[11702] in Kerberos-V5-bugs
Re: [krbdev.mit.edu #6777] Segmentation fault in krb library
daemon@ATHENA.MIT.EDU (Tom Yu via RT)
Fri Sep 17 15:30:49 2010
Mail-followup-to: rt@krbdev.mit.edu
mail-copies-to: never
From: "Tom Yu via RT" <rt-comment@krbdev.MIT.EDU>
In-Reply-To: <rt-6777@krbdev.mit.edu>
Message-ID: <rt-6777-33172.17.1431432390998@krbdev.mit.edu>
To: "'AdminCc of krbdev.mit.edu Ticket #6777'":;"'AdminCc of krbdev.mit.edu Ticket #6777'":;@MIT.EDU
Date: Fri, 17 Sep 2010 15:30:47 -0400 (EDT)
Reply-To: rt-comment@krbdev.MIT.EDU
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
Errors-To: krb5-bugs-bounces@mit.edu
"Greg Hudson via RT" <rt-comment@krbdev.mit.edu> writes:
> [tlyu - Fri Sep 17 14:46:43 2010]:
>> Perhaps strdup() on AIX can return NULL for a zero-length string? For
>> that matter, does malloc() on AIX return NULL for a zero argument?
>
> I don't think it's fruitful to speculate, only to step through in a
> debugger.
>
> (krb5int_build_principal_va() would not invoke malloc(0) if the realm is
> empty. It would invoke strdup("") but a lot more software would break on
> AIX if it's strdup were so broken as to return NULL in that case.)
Actually krb5_build_principal_va() in krb5-1.6.x (the krb5int_ version
is newer than 1.6) does execute malloc(rlen) and check it for NULL.
_______________________________________________
krb5-bugs mailing list
krb5-bugs@mit.edu
https://mailman.mit.edu/mailman/listinfo/krb5-bugs
