Re: [PATCH rdma-rc] RDMA/core: Check that type_attrs is not NULL prior access

[Leon Romanovsky <leon@kernel.org>]

On Tue, Jun 16, 2020 at 04:20:54PM -0300, Jason Gunthorpe wrote:
> On Tue, Jun 16, 2020 at 01:58:13PM +0300, Leon Romanovsky wrote:
> > diff --git a/drivers/infiniband/core/rdma_core.c b/drivers/infiniband/core/rdma_core.c
> > index 38de4942c682..16b86635d752 100644
> > +++ b/drivers/infiniband/core/rdma_core.c
> > @@ -470,40 +470,41 @@ static struct ib_uobject *
> >  alloc_begin_fd_uobject(const struct uverbs_api_object *obj,
> >  		       struct uverbs_attr_bundle *attrs)
> >  {
> > -	const struct uverbs_obj_fd_type *fd_type =
> > -		container_of(obj->type_attrs, struct uverbs_obj_fd_type, type);
> > +	const struct uverbs_obj_fd_type *fd_type;
> >  	int new_fd;
> >  	struct ib_uobject *uobj;
> >  	struct file *filp;
> >
> > +	uobj = alloc_uobj(attrs, obj);
> > +	if (IS_ERR(uobj))
> > +		return uobj;
> > +
> > +	fd_type =
> > +		container_of(obj->type_attrs, struct uverbs_obj_fd_type, type);
> >  	if (WARN_ON(fd_type->fops->release != &uverbs_uobject_fd_release &&
> > -		    fd_type->fops->release != &uverbs_async_event_release))
> > +		    fd_type->fops->release != &uverbs_async_event_release)) {
> > +		uverbs_uobject_put(uobj);
> >  		return ERR_PTR(-EINVAL);
> > +	}
>
> I feel like this is a bit cleaner with a goto unwind ?

I don't have strong opinion about it, will resend.

Thanks

>
> Jason