This is the mail archive of the
glibc-bugs@sourceware.org
mailing list for the glibc project.
[Bug manual/1997] opern_memstream() buffer should be freed
- From: "michael dot kerrisk at gmx dot net" <sourceware-bugzilla at sourceware dot org>
- To: glibc-bugs at sources dot redhat dot com
- Date: 13 Dec 2005 09:05:40 -0000
- Subject: [Bug manual/1997] opern_memstream() buffer should be freed
- References: <20051208155151.1997.michael.kerrisk@gmx.net>
- Reply-to: sourceware-bugzilla at sourceware dot org
------- Additional Comments From michael dot kerrisk at gmx dot net 2005-12-13 09:05 -------
Subject: Re: opern_memstream() buffer should be freed
> Von: "decimal at us dot ibm dot com" <sourceware-bugzilla@sourceware.org>
> I took a look in the linux man pages and in the glibc manual and I found
> that while the man page for strdup(3) does say
>
> The strdup() function returns a pointer to a new string which is a
> duplicate of the string s. Memory for the new string is obtained with
> malloc(3), and can be freed with free(3).
>
> the glibc manual info entry for strdup() says
>
> This function copies the null-terminated string s into a newly allocated
> string.
> The string is allocated using malloc; see Unconstrained Allocation. If
> malloc
> cannot allocate space for the new string, strdup returns a null pointer.
> Otherwise it returns a pointer to the new string.
>
> The section on Unconstrained Allocation includes a section "Freeing after
> Malloc" which discusses free().
>
> The current section in the glibc manual on open_memstream() says
>
> This function opens a stream for writing to a buffer. The buffer is
> allocated
> dynamically (as with malloc; see Unconstrained Allocation) and grown as
> necessary.
>
> So is it really necessary to explicitly mention free() next to every
> function
> which uses malloc()? I'm asking to see if anyone else has a strong
> opinion.
Being the Linux manual page maintainer (but I didn't write that
strdup(2)) text, I'm inclined to the view that it is useful
to mention free() when describing these interfaces. It is just
too easy to create memory leaks in C: giving people more direct
hints (instead of suggesting a hyperlink in the doc, in which it
only becomes clear that free() is needed after quite a bit of
reading) alerts people to the issue.
There is a second reason for doing this in the case of
open_memstream(): the interface is non-standard. I can determine
from any number of places (my own knowledge, the SUSv3 spec,
manual pages on various systems) that glibc's strdup() must be
followed with a free(). However, those sources of information
are not available for open_memstream().
I realise there are differences on documentation philosophy for
"info" and the manual pages, but I do think an explicit mention of
free() could be valuable here (and perhaps in a few other places).
Cheers,
Michael
--
http://sourceware.org/bugzilla/show_bug.cgi?id=1997
------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.