From: "M. Schulter" <mschulter AT value DOT net>
Newsgroups: comp.os.msdos.djgpp
Subject: Re: fsdb question: cltd shown as cwd
Date: 12 Oct 1997 06:44:09 GMT
Organization: Value Net Internetwork Services Inc.
Lines: 31
Message-ID: <61prjp$8j0$1@vnetnews.value.net>
References: <199710100004 DOT RAA14953 AT adit DOT ap DOT net>
NNTP-Posting-Host: value.net
To: djgpp AT delorie DOT com
DJ-Gateway: from newsgroup comp.os.msdos.djgpp
Precedence: bulk

Nate Eldredge <eldredge AT ap DOT net> wrote:
: Yeah, this is fsdb's fault. The problem is that `cdw' and `cdq' have the
: same opcode, and are distinguished by either their presence in a 16- or
: 32-bit segment, or by a prefix byte. (An ugly kludge on Intel's part, IMHO.)

Thank you very much for this explanation, which teaches me something about
assembly generally as well as fsdb.

: This is, AFAIK, the only case in which this actually changes the name of the
: opcode (in Intel syntax). Usually it's a matter of using %ax vs. %eax, 16-
: vs. 32-bit registers. It seems fsdb is not smart enough to be able to change
: the opcode name depending on the operand size. I guess you can either live
: with it, or add the feature to fsdb.

Would that I had that level of competence <grin> -- and I certainly
wouldn't complain about fsdb's performance on this point. It was just a
point of curiosity, which you have amply satisfied.

Incidentally, while I often use fsdb either to debug or to test a program
for possible bugs (suspecting that at times this might prevent a crash if
an error does occur), I also enjoy using it just to watch the registers in
action. The fine point we've been discussing quite aside, a 64-bit
division is very impressive and even beautiful to watch.

: Nate Eldredge
: eldredge AT ap DOT net

Most appreciatively,
Margo Schulter
mschulter AT value DOT net