Hello,

I currently have a very simple application (no RTOS) running on a 
single-core x86 inmate that is unaware of P-states. I am also running with 
Hardware P-states (HWP) disabled on the Linux command line. When the root 
cell gives control of the core over to the inmate, it's unclear to me what 
happens to the P-state of that core. If it has an idle P-state, and then is 
handed off to inmate, will it still be at the same P-state? Or will it get 
reset to the full-throttle P0 P-state?

In my testing, my inmate appears to be running full throttle at the base 
frequency, meaning that the P-state gets reset to P0 somehow. But I'm not 
sure how to verify this beyond calculating the frequency with aperf and 
mperf, which I'm doing.

I guess I could dive into the Linux source code to see what the 
intel_pstate driver is doing, but I am a bit nervous that I might not 
understand it. So I'm hoping someone here knows how this handoff works. 
Maybe there is some kind of watchdog timer that needs to be actively reset 
by the intel_pstate driver to keep a core at a P-state other than P0.

Thanks,
Michael

-- 
You received this message because you are subscribed to the Google Groups 
"Jailhouse" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jailhouse-dev/5cd0fb26-8c4f-482c-9ac9-6f6b1d234f03%40googlegroups.com.

Reply via email to