提交 0685c31b 编写于 作者: M Malli Chilakala 提交者: Jeff Garzik

[PATCH] e100: NAPI performance enhancements

NAPI performance enhancements - Fixed issues with shared interrupts and NAPI resulting in bad performance.
Signed-off-by: NMallikarjuna R Chilakala <mallikarjuna.chilakala@intel.com>
Signed-off-by: NGanesh Venkatesan <ganesh.venkatesan@intel.com>
Signed-off-by: NJohn Ronciak <john.ronciak@intel.com>
上级 f92d8728
...@@ -1671,8 +1671,10 @@ static irqreturn_t e100_intr(int irq, void *dev_id, struct pt_regs *regs) ...@@ -1671,8 +1671,10 @@ static irqreturn_t e100_intr(int irq, void *dev_id, struct pt_regs *regs)
if(stat_ack & stat_ack_rnr) if(stat_ack & stat_ack_rnr)
nic->ru_running = RU_SUSPENDED; nic->ru_running = RU_SUSPENDED;
e100_disable_irq(nic); if(likely(netif_rx_schedule_prep(netdev))) {
netif_rx_schedule(netdev); e100_disable_irq(nic);
__netif_rx_schedule(netdev);
}
return IRQ_HANDLED; return IRQ_HANDLED;
} }
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册