Messages in this thread | | | Date | Fri, 17 Dec 2010 13:18:55 +0100 (CET) | From | Jesper Juhl <> | Subject | Re: [PATCH][rfc] md: Close mem leak in userspace_ctr() |
| |
On Tue, 14 Dec 2010, Mike Snitzer wrote:
> On Mon, Dec 13 2010 at 6:40pm -0500, > Jesper Juhl <jj@chaosbits.net> wrote: > > > Hi, > > > > There's a small memory leak in > > drivers/md/dm-log-userspace-base.c::userspace_ctr(). > > > > The call to build_constructor_string() dynamically allocates memory for > > its last argument, but we do not always clean up that allocated memory. > > I'm not seeing a leak. > > The kfree() that you've added to the build_constructor_string() failure > path isn't needed because build_constructor_string() only returns error > if the kzalloc() failed. >
Hmm, that's true, assuming that there are no bugs (and never will be) in strlen() or sprintf() that result in 'str_size' becoming negative the code is safe. That's probably a safe assumption ;-)
Sorry about the noise.
-- Jesper Juhl <jj@chaosbits.net> http://www.chaosbits.net/ Don't top-post http://www.catb.org/~esr/jargon/html/T/top-post.html Plain text mails only, please.
| |