Bug 10839
| Summary: | Computer does not boot on kernel more recent than 2.6.22-2 | ||
|---|---|---|---|
| Product: | ACPI | Reporter: | Louis-Maurice De Sousa (louis-maurice) |
| Component: | Power-Thermal | Assignee: | Zhang Rui (rui.zhang) |
| Status: | REJECTED INVALID | ||
| Severity: | blocking | CC: | acpi-bugzilla, devzero, max, rui.zhang |
| Priority: | P1 | ||
| Hardware: | All | ||
| OS: | Linux | ||
| Kernel Version: | 2.6.23 to 2.6.25 | Subsystem: | |
| Regression: | Yes | Bisected commit-id: | |
| Attachments: |
acpidump after booting with no modules in /lib/modules/2.6.22-2-686/kernel/drivers/acpi/
acpidump after booting with all modules in /lib/modules/2.6.22-2-686/kernel/drivers/acpi/ |
||
|
Description
Louis-Maurice De Sousa
2008-05-31 01:57:38 UTC
>After some mails and search with Debian team >(http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=482964), we found that >thermal module freezes the computer. >I commented it. can you test and make sure, that the same module loads fine with the older kernel ? is this tested with vanilla kernel? if you have lots of bandwidth, maybe it`s worth booting a recent live cd from another distro just to see if it also exists there and how it behaves there. opensuse 11 rc1 has 2.6.25, iirc. see http://software.opensuse.org/developer you can use git-bisect to identify, which patch introduced this problem. >Boot process goes further, but stopps here : >udevd_event[485]: wait_for_sysfs: file >'/sys/devices/LNXSYSTM:00/device:00//bus' appeared >after 0 loops >udevd_event[485]: pass_event_to_socket: passed -1 bytes to socket >'/org/kernel/udev/monitor' >udevd_event[485]: pass_event_to_socket: finished with 0 : udev_event_num: seq >410 >'/org/freedesktop/hal/udev_event' no clue about this one. (In reply to comment #1) > > >Boot process goes further, but stopps here : > >udevd_event[485]: wait_for_sysfs: file > >'/sys/devices/LNXSYSTM:00/device:00//bus' appeared > >after 0 loops this may load the thermal module automatically. You can rename the thermal module under /lib/modules/ to see if the system doesn't hang when thermal is not loaded. And a git-bisect to narrow down the problem would be great. :) > http://bugzilla.kernel.org/show_bug.cgi?id=10839 > > > devzero@web.de changed: > > What |Removed |Added > ---------------------------------------------------------------------- > ------ > CC| |devzero@web.de > > > Hello, sorry for a so late answer but I hadn't too much time to work on this problem and my machine is now complety crashed after an acpid upgrade. http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=483801 > ------- Comment #1 from devzero@web.de 2008-06-01 13:40 ------- > >After some mails and search with Debian team > >(http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=482964), we found > >that thermal module freezes the computer. I commented it. > > can you test and make sure, that the same module loads fine with the > older kernel ? is this tested with vanilla kernel? if you have lots of > bandwidth, maybe it`s worth booting a recent live cd from another > distro just to see if it also exists there and how it behaves there. > opensuse 11 rc1 has 2.6.25, iirc. see > http://software.opensuse.org/developer I tried opensuse 10.3. LiveCd makes computer reboot after loading acpid. Ubuntu 8.04 makes the computer reboot just after kernel is loaded. SystemRescueCd with kernel 2.6.24, freezes after : pci_hotplug: PCI Hot Plug PCI Core version: 0.5 Same CD launches with kernel 2.6.22.18 Open Suse 11 rc1, freezes after : pci_hotplug: PCI Hot Plug PCI Core version: 0.5 agpgart: Detected VIA PT800 chipset agpgart: AGP aperture is 64M @ 0xe00000000 ACPI: Sleep Button (CM) [SLPB] I'll make further tests on wednesday. -- Cordialement. Louis-Maurice De Sousa any update? :) On my MSI M670X I have this problem too. 2.6.22 work perfectly, but 2.6.24-26 freeze my system on boot. I use vanilla kernels. if you are not using the same hardware as the one here, please file a new bug report and attach the acpidump, dmesg output and a screenshot when the system hangs. > http://bugzilla.kernel.org/show_bug.cgi?id=10839 > > > devzero@web.de changed: > > What |Removed |Added > ---------------------------------------------------------------------- > ------ > CC| |devzero@web.de > > > > > ------- Comment #1 from devzero@web.de 2008-06-01 13:40 ------- > >After some mails and search with Debian team > >(http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=482964), we found > >that thermal module freezes the computer. I commented it. > can you test and make sure, that the same module loads fine with the > older kernel ? You are right, the thermal module doesn't seem to be loaded with the older kernel. orval:~# lsmod |grep th orval:~# > is this tested with vanilla kernel? I use a Debian kernel witch I do not modify. orval:~# cat /proc/version Linux version 2.6.22-2-686 (Debian 2.6.22-4) (waldi@debian.org) (gcc version 4.1.3 20070812 (prerelease) (Debian 4.1.2-15)) #1 SMP Fri Aug 31 00:24:01 UTC 2007 > you can use git-bisect to identify, which patch introduced this > problem. I'm sorry but I'm not good enough to compile the kernel myself :-) > >Boot process goes further, but stopps here : > >udevd_event[485]: wait_for_sysfs: file > >'/sys/devices/LNXSYSTM:00/device:00//bus' appeared > >after 0 loops > >udevd_event[485]: pass_event_to_socket: passed -1 bytes to socket > >'/org/kernel/udev/monitor' udevd_event[485]: pass_event_to_socket: > >finished with 0 : udev_event_num: seq 410 > >'/org/freedesktop/hal/udev_event' > > no clue about this one. > After other answers, I moved all acpi modules orval:~# ls /lib/modules/2.6.25-2-686/kernel/drivers/acpi/ orval:~# Here is the result of booting sequence : udevd-event[479]: run_program: '/sbin/modprobe --use-blacklist acpi :ACPI0007 :' udevd[323]: udev_event_run: seq 407 forked, pid [479], 'add' 'acpi', 1 seconds old udevd-event[481]: wait_for_sysfs: file '/sys/devices/LNXSYSTM:00/LNXPWRBN:00/bus'- appeared after 0 loops udevd-event[481]: pass_env_to_socket: passed -1 bytes to socket '/org/kernel/udev/monitor', udevd-event[481]: run_program: '/sbin/modprobe --use-blacklist acpi:LNXPWRBN:' udevd-event[323]: udev_event_run: seq 408 forked, pid [481], 'add' 'acpi' , 1 seconds old udevd-event[483]: wait_for_sysfs: file '/sys/devices/LNXSYSTM:00/LNXTHERM:00/bus' appeared after 0 loops udevd-event[483]: pass_env_to_socket: passed -1 bytes to socket '/org/kernel/udev/monitor', udevd-event[483]: run_program: '/sbin/modprobe --use-blacklist acpi:LNXTHERM:' udevd-event[323]: udev_event_run: seq 409 forked, pid [483], 'add' 'acpi', 1 seconds old udevd-event[486]: wait_for_sysfs: file '/sys/devices/LNXSYSTM:00/device:00/PNP0A03:00/bus' appeared after 0 loops udevd-event[486]: pass_env_to_socket: passed -1 bytes to socket '/org/kernel/udev/monitor', udevd-event[486]: pass_env_to_socket: passed -1 bytes to socket '/org/freedesktop/hal/udev_event', udevd-event[486]: udev_event_run: seq 410 finished with 0 freeze My motherboard is an ASUS P4V8X-X I made last Debian upgrade for 2.6.25 kernel. Sorry for beeing so late but debugging is quite un-easy for me. I'll be more reactive these days. Thanks for help. -- Cordialement. Louis-Maurice De Sousa so there are two problem here, 1. system hangs when boot with thermal driver, this happens on all kernel versions 2. system still freezes if thermal is not loaded, and this is a regression. right? For the first problem, please attach the acpidump output, you can use the latest pmtools here: http://www.lesswatts.org/patches/linux_acpi/ > http://bugzilla.kernel.org/show_bug.cgi?id=10839 Hello, > > > ------- Comment #8 from rui.zhang@intel.com 2008-08-17 19:27 ------- > so there are two problem here, >1. system hangs when boot with thermal > driver, this happens on all kernel versions Ok >2. system still freezes if > thermal is not loaded, and this is a regression. right? Yes. I have to say that the last two acpid upgrades on Debian Lenny (1.0.6-10) made the computer crash on 2.6.22-2 kernel. I had to move all modules from /lib/modules/2.6.22-2- 686/kernel/drivers/acpi/ to let the upgrade end. > For the first problem, please attach the acpidump output, > you can use the latest pmtools here: > http://www.lesswatts.org/patches/linux_acpi/ Joined two acpidump of running 2.6.22-2 kernel. First after booting with no modules in /lib/modules/2.6.22-2-686/kernel/drivers/acpi/ Second after booting with all modules in /lib/modules/2.6.22-2-686/kernel/drivers/acpi/ In this case, I only can boot in recovery mode and start X server by hand. I'll try later with pmtools you gave and booting with one module after another to try to find wich one makes the computer crash even on 2.6.22-2 kernel. -- Cordialement. Louis-Maurice De Sousa The following section of this message contains a file attachment prepared for transmission using the Internet MIME message format. If you are using Pegasus Mail, or any other MIME-compliant system, you should be able to save it or view it from within your mailer. If you cannot, please ask your system administrator for assistance. ---- File information ----------- File: acpidump.txt Date: 18 Aug 2008, 11:30 Size: 67918 bytes. Type: Text The following section of this message contains a file attachment prepared for transmission using the Internet MIME message format. If you are using Pegasus Mail, or any other MIME-compliant system, you should be able to save it or view it from within your mailer. If you cannot, please ask your system administrator for assistance. ---- File information ----------- File: acpidump2.txt Date: 18 Aug 2008, 11:56 Size: 67918 bytes. Type: Text (In reply to comment #9) > > http://bugzilla.kernel.org/show_bug.cgi?id=10839 > > ------- Comment #8 from rui.zhang@intel.com 2008-08-17 19:27 ------- > > so there are two problem here, > >1. system hangs when boot with thermal > > driver, this happens on all kernel versions > > Ok > > >2. system still freezes if > > thermal is not loaded, and this is a regression. right? > > Yes. > > I have to say that the last two acpid upgrades on Debian Lenny (1.0.6-10) > made > the > computer crash on 2.6.22-2 kernel. I had to move all modules from > /lib/modules/2.6.22-2- > 686/kernel/drivers/acpi/ to let the upgrade end. > > > For the first problem, please attach the acpidump output, > > you can use the latest pmtools here: > > http://www.lesswatts.org/patches/linux_acpi/ > > Joined two acpidump of running 2.6.22-2 kernel. > First after booting with no modules in > /lib/modules/2.6.22-2-686/kernel/drivers/acpi/ > Second after booting with all modules in > /lib/modules/2.6.22-2-686/kernel/drivers/acpi/ > In this case, I only can boot in recovery mode and start X server by hand. > Sorry but I didn't found the acpidump. please attach it to the bugzilla. > I'll try later with pmtools you gave and booting with one module after > another > to try to find > wich one makes the computer crash even on 2.6.22-2 kernel. no, pmtools are used to get the acpidump. :) Created attachment 17568 [details]
acpidump after booting with no modules in /lib/modules/2.6.22-2-686/kernel/drivers/acpi/
Created attachment 17569 [details]
acpidump after booting with all modules in /lib/modules/2.6.22-2-686/kernel/drivers/acpi/
Reply-To: louis-maurice.de-sousa@pi-et-ro.net > http://bugzilla.kernel.org/show_bug.cgi?id=10839 > > > > > > ------- Comment #10 from rui.zhang@intel.com 2008-09-01 23:11 ------- > (In reply to comment #9) > > > http://bugzilla.kernel.org/show_bug.cgi?id=10839 > > > > ------- Comment #8 from rui.zhang@intel.com 2008-08-17 19:27 > > > ------- so there are two problem here, > > >1. system hangs when boot with thermal > > > driver, this happens on all kernel versions > > > > Ok > > > > >2. system still freezes if > > > thermal is not loaded, and this is a regression. right? > > > > Yes. > > > > I have to say that the last two acpid upgrades on Debian Lenny > > (1.0.6-10) made the computer crash on 2.6.22-2 kernel. I had to move > > all modules from /lib/modules/2.6.22-2- 686/kernel/drivers/acpi/ to > > let the upgrade end. > > > > > For the first problem, please attach the acpidump output, > > > you can use the latest pmtools here: > > > http://www.lesswatts.org/patches/linux_acpi/ > > > > Joined two acpidump of running 2.6.22-2 kernel. > > First after booting with no modules in > > /lib/modules/2.6.22-2-686/kernel/drivers/acpi/ > > Second after booting with all modules in > > /lib/modules/2.6.22-2-686/kernel/drivers/acpi/ > > In this case, I only can boot in recovery mode and start X server by > > hand. > > > Sorry but I didn't found the acpidump. please attach it to the > bugzilla. Sorry, I misunderstood "attach to bugzilla". :-) It's done now. > > I'll try later with pmtools you gave and booting with one module > > after another to try to find wich one makes the computer crash even > > on 2.6.22-2 kernel. Under 2.6.22-2 kernel, module processor.ko makes the computer reboot on X start. I don't know if it can help. I tried the new Debian package for 2.6.26-1 kernel and the machine still doesn't boot. > no, pmtools are used to get the acpidump. :) Ok, "autant pour moi" as we say in french. :-) What can I do with pmtools to help you ? -- Cordialement. Louis-Maurice De Sousa Will you please try the boot option of "idle=poll" on the latest kernel and see whether the system still can be booted? Hello,
> http://bugzilla.kernel.org/show_bug.cgi?id=10839
>
>
>
>
>
> ------- Comment #14 from yakui.zhao@intel.com 2008-09-23 02:04
> ------- Will you please try the boot option of "idle=poll" on the
> latest kernel and see whether the system still can be booted?
Boot goes further.
Latest Debian kernel 2.6.26-1-686
Stops on initramfs.
***********************
Begin: Running /scripts/init-bottom... mount: mounting /root/dev on /dev/.static/dev failed: No
such file or directory
Done.
mount: mounting /sys on /root/sys failed: No such file or directory
mount: mounting /proc /root/proc failed: No such file or directory
Target fliesystem doesn't have /sbin/init.
No init found. Try passing init= bootarg.
BusyBox....
/bin/sh: can't access tty; job control turned off
(initramfs)
**********************
--
Cordialement.
Louis-Maurice De Sousa
[ switched to email. Please respond via emailed reply-to-all, not via the bugzilla web interface ] On Tue, 23 Sep 2008, bugme-daemon@bugzilla.kernel.org wrote: > > ------- Will you please try the boot option of "idle=poll" on the > > latest kernel and see whether the system still can be booted? > > Boot goes further. > Latest Debian kernel 2.6.26-1-686 > > Stops on initramfs. > > *********************** > Begin: Running /scripts/init-bottom... mount: mounting /root/dev on > /dev/.static/dev failed: No > such file or directory > Done. > mount: mounting /sys on /root/sys failed: No such file or directory > mount: mounting /proc /root/proc failed: No such file or directory > Target fliesystem doesn't have /sbin/init. > No init found. Try passing init= bootarg. > > BusyBox.... > > /bin/sh: can't access tty; job control turned off > (initramfs) this is *not* a kernel failure, check http://wiki.debian.org/InitramfsDebug > http://bugzilla.kernel.org/show_bug.cgi?id=10839
>
> ------- Comment #16 from max@stro.at 2008-09-24 05:20 -------
> [ switched to email. Please respond via emailed reply-to-all, not via
> the
> bugzilla web interface ]
>
> On Tue, 23 Sep 2008, bugme-daemon@bugzilla.kernel.org wrote:
>
> > > ------- Will you please try the boot option of "idle=poll" on the
> > > latest kernel and see whether the system still can be booted?
> >
> > Boot goes further.
> > Latest Debian kernel 2.6.26-1-686
> >
> > Stops on initramfs.
> >
> > ***********************
> > Begin: Running /scripts/init-bottom... mount: mounting /root/dev on
> > /dev/.static/dev failed: No such file or directory Done. mount:
> > mounting /sys on /root/sys failed: No such file or directory mount:
> > mounting /proc /root/proc failed: No such file or directory Target
> > fliesystem doesn't have /sbin/init. No init found. Try passing init=
> > bootarg.
> >
> > BusyBox....
> >
> > /bin/sh: can't access tty; job control turned off
> > (initramfs)
>
> this is *not* a kernel failure, check
> http://wiki.debian.org/InitramfsDebug
Ok, sorry, that was my faut. Grub call to kernel was wrong. What I don't understand is how
the computer could boot with 2.6.22 kernel. The same error was on each kernel call.
So, 2.6.26 works now. :-) Thank you so much for your help.
orval:~# uname -a
Linux orval 2.6.26-1-686 #1 SMP Wed Sep 10 16:46:13 UTC 2008 i686 GNU/Linux
But it only boots with idle=poll option. Without, boot stops on
"/scripts init-premount"
When using a recent live CD (Suse, Ubuntu, Systemrescuecd...) kernel doesn't boot.
--
Cordialement.
Louis-Maurice De Sousa
this is not a kernel bug, and has already been fixed. close this bug. |