Re: Oops in 2.3.X, 2.2.X, and 2.0.3X with FENRIS and NCR53c875 SCSI Driver

Jeff Merkey (jmerkey@timpanogas.com)
Wed, 9 Jun 1999 16:41:32 -0600

This is a multi-part message in MIME format.

------=_NextPart_000_024F_01BEB296.EE521D70
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

Also shows up single processor (different machine, same chipset 53C875). =
Looks like a possible SCSI Scipts error. FENRIS, unlike UNIX FS's will =
talk to more than one disk or partition concurrently since a volume can =
have multiple segments strip[ed across several drives. May be exposing =
some timing condition with the way I am calling breada abd bread.

Jeff

----- Original Message -----=20
From: Jeff Merkey=20
To: Jeff Merkey ; linux-kernel@vger.rutgers.edu=20
Sent: Wednesday, June 09, 1999 4:36 PM
Subject: Re: Oops in 2.3.X, 2.2.X, and 2.0.3X with FENRIS and =
NCR53c875 SCSI Driver

More on this -- am also getting "..could not get a free page...". =
Tracing with a logic analyzer shows memory corruption. Looks like an =
errant Data Transfer. Machine is a dual processor 500Mhz PIII Server. =
Problem only shows up when SMP is active. Will attempt to reproduce =
using single processor.

Jeff
----- Original Message -----=20
From: Jeff Merkey=20
To: linux-kernel@vger.rutgers.edu=20
Sent: Wednesday, June 09, 1999 4:33 PM
Subject: Oops in 2.3.X, 2.2.X, and 2.0.3X with FENRIS and NCR53c875 =
SCSI Driver

It's busted in all three. Not my bug, I will get Oops specific =
info. Complains that the LRU Block list has been corrupted. Only =
happens with the NCR driver and the 53c875 Chipset. Could be a scripts =
bug since the NCR SCSI chip can program it's own DMA from the SCSI =
scripts.

Jeff

------=_NextPart_000_024F_01BEB296.EE521D70
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">

Also shows up single processor = (different machine,=20 same chipset 53C875).  Looks like a possible SCSI Scipts = error. =20 FENRIS, unlike UNIX FS's will talk to more than one disk or partition=20 concurrently since a volume can have multiple segments strip[ed across = several=20 drives.  May be exposing some timing condition with the way I am = calling=20 breada abd bread.
 
Jeff
 
----- Original Message -----
From:=20 Jeff=20 Merkey
To: Jeff=20 Merkey ; linux-kernel@vger.rutgers.edu =
Sent: Wednesday, June 09, 1999 = 4:36=20 PM
Subject: Re: Oops in 2.3.X, = 2.2.X, and=20 2.0.3X with FENRIS and NCR53c875 SCSI Driver

More on this -- am also getting = "..could not get=20 a free page...".  Tracing with a logic analyzer shows memory=20 corruption.  Looks like an errant Data Transfer.  Machine is = a dual=20 processor 500Mhz PIII Server.  Problem only shows up when SMP is=20 active.  Will attempt to reproduce using single = processor.
 
Jeff
----- Original Message -----
From:=20 Jeff=20 Merkey
To: linux-kernel@vger.rutgers.edu =
Sent: Wednesday, June 09, = 1999 4:33=20 PM
Subject: Oops in 2.3.X, = 2.2.X, and=20 2.0.3X with FENRIS and NCR53c875 SCSI Driver

 
It's busted in all three.  Not = my bug, I=20 will get Oops specific info.  Complains that the LRU Block list = has=20 been corrupted.  Only happens with the NCR driver and the = 53c875=20 Chipset.  Could be a scripts bug since the NCR SCSI chip can = program=20 it's own DMA from the SCSI scripts.
 
Jeff
 
------=_NextPart_000_024F_01BEB296.EE521D70-- - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/