| To: | Henner Eisen <eis@xxxxxxxxxxxxx> |
|---|---|
| Subject: | Re: Q: sock output serialization |
| From: | "Andi Kleen" <ak@xxxxxxx> |
| Date: | Sun, 17 Sep 2000 01:54:35 +0200 |
| Cc: | alan@xxxxxxxxxxxxxxxxxxx, davem@xxxxxxxxxx, kuznet@xxxxxxxxxxxxx, linux-kernel@xxxxxxxxxxxxxxx, netdev@xxxxxxxxxxx |
| In-reply-to: | <200009162139.XAA01092@xxxxxxxxxxxxx>; from eis@xxxxxxxxxxxxx on Sat, Sep 16, 2000 at 11:39:45PM +0200 |
| References: | <E13a3yj-0008E6-00@xxxxxxxxxxxxxxxxx> <200009162139.XAA01092@xxxxxxxxxxxxx> |
| Sender: | owner-netdev@xxxxxxxxxxx |
On Sat, Sep 16, 2000 at 11:39:45PM +0200, Henner Eisen wrote:
> int netif_would_drop(dev)
> {
> return (queue->input_pkt_queue.qlen > netdev_max_backlog)
> || ( (queue->input_pkt_queue.qlen) && (queue->throttle) )
> }
>
> would fulfil those requirements.
It would just be racy. You test, get a not drop and then another different
interrupt would deliver another packet before you can and fill the queue.
Jamal's extended netif_rx probably makes more sense, because it can be
atomic.
-Andi
|
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | Re: Q: sock output serialization, Alan Cox |
|---|---|
| Next by Date: | Re: Q: sock output serialization, Henner Eisen |
| Previous by Thread: | Re: Q: sock output serialization, Henner Eisen |
| Next by Thread: | Re: Q: sock output serialization, Henner Eisen |
| Indexes: | [Date] [Thread] [Top] [All Lists] |