[PATCH v3 08/18] KVM: arm64: Introduce hyp context

Marc Zyngier maz at kernel.org
Mon Sep 7 09:29:11 EDT 2020


On Thu, 03 Sep 2020 14:52:57 +0100,
Andrew Scull <ascull at google.com> wrote:
> 
> During __guest_enter, save and restore from a new hyp context rather
> than the host context. This is preparation for separation of the hyp and
> host context in nVHE.
> 
> Signed-off-by: Andrew Scull <ascull at google.com>
> ---
>  arch/arm64/include/asm/kvm_hyp.h        |  3 ++-
>  arch/arm64/kernel/image-vars.h          |  1 +
>  arch/arm64/kvm/arm.c                    | 10 ++++++++++
>  arch/arm64/kvm/hyp/entry.S              | 10 +++++-----
>  arch/arm64/kvm/hyp/include/hyp/switch.h |  2 +-
>  arch/arm64/kvm/hyp/nvhe/switch.c        |  2 +-
>  arch/arm64/kvm/hyp/vhe/switch.c         |  2 +-
>  7 files changed, 21 insertions(+), 9 deletions(-)
> 
> diff --git a/arch/arm64/include/asm/kvm_hyp.h b/arch/arm64/include/asm/kvm_hyp.h
> index 1e2491da324e..0b525e05e5bf 100644
> --- a/arch/arm64/include/asm/kvm_hyp.h
> +++ b/arch/arm64/include/asm/kvm_hyp.h
> @@ -12,6 +12,7 @@
>  #include <asm/alternative.h>
>  #include <asm/sysreg.h>
>  
> +DECLARE_PER_CPU(struct kvm_cpu_context, kvm_hyp_ctxt);
>  DECLARE_PER_CPU(unsigned long, kvm_hyp_vector);
>  
>  #define read_sysreg_elx(r,nvh,vh)					\
> @@ -89,7 +90,7 @@ void activate_traps_vhe_load(struct kvm_vcpu *vcpu);
>  void deactivate_traps_vhe_put(void);
>  #endif
>  
> -u64 __guest_enter(struct kvm_vcpu *vcpu, struct kvm_cpu_context *host_ctxt);
> +u64 __guest_enter(struct kvm_vcpu *vcpu);
>  
>  void __noreturn hyp_panic(void);
>  #ifdef __KVM_NVHE_HYPERVISOR__
> diff --git a/arch/arm64/kernel/image-vars.h b/arch/arm64/kernel/image-vars.h
> index 54bb0eb34b0f..9f419e4fc66b 100644
> --- a/arch/arm64/kernel/image-vars.h
> +++ b/arch/arm64/kernel/image-vars.h
> @@ -71,6 +71,7 @@ KVM_NVHE_ALIAS(kvm_update_va_mask);
>  /* Global kernel state accessed by nVHE hyp code. */
>  KVM_NVHE_ALIAS(arm64_ssbd_callback_required);
>  KVM_NVHE_ALIAS(kvm_host_data);
> +KVM_NVHE_ALIAS(kvm_hyp_ctxt);
>  KVM_NVHE_ALIAS(kvm_hyp_vector);
>  KVM_NVHE_ALIAS(kvm_vgic_global_state);
>  
> diff --git a/arch/arm64/kvm/arm.c b/arch/arm64/kvm/arm.c
> index b6442c6be5ad..ae4b34f91e94 100644
> --- a/arch/arm64/kvm/arm.c
> +++ b/arch/arm64/kvm/arm.c
> @@ -47,6 +47,7 @@ __asm__(".arch_extension	virt");
>  #endif
>  
>  DEFINE_PER_CPU(struct kvm_host_data, kvm_host_data);
> +DEFINE_PER_CPU(struct kvm_cpu_context, kvm_hyp_ctxt);

[back to this patch after having reviewed a few of the subsequent
ones]

Given the variety of contexts you are introducing, I wonder if the
best course of action for most of this isn't simply to use the EL2
stack rather than defining ad-hoc structures.

The host save/restore you are introducing in a later patch certainly
could do without a separate data structure (hello, struct pt_regs) and
the hyp/host churn.

What do you think?

	M.

-- 
Without deviation from the norm, progress is not possible.



More information about the linux-arm-kernel mailing list