Re: [Xen-devel] [PATCH v3 2/7] xen: credit: (micro) optimize csched_runq_steal().

[George Dunlap <George.Dunlap@xxxxxxxxxxxxx>]

On Fri, Apr 7, 2017 at 5:56 PM, Dario Faggioli
<dario.faggioli@xxxxxxxxxx> wrote:
> Chacking whether or not a vCPU can be 'stolen'

"Checking" -- I'll fix this one on check-in. :-)

> from a peer pCPU's runqueue is relatively cheap.
>
> Therefore, let's do that  as early as possible,
> avoiding potentially useless complex checks, and
> cpumask manipulations.
>
> Signed-off-by: Dario Faggioli <dario.faggioli@xxxxxxxxxx>

Reviewed-by: George Dunlap <george.dunlap@xxxxxxxxxx>

> ---
> Cc: George Dunlap <george.dunlap@xxxxxxxxxxxxx>
> Cc: Anshul Makkar <anshul.makkar@xxxxxxxxxx>
> ---
> Changes from v2:
> * add an assert enforcing the old !is_running check inside the
>   is_migrateable() function.
>
> Changes from v1:
> * fixed a typo in a comment.
> ---
>  xen/common/sched_credit.c |   22 ++++++++++++++--------
>  1 file changed, 14 insertions(+), 8 deletions(-)
>
> diff --git a/xen/common/sched_credit.c b/xen/common/sched_credit.c
> index 63a8675..59ed4ca 100644
> --- a/xen/common/sched_credit.c
> +++ b/xen/common/sched_credit.c
> @@ -708,12 +708,15 @@ static inline int
>  __csched_vcpu_is_migrateable(struct vcpu *vc, int dest_cpu, cpumask_t *mask)
>  {
>      /*
> -     * Don't pick up work that's in the peer's scheduling tail or hot on
> -     * peer PCPU. Only pick up work that prefers and/or is allowed to run
> -     * on our CPU.
> +     * Don't pick up work that's hot on peer PCPU, or that can't (or
> +     * would prefer not to) run on cpu.
> +     *
> +     * The caller is supposed to have already checked that vc is also
> +     * not running.
>       */
> -    return !vc->is_running &&
> -           !__csched_vcpu_is_cache_hot(vc) &&
> +    ASSERT(!vc->is_running);
> +
> +    return !__csched_vcpu_is_cache_hot(vc) &&
>             cpumask_test_cpu(dest_cpu, mask);
>  }
>
> @@ -1622,7 +1625,9 @@ csched_runq_steal(int peer_cpu, int cpu, int pri, int 
> balance_step)
>          BUG_ON( is_idle_vcpu(vc) );
>
>          /*
> -         * If the vcpu has no useful soft affinity, skip this vcpu.
> +         * If the vcpu is still in peer_cpu's scheduling tail, or if it
> +         * has no useful soft affinity, skip it.
> +         *
>           * In fact, what we want is to check if we have any "soft-affine
>           * work" to steal, before starting to look at "hard-affine work".
>           *
> @@ -1633,8 +1638,9 @@ csched_runq_steal(int peer_cpu, int cpu, int pri, int 
> balance_step)
>           * vCPUs with useful soft affinities in some sort of bitmap
>           * or counter.
>           */
> -        if ( balance_step == CSCHED_BALANCE_SOFT_AFFINITY
> -             && !__vcpu_has_soft_affinity(vc, vc->cpu_hard_affinity) )
> +        if ( vc->is_running ||
> +             (balance_step == CSCHED_BALANCE_SOFT_AFFINITY
> +              && !__vcpu_has_soft_affinity(vc, vc->cpu_hard_affinity)) )
>              continue;
>
>          csched_balance_cpumask(vc, balance_step, cpumask_scratch);
>
>
> _______________________________________________
> Xen-devel mailing list
> Xen-devel@xxxxxxxxxxxxx
> https://lists.xen.org/xen-devel

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
https://lists.xen.org/xen-devel