On ppc64 hardware (and IBM ppc32 hardware) there is typically a reset
button that causes a non-maskable reset.  This vectors all cpus into a
single function.  I would like to code it to trap into kdb when I do
this.

I have found that kdb doesn't seem to like this with any of the
currently defined "reasons."  All the processors in this case are going
to call into kdb at precisely the same time.  I currently have code in
the ppc64 reset path that lets the first processor into kdb
(KDB_REASON_ENTER) and forces the rest to resume so that kdb can
eventually snag them with its own IPI.  I don't like this because I lose
perfectly good state, and we don't have any other NMI to stop "stuck"
processors :(.

Anyway, the point of this e-mail was to make sure I am not missing
something.  Otherwise I can code up a new KDB_REASON_RESET or such.

-todd



Reply via email to