only gives a dump_stack(), not  discontinue the program going on, so it will indeed go to the below line in framebuffer_check() and raise an error:
```
for (i = 0; i < info->num_planes; i++) {
```
So maybe it will make  sense of my commited code.

Looking forward to your reply and guidance.
> From: "Ville Syrjälä"<ville.syrjala@linux.intel.com>
> Date:  Fri, Oct 27, 2023, 20:33
> Subject:  Re: [PATCH] gpu/drm/drm_framebuffer.c: Add judgement for return value of drm_get_format_info().
> To: "Peng Hao"<penghao@dingdao.com>
> Cc: <maarten.lankhorst@linux.intel.com>, <mripard@kernel.org>, <tzimmermann@suse.de>, <airlied@gmail.com>, <daniel@ffwll.ch>, <linux-kernel@vger.kernel.org>, <dri-devel@lists.freedesktop.org>
> On Fri, Oct 27, 2023 at 05:19:12PM +0800, Peng Hao wrote:
> > Since drm_get_format_info() may return NULL,
> 
> Not in this case since we already checked it earlier.
> 
> > so a judgement of return
> > value is needed to add.
> > 
> > Signed-off-by: Peng Hao <penghao@dingdao.com>
> > ---
> >  drivers/gpu/drm/drm_framebuffer.c | 4 ++++
> >  1 file changed, 4 insertions(+)
> > 
> > diff --git a/drivers/gpu/drm/drm_framebuffer.c b/drivers/gpu/drm/drm_framebuffer.c
> > index aff3746dedfb..be7dd1998c04 100644
> > --- a/drivers/gpu/drm/drm_framebuffer.c
> > +++ b/drivers/gpu/drm/drm_framebuffer.c
> > @@ -194,6 +194,10 @@ static int framebuffer_check(struct drm_device *dev,
> >  
> >  	/* now let the driver pick its own format info */
> >  	info = drm_get_format_info(dev, r);
> > +	if (!info) {
> > +		drm_dbg_kms(dev, "no matched format info\n");
> > +		return -EFAULT;
> > +	}
> >  
> >  	for (i = 0; i < info->num_planes; i++) {
> >  		unsigned int width = fb_plane_width(r->width, info, i);
> > -- 
> > 2.37.1
> 
> -- 
> Ville Syrjälä
> Intel

[-- Attachment #2: Type: text/html, Size: 8529 bytes --]

Re: [PATCH] gpu/drm/drm_framebuffer.c: Add judgement for return value of drm_get_format_info().

[彭昊]

[-- Attachment #1: Type: text/plain, Size: 2322 bytes --]

Hi, Ville & all
Thanks for quick reply!
According to your reply, the *info has been checked info earlier, and I have read the code of framebuffer_check() and related drm_get_format_info() which it calls, but unfortunatelly, I haven't seen  judgement(check) for it. What I only found related with check is the sentence below (in *drm_format_info() in drivers/gpu/drm/drm_fourcc.c)：
```
info = __drm_format_info(format);
WARN_ON(!info);
``` 
But to the best of my knowledge, "WARN_ON" only gives a dump_stack(), not  discontinue the program going on, so it will indeed go to the below line in framebuffer_check() and raise an error:
```
for (i = 0; i < info->num_planes; i++) {
```
So maybe it will make  sense of my commited code.

Looking forward to your reply and guidance.
> From: "Ville Syrjälä"<ville.syrjala@linux.intel.com>
> Date:  Fri, Oct 27, 2023, 20:33
> Subject:  Re: [PATCH] gpu/drm/drm_framebuffer.c: Add judgement for return value of drm_get_format_info().
> To: "Peng Hao"<penghao@dingdao.com>
> Cc: <maarten.lankhorst@linux.intel.com>, <mripard@kernel.org>, <tzimmermann@suse.de>, <airlied@gmail.com>, <daniel@ffwll.ch>, <linux-kernel@vger.kernel.org>, <dri-devel@lists.freedesktop.org>
> On Fri, Oct 27, 2023 at 05:19:12PM +0800, Peng Hao wrote:
> > Since drm_get_format_info() may return NULL,
> 
> Not in this case since we already checked it earlier.
> 
> > so a judgement of return
> > value is needed to add.
> > 
> > Signed-off-by: Peng Hao <penghao@dingdao.com>
> > ---
> >  drivers/gpu/drm/drm_framebuffer.c | 4 ++++
> >  1 file changed, 4 insertions(+)
> > 
> > diff --git a/drivers/gpu/drm/drm_framebuffer.c b/drivers/gpu/drm/drm_framebuffer.c
> > index aff3746dedfb..be7dd1998c04 100644
> > --- a/drivers/gpu/drm/drm_framebuffer.c
> > +++ b/drivers/gpu/drm/drm_framebuffer.c
> > @@ -194,6 +194,10 @@ static int framebuffer_check(struct drm_device *dev,
> >  
> >  	/* now let the driver pick its own format info */
> >  	info = drm_get_format_info(dev, r);
> > +	if (!info) {
> > +		drm_dbg_kms(dev, "no matched format info\n");
> > +		return -EFAULT;
> > +	}
> >  
> >  	for (i = 0; i < info->num_planes; i++) {
> >  		unsigned int width = fb_plane_width(r->width, info, i);
> > -- 
> > 2.37.1
> 
> -- 
> Ville Syrjälä
> Intel

[-- Attachment #2: Type: text/html, Size: 8529 bytes --]

[PATCH] gpu/drm/drm_framebuffer.c: Add judgement for return value of drm_get_format_info().

[彭昊]

[-- Attachment #1: Type: text/plain, Size: 1566 bytes --]

According to your(Villie) reply, the *info has been checked info earlier, and I have read the code of framebuffer_check() and related drm_get_format_info() which it calls, but unfortunatelly, I haven't seen  judgement(check) for it. What I only found related with check is the sentence below (in *drm_format_info() in drivers/gpu/drm/drm_fourcc.c)：
```
info = __drm_format_info(format);
WARN_ON(!info);
``` 
But to the best of my knowledge, "WARN_ON" only gives a dump_stack(), not  discontinue the program going on, so it will indeed go to the below line in framebuffer_check() and raise an error:
```
for (i = 0; i < info->num_planes; i++) {
```
So maybe it will make  sense of my commited code.

Signed-off-by: Peng Hao <penghao@dingdao.com>
---
 drivers/gpu/drm/drm_framebuffer.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/gpu/drm/drm_framebuffer.c b/drivers/gpu/drm/drm_framebuffer.c
index aff3746dedfb..be7dd1998c04 100644
--- a/drivers/gpu/drm/drm_framebuffer.c
+++ b/drivers/gpu/drm/drm_framebuffer.c
@@ -194,6 +194,10 @@ static int framebuffer_check(struct drm_device *dev,
 
         /* now let the driver pick its own format info */
         info = drm_get_format_info(dev, r);
+        if (!info) {
+                drm_dbg_kms(dev, "no matched format info\n");
+                return -EFAULT;
+        }
 
         for (i = 0; i < info->num_planes; i++) {
                 unsigned int width = fb_plane_width(r->width, info, i);
-- 
2.37.1

[-- Attachment #2: Type: text/html, Size: 15038 bytes --]