From mboxrd@z Thu Jan  1 00:00:00 1970
From: David Miller <davem@davemloft.net>
Subject: Re: [PATCH 3/9] [I/OAT] Setup the networking subsystem as a DMA
 client
Date: Tue, 23 May 2006 18:13:55 -0700 (PDT)
Message-ID: <20060523.181355.115913162.davem@davemloft.net>
References: <20060524001653.19403.31396.stgit@gitlost.site>
	<20060524002014.19403.93371.stgit@gitlost.site>
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Cc: linux-kernel@vger.kernel.org, netdev@vger.kernel.org
Return-path: <netdev-owner@vger.kernel.org>
Received: from dsl027-180-168.sfo1.dsl.speakeasy.net ([216.27.180.168]:36480
	"EHLO sunset.sfo1.dsl.speakeasy.net") by vger.kernel.org with ESMTP
	id S932524AbWEXBNz (ORCPT <rfc822;netdev@vger.kernel.org>);
	Tue, 23 May 2006 21:13:55 -0400
To: christopher.leech@intel.com
In-Reply-To: <20060524002014.19403.93371.stgit@gitlost.site>
Sender: netdev-owner@vger.kernel.org
List-Id: netdev.vger.kernel.org

From: Chris Leech <christopher.leech@intel.com>
Date: Tue, 23 May 2006 17:20:15 -0700

> +static void net_dma_rebalance(void)
> +{
> +	unsigned int cpu, i, n;
> +	struct dma_chan *chan;
> +
> +	lock_cpu_hotplug();

You can't call lock_cpu_hotplug(), because that sleeps and takes
semaphores and we currently hold a spinlock taken here:

> +static void netdev_dma_event(struct dma_client *client, struct dma_chan *chan,
> +	enum dma_event event)
> +{
> +	spin_lock(&net_dma_event_lock);
> +	switch (event) {
> +	case DMA_RESOURCE_ADDED:
> +		net_dma_count++;
> +		net_dma_rebalance();
> +		break;

You'll need to run this DMA rebalancing asynchronously in process
context via keventd or similar to deal with this locking bug.