On Mon, Dec 18, 2023 at 04:48:04PM -0800, Atish Patra wrote:
> On Sun, Dec 17, 2023 at 12:40 PM Andrew Jones <ajones@xxxxxxxxxxxxxxxx> wrote:
...
> > +int __init pv_time_init(void)
> > +{
> > + int ret;
> > +
> > + if (!has_pv_steal_clock())
> > + return 0;
> > +
> > + ret = cpuhp_setup_state(CPUHP_AP_ONLINE_DYN,
> > + "riscv/pv_time:online",
> > + pv_time_cpu_online,
> > + pv_time_cpu_down_prepare);
> > + if (ret < 0)
> > + return ret;
> > +
> > + static_call_update(pv_steal_clock, pv_time_steal_clock);
> > +
> > + static_key_slow_inc(¶virt_steal_enabled);
> > + if (steal_acc)
> > + static_key_slow_inc(¶virt_steal_rq_enabled);
> > +
> > + pr_info("using paravirt steal-time\n");
> > +
>
> Nit comment:
> If the scheduler doesn't take stolen time into consideration (when If
> no-steal-acc is specified in the command line),
> the above log is a bit misleading ? ARM64 also seems to be doing the
> same thing.
>
> Maybe the intention of the log is to steal time extension presence
> rather than actual usage in place ?
You're right that the word "using" isn't great. "Computing" would be
better as, even with no-steal-acc, steal time gets computed and shows
up in /proc/stat.
>
> > + return 0;
> > +}
> > diff --git a/arch/riscv/kernel/time.c b/arch/riscv/kernel/time.c
> > index 23641e82a9df..ba3477197789 100644
> > --- a/arch/riscv/kernel/time.c
> > +++ b/arch/riscv/kernel/time.c
> > @@ -12,6 +12,7 @@
> > #include <asm/sbi.h>
> > #include <asm/processor.h>
> > #include <asm/timex.h>
> > +#include <asm/paravirt.h>
> >
> > unsigned long riscv_timebase __ro_after_init;
> > EXPORT_SYMBOL_GPL(riscv_timebase);
> > @@ -45,4 +46,6 @@ void __init time_init(void)
> > timer_probe();
> >
> > tick_setup_hrtimer_broadcast();
> > +
> > + pv_time_init();
> > }
> > --
> > 2.43.0
> >
>
> Other than that, lgtm.
>
> Reviewed-by: Atish Patra <atishp@xxxxxxxxxxxx>
Thanks,
drew
