[libvirt] [PATCH] test_driver: implement virDomainGetCPUStats

[Erik Skultety]

On Fri, Jul 26, 2019 at 01:24:39PM +0200, Ilias Stamatis wrote:
> On Thu, Jul 25, 2019 at 2:01 PM Erik Skultety <eskultet at redhat.com> wrote:
> >
> > On Thu, Jul 18, 2019 at 12:02:43PM +0200, Ilias Stamatis wrote:
> > > Signed-off-by: Ilias Stamatis <stamatis.iliass at gmail.com>
> > > ---
> > >  src/test/test_driver.c | 131 +++++++++++++++++++++++++++++++++++++++++
> > >  1 file changed, 131 insertions(+)
> > >
> > > diff --git a/src/test/test_driver.c b/src/test/test_driver.c
> > > index fcb80c9e47..2907c043cb 100644
> > > --- a/src/test/test_driver.c
> > > +++ b/src/test/test_driver.c
> > > @@ -3258,6 +3258,136 @@ static int testDomainSetMetadata(virDomainPtr dom,
> > >      return ret;
> > >  }
> > >
> > > +
> > > +static int
> > > +testDomainGetDomainTotalCpuStats(virTypedParameterPtr params,
> > > +                                int nparams)
> > > +{
> > > +    if (nparams == 0) /* return supported number of params */
> > > +        return 3;
> > > +
> > > +    if (virTypedParameterAssign(&params[0], VIR_DOMAIN_CPU_STATS_CPUTIME,
> > > +                                VIR_TYPED_PARAM_ULLONG, 77102913900) < 0)
> > > +        return -1;
> > > +
> > > +    if (nparams > 1 &&
> > > +        virTypedParameterAssign(&params[1],
> > > +                                VIR_DOMAIN_CPU_STATS_USERTIME,
> > > +                                VIR_TYPED_PARAM_ULLONG, 45650000000) < 0)
> > > +        return -1;
> > > +
> > > +    if (nparams > 2 &&
> > > +        virTypedParameterAssign(&params[2],
> > > +                                VIR_DOMAIN_CPU_STATS_SYSTEMTIME,
> > > +                                VIR_TYPED_PARAM_ULLONG, 11390000000) < 0)
> > > +        return -1;
> > > +
> > > +    if (nparams > 3)
> > > +        nparams = 3;
> > > +
> > > +    return nparams;
> > > +}
> > > +
> > > +
> > > +static int
> > > +testDomainGetPercpuStats(virTypedParameterPtr params,
> > > +                         unsigned int nparams,
> > > +                         int start_cpu,
> > > +                         unsigned int ncpus,
> > > +                         int total_cpus)
> > > +{
> > > +    size_t i;
> > > +    int need_cpus;
> > > +    int param_idx;
> > > +    int ret = -1;
> > > +
> > > +    /* return the number of supported params */
> > > +    if (nparams == 0 && ncpus != 0)
> > > +        return 2;
> > > +
> > > +    /* return total number of cpus */
> > > +    if (ncpus == 0) {
> > > +        ret = total_cpus;
> > > +        goto cleanup;
> > > +    }
> > > +
> > > +    if (start_cpu >= total_cpus) {
> > > +        virReportError(VIR_ERR_INVALID_ARG,
> > > +                       _("start_cpu %d larger than maximum of %d"),
> > > +                       start_cpu, total_cpus - 1);
> > > +        goto cleanup;
> > > +    }
> > > +
> > > +    /* return percpu cputime in index 0 */
> > > +    param_idx = 0;
> > > +
> > > +    /* number of cpus to compute */
> > > +    need_cpus = MIN(total_cpus, start_cpu + ncpus);
> > > +
> > > +    for (i = 0; i < need_cpus; i++) {
> > > +        if (i < start_cpu)
> > > +            continue;
> > > +        int idx = (i - start_cpu) * nparams + param_idx;
> > > +        if (virTypedParameterAssign(&params[idx],
> > > +                                    VIR_DOMAIN_CPU_STATS_CPUTIME,
> > > +                                    VIR_TYPED_PARAM_ULLONG,
> > > +                                    202542145062 + 10 * i) < 0)
> >
> > What's the reasoning behind the formula? I'm curious, wouldn't have
> > 202542145062 + i been enough? Anyhow, the CPUTIME should be a portion of the
> > total cputime of all CPUs, your per-CPU time is much bigger by the total,
> > that doesn't sound right. I don't care about the exact numbers, but I'd like to
> > see them to make sense.
> >
> > Erik
>
> The 10 * i was totally random. You are right, let's make them make sense.
>
> We can have a total CPUTIME and then divide that equally (?) by the
> number of CPUs.

I didn't want to do division really (in an ideal world we'd be okay with bit
shifts, but that's not the case unfortunately), but it's test driver, so
doesn't matter at all.

>
> The "problem" is that the number of CPUs can vary. So depending on the
> number of CPUs the test VM has been configured with, this API will
> return different numbers. Is that ok?

Yes it is, that is expected, it would be weird if a configuration with 4 vCPUs
returns data for 2. CPUTIME needs to be static though.

Erik

>
> Because I thought we would prefer to return the same fixed numbers
> every time. However then they won't make sense when the number of CPUs
> changes.
>
>
> Ilias
>
> --
> libvir-list mailing list
> libvir-list at redhat.com
> https://www.redhat.com/mailman/listinfo/libvir-list