panic: Invalid priority <number> on timeshare runq

[Jeff Roberson]

On Mon, 10 Mar 2008, Ed Schouten wrote:

> * David Wolfskill <david at catwhisker.org> wrote:
>> panic: Invalid priority 155 on timeshare runq
>> cpuid = 1
>> KDB: enter: panic
>> [thread pid 718 tid 100068 ]
>> Stopped at      kdb_enter+0x3a: movl    $0,kdb_why
>> db> bt
>> Tracing pid 718 tid 100068 td 0xc58208c0
>> kdb_enter(c0af2a73,c0af2a73,c0af4976,e7a3c938,1,...) at kdb_enter+0x3a
>> panic(c0af4976,9b,c0af449e,183,c0c167c0,...) at panic+0x12c
>> sched_switch(c58208c0,0,6,184,6c4a2d63,...) at sched_switch+0x1d4
>> mi_switch(6,0,c0af483e,b6,0,...) at mi_switch+0x223
>> critical_exit(c4f20b00,0,728,c58208c0,c4f20b00,...) at critical_exit+0x99
>> intr_execute_handlers(c4f08f34,e7a3c9fc,e7a3ca98,c0a46a74,32,...) at intr_execute_handlers+0x14b
>> lapic_handle_intr(32,e7a3c9fc) at lapic_handle_intr+0x3f
>> Xapic_isr1() at Xapic_isr1+0x34
>> --- interrupt, eip = 0xc0a5da09, esp = 0xe7a3ca3c, ebp = 0xe7a3ca98 ---
>
> Same problem here. Good to see I'm not the only one who has this
> problem; I am seeing this panic in my mpsafetty branch (in Perforce) and
> I was starting to suspect my own code... ;-)

http://people.freebsd.org/~jeff/ulerunq.diff

Actually ignore that last diff.  I know it works but this one should be 
more resiliant to further changes in the scheduler.  It tests out fine on 
my systems but I wasn't able to trip the last bug either.  Please let me 
know if this works and I will commit it.

Thanks,
Jeff

>
> -- 
> Ed Schouten <ed at fxq.nl>
> WWW: http://g-rave.nl/
>