On 05/10/2010 09:40 AM, Dan Magenheimer wrote:
>>>>> enlighten.c, 'd0 domain attempted wrmsr', 'Detected 1226681.732 MHz
>>>>> processor.'(a 1.2 TeraHz processor??? - the processor is actually a
>>> I'm wondering if this bothers delay loop timings, etc. ???
>> Dan, thoughts?
> This info is obtained from Xen via the shared_info struct, so
> I'd bet the shared_info struct is getting trashed.
> Or, for awhile, Jeremy had some code that made it possible
> for pvclock to use multiple shared_info structs to detect
> TSC skew.
I had a patch for multiple pvclock structures so that they could be
exported to userspace, but I screwed up the Xen side so that it was
trashing memory on context switch. But the use of pvclock in usermode
is fairly iffy anyway, since there are concerns about how to attain the
monotonicity guarantees required by clock_gettime, etc. So at the
moment, neither Xen nor the .32+ kernels have support for that.
Xen-devel mailing list