| To: | Robert Olsson <Robert.Olsson@xxxxxxxxxxx> |
|---|---|
| Subject: | Re: Intel and TOE in the news |
| From: | jamal <hadi@xxxxxxxxxx> |
| Date: | 21 Feb 2005 10:50:51 -0500 |
| Cc: | Thomas Graf <tgraf@xxxxxxx>, Andi Kleen <ak@xxxxxx>, Leonid Grossman <leonid.grossman@xxxxxxxxxxxx>, "'rick jones'" <rick.jones2@xxxxxx>, netdev@xxxxxxxxxxx, "'Alex Aizman'" <alex@xxxxxxxxxxxx> |
| In-reply-to: | <16922.89.271972.745365@xxxxxxxxxxxx> |
| Organization: | jamalopolous |
| References: | <20050220230713.GA62354@xxxxxx> <200502210332.j1L3WkDD014744@xxxxxxxxxxxxxxxxx> <20050221115006.GB87576@xxxxxx> <20050221132844.GU31837@xxxxxxxxxxxxxx> <1108994621.1089.158.camel@xxxxxxxxxxxxxxxx> <20050221141714.GV31837@xxxxxxxxxxxxxx> <1108996313.1090.178.camel@xxxxxxxxxxxxxxxx> <16922.89.271972.745365@xxxxxxxxxxxx> |
| Reply-to: | hadi@xxxxxxxxxx |
| Sender: | netdev-bounce@xxxxxxxxxxx |
On Mon, 2005-02-21 at 10:38, Robert Olsson wrote: > Yes when queue length/batch increases you're risking to load the L2 > twice for the same skb. Which is the most expensive operation.... You are probably onto something here. I have to rerun with varying batch sizes to see if this is true (although i did i think try with even 2 and still saw performance degradation). I will be revisiting these patches in about a month. cheers, jamal |
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | Re: Intel and TOE in the news, jamal |
|---|---|
| Next by Date: | Re: [PATCH] pktgen: reduce stack usage, Randy.Dunlap |
| Previous by Thread: | Re: Intel and TOE in the news, Robert Olsson |
| Next by Thread: | RE: Intel and TOE in the news, jamal |
| Indexes: | [Date] [Thread] [Top] [All Lists] |