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: Rich Wales <richw at richw.org>
Cc: bug-followup at FreeBSD.org,  freebsd at jdc.parodius.com
Subject: Re: i386/103435: Kernel appears somewhat deadlocked during heavy
 ATA I/O (post-August 4th)
Date: Sat, 07 Oct 2006 18:11:06 -0700

 Per a suggestion from Søren Schmidt, I rechecked my hardware and
 my BIOS settings.  After a bunch of experimentation, I found that
 if I disabled PCI master burst mode in the BIOS setup, the timeout
 problems all went away.  Disk I/O speed was -- not surprisingly --
 noticeably slower with burst mode turned off, but the system was
 stable and reliable.
 
 I'm not sure if the primary culprit here is my Promise card or my
 motherboard, but it does suggest that some systems with iffy PCI
 bus tolerances might require suboptimal BIOS settings in order to
 work properly with some Promise controllers.
 
 As for whether any effective steps can be taken in the driver to
 make it more robust against this problem, I'll have to leave that
 to people who are more familiar with the hardware and the code.
 
 Rich Wales            richw at richw.org            http://www.richw.org