cvs commit: src/sys/dev/fxp if_fxp.c if_fxpvar.h

[Andrew Gallatin]

Warner Losh [imp at FreeBSD.org] wrote:
>           2) Call FXP_UNLOCK() before calling bus_teardown_intr to avoid
>              a possible deadlock reported by jhb.
>           3) add gone to the softc.  Set it to true in detach.

>           5) Return immediately if gone is true in fxp_intr
>   

My workstation has been up for ~5 hours now, and has seen 1.3 million 
fxp interrupts.  That's 1.3 million extra compares that this code
is now executing.

What is this deadlock with bus_teardown_intr?  Could we possibly fix
this some other way than by adding (mostly) useless code to the
critical path? 

Drew