Re: 2.1.108 stress test results

Mike Jagdis (mike@roan.co.uk)
Thu, 16 Jul 1998 09:16:16 +0100 (GMT/BST)


On Wed, 15 Jul 1998, Alan Cox wrote:

> My first suspicion is the ioctls arent all locked right - since gtcd
> was watching a music cd at the same time it was probably doing stuff

Ah... interesting... I had a funny that appeared to be a corruption
in the buffer head lists (or might not have been). This is a system
with 7 SCSI drives, a couple of tape drives, 2 DPT controllers and
a future domain, 3 RAID groups of two drives. It appears solid
even under the most extreme load. When the error occurred I had
just been playing with mt, doing multi-file backups to an Exabyte
(which is also solid normally). mt does ioctls. That's all it does :-).

N.B. I'm using the eata_dma driver for the DPTs. Previously I
tried using the other eata driver but that seemed to trash itself
under load (both RAID1'd and normal partitions) - *very* heavy
load that is. I had tagged queuing etc. turned on. I didn't try
turning it off. Not much chance of trying it at the moment :-).
I think this was a driver problem where as the above is likely
higher level. Maybe.

Mike

-- 
.----------------------------------------------------------------------.
|  Mike Jagdis                  |  Internet:  mailto:mike@roan.co.uk   |
|  Roan Technology Ltd.         |                                      |
|  54A Peach Street, Wokingham  |  Telephone:  +44 118 989 0403        |
|  RG40 1XG, ENGLAND            |  Fax:        +44 118 989 1195        |
`----------------------------------------------------------------------'

- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html