From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ido Schimmel Subject: Re: [net-next 01/10] igb: reduce CPU0 latency when updating statistics Date: Tue, 25 Dec 2018 17:47:15 +0200 Message-ID: <20181225154715.GA20609@splinter.mtl.com> References: <20181220205634.8509-1-jeffrey.t.kirsher@intel.com> <20181220205634.8509-2-jeffrey.t.kirsher@intel.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: davem@davemloft.net, netdev@vger.kernel.org, nhorman@redhat.com, sassmann@redhat.com, Bernhard Kaindl , Jan Jablonsky To: Jeff Kirsher Return-path: Received: from out1-smtp.messagingengine.com ([66.111.4.25]:35449 "EHLO out1-smtp.messagingengine.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1725822AbeLYPrW (ORCPT ); Tue, 25 Dec 2018 10:47:22 -0500 Content-Disposition: inline In-Reply-To: <20181220205634.8509-2-jeffrey.t.kirsher@intel.com> Sender: netdev-owner@vger.kernel.org List-ID: On Thu, Dec 20, 2018 at 12:56:25PM -0800, Jeff Kirsher wrote: > This change is based off of the work and suggestion of Jan Jablonsky > . > > The Watchdog workqueue in igb driver is scheduled every 2s for each > network interface. That includes updating a statistics protected by > spinlock. Function igb_update_stats in this case will be protected > against preemption. According to number of a statistics registers > (cca 60), processing this function might cause additional cpu load > on CPU0. > > In case of statistics spinlock may be replaced with mutex, which > reduce latency on CPU0. > > CC: Bernhard Kaindl > CC: Jan Jablonsky > Signed-off-by: Jeff Kirsher > Tested-by: Aaron Brown > Signed-off-by: Jeff Kirsher Hi, I'm hitting following issue with this commit: https://lists.01.org/pipermail/lkp/2018-September/009139.html