On Thu, Oct 01, 2009 at 03:59:00PM -0700, Sean Hefty wrote:
> >diff --git a/opensm/opensm/osm_drop_mgr.c b/opensm/opensm/osm_drop_mgr.c
> >index 4f98cc9..8fe5129 100644
> >+++ b/opensm/opensm/osm_drop_mgr.c
> >@@ -209,8 +209,13 @@ static void drop_mgr_remove_port(osm_sm_t * sm, IN
> >osm_port_t * p_port)
> > drop_mgr_clean_physp(sm, p_port->p_physp);
> >
> > while (!cl_is_qlist_empty(&p_port->mcm_list)) {
> >+#ifndef __WIN__
> > mcm_port = cl_item_obj(cl_qlist_head(&p_port->mcm_list),
> > mcm_port, list_item);
> >+#else
> >+ mcm_port = cl_item_obj(cl_qlist_head(&p_port->mcm_list),
> >+ mcm_port, list_item, (osm_mcm_port_t*) );
> >+#endif
>
> I'd find this more readable if it were #ifdef rather than #ifndef.
>
> That said, I've got to believe that there's a better way to handle these
> changes. I just don't know what it is.
Hmm
How about
#ifdef __GCC_
#define cl_item_obj(item_ptr, obj_ptr, item_field) (typeof(obj_ptr)) \
((void *)item_ptr - (unsigned long)&((typeof(obj_ptr))0)->item_field)
#else
#define cl_item_obj(item_ptr, obj_ptr, item_field) (typeof(obj_ptr)) \
(void *)((uint8_t*)item_ptr - ((uintptr_t)(&(obj_ptr)->item_field) -
((uintptr_t)&(obj_ptr))))
#endif
And rely on the C implicit cast from void* to an object type, and rely
on compiles with gcc to do the extra type checking.
Jason
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
