Re: realtime-preempt patch-2.6.18-rt7 oops

[Daniel Walker <dwalker@mvista.com>]

Try turning this option in Kernel Hacking
"RT Mutex debugging, deadlock detection"

I'm assuming it's off.

On Sun, 2006-11-05 at 18:38 +0000, Rui Nuno Capela wrote:
> Daniel Walker wrote:
> > On Sun, 2006-11-05 at 16:47 +0000, Rui Nuno Capela wrote:
> >> Rui Nuno Capela wrote:
> >>> After a suggestion from Mike Galbraith, I now turned to pure and
> >>> original 2.6.18-rt7 and configured with:
> >>>
> >>> CONFIG_FRAME_POINTER=y
> >>> CONFIG_UNWIND_INFO=y
> >>> CONFIG_STACK_UNWIND=y
> >>>
> >>> Nasty things still do happen, as the following capture can tell as evidence:
> >>>
> >> Some more evidence, just happening:
> >>
> >> ...
> >> Oops: 0002 [#1]
> >>  [<c0106455>] show_trace_log_lvl+0x185/0x1a0
> >>  [<c0106ae2>] show_trace+0x12/0x20
> >>  [<c0106c49>] dump_stack+0x19/0x20
> >>  [<c02f8d2f>] __schedule+0x63f/0xea0
> >>  [<c02f9700>] schedule+0x30/0x100
> >>  [<c02fa5db>] rt_spin_lock_slowlock+0x9b/0x1b0
> >>  [<c02fadb2>] rt_spin_lock+0x22/0x30
> >>  [<c02fd8b1>] kprobe_flush_task+0x11/0x50
> >>  [<c02f91fa>] __schedule+0xb0a/0xea0
> >>  [<c0103b61>] cpu_idle+0xb1/0x120
> >>  [<c011760f>] start_secondary+0x43f/0x500
> > 
> > Is there more above this log? It looks like your cutting off some stuff
> > above this.
> > 
> > Daniel
> 
> Yes there is. However I didn't find it relevant. Here it goes now (full
> captured serial-console log):
> 
> --