* Re: [PATCH 1/1] v4l: ioctl: Validate num_planes before using it
2019-01-10 12:43 [PATCH 1/1] v4l: ioctl: Validate num_planes before using it Sakari Ailus
@ 2019-01-10 12:51 ` Thierry Reding
2019-01-10 13:02 ` Hans Verkuil
` (2 subsequent siblings)
3 siblings, 0 replies; 11+ messages in thread
From: Thierry Reding @ 2019-01-10 12:51 UTC (permalink / raw)
To: Sakari Ailus; +Cc: linux-media, hverkuil, Ezequiel Garcia
[-- Attachment #1: Type: text/plain, Size: 2836 bytes --]
On Thu, Jan 10, 2019 at 02:43:19PM +0200, Sakari Ailus wrote:
> The for loop to reset the memory of the plane reserved fields runs over
> num_planes provided by the user without validating it. Ensure num_planes
> is no more than VIDEO_MAX_PLANES before the loop.
>
> Fixes: 4e1e0eb0e074 ("media: v4l2-ioctl: Zero v4l2_plane_pix_format reserved fields")
> Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
> ---
> Hi folks,
>
> This patch goes on top of Thierry's patch "media: v4l2-ioctl: Clear only
> per-plane reserved fields".
>
> drivers/media/v4l2-core/v4l2-ioctl.c | 8 ++++++++
> 1 file changed, 8 insertions(+)
>
> diff --git a/drivers/media/v4l2-core/v4l2-ioctl.c b/drivers/media/v4l2-core/v4l2-ioctl.c
> index 392f1228af7b5..9e68a608ac6d3 100644
> --- a/drivers/media/v4l2-core/v4l2-ioctl.c
> +++ b/drivers/media/v4l2-core/v4l2-ioctl.c
> @@ -1551,6 +1551,8 @@ static int v4l_s_fmt(const struct v4l2_ioctl_ops *ops,
> if (unlikely(!ops->vidioc_s_fmt_vid_cap_mplane))
> break;
> CLEAR_AFTER_FIELD(p, fmt.pix_mp.xfer_func);
> + if (p->fmt.pix_mp.num_planes > VIDEO_MAX_PLANES)
> + break;
> for (i = 0; i < p->fmt.pix_mp.num_planes; i++)
> CLEAR_AFTER_FIELD(&p->fmt.pix_mp.plane_fmt[i], bytesperline);
> return ops->vidioc_s_fmt_vid_cap_mplane(file, fh, arg);
> @@ -1581,6 +1583,8 @@ static int v4l_s_fmt(const struct v4l2_ioctl_ops *ops,
> if (unlikely(!ops->vidioc_s_fmt_vid_out_mplane))
> break;
> CLEAR_AFTER_FIELD(p, fmt.pix_mp.xfer_func);
> + if (p->fmt.pix_mp.num_planes > VIDEO_MAX_PLANES)
> + break;
> for (i = 0; i < p->fmt.pix_mp.num_planes; i++)
> CLEAR_AFTER_FIELD(&p->fmt.pix_mp.plane_fmt[i], bytesperline);
> return ops->vidioc_s_fmt_vid_out_mplane(file, fh, arg);
> @@ -1648,6 +1652,8 @@ static int v4l_try_fmt(const struct v4l2_ioctl_ops *ops,
> if (unlikely(!ops->vidioc_try_fmt_vid_cap_mplane))
> break;
> CLEAR_AFTER_FIELD(p, fmt.pix_mp.xfer_func);
> + if (p->fmt.pix_mp.num_planes > VIDEO_MAX_PLANES)
> + break;
> for (i = 0; i < p->fmt.pix_mp.num_planes; i++)
> CLEAR_AFTER_FIELD(&p->fmt.pix_mp.plane_fmt[i], bytesperline);
> return ops->vidioc_try_fmt_vid_cap_mplane(file, fh, arg);
> @@ -1678,6 +1684,8 @@ static int v4l_try_fmt(const struct v4l2_ioctl_ops *ops,
> if (unlikely(!ops->vidioc_try_fmt_vid_out_mplane))
> break;
> CLEAR_AFTER_FIELD(p, fmt.pix_mp.xfer_func);
> + if (p->fmt.pix_mp.num_planes > VIDEO_MAX_PLANES)
> + break;
> for (i = 0; i < p->fmt.pix_mp.num_planes; i++)
> CLEAR_AFTER_FIELD(&p->fmt.pix_mp.plane_fmt[i], bytesperline);
> return ops->vidioc_try_fmt_vid_out_mplane(file, fh, arg);
Do we also want to validate the instance in v4l_print_format() before
using it?
Reviewed-by: Thierry Reding <treding@nvidia.com>
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]
^ permalink raw reply [flat|nested] 11+ messages in thread* Re: [PATCH 1/1] v4l: ioctl: Validate num_planes before using it
2019-01-10 12:43 [PATCH 1/1] v4l: ioctl: Validate num_planes before using it Sakari Ailus
2019-01-10 12:51 ` Thierry Reding
@ 2019-01-10 13:02 ` Hans Verkuil
2019-01-10 13:30 ` Sakari Ailus
2019-01-10 13:41 ` Sakari Ailus
2019-01-10 14:12 ` Hans Verkuil
2019-02-15 15:52 ` Ezequiel Garcia
3 siblings, 2 replies; 11+ messages in thread
From: Hans Verkuil @ 2019-01-10 13:02 UTC (permalink / raw)
To: Sakari Ailus, linux-media; +Cc: Thierry Reding, Ezequiel Garcia
On 01/10/19 13:43, Sakari Ailus wrote:
> The for loop to reset the memory of the plane reserved fields runs over
> num_planes provided by the user without validating it. Ensure num_planes
> is no more than VIDEO_MAX_PLANES before the loop.
>
> Fixes: 4e1e0eb0e074 ("media: v4l2-ioctl: Zero v4l2_plane_pix_format reserved fields")
> Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
> ---
> Hi folks,
>
> This patch goes on top of Thierry's patch "media: v4l2-ioctl: Clear only
> per-plane reserved fields".
>
> drivers/media/v4l2-core/v4l2-ioctl.c | 8 ++++++++
> 1 file changed, 8 insertions(+)
>
> diff --git a/drivers/media/v4l2-core/v4l2-ioctl.c b/drivers/media/v4l2-core/v4l2-ioctl.c
> index 392f1228af7b5..9e68a608ac6d3 100644
> --- a/drivers/media/v4l2-core/v4l2-ioctl.c
> +++ b/drivers/media/v4l2-core/v4l2-ioctl.c
> @@ -1551,6 +1551,8 @@ static int v4l_s_fmt(const struct v4l2_ioctl_ops *ops,
> if (unlikely(!ops->vidioc_s_fmt_vid_cap_mplane))
> break;
> CLEAR_AFTER_FIELD(p, fmt.pix_mp.xfer_func);
> + if (p->fmt.pix_mp.num_planes > VIDEO_MAX_PLANES)
> + break;
I would check this not here but in check_fmt: all *_FMT ioctls call that function
first, so it makes sense to do the check there.
v4l_print_format should also be adjusted (take the minimum of num_planes and
VIDEO_MAX_PLANES), since it can still be called even if check_fmt returns an
error if num_planes is too large.
In fact, the change to v4l_print_format should be a separate patch since that
should be backported. It can leak memory in the kernel log if num_planes is
too large.
Regards,
Hans
> for (i = 0; i < p->fmt.pix_mp.num_planes; i++)
> CLEAR_AFTER_FIELD(&p->fmt.pix_mp.plane_fmt[i], bytesperline);
> return ops->vidioc_s_fmt_vid_cap_mplane(file, fh, arg);
> @@ -1581,6 +1583,8 @@ static int v4l_s_fmt(const struct v4l2_ioctl_ops *ops,
> if (unlikely(!ops->vidioc_s_fmt_vid_out_mplane))
> break;
> CLEAR_AFTER_FIELD(p, fmt.pix_mp.xfer_func);
> + if (p->fmt.pix_mp.num_planes > VIDEO_MAX_PLANES)
> + break;
> for (i = 0; i < p->fmt.pix_mp.num_planes; i++)
> CLEAR_AFTER_FIELD(&p->fmt.pix_mp.plane_fmt[i], bytesperline);
> return ops->vidioc_s_fmt_vid_out_mplane(file, fh, arg);
> @@ -1648,6 +1652,8 @@ static int v4l_try_fmt(const struct v4l2_ioctl_ops *ops,
> if (unlikely(!ops->vidioc_try_fmt_vid_cap_mplane))
> break;
> CLEAR_AFTER_FIELD(p, fmt.pix_mp.xfer_func);
> + if (p->fmt.pix_mp.num_planes > VIDEO_MAX_PLANES)
> + break;
> for (i = 0; i < p->fmt.pix_mp.num_planes; i++)
> CLEAR_AFTER_FIELD(&p->fmt.pix_mp.plane_fmt[i], bytesperline);
> return ops->vidioc_try_fmt_vid_cap_mplane(file, fh, arg);
> @@ -1678,6 +1684,8 @@ static int v4l_try_fmt(const struct v4l2_ioctl_ops *ops,
> if (unlikely(!ops->vidioc_try_fmt_vid_out_mplane))
> break;
> CLEAR_AFTER_FIELD(p, fmt.pix_mp.xfer_func);
> + if (p->fmt.pix_mp.num_planes > VIDEO_MAX_PLANES)
> + break;
> for (i = 0; i < p->fmt.pix_mp.num_planes; i++)
> CLEAR_AFTER_FIELD(&p->fmt.pix_mp.plane_fmt[i], bytesperline);
> return ops->vidioc_try_fmt_vid_out_mplane(file, fh, arg);
>
^ permalink raw reply [flat|nested] 11+ messages in thread* Re: [PATCH 1/1] v4l: ioctl: Validate num_planes before using it
2019-01-10 13:02 ` Hans Verkuil
@ 2019-01-10 13:30 ` Sakari Ailus
2019-01-10 13:41 ` Sakari Ailus
1 sibling, 0 replies; 11+ messages in thread
From: Sakari Ailus @ 2019-01-10 13:30 UTC (permalink / raw)
To: Hans Verkuil; +Cc: linux-media, Thierry Reding, Ezequiel Garcia
Hi Hans,
On Thu, Jan 10, 2019 at 02:02:14PM +0100, Hans Verkuil wrote:
> On 01/10/19 13:43, Sakari Ailus wrote:
> > The for loop to reset the memory of the plane reserved fields runs over
> > num_planes provided by the user without validating it. Ensure num_planes
> > is no more than VIDEO_MAX_PLANES before the loop.
> >
> > Fixes: 4e1e0eb0e074 ("media: v4l2-ioctl: Zero v4l2_plane_pix_format reserved fields")
> > Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
> > ---
> > Hi folks,
> >
> > This patch goes on top of Thierry's patch "media: v4l2-ioctl: Clear only
> > per-plane reserved fields".
> >
> > drivers/media/v4l2-core/v4l2-ioctl.c | 8 ++++++++
> > 1 file changed, 8 insertions(+)
> >
> > diff --git a/drivers/media/v4l2-core/v4l2-ioctl.c b/drivers/media/v4l2-core/v4l2-ioctl.c
> > index 392f1228af7b5..9e68a608ac6d3 100644
> > --- a/drivers/media/v4l2-core/v4l2-ioctl.c
> > +++ b/drivers/media/v4l2-core/v4l2-ioctl.c
> > @@ -1551,6 +1551,8 @@ static int v4l_s_fmt(const struct v4l2_ioctl_ops *ops,
> > if (unlikely(!ops->vidioc_s_fmt_vid_cap_mplane))
> > break;
> > CLEAR_AFTER_FIELD(p, fmt.pix_mp.xfer_func);
> > + if (p->fmt.pix_mp.num_planes > VIDEO_MAX_PLANES)
> > + break;
>
> I would check this not here but in check_fmt: all *_FMT ioctls call that function
> first, so it makes sense to do the check there.
check_fmt() just takes the buffer type, no more parameters than that. But
I'm fine with introducing another function for the purpose, perhaps
renaming check_fmt() as check_buf_type().
I'll send v2.
>
> v4l_print_format should also be adjusted (take the minimum of num_planes and
> VIDEO_MAX_PLANES), since it can still be called even if check_fmt returns an
> error if num_planes is too large.
>
> In fact, the change to v4l_print_format should be a separate patch since that
> should be backported. It can leak memory in the kernel log if num_planes is
> too large.
Yes. I'm trying to figure out the affected versions. The code has been
mangled multiple times over the years so it's not entirely trivial.
>
> Regards,
>
> Hans
>
> > for (i = 0; i < p->fmt.pix_mp.num_planes; i++)
> > CLEAR_AFTER_FIELD(&p->fmt.pix_mp.plane_fmt[i], bytesperline);
> > return ops->vidioc_s_fmt_vid_cap_mplane(file, fh, arg);
> > @@ -1581,6 +1583,8 @@ static int v4l_s_fmt(const struct v4l2_ioctl_ops *ops,
> > if (unlikely(!ops->vidioc_s_fmt_vid_out_mplane))
> > break;
> > CLEAR_AFTER_FIELD(p, fmt.pix_mp.xfer_func);
> > + if (p->fmt.pix_mp.num_planes > VIDEO_MAX_PLANES)
> > + break;
> > for (i = 0; i < p->fmt.pix_mp.num_planes; i++)
> > CLEAR_AFTER_FIELD(&p->fmt.pix_mp.plane_fmt[i], bytesperline);
> > return ops->vidioc_s_fmt_vid_out_mplane(file, fh, arg);
> > @@ -1648,6 +1652,8 @@ static int v4l_try_fmt(const struct v4l2_ioctl_ops *ops,
> > if (unlikely(!ops->vidioc_try_fmt_vid_cap_mplane))
> > break;
> > CLEAR_AFTER_FIELD(p, fmt.pix_mp.xfer_func);
> > + if (p->fmt.pix_mp.num_planes > VIDEO_MAX_PLANES)
> > + break;
> > for (i = 0; i < p->fmt.pix_mp.num_planes; i++)
> > CLEAR_AFTER_FIELD(&p->fmt.pix_mp.plane_fmt[i], bytesperline);
> > return ops->vidioc_try_fmt_vid_cap_mplane(file, fh, arg);
> > @@ -1678,6 +1684,8 @@ static int v4l_try_fmt(const struct v4l2_ioctl_ops *ops,
> > if (unlikely(!ops->vidioc_try_fmt_vid_out_mplane))
> > break;
> > CLEAR_AFTER_FIELD(p, fmt.pix_mp.xfer_func);
> > + if (p->fmt.pix_mp.num_planes > VIDEO_MAX_PLANES)
> > + break;
> > for (i = 0; i < p->fmt.pix_mp.num_planes; i++)
> > CLEAR_AFTER_FIELD(&p->fmt.pix_mp.plane_fmt[i], bytesperline);
> > return ops->vidioc_try_fmt_vid_out_mplane(file, fh, arg);
> >
>
--
Sakari Ailus
sakari.ailus@linux.intel.com
^ permalink raw reply [flat|nested] 11+ messages in thread* Re: [PATCH 1/1] v4l: ioctl: Validate num_planes before using it
2019-01-10 13:02 ` Hans Verkuil
2019-01-10 13:30 ` Sakari Ailus
@ 2019-01-10 13:41 ` Sakari Ailus
2019-01-10 14:11 ` Hans Verkuil
1 sibling, 1 reply; 11+ messages in thread
From: Sakari Ailus @ 2019-01-10 13:41 UTC (permalink / raw)
To: Hans Verkuil; +Cc: linux-media, Thierry Reding, Ezequiel Garcia
On Thu, Jan 10, 2019 at 02:02:14PM +0100, Hans Verkuil wrote:
> On 01/10/19 13:43, Sakari Ailus wrote:
> > The for loop to reset the memory of the plane reserved fields runs over
> > num_planes provided by the user without validating it. Ensure num_planes
> > is no more than VIDEO_MAX_PLANES before the loop.
> >
> > Fixes: 4e1e0eb0e074 ("media: v4l2-ioctl: Zero v4l2_plane_pix_format reserved fields")
> > Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
> > ---
> > Hi folks,
> >
> > This patch goes on top of Thierry's patch "media: v4l2-ioctl: Clear only
> > per-plane reserved fields".
> >
> > drivers/media/v4l2-core/v4l2-ioctl.c | 8 ++++++++
> > 1 file changed, 8 insertions(+)
> >
> > diff --git a/drivers/media/v4l2-core/v4l2-ioctl.c b/drivers/media/v4l2-core/v4l2-ioctl.c
> > index 392f1228af7b5..9e68a608ac6d3 100644
> > --- a/drivers/media/v4l2-core/v4l2-ioctl.c
> > +++ b/drivers/media/v4l2-core/v4l2-ioctl.c
> > @@ -1551,6 +1551,8 @@ static int v4l_s_fmt(const struct v4l2_ioctl_ops *ops,
> > if (unlikely(!ops->vidioc_s_fmt_vid_cap_mplane))
> > break;
> > CLEAR_AFTER_FIELD(p, fmt.pix_mp.xfer_func);
> > + if (p->fmt.pix_mp.num_planes > VIDEO_MAX_PLANES)
> > + break;
>
> I would check this not here but in check_fmt: all *_FMT ioctls call that function
> first, so it makes sense to do the check there.
Even G_FMT? I'm not saying no though. There's just a slight chance of
breaking something as it hasn't been a problem to call G_FMT with incorrect
number of planes in the mplane format; the number would be overwritten
anyway.
Apart from that, this leaves just the four locations --- putting this to a
separate function will reduce the calls to that function to just two.
>
> v4l_print_format should also be adjusted (take the minimum of num_planes and
> VIDEO_MAX_PLANES), since it can still be called even if check_fmt returns an
> error if num_planes is too large.
>
> In fact, the change to v4l_print_format should be a separate patch since that
> should be backported. It can leak memory in the kernel log if num_planes is
> too large.
>
> Regards,
>
> Hans
>
> > for (i = 0; i < p->fmt.pix_mp.num_planes; i++)
> > CLEAR_AFTER_FIELD(&p->fmt.pix_mp.plane_fmt[i], bytesperline);
> > return ops->vidioc_s_fmt_vid_cap_mplane(file, fh, arg);
> > @@ -1581,6 +1583,8 @@ static int v4l_s_fmt(const struct v4l2_ioctl_ops *ops,
> > if (unlikely(!ops->vidioc_s_fmt_vid_out_mplane))
> > break;
> > CLEAR_AFTER_FIELD(p, fmt.pix_mp.xfer_func);
> > + if (p->fmt.pix_mp.num_planes > VIDEO_MAX_PLANES)
> > + break;
> > for (i = 0; i < p->fmt.pix_mp.num_planes; i++)
> > CLEAR_AFTER_FIELD(&p->fmt.pix_mp.plane_fmt[i], bytesperline);
> > return ops->vidioc_s_fmt_vid_out_mplane(file, fh, arg);
> > @@ -1648,6 +1652,8 @@ static int v4l_try_fmt(const struct v4l2_ioctl_ops *ops,
> > if (unlikely(!ops->vidioc_try_fmt_vid_cap_mplane))
> > break;
> > CLEAR_AFTER_FIELD(p, fmt.pix_mp.xfer_func);
> > + if (p->fmt.pix_mp.num_planes > VIDEO_MAX_PLANES)
> > + break;
> > for (i = 0; i < p->fmt.pix_mp.num_planes; i++)
> > CLEAR_AFTER_FIELD(&p->fmt.pix_mp.plane_fmt[i], bytesperline);
> > return ops->vidioc_try_fmt_vid_cap_mplane(file, fh, arg);
> > @@ -1678,6 +1684,8 @@ static int v4l_try_fmt(const struct v4l2_ioctl_ops *ops,
> > if (unlikely(!ops->vidioc_try_fmt_vid_out_mplane))
> > break;
> > CLEAR_AFTER_FIELD(p, fmt.pix_mp.xfer_func);
> > + if (p->fmt.pix_mp.num_planes > VIDEO_MAX_PLANES)
> > + break;
> > for (i = 0; i < p->fmt.pix_mp.num_planes; i++)
> > CLEAR_AFTER_FIELD(&p->fmt.pix_mp.plane_fmt[i], bytesperline);
> > return ops->vidioc_try_fmt_vid_out_mplane(file, fh, arg);
> >
>
--
Sakari Ailus
sakari.ailus@linux.intel.com
^ permalink raw reply [flat|nested] 11+ messages in thread* Re: [PATCH 1/1] v4l: ioctl: Validate num_planes before using it
2019-01-10 13:41 ` Sakari Ailus
@ 2019-01-10 14:11 ` Hans Verkuil
2019-01-10 14:18 ` Sakari Ailus
0 siblings, 1 reply; 11+ messages in thread
From: Hans Verkuil @ 2019-01-10 14:11 UTC (permalink / raw)
To: Sakari Ailus; +Cc: linux-media, Thierry Reding, Ezequiel Garcia
On 01/10/19 14:41, Sakari Ailus wrote:
> On Thu, Jan 10, 2019 at 02:02:14PM +0100, Hans Verkuil wrote:
>> On 01/10/19 13:43, Sakari Ailus wrote:
>>> The for loop to reset the memory of the plane reserved fields runs over
>>> num_planes provided by the user without validating it. Ensure num_planes
>>> is no more than VIDEO_MAX_PLANES before the loop.
>>>
>>> Fixes: 4e1e0eb0e074 ("media: v4l2-ioctl: Zero v4l2_plane_pix_format reserved fields")
>>> Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
>>> ---
>>> Hi folks,
>>>
>>> This patch goes on top of Thierry's patch "media: v4l2-ioctl: Clear only
>>> per-plane reserved fields".
>>>
>>> drivers/media/v4l2-core/v4l2-ioctl.c | 8 ++++++++
>>> 1 file changed, 8 insertions(+)
>>>
>>> diff --git a/drivers/media/v4l2-core/v4l2-ioctl.c b/drivers/media/v4l2-core/v4l2-ioctl.c
>>> index 392f1228af7b5..9e68a608ac6d3 100644
>>> --- a/drivers/media/v4l2-core/v4l2-ioctl.c
>>> +++ b/drivers/media/v4l2-core/v4l2-ioctl.c
>>> @@ -1551,6 +1551,8 @@ static int v4l_s_fmt(const struct v4l2_ioctl_ops *ops,
>>> if (unlikely(!ops->vidioc_s_fmt_vid_cap_mplane))
>>> break;
>>> CLEAR_AFTER_FIELD(p, fmt.pix_mp.xfer_func);
>>> + if (p->fmt.pix_mp.num_planes > VIDEO_MAX_PLANES)
>>> + break;
>>
>> I would check this not here but in check_fmt: all *_FMT ioctls call that function
>> first, so it makes sense to do the check there.
>
> Even G_FMT? I'm not saying no though. There's just a slight chance of
> breaking something as it hasn't been a problem to call G_FMT with incorrect
> number of planes in the mplane format; the number would be overwritten
> anyway.
Not G_FMT since everything after the type field is zeroed.
> Apart from that, this leaves just the four locations --- putting this to a
> separate function will reduce the calls to that function to just two.
I was a bit too quick with my comment about check_fmt. Your original patch
is fine.
I'll Ack your patch.
Regards,
Hans
>
>>
>> v4l_print_format should also be adjusted (take the minimum of num_planes and
>> VIDEO_MAX_PLANES), since it can still be called even if check_fmt returns an
>> error if num_planes is too large.
>>
>> In fact, the change to v4l_print_format should be a separate patch since that
>> should be backported. It can leak memory in the kernel log if num_planes is
>> too large.
>>
>> Regards,
>>
>> Hans
>>
>>> for (i = 0; i < p->fmt.pix_mp.num_planes; i++)
>>> CLEAR_AFTER_FIELD(&p->fmt.pix_mp.plane_fmt[i], bytesperline);
>>> return ops->vidioc_s_fmt_vid_cap_mplane(file, fh, arg);
>>> @@ -1581,6 +1583,8 @@ static int v4l_s_fmt(const struct v4l2_ioctl_ops *ops,
>>> if (unlikely(!ops->vidioc_s_fmt_vid_out_mplane))
>>> break;
>>> CLEAR_AFTER_FIELD(p, fmt.pix_mp.xfer_func);
>>> + if (p->fmt.pix_mp.num_planes > VIDEO_MAX_PLANES)
>>> + break;
>>> for (i = 0; i < p->fmt.pix_mp.num_planes; i++)
>>> CLEAR_AFTER_FIELD(&p->fmt.pix_mp.plane_fmt[i], bytesperline);
>>> return ops->vidioc_s_fmt_vid_out_mplane(file, fh, arg);
>>> @@ -1648,6 +1652,8 @@ static int v4l_try_fmt(const struct v4l2_ioctl_ops *ops,
>>> if (unlikely(!ops->vidioc_try_fmt_vid_cap_mplane))
>>> break;
>>> CLEAR_AFTER_FIELD(p, fmt.pix_mp.xfer_func);
>>> + if (p->fmt.pix_mp.num_planes > VIDEO_MAX_PLANES)
>>> + break;
>>> for (i = 0; i < p->fmt.pix_mp.num_planes; i++)
>>> CLEAR_AFTER_FIELD(&p->fmt.pix_mp.plane_fmt[i], bytesperline);
>>> return ops->vidioc_try_fmt_vid_cap_mplane(file, fh, arg);
>>> @@ -1678,6 +1684,8 @@ static int v4l_try_fmt(const struct v4l2_ioctl_ops *ops,
>>> if (unlikely(!ops->vidioc_try_fmt_vid_out_mplane))
>>> break;
>>> CLEAR_AFTER_FIELD(p, fmt.pix_mp.xfer_func);
>>> + if (p->fmt.pix_mp.num_planes > VIDEO_MAX_PLANES)
>>> + break;
>>> for (i = 0; i < p->fmt.pix_mp.num_planes; i++)
>>> CLEAR_AFTER_FIELD(&p->fmt.pix_mp.plane_fmt[i], bytesperline);
>>> return ops->vidioc_try_fmt_vid_out_mplane(file, fh, arg);
>>>
>>
>
^ permalink raw reply [flat|nested] 11+ messages in thread* Re: [PATCH 1/1] v4l: ioctl: Validate num_planes before using it
2019-01-10 14:11 ` Hans Verkuil
@ 2019-01-10 14:18 ` Sakari Ailus
0 siblings, 0 replies; 11+ messages in thread
From: Sakari Ailus @ 2019-01-10 14:18 UTC (permalink / raw)
To: Hans Verkuil; +Cc: linux-media, Thierry Reding, Ezequiel Garcia
On Thu, Jan 10, 2019 at 03:11:35PM +0100, Hans Verkuil wrote:
> On 01/10/19 14:41, Sakari Ailus wrote:
> > On Thu, Jan 10, 2019 at 02:02:14PM +0100, Hans Verkuil wrote:
> >> On 01/10/19 13:43, Sakari Ailus wrote:
> >>> The for loop to reset the memory of the plane reserved fields runs over
> >>> num_planes provided by the user without validating it. Ensure num_planes
> >>> is no more than VIDEO_MAX_PLANES before the loop.
> >>>
> >>> Fixes: 4e1e0eb0e074 ("media: v4l2-ioctl: Zero v4l2_plane_pix_format reserved fields")
> >>> Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
> >>> ---
> >>> Hi folks,
> >>>
> >>> This patch goes on top of Thierry's patch "media: v4l2-ioctl: Clear only
> >>> per-plane reserved fields".
> >>>
> >>> drivers/media/v4l2-core/v4l2-ioctl.c | 8 ++++++++
> >>> 1 file changed, 8 insertions(+)
> >>>
> >>> diff --git a/drivers/media/v4l2-core/v4l2-ioctl.c b/drivers/media/v4l2-core/v4l2-ioctl.c
> >>> index 392f1228af7b5..9e68a608ac6d3 100644
> >>> --- a/drivers/media/v4l2-core/v4l2-ioctl.c
> >>> +++ b/drivers/media/v4l2-core/v4l2-ioctl.c
> >>> @@ -1551,6 +1551,8 @@ static int v4l_s_fmt(const struct v4l2_ioctl_ops *ops,
> >>> if (unlikely(!ops->vidioc_s_fmt_vid_cap_mplane))
> >>> break;
> >>> CLEAR_AFTER_FIELD(p, fmt.pix_mp.xfer_func);
> >>> + if (p->fmt.pix_mp.num_planes > VIDEO_MAX_PLANES)
> >>> + break;
> >>
> >> I would check this not here but in check_fmt: all *_FMT ioctls call that function
> >> first, so it makes sense to do the check there.
> >
> > Even G_FMT? I'm not saying no though. There's just a slight chance of
> > breaking something as it hasn't been a problem to call G_FMT with incorrect
> > number of planes in the mplane format; the number would be overwritten
> > anyway.
>
> Not G_FMT since everything after the type field is zeroed.
>
> > Apart from that, this leaves just the four locations --- putting this to a
> > separate function will reduce the calls to that function to just two.
>
> I was a bit too quick with my comment about check_fmt. Your original patch
> is fine.
>
> I'll Ack your patch.
Thanks. Could you pick this to your tree as you did Thierry's patch?
I'll submit the other patch in a moment.
>
> Regards,
>
> Hans
>
> >
> >>
> >> v4l_print_format should also be adjusted (take the minimum of num_planes and
> >> VIDEO_MAX_PLANES), since it can still be called even if check_fmt returns an
> >> error if num_planes is too large.
> >>
> >> In fact, the change to v4l_print_format should be a separate patch since that
> >> should be backported. It can leak memory in the kernel log if num_planes is
> >> too large.
> >>
> >> Regards,
> >>
> >> Hans
> >>
> >>> for (i = 0; i < p->fmt.pix_mp.num_planes; i++)
> >>> CLEAR_AFTER_FIELD(&p->fmt.pix_mp.plane_fmt[i], bytesperline);
> >>> return ops->vidioc_s_fmt_vid_cap_mplane(file, fh, arg);
> >>> @@ -1581,6 +1583,8 @@ static int v4l_s_fmt(const struct v4l2_ioctl_ops *ops,
> >>> if (unlikely(!ops->vidioc_s_fmt_vid_out_mplane))
> >>> break;
> >>> CLEAR_AFTER_FIELD(p, fmt.pix_mp.xfer_func);
> >>> + if (p->fmt.pix_mp.num_planes > VIDEO_MAX_PLANES)
> >>> + break;
> >>> for (i = 0; i < p->fmt.pix_mp.num_planes; i++)
> >>> CLEAR_AFTER_FIELD(&p->fmt.pix_mp.plane_fmt[i], bytesperline);
> >>> return ops->vidioc_s_fmt_vid_out_mplane(file, fh, arg);
> >>> @@ -1648,6 +1652,8 @@ static int v4l_try_fmt(const struct v4l2_ioctl_ops *ops,
> >>> if (unlikely(!ops->vidioc_try_fmt_vid_cap_mplane))
> >>> break;
> >>> CLEAR_AFTER_FIELD(p, fmt.pix_mp.xfer_func);
> >>> + if (p->fmt.pix_mp.num_planes > VIDEO_MAX_PLANES)
> >>> + break;
> >>> for (i = 0; i < p->fmt.pix_mp.num_planes; i++)
> >>> CLEAR_AFTER_FIELD(&p->fmt.pix_mp.plane_fmt[i], bytesperline);
> >>> return ops->vidioc_try_fmt_vid_cap_mplane(file, fh, arg);
> >>> @@ -1678,6 +1684,8 @@ static int v4l_try_fmt(const struct v4l2_ioctl_ops *ops,
> >>> if (unlikely(!ops->vidioc_try_fmt_vid_out_mplane))
> >>> break;
> >>> CLEAR_AFTER_FIELD(p, fmt.pix_mp.xfer_func);
> >>> + if (p->fmt.pix_mp.num_planes > VIDEO_MAX_PLANES)
> >>> + break;
> >>> for (i = 0; i < p->fmt.pix_mp.num_planes; i++)
> >>> CLEAR_AFTER_FIELD(&p->fmt.pix_mp.plane_fmt[i], bytesperline);
> >>> return ops->vidioc_try_fmt_vid_out_mplane(file, fh, arg);
> >>>
> >>
> >
>
--
Sakari Ailus
sakari.ailus@linux.intel.com
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [PATCH 1/1] v4l: ioctl: Validate num_planes before using it
2019-01-10 12:43 [PATCH 1/1] v4l: ioctl: Validate num_planes before using it Sakari Ailus
2019-01-10 12:51 ` Thierry Reding
2019-01-10 13:02 ` Hans Verkuil
@ 2019-01-10 14:12 ` Hans Verkuil
2019-02-15 15:52 ` Ezequiel Garcia
3 siblings, 0 replies; 11+ messages in thread
From: Hans Verkuil @ 2019-01-10 14:12 UTC (permalink / raw)
To: Sakari Ailus, linux-media; +Cc: Thierry Reding, Ezequiel Garcia
On 01/10/19 13:43, Sakari Ailus wrote:
> The for loop to reset the memory of the plane reserved fields runs over
> num_planes provided by the user without validating it. Ensure num_planes
> is no more than VIDEO_MAX_PLANES before the loop.
>
> Fixes: 4e1e0eb0e074 ("media: v4l2-ioctl: Zero v4l2_plane_pix_format reserved fields")
> Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Reviewed-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Regards,
Hans
> ---
> Hi folks,
>
> This patch goes on top of Thierry's patch "media: v4l2-ioctl: Clear only
> per-plane reserved fields".
>
> drivers/media/v4l2-core/v4l2-ioctl.c | 8 ++++++++
> 1 file changed, 8 insertions(+)
>
> diff --git a/drivers/media/v4l2-core/v4l2-ioctl.c b/drivers/media/v4l2-core/v4l2-ioctl.c
> index 392f1228af7b5..9e68a608ac6d3 100644
> --- a/drivers/media/v4l2-core/v4l2-ioctl.c
> +++ b/drivers/media/v4l2-core/v4l2-ioctl.c
> @@ -1551,6 +1551,8 @@ static int v4l_s_fmt(const struct v4l2_ioctl_ops *ops,
> if (unlikely(!ops->vidioc_s_fmt_vid_cap_mplane))
> break;
> CLEAR_AFTER_FIELD(p, fmt.pix_mp.xfer_func);
> + if (p->fmt.pix_mp.num_planes > VIDEO_MAX_PLANES)
> + break;
> for (i = 0; i < p->fmt.pix_mp.num_planes; i++)
> CLEAR_AFTER_FIELD(&p->fmt.pix_mp.plane_fmt[i], bytesperline);
> return ops->vidioc_s_fmt_vid_cap_mplane(file, fh, arg);
> @@ -1581,6 +1583,8 @@ static int v4l_s_fmt(const struct v4l2_ioctl_ops *ops,
> if (unlikely(!ops->vidioc_s_fmt_vid_out_mplane))
> break;
> CLEAR_AFTER_FIELD(p, fmt.pix_mp.xfer_func);
> + if (p->fmt.pix_mp.num_planes > VIDEO_MAX_PLANES)
> + break;
> for (i = 0; i < p->fmt.pix_mp.num_planes; i++)
> CLEAR_AFTER_FIELD(&p->fmt.pix_mp.plane_fmt[i], bytesperline);
> return ops->vidioc_s_fmt_vid_out_mplane(file, fh, arg);
> @@ -1648,6 +1652,8 @@ static int v4l_try_fmt(const struct v4l2_ioctl_ops *ops,
> if (unlikely(!ops->vidioc_try_fmt_vid_cap_mplane))
> break;
> CLEAR_AFTER_FIELD(p, fmt.pix_mp.xfer_func);
> + if (p->fmt.pix_mp.num_planes > VIDEO_MAX_PLANES)
> + break;
> for (i = 0; i < p->fmt.pix_mp.num_planes; i++)
> CLEAR_AFTER_FIELD(&p->fmt.pix_mp.plane_fmt[i], bytesperline);
> return ops->vidioc_try_fmt_vid_cap_mplane(file, fh, arg);
> @@ -1678,6 +1684,8 @@ static int v4l_try_fmt(const struct v4l2_ioctl_ops *ops,
> if (unlikely(!ops->vidioc_try_fmt_vid_out_mplane))
> break;
> CLEAR_AFTER_FIELD(p, fmt.pix_mp.xfer_func);
> + if (p->fmt.pix_mp.num_planes > VIDEO_MAX_PLANES)
> + break;
> for (i = 0; i < p->fmt.pix_mp.num_planes; i++)
> CLEAR_AFTER_FIELD(&p->fmt.pix_mp.plane_fmt[i], bytesperline);
> return ops->vidioc_try_fmt_vid_out_mplane(file, fh, arg);
>
^ permalink raw reply [flat|nested] 11+ messages in thread* Re: [PATCH 1/1] v4l: ioctl: Validate num_planes before using it
2019-01-10 12:43 [PATCH 1/1] v4l: ioctl: Validate num_planes before using it Sakari Ailus
` (2 preceding siblings ...)
2019-01-10 14:12 ` Hans Verkuil
@ 2019-02-15 15:52 ` Ezequiel Garcia
2019-02-15 16:05 ` Hans Verkuil
3 siblings, 1 reply; 11+ messages in thread
From: Ezequiel Garcia @ 2019-02-15 15:52 UTC (permalink / raw)
To: Sakari Ailus, linux-media, hverkuil; +Cc: Thierry Reding
On Thu, 2019-01-10 at 14:43 +0200, Sakari Ailus wrote:
> The for loop to reset the memory of the plane reserved fields runs over
> num_planes provided by the user without validating it. Ensure num_planes
> is no more than VIDEO_MAX_PLANES before the loop.
>
> Fixes: 4e1e0eb0e074 ("media: v4l2-ioctl: Zero v4l2_plane_pix_format reserved fields")
> Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
> ---
> Hi folks,
>
> This patch goes on top of Thierry's patch "media: v4l2-ioctl: Clear only
> per-plane reserved fields".
>
> drivers/media/v4l2-core/v4l2-ioctl.c | 8 ++++++++
> 1 file changed, 8 insertions(+)
>
> diff --git a/drivers/media/v4l2-core/v4l2-ioctl.c b/drivers/media/v4l2-core/v4l2-ioctl.c
> index 392f1228af7b5..9e68a608ac6d3 100644
> --- a/drivers/media/v4l2-core/v4l2-ioctl.c
> +++ b/drivers/media/v4l2-core/v4l2-ioctl.c
> @@ -1551,6 +1551,8 @@ static int v4l_s_fmt(const struct v4l2_ioctl_ops *ops,
> if (unlikely(!ops->vidioc_s_fmt_vid_cap_mplane))
> break;
> CLEAR_AFTER_FIELD(p, fmt.pix_mp.xfer_func);
> + if (p->fmt.pix_mp.num_planes > VIDEO_MAX_PLANES)
> + break;
> for (i = 0; i < p->fmt.pix_mp.num_planes; i++)
> CLEAR_AFTER_FIELD(&p->fmt.pix_mp.plane_fmt[i], bytesperline);
> return ops->vidioc_s_fmt_vid_cap_mplane(file, fh, arg);
> @@ -1581,6 +1583,8 @@ static int v4l_s_fmt(const struct v4l2_ioctl_ops *ops,
> if (unlikely(!ops->vidioc_s_fmt_vid_out_mplane))
> break;
> CLEAR_AFTER_FIELD(p, fmt.pix_mp.xfer_func);
> + if (p->fmt.pix_mp.num_planes > VIDEO_MAX_PLANES)
> + break;
> for (i = 0; i < p->fmt.pix_mp.num_planes; i++)
> CLEAR_AFTER_FIELD(&p->fmt.pix_mp.plane_fmt[i], bytesperline);
> return ops->vidioc_s_fmt_vid_out_mplane(file, fh, arg);
> @@ -1648,6 +1652,8 @@ static int v4l_try_fmt(const struct v4l2_ioctl_ops *ops,
> if (unlikely(!ops->vidioc_try_fmt_vid_cap_mplane))
> break;
> CLEAR_AFTER_FIELD(p, fmt.pix_mp.xfer_func);
> + if (p->fmt.pix_mp.num_planes > VIDEO_MAX_PLANES)
> + break;
> for (i = 0; i < p->fmt.pix_mp.num_planes; i++)
> CLEAR_AFTER_FIELD(&p->fmt.pix_mp.plane_fmt[i], bytesperline);
> return ops->vidioc_try_fmt_vid_cap_mplane(file, fh, arg);
> @@ -1678,6 +1684,8 @@ static int v4l_try_fmt(const struct v4l2_ioctl_ops *ops,
> if (unlikely(!ops->vidioc_try_fmt_vid_out_mplane))
> break;
> CLEAR_AFTER_FIELD(p, fmt.pix_mp.xfer_func);
> + if (p->fmt.pix_mp.num_planes > VIDEO_MAX_PLANES)
> + break;
> for (i = 0; i < p->fmt.pix_mp.num_planes; i++)
> CLEAR_AFTER_FIELD(&p->fmt.pix_mp.plane_fmt[i], bytesperline);
> return ops->vidioc_try_fmt_vid_out_mplane(file, fh, arg);
I'm under the impression this commit is causing the following v4l2-compliance warning:
TRY_FMT cannot handle an invalid pixelformat.
This may or may not be a problem. For more information see:
http://www.mail-archive.com/linux-media@vger.kernel.org/msg56550.html
This applies to try_fmt and s_fmt.
After checking the v4l2 spec, it seems we can only error on a bad type field [1]:
"Drivers should not return an error code unless the type field is invalid".
Perhaps we can clamp num_planes to VIDEO_MAX_PLANES, instead of erroring out?
[1] https://www.kernel.org/doc/html/latest/media/uapi/v4l/vidioc-g-fmt.html
Regards,
Eze
^ permalink raw reply [flat|nested] 11+ messages in thread* Re: [PATCH 1/1] v4l: ioctl: Validate num_planes before using it
2019-02-15 15:52 ` Ezequiel Garcia
@ 2019-02-15 16:05 ` Hans Verkuil
2019-02-15 16:14 ` Ezequiel Garcia
0 siblings, 1 reply; 11+ messages in thread
From: Hans Verkuil @ 2019-02-15 16:05 UTC (permalink / raw)
To: Ezequiel Garcia, Sakari Ailus, linux-media; +Cc: Thierry Reding
On 2/15/19 4:52 PM, Ezequiel Garcia wrote:
> On Thu, 2019-01-10 at 14:43 +0200, Sakari Ailus wrote:
>> The for loop to reset the memory of the plane reserved fields runs over
>> num_planes provided by the user without validating it. Ensure num_planes
>> is no more than VIDEO_MAX_PLANES before the loop.
>>
>> Fixes: 4e1e0eb0e074 ("media: v4l2-ioctl: Zero v4l2_plane_pix_format reserved fields")
>> Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
>> ---
>> Hi folks,
>>
>> This patch goes on top of Thierry's patch "media: v4l2-ioctl: Clear only
>> per-plane reserved fields".
>>
>> drivers/media/v4l2-core/v4l2-ioctl.c | 8 ++++++++
>> 1 file changed, 8 insertions(+)
>>
>> diff --git a/drivers/media/v4l2-core/v4l2-ioctl.c b/drivers/media/v4l2-core/v4l2-ioctl.c
>> index 392f1228af7b5..9e68a608ac6d3 100644
>> --- a/drivers/media/v4l2-core/v4l2-ioctl.c
>> +++ b/drivers/media/v4l2-core/v4l2-ioctl.c
>> @@ -1551,6 +1551,8 @@ static int v4l_s_fmt(const struct v4l2_ioctl_ops *ops,
>> if (unlikely(!ops->vidioc_s_fmt_vid_cap_mplane))
>> break;
>> CLEAR_AFTER_FIELD(p, fmt.pix_mp.xfer_func);
>> + if (p->fmt.pix_mp.num_planes > VIDEO_MAX_PLANES)
>> + break;
>> for (i = 0; i < p->fmt.pix_mp.num_planes; i++)
>> CLEAR_AFTER_FIELD(&p->fmt.pix_mp.plane_fmt[i], bytesperline);
>> return ops->vidioc_s_fmt_vid_cap_mplane(file, fh, arg);
>> @@ -1581,6 +1583,8 @@ static int v4l_s_fmt(const struct v4l2_ioctl_ops *ops,
>> if (unlikely(!ops->vidioc_s_fmt_vid_out_mplane))
>> break;
>> CLEAR_AFTER_FIELD(p, fmt.pix_mp.xfer_func);
>> + if (p->fmt.pix_mp.num_planes > VIDEO_MAX_PLANES)
>> + break;
>> for (i = 0; i < p->fmt.pix_mp.num_planes; i++)
>> CLEAR_AFTER_FIELD(&p->fmt.pix_mp.plane_fmt[i], bytesperline);
>> return ops->vidioc_s_fmt_vid_out_mplane(file, fh, arg);
>> @@ -1648,6 +1652,8 @@ static int v4l_try_fmt(const struct v4l2_ioctl_ops *ops,
>> if (unlikely(!ops->vidioc_try_fmt_vid_cap_mplane))
>> break;
>> CLEAR_AFTER_FIELD(p, fmt.pix_mp.xfer_func);
>> + if (p->fmt.pix_mp.num_planes > VIDEO_MAX_PLANES)
>> + break;
>> for (i = 0; i < p->fmt.pix_mp.num_planes; i++)
>> CLEAR_AFTER_FIELD(&p->fmt.pix_mp.plane_fmt[i], bytesperline);
>> return ops->vidioc_try_fmt_vid_cap_mplane(file, fh, arg);
>> @@ -1678,6 +1684,8 @@ static int v4l_try_fmt(const struct v4l2_ioctl_ops *ops,
>> if (unlikely(!ops->vidioc_try_fmt_vid_out_mplane))
>> break;
>> CLEAR_AFTER_FIELD(p, fmt.pix_mp.xfer_func);
>> + if (p->fmt.pix_mp.num_planes > VIDEO_MAX_PLANES)
>> + break;
>> for (i = 0; i < p->fmt.pix_mp.num_planes; i++)
>> CLEAR_AFTER_FIELD(&p->fmt.pix_mp.plane_fmt[i], bytesperline);
>> return ops->vidioc_try_fmt_vid_out_mplane(file, fh, arg);
>
> I'm under the impression this commit is causing the following v4l2-compliance warning:
>
> TRY_FMT cannot handle an invalid pixelformat.
> This may or may not be a problem. For more information see:
> http://www.mail-archive.com/linux-media@vger.kernel.org/msg56550.html
>
> This applies to try_fmt and s_fmt.
>
> After checking the v4l2 spec, it seems we can only error on a bad type field [1]:
>
> "Drivers should not return an error code unless the type field is invalid".
>
> Perhaps we can clamp num_planes to VIDEO_MAX_PLANES, instead of erroring out?
Yes, that would be the right approach.
Ezequiel, can you clamp num_planes and test the patch? And post it if it fixes the
warning?
This really should have been caught by testing with e.g. vivid and v4l2-compliance.
Regards,
Hans
>
> [1] https://www.kernel.org/doc/html/latest/media/uapi/v4l/vidioc-g-fmt.html
>
> Regards,
> Eze
>
^ permalink raw reply [flat|nested] 11+ messages in thread* Re: [PATCH 1/1] v4l: ioctl: Validate num_planes before using it
2019-02-15 16:05 ` Hans Verkuil
@ 2019-02-15 16:14 ` Ezequiel Garcia
0 siblings, 0 replies; 11+ messages in thread
From: Ezequiel Garcia @ 2019-02-15 16:14 UTC (permalink / raw)
To: Hans Verkuil, Sakari Ailus, linux-media; +Cc: Thierry Reding
On Fri, 2019-02-15 at 17:05 +0100, Hans Verkuil wrote:
> On 2/15/19 4:52 PM, Ezequiel Garcia wrote:
> > On Thu, 2019-01-10 at 14:43 +0200, Sakari Ailus wrote:
> > > The for loop to reset the memory of the plane reserved fields runs over
> > > num_planes provided by the user without validating it. Ensure num_planes
> > > is no more than VIDEO_MAX_PLANES before the loop.
> > >
> > > Fixes: 4e1e0eb0e074 ("media: v4l2-ioctl: Zero v4l2_plane_pix_format reserved fields")
> > > Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
> > > ---
> > > Hi folks,
> > >
> > > This patch goes on top of Thierry's patch "media: v4l2-ioctl: Clear only
> > > per-plane reserved fields".
> > >
> > > drivers/media/v4l2-core/v4l2-ioctl.c | 8 ++++++++
> > > 1 file changed, 8 insertions(+)
> > >
> > > diff --git a/drivers/media/v4l2-core/v4l2-ioctl.c b/drivers/media/v4l2-core/v4l2-ioctl.c
> > > index 392f1228af7b5..9e68a608ac6d3 100644
> > > --- a/drivers/media/v4l2-core/v4l2-ioctl.c
> > > +++ b/drivers/media/v4l2-core/v4l2-ioctl.c
> > > @@ -1551,6 +1551,8 @@ static int v4l_s_fmt(const struct v4l2_ioctl_ops *ops,
> > > if (unlikely(!ops->vidioc_s_fmt_vid_cap_mplane))
> > > break;
> > > CLEAR_AFTER_FIELD(p, fmt.pix_mp.xfer_func);
> > > + if (p->fmt.pix_mp.num_planes > VIDEO_MAX_PLANES)
> > > + break;
> > > for (i = 0; i < p->fmt.pix_mp.num_planes; i++)
> > > CLEAR_AFTER_FIELD(&p->fmt.pix_mp.plane_fmt[i], bytesperline);
> > > return ops->vidioc_s_fmt_vid_cap_mplane(file, fh, arg);
> > > @@ -1581,6 +1583,8 @@ static int v4l_s_fmt(const struct v4l2_ioctl_ops *ops,
> > > if (unlikely(!ops->vidioc_s_fmt_vid_out_mplane))
> > > break;
> > > CLEAR_AFTER_FIELD(p, fmt.pix_mp.xfer_func);
> > > + if (p->fmt.pix_mp.num_planes > VIDEO_MAX_PLANES)
> > > + break;
> > > for (i = 0; i < p->fmt.pix_mp.num_planes; i++)
> > > CLEAR_AFTER_FIELD(&p->fmt.pix_mp.plane_fmt[i], bytesperline);
> > > return ops->vidioc_s_fmt_vid_out_mplane(file, fh, arg);
> > > @@ -1648,6 +1652,8 @@ static int v4l_try_fmt(const struct v4l2_ioctl_ops *ops,
> > > if (unlikely(!ops->vidioc_try_fmt_vid_cap_mplane))
> > > break;
> > > CLEAR_AFTER_FIELD(p, fmt.pix_mp.xfer_func);
> > > + if (p->fmt.pix_mp.num_planes > VIDEO_MAX_PLANES)
> > > + break;
> > > for (i = 0; i < p->fmt.pix_mp.num_planes; i++)
> > > CLEAR_AFTER_FIELD(&p->fmt.pix_mp.plane_fmt[i], bytesperline);
> > > return ops->vidioc_try_fmt_vid_cap_mplane(file, fh, arg);
> > > @@ -1678,6 +1684,8 @@ static int v4l_try_fmt(const struct v4l2_ioctl_ops *ops,
> > > if (unlikely(!ops->vidioc_try_fmt_vid_out_mplane))
> > > break;
> > > CLEAR_AFTER_FIELD(p, fmt.pix_mp.xfer_func);
> > > + if (p->fmt.pix_mp.num_planes > VIDEO_MAX_PLANES)
> > > + break;
> > > for (i = 0; i < p->fmt.pix_mp.num_planes; i++)
> > > CLEAR_AFTER_FIELD(&p->fmt.pix_mp.plane_fmt[i], bytesperline);
> > > return ops->vidioc_try_fmt_vid_out_mplane(file, fh, arg);
> >
> > I'm under the impression this commit is causing the following v4l2-compliance warning:
> >
> > TRY_FMT cannot handle an invalid pixelformat.
> > This may or may not be a problem. For more information see:
> > http://www.mail-archive.com/linux-media@vger.kernel.org/msg56550.html
> >
> > This applies to try_fmt and s_fmt.
> >
> > After checking the v4l2 spec, it seems we can only error on a bad type field [1]:
> >
> > "Drivers should not return an error code unless the type field is invalid".
> >
> > Perhaps we can clamp num_planes to VIDEO_MAX_PLANES, instead of erroring out?
>
> Yes, that would be the right approach.
>
> Ezequiel, can you clamp num_planes and test the patch? And post it if it fixes the
> warning?
>
I can try :-)
> This really should have been caught by testing with e.g. vivid and v4l2-compliance.
>
We need a CI!
> Regards,
>
> Hans
>
> > [1] https://www.kernel.org/doc/html/latest/media/uapi/v4l/vidioc-g-fmt.html
> >
> > Regards,
> > Eze
> >
^ permalink raw reply [flat|nested] 11+ messages in thread