diff mbox

[next] usb: typec: unlock dp->lock on error exit path, and also zero ret if successful

Message ID 20180704124611.9852-1-colin.king@canonical.com (mailing list archive)
State New, archived
Headers show

Commit Message

Colin King July 4, 2018, 12:46 p.m. UTC
From: Colin Ian King <colin.king@canonical.com>

One of the error handling paths forgets to unlock dp->lock on the error
exit path leading to a potential lock-up.  Also the return path for a
successful call to the function configuration_store can return an
uninitialized error return code if dp->alt->active is false, so ensure
ret is zeroed on the successful exit path to avoid garbage being returned.

Detected by CoverityScan, CID#1471597 ("Unitialized scalar variable")

Fixes: 0e3bb7d6894d ("usb: typec: Add driver for DisplayPort alternate mode")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
---
 drivers/usb/typec/altmodes/displayport.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

Comments

Heikki Krogerus July 5, 2018, 1:18 p.m. UTC | #1
On Wed, Jul 04, 2018 at 01:46:11PM +0100, Colin King wrote:
> From: Colin Ian King <colin.king@canonical.com>
> 
> One of the error handling paths forgets to unlock dp->lock on the error
> exit path leading to a potential lock-up.  Also the return path for a
> successful call to the function configuration_store can return an
> uninitialized error return code if dp->alt->active is false, so ensure
> ret is zeroed on the successful exit path to avoid garbage being returned.
> 
> Detected by CoverityScan, CID#1471597 ("Unitialized scalar variable")
> 
> Fixes: 0e3bb7d6894d ("usb: typec: Add driver for DisplayPort alternate mode")
> Signed-off-by: Colin Ian King <colin.king@canonical.com>
> ---
>  drivers/usb/typec/altmodes/displayport.c | 7 +++++--
>  1 file changed, 5 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/usb/typec/altmodes/displayport.c b/drivers/usb/typec/altmodes/displayport.c
> index ef12b15bd484..9a637c2b4460 100644
> --- a/drivers/usb/typec/altmodes/displayport.c
> +++ b/drivers/usb/typec/altmodes/displayport.c
> @@ -349,8 +349,10 @@ configuration_store(struct device *dev, struct device_attribute *attr,
>  	cap = DP_CAP_CAPABILITY(dp->alt->vdo);
>  
>  	if ((con == DP_CONF_DFP_D && !(cap & DP_CAP_DFP_D)) ||
> -	    (con == DP_CONF_UFP_D && !(cap & DP_CAP_UFP_D)))
> -		return -EINVAL;
> +	    (con == DP_CONF_UFP_D && !(cap & DP_CAP_UFP_D))) {
> +		ret = -EINVAL;
> +		goto err_unlock;
> +	}
>  
>  	conf = dp->data.conf & ~DP_CONF_DUAL_D;
>  	conf |= con;
> @@ -362,6 +364,7 @@ configuration_store(struct device *dev, struct device_attribute *attr,
>  	}
>  
>  	dp->data.conf = conf;
> +	ret = 0;

Could you set the value when the variable is introduced:

        int ret = 0;


Thank you,
diff mbox

Patch

diff --git a/drivers/usb/typec/altmodes/displayport.c b/drivers/usb/typec/altmodes/displayport.c
index ef12b15bd484..9a637c2b4460 100644
--- a/drivers/usb/typec/altmodes/displayport.c
+++ b/drivers/usb/typec/altmodes/displayport.c
@@ -349,8 +349,10 @@  configuration_store(struct device *dev, struct device_attribute *attr,
 	cap = DP_CAP_CAPABILITY(dp->alt->vdo);
 
 	if ((con == DP_CONF_DFP_D && !(cap & DP_CAP_DFP_D)) ||
-	    (con == DP_CONF_UFP_D && !(cap & DP_CAP_UFP_D)))
-		return -EINVAL;
+	    (con == DP_CONF_UFP_D && !(cap & DP_CAP_UFP_D))) {
+		ret = -EINVAL;
+		goto err_unlock;
+	}
 
 	conf = dp->data.conf & ~DP_CONF_DUAL_D;
 	conf |= con;
@@ -362,6 +364,7 @@  configuration_store(struct device *dev, struct device_attribute *attr,
 	}
 
 	dp->data.conf = conf;
+	ret = 0;
 
 err_unlock:
 	mutex_unlock(&dp->lock);