On Tue, 9 Oct 2007, karvjorm at users dot sourceforge dot net wrote:
------- Additional Comments From karvjorm at users dot sourceforge dot net
2007-10-09 05:09 -------
(In reply to comment #1)
Well, it's 8-bit signed *and* 8-bit unsigned: perhaps saying
"not in 8-bit signed or unsigned range" would sound better to you?
If you have a better wording to suggest, feel free.
brgds, H-P
Yes, that is one way to get around that, but there are still holes in that error
message. If it is a signed number, we don't catch values between 127...255 and
if it is an unsigned number, we don't catch values between -128 and 0. And there
is no way to know, which one we have, I suppose.
That's what I'm trying to say; where this message can be issued,
the operand can be either signed or unsigned and the specific
interpretation is not part of the semantics for this particular
operand. The value at hand is 8 bits, that's all that's known,
and further smartness from gas would not be appropriate. There
are no "holes" here. The test and message is just to catch the
most trivial programmer errors.
brgds, H-P
_______________________________________________
bug-binutils mailing list
bug-binutils@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-binutils
