Re: [PATCH v2] riscv: skip csr restore if vcpu preempted reload

From: Andrew Jones <ajones@ventanamicro.com>
To: Jinyu Tang <tjytimi@163.com>
Cc: Anup Patel <anup@brainfault.org>,
	Atish Patra <atish.patra@linux.dev>,
	 Conor Dooley <conor.dooley@microchip.com>,
	Yong-Xuan Wang <yongxuan.wang@sifive.com>,
	 Paul Walmsley <paul.walmsley@sifive.com>,
	Nutty Liu <nutty.liu@hotmail.com>,
	kvm@vger.kernel.org,  kvm-riscv@lists.infradead.org,
	linux-riscv@lists.infradead.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH v2] riscv: skip csr restore if vcpu preempted reload
Date: Mon, 25 Aug 2025 06:17:51 -0500	[thread overview]
Message-ID: <20250825-a30edf6b9d0301b5fdbcfffa@orel> (raw)
In-Reply-To: <20250825110708.75474-1-tjytimi@163.com>

On Mon, Aug 25, 2025 at 07:07:08PM +0800, Jinyu Tang wrote:
> The kvm_arch_vcpu_load() function is called in two cases for riscv:
> 1. When entering KVM_RUN from userspace ioctl.
> 2. When a preempted VCPU is scheduled back.
> 
> In the second case, if no other KVM VCPU has run on this CPU since the
> current VCPU was preempted, the guest CSR (including AIA CSRS) values 
> are still valid in the hardware and do not need to be restored.
> 
> This patch is to skip the CSR write path when:
> 1. The VCPU was previously preempted
> (vcpu->scheduled_out == 1).
> 2. It is being reloaded on the same physical CPU
> (vcpu->arch.last_exit_cpu == cpu).
> 3. No other KVM VCPU has used this CPU in the meantime
> (vcpu == __this_cpu_read(kvm_former_vcpu)).
> 
> This reduces many CSR writes with frequent preemption on the same CPU.
> 
> Signed-off-by: Jinyu Tang <tjytimi@163.com>
> Reviewed-by: Nutty Liu <nutty.liu@hotmail.com>
> ---
>  v1 -> v2:
>  Apply the logic to aia csr load. Thanks for
>  Andrew Jones's advice.
> 
>  arch/riscv/kvm/vcpu.c | 13 +++++++++++--
>  1 file changed, 11 insertions(+), 2 deletions(-)
> 
> diff --git a/arch/riscv/kvm/vcpu.c b/arch/riscv/kvm/vcpu.c
> index f001e5640..e50d1f76c 100644
> --- a/arch/riscv/kvm/vcpu.c
> +++ b/arch/riscv/kvm/vcpu.c
> @@ -25,6 +25,8 @@
>  #define CREATE_TRACE_POINTS
>  #include "trace.h"
>  
> +static DEFINE_PER_CPU(struct kvm_vcpu *, kvm_former_vcpu);
> +
>  const struct _kvm_stats_desc kvm_vcpu_stats_desc[] = {
>  	KVM_GENERIC_VCPU_STATS(),
>  	STATS_DESC_COUNTER(VCPU, ecall_exit_stat),
> @@ -581,6 +583,10 @@ void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
>  	struct kvm_vcpu_csr *csr = &vcpu->arch.guest_csr;
>  	struct kvm_vcpu_config *cfg = &vcpu->arch.cfg;
>  
> +	if  (vcpu == __this_cpu_read(kvm_former_vcpu) &&
> +		vcpu->arch.last_exit_cpu == cpu)

Why was the vcpu->scheduled_out check dropped? The changelog doesn't
mention that and the commit message still states that it is checked.

Thanks,
drew

> +		goto csr_restore_done;
> +
>  	if (kvm_riscv_nacl_sync_csr_available()) {
>  		nsh = nacl_shmem();
>  		nacl_csr_write(nsh, CSR_VSSTATUS, csr->vsstatus);
> @@ -624,6 +630,9 @@ void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
>  
>  	kvm_riscv_mmu_update_hgatp(vcpu);
>  
> +	kvm_riscv_vcpu_aia_load(vcpu, cpu);
> +
> +csr_restore_done:
>  	kvm_riscv_vcpu_timer_restore(vcpu);
>  
>  	kvm_riscv_vcpu_host_fp_save(&vcpu->arch.host_context);
> @@ -633,8 +642,6 @@ void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
>  	kvm_riscv_vcpu_guest_vector_restore(&vcpu->arch.guest_context,
>  					    vcpu->arch.isa);
>  
> -	kvm_riscv_vcpu_aia_load(vcpu, cpu);
> -
>  	kvm_make_request(KVM_REQ_STEAL_UPDATE, vcpu);
>  
>  	vcpu->cpu = cpu;
> @@ -645,6 +652,8 @@ void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu)
>  	void *nsh;
>  	struct kvm_vcpu_csr *csr = &vcpu->arch.guest_csr;
>  
> +	__this_cpu_write(kvm_former_vcpu, vcpu);
> +
>  	vcpu->cpu = -1;
>  
>  	kvm_riscv_vcpu_aia_put(vcpu);
> -- 
> 2.43.0
> 

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