[PATCH] cptm1217: check if interrupts are masked at probe

[sanjeev sharma]

On Fri, Jan 2, 2015 at 11:47 PM, Devendra Naga <devendra.aaru at gmail.com> wrote:
> the function cp_tm1217_mask_interrupt can return failure.
> added the check and the failure path.
>
> Cc: Ramesh Agarwal <ramesh.agarwal at intel.com>
> Signed-off-by: Devendra Naga <devendra.aaru at gmail.com>
> ---
>
>  build tested only on x86_64. config is allmodconfig.
>
>  drivers/staging/cptm1217/clearpad_tm1217.c | 6 ++++++
>  1 file changed, 6 insertions(+)
>
> diff --git a/drivers/staging/cptm1217/clearpad_tm1217.c b/drivers/staging/cptm1217/clearpad_tm1217.c
> index 7f265ce..54e5953 100644
> --- a/drivers/staging/cptm1217/clearpad_tm1217.c
> +++ b/drivers/staging/cptm1217/clearpad_tm1217.c
> @@ -446,6 +446,12 @@ static int cp_tm1217_probe(struct i2c_client *client,
>
>         /* Mask all the interrupts */
>         retval = cp_tm1217_mask_interrupt(ts);
> +       if (retval) {
> +               dev_err(ts->dev, "failed to mask interrupts, error: %d\n",
> +                       retval);
> +               kfree(ts);

Here you are doing more than what you have specified in change-log.How
did you find out memory leak ? Did you used any facility like kmemleak
facility to find this
problem ?

> +               return retval;
> +       }
>
>         /* Read the controller information */
>         cp_tm1217_init_data(ts);
> --
> 1.9.3
>
> _______________________________________________
> devel mailing list
> devel at linuxdriverproject.org
> http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel