Is this a race in mbuf's refcounting?

[Ed Maste]

On Mon, Sep 21, 2009 at 01:43:33PM +0100, Andrew Brampton wrote:

> I've been reading the FreeBSD source code to understand how mbufs are
> reference counted. However, there are a few bits of code that I'm
> wondering if they would fail under the exactly right timing. Take for
> example in uipc_mbuf.c:
> 
>  286 static void
>  287 mb_dupcl(struct mbuf *n, struct mbuf *m)
>  288 {
> ...
>  293        if (*(m->m_ext.ref_cnt) == 1)
>  294                *(m->m_ext.ref_cnt) += 1;
>  295        else
>  296                atomic_add_int(m->m_ext.ref_cnt, 1);
> ...
>  305 }
> 
> Now, the way I understand this code is, if ref_cnt is 1, then it is
> not shared. In that case non-atomically increment ref_cnt. However, if
> ref_cnt was something else, then it is shared so update the value in
> an atomic way. This seems valid, however what happens if two threads
> call mb_dupcl at the same time with a non-shared m. Could they both
> evaluate the if on line 293 at the same time, and then both
> non-atomically increment ref_cnt?

Your analysis is correct; this issue also has a PR, kern/137145.
http://www.freebsd.org/cgi/query-pr.cgi?pr=kern/137145

As you point out it requires that two threads have a reference to the
same non-shared mbuf.  I had a quick look and didn't find any case of
this in the vanilla FreeBSD tree; if I didn't miss anything it'll
affect only 3rd party src.

We'll need to have a look at this after 8.0 is done.

-Ed