Bug#630360: linux-2.6: sporadic kernel bug while trying to hibernate system
On Tue, 02 Aug 2011 01:58:43 +0100, Ben Hutchings b...@decadent.org.uk wrote: Linux 3.0 is now in unstable; please test it. (Note that the upstream version used for our packages is '3.0.0' and not '3.0'.) It seems to be fixed on my netbook (i386) while my other laptop (amd64) now freezes instead of a kernel panic when trying to hibernate with more than half of my RAM in use. Cheers, Daniel pgpE5T2hZyXtc.pgp Description: PGP signature
Bug#630360: linux-2.6: sporadic kernel bug while trying to hibernate system
On Mon, 2011-07-18 at 17:24 -0500, Jonathan Nieder wrote: Daniel Schoepe wrote: No improvement with linux-image-3.0.0-rc5. Good news: v3.0-rc7~35^2 (PM / Hibernate: Fix free_unnecessary_pages(), 2011-07-06) is said to fix this. Assuming that works, no need to bisect. :) Once 3.0.0~rc7 or later is in experimental, if you get a chance to test it, I'd be happy to hear about the result. Thanks again for your work. Linux 3.0 is now in unstable; please test it. (Note that the upstream version used for our packages is '3.0.0' and not '3.0'.) Ben. signature.asc Description: This is a digitally signed message part
Bug#630360: linux-2.6: sporadic kernel bug while trying to hibernate system
On Mon, 4 Jul 2011 17:27:15 -0500, Jonathan Nieder jrnie...@gmail.com wrote: So it's a regression from squeeze and 100% reproducible? From the point of view of fixing it, that's excellent news. Sorry for replying so late, but yes, I can reproduce it reliably. Except for the list of affected versions, this looks like [*] https://bugzilla.kernel.org/show_bug.cgi?id=15816. If you have time, could you do the following? - first, test the latest kernel from experimental No improvement with linux-image-3.0.0-rc5. - find the first affected upstream release by running a bisection search through packages at snapshot.debian.org The first affected version seems to be 2.6.38-1. - try the debugging patch from Raphael Wysoki's upsteam [*] comment #7, using instructions from here: http://kernel-handbook.alioth.debian.org/ch-common-tasks.html#s-common-building and Documentation/applying-patches.txt and let Raphael know the result and details about your case (e.g., by commenting on the upstream report). - try using git bisect to find out which upstream change introduced the bug, like so: I'll do that when I get some time, which may not be for a few weeks though, since I got some exams coming up. Cheers, Daniel pgpLsQ2OoM2I5.pgp Description: PGP signature
Bug#630360: linux-2.6: sporadic kernel bug while trying to hibernate system
found 630360 linux-2.6/3.0.0~rc5-1~experimental.1 forwarded 630360 https://bugzilla.kernel.org/show_bug.cgi?id=15816 tags 630360 + fixed-upstream quit Hi, Daniel Schoepe wrote: No improvement with linux-image-3.0.0-rc5. Good news: v3.0-rc7~35^2 (PM / Hibernate: Fix free_unnecessary_pages(), 2011-07-06) is said to fix this. Assuming that works, no need to bisect. :) Once 3.0.0~rc7 or later is in experimental, if you get a chance to test it, I'd be happy to hear about the result. Thanks again for your work. -- To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20110718222419.ga7...@elie.gateway.2wire.net
Bug#630360: linux-2.6: sporadic kernel bug while trying to hibernate system
I am also affected by this bug and it seems to occur if and only if I try to hibernate with more than about 50% of my RAM in use. This is the configured Preferred maximum image size for uswsusp on my system, but changing it to a higher value does not prevent it. pgphjUxkJT27W.pgp Description: PGP signature
Bug#630360: linux-2.6: sporadic kernel bug while trying to hibernate system
retitle 630360 linux: intermittent BUG at kernel/power/snapshot.c:528 when trying to hibernate found 630360 linux-2.6/2.6.39-2 # based on the above messages quit Hi Daniel, Daniel Schoepe wrote: I am also affected by this bug and it seems to occur if and only if I try to hibernate with more than about 50% of my RAM in use. This is the configured Preferred maximum image size for uswsusp on my system, but changing it to a higher value does not prevent it. So it's a regression from squeeze and 100% reproducible? From the point of view of fixing it, that's excellent news. Except for the list of affected versions, this looks like [*] https://bugzilla.kernel.org/show_bug.cgi?id=15816. If you have time, could you do the following? - first, test the latest kernel from experimental - find the first affected upstream release by running a bisection search through packages at snapshot.debian.org - try the debugging patch from Raphael Wysoki's upsteam [*] comment #7, using instructions from here: http://kernel-handbook.alioth.debian.org/ch-common-tasks.html#s-common-building and Documentation/applying-patches.txt and let Raphael know the result and details about your case (e.g., by commenting on the upstream report). - try using git bisect to find out which upstream change introduced the bug, like so: $ apt-get install git build-essential $ git clone git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6.git $ cd linux-2.6 $ make localmodconfig; # minimal configuration $ make deb-pkg $ cd ..; dpkg -i produced package; reboot $ ... test test test ... If it works at this point, declare victory and let us know. $ git bisect start $ git bisect bad; # tell git the good news $ git checkout v2.6.32; # or whatever version you think will work $ make silentoldconfig $ make fakeroot -u make deb-pkg $ cd ..; dpkg -i produced package; reboot $ ... test test test ... Hopefully that one does not trigger the BUG. $ git bisect good; # tell git the good news Then it will automatically check out a revision half-way, so: $ make silentoldconfig $ make fakeroot -u make deb-pkg $ cd ..; dpkg -i produced package; reboot $ ... test test test ... $ git bisect bad; # if it triggers the BUG, or $ git bisect good; # if working hard still does not trigger the BUG, or $ git bisect skip; # if some other bug makes it hard to test Rinse and repeat until it tells the first bad commit or you get bored and let us know the outcome plus git bisect log output. If the gitk package is installed, you can use git bisect visualize to see the regression range narrow at each step. See http://www.kernel.org/pub/software/scm/git/docs/git-bisect-lk2009.html for more details. Thanks for reporting and hope that helps. Jonathan -- To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20110704222715.GB8909@elie
Bug#630360: linux-2.6: sporadic kernel bug while trying to hibernate system
Ok, next time I will send photo. Kernel 2.6.36 was perfectly hibernating. I noticed that first affected kernel was 2.6.38. So I tried newest kernel from unstable (2.6.39), but defect remained. haldyr On Tue, Jun 14, 2011 at 00:45, Ben Hutchings b...@decadent.org.uk wrote: On Mon, 2011-06-13 at 13:43 +0200, haldyr wrote: Package: linux-2.6 Severity: normal Tags: upstream When trying to hibernate my system, it works in about 3 out of 5 times but sometimes crashes with a kernel bug. The error message is not store in any log, so I had to rewrite from photo (might be some errors from typing) often. [...] If you ever have to send kernel bug/oops messages again, a photo is acceptable if the text is readable. There is no need to type it all in. What was the last kernel version you used, that did not have this problem? Ben. -- Ben Hutchings Larkinson's Law: All laws are basically false.
Bug#630360: linux-2.6: sporadic kernel bug while trying to hibernate system
Package: linux-2.6 Severity: normal Tags: upstream When trying to hibernate my system, it works in about 3 out of 5 times but sometimes crashes with a kernel bug. The error message is not store in any log, so I had to rewrite from photo (might be some errors from typing) often. -- System Information: Debian Release: wheezy/sid APT prefers testing APT policy: (990, 'testing'), (500, 'testing-proposed-updates'), (500, 'unstable'), (500, 'stable'), (1, 'experimental') Architecture: i386 (i686) Kernel: Linux 2.6.39-1-686-pae (SMP w/2 CPU cores) Locale: LANG=cs_CZ.UTF-8, LC_CTYPE=cs_CZ.UTF-8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/dash *** /media/data/stazene/bug/bug_report.txt [ cut here ] [19437.775417] kernel BUG at /build/buildd- linux-2.6_2.6.39-1-i386-RRBuT6/1inux-2.6-2.6.39/debian/build/source_i386_none/kernel/power/snapshot.c:528! [19437.775417] invalid opcode: [#1] SMP [19437.775417] last sysfs file: /sys/devices/virtua1/sound/timer/uevent [19437.775417] Modules linked in: ip6tab1e_filter ip6_tables iptable_filter ip_tables ebtable_nat ebtables x_tables acpi_cpufreq mperf cpufreq_conservative cpufreq_stats cpufreq_userspace cpufreq_powersave parport_pc ppdev lp parport snd_hrtimer binfmt_misc kvm_intel kvm uinput fuse nfsd nfs lockd fscache auth_rpcgss nfs_acl sunrpc lm90 radeon ttm drm_kms_helper drm i2c_algo_bit coretemp firew1re_sbp2 loop joydev snd_hda_codec_rea1teak snd_hda_codec_si3054 snd_hda_intel arc4 snd_hda_codec snd_hwdep snd_pcm_oss snd_mixer_oss snd_pcm snd_seq_midi snd_seq_midi_event iwl3945 snd_seq snd_timer snd_seq_deviceiwl_legacy mac80211 tpm_tis tpm tpm_bios snd i2c_i801 evdev cfg80211 battery serio_raw pcspkr i2c_core ac rfkill video processor wmi power_supply soundcore snd_pagre_alloc rng_core button ext4 mbcache jbd2 crc16 usbhid hid sg sr_mod sd_mod cdrom crc_t10dif ata_generic ahci libahci ata_piix libata r8169 scsi_mod usbcore firewire_ohci thermal mii firewire_core thermal_sys crc_itu_t [last unloaded: uhci_hcd] [19437.775417] [19437.775417] Pid: 23028, comm: s2disk Not tainted 2.6.39-1-686-pae #1 FUJITSU SIEMENS AMILO Pi 1536/AMILO Pi 1536 [19437.775417] EIP: 0060:[c1062cd4] EFLAGS: 00010282 CPU: 0 [19437.775417] EIP is at memory_bm_clear_bit+0x15/0x28 [19437.775417] EAX: fff2 EBX: c3657e98 ECX: c3657e9c EDX: c3657e98 [19437.775417] ESI: f5d1ffe0 EDI: EBP: 0003cf5a ESP: c3657e98 [19437.775417] DS: 007b ES: 007b FS: 00d8 GS: 00e0 SS: 0068 [19437.775417] Process s2disk (pid: 23028, ti=c3656000 task =ed983240 task.ti=c3656000) [19437.775417] Stack: [19437.775417] c10633df fffe2dd1 fffe2dd0 c1063b74 4f341f90 0001d172 1188 [19437.775417] 00024249 00023911 0003a420 00048492 91bf 0001 0004 4defff99 [19437.775417] 000eb385 0001 bf84179C c14d8764 C106260c 40043311 fff0 [19437.775417] Call trace: [19437.775417] [c10633df] ? memory_bm_next_pfn+0x18/0x4f [19437.776005] [c1063b74] ? hibernate_preallocate_memory+0x3ca/0x458 [19437.776005] [c106260c] ? hibernation_snapshot+0x23/0x210 [19437.776005] [c10659de] ? snapshot_ioctl+0x229/0x4c6 [19437.776005] [c11c352f] ? n_tty.write+0x2b9/0x2ee [19437.776005] [C102dfab] ? __wake_up+0x2c/0x3b [19437.776005] [c10657b5] ? snapshot_deprecated_ioctl+0x25/0x25 [19437.776005] [c10d7733] ? do_vfs_ioct1+0x457/0x4a2 [19437.776005] [c10cc0c21 ? fsnotify_modify+0x48/0x4f [19437.776005] [c11c0615] ? tty.write_1ock+0x35/0x35 [19437.776005] [c10ccc7d] ? vfs_write+0x9c/0xd9 [19437.776005] [c12b6fdf] ? sysenter_do_call+0x12/0x28 [19437.776005] Code: 8d 54 02 04 89 73 04 89 dO 5b 5e c3 ba 01 00 00 00 e9 34 ff ff ff 53 83 ec 08 89 e3 8d 4c 24 04 53 e8 49 fd ff ff 5a 85 c0 74 04 0f 0b ab fe 8b 44 24 04 8b 14 24 f0 0f b3 10 5b 58 5b c3 89 c2 -- To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20110613114303.17858.47243.report...@haldyr-laptop.sh.cvut.cz
Bug#630360: linux-2.6: sporadic kernel bug while trying to hibernate system
On Mon, 2011-06-13 at 13:43 +0200, haldyr wrote: Package: linux-2.6 Severity: normal Tags: upstream When trying to hibernate my system, it works in about 3 out of 5 times but sometimes crashes with a kernel bug. The error message is not store in any log, so I had to rewrite from photo (might be some errors from typing) often. [...] If you ever have to send kernel bug/oops messages again, a photo is acceptable if the text is readable. There is no need to type it all in. What was the last kernel version you used, that did not have this problem? Ben. -- Ben Hutchings Larkinson's Law: All laws are basically false. signature.asc Description: This is a digitally signed message part
