I know that the code will work much better after you add
   your copyright or run the code through indent :-).  BTW, I am suprised
   that you haven't fixed a significant oversight in the machine independent SCSI
   subsystem yet... (just a for instance.)  I guess those TAILQs are really
   important!!!  (BTW, we haven't fully fixed it yet -- I think.)  You probably
   won't manifest the problem until you port one of our drivers...

why don't you start the "nice" ball rolling by telling us what, exactly,
is the problem here ?

.mrg.