Re: Strange IDE performance observations

From: Mike Castle (dalgoda@ix.netcom.com)
Date: Sat Mar 18 2000 - 13:14:07 EST


On Sat, Mar 18, 2000 at 12:57:54AM -0800, Andre Hedrick wrote:
> Move to
>
> linux-2.2.14 + pre-patch-2.2.15-15 + ide.2.2.15-15.20000317.patch.bz2

Ok. Tried it. Same results:

/dev/hda:
 Timing buffered disk reads: 64 MB in 8.50 seconds = 7.53 MB/sec

/dev/hdb:
 Timing buffered disk reads: 64 MB in 6.31 seconds = 10.14 MB/sec

/dev/hdc:
 Timing buffered disk reads: 64 MB in 7.39 seconds = 8.66 MB/sec

/dev/hdd:
 Timing buffered disk reads: 64 MB in 9.50 seconds = 6.74 MB/sec

> Are you setting:
>
> CONFIG_BLK_DEV_IDEPCI=y
> CONFIG_BLK_DEV_IDEDMA=y
> CONFIG_IDEDMA_AUTO=y
> CONFIG_BLK_DEV_PIIX=y
> CONFIG_BLK_DEV_PIIX_TUNING=y

Just verified (I'm assuming you did mean CONFIG_PIIX_TUNING since that is
what is everywhere EXCEPT in Configure.help, where it does have
CONFIG_BLK_DEV_PIIX_TUNING).

> If you have not set these options, the kernel can not correctly setup
> transfer rates. If you have done the above, then I need to go to Maxtor.

Well, I do have those set up. However, don't forget, I was getting regular
DMA errors with hdd. Course, now I'm not seeing them with 2.2.15pre15
using my usual stress test of du -s on one of the partitions. Hmmm...

Also, someone mentioned to me offlist that the 430VX is, well, a piece of
shit. If it's just a crappy motherboard, I'm happy with that. Or if it's
maybe just this harddrive acting up, I can live with that, too. I just
wish I knew how to verify that it's either of those two situations as
opposed to something wrong in the kernel. Alas, I don't have a duplicate
system to test on, though.

mrc

-- 
       Mike Castle       Life is like a clock:  You can work constantly
  dalgoda@ix.netcom.com  and be right all the time, or not work at all
www.netcom.com/~dalgoda/ and be right at least twice a day.  -- mrc
    We are all of us living in the shadow of Manhattan.  -- Watchmen

- 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 : Thu Mar 23 2000 - 21:00:24 EST