Re: setenv memory leak
Re: setenv memory leak
- Subject: Re: setenv memory leak
- From: "Peter O'Gorman" <email@hidden>
- Date: Mon, 30 Oct 2006 23:59:12 +0900
On Oct 30, 2006, at 10:06 PM, email@hidden wrote:
Just documenting a bit for anybody out there running into the same
problem
: g++ has a bit of a memory leak in its setenv function on os X (xcode
2.2)
Fbsd used to have it too, on my FreeBSD 6.1-RELEASE ws at work it
seems to
be patched.
us = mktime(&loctime);
setenv("TZ", "GMT", 1);
struct tm *theirtime = localtime(&us);
setenv("TZ", "CET", 1);
BUGS section of the manpage for setenv(3).
Successive calls to setenv() or putenv() assigning a
differently sized
value to the same name will result in a memory leak. The
FreeBSD seman-
tics for these functions (namely, that the contents of value
are copied
and that old values remain accessible indefinitely) make this bug
unavoidable. Future versions may eliminate one or both of
these semantic
guarantees in order to fix the bug.
Also this comment in the source code: <http://
www.opensource.apple.com/darwinsource/10.4.7.ppc/Libc-391.2.6/stdlib/
FreeBSD/setenv.c.patch>
+ /* In UNIX03, we can't overwrite even if the string is long
+ * enough, because the putenv() string is owned by the user
+ * (ie, always malloc() a new string) */
And <http://www.opensource.apple.com/darwinsource/10.4.7.ppc/
Libc-391.2.6/stdlib/FreeBSD/setenv.c.>:
/* For non Unix03, or UnixO3 setenv(), we make a copy of the
user's
* strings. For Unix03 putenv(), we put the string directly in
* the environment. */
There is definitely a documentation bug in the manpage, and the fact
that it appears to be impossible to set an environment variable
without leaking is either a bug in the standard or in someone's
interpretation of it. Whatever, file a bug :)
Peter
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Xcode-users mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden