em interrupt storm

[Stefan Bethke]

Am 24.11.2005 um 10:36 schrieb Kris Kennaway:

> On Thu, Nov 24, 2005 at 10:28:59AM +0100, Stefan Bethke wrote:
>>
>> Am 24.11.2005 um 07:26 schrieb Kris Kennaway:
>>
>>> On Thu, Nov 24, 2005 at 06:24:17AM +0100, Stefan Bethke wrote:
>>>> Am 24.11.2005 um 00:15 schrieb Scott Long:
>>>>
>>>>> I've directly observed it on 7501WV2 and 7520BD2 boards.  I don't
>>>>> know if it's a problem that will affect other board configurations
>>>>> or cousin chipsets like the 7505 and 7525.
>>>>
>>>> We've got a Tyan i7501 Pro that we have had stability issues with
>>>> since the beginning.  Is it thinkable that the interrupt masking
>>>> could lead to a hard lock (no break to debugger possible)?
>>>
>>> Sounds like a different problem.  Enable KDB_STOP_NMI in your  
>>> kernel,
>>> which should let you break to DDB.
>>
>> Unfortunatly, no luck: can't break into the debugger with the option
>> compiled into the kernel and debug.kdb.stop_cpus_with_nmi=1 set in
>> loader.conf either.
>
> I've never encountered this..is it possible this hardware is faulty?

Quite possible, although we've swapped everything except for the  
CPUs. We've got a replacement mainboard we'll try next.


Stefan

-- 
Stefan Bethke <stb at lassitu.de>   Fon +49 170 346 0140