lkml.org 
[lkml]   [2000]   [Oct]   [2]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: Hard Disk slowdown w/ new mobo?
On Sun, 1 Oct 2000, Alexander Valys wrote:

>Date: Sun, 1 Oct 2000 18:17:30 -0400
>From: Alexander Valys <avalys@optonline.net>
>To: linux-kernel@vger.kernel.org
>Content-Type: text/plain;
> charset="iso-8859-1"
>Subject: Hard Disk slowdown w/ new mobo?
>
>I just purchased a Tyan Tiger 133A motherboard. For those who haven't heard
>of it, it uses the Via Apollo Pro133A chipset, which is ATA66-enabled. I am
>using two maxtor hard drives- hda=20gb hdb=25gb. They are connected with
>ATA66 cable, and are ATA66 capable drives, but they only run at around 2.5
>mb/sec (hdparm -t). Any way to speed these things up? I've heard they
>should operate around 30 mb/sec. I remember seeing a post about this a
>while ago, but I can't seem to find it in the archives. Any help would be
>much appreciated - this upgrade has been hell.

All I can say is _WOW_. This is exactly the same thing that
happened to me just now.. ;o) Well, almost..

I had an ALi Aladdin IV+ motherboard before and was getting
7.6M/s and 9.8M/s out of it with a stock kernel with a few hdparm
tweaks but no DMA.

Both drives are Quantum Fireballs, and SE and an EL.

I just got a DFI K6VB3 +66 motherboard with a VIA MVP3 chipset
which is also ATA66. I was looking forward to UDMA enabling
these drives, but when I booted up I expected at _least_ the
speed that I was getting on the old board. Instead I got a
surprise:

2 root@asdf:/etc/X11# hdparm -t /dev/hd[ab]

/dev/hda:
Timing buffered disk reads: 64 MB in 12.47 seconds = 5.13 MB/sec

/dev/hdb:
Timing buffered disk reads: 64 MB in 13.18 seconds = 4.86 MB/sec


Repeated tests give the same results. ;o(

So my CPU went from being a K6-200 to a K6/2-333 (actually a 350
underclocked to 333 with 66*5 because my memory is 66Mhz). So I
was hoping for a decent speedup. The hard disk slowdown from
9.8M/s to 5.13 sucks the big one pretty badly. It virtually
negates the point of the upgrade. ;o(

Also, I'm using a 2.2.16 kernel with this new motherboard, and
Linux only detects 64M of RAM. I had to force 96M with a lilo
append line. ;o(

So, so far I'm not too impressed with either Linux, or the
motherboard, or both... Also, as if this didn't suck badly
enough already, all the hardware in my machine is exactly the
same as it was, except now X windows will not start. No errors,
it just half-starts, and then fails, with no errors. I
reconfigured it despite the fact it is the same video card,
mouse, everything, and it still fails. ;o(

Are there any known problems with this motherboard/chipset, etc?

Also, this is in my /var/log/messages right at boot time:

Oct 1 23:35:32 asdf kernel: hda: dma_intr: status=0x51 {
DriveReady SeekComplete Error }
Oct 1 23:35:32 asdf kernel: hda: dma_intr: error=0x84 {
DriveStatusError BadCRC }
Oct 1 23:35:32 asdf kernel: hda: dma_intr: status=0x51 {
DriveReady SeekComplete Error }
Oct 1 23:35:32 asdf kernel: hda: dma_intr: error=0x84 {
DriveStatusError BadCRC }
Oct 1 23:35:32 asdf kernel: hda: dma_intr: status=0x51 {
DriveReady SeekComplete Error }
Oct 1 23:35:32 asdf kernel: hda: dma_intr: error=0x84 {
DriveStatusError BadCRC }
Oct 1 23:35:32 asdf kernel: hda: dma_intr: status=0x51 {
DriveReady SeekComplete Error }
Oct 1 23:35:32 asdf kernel: hda: dma_intr: error=0x84 {
DriveStatusError BadCRC }
Oct 1 23:35:32 asdf kernel: hda: DMA disabled
Oct 1 23:35:32 asdf kernel: hdb: DMA disabled
Oct 1 23:35:32 asdf kernel: ide0: reset: success
Oct 1 23:35:32 asdf kernel: CSLIP: code copyright 1989 Regents
of the Universit


I've got these 2 Quantum drives on one cable, and 2 CDROM's on
the other cable. Both cables are exactly 18 inches long
(measured).

Is there a way to get back to at least 9.8M/s with the default
kernel? I mean I was getting that with only:

hdparm -m16c1k1 /dev/hd[ab]

Now with the same setting, I get the crappy 5.x M/s.. ;o(

Any suggestions would be greatly appreciated. I'd like to get
things going with the stock kernel first, and understand the
above errors first before anything else. Should I apply Andre's
IDE patches and try again? I used them before on various other
computers with good success, but not on this machine because it
wasn't properly UDMA compatible before.

Thanks in advance.
TTYL

----------------------------------------------------------------------
Mike A. Harris - Linux advocate - Open source advocate
Computer Consultant - Capslock Consulting
Copyright 2000 all rights reserved
----------------------------------------------------------------------

Be up to date on nerd news and stuff that matters: http://slashdot.org

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

\
 
 \ /
  Last update: 2005-03-22 12:39    [W:0.032 / U:1.344 seconds]
©2003-2018 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site