| To: | Patrick McHardy <kaber@xxxxxxxxx> |
|---|---|
| Subject: | Re: SFQ: Reordering? |
| From: | Andy Furniss <andy.furniss@xxxxxxxxxxxxx> |
| Date: | Tue, 10 May 2005 00:14:47 +0100 |
| Cc: | Thomas Graf <tgraf@xxxxxxx>, Asim Shankar <asimshankar@xxxxxxxxx>, netdev@xxxxxxxxxxx |
| In-reply-to: | <427C19C3.5030304@xxxxxxxxx> |
| References: | <7bca1cb5050506145344d16b1e@xxxxxxxxxxxxxx> <427BEAAE.409@xxxxxxxxx> <427BF3C4.1030105@xxxxxxxxx> <20050506230203.GI28419@xxxxxxxxxxxxxx> <427BFB72.7080407@xxxxxxxxx> <20050507005851.GK28419@xxxxxxxxxxxxxx> <427C19C3.5030304@xxxxxxxxx> |
| Sender: | netdev-bounce@xxxxxxxxxxx |
| User-agent: | Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.8b) Gecko/20050217 |
Patrick McHardy wrote: It would be interesting to see this in real-life, when a single flow is hashed to multiple buckets (it can be even more than two) and each bucket has some packets queued, the result should look pretty chaotic. It certainly hurts latency if used for ingress shaping, perturb nearly always makes the sender backoff and then burst. Andy. |
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | Re: [TG3]: Add tagged status support, David S. Miller |
|---|---|
| Next by Date: | [RFC] TCP congestion infrastructure, Stephen Hemminger |
| Previous by Thread: | Re: SFQ: Reordering?, Thomas Graf |
| Next by Thread: | Re: 3c509 module and 2.6 kernel: not all NICs are recognized?, Michael Tokarev |
| Indexes: | [Date] [Thread] [Top] [All Lists] |