Re: New IDE drive dramatically slow system?

From: Alvin Oga (aoga_at_ns.Linux-Consulting.com)
Date: 07/23/03

  • Next message: Micha Feigin: "changing default runlevel dynamically"
    Date: Tue, 22 Jul 2003 23:07:08 -0700 (PDT)
    To: "Daniel B." <dsb@smart.net>
    
    

    hi ya daniel

    On Wed, 23 Jul 2003, Daniel B. wrote:

    > Alvin Oga wrote:
    > >
    > > On Mon, 14 Jul 2003, Joyce, Matthew wrote:
    > >
    > > ...
    > > > > -- dont mix drives unless they are both ata-100
    > > > > and both have 2MB disk buffers .. etc.etc..
    > > > >
    > > > > c ya
    > > > > alvin
    > > > >
    > > >
    > > > I've never heard of matching buffers before.
    > >
    > > mixing w/ 8MB buffer disk cache is a recent problem
    > > since its been around a year or two
    > >
    > > and you do have to keep those disk buffers the same on the
    > > same ide cable along w/ the same ata-66 vs ata-100 vs ata-133
    >
    > That can't be right.

    you can "assume" it can't be right or wrong all you like...

    - the disk tests is left to the user to figure out if you
      do in fact have 2MB buffer disks ofr 8MB buffer disks
      on the same ide cable... and if so, and you do have slow
      ide responses ... duh.....
            - it's a known problem ...
            ( one can always go fix the ide driver if needed
            ( or not mix the disks

    - how you do the disk tests can affect your results too

    - i dont mix ata speeds, let alone disk buffers...
            ( so i dont see any ide problems

    c ya
    alvin

    > How can the total cache size affect anything in the transfer
    > between the IDE controller and the drives?
    >
    > Are you confusing total cache size with sector size or some
    > transfer size?
    >

    -- 
    To UNSUBSCRIBE, email to debian-user-request@lists.debian.org 
    with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
    

  • Next message: Micha Feigin: "changing default runlevel dynamically"

    Relevant Pages

    • Re: Figured it out: IDE controller had reverted to PIO mode
      ... HP put a 40wire ide cable in mine instead of an 80wire one going to my DVDRW ... IDE (I have a USB enclosure for the disk). ... reader driver to handle 2GB cards. ... I ran the disk diagnotics from Windows, ...
      (comp.sys.hp.hardware)
    • Re: Reihenfolge der Festplattenerkennung
      ... IDE/SCSI würde die erste IDE Platte mit 0x80, ... Erst dann kämen die SCSI mit z.B. 0x83. ... Defines non-standard parameters for the specified disk. ... The BIOS numbers your disks 0x80, 0x81, etc. and it is impossible to decide which Linux disk corresponds to which BIOS disk, so if you have an unusual setup you need to state the correspondence between Linux disks and BIOS disks. ...
      (de.comp.os.unix.linux.hardware)
    • Re: Difference between IDE and SCSI ??
      ... IDE stands for Intelligent Drive ... This was the first attempt to standardize the interface between mass ... owned a Northstar single density hard-sectored 5.25" disk system for my ... where the drives plug directly into the frame. ...
      (Fedora)
    • Re: 2.6.8 stock kernel bug
      ... Among the kernel modules in kernel 2.6.8, ... > deciding whether they are required for mounting an IDE CDROM device. ... >> we're talking about the ide driver that controls the disk where the ... echo Mounted /proc filesystem ...
      (comp.os.linux.hardware)
    • Re: Converting IDE RAID 1 to SATA RAID 1
      ... If it is connected to the IDE controller no other hardware has ... > boot from) the disk created by the IDE controller once it was removed from ... I replace the drives each year. ...
      (comp.sys.ibm.pc.hardware.storage)