Smatch warns:
drivers/media/i2c/ds90ub960.c:1671 ub960_rxport_add_serializer():
err: 'rxport->ser.client' dereferencing possible ERR_PTR()
i2c_new_client_device() returns error pointers on failure and in
dev_dbg statement we are dereferencing error pointer which is a bug.
Fix this by using IS_ERR() which checks for error pointers.
Fixes: afe267f2d368 ("media: i2c: add DS90UB960 driver")
Signed-off-by: Harshit Mogalapalli <[email protected]>
---
Found with static analysis, only compile tested. Although in
drivers/media i2c_client_has_driver() checks are present, IS_ERR() would
probably be sufficient here.
---
drivers/media/i2c/ds90ub960.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/media/i2c/ds90ub960.c b/drivers/media/i2c/ds90ub960.c
index e101bcf2356a..88144e3ec183 100644
--- a/drivers/media/i2c/ds90ub960.c
+++ b/drivers/media/i2c/ds90ub960.c
@@ -1662,7 +1662,7 @@ static int ub960_rxport_add_serializer(struct ub960_data *priv, u8 nport)
ser_info.addr = rxport->ser.alias;
rxport->ser.client =
i2c_new_client_device(priv->client->adapter, &ser_info);
- if (!rxport->ser.client) {
+ if (IS_ERR(rxport->ser.client)) {
dev_err(dev, "rx%u: cannot add %s i2c device", nport,
ser_info.type);
return -EIO;
--
2.39.3
Hi,
On 18/07/2023 11:58, Harshit Mogalapalli wrote:
> Smatch warns:
> drivers/media/i2c/ds90ub960.c:1671 ub960_rxport_add_serializer():
> err: 'rxport->ser.client' dereferencing possible ERR_PTR()
>
> i2c_new_client_device() returns error pointers on failure and in
> dev_dbg statement we are dereferencing error pointer which is a bug.
>
> Fix this by using IS_ERR() which checks for error pointers.
>
> Fixes: afe267f2d368 ("media: i2c: add DS90UB960 driver")
> Signed-off-by: Harshit Mogalapalli <[email protected]>
> ---
> Found with static analysis, only compile tested. Although in
> drivers/media i2c_client_has_driver() checks are present, IS_ERR() would
> probably be sufficient here.
> ---
> drivers/media/i2c/ds90ub960.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/media/i2c/ds90ub960.c b/drivers/media/i2c/ds90ub960.c
> index e101bcf2356a..88144e3ec183 100644
> --- a/drivers/media/i2c/ds90ub960.c
> +++ b/drivers/media/i2c/ds90ub960.c
> @@ -1662,7 +1662,7 @@ static int ub960_rxport_add_serializer(struct ub960_data *priv, u8 nport)
> ser_info.addr = rxport->ser.alias;
> rxport->ser.client =
> i2c_new_client_device(priv->client->adapter, &ser_info);
> - if (!rxport->ser.client) {
> + if (IS_ERR(rxport->ser.client)) {
> dev_err(dev, "rx%u: cannot add %s i2c device", nport,
> ser_info.type);
> return -EIO;
Good catch. I think this should be modified to return
PTR_ERR(rxport->ser.client) instead of -EIO. Do you want to update the
patch or shall I do the change?
Tomi
Hi Tomi,
On 18/07/23 7:01 pm, Tomi Valkeinen wrote:
> Hi,
>
> On 18/07/2023 11:58, Harshit Mogalapalli wrote:
>> Smatch warns:
>> drivers/media/i2c/ds90ub960.c:1671 ub960_rxport_add_serializer():
>> err: 'rxport->ser.client' dereferencing possible ERR_PTR()
>>
>> i2c_new_client_device() returns error pointers on failure and in
>> dev_dbg statement we are dereferencing error pointer which is a bug.
>>
>> Fix this by using IS_ERR() which checks for error pointers.
>>
>> Fixes: afe267f2d368 ("media: i2c: add DS90UB960 driver")
>> Signed-off-by: Harshit Mogalapalli <[email protected]>
>> ---
>> Found with static analysis, only compile tested. Although in
>> drivers/media i2c_client_has_driver() checks are present, IS_ERR() would
>> probably be sufficient here.
>> ---
>> drivers/media/i2c/ds90ub960.c | 2 +-
>> 1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/drivers/media/i2c/ds90ub960.c
>> b/drivers/media/i2c/ds90ub960.c
>> index e101bcf2356a..88144e3ec183 100644
>> --- a/drivers/media/i2c/ds90ub960.c
>> +++ b/drivers/media/i2c/ds90ub960.c
>> @@ -1662,7 +1662,7 @@ static int ub960_rxport_add_serializer(struct
>> ub960_data *priv, u8 nport)
>> ser_info.addr = rxport->ser.alias;
>> rxport->ser.client =
>> i2c_new_client_device(priv->client->adapter, &ser_info);
>> - if (!rxport->ser.client) {
>> + if (IS_ERR(rxport->ser.client)) {
>> dev_err(dev, "rx%u: cannot add %s i2c device", nport,
>> ser_info.type);
>> return -EIO;
>
> Good catch. I think this should be modified to return
> PTR_ERR(rxport->ser.client) instead of -EIO. Do you want to update the
> patch or shall I do the change?
Thanks,
I think it is easier for me to do that change PTR_ERR change. Will send
V2 with that change added.
Regards,
Harshit
>
> Tomi
>
On Tue, Jul 18, 2023 at 01:58:46AM -0700, Harshit Mogalapalli wrote:
> Smatch warns:
> drivers/media/i2c/ds90ub960.c:1671 ub960_rxport_add_serializer():
> err: 'rxport->ser.client' dereferencing possible ERR_PTR()
>
> i2c_new_client_device() returns error pointers on failure and in
> dev_dbg statement we are dereferencing error pointer which is a bug.
>
> Fix this by using IS_ERR() which checks for error pointers.
...
> - if (!rxport->ser.client) {
> + if (IS_ERR(rxport->ser.client)) {
> dev_err(dev, "rx%u: cannot add %s i2c device", nport,
> ser_info.type);
> return -EIO;
It should be changed to return PTR_ERR(...);
--
With Best Regards,
Andy Shevchenko
Hi Andy,
On 18/07/23 7:28 pm, Andy Shevchenko wrote:
> On Tue, Jul 18, 2023 at 01:58:46AM -0700, Harshit Mogalapalli wrote:
>> Smatch warns:
>> drivers/media/i2c/ds90ub960.c:1671 ub960_rxport_add_serializer():
>> err: 'rxport->ser.client' dereferencing possible ERR_PTR()
>>
>> i2c_new_client_device() returns error pointers on failure and in
>> dev_dbg statement we are dereferencing error pointer which is a bug.
>>
>> Fix this by using IS_ERR() which checks for error pointers.
>
> ...
>
>> - if (!rxport->ser.client) {
>> + if (IS_ERR(rxport->ser.client)) {
>
>> dev_err(dev, "rx%u: cannot add %s i2c device", nport,
>> ser_info.type);
>> return -EIO;
>
> It should be changed to return PTR_ERR(...);
I just sent a V2 with this suggestion included.
Thanks,
Harshit
>