When I break the ICE, when the system is hung in the CPU idle loop, it looks to me
like interrupts are enabled (eflags bit 9 = 1).  

What other conditions could cause the system to get stuck at the HLT, at the end
of the CPU idle loop?

Frank van der Linden wrote:
> 
> On Tue, Oct 16, 2001 at 11:31:30AM -0400, Brian Clough wrote:
> > Actually, after the call, the system continues for a short time then stops dead,
> > no error messages or anything.  I can output characters to the console so that's
> > how I know it's returning after the set mode call.  I do the actual BIOSCALL
> > inside an IOCTL in the video driver, then call the driver from an app.  I get back
> > to the app, but when I do get back I don't have long to live.
> >
> > I've got an ICE connected to the system, and I can see where execution is hitting
> > a HLT, but how can I find the corresponding location in the source code?
> 
> The hlt is likely the idle CPU loop.. which means that you might not
> be getting any interrupts anymore at that point.
> 
> Can you check if the call in question perhaps disables interrupts and
> not enables them again?
> 
> - Frank
> 
> --
> Frank van der Linden                           fvdl@wasabisystems.com
> ======================================================================
> Quality NetBSD CDs, Support & Service.   http://www.wasabisystems.com/

-- 
Brian Clough
OS Development Group, Norwood, MA
x(82)5015
(781)551-5015
mailto:brian_clough@phoenix.com