From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752953Ab2DUIYf (ORCPT ); Sat, 21 Apr 2012 04:24:35 -0400 Received: from nk11p00mm-asmtp003.mac.com ([17.158.161.2]:52724 "EHLO nk11p00mm-asmtp003.mac.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750709Ab2DUIYb (ORCPT ); Sat, 21 Apr 2012 04:24:31 -0400 X-Greylist: delayed 3600 seconds by postgrey-1.27 at vger.kernel.org; Sat, 21 Apr 2012 04:24:31 EDT MIME-version: 1.0 Content-transfer-encoding: 7BIT Content-type: text/plain; CHARSET=US-ASCII X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10432:5.6.7580,1.0.260,0.0.0000 definitions=2012-04-21_02:2012-04-20,2012-04-21,1970-01-01 signatures=0 X-Proofpoint-Spam-Details: rule=notspam policy=default score=0 spamscore=0 ipscore=0 suspectscore=1 phishscore=0 bulkscore=0 adultscore=0 classifier=spam adjust=0 reason=mlx scancount=1 engine=6.0.2-1012030000 definitions=main-1204210007 Subject: Re: [PATCH] fix oops in updating thread cputime and task time From: Fawzi Mohamed In-reply-to: Date: Sat, 21 Apr 2012 09:23:52 +0200 Cc: Ingo Molnar , Thomas Dargel , Hillf Danton Message-id: <301628B7-B058-4C01-9A65-F2A4AAF05EE6@me.com> References: To: linux-kernel@vger.kernel.org X-Mailer: Apple Mail (2.1251.1) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org An update on this, it successfully passed our testing (more than 35 days of uptime/computation without problems), so we rebased, added tested-by it and submit it again. >>From d0a7d56c283205806d122442448063d2c8812669 Mon Sep 17 00:00:00 2001 From: Fawzi Mohamed Date: Mon, 12 Mar 2012 23:10:57 +0100 Subject: [PATCH] fix oops in updating thread cputime and task time use div64_u64 instead of do_div to divide cputime_t with each other because if cputime_t is u64 it does crash when overflowing u32. KERNEL: ./vmlinux-2.6.32.54-0.3-default.gz DEBUGINFO: .//vmlinux-2.6.32.54-0.3-default.debug DUMPFILE: ./vmcore CPUS: 24 DATE: Mon Mar 5 00:58:10 2012 UPTIME: 19 days, 11:01:56 LOAD AVERAGE: 0.74, 0.42, 0.40 TASKS: 731 NODENAME: node65 RELEASE: 2.6.32.54-0.3-default VERSION: #1 SMP 2012-01-27 17:38:56 +0100 MACHINE: x86_64 (3067 Mhz) MEMORY: 96 GB PANIC: "" PID: 11234 COMMAND: "ricc2_smp" TASK: ffff881813de2240 [THREAD_INFO: ffff8818153c2000] CPU: 8 STATE: TASK_RUNNING (PANIC) crash> bt PID: 11234 TASK: ffff881813de2240 CPU: 8 COMMAND: "ricc2_smp" RIP: 00007f5e4f418617 RSP: 00007f5e4dd3ebd0 RFLAGS: 00000202 RAX: 0000000000000062 RBX: ffffffff81002f7b RCX: 0000000000000179 RDX: 00007f5e4dd3edb4 RSI: 00007f5e4dd3ec70 RDI: 0000000000000000 RBP: 00007f5e4dd3ecf0 R8: 0000000000000001 R9: 00007f5e4dd3edc4 R10: 0000000000000001 R11: 0000000000000246 R12: ffffffff8105eb81 R13: 00007f5e4dd3ed30 R14: 00007f5e4dd3f2f0 R15: 0000000000000000 ORIG_RAX: 0000000000000062 CS: 0033 SS: 002b Basically the same problem was found earlier in kernel vmlinux-2.6.32.12-0.7-default.gz. This patch was tested on 2.6.32 (as some kernel modules we need require it) and the computation ran for more than 35 days without any problems, before it would crash after 19 days of that ricc2 computation. Signed-off-by: Fawzi Mohamed Signed-off-by: Thomas Dargel Tested-by: Thomas Dargel Cc: Hillf Danton Cc: Ingo Molnar --- kernel/sched/core.c | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 4603b9d..03a2d89 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -2966,7 +2966,7 @@ void task_times(struct task_struct *p, cputime_t *ut, cputime_t *st) u64 temp = (__force u64) rtime; temp *= (__force u64) utime; - do_div(temp, (__force u32) total); + temp = div64_u64(temp, total); utime = (__force cputime_t) temp; } else utime = rtime; @@ -2999,7 +2999,7 @@ void thread_group_times(struct task_struct *p, cputime_t *ut, cputime_t *st) u64 temp = (__force u64) rtime; temp *= (__force u64) cputime.utime; - do_div(temp, (__force u32) total); + temp = div64_u64(temp, total); utime = (__force cputime_t) temp; } else utime = rtime; -- 1.7.0.4