Skip to contentSkip to conversation

net/9p/mux.c: use-after-free

DISCUSSION Adrian Bunk 3 participants 5 messages Jul 23, 2007 – Jul 26, 2007 1 message not in the archive

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

The Coverity checker spotted the following use-after-free in net/9p/mux.c:

<-- snip -->

...
struct p9_conn *p9_conn_create(struct p9_transport *trans, int msize,
unsigned char *extended)
{
...

        if (!m->tagpool) {
                kfree(m);
                return ERR_PTR(PTR_ERR(m->tagpool));
        }

...

<-- snip -->

cu
Adrian

Signature · 4 lines
       "Is there not promise of rain?" Ling Tan asked suddenly out
        of the darkness. There had been need of rain for many days.
       "Only a promise," Lao Er said.
                                       Pearl S. Buck - Dragon Seed
#1 lore raw
Quoted 10 lines · Adrian Bunk

The Coverity checker spotted the following use-after-free in net/9p/mux.c:

<-- snip -->

...
struct p9_conn *p9_conn_create(struct p9_transport *trans, int msize,
unsigned char *extended)
{
...

Quoted 8 lines
        if (!m->tagpool) {
                kfree(m);
                return ERR_PTR(PTR_ERR(m->tagpool));
        }

...

<-- snip -->

I've got a fix for this one:

       if (!m->tagpool) {
               mtmp = ERR_PTR(PTR_ERR(m->tagpool));
               kfree(m);
               return mtmp;
       }

but I was wondering about one of the other returns further down the function:

...

       memset(&m->poll_waddr, 0, sizeof(m->poll_waddr));
       m->poll_task = NULL;
       n = p9_mux_poll_start(m);
       if (n)
               return ERR_PTR(n);

       n = trans->poll(trans, &m->pt);

...

lucho: doesn't that constitute a leak? Shouldn't we be doing:

       if (n) {
               kfree(m);
               return ERR_PTR(n);
       }

        -eric
#2 lore raw

Yep, it's a leak.

Thanks,
Lucho

Quoted 48 lines · Eric Van Hensbergen
Quoted 10 lines · Adrian Bunk

The Coverity checker spotted the following use-after-free in net/9p/mux.c:

<-- snip -->

...
struct p9_conn *p9_conn_create(struct p9_transport *trans, int msize,
unsigned char *extended)
{
...

Quoted 8 lines
        if (!m->tagpool) {
                kfree(m);
                return ERR_PTR(PTR_ERR(m->tagpool));
        }

...

<-- snip -->

I've got a fix for this one:

        if (!m->tagpool) {
                mtmp = ERR_PTR(PTR_ERR(m->tagpool));
                kfree(m);
                return mtmp;
        }

but I was wondering about one of the other returns further down the function:

...

        memset(&m->poll_waddr, 0, sizeof(m->poll_waddr));
        m->poll_task = NULL;
        n = p9_mux_poll_start(m);
        if (n)
                return ERR_PTR(n);

        n = trans->poll(trans, &m->pt);

...

lucho: doesn't that constitute a leak? Shouldn't we be doing:

        if (n) {
                kfree(m);
                return ERR_PTR(n);
        }

         -eric
#3 lore raw
Quoted 2 lines · Latchesar Ionkov

Yep, it's a leak.

Okay, I'll roll that into the patch as well.

-eric

#5 lore raw
In reply to a message not in the archive
Quoted 6 lines · Andrew Morton
Quoted 1 line · On Wed, 25 Jul 2007 13:43:16 -0500 "Eric Van Hensbergen" <[email protected]>…
                mtmp = ERR_PTR(PTR_ERR(m->tagpool));

odd. What does ERR_PTR(PTR_ERR(...)) do?

I kind of assumed it was a necessry evil to get the casting right. A quick grep shows it in 42 other places within the kernel. Unpacking the macros it looks like:

(void *)(long)(struct p9_idpool *)

So all that you would really need is (void *) or ERR_PTR -- but that might look confusing in the code. Of course, broadening the context a bit:

       m->tagpool = p9_idpool_create();
       if (!m->tagpool) {
               mtmp = ERR_PTR(PTR_ERR(m->tagpool));
               kfree(m);
               return mtmp;
       }

m->tagpool must be zero to enter the code at all, so we are returning a NULL pointer, not really an error -- which is probably wrong (I don't think it will properly trigger IS_ERR_VALUE) -- so we should probably be returning -ENOMEM.

Of course, we really should be seeing an ERR_PTR returned from p9_idpool_create, not 0 -- checking that code, it either returns -ENOMEM or the correct value, never 0, so the check is wrong as well. It should be:

       m->tagpool = p9_idpool_create();
       if (IS_ERR(m->tagpool)) {
               mtmp = ERR_PTR(-ENOMEM);
               kfree(m);
               return mtmp;
       }

We could have done:
ERR_PTR(m->tagpool);
or kept the long:
ERR_PTR(PTR_ERR(m->tagpool));
but I think returning an explicit error code keeps the code more clear.

So, which is the correct approach?

-eric