2940U2W under Linux 2.4.x

[Justin T. Gibbs]

>	Ok, things seem to behave a little differently with the modification
>you asked for (but without KDB compiled in).  I've included two bootlogs 
>from that as they seem to have failed in marginally different ways. 
>I've seen this type of problem in other 2.4.3+ kernels, and I'm not sure 
>what triggers it vs. the instant failure we've been seeing.  Anyway, 
>I've also attached a bootlog with the modification and KDB, which fails 
>in the original way.  One problem that cropped up is you asked for a 
>backtrace (bt) which doesn't work-- it says it's not in kernel space.  I 
>did a process back trace on the active task instead, although I'm not 
>sure that's helpful.

Have you verified that there are no new BIOS versions available for your
motherboard?  From what I can tell, when we write to registers over the
PCI bus, we are not getting the expected results.  This may mean that
the PCI chipset has a non-standard caching feature enabled that violates
the PCI spec.  I've seen this on some other VIA based boards.

--
Justin

To Unsubscribe: send mail to majordomo at FreeBSD.org
with "unsubscribe aic7xxx" in the body of the message