| To: | kuznet@xxxxxxxxxxxxx |
|---|---|
| Subject: | Re: [Fwd: Re: possible bug x86 2.4.2 SMP in IP receive stack] |
| From: | Bob Felderman <feldy@xxxxxxxx> |
| Date: | Fri, 9 Mar 2001 12:48:26 -0800 (PST) |
| Cc: | "David S. Miller" <davem@xxxxxxxxxx>, hadi@xxxxxxxxxx, ak@xxxxxx, andrewm@xxxxxxxxxx, netdev@xxxxxxxxxxx, Bob Felderman <feldy@xxxxxxxx>, pp@xxxxxxxxxxxxxx |
| In-reply-to: | <200103092035.XAA27248@xxxxxxxxxxxxx> |
| Sender: | owner-netdev@xxxxxxxxxxx |
On Fri, 9 Mar 2001 kuznet@xxxxxxxxxxxxx wrote: > Hello! > > > He's using fragmented UDP packets, so the lazy receive defrag stuff > > (ie. 2 copies turn into 1) work wonders. > > _No_ copy on softirq. It is more important. > > It is the reason why all the tricks with backlog are not essential. Can you explain this a bit more? Are you saying that a softirq will reduce the number of copies? Should the driver be calling softirq to deliver the packets using netif_rx? |
| Previous by Date: | Re: [Fwd: Re: possible bug x86 2.4.2 SMP in IP receive stack], kuznet |
|---|---|
| Next by Date: | Re: [Fwd: Re: possible bug x86 2.4.2 SMP in IP receive stack], David S. Miller |
| Previous by Thread: | Re: [Fwd: Re: possible bug x86 2.4.2 SMP in IP receive stack], kuznet |
| Next by Thread: | Re: [Fwd: Re: possible bug x86 2.4.2 SMP in IP receive stack], David S. Miller |
| Indexes: | [Date] [Thread] [Top] [All Lists] |