Re: [PATCH v6 2/4] arm64: KVM: add accessors to track guest/host only counters

From: Christoffer Dall <christoffer.dall@arm.com>
To: Andrew Murray <andrew.murray@arm.com>
Cc: Mark Rutland <mark.rutland@arm.com>,
	Julien Thierry <julien.thierry@arm.com>,
	Marc Zyngier <marc.zyngier@arm.com>,
	Catalin Marinas <catalin.marinas@arm.com>,
	Suzuki K Poulose <suzuki.poulose@arm.com>,
	Will Deacon <will.deacon@arm.com>,
	kvmarm@lists.cs.columbia.edu,
	linux-arm-kernel@lists.infradead.org
Subject: Re: [PATCH v6 2/4] arm64: KVM: add accessors to track guest/host only counters
Date: Mon, 10 Dec 2018 11:26:34 +0100	[thread overview]
Message-ID: <20181210102634.GL30263@e113682-lin.lund.arm.com> (raw)
In-Reply-To: <1544435159-54781-3-git-send-email-andrew.murray@arm.com>

On Mon, Dec 10, 2018 at 09:45:57AM +0000, Andrew Murray wrote:
> In order to effeciently enable/disable guest/host only perf counters
> at guest entry/exit we add bitfields to kvm_cpu_context for guest and
> host events as well as accessors for updating them.
> 
> Signed-off-by: Andrew Murray <andrew.murray@arm.com>
> ---
>  arch/arm64/include/asm/kvm_host.h | 24 ++++++++++++++++++++++++
>  1 file changed, 24 insertions(+)
> 
> diff --git a/arch/arm64/include/asm/kvm_host.h b/arch/arm64/include/asm/kvm_host.h
> index 1550192..800c87b 100644
> --- a/arch/arm64/include/asm/kvm_host.h
> +++ b/arch/arm64/include/asm/kvm_host.h
> @@ -203,6 +203,8 @@ struct kvm_cpu_context {
>  	};
>  
>  	struct kvm_vcpu *__hyp_running_vcpu;
> +	u32 events_host;
> +	u32 events_guest;

This is confusing to me.

These values appear only to be used for the host instance, which makes
me wonder why we add them to kvm_cpu_context, which is also used for the
guest state?  Should we not instead consider moving them to their own
data structure and add a per-cpu data structure or something more fancy
like having a new data structure, kvm_percpu_host_data, which contains
the kvm_cpu_context and the events flags?

I don't know much about perf, but doesn't this design also imply that
you can only set these modifiers at a per-cpu level, and not attach
the modifiers to a task/vcpu or vm ?  Is that by design?

Thanks,

    Christoffer

>  };
>  
>  typedef struct kvm_cpu_context kvm_cpu_context_t;
> @@ -467,11 +469,33 @@ void kvm_arch_vcpu_load_fp(struct kvm_vcpu *vcpu);
>  void kvm_arch_vcpu_ctxsync_fp(struct kvm_vcpu *vcpu);
>  void kvm_arch_vcpu_put_fp(struct kvm_vcpu *vcpu);
>  
> +#define KVM_PMU_EVENTS_HOST	1
> +#define KVM_PMU_EVENTS_GUEST	2
> +
>  #ifdef CONFIG_KVM /* Avoid conflicts with core headers if CONFIG_KVM=n */
>  static inline int kvm_arch_vcpu_run_pid_change(struct kvm_vcpu *vcpu)
>  {
>  	return kvm_arch_vcpu_run_map_fp(vcpu);
>  }
> +static inline void kvm_set_pmu_events(u32 set, int flags)
> +{
> +	kvm_cpu_context_t *ctx = this_cpu_ptr(&kvm_host_cpu_state);
> +
> +	if (flags & KVM_PMU_EVENTS_HOST)
> +		ctx->events_host |= set;
> +	if (flags & KVM_PMU_EVENTS_GUEST)
> +		ctx->events_guest |= set;
> +}
> +static inline void kvm_clr_pmu_events(u32 clr)
> +{
> +	kvm_cpu_context_t *ctx = this_cpu_ptr(&kvm_host_cpu_state);
> +
> +	ctx->events_host &= ~clr;
> +	ctx->events_guest &= ~clr;
> +}
> +#else
> +static inline void kvm_set_pmu_events(u32 set, int flags) {}
> +static inline void kvm_clr_pmu_events(u32 clr) {}
>  #endif
>  
>  static inline void kvm_arm_vhe_guest_enter(void)
> -- 
> 2.7.4
> 

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel