[issue2092] Panic: Bad link elm 0x... next->prev != elm

["Magliano Andrea \(via DragonFly issue tracker\)" <sinknull@xxxxxxxxxxxxxxxxxxxxx>]

Magliano Andrea <masterblaster@tiscali.it> added the comment:

After a stress2 disk test, got panic below (of course no kernel dump).
Timeout msg look like a well known problem (logged either by FreeBSD and Linux
kernel msg) of WD Raptor hard disk series (in my case a WD3000HLFS).

I'm migrating right now to an SSD installation, i'm looking forward to see if
it's HD fault...

Anyway it would be nice to force a kernel dump (as a list elem contention
problem is still present); but here the point of failure is the very piece of
hardware where the dump should be written... Should i dump, say, to an usb
stick? Any ideas?




ahci0.1: CMD TIMEOUT state=5 slot=20
        cmd-reg 0xd417<CR,FR,FRE,POD,SUD,ST>
        sactive=00100000 active=00000000 expired=00000000
           sact=00000000     ci=00000000
            STS=40
ahci0.1: disk_rw: timeout
ahci0.1: CMD TIMEOUT state=5 slot=9
        cmd-reg 0xc917<CR,FR,FRE,POD,SUD,ST>
        sactive=00000200 active=00000000 expired=00000000
           sact=00000000     ci=00000000
            STS=40
ahci0.1: disk_rw: timeout
[diagnostic] cache_lock: blocked on 0xffffffe04d25bca8 "stressX"
[diagnostic] cache_lock: unblocked stressX after 1 secs
[diagnostic] cache_lock: blocked on 0xffffffe04d25bca8[diagnostic] cache_lock:
blocked on 0xffffffe04d25bca8 "stressX"
 "stressX"
[diagnostic] cache_lock: blocked on 0xffffffe04d25bca8 "stressX"
[diagnostic] cache_lock: unblocked stressX after 0 secs
[diagnostic] cache_lock: unblocked stressX after 0 secs
[diagnostic] cache_lock: unblocked stressX after 0 secs
ahci0.1: CMD TIMEOUT state=5 slot=28
        cmd-reg 0xdc17<CR,FR,FRE,POD,SUD,ST>
        sactive=10000000 active=00000000 expired=00000000
           sact=00000000     ci=00000000
            STS=40
ahci0.1: disk_rw: timeout
ahci0.1: CMD TIMEOUT state=5 slot=28
        cmd-reg 0xdc17<CR,FR,FRE,POD,SUD,ST>
        sactive=10000000 active=00000000 expired=00000000
           sact=00000000     ci=00000000
            STS=40
ahci0.1: disk_rw: timeout
[diagnostic] cache_lock: blocked on 0xffffffe04d25bca8[diagnostic] cache_lock:
blocked on 0xffffffe04d25bca8 "stressX"
 "stressX"
[diagnostic] cache_lock: unblocked stressX after 0 secs
[diagnostic] cache_lock: unblocked stressX after 0 secs
ahci0.1: CMD TIMEOUT state=5 slot=6
        cmd-reg 0xc617<CR,FR,FRE,POD,SUD,ST>
        sactive=00000040 active=00000000 expired=00000000
           sact=00000000     ci=00000000
            STS=40
ahci0.1: disk_rw: timeout
ahci0.1: CMD TIMEOUT state=5 slot=0
        cmd-reg 0xc017<CR,FR,FRE,POD,SUD,ST>
        sactive=00000001 active=00000000 expired=00000000
           sact=00000000     ci=00000000
            STS=40
ahci0.1: disk_rw: timeout
ahci0.1: disk_rw: unknown state 6
ahci0.1: disk_rw: unknown state 6
panic: Bad link elm 0xffffffe01e2faa78 next->prev != elm
cpuid = 0
Trace beginning at frame 0xffffffe01e10fa00
panic() at panic+0x1ed
panic() at panic+0x1ed
camisr_runqueue() at camisr_runqueue+0x87
swi_cambio() at swi_cambio+0x167
ithread_handler() at ithread_handler+0x1d4
boot() called on cpu#0
Uptime: 8h50m24s
Physical memory: 2019 MB
Dumping 1064 MB:spin_lock: 0xffffffe005e0b5a0, indefinite wait!
panic: Bad link elm 0xffffffe01e2faa78 next->prev != elm
cpuid = 0
boot() called on cpu#0

_____________________________________________________
DragonFly issue tracker <bugs@lists.dragonflybsd.org>
<http://bugs.dragonflybsd.org/issue2092>
_____________________________________________________