linux-kernel - Re: [PATCH 1/1] infiniband: ulp/iser, fix error retval in iser_create_ib_conn

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <adak4ssi4ky.fsf@roland-alpha.cisco.com>
Date:	Wed, 31 Mar 2010 14:06:21 -0700
From:	Roland Dreier <rdreier@...co.com>
To:	Dan Carpenter <error27@...il.com>
Cc:	Jiri Slaby <jslaby@...e.cz>, linux-kernel@...r.kernel.org,
	jirislaby@...il.com, Or Gerlitz <ogerlitz@...taire.com>
Subject: Re: [PATCH 1/1] infiniband: ulp/iser, fix error retval in iser_create_ib_conn_res

So looking at merging this finally, I think I see one problem with the
proposed patch.  We have:

 > @@ -183,7 +180,7 @@  static int iser_create_ib_conn_res(struct iser_conn *ib_conn)
 >  	ib_conn->fmr_pool = ib_create_fmr_pool(device->pd, &params);
 >  	if (IS_ERR(ib_conn->fmr_pool)) {
 >  		ret = PTR_ERR(ib_conn->fmr_pool);
 > -		goto fmr_pool_err;
 > +		goto out_err;
 >  	}

and

 > @@ -209,12 +206,7 @@  static int iser_create_ib_conn_res(struct iser_conn *ib_conn)
 >  		 ib_conn->fmr_pool, ib_conn->cma_id->qp);
 >  	return ret;
 >  
 > -qp_err:
 > -	(void)ib_destroy_fmr_pool(ib_conn->fmr_pool);
 > -fmr_pool_err:
 > -	kfree(ib_conn->page_vec);
 > -	kfree(ib_conn->login_buf);
 > -alloc_err:
 > +out_err:
 >  	iser_err("unable to alloc mem or create resource, err %d\n", ret);
 >  	return ret;
 >  }

so if ib_create_fmr_pool() fails, we're left with ib_conn->fmr_pool
holding an error pointer, right?  But we're relying on
iser_free_ib_conn_res() to clean up after us, and that has:

	if (ib_conn->fmr_pool != NULL)
		ib_destroy_fmr_pool(ib_conn->fmr_pool);

so we're going to end up trying to free an error pointer, which will
probably crash.

I think.

Dan or Or, am I wrong here or do we need another iteration of this
patch? (the login_buf and page_vec changes do look correct to me, since
a failed kmalloc() will leave us with a NULL pointer that it is safe to
kfree() later)

 - R.
-- 
Roland Dreier <rolandd@...co.com> || For corporate legal information go to:
http://www.cisco.com/web/about/doing_business/legal/cri/index.html
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/