Re: [announce] HVCS for inclusion in 2.6 tree

[Andrew Morton <akpm@osdl.org>]

Ryan Arnold <rsa@us.ibm.com> wrote:
>
> +static int hvcs_get_pi(struct hvcs_struct *hvcsd)
> +{
> +	/* struct hvcs_partner_info *head_pi = NULL; */
> +	struct hvcs_partner_info *pi = NULL;
> +	unsigned int unit_address = hvcsd->vdev->unit_address;
> +	struct list_head head;
> +	unsigned long flags;
> +	int retval;
> +
> +	spin_lock_irqsave(&hvcs_pi_lock, flags);
> +	if (!hvcs_pi_buff) {
> +		spin_unlock_irqrestore(&hvcs_pi_lock, flags);
> +		return -EFAULT;
> +	}
> +	retval = hvcs_get_partner_info(unit_address, &head, hvcs_pi_buff);
> +	if (retval) {
> +		printk(KERN_ERR "HVCS: Failed to fetch partner"
> +			" info for vty-server@%x.\n",unit_address);
> +		spin_lock_irqsave(&hvcs_pi_lock, flags);

                ^^ spin_unlock

> +		return retval;
> +	}
> +	spin_unlock_irqrestore(&hvcs_pi_lock, flags);
> 

(Just move the spin_unlock() to be prior to the test of retval).

Suggest you review all similar code in the patch for the same copy-n-paste
bug.