From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1758733Ab0BRRBk (ORCPT ); Thu, 18 Feb 2010 12:01:40 -0500 Received: from bombadil.infradead.org ([18.85.46.34]:55819 "EHLO bombadil.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1758723Ab0BRRBj (ORCPT ); Thu, 18 Feb 2010 12:01:39 -0500 Subject: Re: Stupid futex question - 2.6.33-rc7-mmotm0210 From: Peter Zijlstra To: Valdis.Kletnieks@vt.edu Cc: Andrew Morton , Ingo Molnar , Thomas Gleixner , linux-kernel@vger.kernel.org In-Reply-To: <7380.1266511335@localhost> References: <10057.1266501862@localhost> <1266503879.26719.702.camel@laptop> <7380.1266511335@localhost> Content-Type: text/plain; charset="UTF-8" Date: Thu, 18 Feb 2010 18:01:26 +0100 Message-ID: <1266512486.26719.856.camel@laptop> Mime-Version: 1.0 X-Mailer: Evolution 2.28.2 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Thu, 2010-02-18 at 11:42 -0500, Valdis.Kletnieks@vt.edu wrote: > 1) should it get killed for RLIMIT_RTTIME because it's been in a futex() > for multiple seconds? It seems suspicious - docs say a blocking syscall > resets RTTIME - so if futex() blocks it shouldn't kill, and if it's in the > kernel for a second without blocking it's a bug too. sys_futex() shouldn't be taking that long indeed, and if its not caused by the sysrq interrupt then something funny is going on indeed. You could poke around in kernel/futex.c:futex_wait() to see where it thinks its spending it time? There's a single retry loop in there, maybe instrument that to see if its retrying too many times..