STI, HLT in acpi_cpu_idle_c1

[Don Bowman]

From: John Baldwin [mailto:jhb at FreeBSD.org]
> On Friday 25 June 2004 05:05 pm, Gerrit Nagelhout wrote:
> > John Baldwin wrote:
> > > Odd, all it does is eoi fast interrupts earlier.  Oh, there's
> > > a bug. :(  In
> > > the second hunk, change 'pic_disable_source' to 
> 'pic_enable_source'.
> >
> > sigh...it just locked up again.  I have included the current apic
> > dump for completeness.  Is there anything other information that
> > would be useful to debug this problem?
> 
> Hmm, it appears it is consistently CPU1 that thinks that 
> IPI_HARDCLOCK is not 
> being EOI'd.  Can you try swapping the CPUs and see if the 
> lockup moves from 
> CPU1 to CPU3?

There is more than one device that does this, so
its probably not worth it (its a huge huge effort
on these systems due to the way the heatsink
is bonded on: it requires breaking the heatsink
glue with a heat gun, disassembling the motherboard,
etc).