From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 24458 invoked by alias); 12 Feb 2013 21:41:25 -0000 Received: (qmail 24448 invoked by uid 22791); 12 Feb 2013 21:41:24 -0000 X-SWARE-Spam-Status: No, hits=-4.7 required=5.0 tests=AWL,BAYES_00,DKIM_SIGNED,DKIM_VALID,FREEMAIL_FROM,KHOP_RCVD_TRUST,KHOP_THREADED,RCVD_IN_DNSWL_LOW,RCVD_IN_HOSTKARMA_YE X-Spam-Check-By: sourceware.org Received: from mail-wi0-f172.google.com (HELO mail-wi0-f172.google.com) (209.85.212.172) by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Tue, 12 Feb 2013 21:41:17 +0000 Received: by mail-wi0-f172.google.com with SMTP id ez12so4971750wid.17 for ; Tue, 12 Feb 2013 13:41:16 -0800 (PST) MIME-Version: 1.0 X-Received: by 10.180.100.169 with SMTP id ez9mr6146931wib.3.1360705275905; Tue, 12 Feb 2013 13:41:15 -0800 (PST) Received: by 10.216.179.134 with HTTP; Tue, 12 Feb 2013 13:41:15 -0800 (PST) In-Reply-To: References: <20130209.014909.573423212502453699.davem@davemloft.net> <5116636F.1040003@redhat.com> <20130209.235603.1409377113964639273.davem@davemloft.net> Date: Tue, 12 Feb 2013 21:41:00 -0000 Message-ID: Subject: Re: [PATCH] Avoid unnecessary busy loop in __lll_timedlock_wait on ARM. From: Daniel Jacobowitz To: "Carlos O'Donell" Cc: David Miller , carlos@redhat.com, "Joseph S. Myers" , schwab@linux-m68k.org, thomas_schwinge@mentor.com, kkojima@rr.iij4u.or.jp, marcus.shawcroft@linaro.org, katsuki.uwatoko@toshiba.co.jp, libc-ports@sourceware.org Content-Type: text/plain; charset=ISO-8859-1 Mailing-List: contact libc-ports-help@sourceware.org; run by ezmlm Precedence: bulk List-Id: List-Subscribe: List-Post: List-Help: , Sender: libc-ports-owner@sourceware.org X-SW-Source: 2013-02/txt/msg00027.txt.bz2 It's been too long, sorry. It may have been necessary solely to provide the separate EABI and NPTL versions in sysdeps; you'd have to look at e.g. the sysdeps selection order for the LinuxThreads version. It may also be related to the lack of usable atomic primitives, pre-EABI. On Sun, Feb 10, 2013 at 12:54 PM, Carlos O'Donell wrote: > On Sat, Feb 9, 2013 at 11:56 PM, David Miller wrote: >> From: "Carlos O'Donell" >> Date: Sat, 09 Feb 2013 09:55:43 -0500 >> >>> I'd seen the *other* sparc pre-v9 implementation that used 64 global >>> locks per-library and that seemed signal unsafe and prone to deadlocks. >> >> All of these pre-v9 things are signal unsafe and deadlock. >> >> I thought about doing the kernel atomic emulation other platforms have >> adopted, but frankly these cpus are so old and deprecated that they're >> not worth doing the work for. >> >> And by the time we'd propagate all of this infrastructure necessary to >> support this kind of scheme, those cpus would be even more outdated. >> >> Even debian does v9-only build on 32-bit. > > Eminently practical. Just curious. Thanks for verifying what I suspected. > > Cheers, > Carlos. -- Thanks, Daniel