RE: [suse-amd64] many lost ticks
Last thing when I look at the output of dmesg I get the
following (see
below), I am slightly worried, since I have read somewhere that the kernel needs to be patched to include a more recent version of powernow-k8. Is this error expected, or should I patch?
Known problem, but harmless. Just ignore for now. Some future update kernel will likely fix it.
Andi - I can submit a patch, but I'm not sure what to do. The code loops through NR_CPUS and ignores the ones that are not online: if (!cpu_online(i)) continue; This code used to work, but the hotplug changes broke it. What's the correct code now to check if a cpu isn't available? -Mark Langsdorf AMD, Inc.
On Friday 28 October 2005 17:54, Langsdorf, Mark wrote:
I can submit a patch, but I'm not sure what to do.
The code loops through NR_CPUS and ignores the ones that are not online: if (!cpu_online(i)) continue;
This code used to work, but the hotplug changes broke it. What's the correct code now to check if a cpu isn't available?
cpu_online should still work as before. The current theory is that something in sysfs uses the possible map (which has all 128 possible hotplug CPUs) and that causes the messages. -Andi
Hi Andi, Mark, Can confirm that if I remove the nostc argument to the boot command, I do start to see the following in the output from dmesg: warning: many lost ticks. Your time source seems to be instable or some driver is hogging interupts rip acpi_processor_idle+0x12f/0x37f [processor] So as you suspected the only thing that seems to make a difference is nostc. Though I wonder if the error message is use dependent, I left the machine running over the weekend, and didn't gain too much time on the clock, and only had the one warning with dmesg. Still wish I knew what on earth was going on :) Best wishes, Jon. Langsdorf, Mark wrote:
Last thing when I look at the output of dmesg I get the following (see below), I am slightly worried, since I have read somewhere that the kernel needs to be patched to include a more recent version of powernow-k8. Is this error expected, or should I patch? Known problem, but harmless. Just ignore for now. Some future update kernel will likely fix it.
Andi -
I can submit a patch, but I'm not sure what to do.
The code loops through NR_CPUS and ignores the ones that are not online: if (!cpu_online(i)) continue;
This code used to work, but the hotplug changes broke it. What's the correct code now to check if a cpu isn't available?
-Mark Langsdorf AMD, Inc.
-- Jonathan Brooks (Ph.D.) Research Assistant. PaIN Group, Department of Human Anatomy & Genetics, University of Oxford, South Parks Road, Oxford, OX1 3QX tel: +44(0)1865-282654 fax: +44(0)1865-282656 web: http://www.fmrib.ox.ac.uk/~jon
Am Mo 31.10.2005 11:27 schrieb Jonathan Brooks <jonathan.brooks@human-anatomy.oxford.ac.uk>:
Hi Andi, Mark,
Can confirm that if I remove the nostc argument to the boot command, I do start to see the following in the output from dmesg:
warning: many lost ticks. Your time source seems to be instable or some driver is hogging interupts rip acpi_processor_idle+0x12f/0x37f [processor]
So as you suspected the only thing that seems to make a difference is nostc. Though I wonder if the error message is use dependent, I left the machine running over the weekend, and didn't gain too much time on the clock, and only had the one warning with dmesg. Still wish I knew what on earth was going on :)
You actually used notsc, not nostc, right? Just double checking because you consistently misspelled it. -Andi
participants (4)
-
Andi Kleen -
Andreas Kleen
-
Jonathan Brooks -
Langsdorf, Mark