The crash I seen without this patch was always something like this: Trace: [<fffffc000031a164>] handle_IRQ_event+0x74/0xf0 [<fffffc000031ab50>] handle_irq+0xe0/0x1c0 [<fffffc0000329b04>] srm_device_interrupt+0x24/0x40 [<fffffc000031b1f4>] do_entInt+0xf4/0x140 [<fffffc0000315260>] ret_from_sys_call+0x0/0x10 [<fffffc0000316e30>] default_idle+0x0/0x10 [<fffffc0000316e98>] cpu_idle+0x58/0x80 [<fffffc0000316e30>] default_idle+0x0/0x10 [<fffffc0000316e30>] default_idle+0x0/0x10 [<fffffc0000310234>] rest_init+0x34/0x50 [<fffffc000031001c>] __start+0x1c/0x20Code: 243f0010 245f0020 21c10100 21a20200 a4490008 a4290000 <b4410008> b4220000
Kernel panic: Aiee, killing interrupt handler! In interrupt handler - not syncingOnly the trace was sometiomes a little bit different, but handle_IRQ_event was alwas the failing function.
At 15:42 02.04.2005, Steve Langasek wrote:
> The only problem with 2.6.8 and 2.6.10 here was a problem with irq > handling. In idle mode the kernel crashes sometimes in function > handle_IRQ_event (not in a driver!). Looking into the code I find out that > alpha was the only platform where this function diffs from all other > architectures (relict from 2.4.x?). After replacing the C code in this > function with the x86 code the kernel runs fine (file: > arch/alpha/kernel/irq.c). Ah, that sounds like the crash I was seeing, though it triggered only when running hotplug. Do you have a patch for this that could be sent to the Debian kernel maintainers? Cheers, -- Steve Langasek postmodern programmer
----- Uwe Schindler H.-H.-Meier-Allee 63, D-28213 Bremen http://www.thetaphi.de - http://www.schindlers-software.de eMails: uwe@thetaphi.de (private); info@schindlers-software.de (company) Tel./Fax: +49 700 PCLATEIN (+49 700 72528346) Schindlers Software - Home of Schindlers PC-LATEIN 3.10 DIE Software zum Lateinlernen!
Attachment:
kernel-irq.patch
Description: Binary data