From mboxrd@z Thu Jan 1 00:00:00 1970 From: Carsten Otto Subject: Re: [2.6.23-rc3] Regression: "spurious completions during NCQ" Date: Sat, 22 Sep 2007 15:11:04 +0200 Message-ID: <20070922131104.GG32509@server.c-otto.de> Reply-To: Carsten Otto Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="a1QUDc0q7S3U7/Jg" Return-path: Received: from mail.gmx.net ([213.165.64.20]:54636 "HELO mail.gmx.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with SMTP id S1751530AbXIVNLI (ORCPT ); Sat, 22 Sep 2007 09:11:08 -0400 Received: from carsten-otto.halifax.rwth-aachen.de ([137.226.108.169] helo=server.c-otto.de) by server.c-otto.de with smtp (Exim 4.67) (envelope-from ) id 1IZ4lE-0007kR-Os for linux-ide@vger.kernel.org; Sat, 22 Sep 2007 15:11:05 +0200 Content-Disposition: inline Sender: linux-ide-owner@vger.kernel.org List-Id: linux-ide@vger.kernel.org To: linux-ide@vger.kernel.org --a1QUDc0q7S3U7/Jg Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable Hi! (I am referring to the thread at http://marc.info/?l=3Dlinux-ide&m=3D118703097726277&w=3D2) I also experience a lot of NCQ errors which look like these: Sep 22 10:11:53 server kernel: ata1.00: exception Emask 0x2 SAct 0x7 SErr 0= x0 action 0x2 frozen Sep 22 10:11:53 server kernel: ata1.00: (spurious completions during NCQ is= sue=3D0x0 SAct=3D0x7 FIS=3D004040a1:00000010) Sep 22 10:11:53 server kernel: ata1.00: cmd 60/38:00:bd:ee:01/00:00:13:00:0= 0/40 tag 0 cdb 0x0 data 28672 in Sep 22 10:11:53 server kernel: res 40/00:20:5d:1d:eb/00:00:13:00:0= 0/40 Emask 0x2 (HSM violation) Sep 22 10:11:53 server kernel: ata1.00: cmd 60/40:08:f5:ee:01/00:00:13:00:0= 0/40 tag 1 cdb 0x0 data 32768 in Sep 22 10:11:53 server kernel: res 40/00:20:5d:1d:eb/00:00:13:00:0= 0/40 Emask 0x2 (HSM violation) Sep 22 10:11:53 server kernel: ata1.00: cmd 60/08:10:35:ef:01/00:00:13:00:0= 0/40 tag 2 cdb 0x0 data 4096 in Sep 22 10:11:53 server kernel: res 40/00:20:5d:1d:eb/00:00:13:00:0= 0/40 Emask 0x2 (HSM violation) Sep 22 10:11:53 server kernel: ata1: soft resetting port Sep 22 10:11:53 server kernel: ata1: SATA link up 3.0 Gbps (SStatus 123 SCo= ntrol 300) Sep 22 10:11:53 server kernel: ata1.00: configured for UDMA/133 Sep 22 10:11:53 server kernel: ata1: EH complete Sep 22 10:11:53 server kernel: sd 0:0:0:0: [sda] 586114704 512-byte hardwar= e sectors (300091 MB) Sep 22 10:11:53 server kernel: sd 0:0:0:0: [sda] Write Protect is off Sep 22 10:11:53 server kernel: sd 0:0:0:0: [sda] Mode Sense: 00 3a 00 00 Sep 22 10:11:53 server kernel: sd 0:0:0:0: [sda] Write cache: disabled, rea= d cache: enabled, doesn't support DPO or FUA In total I have 355 such entries regarding my SATA disks starting on September 15 when I installed kernel 2.6.22.6 (before I had 2.6.18-rc4-mm1 or so). In addition I also see messages like this one: kern.log:Sep 20 00:42:23 server kernel: ata2.00: NCQ disabled due to excess= ive errors Please tell me how I can help resolving the issue (if it is a bug) or, if possible, tell me how to fix it myself. Information about my system: - 4x Maxtor 7V300F0 (MaxLine III, 300GB, 7200rpm, 16MB cache, SATA2), firmware VA111900 - 1x Samsung 2SP2014N (SpinPoint P, 200GB, 7200rpm, 8MB cache, PATA), firmware VC100-33 - 4 software RAIDs on the SATA disks, 1x RAID1, 3x RAID5 - heavy usage of the RAIDs - ICH7R - 2.6.22.6 - Pentium D 805, 2GB RAM Thanks a lot, --=20 Carsten Otto c-otto@gmx.de www.c-otto.de --a1QUDc0q7S3U7/Jg Content-Type: application/pgp-signature; name="signature.asc" Content-Description: Digital signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) iD8DBQFG9RRojUF4jpCSQBQRAotNAJ9UQ3FuTRiZqLn+NRZukW/e+rHpkQCfdQ29 FJ5Dk4Wdh03GH+MGCizv0o8= =T3p+ -----END PGP SIGNATURE----- --a1QUDc0q7S3U7/Jg--