6 files changed, 76 insertions, 59 deletions

diff --git a/drivers/net/igb/e1000_82575.h b/drivers/net/igb/e1000_82575.h
index eaf977050368..0f16abab2565 100644
--- a/drivers/net/igb/e1000_82575.h
+++ b/drivers/net/igb/e1000_82575.h
@@ -130,6 +130,7 @@ struct e1000_adv_tx_context_desc {
 #define E1000_ADVTXD_MACLEN_SHIFT    9  /* Adv ctxt desc mac len shift */
 #define E1000_ADVTXD_TUCMD_IPV4    0x00000400  /* IP Packet Type: 1=IPv4 */
 #define E1000_ADVTXD_TUCMD_L4T_TCP 0x00000800  /* L4 Packet TYPE of TCP */
+#define E1000_ADVTXD_TUCMD_L4T_SCTP 0x00001000 /* L4 packet TYPE of SCTP */
 /* IPSec Encrypt Enable for ESP */
 #define E1000_ADVTXD_L4LEN_SHIFT     8  /* Adv ctxt L4LEN shift */
 #define E1000_ADVTXD_MSS_SHIFT      16  /* Adv ctxt MSS shift */
diff --git a/drivers/net/igb/e1000_defines.h b/drivers/net/igb/e1000_defines.h
index ad2d319d0f8b..3bda3db73f1f 100644
--- a/drivers/net/igb/e1000_defines.h
+++ b/drivers/net/igb/e1000_defines.h
@@ -289,8 +289,9 @@
 #define E1000_SCTL_DISABLE_SERDES_LOOPBACK 0x0400
 
 /* Receive Checksum Control */
+#define E1000_RXCSUM_IPOFL     0x00000100   /* IPv4 checksum offload */
 #define E1000_RXCSUM_TUOFL     0x00000200   /* TCP / UDP checksum offload */
-#define E1000_RXCSUM_IPPCSE    0x00001000   /* IP payload checksum enable */
+#define E1000_RXCSUM_CRCOFL    0x00000800   /* CRC32 offload enable */
 #define E1000_RXCSUM_PCSD      0x00002000   /* packet checksum disabled */
 
 /* Header split receive */
diff --git a/drivers/net/igb/e1000_mbx.c b/drivers/net/igb/e1000_mbx.c
index 840782fb5736..ed9058eca45c 100644
--- a/drivers/net/igb/e1000_mbx.c
+++ b/drivers/net/igb/e1000_mbx.c
@@ -140,13 +140,13 @@ static s32 igb_poll_for_msg(struct e1000_hw *hw, u16 mbx_id)
         struct e1000_mbx_info *mbx = &hw->mbx;
         int countdown = mbx->timeout;
 
-        if (!mbx->ops.check_for_msg)
+        if (!countdown || !mbx->ops.check_for_msg)
                 goto out;
 
         while (mbx->ops.check_for_msg(hw, mbx_id)) {
+                countdown--;
                 if (!countdown)
                         break;
-                countdown--;
                 udelay(mbx->usec_delay);
         }
 out:
@@ -165,13 +165,13 @@ static s32 igb_poll_for_ack(struct e1000_hw *hw, u16 mbx_id)
         struct e1000_mbx_info *mbx = &hw->mbx;
         int countdown = mbx->timeout;
 
-        if (!mbx->ops.check_for_ack)
+        if (!countdown || !mbx->ops.check_for_ack)
                 goto out;
 
         while (mbx->ops.check_for_ack(hw, mbx_id)) {
+                countdown--;
                 if (!countdown)
                         break;
-                countdown--;
                 udelay(mbx->usec_delay);
         }
 out:
diff --git a/drivers/net/igb/igb.h b/drivers/net/igb/igb.h
index 4e8464b9df2e..154c5acc6fce 100644
--- a/drivers/net/igb/igb.h
+++ b/drivers/net/igb/igb.h
@@ -238,7 +238,6 @@ struct igb_adapter {
         u64 hw_csum_err;
         u64 hw_csum_good;
         u32 alloc_rx_buff_failed;
-        bool rx_csum;
         u32 gorc;
         u64 gorc_old;
         u16 rx_ps_hdr_size;
@@ -286,6 +285,7 @@ struct igb_adapter {
 #define IGB_FLAG_DCA_ENABLED       (1 << 1)
 #define IGB_FLAG_QUAD_PORT_A       (1 << 2)
 #define IGB_FLAG_NEED_CTX_IDX      (1 << 3)
+#define IGB_FLAG_RX_CSUM_DISABLED  (1 << 4)
 
 enum e1000_state_t {
         __IGB_TESTING,
diff --git a/drivers/net/igb/igb_ethtool.c b/drivers/net/igb/igb_ethtool.c
index 27eae49e79c2..b8551a57dd3f 100644
--- a/drivers/net/igb/igb_ethtool.c
+++ b/drivers/net/igb/igb_ethtool.c
@@ -275,13 +275,17 @@ static int igb_set_pauseparam(struct net_device *netdev,
 static u32 igb_get_rx_csum(struct net_device *netdev)
 {
         struct igb_adapter *adapter = netdev_priv(netdev);
-        return adapter->rx_csum;
+        return !(adapter->flags & IGB_FLAG_RX_CSUM_DISABLED);
 }
 
 static int igb_set_rx_csum(struct net_device *netdev, u32 data)
 {
         struct igb_adapter *adapter = netdev_priv(netdev);
-        adapter->rx_csum = data;
+
+        if (data)
+                adapter->flags &= ~IGB_FLAG_RX_CSUM_DISABLED;
+        else
+                adapter->flags |= IGB_FLAG_RX_CSUM_DISABLED;
 
         return 0;
 }
@@ -293,10 +297,16 @@ static u32 igb_get_tx_csum(struct net_device *netdev)
 
 static int igb_set_tx_csum(struct net_device *netdev, u32 data)
 {
-        if (data)
+        struct igb_adapter *adapter = netdev_priv(netdev);
+
+        if (data) {
                 netdev->features |= (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM);
-        else
-                netdev->features &= ~(NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM);
+                if (adapter->hw.mac.type == e1000_82576)
+                        netdev->features |= NETIF_F_SCTP_CSUM;
+        } else {
+                netdev->features &= ~(NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
+                                      NETIF_F_SCTP_CSUM);
+        }
 
         return 0;
 }
diff --git a/drivers/net/igb/igb_main.c b/drivers/net/igb/igb_main.c
index e25343588fc7..ffd731539997 100644
--- a/drivers/net/igb/igb_main.c
+++ b/drivers/net/igb/igb_main.c
@@ -942,6 +942,8 @@ int igb_up(struct igb_adapter *adapter)
         rd32(E1000_ICR);
         igb_irq_enable(adapter);
 
+        netif_tx_start_all_queues(adapter->netdev);
+
         /* Fire a link change interrupt to start the watchdog. */
         wr32(E1000_ICS, E1000_ICS_LSC);
         return 0;
@@ -1343,6 +1345,9 @@ static int __devinit igb_probe(struct pci_dev *pdev,
         if (pci_using_dac)
                 netdev->features |= NETIF_F_HIGHDMA;
 
+        if (adapter->hw.mac.type == e1000_82576)
+                netdev->features |= NETIF_F_SCTP_CSUM;
+
         adapter->en_mng_pt = igb_enable_mng_pass_thru(&adapter->hw);
 
         /* before reading the NVM, reset the controller to put the device in a
@@ -1390,8 +1395,6 @@ static int __devinit igb_probe(struct pci_dev *pdev,
 
         igb_validate_mdi_setting(hw);
 
-        adapter->rx_csum = 1;
-
         /* Initial Wake on LAN setting If APM wake is enabled in the EEPROM,
          * enable the ACPI Magic Packet filter
          */
@@ -1442,15 +1445,14 @@ static int __devinit igb_probe(struct pci_dev *pdev,
          * driver. */
         igb_get_hw_control(adapter);
 
-        /* tell the stack to leave us alone until igb_open() is called */
-        netif_carrier_off(netdev);
-        netif_tx_stop_all_queues(netdev);
-
         strcpy(netdev->name, "eth%d");
         err = register_netdev(netdev);
         if (err)
                 goto err_register;
 
+        /* carrier off reporting is important to ethtool even BEFORE open */
+        netif_carrier_off(netdev);
+
 #ifdef CONFIG_IGB_DCA
         if (dca_add_requester(&pdev->dev) == 0) {
                 adapter->flags |= IGB_FLAG_DCA_ENABLED;
@@ -1699,6 +1701,8 @@ static int igb_open(struct net_device *netdev)
         if (test_bit(__IGB_TESTING, &adapter->state))
                 return -EBUSY;
 
+        netif_carrier_off(netdev);
+
         /* allocate transmit descriptors */
         err = igb_setup_all_tx_resources(adapter);
         if (err)
@@ -2231,29 +2235,24 @@ static void igb_configure_rx(struct igb_adapter *adapter)
                 mrqc |= (E1000_MRQC_RSS_FIELD_IPV6_UDP_EX |
                          E1000_MRQC_RSS_FIELD_IPV6_TCP_EX);
 
-
                 wr32(E1000_MRQC, mrqc);
-
-                /* Multiqueue and raw packet checksumming are mutually
-                 * exclusive.  Note that this not the same as TCP/IP
-                 * checksumming, which works fine. */
-                rxcsum = rd32(E1000_RXCSUM);
-                rxcsum |= E1000_RXCSUM_PCSD;
-                wr32(E1000_RXCSUM, rxcsum);
-        } else {
+        } else if (adapter->vfs_allocated_count) {
                 /* Enable multi-queue for sr-iov */
-                if (adapter->vfs_allocated_count)
-                        wr32(E1000_MRQC, E1000_MRQC_ENABLE_VMDQ);
-                /* Enable Receive Checksum Offload for TCP and UDP */
-                rxcsum = rd32(E1000_RXCSUM);
-                if (adapter->rx_csum)
-                        rxcsum |= E1000_RXCSUM_TUOFL | E1000_RXCSUM_IPPCSE;
-                else
-                        rxcsum &= ~(E1000_RXCSUM_TUOFL | E1000_RXCSUM_IPPCSE);
-
-                wr32(E1000_RXCSUM, rxcsum);
+                wr32(E1000_MRQC, E1000_MRQC_ENABLE_VMDQ);
         }
 
+        /* Enable Receive Checksum Offload for TCP and UDP */
+        rxcsum = rd32(E1000_RXCSUM);
+        /* Disable raw packet checksumming */
+        rxcsum |= E1000_RXCSUM_PCSD;
+
+        if (adapter->hw.mac.type == e1000_82576)
+                /* Enable Receive Checksum Offload for SCTP */
+                rxcsum |= E1000_RXCSUM_CRCOFL;
+
+        /* Don't need to set TUOFL or IPOFL, they default to 1 */
+        wr32(E1000_RXCSUM, rxcsum);
+
         /* Set the default pool for the PF's first queue */
         igb_configure_vt_default_pool(adapter);
 
@@ -2661,7 +2660,6 @@ static void igb_watchdog_task(struct work_struct *work)
                         }
 
                         netif_carrier_on(netdev);
-                        netif_tx_wake_all_queues(netdev);
 
                         igb_ping_all_vfs(adapter);
 
@@ -2678,7 +2676,6 @@ static void igb_watchdog_task(struct work_struct *work)
                         printk(KERN_INFO "igb: %s NIC Link is Down\n",
                                netdev->name);
                         netif_carrier_off(netdev);
-                        netif_tx_stop_all_queues(netdev);
 
                         igb_ping_all_vfs(adapter);
 
@@ -2712,6 +2709,8 @@ link_up:
                          * (Do the reset outside of interrupt context). */
                         adapter->tx_timeout_count++;
                         schedule_work(&adapter->reset_task);
+                        /* return immediately since reset is imminent */
+                        return;
                 }
         }
 
@@ -2895,13 +2894,13 @@ static void igb_set_itr(struct igb_adapter *adapter)
         switch (current_itr) {
         /* counts and packets in update_itr are dependent on these numbers */
         case lowest_latency:
-                new_itr = 70000;
+                new_itr = 56;  /* aka 70,000 ints/sec */
                 break;
         case low_latency:
-                new_itr = 20000; /* aka hwitr = ~200 */
+                new_itr = 196; /* aka 20,000 ints/sec */
                 break;
         case bulk_latency:
-                new_itr = 4000;
+                new_itr = 980; /* aka 4,000 ints/sec */
                 break;
         default:
                 break;
@@ -2920,7 +2919,8 @@ set_itr_now:
                  * by adding intermediate steps when interrupt rate is
                  * increasing */
                 new_itr = new_itr > adapter->itr ?
-                             min(adapter->itr + (new_itr >> 2), new_itr) :
+                             max((new_itr * adapter->itr) /
+                                 (new_itr + (adapter->itr >> 2)), new_itr) :
                              new_itr;
                 /* Don't write the value here; it resets the adapter's
                  * internal timer, and causes us to delay far longer than
@@ -2929,7 +2929,7 @@ set_itr_now:
                  * ends up being correct.
                  */
                 adapter->itr = new_itr;
-                adapter->rx_ring->itr_val = 1000000000 / (new_itr * 256);
+                adapter->rx_ring->itr_val = new_itr;
                 adapter->rx_ring->set_itr = 1;
         }
 
@@ -3068,11 +3068,15 @@ static inline bool igb_tx_csum_adv(struct igb_adapter *adapter,
                                 tu_cmd |= E1000_ADVTXD_TUCMD_IPV4;
                                 if (ip_hdr(skb)->protocol == IPPROTO_TCP)
                                         tu_cmd |= E1000_ADVTXD_TUCMD_L4T_TCP;
+                                else if (ip_hdr(skb)->protocol == IPPROTO_SCTP)
+                                        tu_cmd |= E1000_ADVTXD_TUCMD_L4T_SCTP;
                                 break;
                         case cpu_to_be16(ETH_P_IPV6):
                                 /* XXX what about other V6 headers?? */
                                 if (ipv6_hdr(skb)->nexthdr == IPPROTO_TCP)
                                         tu_cmd |= E1000_ADVTXD_TUCMD_L4T_TCP;
+                                else if (ipv6_hdr(skb)->nexthdr == IPPROTO_SCTP)
+                                        tu_cmd |= E1000_ADVTXD_TUCMD_L4T_SCTP;
                                 break;
                         default:
                                 if (unlikely(net_ratelimit()))
@@ -4434,20 +4438,12 @@ static void igb_receive_skb(struct igb_ring *ring, u8 status,
         bool vlan_extracted = (adapter->vlgrp && (status & E1000_RXD_STAT_VP));
 
         skb_record_rx_queue(skb, ring->queue_index);
-        if (skb->ip_summed == CHECKSUM_UNNECESSARY) {
-                if (vlan_extracted)
-                        vlan_gro_receive(&ring->napi, adapter->vlgrp,
-                                         le16_to_cpu(rx_desc->wb.upper.vlan),
-                                         skb);
-                else
-                        napi_gro_receive(&ring->napi, skb);
-        } else {
-                if (vlan_extracted)
-                        vlan_hwaccel_receive_skb(skb, adapter->vlgrp,
-                                          le16_to_cpu(rx_desc->wb.upper.vlan));
-                else
-                        netif_receive_skb(skb);
-        }
+        if (vlan_extracted)
+                vlan_gro_receive(&ring->napi, adapter->vlgrp,
+                                 le16_to_cpu(rx_desc->wb.upper.vlan),
+                                 skb);
+        else
+                napi_gro_receive(&ring->napi, skb);
 }
 
 static inline void igb_rx_checksum_adv(struct igb_adapter *adapter,
@@ -4456,19 +4452,28 @@ static inline void igb_rx_checksum_adv(struct igb_adapter *adapter,
         skb->ip_summed = CHECKSUM_NONE;
 
         /* Ignore Checksum bit is set or checksum is disabled through ethtool */
-        if ((status_err & E1000_RXD_STAT_IXSM) || !adapter->rx_csum)
+        if ((status_err & E1000_RXD_STAT_IXSM) ||
+            (adapter->flags & IGB_FLAG_RX_CSUM_DISABLED))
                 return;
         /* TCP/UDP checksum error bit is set */
         if (status_err &
             (E1000_RXDEXT_STATERR_TCPE | E1000_RXDEXT_STATERR_IPE)) {
+                /*
+                 * work around errata with sctp packets where the TCPE aka
+                 * L4E bit is set incorrectly on 64 byte (60 byte w/o crc)
+                 * packets, (aka let the stack check the crc32c)
+                 */
+                if (!((adapter->hw.mac.type == e1000_82576) &&
+                      (skb->len == 60)))
+                        adapter->hw_csum_err++;
                 /* let the stack verify checksum errors */
-                adapter->hw_csum_err++;
                 return;
         }
         /* It must be a TCP or UDP packet with a valid checksum */
         if (status_err & (E1000_RXD_STAT_TCPCS | E1000_RXD_STAT_UDPCS))
                 skb->ip_summed = CHECKSUM_UNNECESSARY;
 
+        dev_dbg(&adapter->pdev->dev, "cksum success: bits %08X\n", status_err);
         adapter->hw_csum_good++;
 }