Re: Crash in today's -CURRENT

On Sun, 2004-12-12 at 23:29 -0500, Joe Marcus Clarke wrote:
> I have a Dell Precision workstation 340 that has been running -CURRENT
> happily now for a few months.  I upgraded to today's -CURRENT, and it
> will no longer boot (even in SAFE mode).  The crash is similar to the
> one reported by Alexandre Kovalenko (i.e. in the CAM code), but not the
> same:
> 
> acd0: DVDROM <LITEON DVD-ROM LTD163/GDMF> at ata1-master PIO4
> ata3-master: FAILURE - ATA_IDENTIFY timed out
> ata3-master: FAILURE - ATA_IDENTIFY timed out
> ata3-master: FAILURE - ATA_IDENTIFY timed out
> Waiting 8 seconds for SCSI devices to settle
> ata3-master: FAILURE - ATA_IDENTIFY timed out
> ata3-master: FAILURE - ATA_IDENTIFY timed out
> 
> Fatal trap 12: page fault while in kernel mode
> fault virtual address	= 0x4
> fault code		= supervisor read, page not present
> instruction pointer	= 0x8:0xc0431dd2
> stack pointer		= 0x10:0xe9ce6ca8
> frame pointer		= 0x10:0xe9ce6ca8
> code segment		= base 0x0, limit 0xfffff, type = 0x1b
> 			= DPL 0, pres 1, def32 1, gran 1
> processor eflags	= interrupt enabled, resume, IOPL = 0
> current process		= 20 (irq9:mpt0)
> Stopped at      xpt_freeze_devq+0x1e:   movl    0x4(%eax),%eax
> 
> xpt_freeze_devq(c3064580) at xpt_freeze_devq+0x1e
> mpt_done(c315a800,bebec5c0,c318f380,4,c3076880) at mpt_done+0x691
> mpt_intr(c315a800,c307da80,0,0,c30c1000) at mpt_intr+0x69
> ithread_loop(c3076880,e9ce7d48,c04d9530,0) at ithread_loop+0xad
> fork_exit(c04d9530,c3076880,e9ce7d48) at fork_exit+0x54
> fork_trampoline() at fork_trampoline+0x8
> --- trap 0x1, eip = 0, esp = 0xe9ce7d7c, ebp = 0 ---
> 
> (gdb) l *xpt_freeze_devq+0x1e
> 0xc0431dd2 is in xpt_freeze_devq (/usr/src/sys/cam/cam_xpt.c:4671).
> 4666             * if a ccb's status is still CAM_REQ_INPROG
> 4667             * under spl protection just before they queue
> 4668             * the CCB.  See ahc_action/ahc_freeze_devq for
> 4669             * an example.
> 4670             */
> 4671            ccbh = TAILQ_LAST(&path->device->ccbq.active_ccbs,
> ccb_hdr_tailq);
> 4672            if (ccbh && ccbh->status == CAM_REQ_INPROG)
> 4673                    ccbh->status = CAM_REQUEUE_REQ;
> 4674            splx(s);
> 4675            return (path->device->qfrozen_cnt);
> 
> 
> The only new output (besides the panic) is the ATA_IDENTIFY errors.
> This device corresponds to my Maxtor SATA drive on my Promise SATA
> controller (see http://www.marcuscom.com/downloads/FUGU.dmesg for a
> sample dmesg).  The kernel that fails is:
> 
> FreeBSD 6.0-CURRENT #17: Sun Dec 12 19:08:58 EST 2004
> 
> The kernel that last worked was:
> 
> FreeBSD 6.0-CURRENT #16: Fri Dec  3 03:03:39 EST 2004
> 
> Besides fixing the crash, I'd also like to get my SATA drive back.
> Thanks.
> 
> Joe
> 
Removing everything remotely SCSI from my kernel config got me over this
hump. I don't know if it is an option for you...

-- 
Alexandre "Sunny" Kovalenko (Олександр Коваленко)