From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Miller Subject: Re: [PATCH 2/8 2.6.28] cxgb3 - Allocate multiqueues at init time Date: Wed, 08 Oct 2008 17:42:02 -0700 (PDT) Message-ID: <20081008.174202.160320028.davem@davemloft.net> References: <20081009003402.13315.18192.stgit@speedy5> Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Cc: jeff@garzik.org, netdev@vger.kernel.org, linux-kernel@vger.kernel.org, swise@opengridcomputing.com To: divy@chelsio.com Return-path: Received: from 74-93-104-97-Washington.hfc.comcastbusiness.net ([74.93.104.97]:32848 "EHLO sunset.davemloft.net" rhost-flags-OK-FAIL-OK-OK) by vger.kernel.org with ESMTP id S1755845AbYJIAm0 (ORCPT ); Wed, 8 Oct 2008 20:42:26 -0400 In-Reply-To: <20081009003402.13315.18192.stgit@speedy5> Sender: netdev-owner@vger.kernel.org List-ID: From: Divy Le Ray Date: Wed, 08 Oct 2008 17:34:02 -0700 > Allocate a queue set per core, up to the maximum of available qsets. > Share the queue sets on multi port adapters. > Rename MSI-X interrupt vectors ethX-N, N being the queue set number. > > Signed-off-by: Divy Le Ray Applied, but... This "num processors" logic and policy belongs in some generic place. Perhaps "num_possible_cpus()" is more appropriate, who knows right? Otherwise every stinking driver is going to implement this in their own way and use their own policy, and from a user experience perspective that totally sucks.