From mboxrd@z Thu Jan 1 00:00:00 1970 From: Robin Hill Subject: Re: Spurious HD convictions Date: Wed, 16 Dec 2009 09:13:31 +0000 Message-ID: <20091216091331.GA15689@cthulhu.home.robinhill.me.uk> References: <70ed7c3e0912141206p65005aeegf663183c946dec7b@mail.gmail.com> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="xHFwDpU9dbj6ez1V" Return-path: Content-Disposition: inline In-Reply-To: Sender: linux-raid-owner@vger.kernel.org To: linux-raid@vger.kernel.org List-Id: linux-raid.ids --xHFwDpU9dbj6ez1V Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Tue Dec 15, 2009 at 11:41:03PM -0600, Leslie Rhorer wrote: > It seems to have done so, yes. I've looked around the web trying to > find some additional info, but I've come up empty handed. Perhaps someone > here can answer at least one of my questions? I know that putting a value > of 32 into /sys/block//device/queue_depth fully enables NCQ, and > putting a value of 1 there disables NCQ. What do all the numbers in betw= een > do? >=20 I believe the value sets the allowed queue length. A value of 1 thus effectively disables queueing. Most ATA drives have a maximum queue depth of 32 so this is usually set for fully enabling it (though I believe it's recommended only to use 31 - this seems to be the default). If you set it to 32 and the drive has a shorter maximum queue, the kernel will use the drive's maximum instead. Setting any value is permitted, there's just generally little point in doing so. Cheers, Robin --=20 ___ =20 ( ' } | Robin Hill | / / ) | Little Jim says .... | // !! | "He fallen in de water !!" | --xHFwDpU9dbj6ez1V Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.11 (GNU/Linux) iEYEARECAAYFAksopLoACgkQShxCyD40xBLlGQCglin5flE0lbmMQRNslM1/Wiy5 1t4AniRMODPkHzY9bHH0Hbk5x30DFyjs =3zgp -----END PGP SIGNATURE----- --xHFwDpU9dbj6ez1V--