On Fri, Aug 11, 2000 at 05:11:59PM +0100, Corin Hartland-Swann wrote:
> > > ==> 2.2.16 <==
> > > Dir Size BlkSz Thr# Read (CPU%) Write (CPU%) Seeks (CPU%)
> > > ----- ------ ------- ---- ------------- -------------- --------------
> > > /mnt/ 256 4096 1 27.0948 10.5% 25.9272 22.5% 147.264 0.83%
> > > /mnt/ 256 4096 2 20.5753 8.42% 25.9316 22.7% 143.113 0.56%
> > > /mnt/ 256 4096 4 13.0397 6.09% 25.7723 22.7% 143.619 0.51%
> > > /mnt/ 256 4096 16 9.38573 5.40% 25.6285 23.1% 147.815 0.51%
> > > /mnt/ 256 4096 32 6.96358 5.90% 25.2889 22.9% 151.182 0.57%
I�ve seen the same behaviour testing SCSI disks with 2.2.16 as well.
2.4.0-t3 scaled dropped from 20 to 12 for two threads and stayed stable
there, while 2.2.16(-SuSE) went from 17 down to 6 (for 8 threads).
I'm testing 2.2.14 (with an IDE disk -d1 IBM DJNA) currently. Wait. . . . . .
It shows the same behaviour as 2.2.16:
File Block Num Seq Read Rand Read Seq Write Rand Write
Dir Size Size Thr Rate (CPU%) Rate (CPU%) Rate (CPU%) Rate (CPU%)
------- ------ ------- --- ----------- ----------- ----------- -----------
. 512 4096 1 15.77 11.8% 0.558 0.89% 11.47 10.4% 0.759 2.33%
. 512 4096 2 9.186 7.12% 0.550 0.66% 12.00 10.4% 0.770 2.07%
. 512 4096 4 7.069 6.09% 0.568 0.69% 12.04 10.5% 0.768 2.03%
. 512 4096 8 2.578 2.55% 0.577 0.64% 12.01 10.5% 0.774 2.10%
I have to say that I did do some work in the background in the end, so the
8 thread numbers should be slightly better in reality.
I wonder, why you had such a flat profile for 2.2.15.
Does anybody have an explanation for this?
> I think I've located the problem... (with kernel 2.4, anyway) - it is
> refusing to use DMA.
>
> When I try hdparm -m -c -d1 -a, I get the following output:
Oops. PIIX4 should be supported by 2.4 kernels.
> hdc: [PTBL] [4982/255/63] hdc1
>
> Can anyone explain to me what the [PTBL] bit means? I've been wondering
> this for about 4 years now, and still don't know :)
Partition table. The mapping is reported by the BIOS, but also stored in the
partition table. The kernel uses the latter.
> Anybody have any suggestions about how to get DMA working? Is it a problem
> with the IDE controller?
I don't think so.
Did you enable DMA in the BIOS?
Anyway, using Andre Hedrick�s patches for IDE is a good idea in general.
They are on the kernel mirrors.
Regards,
-- Kurt Garloff <garloff@suse.de> Eindhoven, NL GPG key: See mail header, key servers Linux kernel development SuSE GmbH, Nuernberg, FRG SCSI, Security
- 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.tux.org/lkml/
This archive was generated by hypermail 2b29 : Tue Aug 15 2000 - 21:00:24 EST