| To: | ak@xxxxxxx |
|---|---|
| Subject: | Re: Early SPECWeb99 results on 2.5.33 with TSO on e1000 |
| From: | "David S. Miller" <davem@xxxxxxxxxx> |
| Date: | Fri, 06 Sep 2002 12:24:05 -0700 (PDT) |
| Cc: | niv@xxxxxxxxxx, linux-kernel@xxxxxxxxxxxxxxx, netdev@xxxxxxxxxxx |
| In-reply-to: | <20020906212619.A28172@wotan.suse.de> |
| References: | <20020906202646.A2185@wotan.suse.de> <1031339954.3d78ffb257d22@imap.linux.ibm.com> <20020906212619.A28172@wotan.suse.de> |
| Sender: | netdev-bounce@xxxxxxxxxxx |
From: Andi Kleen <ak@xxxxxxx> Date: Fri, 6 Sep 2002 21:26:19 +0200 I'm not entirely sure it is worth it in this case. The locks are probably the majority of the cost. You can more localize the lock accesses (since we use per-chain locks) by applying a cpu salt to the port numbers you allocate. See my other email. |
| Previous by Date: | Re: Early SPECWeb99 results on 2.5.33 with TSO on e1000, David S. Miller |
|---|---|
| Next by Date: | Re: Early SPECWeb99 results on 2.5.33 with TSO on e1000, Manfred Spraul |
| Previous by Thread: | Re: Early SPECWeb99 results on 2.5.33 with TSO on e1000, Andi Kleen |
| Next by Thread: | Re: Early SPECWeb99 results on 2.5.33 with TSO on e1000, David S. Miller |
| Indexes: | [Date] [Thread] [Top] [All Lists] |