MFC: r264630

For NFS mounts using rsize,wsize=65536 over TSO enabled
network interfaces limited to 32 transmit segments, there
are two known issues.
The more serious one is that for an I/O of slightly less than 64K,
the net device driver prepends an ethernet header, resulting in a
TSO segment slightly larger than 64K. Since m_defrag() copies this
into 33 mbuf clusters, the transmit fails with EFBIG.
A tester indicated observing a similar failure using iSCSI.

The second less critical problem is that the network
device driver must copy the mbuf chain via m_defrag()
(m_collapse() is not sufficient), resulting in measurable overhead.

This patch reduces the default size of if_hw_tsomax
slightly, so that the first issue is avoided.
Fixing the second issue will require a way for the
network device driver to inform tcp_output() that it
is limited to 32 transmit segments.

1 files changed, 3 insertions, 1 deletions

diff --git a/sys/net/if.c b/sys/net/if.c
index 0356ec7..ea2aa85 100644
--- a/sys/net/if.c
+++ b/sys/net/if.c
@@ -74,6 +74,7 @@
 #include <net/vnet.h>
 
 #if defined(INET) || defined(INET6)
+#include <net/ethernet.h>
 #include <netinet/in.h>
 #include <netinet/in_var.h>
 #include <netinet/ip.h>
@@ -658,7 +659,8 @@ if_attach_internal(struct ifnet *ifp, int vmove)
 #if defined(INET) || defined(INET6)
 		/* Initialize to max value. */
 		if (ifp->if_hw_tsomax == 0)
-			ifp->if_hw_tsomax = IP_MAXPACKET;
+			ifp->if_hw_tsomax = min(IP_MAXPACKET, 32 * MCLBYTES -
+			    (ETHER_HDR_LEN + ETHER_VLAN_ENCAP_LEN));
 		KASSERT(ifp->if_hw_tsomax <= IP_MAXPACKET &&
 		    ifp->if_hw_tsomax >= IP_MAXPACKET / 8,
 		    ("%s: tsomax outside of range", __func__));