I: [PATCH] few asprintf error handling fixes

James Antill james@and.org
Thu Jun 3 14:38:00 GMT 2004


Roland McGrath <roland@redhat.com> writes:

>> The current behavior is undefined behavior, the result pointer after
>> failed asprintf call may contain any value, e.g. pointer to already freed
>> memory.  I'm doubt that some programs could depend on this behavior.
>
> You mean that actually happens now?

 Very much so, but only on glibc, this was "discussed" in late 2001...

http://sources.redhat.com/ml/libc-alpha/2001-12/msg00021.html
http://sources.redhat.com/ml/libc-alpha/2001-12/msg00029.html
http://sources.redhat.com/ml/libc-alpha/2001-12/msg00029.html
http://sources.redhat.com/ml/libc-alpha/2001-12/msg00041.html

http://sources.redhat.com/ml/libc-alpha/2001-12/msg00045.html
http://sources.redhat.com/ml/libc-alpha/2001-12/msg00050.html
http://sources.redhat.com/ml/libc-alpha/2001-12/msg00051.html

http://sources.redhat.com/ml/libc-alpha/2001-12/msg00100.html
http://sources.redhat.com/ml/libc-alpha/2001-12/msg00121.html

...etc.

-- 
# James Antill -- james@and.org
:0:
* ^From: .*james@and\.org
/dev/null



More information about the Libc-alpha mailing list