From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S2992703AbXDDHQM (ORCPT ); Wed, 4 Apr 2007 03:16:12 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S2992709AbXDDHQM (ORCPT ); Wed, 4 Apr 2007 03:16:12 -0400 Received: from an-out-0708.google.com ([209.85.132.247]:47631 "EHLO an-out-0708.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S2992703AbXDDHQL (ORCPT ); Wed, 4 Apr 2007 03:16:11 -0400 DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=beta; h=received:message-id:date:from:to:subject:cc:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references; b=OMlorb0+MuSF90ROxuAgg9T6UIXDM6KUAvbTLD8Gp5+Ezwpe2pjTuECRgbEuZcK/D6mI4P41XLVaFaZ4V61i+O5cSWW7C4vSZfdNyCnfggfNzilbz8ffQOMjEAOT/vOzAoBDtKwDzaxUxJvZJ53ni1/A2xTDN76N+qNZNW6ouUI= Message-ID: Date: Wed, 4 Apr 2007 00:16:09 -0700 From: "Ulrich Drepper" To: "Eric Dumazet" Subject: Re: [PATCH 0/3] FUTEX : new PRIVATE futexes, SMP and NUMA improvements Cc: "Nick Piggin" , "Andrew Morton" , "Ingo Molnar" , "Andi Kleen" , "Ravikiran G Thirumalai" , "Shai Fultheim (Shai@scalex86.org)" , "pravin b shelar" , linux-kernel@vger.kernel.org In-Reply-To: <200703152010.35614.dada1@cosmosbay.com> MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Content-Disposition: inline References: <20060808070708.GA3931@localhost.localdomain> <200608090826.28249.dada1@cosmosbay.com> <200608090843.52893.dada1@cosmosbay.com> <200703152010.35614.dada1@cosmosbay.com> Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On 3/15/07, Eric Dumazet wrote: > But fact is POSIX defined PRIVATE and SHARED, allowing clear separation, and > optimal performance if carefuly implemented. Time has come for linux to have > better threading performance. Now that I've been pointed to the thread I can comment on it. Yes, this approach makes a lot of sense. Programs which shared syn objects without declaring them correctly are broken and deserve to fail. It would be quite a lot of change in libpthread but it's manageable. I haven't tested the code nor will I likely do until I get a Fedora kernel with it. So, convince DaveJ to take it for testing.