On Mon, Aug 29, 2022 at 12:02:42PM -0500, Scott Cheloha wrote: > If hv_delay() never causes a vm exit, but tsc_delay() *might* cause a > vm exit, and both have microsecond or better resolution, then > hv_delay() is the preferable delay(9) implementation where it is > available because vm exits have ambiguous overhead.
with hv_delay() currently doing rdmsr I wouldn't say never > > If that seems sensible to you, I'll commit this switch. There is a msr to allow cpuid to report invariant tsc (0x40000118). Used by linux but not documented in the hyper-v tlfs. Without that, tsc delay is never used on hyper-v. So leave it as is until someone running hyper-v/azure would like it changed? > > Index: arch/amd64/amd64/tsc.c > =================================================================== > RCS file: /cvs/src/sys/arch/amd64/amd64/tsc.c,v > retrieving revision 1.26 > diff -u -p -r1.26 tsc.c > --- arch/amd64/amd64/tsc.c 25 Aug 2022 17:38:16 -0000 1.26 > +++ arch/amd64/amd64/tsc.c 29 Aug 2022 16:58:25 -0000 > @@ -109,7 +109,7 @@ tsc_identify(struct cpu_info *ci) > > tsc_frequency = tsc_freq_cpuid(ci); > if (tsc_frequency > 0) > - delay_init(tsc_delay, 5000); > + delay_init(tsc_delay, 4000); > } > > static inline int > Index: dev/pv/pvbus.c > =================================================================== > RCS file: /cvs/src/sys/dev/pv/pvbus.c,v > retrieving revision 1.25 > diff -u -p -r1.25 pvbus.c > --- dev/pv/pvbus.c 25 Aug 2022 17:38:16 -0000 1.25 > +++ dev/pv/pvbus.c 29 Aug 2022 16:58:26 -0000 > @@ -320,7 +320,7 @@ pvbus_hyperv(struct pvbus_hv *hv) > > #if NHYPERV > 0 > if (hv->hv_features & CPUID_HV_MSR_TIME_REFCNT) > - delay_init(hv_delay, 4000); > + delay_init(hv_delay, 5000); > #endif > } > > >
