Re: [patch v2 32/37] add rxe_net.h

On Sun, Jul 24, 2011 at 9:43 PM,  [off-list ref] wrote:

+/*
+ * this should be defined in .../include/linux/if_ether.h
+ */
+#define ETH_P_RXE			(0x8915)
+
+/*
+ * this should be defined in .../include/linux/netfilter.h
+ * to a specific value
+ */
+#define NFPROTO_RXE			(0)
+
+/*
+ * these should be defined in .../include/linux/netfilter_rxe.h
+ */
+#define NF_RXE_IN			(0)
+#define NF_RXE_OUT			(1)

Please update the cited header files instead of adding these symbols here.

+/* Should probably move to something other than an array...these can be big */
+#define RXE_MAX_IF_INDEX	(384)

Has it been considered to change net_info[] from an array into a
linked list ? As far as I can see none of the net_info[] accesses are
in the fast path and the number of entries used in the (sparse)
net_info[] array is small. Such a change would allow to eliminate the
RXE_MAX_IF_INDEX artificial limit.

+struct rxe_net_info {
+	struct rxe_dev		*rxe;
+	u8			port;
+	struct net_device	*ndev;
+	int			status;
+};

Is there a reason why status and port are in struct rxe_net_info
instead of struct rxe_dev ? Moving these two fields would allow
several code simplifications IMHO.

Bart.
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html