Re: lib/43899: setenv(3)/unsetenv(3) memory leak

[Nicolas Joly <njoly%pasteur.fr@localhost>]

The following reply was made to PR lib/43899; it has been noted by GNATS.

From: Nicolas Joly <njoly%pasteur.fr@localhost>
To: gnats-bugs%NetBSD.org@localhost, netbsd-bugs%NetBSD.org@localhost
Cc: 
Subject: Re: lib/43899: setenv(3)/unsetenv(3) memory leak
Date: Mon, 27 Sep 2010 18:24:07 +0200

 On Sat, Sep 25, 2010 at 05:16:33PM +0900, ITOH Yasufumi wrote:
 > In article <20100923131001.5C93F63B995%www.NetBSD.org@localhost>
 > njoly%pasteur.fr@localhost writes:
 > 
 > > >Description:
 > > I just noticed that unsetenv(3) does not seems to reclaim memory that was
 > > allocated by setenv(3) for a single variable.
 > 
 > setenv(3) leaks memory by its nature.  It's a spec.
 
 True. While i can understand that setenv(3) only in a loop do/can leak
 memory, but if associated with unsetenv(3) in the same loop it should
 not.
 
 > The problem is that our implementation of putenv(3) leaks memory.
 > The standard says we can use putenv(3) without memory leaks.
 > http://www.opengroup.org/onlinepubs/9699919799/functions/putenv.html
 
 Looks like it should try to unset the variable if it exists before
 setting it again.
 
 -- 
 Nicolas Joly
 
 Biological Software and Databanks.
 Institut Pasteur, Paris.