From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752296Ab1KNG2M (ORCPT ); Mon, 14 Nov 2011 01:28:12 -0500 Received: from mail-wy0-f174.google.com ([74.125.82.174]:39986 "EHLO mail-wy0-f174.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750993Ab1KNG2K (ORCPT ); Mon, 14 Nov 2011 01:28:10 -0500 Message-ID: <1321252086.17837.56.camel@edumazet-laptop> Subject: Re: [patch 0/7] timekeeping: Decouple xtime_lock from xtime seqcount From: Eric Dumazet To: Thomas Gleixner Cc: LKML , John Stultz , Richard Cochran Date: Mon, 14 Nov 2011 07:28:06 +0100 In-Reply-To: <20111113230921.728083714@linutronix.de> References: <20111113230921.728083714@linutronix.de> Content-Type: text/plain; charset="UTF-8" X-Mailer: Evolution 3.2.1- Content-Transfer-Encoding: 8bit Mime-Version: 1.0 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Le dimanche 13 novembre 2011 à 23:19 +0000, Thomas Gleixner a écrit : > xtime_lock which serializes the update of time relevant data > structures is held over a large code pathes which results in extended > seqlock contention times on the reader side. This series decouples the > lock from the seqcount and reduces the seqcount protected sections to > those which actually update the reader visible data. > > Thanks, Thanks a lot Thomas, I'll take a look on this serie ASAP.