easy to reproduce unkillable threads

[Andrew Gallatin]

David Xu writes:
 > Now, I can reproduce it on SMP.  However, if I :

Hurray!  I'm not crazy!

 > sysctl -w kern.smp.forward_signal_enabled=0
 > then, I can always kill it. it sounds like another IPI bug.
 > 

I thought the IPI bugs were IPIs getting stuck.  But this IPI was
delivered.  Naively, I would say that it seems more like an MP-safety
issue in the kernel thread exit code..


Drew