Commit 6006f7f5 authored by Sergei Shtylyov's avatar Sergei Shtylyov Committed by Jeff Garzik

natsemi: netpoll fixes

Fix two issues in this driver's netpoll path: one usual, with spin_unlock_irq()
enabling interrupts which nobody asks it to do (that has been fixed recently in
a number of drivers) and one unusual, with poll_controller() method possibly
causing loss of interrupts due to the interrupt status register being cleared
by a simple read and the interrpupt handler simply storing it, not accumulating.
Signed-off-by: default avatarSergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: default avatarJeff Garzik <jeff@garzik.org>
parent a816c7c7
...@@ -2024,6 +2024,7 @@ static int start_tx(struct sk_buff *skb, struct net_device *dev) ...@@ -2024,6 +2024,7 @@ static int start_tx(struct sk_buff *skb, struct net_device *dev)
struct netdev_private *np = netdev_priv(dev); struct netdev_private *np = netdev_priv(dev);
void __iomem * ioaddr = ns_ioaddr(dev); void __iomem * ioaddr = ns_ioaddr(dev);
unsigned entry; unsigned entry;
unsigned long flags;
/* Note: Ordering is important here, set the field with the /* Note: Ordering is important here, set the field with the
"ownership" bit last, and only then increment cur_tx. */ "ownership" bit last, and only then increment cur_tx. */
...@@ -2037,7 +2038,7 @@ static int start_tx(struct sk_buff *skb, struct net_device *dev) ...@@ -2037,7 +2038,7 @@ static int start_tx(struct sk_buff *skb, struct net_device *dev)
np->tx_ring[entry].addr = cpu_to_le32(np->tx_dma[entry]); np->tx_ring[entry].addr = cpu_to_le32(np->tx_dma[entry]);
spin_lock_irq(&np->lock); spin_lock_irqsave(&np->lock, flags);
if (!np->hands_off) { if (!np->hands_off) {
np->tx_ring[entry].cmd_status = cpu_to_le32(DescOwn | skb->len); np->tx_ring[entry].cmd_status = cpu_to_le32(DescOwn | skb->len);
...@@ -2056,7 +2057,7 @@ static int start_tx(struct sk_buff *skb, struct net_device *dev) ...@@ -2056,7 +2057,7 @@ static int start_tx(struct sk_buff *skb, struct net_device *dev)
dev_kfree_skb_irq(skb); dev_kfree_skb_irq(skb);
np->stats.tx_dropped++; np->stats.tx_dropped++;
} }
spin_unlock_irq(&np->lock); spin_unlock_irqrestore(&np->lock, flags);
dev->trans_start = jiffies; dev->trans_start = jiffies;
...@@ -2222,6 +2223,8 @@ static void netdev_rx(struct net_device *dev, int *work_done, int work_to_do) ...@@ -2222,6 +2223,8 @@ static void netdev_rx(struct net_device *dev, int *work_done, int work_to_do)
pkt_len = (desc_status & DescSizeMask) - 4; pkt_len = (desc_status & DescSizeMask) - 4;
if ((desc_status&(DescMore|DescPktOK|DescRxLong)) != DescPktOK){ if ((desc_status&(DescMore|DescPktOK|DescRxLong)) != DescPktOK){
if (desc_status & DescMore) { if (desc_status & DescMore) {
unsigned long flags;
if (netif_msg_rx_err(np)) if (netif_msg_rx_err(np))
printk(KERN_WARNING printk(KERN_WARNING
"%s: Oversized(?) Ethernet " "%s: Oversized(?) Ethernet "
...@@ -2236,12 +2239,12 @@ static void netdev_rx(struct net_device *dev, int *work_done, int work_to_do) ...@@ -2236,12 +2239,12 @@ static void netdev_rx(struct net_device *dev, int *work_done, int work_to_do)
* reset procedure documented in * reset procedure documented in
* AN-1287. */ * AN-1287. */
spin_lock_irq(&np->lock); spin_lock_irqsave(&np->lock, flags);
reset_rx(dev); reset_rx(dev);
reinit_rx(dev); reinit_rx(dev);
writel(np->ring_dma, ioaddr + RxRingPtr); writel(np->ring_dma, ioaddr + RxRingPtr);
check_link(dev); check_link(dev);
spin_unlock_irq(&np->lock); spin_unlock_irqrestore(&np->lock, flags);
/* We'll enable RX on exit from this /* We'll enable RX on exit from this
* function. */ * function. */
...@@ -2396,8 +2399,19 @@ static struct net_device_stats *get_stats(struct net_device *dev) ...@@ -2396,8 +2399,19 @@ static struct net_device_stats *get_stats(struct net_device *dev)
#ifdef CONFIG_NET_POLL_CONTROLLER #ifdef CONFIG_NET_POLL_CONTROLLER
static void natsemi_poll_controller(struct net_device *dev) static void natsemi_poll_controller(struct net_device *dev)
{ {
struct netdev_private *np = netdev_priv(dev);
disable_irq(dev->irq); disable_irq(dev->irq);
/*
* A real interrupt might have already reached us at this point
* but NAPI might still haven't called us back. As the interrupt
* status register is cleared by reading, we should prevent an
* interrupt loss in this case...
*/
if (!np->intr_status)
intr_handler(dev->irq, dev); intr_handler(dev->irq, dev);
enable_irq(dev->irq); enable_irq(dev->irq);
} }
#endif #endif
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment