mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Andries.Brouwer@cwi.nl
To: Andries.Brouwer@cwi.nl, john@mwk.co.nz,
	linux-kernel@vger.kernel.org, martin.bene@icomedias.com
Cc: andre@linux-ide.org, hugo@firstlinux.net
Subject: Re: AW: Need to force IDE geometry
Date: Sat, 16 Feb 2002 11:50:15 GMT	[thread overview]
Message-ID: <UTC200202161150.LAA30214.aeb@cwi.nl> (raw)

    From martin.bene@icomedias.com Sat Feb 16 10:13:30 2002

    Hi Andries,

    For some reasons the c/h/s settings reported for LBA disks depend on ide =
    device number: /hda uses 255 heads, 63 sectors while /dev/hdc and above =
    use 16 head, 63 sectors.

    hda: 150136560 sectors (76870 MB) w/1916KiB Cache, CHS=3D9345/255/63, =
    UDMA(33)
    hdb: 150136560 sectors (76870 MB) w/1916KiB Cache, CHS=3D148945/16/63, =
    UDMA(33)
    hdc: 150136560 sectors (76870 MB) w/1916KiB Cache, CHS=3D148945/16/63, =
    UDMA(33)

    As you can see, this means you end up with different reported drive =
    geometries for identical disks. Esp. if you want to use software raid =
    this is a major nuisance. The usual workaround is to change =
    head/cylinder settings when first partitioning the drive and let linux =
    change the geometry during partition table check.

    Partition check:
     hda: hda1 hda2 < hda5 hda6 > hda3
     hdb: [PTBL] [9345/255/63] hdb1 hdb2 < hdb5 hdb6 > hdb3
     hdc: [PTBL] [9345/255/63] hdc1 hdc2 < hdc5 hdc6 > hdc3

    While this works, it's quite unintuitive and confusing; correct =
    behaviour would be to treat all disks identicaly regardless of device =
    number.

Yes, this is a FAQ. See
	http://www.win.tue.nl/~aeb/linux/Large-Disk-14.html#ss14.2
"Identical disks have different geometry?".

You give as workaround bootparameters to Linux. My solution would
probably be to set the disks to "Normal" in the BIOS.
There is no need to tell the BIOS to do stupid tricks in order
to avoid DOS problems since we are not running DOS.
Moreover, with "Normal" the disk is slightly larger - for me the
difference is 7 MB.

(There is an unfortunate confusion here:
LBA is used in two very different meanings:
1) LBA "linear block addressing" is an access mode of disks.
Every disk is accessed this way by Linux, unless it is really old.
2) LBA "LBA assist" is a translation of disk geometry by the BIOS
in order to bypass DOS deficiencies.
Everyone wants the first, and gets it automatically.
Nobody running Linux wants the second, but people choosing LBA
in the BIOS setup usually think they are choosing the first.)

Andries

             reply	other threads:[~2002-02-16 11:50 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2002-02-16 11:50 Andries.Brouwer [this message]
  -- strict thread matches above, loose matches on Subject: below --
2002-02-16  9:13 Martin Bene

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=UTC200202161150.LAA30214.aeb@cwi.nl \
    --to=andries.brouwer@cwi.nl \
    --cc=andre@linux-ide.org \
    --cc=hugo@firstlinux.net \
    --cc=john@mwk.co.nz \
    --cc=linux-kernel@vger.kernel.org \
    --cc=martin.bene@icomedias.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®