Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 21 Sep 1998 14:38:12 -0700 (PDT)
From:      Julian Elischer <julian@whistle.com>
To:        Ben Smithurst <ben@scientia.demon.co.uk>
Cc:        Mike Smith <mike@smith.net.au>, FreeBSD Current <freebsd-current@FreeBSD.ORG>
Subject:   Re: system hangs with soft updates
Message-ID:  <Pine.BSF.3.95.980921143212.11176C-100000@current1.whistle.com>
In-Reply-To: <19980917173651.A567@scientia.demon.co.uk>

next in thread | previous in thread | raw e-mail | index | archive | help


On Thu, 17 Sep 1998, Ben Smithurst wrote:

> Mike Smith wrote:
> 
> > > _wdstart(ctrlr=1) at _wdstart+0x4af [../../i386/isa/wd.c:332]
> > 
> > This is where you were when the interrupt occurred.  Unfortunately the 
> > line number above is wrong; wdstart() begins around line 900.
> 
> I've repeated it, and it definitely says 332, so don't blame me for that
> one <g>
> 
> > Did you see any "dummy wdunwedge" messages before you dropped into DDB, 
> > or any other messages at all?  
> 
> No, I'm pretty sure I didn't see anything like that.
> 
> > If you can reproduce the situation again, it would be useful if you 
> > could drop into DDB, copy the trace output, then use 'c' to continue 
> > and repeat the process a couple of times.  This would clarify whether 
> > you're stuck in the IDE driver, or whether you just happened to catch
> > it there on a chance.
> 
> When I press c, nothing happens. I have to ctrl-alt-esc to get back into
> the debugger.

if you re-enter the debugger many times doe sthe stack trace ever change
at all?
<CTL><ALT><DEL>
[get stack]
Cont
<CTL><ALT><DEL>
[get stack]
Cont
<CTL><ALT><DEL>
[get stack]
Cont
<CTL><ALT><DEL>
[get stack]
Cont

> 
> A few other lines which may be relevant, everything else was pretty much
> the same as last time:
> 
> _worklist_insert(head=f1a049a0,item=f0fa0280) at _worklist_insert+0x10 [../../ufs/ffs/ffs_softdep.c:402]
> _softdep_disk_write_complete(bp=f1a049a0) at _softdep_disk_write_complete+0x29d [../../ufs/ffs/ffs_softdep.c:3012]
> _biodone(bp=f1a049a0) at _biodone+0xab [../../kern/vfs_bio.c:1915]

Very interesting...
I have to upgrade my test systems as they've been 
'suspended' during all the ELF/CAM transition.

> 
> The disk light did actually go off after a while, but it came on again
> as soon as I typed (c)ontinue, so it's stuck somewhere.
Well you can single-step..
try 's' instead of 'c'

julian

> 
> -- 
> Ben Smithurst : ben@scientia.demon.co.uk : http://www.scientia.demon.co.uk/
> 
> PGP: 0x99392F7D - 3D 89 87 42 CE CA 93 4C  68 32 0E D5 36 05 3D 16
>      http://www.scientia.demon.co.uk/ben/pgp-key.html (or use keyservers)
> 
> To Unsubscribe: send mail to majordomo@FreeBSD.org
> with "unsubscribe freebsd-current" in the body of the message
> 


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-current" in the body of the message



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.3.95.980921143212.11176C-100000>