On 07/13/2018 12:40 PM, Arjan van de Ven wrote:
> On 7/13/2018 12:19 PM, patrickg wrote:
>> This RFC patch is intended to allow bypass CPUID, MSR and QuickPIT
>> calibration methods should the user desire to.
>>
>> The current ordering in ML x86 tsc is to calibrate in the order listed
>> above; returning whenever there's a successful calibration. However there
>> are certain BIOS/HW Designs for overclocking that cause the TSC to change
>> along with the max core clock; and simple 'trusting' calibration
>> methodologies will lead to the TSC running 'faster' and eventually, TSC
>> instability.
>>
>
>
> that would be a real violation of the contract between cpu and OS: tsc is not
> supposed to change for the duration of the boot
With the methodology used; the TSC is still invariant; it's just running faster
than the CPUID math calculates.
>
>> I only know that there's a use-case for me to want to be able to skip CPUID
>> calibration, however I included args for skipping all the rest just so that
>> all functionality is covered in the long run instead of just one use-case.
>
> wouldn't it be better to start the detailed calibration with the value from
> CPUID instead; that way we also properly calibrate spread spectrum etc...
>
> I thought we switched to that recently to be honest...
Are you referencing:
1bf8915ae5156dff439d2c65314bd8fdde1b83bf - x86/tsc: Enumerate SKL cpu_khz and
tsc_khz via CPUID
However since it's returning at CPUID calibration during
native_calibrate_cpu(); it's not compared after-the-fact, leading to the TSC to
use the 'slower' number returned by CPUID.
Now keep in mind; I dunno if there was any reason to explicitly not want to
utilize the PIT calib sequences on SKL. That'd be a factor for this.
Would comparing the number after the fact; then if there's a significant
difference between PIT and MSR/CPUID, defaulting to the 'faster' value be a
better solution?