Re: [PATCH v3 2/2] timer/linux/x86: override TSC freq if no tsc_known_freq

[Bruce Richardson <bruce.richardson@intel.com>]

On Wed, Oct 02, 2024 at 12:59:58AM +0300, Isaac Boukris wrote:
> On Tue, Oct 1, 2024 at 11:01 PM Bruce Richardson
> <bruce.richardson@intel.com> wrote:
> >
> > On Tue, Oct 01, 2024 at 03:22:51AM +0300, Isaac Boukris wrote:
> > > If the tsc_known_freq cpu flag is missing, it means the kernel doesn't
> > > trust it and calculates its own. We should do the same to avoid drift.
> > >
> > > Signed-off-by: Isaac Boukris <iboukris@gmail.com>
> > > ---
> > >  lib/eal/common/eal_common_timer.c |  3 +-
> > >  lib/eal/common/eal_private.h      |  2 +-
> > >  lib/eal/freebsd/eal_timer.c       |  5 ++-
> > >  lib/eal/linux/eal_timer.c         | 53 +++++++++++++++++++++++++++++--
> > >  lib/eal/windows/eal_timer.c       |  5 ++-
> > >  5 files changed, 60 insertions(+), 8 deletions(-)
> > >
> > > diff --git a/lib/eal/common/eal_common_timer.c b/lib/eal/common/eal_common_timer.c
> > > index c5c4703f15..e00be0a5c8 100644
> > > --- a/lib/eal/common/eal_common_timer.c
> > > +++ b/lib/eal/common/eal_common_timer.c
> > > @@ -66,8 +66,7 @@ set_tsc_freq(void)
> > >       }
> > >
> > >       freq = get_tsc_freq_arch();
> > > -     if (!freq)
> > > -             freq = get_tsc_freq();
> > > +     freq = get_tsc_freq(freq);
> > >       if (!freq)
> > >               freq = estimate_tsc_freq();
> > >
> > > diff --git a/lib/eal/common/eal_private.h b/lib/eal/common/eal_private.h
> > > index af09620426..bb315dab04 100644
> > > --- a/lib/eal/common/eal_private.h
> > > +++ b/lib/eal/common/eal_private.h
> > > @@ -374,7 +374,7 @@ void set_tsc_freq(void);
> > >   *
> > >   * This function is private to the EAL.
> > >   */
> > > -uint64_t get_tsc_freq(void);
> > > +uint64_t get_tsc_freq(uint64_t arch_hz);
> > >
> > >  /**
> > >   * Get TSC frequency if the architecture supports.
> > > diff --git a/lib/eal/freebsd/eal_timer.c b/lib/eal/freebsd/eal_timer.c
> > > index 3dd70e24ba..5a8aea03e1 100644
> > > --- a/lib/eal/freebsd/eal_timer.c
> > > +++ b/lib/eal/freebsd/eal_timer.c
> > > @@ -26,12 +26,15 @@
> > >  enum timer_source eal_timer_source = EAL_TIMER_TSC;
> > >
> > >  uint64_t
> > > -get_tsc_freq(void)
> > > +get_tsc_freq(uint64_t arch_hz)
> > >  {
> > >       size_t sz;
> > >       int tmp;
> > >       uint64_t tsc_hz;
> > >
> > > +     if (arch_hz)
> > > +             return arch_hz;
> > > +
> > >       sz = sizeof(tmp);
> > >       tmp = 0;
> > >
> >
> > On FreeBSD I'm not sure this is the best behaviour. On BSD we read the TSC
> > value from the kernel, which, one assumes, has measured it accurately.
> > Therefore I'd tend toward just using the kernel value in all cases, maybe
> > check the arch value (if non-zero) against that and warning if they have
> > significant divergence. WDYT?
> 
> Makes sense, I'll add a patch for that. We could also use the arch
> value if for some reason the sysctlbyname() failed.

Yep, +1 to that.

/Bruce