Date: Fri, 26 Feb 2010 12:06:49 -0800 (PST)
From: Linus Torvalds <torvalds@linux-foundation.org>
To: Larry Finger <Larry.Finger@lwfinger.net>
cc: "John W. Linville" <linville@tuxdriver.com>,
	"David S. Miller" <davem@davemloft.net>,
	wireless <linux-wireless@vger.kernel.org>,
	Greg Kroah-Hartman <gregkh@suse.de>
Subject: Re: Make b43 driver fall back gracefully to PIO mode after fatal
 DMA errors
In-Reply-To: <alpine.LFD.2.00.1002261112590.4513@localhost.localdomain>
Message-ID: <alpine.LFD.2.00.1002261205160.4513@localhost.localdomain>
References: <alpine.LFD.2.00.1002261034140.4513@localhost.localdomain> <4B881C70.9030004@lwfinger.net> <alpine.LFD.2.00.1002261112590.4513@localhost.localdomain>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: linux-wireless-owner@vger.kernel.org


On Fri, 26 Feb 2010, Linus Torvalds wrote:
> On Fri, 26 Feb 2010, Larry Finger wrote:
> > 
> > Incidentally, we have learned today that the fault does not occur when running
> > MMIO traces, and that it is not SMP related.
> 
> Timing? I guess MMIO tracing will slow down the actual command setup part 
> a lot.

Adding a "udelay(10)" to the "b43_write[16|32]()" functions doesn't make 
any difference on the machine I have access to, though.

Ill happily test other ideas or patches. Automatically falling back to PIO 
makes things work, but I'd obviously _much_ rather have working DMA.

			Linus