R> I see 12 pushes and 12 pops, so SP will have the same value on exit as it 
R> did on entry. BX will also be set correctly; the first pop sets it to the 
R> pushed value of SP, then the second pop the pushed value of BX.

R> The order of pushes matches exactly that of the 186+ opcode PUSHA; POPA also 
R> discards SP, same as the above code.

R> I don't see any problems with the above code at all ... it appears to 
R> correctly emulate PUSHA and POPA.

Nobody likes a smarty-pants, Rowan.  BFN.  Paul.
@EOT:

---