Subject: Re: bufcache cancer in -current?
To: Frank van der Linden <frank@wins.uva.nl>
From: Patrick Welche <prlw1@newn.cam.ac.uk>
List: current-users
Date: 05/25/2000 14:57:21
On Thu, May 25, 2000 at 10:04:25AM +0200, Frank van der Linden wrote:
> On Wed, May 24, 2000 at 08:15:27PM +0200, Sean Doran wrote:
> > sd1(ahc2:6:0): queue full
> > smd: sc_link->active = 4; setting sc_link->openings to that - 1
> > sd1(ahc2:6:0): queue full
> > smd: sc_link->active = 3; setting sc_link->openings to that - 1
> 
> Yes, I've thought about adding such a piece of code myself.
> 
> The other problem that was seen sometimes was that some things were
> sent with tags that shouldn't have. I'll fix that today.
> 
> Now we need to find out what's going wrong in the QUEUE FULL case.
> 
> Thanks for testing so far, even though it meant potentially
> thrashing some data :-)

As this all seems related, I rebuilt a kernel without NOTAG, but with
AHC_DEBUG, with code from 11:04GMT today, did a build and found

sd1(ahc0:2:0): queue full

and later the evidence:
/usr/bin/nm: itime.o: File truncated
etc.

ahc0: target 0 synchronous at 20.0MHz, offset = 0xf
ahc0: target 0 using tagged queuing
sd0 at scsibus0 target 0 lun 0: <QUANTUM, XP34550S, LYK8> SCSI2 0/direct fixed
sd0: 4341 MB, 5899 cyl, 10 head, 150 sec, 512 bytes/sect x 8890760 sectors
ahc0: target 2 synchronous at 20.0MHz, offset = 0xf
ahc0: target 2 using tagged queuing
sd1 at scsibus0 target 2 lun 0: <QUANTUM, XP34550S, LYK8> SCSI2 0/direct fixed
sd1: 4341 MB, 5899 cyl, 10 head, 150 sec, 512 bytes/sect x 8890760 sectors

Cheers,

Patrick