i386/103435: Kernel appears somewhat deadlocked during heavy ATA I/O (post-August 4th)

[Rich Wales]

The following reply was made to PR i386/103435; it has been noted by GNATS.

From: Rich Wales <richw at richw.org>
To: bug-followup at FreeBSD.org
Cc: freebsd at jdc.parodius.com
Subject: Re: i386/103435: Kernel appears somewhat deadlocked during heavy
 ATA I/O (post-August 4th)
Date: Sun, 01 Oct 2006 21:10:09 -0700

 I'm seeing a similar problem on an old (800-MHz original "Slot A"
 Athlon) system with a Promise SATA300 TX4 card (PDC40718) and two
 Seagate 300GB SATA drives (ST3300831AS), running 6.1-RELEASE-p9.
 
 This experimental system locked up twice with the same kinds of
 messages that Jeremy Chadwick reported -- once when I did a "sync"
 after copying the /usr file system from an old PATA disk to one
 of the new SATA disks, and again as I was trying to synchronize
 the second SATA drive to the first via "gmirror".
 
 This is admittedly an old CPU, motherboard, and RAM, but the system
 has been running quite reliably until now.
 
 Rich Wales, Palo Alto, CA, USA, richw at richw.org, http://www.richw.org