From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751960Ab1AEKX7 (ORCPT ); Wed, 5 Jan 2011 05:23:59 -0500 Received: from canuck.infradead.org ([134.117.69.58]:50025 "EHLO canuck.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751398Ab1AEKX6 convert rfc822-to-8bit (ORCPT ); Wed, 5 Jan 2011 05:23:58 -0500 Subject: Re: [RFC -v3 PATCH 2/3] sched: add yield_to function From: Peter Zijlstra To: KOSAKI Motohiro Cc: Avi Kivity , Rik van Riel , kvm@vger.kernel.org, linux-kernel@vger.kernel.org, Srivatsa Vaddagiri , Mike Galbraith , Chris Wright In-Reply-To: <20110105173823.B658.A69D9226@jp.fujitsu.com> References: <20110105110837.B62A.A69D9226@jp.fujitsu.com> <4D242D60.9060301@redhat.com> <20110105173823.B658.A69D9226@jp.fujitsu.com> Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 8BIT Date: Wed, 05 Jan 2011 11:24:14 +0100 Message-ID: <1294223054.2016.235.camel@laptop> Mime-Version: 1.0 X-Mailer: Evolution 2.30.3 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wed, 2011-01-05 at 17:40 +0900, KOSAKI Motohiro wrote: > > > If you are interesting GIL mess and issue, please feel free to ask more. > > > > I suggest looking into an explicit round-robin scheme, where each thread > > adds itself to a queue and an unlock wakes up the first waiter. > > I'm sure you haven't try your scheme. but I did. It's slow. Of course it is, but then your locking scheme (GIL) is the problem, not the locking constructs.