From mboxrd@z Thu Jan  1 00:00:00 1970
From: keith.busch@intel.com (Keith Busch)
Date: Tue, 21 Jan 2014 12:06:20 -0700 (MST)
Subject: [PATCH 2/2] nvme: Cleanup nvme_dev_start() and fix IRQ leak
In-Reply-To: <20140121100719.GC8847@dhcp-26-207.brq.redhat.com>
References: <cover.1389904166.git.agordeev@redhat.com>
 <1c441f670f33375b6c41e074baf6e84e6c7bb0c2.1389904166.git.agordeev@redhat.com>
 <CAErSpo4g_xm4vZWZDgyjtJu+z5ARSerQLrtJ6W_peXv4bKwcgg@mail.gmail.com>
 <alpine.LRH.2.03.1401171419370.25844@AMR>
 <20140120083835.GA19068@dhcp-26-207.brq.redhat.com>
 <20140120084212.GC19068@dhcp-26-207.brq.redhat.com>
 <alpine.LRH.2.03.1401201541000.25844@AMR>
 <20140121100311.GA8847@dhcp-26-207.brq.redhat.com>
 <20140121100719.GC8847@dhcp-26-207.brq.redhat.com>
Message-ID: <alpine.LRH.2.03.1401211151110.25844@AMR>

On Tue, 21 Jan 2014, Alexander Gordeev wrote:
> This is an attempt to make handling of admin queue in a
> single scope. This update also fixes a IRQ leak in case
> nvme_setup_io_queues() failed to allocate enough iomem
> and bailed out with -ENOMEM errno.
>
> Signed-off-by: Alexander Gordeev <agordeev at redhat.com>
> ---

> +static void nvme_teardown_admin_queue(struct nvme_dev *dev)
> +{
> +	nvme_disable_queue(dev, 0);
> +	nvme_free_queue(dev->queues[0]);
> +}

> @@ -2402,11 +2398,20 @@ static int nvme_dev_start(struct nvme_dev *dev)
> 	list_add(&dev->node, &dev_list);
> 	spin_unlock(&dev_list_lock);
>
> -	result = nvme_setup_io_queues(dev);
> -	if (result && result != -EBUSY)
> +	result = set_queue_count(dev, num_online_cpus());
> +	if (result == -EBUSY)
> +		return -EBUSY;
> +
> +	nvme_teardown_admin_queue(dev);

Oh no! Your new teardown function is freeing the admin queue, but it
would be used immediatly after that in nvme_setup_io_queues ...

> +
> +	if (result)
> 		goto disable;

... but you'll never actually get to setup io queues because the 'result'
here is non-zero if we were successful, and is the number of queues the
controller can allocate. I think you meant to do this instead:

+	if (result < 0)

>
> -	return result;
> +	result = nvme_setup_io_queues(dev, result);
> +	if (result)
> +		goto disable;
> +
> +	return 0;