Re: System hang with ASUS Socket-A motherboard

[Forrest L Norvell]

On Mon, May 19, 2003 at 10:14:28AM +0100, [e-mail] wrote:
> Sven,
> 
> Both Forrest and myself have very similar problems with
> the system just completely freezing ie kernel goes catatonic.
> 
> Originally I though this was because of a bug in the AIC7xxx
> driver but I have had my system lock up during a file system
> check (after rebooting from a previous lockup :-(.
> 
> The SCSI driver is a module so that isn't to be faulted (but the
> card could be).  My (very unscientific) conclusion is that it
> probably is a bug in the kernel interrupt handling code.  It
> does tend to happen much more often under high IDE loads,
> for example CD ripping or file system checking so it could
> be getting an interrupt while still dealing with a previous disk
> interrupt but I could well be talking bullsh*t.  Things I haven't
> yet tried include disabling the SATA chipset and pulling
> out the PCI ethernet card (since I don't actually need it
> any more as linux-2.4.21 supports the builtin 3Com device).
> 
> For the mailing list here is my system details.
> 
> Processor: AMD 2800+.
> Motherboard: AUS A7N8X Deluxe
> Chipset: nforce2 SPP + MCP2-T
> Memory: 1.5GB (3 x 512MB)

I'd like to state for the good of the order that since I updated to
2.4.21-rc2 and stopped trying to use the ac patch, I haven't had any
hangs, although I have had a few "8259A Interrupt" messages logged to
kmsg. No uniprocessor APIC support, ACPI + APM compiled in, SCSI
compiled into kernel instead of modularly, all NForce gunch tainting
up the kernel and working just fine.

Forrest

-- 
       . . . the self-reflecting image of a narcotized mind . . .
ozymandias G desiderata     [e-mail]     desperate, deathless
[phone]       http://www.pushby.com/forrest/       ::AOAIOXXYSZ::