| To: | Eric Lemoine <eric.lemoine@xxxxxxxxx> |
|---|---|
| Subject: | Re: [openib-general] Re: LLTX and netif_stop_queue |
| From: | Grant Grundler <iod00d@xxxxxx> |
| Date: | Mon, 3 Jan 2005 09:12:27 -0800 |
| Cc: | Roland Dreier <roland@xxxxxxxxxxx>, netdev@xxxxxxxxxxx, hadi@xxxxxxxxxx, Andi Kleen <ak@xxxxxxx>, openib-general@xxxxxxxxxx, Patrick McHardy <kaber@xxxxxxxxx>, "David S. Miller" <davem@xxxxxxxxxxxxx> |
| In-reply-to: | <5cac192f0501030907c755135@xxxxxxxxxxxxxx> |
| References: | <5cac192f0412230110628749e3@xxxxxxxxxxxxxx> <41CAF444.3000305@xxxxxxxxx> <5cac192f04122408102129af43@xxxxxxxxxxxxxx> <1104240717.1100.66.camel@xxxxxxxxxxxxxxxx> <5cac192f0501021530672a908a@xxxxxxxxxxxxxx> <1104764660.1048.578.camel@xxxxxxxxxxxxxxxx> <52brc68q05.fsf@xxxxxxxxxxx> <5cac192f05010308414a25b548@xxxxxxxxxxxxxx> <527jmu8nbw.fsf@xxxxxxxxxxx> <5cac192f0501030907c755135@xxxxxxxxxxxxxx> |
| Sender: | netdev-bounce@xxxxxxxxxxx |
| User-agent: | Mutt/1.5.6+20040907i |
On Mon, Jan 03, 2005 at 06:07:00PM +0100, Eric Lemoine wrote: > If I understand correctly, LLTX aims at avoiding cache misses on lock > variables (because of cacheline bouncing). So the effect of LLTX > should increase as the number of CPUs not sharing the same cache > increases. And two CPUs might not be enough... Cacheline bouncing usually starts to show up with > 4 CPUs (ie 8 or more). Some workloads that Jamal cares about (routing) only need 2 cpus. grant |
| Previous by Date: | Re: LLTX and netif_stop_queue, Eric Lemoine |
|---|---|
| Next by Date: | Sneak preview: IPv6-IPsec now described in Linux+IPv6-HOWTO, Peter Bieringer |
| Previous by Thread: | Re: LLTX and netif_stop_queue, Eric Lemoine |
| Next by Thread: | Re: [openib-general] Re: LLTX and netif_stop_queue, jamal |
| Indexes: | [Date] [Thread] [Top] [All Lists] |