Re: [PATCH] regulator: core: avoid memory access after freeing it

[Charles Keepax]

On Tue, Nov 06, 2012 at 04:04:09PM +0530, Laxman Dewangan wrote:
> When regulator_register() failed due to non availability of
>       mutex_unlock(&regulator_list_mutex);
...
>  }
>  EXPORT_SYMBOL_GPL(regulator_put);
> @@ -3453,11 +3460,10 @@ scrub:
>               gpio_free(rdev->ena_gpio);
>       kfree(rdev->constraints);
>  wash:
> +     if (rdev->supply)
> +             __regulator_put_locked(rdev->supply);
>       device_unregister(&rdev->dev);
> -
>       mutex_unlock(&regulator_list_mutex);
> -     if (rdev->supply)
> -             regulator_put(rdev->supply);
>  
>       /* device core frees rdev */
>       rdev = ERR_PTR(ret);

Here would it not make sense to go slightly further and basically
revert my original change. This feels sensible as it keeps it at
only one location where the mutex is unlocked and removes the
second exit point I added. This would effectively make the code
look like:

scrub:
        if (rdev->supply)
                __regulator_put_locked(rdev->supply);
        if (rdev->ena_gpio)
                gpio_free(rdev->ena_gpio);
        kfree(rdev->constraints);
        device_unregister(&rdev->dev);
        /* device core frees rdev */
        rdev = ERR_PTR(ret);
        goto out;

Otherwise the change looks great to me.

> -- 
> 1.7.1.1
> 
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/