recursive panics

recursive panics

[Jan Beulich]

With crash_debug enabled, we're seeing recursive panics because of the
embedded int3 in panic(), as long as interrupts were disabled at the top
level exception's origin (in the given case, in the context switch code). It
would seem to me that even in this case, reporting back to the debugger,
if any, or ignoring the breakpoint (if there's no debugger) should be the
normal operation, but I'm somewhat hesitant to add more special case
code to entry.S. Are there any ideas how to solve this in a more elegant
way?

Thanks, Jan

Re: recursive panics

[Keir Fraser]

I don't think this has been the case since Xen 3.0.4. In newer Xen, the
entry.S asm code happily passes through exceptions with interrupts disabled.

 -- Keir

On 15/11/07 13:37, "Jan Beulich" <jbeulich@novell.com> wrote:

> With crash_debug enabled, we're seeing recursive panics because of the
> embedded int3 in panic(), as long as interrupts were disabled at the top
> level exception's origin (in the given case, in the context switch code). It
> would seem to me that even in this case, reporting back to the debugger,
> if any, or ignoring the breakpoint (if there's no debugger) should be the
> normal operation, but I'm somewhat hesitant to add more special case
> code to entry.S. Are there any ideas how to solve this in a more elegant
> way?
> 
> Thanks, Jan
> 
> 
> _______________________________________________
> Xen-devel mailing list
> Xen-devel@lists.xensource.com
> http://lists.xensource.com/xen-devel