[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Your changes in revision 106240

From: Eli Zaretskii
Subject: Re: Your changes in revision 106240
Date: Sun, 30 Oct 2011 02:16:31 -0400

> From: Stefan Monnier <address@hidden>
> Cc: Eli Zaretskii <address@hidden>,  address@hidden
> Date: Sun, 30 Oct 2011 01:06:16 -0400
> >> what possible harm can be done by using an int here?
> > On a typical 64-bit host, using 'int' breaks a GDB command like
> > "xwhichsymbols Qnil 4294967297", by causing xwhichsymbols
> > to silently treat the 4294967297 as if it were 1.
> But this has nothing to do with EMACS_INT.  It's at best a misfeature in
> the GDB->C interfacing that doesn't warn of such rounding, tho one might
> argue that they are intentional and that a programmer should expect them.

Even if we wanted to support such ridiculously large values, TRT would
be to make that argument `long', not EMACS_INT, because we have the
"--with-wide-int" configuration where EMACS_INT is a 64-bit type
whereas GDB will pass a 32-bit value.

> In any case, this arg is *never* going to be that large.

Right.  And moreover, a GDB user who stupidly passes such a large
value will hurt no one but herself, so we don't need to protect Emacs
from that.

reply via email to

[Prev in Thread] Current Thread [Next in Thread]