2019-03-09 07:06:16

by Kangjie Lu

[permalink] [raw]
Subject: [PATCH] media: rcar-vin: fix a potential NULL pointer dereference

In case of_match_node cannot find a match, the fix returns
-EINVAL to avoid NULL pointer dereference.

Signed-off-by: Kangjie Lu <[email protected]>
---
drivers/media/platform/rcar-vin/rcar-core.c | 2 ++
1 file changed, 2 insertions(+)

diff --git a/drivers/media/platform/rcar-vin/rcar-core.c b/drivers/media/platform/rcar-vin/rcar-core.c
index f0719ce24b97..a058e2023ca8 100644
--- a/drivers/media/platform/rcar-vin/rcar-core.c
+++ b/drivers/media/platform/rcar-vin/rcar-core.c
@@ -266,6 +266,8 @@ static int rvin_group_init(struct rvin_group *group, struct rvin_dev *vin)

match = of_match_node(vin->dev->driver->of_match_table,
vin->dev->of_node);
+ if (unlikely(!match))
+ return -EINVAL;

strscpy(mdev->driver_name, KBUILD_MODNAME, sizeof(mdev->driver_name));
strscpy(mdev->model, match->compatible, sizeof(mdev->model));
--
2.17.1



2019-03-09 11:03:00

by Niklas Söderlund

[permalink] [raw]
Subject: Re: [PATCH] media: rcar-vin: fix a potential NULL pointer dereference

Hi Kangjie,

Thanks for your patch.

On 2019-03-09 01:05:27 -0600, Kangjie Lu wrote:
> In case of_match_node cannot find a match, the fix returns
> -EINVAL to avoid NULL pointer dereference.
>
> Signed-off-by: Kangjie Lu <[email protected]>
> ---
> drivers/media/platform/rcar-vin/rcar-core.c | 2 ++
> 1 file changed, 2 insertions(+)
>
> diff --git a/drivers/media/platform/rcar-vin/rcar-core.c b/drivers/media/platform/rcar-vin/rcar-core.c
> index f0719ce24b97..a058e2023ca8 100644
> --- a/drivers/media/platform/rcar-vin/rcar-core.c
> +++ b/drivers/media/platform/rcar-vin/rcar-core.c
> @@ -266,6 +266,8 @@ static int rvin_group_init(struct rvin_group *group, struct rvin_dev *vin)
>
> match = of_match_node(vin->dev->driver->of_match_table,
> vin->dev->of_node);
> + if (unlikely(!match))
> + return -EINVAL;

I don't think this is needed. The driver depends on selects OF and if we
get this far we it is because we had a match already. The reason to call
of_match_node() here is simply to retrieve which of the possible
compatible strings was matched.

Am I missing something? What scenario do you see where this can fail?

>
> strscpy(mdev->driver_name, KBUILD_MODNAME, sizeof(mdev->driver_name));
> strscpy(mdev->model, match->compatible, sizeof(mdev->model));
> --
> 2.17.1
>

--
Regards,
Niklas S?derlund

2019-03-28 15:27:05

by Hans Verkuil

[permalink] [raw]
Subject: Re: [PATCH] media: rcar-vin: fix a potential NULL pointer dereference

On 3/9/19 12:02 PM, Niklas Söderlund wrote:
> Hi Kangjie,
>
> Thanks for your patch.
>
> On 2019-03-09 01:05:27 -0600, Kangjie Lu wrote:
>> In case of_match_node cannot find a match, the fix returns
>> -EINVAL to avoid NULL pointer dereference.
>>
>> Signed-off-by: Kangjie Lu <[email protected]>
>> ---
>> drivers/media/platform/rcar-vin/rcar-core.c | 2 ++
>> 1 file changed, 2 insertions(+)
>>
>> diff --git a/drivers/media/platform/rcar-vin/rcar-core.c b/drivers/media/platform/rcar-vin/rcar-core.c
>> index f0719ce24b97..a058e2023ca8 100644
>> --- a/drivers/media/platform/rcar-vin/rcar-core.c
>> +++ b/drivers/media/platform/rcar-vin/rcar-core.c
>> @@ -266,6 +266,8 @@ static int rvin_group_init(struct rvin_group *group, struct rvin_dev *vin)
>>
>> match = of_match_node(vin->dev->driver->of_match_table,
>> vin->dev->of_node);
>> + if (unlikely(!match))
>> + return -EINVAL;
>
> I don't think this is needed. The driver depends on selects OF and if we
> get this far we it is because we had a match already. The reason to call
> of_match_node() here is simply to retrieve which of the possible
> compatible strings was matched.
>
> Am I missing something? What scenario do you see where this can fail?

And even if it can fail, then please drop the 'unlikely'.

Regards,

Hans

>
>>
>> strscpy(mdev->driver_name, KBUILD_MODNAME, sizeof(mdev->driver_name));
>> strscpy(mdev->model, match->compatible, sizeof(mdev->model));
>> --
>> 2.17.1
>>
>