[Bug 436384] New: ACPI powerdown fails (regression from 10.2)
https://bugzilla.novell.com/show_bug.cgi?id=436384 Summary: ACPI powerdown fails (regression from 10.2) Product: openSUSE 11.0 Version: Final Platform: i686 OS/Version: openSUSE 11.0 Status: NEW Severity: Major Priority: P5 - None Component: Kernel AssignedTo: bnc-team-screening@forge.provo.novell.com ReportedBy: matthias.andree@gmx.de QAContact: qa@suse.de Found By: --- Since the openSUSE 10.2 -> 11.0 upgrade, my machine hangs during most "poweroff" attempts. It shuts down Linux itself properly, unmounts all file systems, shuts down hard drives, but when trying to power off, the green power LED shuts off, the red IDE HDD LED lights steadily and the computer is not powered down. Upon next boot, the BIOS sometimes complains that CPU frequency settings were wrong and calls setup. openSUSE 10.2 (no longer installed), FreeBSD 6-STABLE and 7-STABLE are capable of shutting down this hardware properly. This happens with kernel-pae-2.6.25.16-0.1 and with vanilla 2.6.27. Are there any weakly documented ACPI or kernel parameters worth trying to aid debugging? -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.
https://bugzilla.novell.com/show_bug.cgi?id=436384
User matthias.andree@gmx.de added comment
https://bugzilla.novell.com/show_bug.cgi?id=436384#c1
--- Comment #1 from Matthias Andree
https://bugzilla.novell.com/show_bug.cgi?id=436384
Michal Seben
https://bugzilla.novell.com/show_bug.cgi?id=436384
Jeff Mahoney
https://bugzilla.novell.com/show_bug.cgi?id=436384
User trenn@novell.com added comment
https://bugzilla.novell.com/show_bug.cgi?id=436384#c2
Thomas Renninger
Are there any weakly documented ACPI or kernel parameters worth trying to aid debugging? With newer (11.1, not sure about 11.0) kernels you can increase the acpi debug level. E.g.: echo 0x1f /sys/modules/acpi/parameters/debug_level and then shut down. The last line you see is the interesting one.
-- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.
https://bugzilla.novell.com/show_bug.cgi?id=436384
User matthias.andree@gmx.de added comment
https://bugzilla.novell.com/show_bug.cgi?id=436384#c3
--- Comment #3 from Matthias Andree
https://bugzilla.novell.com/show_bug.cgi?id=436384
User matthias.andree@gmx.de added comment
https://bugzilla.novell.com/show_bug.cgi?id=436384#c4
--- Comment #4 from Matthias Andree
Question: do Linux and FreeBSD share ACPI code and/or updates? Yes. So your report that BSD also broke is interesting. It seems the problem
https://bugzilla.novell.com/show_bug.cgi?id=436384
User trenn@novell.com added comment
https://bugzilla.novell.com/show_bug.cgi?id=436384#c5
Thomas Renninger
Followup-Question: is that my BIOS at fault or an ACPI regression shared by Linux and FreeBSD? Hard to say. Maybe both. As it already worked it definitely looks like a kernel regression. Even if it was a bad BIOS it could handle, it should still be able to do so.
I am setting the bug to need-info for you. Remove the flag as soon as you have tested 11.1. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.
https://bugzilla.novell.com/show_bug.cgi?id=436384
User matthias.andree@gmx.de added comment
https://bugzilla.novell.com/show_bug.cgi?id=436384#c6
Matthias Andree
https://bugzilla.novell.com/show_bug.cgi?id=436384
User trenn@novell.com added comment
https://bugzilla.novell.com/show_bug.cgi?id=436384#c7
Thomas Renninger
participants (1)
-
bugzilla_noreply@novell.com