qemu-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [PATCH v11 4/4] softmmu/dirtylimit: implement dirty page rate limit


From: Peter Xu
Subject: Re: [PATCH v11 4/4] softmmu/dirtylimit: implement dirty page rate limit
Date: Mon, 17 Jan 2022 15:35:55 +0800

On Wed, Jan 05, 2022 at 01:14:09AM +0800, huangy81@chinatelecom.cn wrote:
>  ##
> +# @set-vcpu-dirty-limit:
> +#
> +# Set the upper limit of dirty page rate for virtual CPU.
> +#
> +# Requires KVM with accelerator property "dirty-ring-size" set.
> +# A virtual CPU's dirty page rate is a measure of its memory load.
> +# To observe dirty page rates, use @calc-dirty-rate.
> +#
> +# @cpu-index: index of virtual CPU, default is all.
> +#
> +# @dirty-rate: upper limit of dirty page rate for virtual CPU.

Same here, we could add the unit of dirty rate (MB/s), and in all the rest of
the places around dirty rate.

> +void qmp_set_vcpu_dirty_limit(bool has_cpu_index,
> +                              uint64_t cpu_index,
> +                              uint64_t dirty_rate,
> +                              Error **errp)
> +{
> +    if (!kvm_enabled() || !kvm_dirty_ring_enabled()) {
> +        error_setg(errp, "dirty page limit feature requires KVM with"
> +                   " accelerator property 'dirty-ring-size' set'");
> +        return;
> +    }
> +
> +    if (!dirtylimit_in_service()) {
> +        vcpu_dirty_rate_stat_initialize();
> +        vcpu_dirty_rate_stat_start();
> +    }
> +
> +    if (has_cpu_index && !dirtylimit_vcpu_index_valid(cpu_index)) {
> +        error_setg(errp, "incorrect cpu index specified");
> +        return;
> +    }

This could be moved before above vcpu_dirty_rate_stat_initialize() calls.  I've
left some comments in previous patch on this; please refer to the pesudo code.

The rest looks good to me.

Thanks,

> +
> +    if (has_cpu_index) {
> +        dirtylimit_set_vcpu(cpu_index, dirty_rate, true);
> +    } else {
> +        dirtylimit_set_all(dirty_rate, true);
> +    }
> +}

-- 
Peter Xu




reply via email to

[Prev in Thread] Current Thread [Next in Thread]