diff options
| author | Eric Dumazet <eric.dumazet@gmail.com> | 2010-05-10 05:01:31 -0700 |
|---|---|---|
| committer | David S. Miller <davem@davemloft.net> | 2010-05-10 05:01:31 -0700 |
| commit | 1ae5dc342ac78d7a42965fd1f323815f6f5ef2c1 (patch) | |
| tree | d1955a7639e99832590df26466a34d5786a880ae /drivers/net/pcnet32.c | |
| parent | 2b0b05ddc04b6d45e71cd36405df512075786f1e (diff) | |
| download | linux-1ae5dc342ac78d7a42965fd1f323815f6f5ef2c1.tar.gz | |
net: trans_start cleanups
Now that core network takes care of trans_start updates, dont do it in drivers themselves, if possible. Drivers can avoid one cache miss (on dev->trans_start) in their start_xmit() handler. Exceptions are NETIF_F_LLTX drivers Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'drivers/net/pcnet32.c')
| -rw-r--r-- | drivers/net/pcnet32.c | 6 |
1 files changed, 2 insertions, 4 deletions
diff --git a/drivers/net/pcnet32.c b/drivers/net/pcnet32.c index a2254f749a9a..566fd89da861 100644 --- a/drivers/net/pcnet32.c +++ b/drivers/net/pcnet32.c @@ -448,7 +448,7 @@ static void pcnet32_netif_stop(struct net_device *dev) { struct pcnet32_private *lp = netdev_priv(dev); - dev->trans_start = jiffies; + dev->trans_start = jiffies; /* prevent tx timeout */ napi_disable(&lp->napi); netif_tx_disable(dev); } @@ -2398,7 +2398,7 @@ static void pcnet32_tx_timeout(struct net_device *dev) } pcnet32_restart(dev, CSR0_NORMAL); - dev->trans_start = jiffies; + dev->trans_start = jiffies; /* prevent tx timeout */ netif_wake_queue(dev); spin_unlock_irqrestore(&lp->lock, flags); @@ -2449,8 +2449,6 @@ static netdev_tx_t pcnet32_start_xmit(struct sk_buff *skb, /* Trigger an immediate send poll. */ lp->a.write_csr(ioaddr, CSR0, CSR0_INTEN | CSR0_TXPOLL); - dev->trans_start = jiffies; - if (lp->tx_ring[(entry + 1) & lp->tx_mod_mask].base != 0) { lp->tx_full = 1; netif_stop_queue(dev); |