[Bug 410612] New: kacpid goes wild after s2disk
https://bugzilla.novell.com/show_bug.cgi?id=410612 Summary: kacpid goes wild after s2disk Product: openSUSE 11.0 Version: Final Platform: Other OS/Version: Other Status: NEW Severity: Normal Priority: P5 - None Component: Kernel AssignedTo: trenn@novell.com ReportedBy: pablo@blueoakdb.com QAContact: qa@suse.de CC: astarikovskiy@novell.com Found By: --- Per Thomas (otherwise I wouldn't want someone to think I knew what the heck I'm saying here! :) kacpid goes into some sort of loop after resuming from an s2disk. It seems to be a power status getting evaluated in the battery's resume function. Thomas believes he can work on the fix so astarikvskiy, please stand by. -- 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=410612
User pablo@blueoakdb.com added comment
https://bugzilla.novell.com/show_bug.cgi?id=410612#c1
--- Comment #1 from Pablo Sanchez
https://bugzilla.novell.com/show_bug.cgi?id=410612
User pablo@blueoakdb.com added comment
https://bugzilla.novell.com/show_bug.cgi?id=410612#c2
Pablo Sanchez
https://bugzilla.novell.com/show_bug.cgi?id=410612
User pablo@blueoakdb.com added comment
https://bugzilla.novell.com/show_bug.cgi?id=410612#c3
--- Comment #3 from Pablo Sanchez
https://bugzilla.novell.com/show_bug.cgi?id=410612
User trenn@novell.com added comment
https://bugzilla.novell.com/show_bug.cgi?id=410612#c4
Thomas Renninger
Thomas believes he can work on the fix Thomas is not that sure anymore he knows what is going wrong or better say why... I'd say battery reads seem to happen a way too often, but not that often that kacpid should eat 100% CPU?
Can you also attach full dmesg and acpidump, best unzipped as plain/text. It should also be a good idea to try to move the battery.ko driver away first, to be sure it is that: mv /lib/modules/`uname -r`/kernel/drivers/acpi/battery.ko /lib/modules try again. Best with the ACPI debug hooks again, maybe then SATA is looping again? If yes it is related and the on top fix should also not help Michal. If not, then there is another bug. Do not forget to move the battery.ko driver back: mv /lib/modules/battery.ko /lib/modules/`uname -r`/kernel/drivers/acpi/battery.ko
I should mention I'm leaving on vacation for a week I am going to lower the severity a bit, having a critical bug in my neck, knowing there won't be action the next time makes me nervous :)
-- 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=410612
User pablo@blueoakdb.com added comment
https://bugzilla.novell.com/show_bug.cgi?id=410612#c5
Pablo Sanchez
https://bugzilla.novell.com/show_bug.cgi?id=410612
User pablo@blueoakdb.com added comment
https://bugzilla.novell.com/show_bug.cgi?id=410612#c6
Pablo Sanchez
https://bugzilla.novell.com/show_bug.cgi?id=410612
User pablo@blueoakdb.com added comment
https://bugzilla.novell.com/show_bug.cgi?id=410612#c7
--- Comment #7 from Pablo Sanchez
https://bugzilla.novell.com/show_bug.cgi?id=410612
User pablo@blueoakdb.com added comment
https://bugzilla.novell.com/show_bug.cgi?id=410612#c8
--- Comment #8 from Pablo Sanchez
https://bugzilla.novell.com/show_bug.cgi?id=410612
User trenn@novell.com added comment
https://bugzilla.novell.com/show_bug.cgi?id=410612#c9
Thomas Renninger
https://bugzilla.novell.com/show_bug.cgi?id=410612
User trenn@novell.com added comment
https://bugzilla.novell.com/show_bug.cgi?id=410612#c10
--- Comment #10 from Thomas Renninger
Unfortunately the ACPI debug output when running in a loop after s2disk cannot be written to disk or be caught by the reporter To clarify: s2disk works, but after s2disk kacpi process produces high CPU load right? (you should see this process in top, or do you only see increased acpi interrupts with powertop)? I expect: If you increase acpi debug level to 0x1f there is so much output that it just takes very long until it goes through. You could try again, if you let it run for a while, s2disk (maybe 10 minutes or so?) could finish at some time even with increase output.
-- 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=410612
User hmacht@novell.com added comment
https://bugzilla.novell.com/show_bug.cgi?id=410612#c11
--- Comment #11 from Holger Macht
https://bugzilla.novell.com/show_bug.cgi?id=410612
User pablo@blueoakdb.com added comment
https://bugzilla.novell.com/show_bug.cgi?id=410612#c13
--- Comment #13 from Pablo Sanchez
To clarify: s2disk works, but after s2disk kacpi process produces high CPU load right?
You're correct. I'm able to resume and that's when I see `kacpi_notify' consume 41% of my CPU and `kacpid' consumes 31%; I have a duo core.
I expect: If you increase acpi debug level to 0x1f there is so much output that it just takes very long until it goes through. You could try again, if you let it run for a while, s2disk (maybe 10 minutes or so?) could finish at some time even with increase output.
I'll see if I can squeeze this test out before I go on vacation (less than 60 minutes from now .. and I have to pack! :) -- 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=410612
User pablo@blueoakdb.com added comment
https://bugzilla.novell.com/show_bug.cgi?id=410612#c14
--- Comment #14 from Pablo Sanchez
https://bugzilla.novell.com/show_bug.cgi?id=410612
User trenn@novell.com added comment
https://bugzilla.novell.com/show_bug.cgi?id=410612#c15
--- Comment #15 from Thomas Renninger
https://bugzilla.novell.com/show_bug.cgi?id=410612
User pablo@blueoakdb.com added comment
https://bugzilla.novell.com/show_bug.cgi?id=410612#c16
--- Comment #16 from Pablo Sanchez
https://bugzilla.novell.com/show_bug.cgi?id=410612
User trenn@novell.com added comment
https://bugzilla.novell.com/show_bug.cgi?id=410612#c17
Thomas Renninger
https://bugzilla.novell.com/show_bug.cgi?id=410612
User pablo@blueoakdb.com added comment
https://bugzilla.novell.com/show_bug.cgi?id=410612#c18
--- Comment #18 from Pablo Sanchez
participants (1)
-
bugzilla_noreply@novell.com