Re: Another illegal instruction awk problem

[der Mouse <mouse%Rodents-Montreal.ORG@localhost>]

>> However, I have trouble seeing how "movd r9,r0" could fault showing
>> SIGILL.  While I don't have the -5 source at ready hand, the 4.0.1
> My guess would be that it's trying to move a 'NaN', which someone had
> decided should be a value that causes a reserved operand fault on the
> vax.

Yeah, the VAX doesn't have NANs; it wouldn't surprise me if a reserved
operand were used instead of signaling NANs, since the semantics are
moderately close.  (There isn't anything at all close to quiet NANs, as
I daresay you know but possibly other readers don't.)

I thought of that but didn't mention it because I _thought_ MOVD was
defined to be data-blind, functionally equivalent to MOVQ.  On checking
my VARM, I find you're right, it does test for reserved operands, and,
given that, I agree that that's the most likely basis for the SIGILL.

> I'd have to see the contents of the source register to be sure.

John, could you maybe fire up "gdb awk awk.core" again and show us the
output of "info reg"?

/~\ The ASCII                             Mouse
\ / Ribbon Campaign
 X  Against HTML                mouse%rodents-montreal.org@localhost
/ \ Email!           7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B