From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755407AbYDWN1U (ORCPT ); Wed, 23 Apr 2008 09:27:20 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1752968AbYDWN1M (ORCPT ); Wed, 23 Apr 2008 09:27:12 -0400 Received: from mail.gmx.net ([213.165.64.20]:57297 "HELO mail.gmx.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with SMTP id S1752952AbYDWN1L convert rfc822-to-8bit (ORCPT ); Wed, 23 Apr 2008 09:27:11 -0400 X-Authenticated: #14349625 X-Provags-ID: V01U2FsdGVkX18/XuFe++M/sjHUYhLP3vk/zyB05M8JBUhrV3JdCm UHf8TVF3mCpIOd Subject: Re: [git pull] scheduler changes for v2.6.26 From: Mike Galbraith To: Peter Zijlstra Cc: Ingo Molnar , Frans Pop , torvalds@linux-foundation.org, linux-kernel@vger.kernel.org, akpm@linux-foundation.org, Richard Jonsson , "Rafael J. Wysocki" In-Reply-To: <1208949715.7115.334.camel@twins> References: <20080419181304.GB21353@elte.hu> <200804192147.43719.elendil@planet.nl> <20080421123903.GE9554@elte.hu> <200804211831.29976.elendil@planet.nl> <20080421194359.GD8770@elte.hu> <1208854300.4695.7.camel@marge.simson.net> <20080422085905.GA9939@elte.hu> <1208867154.4873.4.camel@marge.simson.net> <1208938303.4905.23.camel@marge.simson.net> <1208946604.4803.9.camel@marge.simson.net> <1208949715.7115.334.camel@twins> Content-Type: text/plain; charset=US-ASCII Date: Wed, 23 Apr 2008 15:27:06 +0200 Message-Id: <1208957226.4223.14.camel@marge.simson.net> Mime-Version: 1.0 X-Mailer: Evolution 2.12.0 Content-Transfer-Encoding: 7BIT X-Y-GMX-Trusted: 0 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wed, 2008-04-23 at 13:21 +0200, Peter Zijlstra wrote: > this should fix it: Did. But... Continuing testing, x86/latest doesn't seem to hard lock (though i'm going to run a while longer to be sure). However, it does do something else. watch-rq-clock.sh is triggering an occasional oops. After the oops, instead of heading off to bit heaven, grep hangs around eating 100% cpu, _possibly_ until it's parent is killed. I say possibly, because that's what it looked like the first time. Also, amarok suddenly ceased playing and exited without a peep... weird. I restarted amarok, kept testing and repeated the oops, but just as luck would have it, my keyboard went away again, so I had to reboot before I was able to see if killing watch-rq-clock.sh would kill the runaway grep again, or if amarok was going to go awol again. [ 399.477589] BUG: unable to handle kernel NULL pointer dereference at 0000000000000418 [ 399.481049] IP: [] calc_delta_weight+0x17/0x37 [ 399.481049] PGD 7b0ba067 PUD 7b1a2067 PMD 0 [ 399.481049] Oops: 0000 [1] SMP [ 399.481049] CPU 2 [ 399.481049] Modules linked in: ip6t_LOG nf_conntrack_ipv6 xt_pkttype ipt_LOG xt_limit af_packet snd_pcm_oss snd_mixer_oss snd_seq snd_seq_device ip6t_REJECT xt_tcpudp ipt_REJECT xt_state iptable_mangle iptable_nat nf_nat iptable_filter ip6table_mangle nf_conntrack_ipv4 nf_conntrack ip_tables ip6table_filter ip6_tables x_tables cpufreq_conservative cpufreq_userspace ipv6 cpufreq_powersave acpi_cpufreq microcode nls_iso8859_1 nls_cp437 vfat fat fuse loop dm_mod tda1004x saa7134_dvb videobuf_dvb dvb_core tuner tea5767 tda8290 tda827x tuner_xc2028 tda9887 tuner_simple mt20xx tea5761 saa7134 snd_hda_intel compat_ioctl32 videodev v4l1_compat v4l2_common snd_pcm firewire_ohci videobuf_dma_sg firewire_core snd_timer videobuf_core ir_kbd_i2c usbhid snd_page_alloc ir_common crc_itu_t snd_hwdep usb_storage snd tveeprom e1000e ohci1394 hid iTCO_wdt rtc_cmos i2c_i801 ieee1394 iTCO_vendor_support intel_agp rtc_core sr_mod i2c_core rtc_lib button serio_raw ff_memless sg cdrom soundcore ehci_hcd uhci_hcd sd_mod usbcore edd ext3 mbcache jbd fan ahci libata scsi_mod dock thermal processor [ 399.481049] Pid: 20646, comm: grep Not tainted 2.6.26-devel-smp #2 [ 399.481049] RIP: 0010:[] [] calc_delta_weight+0x17/0x37 [ 399.481049] RSP: 0018:ffff8100994bdd88 EFLAGS: 00010286 [ 399.481049] RAX: 0000000000000400 RBX: ffff81000102be30 RCX: 0000000000000000 [ 399.481049] RDX: ffffffff805b1dc0 RSI: 0000000000000001 RDI: 0000000000000400 [ 399.481049] RBP: ffff8100994bdd88 R08: ffff81007b1606af R09: ffff81008c47b2a0 [ 399.481049] R10: ffffffffffffffff R11: 0000000000000000 R12: ffff810001022e58 [ 399.481049] R13: ffff81008c47b2a0 R14: ffff810001022dc0 R15: 0000000000000000 [ 399.481049] FS: 00007fe5fcefa6f0(0000) GS:ffff8100bf80e500(0000) knlGS:0000000000000000 [ 399.481049] CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b [ 399.481049] CR2: 0000000000000418 CR3: 000000007b015000 CR4: 00000000000006e0 [ 399.481049] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 399.481049] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400 [ 399.481049] Process grep (pid: 20646, threadinfo ffff8100994bc000, task ffff81008c5bc000) [ 399.481049] Stack: ffff8100994bde58 ffffffff8022d46b ffffffff805348e8 00000010000240d0 [ 399.481049] ffffe20001aecd00 ffffffff805b1dc0 ffffffff805b1dc0 ffff8100acc36441 [ 399.481049] 0000000000008001 0000000000000001 00000000000040d0 00000000000040d0 [ 399.481049] Call Trace: [ 399.481049] [] sched_debug_show+0x9b3/0xd2b [ 399.481049] [] ? alloc_pages_node+0x31/0x36 [ 399.481049] [] ? __get_free_pages+0xf/0x4c [ 399.481049] [] seq_read+0x10b/0x29a [ 399.481049] [] proc_reg_read+0x7b/0x95 [ 399.481049] [] vfs_read+0xa8/0x102 [ 399.481049] [] sys_read+0x47/0x6d [ 399.481049] [] system_call_after_swapgs+0x7b/0x80 [ 399.481049] [ 399.481049] [ 399.481049] Code: 89 da e8 a7 fe ff ff 48 01 43 50 4c 89 63 40 5b 41 5c c9 c3 55 48 85 f6 48 89 e5 74 29 48 8b 46 d0 48 c7 c2 c0 1d 5b 80 48 8b 36 <8b> 40 18 48 8b 04 c5 a0 41 56 80 48 03 50 08 48 83 c2 60 e8 6b [ 399.481049] RIP [] calc_delta_weight+0x17/0x37 [ 399.481049] RSP [ 399.481049] CR2: 0000000000000418 [ 399.481052] ---[ end trace 36afd5990b2873de ]---