From mboxrd@z Thu Jan  1 00:00:00 1970
Message-ID: <51CB1E2C.2000302@xenomai.org>
Date: Wed, 26 Jun 2013 19:00:28 +0200
From: Philippe Gerum <rpm@xenomai.org>
MIME-Version: 1.0
References: <1372258188.12144.161.camel@lx-henry>
In-Reply-To: <1372258188.12144.161.camel@lx-henry>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Subject: Re: [Xenomai] Kernel Panic not syncing kernel stack overflow on
 AMCC PowerPC 460EX upon upgrade
List-Id: Discussions about the Xenomai project <xenomai.xenomai.org>
List-Unsubscribe: <http://www.xenomai.org/mailman/options/xenomai>,
	<mailto:xenomai-request@xenomai.org?subject=unsubscribe>
List-Archive: <http://www.xenomai.org/pipermail/xenomai>
List-Post: <mailto:xenomai@xenomai.org>
List-Help: <mailto:xenomai-request@xenomai.org?subject=help>
List-Subscribe: <http://www.xenomai.org/mailman/listinfo/xenomai>,
	<mailto:xenomai-request@xenomai.org?subject=subscribe>
To: Henry Bausley <hbausley@deltatau.com>
Cc: xenomai@xenomai.org

On 06/26/2013 04:49 PM, Henry Bausley wrote:
>
>
> I am attempting to upgrade to Xeno 2.6.2.1 / Ipipe 3.2.21 from
> a working Xeno 2.5.6 / Ipipe 2.6.30.3 configuration
>
> I am getting a Kernel Panic not syncing kernel stack overflow errors
> with the new kernel.
>
> The application is a kernel module that operates soley from interrupts
> and communicates to user space with applications with shared memory.  It
> has been running with the earlier kernels without issues for many years.
> Below are the messages I get when the system crashes.
>
> If anyone has any hints based upon the data when it crashes  below it
> would be greatly appreciated.
>
>

<snip>

> Call
> Trace:
> [8049bef0] [80006c00] show_stack+0x54/0x154
> (unreliable)
> [8049bf30] [8001a26c] __schedule_bug
> +0x68/0x6c
> [8049bf40] [80372b1c] __schedule
> +0x3e8/0x430
> [8049bf90] [80007738] cpu_idle
> +0xd8/0xdc
> [8049bfb0] [80001bd4] rest_init
> +0x64/0x78
> [8049bfc0] [8044a770] start_kernel
> +0x2d0/0x2e4
> [8049bff0] [80000044] _start
> +0x44/0x88
> Kernel panic - not syncing: kernel stack overflow

This is typical of some code badly trashing the linux interrupt state 
when running in primary domain. The idle loop breakage may be a 
collateral damage of such issue.

Could you tell us a bit more about your kernel config, e.g.

- which preemption model?
- did you enable CONFIG_IPIPE_DEBUG_INTERNAL? otherwise, do you have any 
warning notice when set?

-- 
Philippe.