Re: ULE/SCHED_SMP diff for 7.0

On Thu, 19 Jul 2007, Jiawei Ye wrote:

> After running the new ULE 3.0 for 2 days, I got this panic when try to
> install diablo-jdk

Hi,

Can you try the file at http://people.freebsd.org/~jeff/sched_ule.c

I fixed a couple of bugs that might impact you.

Thanks,
Jeff

>
> spin lock 0xc0baa380 (sched lock 1) held by 0xc4b1aaa0 (tid 100189) too long
> panic: spin lock held too long
> cpuid = 0
> Uptime: 7h52m15s
> Physical memory: 1003 MB
> Dumping 213 MB: 198 182 166 150 134 118 102 86 70 54 38 22 6
>
>
> (kgdb) bt
> #0  doadump () at pcpu.h:195
> #1  0xc074ffe7 in boot (howto=260) at /usr/src/sys/kern/kern_shutdown.c:409
> #2  0xc07502db in panic (fmt=Variable "fmt" is not available.
> ) at /usr/src/sys/kern/kern_shutdown.c:563
> #3  0xc0743c1f in _mtx_lock_spin_failed (m=0x0)
>   at /usr/src/sys/kern/kern_mutex.c:424
> #4  0xc0743ca5 in _mtx_lock_spin (m=0xc0baa380, tid=3366110272, opts=0,
>   file=0x0, line=0) at /usr/src/sys/kern/kern_mutex.c:457
> #5  0xc076bc84 in sched_add (td=0xc4b1aaa0, flags=0)
>   at /usr/src/sys/kern/sched_ule.c:943
> #6  0xc076c046 in sched_wakeup (td=0xc4b1aaa0)
>   at /usr/src/sys/kern/sched_ule.c:1858
> #7  0xc0758158 in setrunnable (td=0xc4b1aaa0)
>   at /usr/src/sys/kern/kern_synch.c:505
> #8  0xc077f0a9 in sleepq_resume_thread (sq=0xc3ee05e0, td=0xc4b1aaa0,
> pri=Variable "pri" is not available.
> )
>   at /usr/src/sys/kern/subr_sleepqueue.c:658
> #9  0xc077fbe3 in sleepq_broadcast (wchan=0xc8b07784, flags=3, pri=-1, 
> queue=0)
>   at /usr/src/sys/kern/subr_sleepqueue.c:753
> #10 0xc07576e8 in _sx_xunlock_hard (sx=0xc8b07784, tid=3366110272, file=0x0,
>   line=0) at /usr/src/sys/kern/kern_sx.c:623
> #11 0xc0757edf in _sx_xunlock (sx=0xc8b07784, file=0x0, line=0) at sx.h:168
> #12 0xc096ec58 in _vm_map_unlock_read (map=0xc8b07740, file=0x0, line=0)
>   at /usr/src/sys/vm/vm_map.c:457
> #13 0xc096ec81 in vm_map_lookup_done (map=0xc8b07740, entry=0xc8a20bb0)
>   at /usr/src/sys/vm/vm_map.c:3317
> #14 0xc096805a in unlock_and_deallocate (fs=0xe6a97ac4)
>   at /usr/src/sys/vm/vm_fault.c:147
> #15 0xc096a234 in vm_fault (map=0xc8b07740, vaddr=135495680,
>   fault_type=2 '\002', fault_flags=8) at /usr/src/sys/vm/vm_fault.c:922
> #16 0xc0a1295b in trap_pfault (frame=0xe6a97c24, usermode=0, eva=135499760)
>   at /usr/src/sys/i386/i386/trap.c:761
> #17 0xc0a133f9 in trap (frame=0xe6a97c24) at 
> /usr/src/sys/i386/i386/trap.c:462
> #18 0xc09f944b in calltrap () at /usr/src/sys/i386/i386/exception.s:139
> #19 0xc0a10a51 in suword32 () at /usr/src/sys/i386/i386/support.s:1249
> Previous frame inner to this frame (corrupt stack?)
>
> Jiawei
> -- 
> "If it looks like a duck, walks like a duck, and quacks like a duck,
> then to the end user it's a duck, and end users have made it pretty
> clear they want a duck; whether the duck drinks hot chocolate or
> coffee is irrelevant."
>