Automatic merge of /spare/repo/netdev-2.6 branch e100

1 files changed, 139 insertions, 26 deletions

diff --git a/drivers/net/e100.c b/drivers/net/e100.c
index 1b68dd5a49b6..4a47df5a9ff9 100644
--- a/drivers/net/e100.c
+++ b/drivers/net/e100.c
@@ -155,9 +155,9 @@
 
 #define DRV_NAME                "e100"
 #define DRV_EXT         "-NAPI"
-#define DRV_VERSION             "3.3.6-k2"DRV_EXT
+#define DRV_VERSION             "3.4.8-k2"DRV_EXT
 #define DRV_DESCRIPTION         "Intel(R) PRO/100 Network Driver"
-#define DRV_COPYRIGHT           "Copyright(c) 1999-2004 Intel Corporation"
+#define DRV_COPYRIGHT           "Copyright(c) 1999-2005 Intel Corporation"
 #define PFX                     DRV_NAME ": "
 
 #define E100_WATCHDOG_PERIOD    (2 * HZ)
@@ -210,11 +210,17 @@ static struct pci_device_id e100_id_table[] = {
         INTEL_8255X_ETHERNET_DEVICE(0x1069, 6),
         INTEL_8255X_ETHERNET_DEVICE(0x106A, 6),
         INTEL_8255X_ETHERNET_DEVICE(0x106B, 6),
+        INTEL_8255X_ETHERNET_DEVICE(0x1091, 7),
+        INTEL_8255X_ETHERNET_DEVICE(0x1092, 7),
+        INTEL_8255X_ETHERNET_DEVICE(0x1093, 7),
+        INTEL_8255X_ETHERNET_DEVICE(0x1094, 7),
+        INTEL_8255X_ETHERNET_DEVICE(0x1095, 7),
         INTEL_8255X_ETHERNET_DEVICE(0x1209, 0),
         INTEL_8255X_ETHERNET_DEVICE(0x1229, 0),
         INTEL_8255X_ETHERNET_DEVICE(0x2449, 2),
         INTEL_8255X_ETHERNET_DEVICE(0x2459, 2),
         INTEL_8255X_ETHERNET_DEVICE(0x245D, 2),
+        INTEL_8255X_ETHERNET_DEVICE(0x27DC, 7),
         { 0, }
 };
 MODULE_DEVICE_TABLE(pci, e100_id_table);
@@ -269,6 +275,12 @@ enum scb_status {
         rus_mask         = 0x3C,
 };
 
+enum ru_state  {
+        RU_SUSPENDED = 0,
+        RU_RUNNING       = 1,
+        RU_UNINITIALIZED = -1,
+};
+
 enum scb_stat_ack {
         stat_ack_not_ours    = 0x00,
         stat_ack_sw_gen      = 0x04,
@@ -510,7 +522,7 @@ struct nic {
         struct rx *rx_to_use;
         struct rx *rx_to_clean;
         struct rfd blank_rfd;
-        int ru_running;
+        enum ru_state ru_running;
 
         spinlock_t cb_lock                      ____cacheline_aligned;
         spinlock_t cmd_lock;
@@ -539,6 +551,7 @@ struct nic {
         struct timer_list watchdog;
         struct timer_list blink_timer;
         struct mii_if_info mii;
+        struct work_struct tx_timeout_task;
         enum loopback loopback;
 
         struct mem *mem;
@@ -770,7 +783,7 @@ static int e100_eeprom_save(struct nic *nic, u16 start, u16 count)
         return 0;
 }
 
-#define E100_WAIT_SCB_TIMEOUT 40
+#define E100_WAIT_SCB_TIMEOUT 20000 /* we might have to wait 100ms!!! */
 static inline int e100_exec_cmd(struct nic *nic, u8 cmd, dma_addr_t dma_addr)
 {
         unsigned long flags;
@@ -840,6 +853,10 @@ static inline int e100_exec_cb(struct nic *nic, struct sk_buff *skb,
                          * because the controller is too busy, so
                          * let's just queue the command and try again
                          * when another command is scheduled. */
+                        if(err == -ENOSPC) {
+                                //request a reset
+                                schedule_work(&nic->tx_timeout_task);
+                        }
                         break;
                 } else {
                         nic->cuc_cmd = cuc_resume;
@@ -884,7 +901,7 @@ static void mdio_write(struct net_device *netdev, int addr, int reg, int data)
 
 static void e100_get_defaults(struct nic *nic)
 {
-        struct param_range rfds = { .min = 64, .max = 256, .count = 64 };
+        struct param_range rfds = { .min = 16, .max = 256, .count = 64 };
         struct param_range cbs  = { .min = 64, .max = 256, .count = 64 };
 
         pci_read_config_byte(nic->pdev, PCI_REVISION_ID, &nic->rev_id);
@@ -899,8 +916,9 @@ static void e100_get_defaults(struct nic *nic)
         /* Quadwords to DMA into FIFO before starting frame transmit */
         nic->tx_threshold = 0xE0;
 
-        nic->tx_command = cpu_to_le16(cb_tx | cb_i | cb_tx_sf |
-                ((nic->mac >= mac_82558_D101_A4) ? cb_cid : 0));
+        /* no interrupt for every tx completion, delay = 256us if not 557*/
+        nic->tx_command = cpu_to_le16(cb_tx | cb_tx_sf |
+                ((nic->mac >= mac_82558_D101_A4) ? cb_cid : cb_i));
 
         /* Template for a freshly allocated RFD */
         nic->blank_rfd.command = cpu_to_le16(cb_el);
@@ -964,7 +982,8 @@ static void e100_configure(struct nic *nic, struct cb *cb, struct sk_buff *skb)
         if(nic->flags & multicast_all)
                 config->multicast_all = 0x1;            /* 1=accept, 0=no */
 
-        if(!(nic->flags & wol_magic))
+        /* disable WoL when up */
+        if(netif_running(nic->netdev) || !(nic->flags & wol_magic))
                 config->magic_packet_disable = 0x1;     /* 1=off, 0=on */
 
         if(nic->mac >= mac_82558_D101_A4) {
@@ -1203,7 +1222,9 @@ static void e100_update_stats(struct nic *nic)
                 }
         }
 
-        e100_exec_cmd(nic, cuc_dump_reset, 0);
+        
+        if(e100_exec_cmd(nic, cuc_dump_reset, 0))
+                DPRINTK(TX_ERR, DEBUG, "exec cuc_dump_reset failed\n");
 }
 
 static void e100_adjust_adaptive_ifs(struct nic *nic, int speed, int duplex)
@@ -1279,12 +1300,15 @@ static inline void e100_xmit_prepare(struct nic *nic, struct cb *cb,
         struct sk_buff *skb)
 {
         cb->command = nic->tx_command;
+        /* interrupt every 16 packets regardless of delay */
+        if((nic->cbs_avail & ~15) == nic->cbs_avail) cb->command |= cb_i;
         cb->u.tcb.tbd_array = cb->dma_addr + offsetof(struct cb, u.tcb.tbd);
         cb->u.tcb.tcb_byte_count = 0;
         cb->u.tcb.threshold = nic->tx_threshold;
         cb->u.tcb.tbd_count = 1;
         cb->u.tcb.tbd.buf_addr = cpu_to_le32(pci_map_single(nic->pdev,
                 skb->data, skb->len, PCI_DMA_TODEVICE));
+        // check for mapping failure?
         cb->u.tcb.tbd.size = cpu_to_le16(skb->len);
 }
 
@@ -1297,7 +1321,8 @@ static int e100_xmit_frame(struct sk_buff *skb, struct net_device *netdev)
                 /* SW workaround for ICH[x] 10Mbps/half duplex Tx hang.
                    Issue a NOP command followed by a 1us delay before
                    issuing the Tx command. */
-                e100_exec_cmd(nic, cuc_nop, 0);
+                if(e100_exec_cmd(nic, cuc_nop, 0))
+                        DPRINTK(TX_ERR, DEBUG, "exec cuc_nop failed\n");
                 udelay(1);
         }
 
@@ -1415,12 +1440,18 @@ static int e100_alloc_cbs(struct nic *nic)
         return 0;
 }
 
-static inline void e100_start_receiver(struct nic *nic)
+static inline void e100_start_receiver(struct nic *nic, struct rx *rx)
 {
+        if(!nic->rxs) return;
+        if(RU_SUSPENDED != nic->ru_running) return;
+
+        /* handle init time starts */
+        if(!rx) rx = nic->rxs;
+
         /* (Re)start RU if suspended or idle and RFA is non-NULL */
-        if(!nic->ru_running && nic->rx_to_clean->skb) {
-                e100_exec_cmd(nic, ruc_start, nic->rx_to_clean->dma_addr);
-                nic->ru_running = 1;
+        if(rx->skb) {
+                e100_exec_cmd(nic, ruc_start, rx->dma_addr);
+                nic->ru_running = RU_RUNNING;
         }
 }
 
@@ -1437,6 +1468,13 @@ static inline int e100_rx_alloc_skb(struct nic *nic, struct rx *rx)
         rx->dma_addr = pci_map_single(nic->pdev, rx->skb->data,
                 RFD_BUF_LEN, PCI_DMA_BIDIRECTIONAL);
 
+        if(pci_dma_mapping_error(rx->dma_addr)) {
+                dev_kfree_skb_any(rx->skb);
+                rx->skb = 0;
+                rx->dma_addr = 0;
+                return -ENOMEM;
+        }
+
         /* Link the RFD to end of RFA by linking previous RFD to
          * this one, and clearing EL bit of previous.  */
         if(rx->prev->skb) {
@@ -1471,7 +1509,7 @@ static inline int e100_rx_indicate(struct nic *nic, struct rx *rx,
 
         /* If data isn't ready, nothing to indicate */
         if(unlikely(!(rfd_status & cb_complete)))
-                return -EAGAIN;
+                return -ENODATA;
 
         /* Get actual data size */
         actual_size = le16_to_cpu(rfd->actual_size) & 0x3FFF;
@@ -1482,6 +1520,10 @@ static inline int e100_rx_indicate(struct nic *nic, struct rx *rx,
         pci_unmap_single(nic->pdev, rx->dma_addr,
                 RFD_BUF_LEN, PCI_DMA_FROMDEVICE);
 
+        /* this allows for a fast restart without re-enabling interrupts */
+        if(le16_to_cpu(rfd->command) & cb_el)
+                nic->ru_running = RU_SUSPENDED;
+
         /* Pull off the RFD and put the actual data (minus eth hdr) */
         skb_reserve(skb, sizeof(struct rfd));
         skb_put(skb, actual_size);
@@ -1514,20 +1556,45 @@ static inline void e100_rx_clean(struct nic *nic, unsigned int *work_done,
         unsigned int work_to_do)
 {
         struct rx *rx;
+        int restart_required = 0;
+        struct rx *rx_to_start = NULL;
+
+        /* are we already rnr? then pay attention!!! this ensures that
+         * the state machine progression never allows a start with a 
+         * partially cleaned list, avoiding a race between hardware
+         * and rx_to_clean when in NAPI mode */
+        if(RU_SUSPENDED == nic->ru_running)
+                restart_required = 1;
 
         /* Indicate newly arrived packets */
         for(rx = nic->rx_to_clean; rx->skb; rx = nic->rx_to_clean = rx->next) {
-                if(e100_rx_indicate(nic, rx, work_done, work_to_do))
+                int err = e100_rx_indicate(nic, rx, work_done, work_to_do);
+                if(-EAGAIN == err) {
+                        /* hit quota so have more work to do, restart once
+                         * cleanup is complete */
+                        restart_required = 0;
+                        break;
+                } else if(-ENODATA == err)
                         break; /* No more to clean */
         }
 
+        /* save our starting point as the place we'll restart the receiver */
+        if(restart_required)
+                rx_to_start = nic->rx_to_clean;
+
         /* Alloc new skbs to refill list */
         for(rx = nic->rx_to_use; !rx->skb; rx = nic->rx_to_use = rx->next) {
                 if(unlikely(e100_rx_alloc_skb(nic, rx)))
                         break; /* Better luck next time (see watchdog) */
         }
 
-        e100_start_receiver(nic);
+        if(restart_required) {
+                // ack the rnr?
+                writeb(stat_ack_rnr, &nic->csr->scb.stat_ack);
+                e100_start_receiver(nic, rx_to_start);
+                if(work_done)
+                        (*work_done)++;
+        }
 }
 
 static void e100_rx_clean_list(struct nic *nic)
@@ -1535,6 +1602,8 @@ static void e100_rx_clean_list(struct nic *nic)
         struct rx *rx;
         unsigned int i, count = nic->params.rfds.count;
 
+        nic->ru_running = RU_UNINITIALIZED;
+
         if(nic->rxs) {
                 for(rx = nic->rxs, i = 0; i < count; rx++, i++) {
                         if(rx->skb) {
@@ -1548,7 +1617,6 @@ static void e100_rx_clean_list(struct nic *nic)
         }
 
         nic->rx_to_use = nic->rx_to_clean = NULL;
-        nic->ru_running = 0;
 }
 
 static int e100_rx_alloc_list(struct nic *nic)
@@ -1557,6 +1625,7 @@ static int e100_rx_alloc_list(struct nic *nic)
         unsigned int i, count = nic->params.rfds.count;
 
         nic->rx_to_use = nic->rx_to_clean = NULL;
+        nic->ru_running = RU_UNINITIALIZED;
 
         if(!(nic->rxs = kmalloc(sizeof(struct rx) * count, GFP_ATOMIC)))
                 return -ENOMEM;
@@ -1572,6 +1641,7 @@ static int e100_rx_alloc_list(struct nic *nic)
         }
 
         nic->rx_to_use = nic->rx_to_clean = nic->rxs;
+        nic->ru_running = RU_SUSPENDED;
 
         return 0;
 }
@@ -1593,7 +1663,7 @@ static irqreturn_t e100_intr(int irq, void *dev_id, struct pt_regs *regs)
 
         /* We hit Receive No Resource (RNR); restart RU after cleaning */
         if(stat_ack & stat_ack_rnr)
-                nic->ru_running = 0;
+                nic->ru_running = RU_SUSPENDED;
 
         e100_disable_irq(nic);
         netif_rx_schedule(netdev);
@@ -1663,6 +1733,7 @@ static int e100_change_mtu(struct net_device *netdev, int new_mtu)
         return 0;
 }
 
+#ifdef CONFIG_PM
 static int e100_asf(struct nic *nic)
 {
         /* ASF can be enabled from eeprom */
@@ -1671,6 +1742,7 @@ static int e100_asf(struct nic *nic)
            !(nic->eeprom[eeprom_config_asf] & eeprom_gcl) &&
            ((nic->eeprom[eeprom_smbus_addr] & 0xFF) != 0xFE));
 }
+#endif
 
 static int e100_up(struct nic *nic)
 {
@@ -1683,13 +1755,16 @@ static int e100_up(struct nic *nic)
         if((err = e100_hw_init(nic)))
                 goto err_clean_cbs;
         e100_set_multicast_list(nic->netdev);
-        e100_start_receiver(nic);
+        e100_start_receiver(nic, 0);
         mod_timer(&nic->watchdog, jiffies);
         if((err = request_irq(nic->pdev->irq, e100_intr, SA_SHIRQ,
                 nic->netdev->name, nic->netdev)))
                 goto err_no_irq;
-        e100_enable_irq(nic);
         netif_wake_queue(nic->netdev);
+        netif_poll_enable(nic->netdev);
+        /* enable ints _after_ enabling poll, preventing a race between
+         * disable ints+schedule */
+        e100_enable_irq(nic);
         return 0;
 
 err_no_irq:
@@ -1703,11 +1778,13 @@ err_rx_clean_list:
 
 static void e100_down(struct nic *nic)
 {
+        /* wait here for poll to complete */
+        netif_poll_disable(nic->netdev);
+        netif_stop_queue(nic->netdev);
         e100_hw_reset(nic);
         free_irq(nic->pdev->irq, nic->netdev);
         del_timer_sync(&nic->watchdog);
         netif_carrier_off(nic->netdev);
-        netif_stop_queue(nic->netdev);
         e100_clean_cbs(nic);
         e100_rx_clean_list(nic);
 }
@@ -1716,6 +1793,15 @@ static void e100_tx_timeout(struct net_device *netdev)
 {
         struct nic *nic = netdev_priv(netdev);
 
+        /* Reset outside of interrupt context, to avoid request_irq 
+         * in interrupt context */
+        schedule_work(&nic->tx_timeout_task);
+}
+
+static void e100_tx_timeout_task(struct net_device *netdev)
+{
+        struct nic *nic = netdev_priv(netdev);
+
         DPRINTK(TX_ERR, DEBUG, "scb.status=0x%02X\n",
                 readb(&nic->csr->scb.status));
         e100_down(netdev_priv(netdev));
@@ -1749,7 +1835,7 @@ static int e100_loopback_test(struct nic *nic, enum loopback loopback_mode)
                 mdio_write(nic->netdev, nic->mii.phy_id, MII_BMCR,
                         BMCR_LOOPBACK);
 
-        e100_start_receiver(nic);
+        e100_start_receiver(nic, 0);
 
         if(!(skb = dev_alloc_skb(ETH_DATA_LEN))) {
                 err = -ENOMEM;
@@ -1869,7 +1955,6 @@ static int e100_set_wol(struct net_device *netdev, struct ethtool_wolinfo *wol)
         else
                 nic->flags &= ~wol_magic;
 
-        pci_enable_wake(nic->pdev, 0, nic->flags & (wol_magic | e100_asf(nic)));
         e100_exec_cb(nic, NULL, e100_configure);
 
         return 0;
@@ -2223,6 +2308,7 @@ static int __devinit e100_probe(struct pci_dev *pdev,
 
         e100_get_defaults(nic);
 
+        /* locks must be initialized before calling hw_reset */
         spin_lock_init(&nic->cb_lock);
         spin_lock_init(&nic->cmd_lock);
 
@@ -2240,6 +2326,9 @@ static int __devinit e100_probe(struct pci_dev *pdev,
         nic->blink_timer.function = e100_blink_led;
         nic->blink_timer.data = (unsigned long)nic;
 
+        INIT_WORK(&nic->tx_timeout_task,
+                (void (*)(void *))e100_tx_timeout_task, netdev);
+
         if((err = e100_alloc(nic))) {
                 DPRINTK(PROBE, ERR, "Cannot alloc driver memory, aborting.\n");
                 goto err_out_iounmap;
@@ -2263,7 +2352,8 @@ static int __devinit e100_probe(struct pci_dev *pdev,
            (nic->eeprom[eeprom_id] & eeprom_id_wol))
                 nic->flags |= wol_magic;
 
-        pci_enable_wake(pdev, 0, nic->flags & (wol_magic | e100_asf(nic)));
+        /* ack any pending wake events, disable PME */
+        pci_enable_wake(pdev, 0, 0);
 
         strcpy(netdev->name, "eth%d");
         if((err = register_netdev(netdev))) {
@@ -2335,7 +2425,10 @@ static int e100_resume(struct pci_dev *pdev)
 
         pci_set_power_state(pdev, PCI_D0);
         pci_restore_state(pdev);
-        e100_hw_init(nic);
+        /* ack any pending wake events, disable PME */
+        pci_enable_wake(pdev, 0, 0);
+        if(e100_hw_init(nic))
+                DPRINTK(HW, ERR, "e100_hw_init failed\n");
 
         netif_device_attach(netdev);
         if(netif_running(netdev))
@@ -2345,6 +2438,21 @@ static int e100_resume(struct pci_dev *pdev)
 }
 #endif
 
+
+static void e100_shutdown(struct device *dev)
+{
+        struct pci_dev *pdev = container_of(dev, struct pci_dev, dev);
+        struct net_device *netdev = pci_get_drvdata(pdev);
+        struct nic *nic = netdev_priv(netdev);
+
+#ifdef CONFIG_PM
+        pci_enable_wake(pdev, 0, nic->flags & (wol_magic | e100_asf(nic)));
+#else
+        pci_enable_wake(pdev, 0, nic->flags & (wol_magic));
+#endif
+}
+
+
 static struct pci_driver e100_driver = {
         .name =         DRV_NAME,
         .id_table =     e100_id_table,
@@ -2354,6 +2462,11 @@ static struct pci_driver e100_driver = {
         .suspend =      e100_suspend,
         .resume =       e100_resume,
 #endif
+
+        .driver = {
+                .shutdown = e100_shutdown,
+        }
+
 };
 
 static int __init e100_init_module(void)