[lkml]   [1999]   [Oct]   [19]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: ISSUE: (ext2fs?) disk I/O and responsivness
On Tue, 19 Oct 1999, David Ford wrote:

> Date: Tue, 19 Oct 1999 00:00:43 -0700
> From: David Ford <>
> Reply-To:
> To: Mattias Sandgren <>
> Cc:, Ronald Wahl <>
> Subject: Re: ISSUE: (ext2fs?) disk I/O and responsivness
> Let me chime in and say "me too". I have a system that responds precisely like
> this. Actually, it's not "a large file", it is "a large amount of writes".
> You can see this by untarring linux kernel source with -v. it _creeps_ by. If
> done immediately after boot, it's quite fast, then it slows to an intolerable crawl
> and there is -no- indication as to _why_ it's slow. At first I thought it might be
> because I compiled it SMP, nope, that isn't a fix. Next I swapped from adaptec
> 2940 to a symbios card. Nope, that didn't have any affect either.
> Here's a kicker. IDE is still fast. I can zip through an IDE disk like no
> tomorrow. Even when the scsi drive is slogging, the IDE drive is a wee bit faster
> (but terribly slow due to the entire system being killed). Even gpm movements are
> jerky and that's scary. X...forget about it. Takes 90 seconds to move 1/2 across
> the screen, it's easier to close your eyes and calculate the correct mouse motion
> to get where you want to go.
> And no. Absolutely no load shows on the machine, no kernel messages, no NOTHING
> which gives the slightest clue what's wrong.
> -d
Well, I have recently changed machine at work from Compaq DeskPro 5100 to
Compaq DeskPro EN 6450 and OS from RedHat Linux 4.2 to RedHat Linux 6.0;
Both systems are IDE, but with PD-CD connected to BusLogic SCSI;
While my previous system behaved normal when writing to PD, the new one
was completely unusable during write (I was not even able to switch
between screens on X);
Then I went to BIOS settings and changed PCI Busmastering enable
option from default NO to YES (and also disabled serial 2, so I had a
separate IRQ for SCSI card); After this change the system work quite well
during transfers to PD.

I am not sure how much relevant it is to your problems, but as symptoms
were similar, I thought I would post this.

Best regards,


To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to
Please read the FAQ at

 \ /
  Last update: 2005-03-22 13:54    [W:0.037 / U:1.628 seconds]
©2003-2018 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site