Created attachment 494449 [details] output of vmstat, dmesg, dmidecode and contents of /proc interrupts Description of problem: Getting tons of acpi interrupts (attached /proc/interrupts contents after 1-2 minutes of uptime) makes system really slow. Affected hardware is Lenovo IdeaPad Y560p laptop with Intel sandy bridge cpu and hm65 express chipset. Haven't checked BIOS revision but there are no officially available updates. Version-Release number of selected component (if applicable): Any 2.6.38 and later fedora kernels (even in --with vanilla option). 2.6.37 and older from koji works fine. How reproducible: Always. But for me it seems like hardware-specific issue (had no possibility to check with another similar laptop though). Steps to Reproduce: 1. Just boot any affected kernel with any userspace (tried installing fc15 kernels on fc14 and booting fc15 beta livecd). 2. Check vmstat output. Actual results: In attached vmstat output you can see about 25000 interrupts per second, and 50000 context switches (in idle). Most of them are from acpi. Expected results: Low amount of acpi interrupts (irq9). Reasonable values of in and cs counters in vmstat output. Additional info: Checked all acpi-related kernel parameters. Only effective was acpi=off. Feel free to ask me for any additional data or testing.
Created attachment 494484 [details] /proc/interrupts Please don't upload tarballs, it just means someone has to download them, untar, and re-upload as text.
Created attachment 494485 [details] dmesg
This looks wrong: [ 1.099139] \_SB_.PCI0:_OSC invalid UUID [ 1.099140] _OSC request data:1 8 1f [ 1.111213] \_SB_.PCI0:_OSC invalid UUID [ 1.111215] _OSC request data:1 1f 1f [ 1.111219] pci0000:00: Requesting ACPI _OSC control (0x1d) [ 1.111262] \_SB_.PCI0:_OSC invalid UUID [ 1.111263] _OSC request data:1 0 1d
Sorry, didn't know about tarballs. It's my first time using bugzilla. Found first affected commit: http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.38.y.git;a=commit;h=bba63a296ffab20e08d9e8252d2f0d99050ac859
(In reply to comment #4) > Sorry, didn't know about tarballs. It's my first time using bugzilla. > > Found first affected commit: > http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.38.y.git;a=commit;h=bba63a296ffab20e08d9e8252d2f0d99050ac859 That's: Subject: ACPICA: Implicit notify support X-Git-Tag: v2.6.38-rc2~28^2~9 Which is certainly plausible.
I have this same bug. Still exists as of 2.6.39-rc5, which includes Matthew Garett's previous patch for the same issue on a different laptop.
Created attachment 500866 [details] lspci -vvxxx
Created attachment 500869 [details] acpidump
Could you please also attach the output of grep -r . /sys/firmware/acpi ?
Created attachment 500874 [details] grep -r . /sys/firmware/acpi
(In reply to comment #4) > Found first affected commit: > http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.38.y.git;a=commit;h=bba63a296ffab20e08d9e8252d2f0d99050ac859 I tried using a 2.6.37.6 kernel which does not have the above mentioned implicit notify support, but I am still seeing the interrupt storm. How did you trace it to that particular commit?
Does this still happen on 2.6.43/3.3 in F15/F16?
Fedora 15 has reached it's end of life as of June 26, 2012. As a result, we will not be fixing any remaining bugs found in Fedora 15. In the event that you have upgraded to a newer release and the bug you reported is still present, please reopen the bug and set the version field to the newest release you have encountered the issue with. Before doing so, please ensure you are testing the latest kernel update in that release and attach any new and relevant information you may have gathered. Thank you for taking the time to file a report. We hope newer versions of Fedora suit your needs.