enable the full use of jumbos and remove IFCAP_VLAN_MTU.

The chip can do 9008 byte packets (not including the ethernet
header), but only 9004 if you want to enable IFCAP_VLAN_MTU. by not
enabling IFCAP_VLAN_MTU, we let other protocols (eg, mpls or svlan)
use the extra bytes if they want.

using the extra bytes for the hardmtu instead of for IFCAP_VLAN_MTU
works a bit better with how aggr(4) is set up at the moment because
aggr does not pass IFCAP_VLAN_MTU through from its ports, which
means vlan(4) on aggr(4) cannot see the flag and use the extra
bytes.

this was figured out by hrvoje popovski in a discussion with pedro
caetano on the "issues configuring vlan on top of aggr device" on
misc@.
hrvoje also tested the diff and made sure the full use of jumbos
works for things like ping packets with DF set.
jmatthew skimmed the diff and didnt see anything obviously wrong too

1 files changed, 11 insertions, 10 deletions

diff --git a/sys/dev/pci/if_bnx.c b/sys/dev/pci/if_bnx.c
index 3d7018d81ab..95a74a164ea 100644
--- a/sys/dev/pci/if_bnx.c
+++ b/sys/dev/pci/if_bnx.c
@@ -1,4 +1,4 @@
-/*	$OpenBSD: if_bnx.c,v 1.125 2018/03/10 10:51:46 sthen Exp $	*/
+/*	$OpenBSD: if_bnx.c,v 1.126 2019/12/06 01:58:47 dlg Exp $	*/
 
 /*-
  * Copyright (c) 2006 Broadcom Corporation
@@ -875,12 +875,13 @@ bnx_attachhook(struct device *self)
 	ifp->if_ioctl = bnx_ioctl;
 	ifp->if_qstart = bnx_start;
 	ifp->if_watchdog = bnx_watchdog;
+	ifp->if_hardmtu = BNX_MAX_JUMBO_ETHER_MTU_VLAN -
+	    sizeof(struct ether_header);
 	IFQ_SET_MAXLEN(&ifp->if_snd, USABLE_TX_BD - 1);
 	bcopy(sc->eaddr, sc->arpcom.ac_enaddr, ETHER_ADDR_LEN);
 	bcopy(sc->bnx_dev.dv_xname, ifp->if_xname, IFNAMSIZ);
 
-	ifp->if_capabilities = IFCAP_VLAN_MTU | IFCAP_CSUM_TCPv4 |
-	    IFCAP_CSUM_UDPv4;
+	ifp->if_capabilities = IFCAP_CSUM_TCPv4 | IFCAP_CSUM_UDPv4;
 
 #if NVLAN > 0
 	ifp->if_capabilities |= IFCAP_VLAN_HWTAGGING;
@@ -2417,7 +2418,7 @@ bnx_dma_alloc(struct bnx_softc *sc)
 	 */
 	for (i = 0; i < TOTAL_TX_BD; i++) {
 		if (bus_dmamap_create(sc->bnx_dmatag,
-		    MCLBYTES * BNX_MAX_SEGMENTS, BNX_MAX_SEGMENTS,
+		    BNX_MAX_JUMBO_ETHER_MTU_VLAN, BNX_MAX_SEGMENTS,
 		    MCLBYTES, 0, BUS_DMA_NOWAIT, &sc->tx_mbuf_map[i])) {
 			printf(": Could not create Tx mbuf %d DMA map!\n", 1);
 			rc = ENOMEM;
@@ -2650,8 +2651,8 @@ bnx_dma_alloc(struct bnx_softc *sc)
 	 * Create DMA maps for the Rx buffer mbufs.
 	 */
 	for (i = 0; i < TOTAL_RX_BD; i++) {
-		if (bus_dmamap_create(sc->bnx_dmatag, BNX_MAX_MRU,
-		    BNX_MAX_SEGMENTS, BNX_MAX_MRU, 0, BUS_DMA_NOWAIT,
+		if (bus_dmamap_create(sc->bnx_dmatag, BNX_MAX_JUMBO_MRU,
+		    1, BNX_MAX_JUMBO_MRU, 0, BUS_DMA_NOWAIT,
 		    &sc->rx_mbuf_map[i])) {
 			printf(": Could not create Rx mbuf %d DMA map!\n", i);
 			rc = ENOMEM;
@@ -3670,10 +3671,10 @@ bnx_get_buf(struct bnx_softc *sc, u_int16_t *prod,
 	    *prod_bseq);
 
 	/* This is a new mbuf allocation. */
-	m = MCLGETI(NULL, M_DONTWAIT, NULL, MCLBYTES);
+	m = MCLGETI(NULL, M_DONTWAIT, NULL, BNX_MAX_JUMBO_MRU);
 	if (!m)
 		return (0);
-	m->m_len = m->m_pkthdr.len = MCLBYTES;
+	m->m_len = m->m_pkthdr.len = BNX_MAX_JUMBO_MRU;
 	/* the chip aligns the ip header for us, no need to m_adj */
 
 	/* Map the mbuf cluster into device memory. */
@@ -4655,7 +4656,7 @@ bnx_init(void *xsc)
 	bnx_set_mac_addr(sc);
 
 	/* Calculate and program the Ethernet MRU size. */
-	ether_mtu = BNX_MAX_STD_ETHER_MTU_VLAN;
+	ether_mtu = BNX_MAX_JUMBO_ETHER_MTU_VLAN;
 
 	DBPRINT(sc, BNX_INFO, "%s(): setting MRU = %d\n",
 	    __FUNCTION__, ether_mtu);
@@ -5005,7 +5006,7 @@ bnx_ioctl(struct ifnet *ifp, u_long command, caddr_t data)
 
 	case SIOCGIFRXR:
 		error = if_rxr_ioctl((struct if_rxrinfo *)ifr->ifr_data,
-		    NULL, MCLBYTES, &sc->rx_ring);
+		    NULL, BNX_MAX_JUMBO_MRU, &sc->rx_ring);
 		break;
 
 	default: