[9251] in linux-scsi channel archive
Re: concurrent commands to same LUN ?
daemon@ATHENA.MIT.EDU (Kurt Garloff)
Wed Jul 26 13:58:38 2000
Date: Wed, 26 Jul 2000 19:50:56 +0200
From: Kurt Garloff <garloff@suse.de>
To: Martin Peschke <peschke@fh-brandenburg.de>
Cc: Linux SCSI list <linux-scsi@vger.rutgers.edu>
Message-ID: <20000726195056.S32091@garloff.etpnet.phys.tue.nl>
Mail-Followup-To: Kurt Garloff <garloff@suse.de>,
Martin Peschke <peschke@fh-brandenburg.de>,
Linux SCSI list <linux-scsi@vger.rutgers.edu>
Mime-Version: 1.0
Content-Type: multipart/signed; micalg=pgp-md5;
protocol="application/pgp-signature"; boundary="ScUgq5oMe+fJq4F1"
Content-Disposition: inline
In-Reply-To: <Pine.GSO.4.10.10007241753330.5872-100000@zeus.fh-brandenburg.de>; from peschke@fh-brandenburg.de on Mon, Jul 24, 2000 at 06:07:39PM +0200
--ScUgq5oMe+fJq4F1
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable
On Mon, Jul 24, 2000 at 06:07:39PM +0200, Martin Peschke wrote:
> I am just wondering about select_queue_depths() and cmd_per_lun.
> Is a HBA driver allowed to have more than one outstanding command at the
> same LUN?
Tagged Command Queueing.
> Or is it responsible for command serialization?
If you don't advertise more the 1 in the cmd_per_lun field in your
host_template, the mid-level should not send another command, before the
first one completed.
> I am just thinking about multiple paths in Fibre Channel environments
> because of several FC adapters within a single host connected to the same
> switched fabric.
Be careful. As Linux does not know that the devices it sees multiple times
(on each host adapter once) are actually the same, it will not ense data
integrity. If you change the contents of a device (filesystem) on one
adapter and read back from the other one, your buffer cache and your FS will
look inconsistent.
> Are there cases where I can spread several commands to
> the same LUN over multiple ports at the same time to (maybe) increase
> performance?
Just do sequential writes ...
But better use TCQ instead of spreading commands to different hosts.
Without TCQ, performance will be bad (unless you use the write cache).
Regards,
--=20
Kurt Garloff <garloff@suse.de> Eindhoven, NL
GPG key: See mail header, key servers Linux kernel development
SuSE GmbH, Nuernberg, FRG SCSI, Security
--ScUgq5oMe+fJq4F1
Content-Type: application/pgp-signature
Content-Disposition: inline
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.1e-SuSE (GNU/Linux)
Comment: Weitere Infos: siehe http://www.gnupg.org
iD8DBQE5fyUAxmLh6hyYd04RAoGPAKCZurrLQDCxMzecyeT1Ojm+EY+bRACfXq+a
4RsOREfNNl48GMuDKVsno1g=
=gW+K
-----END PGP SIGNATURE-----
--ScUgq5oMe+fJq4F1--
-
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@vger.rutgers.edu
