https://bugzilla.novell.com/show_bug.cgi?id=684751 https://bugzilla.novell.com/show_bug.cgi?id=684751#c Jeff Mahoney changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |jeffm@novell.com AssignedTo|kernel-maintainers@forge.pr |tonyj@novell.com |ovo.novell.com | -- 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=684751 https://bugzilla.novell.com/show_bug.cgi?id=684751#c2 Tony Jones changed: What |Removed |Added ---------------------------------------------------------------------------- Status|ASSIGNED |NEEDINFO InfoProvider| |ensonic@sonicpulse.de --- Comment #2 from Tony Jones 2011-09-05 03:13:32 UTC --- What does dmesg say? Counter 0 not available is usually because it believes the bios has "stolen" that counter. There should be a message. -- 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=684751 https://bugzilla.novell.com/show_bug.cgi?id=684751#c4 --- Comment #4 from Stefan Kost 2011-09-22 20:10:19 UTC --- Sep 22 22:10:22 ada sudo: ensonic : TTY=pts/1 ; PWD=/home/ensonic/projects/buzztard/build-tools/glick ; USER=root ; COMMAND=/usr/bin/opcontrol --start Sep 22 22:10:23 ada kernel: [40546.560929] [Firmware Bug]: cpu 0, try to setup vector 0x10400, but vector 0xf9 was already reserved by another core, APIC500=0xf9 Sep 22 22:10:23 ada kernel: [40546.560932] [Firmware Bug]: cpu 0, IBS interrupt offset 0 not available (MSRC001103A=0x0000000000000100) Sep 22 22:10:23 ada kernel: [40546.560943] [Firmware Bug]: workaround enabled for IBS LVT offset Sep 22 22:10:23 ada kernel: [40546.560945] oprofile: AMD IBS detected (0x0000001f) Sep 22 22:10:23 ada kernel: [40546.560954] oprofile: using NMI interrupt. -- 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=684751 https://bugzilla.novell.com/show_bug.cgi?id=684751#c5 --- Comment #5 from Tony Jones 2011-10-07 00:07:29 UTC --- Hmmn. I think I see the issue. Not sure what is happening kernel side yet. assuming: # cat /proc/sys/kernel/watchdog produces 1 try: # echo 0 > /proc/sys/kernel/watchdog # opcontrol --deinit # opcontrol --start ..... If the above is correct, the issue is that nmi_watchdog appears to have disappeared. The opcontrol script is trying to adjust the file to 0 but it doesn't exist. -- 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=684751 https://bugzilla.novell.com/show_bug.cgi?id=684751#c7 --- Comment #7 from Tony Jones 2012-01-05 16:06:51 UTC --- (In reply to comment #6)

If you do sed -i 's/nmi_watchdog/watchdog/g' /usr/bin/opcontrol, at least opcontrol will tell you to do that. I'm submitting a request to devel:tools:oprofile right now with the patch.

How did this patch get accepted to devel_tools? nmi_watchdog dissapearing was a kernel blip and it is now once again present in the latest factory kernel. reverting change out of devel_tools. -- 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.