cpmac bugfixes and enhancements
authorMatteo Croce <matteo@openwrt.org>
Tue, 13 May 2008 22:58:32 +0000 (00:58 +0200)
committerJeff Garzik <jgarzik@redhat.com>
Thu, 22 May 2008 10:20:02 +0000 (06:20 -0400)
* Resolve some locking issues using atomic_inc/atomic_dec
* move status code in cpmac_check_status
* unmark the BROKEN flag in Kconfig
* move code which should have been in platform code in
  arch/mips/ar7/platform.c
* fixed an IRQ storm which lets the kernel hang
* fixed a double call to netif_start_queue which causes a kernel panic
* don't fail to register the PHY, works on many devices now

Signed-off-by: Matteo Croce <matteo@openwrt.org>
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
drivers/net/cpmac.c

index 2b5740b3d1825dee5b91f9b3862060e71c266f8b..7f3f62e1b1134f2c816770f61fcc4dd0d54139b7 100644 (file)
@@ -38,6 +38,7 @@
 #include <linux/platform_device.h>
 #include <linux/dma-mapping.h>
 #include <asm/gpio.h>
+#include <asm/atomic.h>
 
 MODULE_AUTHOR("Eugene Konev <ejka@imfi.kspu.ru>");
 MODULE_DESCRIPTION("TI AR7 ethernet driver (CPMAC)");
@@ -187,6 +188,7 @@ struct cpmac_desc {
 #define CPMAC_EOQ                      0x1000
        struct sk_buff *skb;
        struct cpmac_desc *next;
+       struct cpmac_desc *prev;
        dma_addr_t mapping;
        dma_addr_t data_mapping;
 };
@@ -208,6 +210,7 @@ struct cpmac_priv {
        struct work_struct reset_work;
        struct platform_device *pdev;
        struct napi_struct napi;
+       atomic_t reset_pending;
 };
 
 static irqreturn_t cpmac_irq(int, void *);
@@ -241,6 +244,16 @@ static void cpmac_dump_desc(struct net_device *dev, struct cpmac_desc *desc)
        printk("\n");
 }
 
+static void cpmac_dump_all_desc(struct net_device *dev)
+{
+       struct cpmac_priv *priv = netdev_priv(dev);
+       struct cpmac_desc *dump = priv->rx_head;
+       do {
+               cpmac_dump_desc(dev, dump);
+               dump = dump->next;
+       } while (dump != priv->rx_head);
+}
+
 static void cpmac_dump_skb(struct net_device *dev, struct sk_buff *skb)
 {
        int i;
@@ -412,21 +425,42 @@ static struct sk_buff *cpmac_rx_one(struct cpmac_priv *priv,
 static int cpmac_poll(struct napi_struct *napi, int budget)
 {
        struct sk_buff *skb;
-       struct cpmac_desc *desc;
-       int received = 0;
+       struct cpmac_desc *desc, *restart;
        struct cpmac_priv *priv = container_of(napi, struct cpmac_priv, napi);
+       int received = 0, processed = 0;
 
        spin_lock(&priv->rx_lock);
        if (unlikely(!priv->rx_head)) {
                if (netif_msg_rx_err(priv) && net_ratelimit())
                        printk(KERN_WARNING "%s: rx: polling, but no queue\n",
                               priv->dev->name);
+               spin_unlock(&priv->rx_lock);
                netif_rx_complete(priv->dev, napi);
                return 0;
        }
 
        desc = priv->rx_head;
+       restart = NULL;
        while (((desc->dataflags & CPMAC_OWN) == 0) && (received < budget)) {
+               processed++;
+
+               if ((desc->dataflags & CPMAC_EOQ) != 0) {
+                       /* The last update to eoq->hw_next didn't happen
+                       * soon enough, and the receiver stopped here.
+                       *Remember this descriptor so we can restart
+                       * the receiver after freeing some space.
+                       */
+                       if (unlikely(restart)) {
+                               if (netif_msg_rx_err(priv))
+                                       printk(KERN_ERR "%s: poll found a"
+                                               " duplicate EOQ: %p and %p\n",
+                                               priv->dev->name, restart, desc);
+                               goto fatal_error;
+                       }
+
+                       restart = desc->next;
+               }
+
                skb = cpmac_rx_one(priv, desc);
                if (likely(skb)) {
                        netif_receive_skb(skb);
@@ -435,19 +469,90 @@ static int cpmac_poll(struct napi_struct *napi, int budget)
                desc = desc->next;
        }
 
+       if (desc != priv->rx_head) {
+               /* We freed some buffers, but not the whole ring,
+                * add what we did free to the rx list */
+               desc->prev->hw_next = (u32)0;
+               priv->rx_head->prev->hw_next = priv->rx_head->mapping;
+       }
+
+       /* Optimization: If we did not actually process an EOQ (perhaps because
+        * of quota limits), check to see if the tail of the queue has EOQ set.
+       * We should immediately restart in that case so that the receiver can
+       * restart and run in parallel with more packet processing.
+       * This lets us handle slightly larger bursts before running
+       * out of ring space (assuming dev->weight < ring_size) */
+
+       if (!restart &&
+            (priv->rx_head->prev->dataflags & (CPMAC_OWN|CPMAC_EOQ))
+                   == CPMAC_EOQ &&
+            (priv->rx_head->dataflags & CPMAC_OWN) != 0) {
+               /* reset EOQ so the poll loop (above) doesn't try to
+               * restart this when it eventually gets to this descriptor.
+               */
+               priv->rx_head->prev->dataflags &= ~CPMAC_EOQ;
+               restart = priv->rx_head;
+       }
+
+       if (restart) {
+               priv->dev->stats.rx_errors++;
+               priv->dev->stats.rx_fifo_errors++;
+               if (netif_msg_rx_err(priv) && net_ratelimit())
+                       printk(KERN_WARNING "%s: rx dma ring overrun\n",
+                              priv->dev->name);
+
+               if (unlikely((restart->dataflags & CPMAC_OWN) == 0)) {
+                       if (netif_msg_drv(priv))
+                               printk(KERN_ERR "%s: cpmac_poll is trying to "
+                                       "restart rx from a descriptor that's "
+                                       "not free: %p\n",
+                                       priv->dev->name, restart);
+                               goto fatal_error;
+               }
+
+               cpmac_write(priv->regs, CPMAC_RX_PTR(0), restart->mapping);
+       }
+
        priv->rx_head = desc;
        spin_unlock(&priv->rx_lock);
        if (unlikely(netif_msg_rx_status(priv)))
                printk(KERN_DEBUG "%s: poll processed %d packets\n",
                       priv->dev->name, received);
-       if (desc->dataflags & CPMAC_OWN) {
+       if (processed == 0) {
+               /* we ran out of packets to read,
+                * revert to interrupt-driven mode */
                netif_rx_complete(priv->dev, napi);
-               cpmac_write(priv->regs, CPMAC_RX_PTR(0), (u32)desc->mapping);
                cpmac_write(priv->regs, CPMAC_RX_INT_ENABLE, 1);
                return 0;
        }
 
        return 1;
+
+fatal_error:
+       /* Something went horribly wrong.
+        * Reset hardware to try to recover rather than wedging. */
+
+       if (netif_msg_drv(priv)) {
+               printk(KERN_ERR "%s: cpmac_poll is confused. "
+                               "Resetting hardware\n", priv->dev->name);
+               cpmac_dump_all_desc(priv->dev);
+               printk(KERN_DEBUG "%s: RX_PTR(0)=0x%08x RX_ACK(0)=0x%08x\n",
+                       priv->dev->name,
+                       cpmac_read(priv->regs, CPMAC_RX_PTR(0)),
+                       cpmac_read(priv->regs, CPMAC_RX_ACK(0)));
+       }
+
+       spin_unlock(&priv->rx_lock);
+       netif_rx_complete(priv->dev, napi);
+       netif_stop_queue(priv->dev);
+       napi_disable(&priv->napi);
+
+       atomic_inc(&priv->reset_pending);
+       cpmac_hw_stop(priv->dev);
+       if (!schedule_work(&priv->reset_work))
+               atomic_dec(&priv->reset_pending);
+       return 0;
+
 }
 
 static int cpmac_start_xmit(struct sk_buff *skb, struct net_device *dev)
@@ -456,6 +561,9 @@ static int cpmac_start_xmit(struct sk_buff *skb, struct net_device *dev)
        struct cpmac_desc *desc;
        struct cpmac_priv *priv = netdev_priv(dev);
 
+       if (unlikely(atomic_read(&priv->reset_pending)))
+               return NETDEV_TX_BUSY;
+
        if (unlikely(skb_padto(skb, ETH_ZLEN)))
                return NETDEV_TX_OK;
 
@@ -621,8 +729,10 @@ static void cpmac_clear_rx(struct net_device *dev)
                        desc->dataflags = CPMAC_OWN;
                        dev->stats.rx_dropped++;
                }
+               desc->hw_next = desc->next->mapping;
                desc = desc->next;
        }
+       priv->rx_head->prev->hw_next = 0;
 }
 
 static void cpmac_clear_tx(struct net_device *dev)
@@ -635,14 +745,14 @@ static void cpmac_clear_tx(struct net_device *dev)
                priv->desc_ring[i].dataflags = 0;
                if (priv->desc_ring[i].skb) {
                        dev_kfree_skb_any(priv->desc_ring[i].skb);
-                       if (netif_subqueue_stopped(dev, i))
-                           netif_wake_subqueue(dev, i);
+                       priv->desc_ring[i].skb = NULL;
                }
        }
 }
 
 static void cpmac_hw_error(struct work_struct *work)
 {
+       int i;
        struct cpmac_priv *priv =
                container_of(work, struct cpmac_priv, reset_work);
 
@@ -651,8 +761,48 @@ static void cpmac_hw_error(struct work_struct *work)
        spin_unlock(&priv->rx_lock);
        cpmac_clear_tx(priv->dev);
        cpmac_hw_start(priv->dev);
-       napi_enable(&priv->napi);
-       netif_start_queue(priv->dev);
+       barrier();
+       atomic_dec(&priv->reset_pending);
+
+       for (i = 0; i < CPMAC_QUEUES; i++)
+               netif_wake_subqueue(priv->dev, i);
+       netif_wake_queue(priv->dev);
+       cpmac_write(priv->regs, CPMAC_MAC_INT_ENABLE, 3);
+}
+
+static void cpmac_check_status(struct net_device *dev)
+{
+       struct cpmac_priv *priv = netdev_priv(dev);
+
+       u32 macstatus = cpmac_read(priv->regs, CPMAC_MAC_STATUS);
+       int rx_channel = (macstatus >> 8) & 7;
+       int rx_code = (macstatus >> 12) & 15;
+       int tx_channel = (macstatus >> 16) & 7;
+       int tx_code = (macstatus >> 20) & 15;
+
+       if (rx_code || tx_code) {
+               if (netif_msg_drv(priv) && net_ratelimit()) {
+                       /* Can't find any documentation on what these
+                        *error codes actually are. So just log them and hope..
+                        */
+                       if (rx_code)
+                               printk(KERN_WARNING "%s: host error %d on rx "
+                                    "channel %d (macstatus %08x), resetting\n",
+                                    dev->name, rx_code, rx_channel, macstatus);
+                       if (tx_code)
+                               printk(KERN_WARNING "%s: host error %d on tx "
+                                    "channel %d (macstatus %08x), resetting\n",
+                                    dev->name, tx_code, tx_channel, macstatus);
+               }
+
+               netif_stop_queue(dev);
+               cpmac_hw_stop(dev);
+               if (schedule_work(&priv->reset_work))
+                       atomic_inc(&priv->reset_pending);
+               if (unlikely(netif_msg_hw(priv)))
+                       cpmac_dump_regs(dev);
+       }
+       cpmac_write(priv->regs, CPMAC_MAC_INT_CLEAR, 0xff);
 }
 
 static irqreturn_t cpmac_irq(int irq, void *dev_id)
@@ -683,49 +833,32 @@ static irqreturn_t cpmac_irq(int irq, void *dev_id)
 
        cpmac_write(priv->regs, CPMAC_MAC_EOI_VECTOR, 0);
 
-       if (unlikely(status & (MAC_INT_HOST | MAC_INT_STATUS))) {
-               if (netif_msg_drv(priv) && net_ratelimit())
-                       printk(KERN_ERR "%s: hw error, resetting...\n",
-                              dev->name);
-               netif_stop_queue(dev);
-               napi_disable(&priv->napi);
-               cpmac_hw_stop(dev);
-               schedule_work(&priv->reset_work);
-               if (unlikely(netif_msg_hw(priv)))
-                       cpmac_dump_regs(dev);
-       }
+       if (unlikely(status & (MAC_INT_HOST | MAC_INT_STATUS)))
+               cpmac_check_status(dev);
 
        return IRQ_HANDLED;
 }
 
 static void cpmac_tx_timeout(struct net_device *dev)
 {
-       struct cpmac_priv *priv = netdev_priv(dev);
        int i;
+       struct cpmac_priv *priv = netdev_priv(dev);
 
        spin_lock(&priv->lock);
        dev->stats.tx_errors++;
        spin_unlock(&priv->lock);
        if (netif_msg_tx_err(priv) && net_ratelimit())
                printk(KERN_WARNING "%s: transmit timeout\n", dev->name);
-       /* 
-        * FIXME: waking up random queue is not the best thing to
-        * do... on the other hand why we got here at all?
-        */
-#ifdef CONFIG_NETDEVICES_MULTIQUEUE
+
+       atomic_inc(&priv->reset_pending);
+       barrier();
+       cpmac_clear_tx(dev);
+       barrier();
+       atomic_dec(&priv->reset_pending);
+
+       netif_wake_queue(priv->dev);
        for (i = 0; i < CPMAC_QUEUES; i++)
-               if (priv->desc_ring[i].skb) {
-                       priv->desc_ring[i].dataflags = 0;
-                       dev_kfree_skb_any(priv->desc_ring[i].skb);
-                       netif_wake_subqueue(dev, i);
-                       break;
-               }
-#else
-       priv->desc_ring[0].dataflags = 0;
-       if (priv->desc_ring[0].skb)
-               dev_kfree_skb_any(priv->desc_ring[0].skb);
-       netif_wake_queue(dev);
-#endif
+               netif_wake_subqueue(dev, i);
 }
 
 static int cpmac_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
@@ -901,9 +1034,12 @@ static int cpmac_open(struct net_device *dev)
                desc->buflen = CPMAC_SKB_SIZE;
                desc->dataflags = CPMAC_OWN;
                desc->next = &priv->rx_head[(i + 1) % priv->ring_size];
+               desc->next->prev = desc;
                desc->hw_next = (u32)desc->next->mapping;
        }
 
+       priv->rx_head->prev->hw_next = (u32)0;
+
        if ((res = request_irq(dev->irq, cpmac_irq, IRQF_SHARED,
                               dev->name, dev))) {
                if (netif_msg_drv(priv))
@@ -912,6 +1048,7 @@ static int cpmac_open(struct net_device *dev)
                goto fail_irq;
        }
 
+       atomic_set(&priv->reset_pending, 0);
        INIT_WORK(&priv->reset_work, cpmac_hw_error);
        cpmac_hw_start(dev);
 
@@ -1007,21 +1144,10 @@ static int __devinit cpmac_probe(struct platform_device *pdev)
 
        if (phy_id == PHY_MAX_ADDR) {
                if (external_switch || dumb_switch) {
-                       struct fixed_phy_status status = {};
-
-                       /*
-                        * FIXME: this should be in the platform code!
-                        * Since there is not platform code at all (that is,
-                        * no mainline users of that driver), place it here
-                        * for now.
-                        */
-                       phy_id = 0;
-                       status.link = 1;
-                       status.duplex = 1;
-                       status.speed = 100;
-                       fixed_phy_add(PHY_POLL, phy_id, &status);
+                       mdio_bus_id = 0; /* fixed phys bus */
+                       phy_id = pdev->id;
                } else {
-                       printk(KERN_ERR "cpmac: no PHY present\n");
+                       dev_err(&pdev->dev, "no PHY present\n");
                        return -ENODEV;
                }
        }
@@ -1064,10 +1190,8 @@ static int __devinit cpmac_probe(struct platform_device *pdev)
        priv->msg_enable = netif_msg_init(debug_level, 0xff);
        memcpy(dev->dev_addr, pdata->dev_addr, sizeof(dev->dev_addr));
 
-       snprintf(priv->phy_name, BUS_ID_SIZE, PHY_ID_FMT, mdio_bus_id, phy_id);
-
-       priv->phy = phy_connect(dev, priv->phy_name, &cpmac_adjust_link, 0,
-                               PHY_INTERFACE_MODE_MII);
+       priv->phy = phy_connect(dev, cpmac_mii.phy_map[phy_id]->dev.bus_id,
+                               &cpmac_adjust_link, 0, PHY_INTERFACE_MODE_MII);
        if (IS_ERR(priv->phy)) {
                if (netif_msg_drv(priv))
                        printk(KERN_ERR "%s: Could not attach to PHY\n",