From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from sunset.davemloft.net (unknown [74.93.104.97]) by ozlabs.org (Postfix) with ESMTP id 5A252DF4A9 for ; Wed, 29 Aug 2007 06:27:43 +1000 (EST) Date: Tue, 28 Aug 2007 13:27:41 -0700 (PDT) Message-Id: <20070828.132741.118957759.davem@davemloft.net> To: ossthema@de.ibm.com Subject: Re: RFC: issues concerning the next NAPI interface From: David Miller In-Reply-To: <200708281321.10679.ossthema@de.ibm.com> References: <46D2F301.7050105@katalix.com> <20070827.140251.95055210.davem@davemloft.net> <200708281321.10679.ossthema@de.ibm.com> Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Cc: tklein@de.ibm.com, themann@de.ibm.com, stefan.roscher@de.ibm.com, netdev@vger.kernel.org, jchapman@katalix.com, raisch@de.ibm.com, linux-kernel@vger.kernel.org, linuxppc-dev@ozlabs.org, akepner@sgi.com, meder@de.ibm.com, shemminger@linux-foundation.org List-Id: Linux on PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , From: Jan-Bernd Themann Date: Tue, 28 Aug 2007 13:21:09 +0200 > So I guess one solution is to "force" an HW interrupt when two many > RQs are processed on the same CPU (when no IRQ pinning is > used). This is something the driver has to handle. No, the solution is to lock the interrupts onto one specific processor and don't move it around. That's what's causing all of the problems. And you can enforce this policy now in the driver even if just for testing by calling the set_affinity() interfaces on the interrupts your driver has. You can even walk over the cpu_online_map and choose a load distribution of your liking.