summaryrefslogtreecommitdiffstats
path: root/dpdk/dpdk-2.1.0_patches
diff options
context:
space:
mode:
authorEd Warnicke <eaw@cisco.com>2015-12-08 15:45:58 -0700
committerEd Warnicke <eaw@cisco.com>2015-12-08 15:47:27 -0700
commitcb9cadad578297ffd78fa8a33670bdf1ab669e7e (patch)
tree6ac2be912482cc7849a26f0ab845561c3d7f4e26 /dpdk/dpdk-2.1.0_patches
parentfb0815d4ae4bb0fe27bd9313f34b45c8593b907e (diff)
Initial commit of vpp code.v1.0.0
Change-Id: Ib246f1fbfce93274020ee93ce461e3d8bd8b9f17 Signed-off-by: Ed Warnicke <eaw@cisco.com>
Diffstat (limited to 'dpdk/dpdk-2.1.0_patches')
-rw-r--r--dpdk/dpdk-2.1.0_patches/0001-kni-fix-igb-build-with-kernel-4.2.patch78
-rw-r--r--dpdk/dpdk-2.1.0_patches/0002-mbuf-rearrange-rte_mbuf-metadata-to-suit-vpp.patch107
-rw-r--r--dpdk/dpdk-2.1.0_patches/0003-e1000-Set-VLAN-Rx-Offload-tag-correctly.patch75
-rw-r--r--dpdk/dpdk-2.1.0_patches/0004-ixgbe-Wait-a-bit-longer-for-autonegotiation-to-leave.patch26
-rw-r--r--dpdk/dpdk-2.1.0_patches/0005-eal-Temporarily-turn-off-unthrottled-RTE_LOG.patch29
-rw-r--r--dpdk/dpdk-2.1.0_patches/0006-virtio-Cleanup-virtio-pmd-debug-log-output-reset-off.patch77
-rw-r--r--dpdk/dpdk-2.1.0_patches/0007-igb_uio-Reinstate-PCI-device-id-table-build-system-c.patch49
7 files changed, 441 insertions, 0 deletions
diff --git a/dpdk/dpdk-2.1.0_patches/0001-kni-fix-igb-build-with-kernel-4.2.patch b/dpdk/dpdk-2.1.0_patches/0001-kni-fix-igb-build-with-kernel-4.2.patch
new file mode 100644
index 00000000000..09bca06ed69
--- /dev/null
+++ b/dpdk/dpdk-2.1.0_patches/0001-kni-fix-igb-build-with-kernel-4.2.patch
@@ -0,0 +1,78 @@
+From 2de9d1629312a32f82c43167467640bc793805a6 Mon Sep 17 00:00:00 2001
+From: Damjan Marion <damarion@cisco.com>
+Date: Mon, 12 Oct 2015 14:23:30 +0200
+Subject: [PATCH 1/9] kni: fix igb build with kernel 4.2
+
+Kernel 4.2 has introduced two new parameters in ndo_bridge_getlink,
+which breaks DPDK compilation.
+
+Linux: 7d4f8d87 ("switchdev: ad VLAN support for ports bridge-getlink")
+
+This patch adds the necessary checks to fix it.
+
+Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
+---
+ lib/librte_eal/linuxapp/kni/ethtool/igb/igb_main.c | 13 +++++++++----
+ lib/librte_eal/linuxapp/kni/ethtool/igb/kcompat.h | 7 ++++++-
+ 2 files changed, 15 insertions(+), 5 deletions(-)
+
+diff --git a/lib/librte_eal/linuxapp/kni/ethtool/igb/igb_main.c b/lib/librte_eal/linuxapp/kni/ethtool/igb/igb_main.c
+index eed8df6..b330b20 100644
+--- a/lib/librte_eal/linuxapp/kni/ethtool/igb/igb_main.c
++++ b/lib/librte_eal/linuxapp/kni/ethtool/igb/igb_main.c
+@@ -2250,14 +2250,14 @@ static int igb_ndo_bridge_setlink(struct net_device *dev,
+ }
+
+ #ifdef HAVE_BRIDGE_FILTER
+-#ifdef HAVE_NDO_BRIDGE_GETLINK_FILTER_MASK
++#ifdef HAVE_NDO_BRIDGE_GETLINK_NLFLAGS
+ static int igb_ndo_bridge_getlink(struct sk_buff *skb, u32 pid, u32 seq,
+ struct net_device *dev, u32 filter_mask,
+ int nlflags)
+ #else
+ static int igb_ndo_bridge_getlink(struct sk_buff *skb, u32 pid, u32 seq,
+ struct net_device *dev, u32 filter_mask)
+-#endif /* HAVE_NDO_BRIDGE_GETLINK_FILTER_MASK */
++#endif /* HAVE_NDO_BRIDGE_GETLINK_NLFLAGS */
+ #else
+ static int igb_ndo_bridge_getlink(struct sk_buff *skb, u32 pid, u32 seq,
+ struct net_device *dev)
+@@ -2275,11 +2275,16 @@ static int igb_ndo_bridge_getlink(struct sk_buff *skb, u32 pid, u32 seq,
+ mode = BRIDGE_MODE_VEPA;
+
+ #ifdef HAVE_NDO_FDB_ADD_VID
+-#ifdef HAVE_NDO_BRIDGE_GETLINK_FILTER_MASK
++#ifdef HAVE_NDO_BRIDGE_GETLINK_NLFLAGS
++#ifdef HAVE_NDO_BRIDGE_GETLINK_FILTER_MASK_VLAN_FILL
++ return ndo_dflt_bridge_getlink(skb, pid, seq, dev, mode, 0, 0,
++ nlflags, filter_mask, NULL);
++#else
+ return ndo_dflt_bridge_getlink(skb, pid, seq, dev, mode, 0, 0, nlflags);
++#endif /* HAVE_NDO_BRIDGE_GETLINK_FILTER_MASK_VLAN_FILL */
+ #else
+ return ndo_dflt_bridge_getlink(skb, pid, seq, dev, mode, 0, 0);
+-#endif /* HAVE_NDO_BRIDGE_GETLINK_FILTER_MASK */
++#endif /* HAVE_NDO_BRIDGE_GETLINK_NLFLAGS */
+ #else
+ return ndo_dflt_bridge_getlink(skb, pid, seq, dev, mode);
+ #endif /* HAVE_NDO_FDB_ADD_VID */
+diff --git a/lib/librte_eal/linuxapp/kni/ethtool/igb/kcompat.h b/lib/librte_eal/linuxapp/kni/ethtool/igb/kcompat.h
+index 852f80f..5f45b8b 100644
+--- a/lib/librte_eal/linuxapp/kni/ethtool/igb/kcompat.h
++++ b/lib/librte_eal/linuxapp/kni/ethtool/igb/kcompat.h
+@@ -3899,6 +3899,11 @@ skb_set_hash(struct sk_buff *skb, __u32 hash, __always_unused int type)
+
+ #if ( LINUX_VERSION_CODE >= KERNEL_VERSION(4,1,0) )
+ /* ndo_bridge_getlink adds new nlflags parameter */
+-#define HAVE_NDO_BRIDGE_GETLINK_FILTER_MASK
++#define HAVE_NDO_BRIDGE_GETLINK_NLFLAGS
+ #endif /* >= 4.1.0 */
++
++#if ( LINUX_VERSION_CODE >= KERNEL_VERSION(4,2,0) )
++/* ndo_bridge_getlink adds new filter_mask and vlan_fill parameters */
++#define HAVE_NDO_BRIDGE_GETLINK_FILTER_MASK_VLAN_FILL
++#endif /* >= 4.2.0 */
+ #endif /* _KCOMPAT_H_ */
+--
+2.5.0
+
diff --git a/dpdk/dpdk-2.1.0_patches/0002-mbuf-rearrange-rte_mbuf-metadata-to-suit-vpp.patch b/dpdk/dpdk-2.1.0_patches/0002-mbuf-rearrange-rte_mbuf-metadata-to-suit-vpp.patch
new file mode 100644
index 00000000000..2ce5004a421
--- /dev/null
+++ b/dpdk/dpdk-2.1.0_patches/0002-mbuf-rearrange-rte_mbuf-metadata-to-suit-vpp.patch
@@ -0,0 +1,107 @@
+From 3609c4fb4d07d4285e96187598f54cb21e9e9b08 Mon Sep 17 00:00:00 2001
+From: Shesha Sreenivasamurthy <shesha@cisco.com>
+Date: Wed, 2 Sep 2015 08:57:24 -0700
+Subject: [PATCH 2/9] mbuf: rearrange rte_mbuf metadata to suit vpp
+
+Offload structure in the second cache line, next pointer in the
+first cache line. Issue reported to Intel.
+---
+ .../linuxapp/eal/include/exec-env/rte_kni_common.h | 10 +++++++--
+ lib/librte_mbuf/rte_mbuf.h | 25 ++++++++++++++--------
+ 2 files changed, 24 insertions(+), 11 deletions(-)
+
+diff --git a/lib/librte_eal/linuxapp/eal/include/exec-env/rte_kni_common.h b/lib/librte_eal/linuxapp/eal/include/exec-env/rte_kni_common.h
+index e9f38bd..d327f71 100644
+--- a/lib/librte_eal/linuxapp/eal/include/exec-env/rte_kni_common.h
++++ b/lib/librte_eal/linuxapp/eal/include/exec-env/rte_kni_common.h
+@@ -111,6 +111,10 @@ struct rte_kni_fifo {
+ * The kernel image of the rte_mbuf struct, with only the relevant fields.
+ * Padding is necessary to assure the offsets of these fields
+ */
++/*
++ * offload in the second cache line, next in the first. Better for vpp
++ * at least as of right now.
++ */
+ struct rte_kni_mbuf {
+ void *buf_addr __attribute__((__aligned__(RTE_CACHE_LINE_SIZE)));
+ char pad0[10];
+@@ -121,16 +125,18 @@ struct rte_kni_mbuf {
+ char pad2[4];
+ uint32_t pkt_len; /**< Total pkt len: sum of all segment data_len. */
+ uint16_t data_len; /**< Amount of data in segment buffer. */
++ char pad3[2];
+ #else
+ char pad2[2];
+ uint16_t data_len; /**< Amount of data in segment buffer. */
+ uint32_t pkt_len; /**< Total pkt len: sum of all segment data_len. */
++ char pad3[4];
+ #endif
++ void *next;
+
+ /* fields on second cache line */
+- char pad3[8] __attribute__((__aligned__(RTE_CACHE_LINE_SIZE)));
++ char pad4[12] __attribute__((__aligned__(RTE_CACHE_LINE_SIZE)));
+ void *pool;
+- void *next;
+ };
+
+ /*
+diff --git a/lib/librte_mbuf/rte_mbuf.h b/lib/librte_mbuf/rte_mbuf.h
+index 8c2db1b..61cbbd7 100644
+--- a/lib/librte_mbuf/rte_mbuf.h
++++ b/lib/librte_mbuf/rte_mbuf.h
+@@ -743,6 +743,12 @@ typedef uint64_t MARKER64[0]; /**< marker that allows us to overwrite 8 bytes
+ /**
+ * The generic rte_mbuf, containing a packet mbuf.
+ */
++/*
++ * offload in the second cache line, next in the first. Better for vpp
++ * at least as of right now.
++ * If you change this structure, you must change the user-mode
++ * version in rte_mbuf.h
++ */
+ struct rte_mbuf {
+ MARKER cacheline0;
+
+@@ -809,6 +815,16 @@ struct rte_mbuf {
+ uint16_t vlan_tci; /**< VLAN Tag Control Identifier (CPU order) */
+ uint16_t vlan_tci_outer; /**< Outer VLAN Tag Control Identifier (CPU order) */
+ #endif /* RTE_NEXT_ABI */
++ struct rte_mbuf *next; /**< Next segment of scattered packet. */
++
++ uint32_t seqn; /**< Sequence number. See also rte_reorder_insert() */
++#ifdef RTE_NEXT_ABI
++ uint16_t vlan_tci_outer; /**< Outer VLAN Tag Control Identifier (CPU order) */
++#endif /* RTE_NEXT_ABI */
++
++ /* second cache line - fields only used in slow path or on TX */
++ MARKER cacheline1 __rte_cache_aligned;
++
+ union {
+ uint32_t rss; /**< RSS hash result if RSS enabled */
+ struct {
+@@ -828,21 +844,12 @@ struct rte_mbuf {
+ uint32_t usr; /**< User defined tags. See rte_distributor_process() */
+ } hash; /**< hash information */
+
+- uint32_t seqn; /**< Sequence number. See also rte_reorder_insert() */
+-#ifdef RTE_NEXT_ABI
+- uint16_t vlan_tci_outer; /**< Outer VLAN Tag Control Identifier (CPU order) */
+-#endif /* RTE_NEXT_ABI */
+-
+- /* second cache line - fields only used in slow path or on TX */
+- MARKER cacheline1 __rte_cache_aligned;
+-
+ union {
+ void *userdata; /**< Can be used for external metadata */
+ uint64_t udata64; /**< Allow 8-byte userdata on 32-bit */
+ };
+
+ struct rte_mempool *pool; /**< Pool from which mbuf was allocated. */
+- struct rte_mbuf *next; /**< Next segment of scattered packet. */
+
+ /* fields to support TX offloads */
+ union {
+--
+2.5.0
+
diff --git a/dpdk/dpdk-2.1.0_patches/0003-e1000-Set-VLAN-Rx-Offload-tag-correctly.patch b/dpdk/dpdk-2.1.0_patches/0003-e1000-Set-VLAN-Rx-Offload-tag-correctly.patch
new file mode 100644
index 00000000000..d1ea27ad778
--- /dev/null
+++ b/dpdk/dpdk-2.1.0_patches/0003-e1000-Set-VLAN-Rx-Offload-tag-correctly.patch
@@ -0,0 +1,75 @@
+From 699252f0b685db4cd298e90f0e1d64e4792356f2 Mon Sep 17 00:00:00 2001
+From: Damjan Marion <damarion@cisco.com>
+Date: Wed, 21 Oct 2015 14:46:12 +0200
+Subject: [PATCH 3/9] e1000: Set VLAN Rx Offload tag correctly
+
+---
+ drivers/net/e1000/igb_rxtx.c | 30 ++++++++++++++++++++++++++++++
+ lib/librte_ether/rte_ether.h | 3 +++
+ 2 files changed, 33 insertions(+)
+
+diff --git a/drivers/net/e1000/igb_rxtx.c b/drivers/net/e1000/igb_rxtx.c
+index b13930e..7fe76c8 100644
+--- a/drivers/net/e1000/igb_rxtx.c
++++ b/drivers/net/e1000/igb_rxtx.c
+@@ -885,6 +885,21 @@ eth_igb_recv_pkts(void *rx_queue, struct rte_mbuf **rx_pkts,
+ pkt_flags = rx_desc_hlen_type_rss_to_pkt_flags(hlen_type_rss);
+ pkt_flags = pkt_flags | rx_desc_status_to_pkt_flags(staterr);
+ pkt_flags = pkt_flags | rx_desc_error_to_pkt_flags(staterr);
++ {
++ /*
++ * Check packet for VLAN ethernet types and set
++ * RX Offload flag PKT_RX_VLAN_PKT accordingly.
++ */
++ struct ether_hdr *eth_hdr =
++ rte_pktmbuf_mtod(rxm, struct ether_hdr *);
++ u16 eth_type = rte_be_to_cpu_16(eth_hdr->ether_type);
++
++ if ((eth_type == ETHER_TYPE_VLAN) ||
++ (eth_type == ETHER_TYPE_VLAN_AD) ||
++ (eth_type == ETHER_TYPE_VLAN_9100) ||
++ (eth_type == ETHER_TYPE_VLAN_9200))
++ pkt_flags |= PKT_RX_VLAN_PKT;
++ }
+ rxm->ol_flags = pkt_flags;
+ #ifdef RTE_NEXT_ABI
+ rxm->packet_type = igb_rxd_pkt_info_to_pkt_type(rxd.wb.lower.
+@@ -1123,6 +1138,21 @@ eth_igb_recv_scattered_pkts(void *rx_queue, struct rte_mbuf **rx_pkts,
+ pkt_flags = rx_desc_hlen_type_rss_to_pkt_flags(hlen_type_rss);
+ pkt_flags = pkt_flags | rx_desc_status_to_pkt_flags(staterr);
+ pkt_flags = pkt_flags | rx_desc_error_to_pkt_flags(staterr);
++ {
++ /*
++ * Check packet for VLAN ethernet types and set
++ * RX Offload flag PKT_RX_VLAN_PKT accordingly.
++ */
++ struct ether_hdr *eth_hdr =
++ rte_pktmbuf_mtod(rxm, struct ether_hdr *);
++ u16 eth_type = rte_be_to_cpu_16(eth_hdr->ether_type);
++
++ if ((eth_type == ETHER_TYPE_VLAN) ||
++ (eth_type == ETHER_TYPE_VLAN_AD) ||
++ (eth_type == ETHER_TYPE_VLAN_9100) ||
++ (eth_type == ETHER_TYPE_VLAN_9200))
++ pkt_flags |= PKT_RX_VLAN_PKT;
++ }
+ first_seg->ol_flags = pkt_flags;
+ #ifdef RTE_NEXT_ABI
+ first_seg->packet_type = igb_rxd_pkt_info_to_pkt_type(rxd.wb.
+diff --git a/lib/librte_ether/rte_ether.h b/lib/librte_ether/rte_ether.h
+index 07c17d7..fd646ec 100644
+--- a/lib/librte_ether/rte_ether.h
++++ b/lib/librte_ether/rte_ether.h
+@@ -332,6 +332,9 @@ struct vxlan_hdr {
+ #define ETHER_TYPE_1588 0x88F7 /**< IEEE 802.1AS 1588 Precise Time Protocol. */
+ #define ETHER_TYPE_SLOW 0x8809 /**< Slow protocols (LACP and Marker). */
+ #define ETHER_TYPE_TEB 0x6558 /**< Transparent Ethernet Bridging. */
++#define ETHER_TYPE_VLAN_AD 0x88a8 /**< IEEE 802.1AD VLAN tagging. */
++#define ETHER_TYPE_VLAN_9100 0x9100 /**< VLAN 0x9100 tagging. */
++#define ETHER_TYPE_VLAN_9200 0x9200 /**< VLAN 0x9200 tagging. */
+
+ #define ETHER_VXLAN_HLEN (sizeof(struct udp_hdr) + sizeof(struct vxlan_hdr))
+ /**< VXLAN tunnel header length. */
+--
+2.5.0
+
diff --git a/dpdk/dpdk-2.1.0_patches/0004-ixgbe-Wait-a-bit-longer-for-autonegotiation-to-leave.patch b/dpdk/dpdk-2.1.0_patches/0004-ixgbe-Wait-a-bit-longer-for-autonegotiation-to-leave.patch
new file mode 100644
index 00000000000..75241173afd
--- /dev/null
+++ b/dpdk/dpdk-2.1.0_patches/0004-ixgbe-Wait-a-bit-longer-for-autonegotiation-to-leave.patch
@@ -0,0 +1,26 @@
+From 67d1c25af7fa16df40a8305405066ba6a40ac659 Mon Sep 17 00:00:00 2001
+From: Shesha Sreenivasamurthy <shesha@cisco.com>
+Date: Wed, 2 Sep 2015 08:46:39 -0700
+Subject: [PATCH 4/9] ixgbe: Wait a bit longer for autonegotiation to leave
+ state 0
+
+---
+ drivers/net/ixgbe/base/ixgbe_82599.c | 2 +-
+ 1 file changed, 1 insertion(+), 1 deletion(-)
+
+diff --git a/drivers/net/ixgbe/base/ixgbe_82599.c b/drivers/net/ixgbe/base/ixgbe_82599.c
+index f0deb59..ae66380 100644
+--- a/drivers/net/ixgbe/base/ixgbe_82599.c
++++ b/drivers/net/ixgbe/base/ixgbe_82599.c
+@@ -2442,7 +2442,7 @@ s32 ixgbe_reset_pipeline_82599(struct ixgbe_hw *hw)
+ IXGBE_WRITE_REG(hw, IXGBE_AUTOC,
+ autoc_reg ^ (0x4 << IXGBE_AUTOC_LMS_SHIFT));
+ /* Wait for AN to leave state 0 */
+- for (i = 0; i < 10; i++) {
++ for (i = 0; i < 50; i++) {
+ msec_delay(4);
+ anlp1_reg = IXGBE_READ_REG(hw, IXGBE_ANLP1);
+ if (anlp1_reg & IXGBE_ANLP1_AN_STATE_MASK)
+--
+2.5.0
+
diff --git a/dpdk/dpdk-2.1.0_patches/0005-eal-Temporarily-turn-off-unthrottled-RTE_LOG.patch b/dpdk/dpdk-2.1.0_patches/0005-eal-Temporarily-turn-off-unthrottled-RTE_LOG.patch
new file mode 100644
index 00000000000..245b43c196e
--- /dev/null
+++ b/dpdk/dpdk-2.1.0_patches/0005-eal-Temporarily-turn-off-unthrottled-RTE_LOG.patch
@@ -0,0 +1,29 @@
+From 9e28214eb784b9f68af6e0503f8cefe861f13440 Mon Sep 17 00:00:00 2001
+From: Shesha Sreenivasamurthy <shesha@cisco.com>
+Date: Wed, 2 Sep 2015 08:55:43 -0700
+Subject: [PATCH 5/9] eal: Temporarily turn off unthrottled RTE_LOG(...)
+
+Otherwise, /var/log/syslog eventually fills the disk. The error
+condition seems only to affect ESXi VM's. It'd be worth suggesting log
+throttling to the DPDK community. Much better to avoid making syslog
+(...) calls in the first place.
+---
+ lib/librte_eal/linuxapp/eal/eal_interrupts.c | 2 +-
+ 1 file changed, 1 insertion(+), 1 deletion(-)
+
+diff --git a/lib/librte_eal/linuxapp/eal/eal_interrupts.c b/lib/librte_eal/linuxapp/eal/eal_interrupts.c
+index 3f87875..29a3539 100644
+--- a/lib/librte_eal/linuxapp/eal/eal_interrupts.c
++++ b/lib/librte_eal/linuxapp/eal/eal_interrupts.c
+@@ -709,7 +709,7 @@ eal_intr_process_interrupts(struct epoll_event *events, int nfds)
+ * for epoll_wait.
+ */
+ bytes_read = read(events[n].data.fd, &buf, bytes_read);
+- if (bytes_read < 0) {
++ if (0 && bytes_read < 0) {
+ if (errno == EINTR || errno == EWOULDBLOCK)
+ continue;
+
+--
+2.5.0
+
diff --git a/dpdk/dpdk-2.1.0_patches/0006-virtio-Cleanup-virtio-pmd-debug-log-output-reset-off.patch b/dpdk/dpdk-2.1.0_patches/0006-virtio-Cleanup-virtio-pmd-debug-log-output-reset-off.patch
new file mode 100644
index 00000000000..9306f112a2c
--- /dev/null
+++ b/dpdk/dpdk-2.1.0_patches/0006-virtio-Cleanup-virtio-pmd-debug-log-output-reset-off.patch
@@ -0,0 +1,77 @@
+From 21a9bf50270f71ebda5acb5fc233b8279cec56a7 Mon Sep 17 00:00:00 2001
+From: Shesha Sreenivasamurthy <shesha@cisco.com>
+Date: Wed, 2 Sep 2015 08:48:09 -0700
+Subject: [PATCH 6/9] virtio: Cleanup virtio pmd debug log output, reset
+ offload field
+
+---
+ drivers/net/virtio/virtio_ethdev.c | 10 +++++-----
+ drivers/net/virtio/virtio_rxtx.c | 4 +++-
+ 2 files changed, 8 insertions(+), 6 deletions(-)
+
+diff --git a/drivers/net/virtio/virtio_ethdev.c b/drivers/net/virtio/virtio_ethdev.c
+index 465d3cd..6a686fe 100644
+--- a/drivers/net/virtio/virtio_ethdev.c
++++ b/drivers/net/virtio/virtio_ethdev.c
+@@ -1521,24 +1521,24 @@ virtio_dev_link_update(struct rte_eth_dev *dev, __rte_unused int wait_to_complet
+ link.link_speed = SPEED_10G;
+
+ if (vtpci_with_feature(hw, VIRTIO_NET_F_STATUS)) {
+- PMD_INIT_LOG(DEBUG, "Get link status from hw");
+ vtpci_read_dev_config(hw,
+ offsetof(struct virtio_net_config, status),
+ &status, sizeof(status));
+ if ((status & VIRTIO_NET_S_LINK_UP) == 0) {
+ link.link_status = 0;
+- PMD_INIT_LOG(DEBUG, "Port %d is down",
+- dev->data->port_id);
+ } else {
+ link.link_status = 1;
+- PMD_INIT_LOG(DEBUG, "Port %d is up",
+- dev->data->port_id);
+ }
+ } else {
+ link.link_status = 1; /* Link up */
+ }
+ virtio_dev_atomic_write_link_status(dev, &link);
+
++ /* This message is far too noisy for normal use */
++ if (0)
++ PMD_INIT_LOG(DEBUG, "Port %d is %s\n", dev->data->port_id,
++ link.link_status ? "up" : "down");
++
+ return (old.link_status == link.link_status) ? -1 : 0;
+ }
+
+diff --git a/drivers/net/virtio/virtio_rxtx.c b/drivers/net/virtio/virtio_rxtx.c
+index c5b53bb..9f0b759 100644
+--- a/drivers/net/virtio/virtio_rxtx.c
++++ b/drivers/net/virtio/virtio_rxtx.c
+@@ -536,6 +536,7 @@ virtio_recv_pkts(void *rx_queue, struct rte_mbuf **rx_pkts, uint16_t nb_pkts)
+ rxm->next = NULL;
+ rxm->pkt_len = (uint32_t)(len[i] - hdr_size);
+ rxm->data_len = (uint16_t)(len[i] - hdr_size);
++ rxm->ol_flags = 0;
+
+ if (hw->vlan_strip)
+ rte_vlan_strip(rxm);
+@@ -651,6 +652,7 @@ virtio_recv_mergeable_pkts(void *rx_queue,
+ rxm->next = NULL;
+ rxm->pkt_len = (uint32_t)(len[0] - hdr_size);
+ rxm->data_len = (uint16_t)(len[0] - hdr_size);
++ rxm->ol_flags = 0;
+
+ rxm->port = rxvq->port_id;
+ rx_pkts[nb_rx] = rxm;
+@@ -752,7 +754,7 @@ virtio_xmit_pkts(void *tx_queue, struct rte_mbuf **tx_pkts, uint16_t nb_pkts)
+ if (unlikely(nb_pkts < 1))
+ return nb_pkts;
+
+- PMD_TX_LOG(DEBUG, "%d packets to xmit", nb_pkts);
++ PMD_TX_LOG(DEBUG, "%d packets to xmit\n", nb_pkts);
+ nb_used = VIRTQUEUE_NUSED(txvq);
+
+ virtio_rmb();
+--
+2.5.0
+
diff --git a/dpdk/dpdk-2.1.0_patches/0007-igb_uio-Reinstate-PCI-device-id-table-build-system-c.patch b/dpdk/dpdk-2.1.0_patches/0007-igb_uio-Reinstate-PCI-device-id-table-build-system-c.patch
new file mode 100644
index 00000000000..b92fca4cb5d
--- /dev/null
+++ b/dpdk/dpdk-2.1.0_patches/0007-igb_uio-Reinstate-PCI-device-id-table-build-system-c.patch
@@ -0,0 +1,49 @@
+From 3ac6cb020c0f2ddb42dcd790cf64118453e966a2 Mon Sep 17 00:00:00 2001
+From: Shesha Sreenivasamurthy <shesha@cisco.com>
+Date: Wed, 2 Sep 2015 08:48:41 -0700
+Subject: [PATCH 7/9] igb_uio: Reinstate PCI device id table, build system
+ changes
+
+---
+ lib/librte_eal/linuxapp/igb_uio/igb_uio.c | 18 +++++++++++++++++-
+ 3 files changed, 34 insertions(+), 1 deletion(-)
+
+diff --git a/lib/librte_eal/linuxapp/igb_uio/igb_uio.c b/lib/librte_eal/linuxapp/igb_uio/igb_uio.c
+index 865a276..16d2a55 100644
+--- a/lib/librte_eal/linuxapp/igb_uio/igb_uio.c
++++ b/lib/librte_eal/linuxapp/igb_uio/igb_uio.c
+@@ -61,6 +61,22 @@ struct rte_uio_pci_dev {
+ static char *intr_mode = NULL;
+ static enum rte_intr_mode igbuio_intr_mode_preferred = RTE_INTR_MODE_MSIX;
+
++/* PCI device id table */
++static struct pci_device_id igbuio_pci_ids[] = {
++#define RTE_PCI_DEV_ID_DECL_EM(vend, dev) {PCI_DEVICE(vend, dev)},
++#define RTE_PCI_DEV_ID_DECL_IGB(vend, dev) {PCI_DEVICE(vend, dev)},
++#define RTE_PCI_DEV_ID_DECL_IGBVF(vend, dev) {PCI_DEVICE(vend, dev)},
++#define RTE_PCI_DEV_ID_DECL_IXGBE(vend, dev) {PCI_DEVICE(vend, dev)},
++#define RTE_PCI_DEV_ID_DECL_IXGBEVF(vend, dev) {PCI_DEVICE(vend, dev)},
++#define RTE_PCI_DEV_ID_DECL_VIRTIO(vend, dev) {PCI_DEVICE(vend, dev)},
++#define RTE_PCI_DEV_ID_DECL_VMXNET3(vend, dev) {PCI_DEVICE(vend, dev)},
++#define RTE_PCI_DEV_ID_DECL_VICE(vend, dev) {PCI_DEVICE(vend, dev)},
++#include <rte_pci_dev_ids.h>
++{ 0, },
++};
++
++MODULE_DEVICE_TABLE(pci, igbuio_pci_ids);
++
+ static inline struct rte_uio_pci_dev *
+ igbuio_get_uio_pci_dev(struct uio_info *info)
+ {
+@@ -606,7 +622,7 @@ igbuio_config_intr_mode(char *intr_str)
+
+ static struct pci_driver igbuio_pci_driver = {
+ .name = "igb_uio",
+- .id_table = NULL,
++ .id_table = igbuio_pci_ids,
+ .probe = igbuio_pci_probe,
+ .remove = igbuio_pci_remove,
+ };
+--
+2.5.0
+
/* Comment.Single */ .highlight .cs { color: #cc0000; font-weight: bold; background-color: #fff0f0 } /* Comment.Special */ .highlight .gd { color: #000000; background-color: #ffdddd } /* Generic.Deleted */ .highlight .ge { font-style: italic } /* Generic.Emph */ .highlight .gr { color: #aa0000 } /* Generic.Error */ .highlight .gh { color: #333333 } /* Generic.Heading */ .highlight .gi { color: #000000; background-color: #ddffdd } /* Generic.Inserted */ .highlight .go { color: #888888 } /* Generic.Output */ .highlight .gp { color: #555555 } /* Generic.Prompt */ .highlight .gs { font-weight: bold } /* Generic.Strong */ .highlight .gu { color: #666666 } /* Generic.Subheading */ .highlight .gt { color: #aa0000 } /* Generic.Traceback */ .highlight .kc { color: #008800; font-weight: bold } /* Keyword.Constant */ .highlight .kd { color: #008800; font-weight: bold } /* Keyword.Declaration */ .highlight .kn { color: #008800; font-weight: bold } /* Keyword.Namespace */ .highlight .kp { color: #008800 } /* Keyword.Pseudo */ .highlight .kr { color: #008800; font-weight: bold } /* Keyword.Reserved */ .highlight .kt { color: #888888; font-weight: bold } /* Keyword.Type */ .highlight .m { color: #0000DD; font-weight: bold } /* Literal.Number */ .highlight .s { color: #dd2200; background-color: #fff0f0 } /* Literal.String */ .highlight .na { color: #336699 } /* Name.Attribute */ .highlight .nb { color: #003388 } /* Name.Builtin */ .highlight .nc { color: #bb0066; font-weight: bold } /* Name.Class */ .highlight .no { color: #003366; font-weight: bold } /* Name.Constant */ .highlight .nd { color: #555555 } /* Name.Decorator */ .highlight .ne { color: #bb0066; font-weight: bold } /* Name.Exception */ .highlight .nf { color: #0066bb; font-weight: bold } /* Name.Function */ .highlight .nl { color: #336699; font-style: italic } /* Name.Label */ .highlight .nn { color: #bb0066; font-weight: bold } /* Name.Namespace */ .highlight .py { color: #336699; font-weight: bold } /* Name.Property */ .highlight .nt { color: #bb0066; font-weight: bold } /* Name.Tag */ .highlight .nv { color: #336699 } /* Name.Variable */ .highlight .ow { color: #008800 } /* Operator.Word */ .highlight .w { color: #bbbbbb } /* Text.Whitespace */ .highlight .mb { color: #0000DD; font-weight: bold } /* Literal.Number.Bin */ .highlight .mf { color: #0000DD; font-weight: bold } /* Literal.Number.Float */ .highlight .mh { color: #0000DD; font-weight: bold } /* Literal.Number.Hex */ .highlight .mi { color: #0000DD; font-weight: bold } /* Literal.Number.Integer */ .highlight .mo { color: #0000DD; font-weight: bold } /* Literal.Number.Oct */ .highlight .sa { color: #dd2200; background-color: #fff0f0 } /* Literal.String.Affix */ .highlight .sb { color: #dd2200; background-color: #fff0f0 } /* Literal.String.Backtick */ .highlight .sc { color: #dd2200; background-color: #fff0f0 } /* Literal.String.Char */ .highlight .dl { color: #dd2200; background-color: #fff0f0 } /* Literal.String.Delimiter */ .highlight .sd { color: #dd2200; background-color: #fff0f0 } /* Literal.String.Doc */ .highlight .s2 { color: #dd2200; background-color: #fff0f0 } /* Literal.String.Double */ .highlight .se { color: #0044dd; background-color: #fff0f0 } /* Literal.String.Escape */ .highlight .sh { color: #dd2200; background-color: #fff0f0 } /* Literal.String.Heredoc */ .highlight .si { color: #3333bb; background-color: #fff0f0 } /* Literal.String.Interpol */ .highlight .sx { color: #22bb22; background-color: #f0fff0 } /* Literal.String.Other */ .highlight .sr { color: #008800; background-color: #fff0ff } /* Literal.String.Regex */ .highlight .s1 { color: #dd2200; background-color: #fff0f0 } /* Literal.String.Single */ .highlight .ss { color: #aa6600; background-color: #fff0f0 } /* Literal.String.Symbol */ .highlight .bp { color: #003388 } /* Name.Builtin.Pseudo */ .highlight .fm { color: #0066bb; font-weight: bold } /* Name.Function.Magic */ .highlight .vc { color: #336699 } /* Name.Variable.Class */ .highlight .vg { color: #dd7700 } /* Name.Variable.Global */ .highlight .vi { color: #3333bb } /* Name.Variable.Instance */ .highlight .vm { color: #336699 } /* Name.Variable.Magic */ .highlight .il { color: #0000DD; font-weight: bold } /* Literal.Number.Integer.Long */ } 
/*
 * Copyright (c) 2015 Cisco and/or its affiliates.
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at:
 *
 *     http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */
/*
  Copyright (c) 2001, 2002, 2003 Eliot Dresselhaus

  Permission is hereby granted, free of charge, to any person obtaining
  a copy of this software and associated documentation files (the
  "Software"), to deal in the Software without restriction, including
  without limitation the rights to use, copy, modify, merge, publish,
  distribute, sublicense, and/or sell copies of the Software, and to
  permit persons to whom the Software is furnished to do so, subject to
  the following conditions:

  The above copyright notice and this permission notice shall be
  included in all copies or substantial portions of the Software.

  THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
  EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
  MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
  NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
  LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
  OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
  WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
*/

#ifdef __KERNEL__

#if __linux__
# include <linux/unistd.h>
# include <linux/signal.h>
#endif

#else /* ! __KERNEL__ */

#ifdef __APPLE__
#define _XOPEN_SOURCE
#endif

#define _GNU_SOURCE		/* to get REG_* in ucontext.h */
#include <ucontext.h>
#undef _GNU_SOURCE
#undef __USE_GNU

#include <unistd.h>
#include <signal.h>
#include <grp.h>

#include <time.h>
#include <sys/socket.h>
#include <netdb.h>
#include <math.h>

#include <vppinfra/time.h>
#if __linux__
#include <vppinfra/linux/syscall.h>

#ifdef AF_NETLINK
#include <linux/types.h>
#include <linux/netlink.h>
#endif
#endif

#endif /* ! __KERNEL__ */


#ifdef __KERNEL__
# include <linux/socket.h>
# include <linux/in.h>
# include <linux/ip.h>
# include <linux/tcp.h>
# include <linux/udp.h>
# include <linux/icmp.h>
# include <linux/if_ether.h>
# include <linux/if_arp.h>
#else
# include <net/if.h>            /* struct ifnet may live here */
# include <netinet/in.h>
# include <netinet/ip.h>
# include <netinet/tcp.h>
# include <netinet/udp.h>
# include <netinet/ip_icmp.h>
# include <netinet/if_ether.h>
#endif /* __KERNEL__ */

#include <vppinfra/bitops.h> /* foreach_set_bit */
#include <vppinfra/format.h>
#include <vppinfra/error.h>

/* Format unix network address family (e.g. AF_INET). */
u8 * format_address_family (u8 * s, va_list * va)
{
  uword family = va_arg (*va, uword);
  u8 * t = (u8 *) "UNKNOWN";
  switch (family)
    {
#define _(x) case PF_##x: t = (u8 *) #x; break
      _ (UNSPEC);
      _ (UNIX);			/* Unix domain sockets 		*/
      _ (INET);			/* Internet IP Protocol 	*/
#ifdef PF_AX25
      _ (AX25);			/* Amateur Radio AX.25 		*/
#endif
#ifdef PF_IPX
      _ (IPX);			/* Novell IPX 			*/
#endif
#ifdef PF_APPLETALK
      _ (APPLETALK);		/* AppleTalk DDP 		*/
#endif
#ifdef PF_NETROM
      _ (NETROM);		/* Amateur Radio NET/ROM 	*/
#endif
#ifdef PF_BRIDGE
      _ (BRIDGE);		/* Multiprotocol bridge 	*/
#endif
#ifdef PF_ATMPVC
      _ (ATMPVC);		/* ATM PVCs			*/
#endif
#ifdef PF_X25
      _ (X25);			/* Reserved for X.25 project 	*/
#endif
#ifdef PF_INET6
      _ (INET6);		/* IP version 6			*/
#endif
#ifdef PF_ROSE
      _ (ROSE);			/* Amateur Radio X.25 PLP	*/
#endif
#ifdef PF_DECnet
      _ (DECnet);		/* Reserved for DECnet project	*/
#endif
#ifdef PF_NETBEUI
      _ (NETBEUI);		/* Reserved for 802.2LLC project*/
#endif
#ifdef PF_SECURITY
      _ (SECURITY);		/* Security callback pseudo AF */
#endif
#ifdef PF_KEY
      _ (KEY);			/* PF_KEY key management API */
#endif
#ifdef PF_NETLINK
      _ (NETLINK);
#endif
#ifdef PF_PACKET
      _ (PACKET);		/* Packet family		*/
#endif
#ifdef PF_ASH
      _ (ASH);			/* Ash				*/
#endif
#ifdef PF_ECONET
      _ (ECONET);		/* Acorn Econet			*/
#endif
#ifdef PF_ATMSVC
      _ (ATMSVC);		/* ATM SVCs			*/
#endif
#ifdef PF_SNA
      _ (SNA);			/* Linux SNA Project */
#endif
#ifdef PF_IRDA
      _ (IRDA);			/* IRDA sockets			*/
#endif
#undef _
    }
  vec_add (s, t, strlen ((char *) t));
  return s;
}

u8 * format_network_protocol (u8 * s, va_list * args)
{
  uword family = va_arg (*args, uword);
  uword protocol = va_arg (*args, uword);

#ifndef __KERNEL__
  struct protoent * p = getprotobynumber (protocol);

  ASSERT (family == AF_INET);
  if (p)
    return format (s, "%s", p->p_name);
  else
    return format (s, "%d", protocol);
#else
  return format (s, "%d/%d", family, protocol);
#endif
}

u8 * format_network_port (u8 * s, va_list * args)
{
  uword proto = va_arg (*args, uword);
  uword port = va_arg (*args, uword);

#ifndef __KERNEL__
  struct servent * p = getservbyport (port, proto == IPPROTO_UDP ? "udp" : "tcp");

  if (p)
    return format (s, "%s", p->s_name);
  else
    return format (s, "%d", port);
#else
  return format (s, "%s/%d", proto == IPPROTO_UDP ? "udp" : "tcp", port);
#endif
}

/* Format generic network address: takes two arguments family and address.
   Assumes network byte order. */
u8 * format_network_address (u8 * s, va_list * args)
{
  uword family = va_arg (*args, uword);
  u8 * addr    = va_arg (*args, u8 *);

  switch (family)
    {
    case AF_INET:
      s = format (s, "%d.%d.%d.%d", addr[0], addr[1], addr[2], addr[3]);
      break;

    case AF_UNSPEC:
      /* We use AF_UNSPEC for ethernet addresses. */
      s = format (s, "%02x:%02x:%02x:%02x:%02x:%02x",
		  addr[0], addr[1], addr[2], addr[3], addr[4], addr[5]);
      break;

    default:
      clib_error ("unsupported address family %d", family);
    }

  return s;
}

u8 * format_sockaddr (u8 * s, va_list * args)
{
  void * v = va_arg (*args, void *);
  struct sockaddr * sa = v;
  static u32 local_counter;

  switch (sa->sa_family)
    {
    case AF_INET:
      {
	struct sockaddr_in * i = v;
	s = format (s, "%U:%U",
		    format_network_address, AF_INET, &i->sin_addr.s_addr,
		    format_network_port, IPPROTO_TCP, ntohs (i->sin_port));
      }
      break;

    case AF_LOCAL:
      {
        /* 
         * There isn't anything useful to print.
         * The unix cli world uses the output to make a node name,
         * so we need to return a unique name. 
         */
        s = format (s, "local:%u", local_counter++);
      }
      break;

#ifndef __KERNEL__
#ifdef AF_NETLINK
    case AF_NETLINK:
      {
	struct sockaddr_nl * n = v;
	s = format (s, "KERNEL-NETLINK");
	if (n->nl_groups)
	  s = format (s, " (groups 0x%x)", n->nl_groups);
	break;
      }
#endif
#endif

    default:
      s = format (s, "sockaddr family %d", sa->sa_family);
      break;
    }

  return s;
}

#ifndef __APPLE__
u8 * format_tcp4_packet (u8 * s, va_list * args)
{
  u8 * p = va_arg (*args, u8 *);
  struct iphdr * ip = (void *) p;
  struct tcphdr * tcp = (void *) (ip + 1);

  s = format (s, "tcp %U:%U -> %U:%U",
	      format_network_address, AF_INET,  &ip->saddr,
	      format_network_port, IPPROTO_TCP, ntohs (tcp->source),
	      format_network_address, AF_INET,  &ip->daddr,
	      format_network_port, IPPROTO_TCP, ntohs (tcp->dest));

  s = format (s, ", seq 0x%08x -> 0x%08x", tcp->seq, tcp->ack_seq);
#define _(f) if (tcp->f) s = format (s, ", " #f);
  _ (syn); _ (ack); _ (fin); _ (rst); _ (psh); _ (urg);
#undef _

  if (tcp->window)
    s = format (s, ", window 0x%04x", tcp->window);
  if (tcp->urg)
    s = format (s, ", urg 0x%04x", tcp->urg_ptr);

  return s;
}

u8 * format_udp4_packet (u8 * s, va_list * args)
{
  u8 * p = va_arg (*args, u8 *);
  struct iphdr * ip = (void *) p;
  struct udphdr * udp = (void *) (ip + 1);

  s = format (s, "udp %U:%U -> %U:%U",
	      format_network_address, AF_INET,  &ip->saddr,
	      format_network_port, IPPROTO_UDP, ntohs (udp->source),
	      format_network_address, AF_INET,  &ip->daddr,
	      format_network_port, IPPROTO_UDP, ntohs (udp->dest));

  return s;
}

u8 * format_icmp4_type_and_code (u8 * s, va_list * args)
{
  uword icmp_type = va_arg (*args, uword);
  uword icmp_code = va_arg (*args, uword);

  switch (icmp_type)
    {
#define _(f,str) case ICMP_##f: s = format (s, str); break;
      _ (ECHOREPLY, "echo reply");
      _ (DEST_UNREACH, "unreachable");
      _ (SOURCE_QUENCH, "source quench");
      _ (REDIRECT, "redirect");
      _ (ECHO, "echo request");
      _ (TIME_EXCEEDED, "time exceeded");
      _ (PARAMETERPROB, "parameter problem");
      _ (TIMESTAMP, "timestamp request");
      _ (TIMESTAMPREPLY, "timestamp reply");
      _ (INFO_REQUEST, "information request");
      _ (INFO_REPLY, "information reply");
      _ (ADDRESS, "address mask request");
      _ (ADDRESSREPLY, "address mask reply");
#undef _
    default:
      s = format (s, "unknown type 0x%x", icmp_type);
    }

  if (icmp_type == ICMP_DEST_UNREACH)
    {
      switch (icmp_code)
	{
#define _(f,str) case ICMP_##f: s = format (s, " " # str); break;
	  _ (NET_UNREACH, "network");
	  _ (HOST_UNREACH, "host");
	  _ (PROT_UNREACH, "protocol");
	  _ (PORT_UNREACH, "port");
	  _ (FRAG_NEEDED, ": fragmentation needed/DF set");
	  _ (SR_FAILED, "source route failed");
	  _ (NET_UNKNOWN, "network unknown");
	  _ (HOST_UNKNOWN, "host unknown");
	  _ (HOST_ISOLATED, "host isolated");
	  _ (NET_ANO, "network: admin. prohibited");
	  _ (HOST_ANO, "host: admin. prohibited");
	  _ (NET_UNR_TOS, "network for type-of-service");
	  _ (HOST_UNR_TOS, "host for type-of-service");
	  _ (PKT_FILTERED, ": packet filtered");
	  _ (PREC_VIOLATION, "precedence violation");
	  _ (PREC_CUTOFF, "precedence cut off");
#undef _
	default:
	  s = format (s, "unknown code 0x%x", icmp_code);
	}
    }
  else if (icmp_type == ICMP_REDIRECT)
    {
      switch (icmp_code)
	{
#define _(f,str) case ICMP_##f: s = format (s, " " # str); break;
	  _ (REDIR_NET, "network");
	  _ (REDIR_HOST, "host");
	  _ (REDIR_NETTOS, "network for type-of-service");
	  _ (REDIR_HOSTTOS, "host for type-of-service");
#undef _
	default:
	  s = format (s, "unknown code 0x%x", icmp_code);
	}
    }
  else if (icmp_type == ICMP_TIME_EXCEEDED)
    {
      switch (icmp_code)
	{
#define _(f,str) case ICMP_##f: s = format (s, " " # str); break;
	  _ (EXC_TTL, "time-to-live zero in transit");
	  _ (EXC_FRAGTIME, "time-to-live zero during reassembly");
#undef _
	default:
	  s = format (s, "unknown code 0x%x", icmp_code);
	}
    }

  return s;
}

typedef struct {
  u8 type;
  u8 code;
  u16 checksum;
} icmp4_t;

u8 * format_icmp4_packet (u8 * s, va_list * args)
{
  u8 * p = va_arg (*args, u8 *);
  struct iphdr * ip = (void *) p;
  icmp4_t * icmp = (void *) (ip + 1);
  s = format (s, "icmp %U %U -> %U",
	      format_icmp4_type_and_code, icmp->type, icmp->code,
	      format_network_address, AF_INET,  &ip->saddr,
	      format_network_address, AF_INET,  &ip->daddr);

  return s;
}

u8 * format_ip4_tos_byte (u8 * s, va_list * args)
{
  uword tos = va_arg (*args, uword);

  if (tos & IPTOS_LOWDELAY)
    s = format (s, "minimize-delay, ");
  if (tos & IPTOS_MINCOST)
    s = format (s, "minimize-cost, ");
  if (tos & IPTOS_THROUGHPUT)
    s = format (s, "maximize-throughput, ");
  if (tos & IPTOS_RELIABILITY)
    s = format (s, "maximize-reliability, ");

  switch (IPTOS_PREC (tos))
    {
#define _(x,y) case IPTOS_PREC_##x: s = format (s, y); break
      _ (NETCONTROL, "network");
      _ (INTERNETCONTROL, "internet");
      _ (CRITIC_ECP, "critical");
      _ (FLASH, "flash");
      _ (FLASHOVERRIDE, "flash-override");
      _ (IMMEDIATE, "immediate");
      _ (PRIORITY, "priority");
      _ (ROUTINE, "routine");
#undef _
    }

  return s;
}

u8 * format_ip4_packet (u8 * s, va_list * args)
{
  u8 * p = va_arg (*args, u8 *);
  struct iphdr * ip = (void *) p;

  static format_function_t * f[256];

  if (! f[IPPROTO_TCP])
    {
      f[IPPROTO_TCP] = format_tcp4_packet;
      f[IPPROTO_UDP] = format_udp4_packet;
      f[IPPROTO_ICMP] = format_icmp4_packet;
    }

  if (f[ip->protocol])
    return format (s, "%U", f[ip->protocol], p);

  s = format (s, "%U: %U -> %U",
	      format_network_protocol, AF_INET, ip->protocol,
	      format_network_address, AF_INET,  &ip->saddr,
	      format_network_address, AF_INET,  &ip->daddr);

  return s;
}

#define foreach_unix_arphrd_type		\
  _ (NETROM, 0)					\
  _ (ETHER, 1)					\
  _ (EETHER, 2)					\
  _ (AX25, 3)					\
  _ (PRONET, 4)					\
  _ (CHAOS, 5)					\
  _ (IEEE802, 6)				\
  _ (ARCNET, 7)					\
  _ (APPLETLK, 8)				\
  _ (DLCI, 15)					\
  _ (ATM, 19)					\
  _ (METRICOM, 23)				\
  _ (IEEE1394, 24)				\
  _ (EUI64, 27)					\
  _ (INFINIBAND, 32)				\
  _ (SLIP, 256)					\
  _ (CSLIP, 257)				\
  _ (SLIP6, 258)				\
  _ (CSLIP6, 259)				\
  _ (RSRVD, 260)				\
  _ (ADAPT, 264)				\
  _ (ROSE, 270)					\
  _ (X25, 271)					\
  _ (HWX25, 272)				\
  _ (PPP, 512)					\
  _ (HDLC, 513)					\
  _ (LAPB, 516)					\
  _ (DDCMP, 517)				\
  _ (RAWHDLC, 518)				\
  _ (TUNNEL, 768)				\
  _ (TUNNEL6, 769)				\
  _ (FRAD, 770)					\
  _ (SKIP, 771)					\
  _ (LOOPBACK, 772)				\
  _ (LOCALTLK, 773)				\
  _ (FDDI, 774)					\
  _ (BIF, 775)					\
  _ (SIT, 776)					\
  _ (IPDDP, 777)				\
  _ (IPGRE, 778)				\
  _ (PIMREG, 779)				\
  _ (HIPPI, 780)				\
  _ (ASH, 781)					\
  _ (ECONET, 782)				\
  _ (IRDA, 783)					\
  _ (FCPP, 784)					\
  _ (FCAL, 785)					\
  _ (FCPL, 786)					\
  _ (FCFABRIC, 787)				\
  _ (IEEE802_TR, 800)				\
  _ (IEEE80211, 801)				\
  _ (IEEE80211_PRISM, 802)			\
  _ (IEEE80211_RADIOTAP, 803)			\
  _ (VOID, 0xFFFF)				\
  _ (NONE, 0xFFFE)

u8 * format_unix_arphrd (u8 * s, va_list * args)
{
#ifndef __COVERITY__ /* doesn't understand this at all... */
  u32 x = va_arg (*args, u32);
  char * t;
  switch (x)
    {
#define _(f,n) case ARPHRD_##f: t = #f; break;
      foreach_unix_arphrd_type
#undef _
    default:
      t = 0;
      break;
    }

  if (t)
    s = format (s, "%s", t);
  else
    s = format (s, "unknown 0x%x", x);
#endif
  return s;
}

#define foreach_unix_interface_flag		\
  _ (up)					\
  _ (broadcast)					\
  _ (debug)					\
  _ (loopback)					\
  _ (pointopoint)				\
  _ (notrailers)				\
  _ (running)					\
  _ (noarp)					\
  _ (promisc)					\
  _ (allmulti)					\
  _ (master)					\
  _ (slave)					\
  _ (multicast)					\
  _ (portsel)					\
  _ (automedia)					\
  _ (dynamic)					\
  _ (lower_up)					\
  _ (dormant)					\
  _ (echo)

static char * unix_interface_flag_names[] = {
#define _(f) #f,
  foreach_unix_interface_flag
#undef _
};

u8 * format_unix_interface_flags (u8 * s, va_list * args)
{
  u32 x = va_arg (*args, u32);
  u32 i;

  if (x == 0)
    s = format (s, "none");
  else foreach_set_bit (i, x, ({
    if (i < ARRAY_LEN (unix_interface_flag_names))
      s = format (s, "%s", unix_interface_flag_names[i]);
    else
      s = format (s, "unknown %d", i);
    if (x >> (i + 1))
      s = format (s, ", ");
  }));
  return s;
}

typedef struct {
  u16 ar_hrd;			/* format of hardware address	*/
  u16 ar_pro;			/* format of protocol address	*/
  u8  ar_hln;			/* length of hardware address	*/
  u8  ar_pln;			/* length of protocol address	*/
  u16 ar_op;			/* ARP opcode (command)		*/
  u8  ar_sha[6];		/* sender hardware address	*/
  u8  ar_spa[4];		/* sender IP address		*/
  u8  ar_tha[6];		/* target hardware address	*/
  u8  ar_tpa[4];		/* target IP address		*/
} arp_ether_ip4_t;

u8 * format_arp_packet (u8 * s, va_list * args)
{
  arp_ether_ip4_t * a = va_arg (*args, arp_ether_ip4_t *);
  char * op = "unknown";

  if (a->ar_pro != ETH_P_IP ||
      a->ar_hrd != ARPHRD_ETHER)
    return s;

  switch (a->ar_op)
    {
#define _(f) case ARPOP_##f: op = #f; break;
      _ (REQUEST);
      _ (REPLY);
      _ (RREQUEST);
      _ (RREPLY);
#undef _
    }

  s = format (s, "%s %U %U -> %U %U",
	      op,
	      format_network_address, AF_INET,   a->ar_spa,
	      format_network_address, AF_UNSPEC, a->ar_sha,
	      format_network_address, AF_INET,   a->ar_tpa,
	      format_network_address, AF_UNSPEC, a->ar_tha);
  return s;
}

u8 * format_ethernet_proto (u8 * s, va_list * args)
{
  uword type = va_arg (*args, uword);
  char * t = 0;

  switch (type)
    {
    case 0: t = "BPDU"; break;
#define _(f) case ETH_P_##f: t = #f; break;
      _ (LOOP);
      _ (PUP);
#ifdef ETH_P_PUPAT
      _ (PUPAT);
#endif
      _ (IP);
      _ (X25);
      _ (ARP);
      _ (BPQ);
#ifdef ETH_P_PUPAT
      _ (IEEEPUP);
      _ (IEEEPUPAT);
#endif
      _ (DEC);
      _ (DNA_DL);
      _ (DNA_RC);
      _ (DNA_RT);
      _ (LAT);
      _ (DIAG);
      _ (CUST);
      _ (SCA);
      _ (RARP);
      _ (ATALK);
      _ (AARP);
      _ (IPX);
      _ (IPV6);
#ifdef ETH_P_PPP_DISC
      _ (PPP_DISC);
      _ (PPP_SES);
#endif
#ifdef ETH_P_ATMMPOA
      _ (ATMMPOA);
      _ (ATMFATE);
#endif
      _ (802_3);
      _ (AX25);
      _ (ALL);
      _ (802_2);
      _ (SNAP);
      _ (DDCMP);
      _ (WAN_PPP);
      _ (PPP_MP);
      _ (LOCALTALK);
      _ (PPPTALK);
      _ (TR_802_2);
      _ (MOBITEX);
      _ (CONTROL);
      _ (IRDA);
#ifdef ETH_P_ECONET
      _ (ECONET);
#endif
#undef _
    }

  if (t)
    vec_add (s, t, strlen (t));
  else
    s = format (s, "ether-type 0x%x", type);
  return s;
}

u8 * format_ethernet_packet (u8 * s, va_list * args)
{
  struct ethhdr * h = va_arg (*args, struct ethhdr *);
  uword proto = h->h_proto;
  u8 * payload = (void *) (h + 1);
  u32 indent;

  /* Check for 802.2/802.3 encapsulation. */
  if (proto < ETH_DATA_LEN)
    {
      typedef struct {
	u8 dsap, ssap, control;
	u8 orig_code[3];
	u16 proto;
      } ethhdr_802_t;
      ethhdr_802_t * h1 = (void *) (h + 1);
      proto = h1->proto;
      payload = (void *) (h1 + 1);
    }

  indent = format_get_indent (s);

  s = format (s, "%U: %U -> %U",
	      format_ethernet_proto, proto,
	      format_network_address, AF_UNSPEC, h->h_source,
	      format_network_address, AF_UNSPEC, h->h_dest);

  switch (proto)
    {
    case ETH_P_ARP:
      s = format (s, "\n%U%U",
		  format_white_space, indent,
		  format_arp_packet, payload);
      break;
    }

  return s;
}

#ifndef __KERNEL__
u8 * format_hostname (u8 * s, va_list * args)
{
  char buffer[1024];
  char * b = buffer;
  if (gethostname (b, sizeof (buffer)) < 0)
    b = "noname";
  return format (s, "%s", b);
}
#endif

#ifndef __KERNEL__
u8 * format_timeval (u8 * s, va_list * args)
{
  char * fmt = va_arg (*args, char *);
  struct timeval * tv = va_arg (*args, struct timeval *);
  struct tm * tm;
  word msec;
  char * f, c;

  if (! fmt)
    fmt = "y/m/d H:M:S:F";

  if (! tv)
    {
      static struct timeval now;
      gettimeofday (&now, 0);
      tv = &now;
    }

  msec = flt_round_nearest (1e-3 * tv->tv_usec);
  if (msec >= 1000)
    { msec = 0; tv->tv_sec++; }

  {
    time_t t = tv->tv_sec;
    tm = localtime (&t);
  }

  for (f = fmt; *f; f++)
    {
      uword what;
      char * what_fmt = "%d";

      switch (c = *f)
	{
	default:
	  vec_add1 (s, c);
	  continue;

	case 'y':
	  what = 1900 + tm->tm_year;
	  what_fmt = "%4d";
	  break;
	case 'm':
	  what = tm->tm_mon + 1;
	  what_fmt = "%2d";
	  break;
	case 'd':
	  what = tm->tm_mday;
	  what_fmt = "%2d";
	  break;
	case 'H':
	  what = tm->tm_hour;
	  what_fmt = "%02d";
	  break;
	case 'M':
	  what = tm->tm_min;
	  what_fmt = "%02d";
	  break;
	case 'S':
	  what = tm->tm_sec;
	  what_fmt = "%02d";
	  break;
	case 'F':
	  what = msec;
	  what_fmt = "%03d";
	  break;
	}

      s = format (s, what_fmt, what);
    }

  return s;
}
#endif

u8 * format_time_float (u8 * s, va_list * args)
{
  u8 * fmt = va_arg (*args, u8 *);
  f64 t = va_arg (*args, f64);
  struct timeval tv;
  if (t <= 0)
    t = unix_time_now ();
  tv.tv_sec = t;
  tv.tv_usec = 1e6*(t - tv.tv_sec);
  return format (s, "%U", format_timeval, fmt, &tv);
}

u8 * format_signal (u8 * s, va_list * args)
{
  uword signum = va_arg (*args, uword);
  char * t = 0;
  switch (signum)
    {
#define _(x) case x: t = #x; break;
      _ (SIGHUP);
      _ (SIGINT);
      _ (SIGQUIT);
      _ (SIGILL);
      _ (SIGTRAP);
      _ (SIGABRT);
      _ (SIGBUS);
      _ (SIGFPE);
      _ (SIGKILL);
      _ (SIGUSR1);
      _ (SIGSEGV);
      _ (SIGUSR2);
      _ (SIGPIPE);
      _ (SIGALRM);
      _ (SIGTERM);
#ifdef SIGSTKFLT
      _ (SIGSTKFLT);
#endif
      _ (SIGCHLD);
      _ (SIGCONT);
      _ (SIGSTOP);
      _ (SIGTSTP);
      _ (SIGTTIN);
      _ (SIGTTOU);
      _ (SIGURG);
      _ (SIGXCPU);
      _ (SIGXFSZ);
      _ (SIGVTALRM);
      _ (SIGPROF);
      _ (SIGWINCH);
      _ (SIGIO);
      _ (SIGPWR);
#ifdef SIGSYS
      _ (SIGSYS);
#endif
#undef _
    default:
      return format (s, "unknown %d", signum);
    }

  vec_add (s, t, strlen (t));
  return s;
}

u8 * format_ucontext_pc (u8 * s, va_list * args)
{
  ucontext_t * uc __attribute__((unused));
  unsigned long * regs = 0;
  uword reg_no = 0;

  uc = va_arg (*args, ucontext_t *);

#if defined (powerpc)
  regs = &uc->uc_mcontext.uc_regs->gregs[0];
#elif defined (powerpc64)
  regs = &uc->uc_mcontext.uc_regs->gp_regs[0];
#elif defined (i386) || defined (__x86_64__)
  regs = (void *) &uc->uc_mcontext.gregs[0];
#endif

#if defined (powerpc) || defined (powerpc64)
  reg_no = PT_NIP;
#elif defined (i386)
  reg_no = REG_EIP;
#elif defined (__x86_64__)
  reg_no = REG_RIP;
#else
  reg_no = 0;
  regs = 0;
#endif

  if (! regs)
    return format (s, "unsupported");
  else
    return format (s, "%p", regs[reg_no]);
}

uword
unformat_unix_gid (unformat_input_t * input, va_list * args)
{
  gid_t *gid = va_arg (*args, gid_t *);
  struct group *grp = 0;
  int r;
  u8 *s;

  if (unformat (input, "%d", &r))
    {
      grp = getgrgid (r);
    }
  else if (unformat (input, "%s", &s))
    {
      grp = getgrnam ((char *) s);
      vec_free (s);
    }
  if (grp)
    {
      *gid = grp->gr_gid;
      return 1;
    }
  return 0;
}

#define MAX_NUMNODES 16
u8 *
format_page_map (u8 * s, va_list * args)
{
  uword va = va_arg (*args, uword);
  uword size = va_arg (*args, uword);
  uword page_size = clib_mem_get_page_size ();
  u32 indent = format_get_indent (s);
  uword n_pages = size / page_size;
  uword pages_per_numa[MAX_NUMNODES] = { 0 };
  uword pages_not_mapped = 0;
  uword pages_unknown = 0;
  int *status = 0;
  void **ptr = 0;
  int i;

  s = format (s, "virtual memory start 0x%llx, size %lluk, %u pages, "
	      "page size %uk", va, size / 1024, n_pages, page_size / 1024);

  vec_validate (status, n_pages - 1);
  vec_validate (ptr, n_pages - 1);

  for (i = 0; i < n_pages; i++)
    ptr[i] = uword_to_pointer (va + i * page_size, void *);

  if (move_pages (0, n_pages, ptr, 0, status, 0) != 0)
    {
      s = format (s, "\n%Upage information not available (errno %u)",
		  format_white_space, indent + 2, errno);
      goto done;
    }

  for (i = 0; i < n_pages; i++)
    {
      if (status[i] >= 0 && status[i] < MAX_NUMNODES)
	pages_per_numa[status[i]]++;
      else if (status[i] == -EFAULT)
	pages_not_mapped++;
      else
	pages_unknown++;
    }

  for (i = 0; i < MAX_NUMNODES; i++)
    if (pages_per_numa[i])
      s = format (s, "\n%Unuma %u: %d pages, %luk", format_white_space,
		  indent + 2, i, pages_per_numa[i], pages_per_numa[i] *
		  page_size / 1024);

  s = format (s, "\n%Unot mapped: %u pages, %luk", format_white_space,
	      indent + 2, pages_not_mapped, pages_not_mapped *
	      page_size / 1024);

  if (pages_unknown)
    s = format (s, "\n%Uunknown: %u pages, %luk", format_white_space,
		indent + 2, pages_unknown, pages_unknown * page_size / 1024);

done:
  vec_free (status);
  vec_free (ptr);
  return s;
}

#endif /* __KERNEL__ */