[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <1207632810.5728.15.camel@moss.renham>
Date: Tue, 08 Apr 2008 15:33:30 +1000
From: Ben Nizette <bn@...sdigital.com>
To: Szilveszter Ordog <slipszi@...il.com>
Cc: kernel <kernel@...32linux.org>,
Haavard Skinnemoen <hskinnemoen@...el.com>,
linux-kernel <linux-kernel@...r.kernel.org>, ed@...rson.com
Subject: Re: [BUG, bisected] atmel_spi: hard lockups in transfer
On Mon, 2008-04-07 at 02:24 +0200, Szilveszter Ordog wrote:
> However, if you add the following line
>
> #define DEBUG
>
> to the top of drivers/spi/atmel_spi.c (before the #include directives)
> and recompile the kernel, then you would get lots of debug output
> which could help us pinpoint the problem.
>
Right, I've attached the output from dmesg from the tester after he
added #define DEBUG to the top of atmel_spi.c. I don't know how helpful
it's going to be though, to my (untrained) eye everything looks in
order. I've also attached the config.
We have seen the problem reported across many SPI usages though in this
case the tester has serial flash connected over SPI as an MTD. Without
the #define DEBUG in there the board would hang while trying to mount
that device. With the #define in place the mount succeeds but a hang is
observed when trying to do anything with the device, eg ls.
And to clarify, I don't have the issue myself, I've just been helping
the tester debug the problem. He's now be added to the CC so we can get
some more direct communications :-)
Thanks,
--Ben.
View attachment "config.txt" of type "text/plain" (24203 bytes)
View attachment "dmesg.txt" of type "text/plain" (15674 bytes)
Powered by blists - more mailing lists