[2240] in Kerberos_V5_Development
Re: Memory allocation/freeing
daemon@ATHENA.MIT.EDU (Theodore Y. Ts'o)
Wed Feb 19 01:52:17 1997
Date: Wed, 19 Feb 1997 01:51:58 -0500
From: "Theodore Y. Ts'o" <tytso@MIT.EDU>
To: Ken Hornstein <kenh@cmf.nrl.navy.mil>
Cc: "Theodore Y. Ts'o" <tytso@MIT.EDU>, "Richard Basch"
<basch@lehman.com>,
krbdev@MIT.EDU
In-Reply-To: Ken Hornstein's message of Wed, 19 Feb 1997 01:39:51 -0500,
<199702190639.BAA06489@ginger.cmf.nrl.navy.mil>
Date: Wed, 19 Feb 1997 01:39:51 -0500
From: Ken Hornstein <kenh@cmf.nrl.navy.mil>
>If we had an API document, it'd say something about "use this only if
>you really know what you are doing it." :-) Actually, as long as you
>follow a few simple rules, it's not all that hard.
Well ... I thought that current API document _does_ talk about the way
krb5_build_principal_ext works.
It does; I hadn't realized that. I didn't think it had been updated to
include it. The description isn't completely accurate, though, and
should be fixed.
Speaking of that .... exactly how come we are not supposed to be using
krb5_xfree anymore? Since it was mentioned in the API document, I figured
we _were_, and I've been using it.
Why? Because it's just an ugly kludge to keep lint happy; it shouldn't
have been documented in the API document. The original reason for it is
actually long gone, by the way. It was only needed on platforms where
free() was declared free(char *), instead of free(void *).
We don't need krb5_xfree() at all these days, actually.....
- Ted
