Re: [PATCH v20 04/14] KVM: arm64: Ensure GCS memory effects are visible

["Lorenzo Stoakes (ARM)" <ljs@kernel.org>]

On Tue, Sep 01, 2026 at 10:47:02PM +0100, Mark Brown wrote:
> In order to allow efficient implementation of GCS implementations are
> permitted to do GCS specific caching, with barriers implicit in stack
> switch operations and GCSB DSYNC as an explicit barrier.  Since we do
> not use GCS in the hypervisor or host kernel the hypervisor cannot rely
> on any barriers being present in vCPU context switches other than those
> it explicitly inserts.  Add explicit barriers when loading and saving
> vCPU state.
>
> Signed-off-by: Mark Brown <broonie@kernel.org>

LGTM so:

Reviewed-by: Lorenzo Stoakes (ARM) <ljs@kernel.org>

> Reviewed-by: Yuan Yao <yaoyuan@linux.alibaba.com>
> ---
>  arch/arm64/kvm/arm.c               | 7 +++++++
>  arch/arm64/kvm/hyp/nvhe/hyp-main.c | 7 +++++++
>  2 files changed, 14 insertions(+)
>
> diff --git a/arch/arm64/kvm/arm.c b/arch/arm64/kvm/arm.c
> index 8b080804bc90..52f53f682e6a 100644
> --- a/arch/arm64/kvm/arm.c
> +++ b/arch/arm64/kvm/arm.c
> @@ -755,6 +755,13 @@ void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu)

So this seems to be where -> vCPU unscheduled so the right place to do it.

>  			vcpu_set_flag(vcpu, PKVM_HOST_STATE_DIRTY);
>  	}
>
> +	/*
> +	 * Ensure any GCS memory effects from the outgoing vCPU are
> +	 * visible elsewhere.
> +	 */
> +	if (kvm_has_gcs(vcpu->kvm))
> +		gcsb_dsync();

Makes sense.

I did notice the curious:

static inline void gcsb_dsync(void)
{
	asm volatile(".inst 0xd503227f" : : : "memory");
}

And various sludging around suggests this is HINT #19 and guaranteed to be a nop
on arches that don't implement the feature :)

> +
>  	kvm_vcpu_put_debug(vcpu);
>  	kvm_arch_vcpu_put_fp(vcpu);
>  	if (has_vhe())
> diff --git a/arch/arm64/kvm/hyp/nvhe/hyp-main.c b/arch/arm64/kvm/hyp/nvhe/hyp-main.c
> index 9a3b92e626ad..534db037c5f6 100644
> --- a/arch/arm64/kvm/hyp/nvhe/hyp-main.c
> +++ b/arch/arm64/kvm/hyp/nvhe/hyp-main.c
> @@ -261,6 +261,13 @@ static void sync_hyp_vcpu(struct pkvm_hyp_vcpu *hyp_vcpu)

And here similar for pKVM.

>  	fpsimd_sve_sync(&hyp_vcpu->vcpu);
>  	sync_debug_state(hyp_vcpu);
>
> +	/*
> +	 * Ensure any GCS memory effects from the outgoing vCPU are
> +	 * visible elsewhere even if the host skips syncing.
> +	 */
> +	if (kvm_has_gcs(hyp_vcpu->vcpu.kvm))
> +		gcsb_dsync();
> +
>  	if (pkvm_hyp_vcpu_is_protected(hyp_vcpu)) {
>  		host_vcpu->arch.ctxt = hyp_vcpu->vcpu.arch.ctxt;
>  	} else {
>
> --
> 2.47.3
>
>

--
Cheers, Lorenzo