Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1699607

Re: [RFC PATCH v2 22/38] KVM: arm64: Handle PSCI call via smc from the guest

From Christoffer Dall <cdall@linaro.org>
Newsgroups linux.kernel
Subject Re: [RFC PATCH v2 22/38] KVM: arm64: Handle PSCI call via smc from the guest
Date 2017-07-30 22:10 +0200
Message-ID <u92Hg-4Vv-13@gated-at.bofh.it> (permalink)
References <u4E0O-4S6-19@gated-at.bofh.it> <u4Eax-5aT-93@gated-at.bofh.it>
Organization linux.* mail to news gateway

Show all headers | View raw


On Tue, Jul 18, 2017 at 11:58:48AM -0500, Jintack Lim wrote:
> VMs used to execute hvc #0 for the psci call if EL3 is not implemented.
> However, when we come to provide the virtual EL2 mode to the VM, the
> host OS inside the VM calls kvm_call_hyp() which is also hvc #0. So,
> it's hard to differentiate between them from the host hypervisor's point
> of view.

This is a bit confusing.  I think you should just refer to the fact that
the architecture requires HVC calls to be handled at EL2, and when
emulating EL2 inside the VM, HVC calls from the VM are handled by the VM
itself, and therefore we add the support for SMC as the conduit for PSCI
calls.

> 
> So, let the VM execute smc instruction for the psci call. On ARMv8.3,
> even if EL3 is not implemented, a smc instruction executed at non-secure
> EL1 is trapped to EL2 if HCR_EL2.TSC==1, rather than being treated as
> UNDEFINED. So, the host hypervisor can handle this psci call without any
> confusion.
> 
> Signed-off-by: Jintack Lim <jintack.lim@linaro.org>
> ---
>  arch/arm64/kvm/handle_exit.c | 24 ++++++++++++++++++++++--
>  1 file changed, 22 insertions(+), 2 deletions(-)
> 
> diff --git a/arch/arm64/kvm/handle_exit.c b/arch/arm64/kvm/handle_exit.c
> index d19e253..6cf6b93 100644
> --- a/arch/arm64/kvm/handle_exit.c
> +++ b/arch/arm64/kvm/handle_exit.c
> @@ -53,8 +53,28 @@ static int handle_hvc(struct kvm_vcpu *vcpu, struct kvm_run *run)
>  
>  static int handle_smc(struct kvm_vcpu *vcpu, struct kvm_run *run)
>  {
> -	kvm_inject_undefined(vcpu);
> -	return 1;
> +	int ret;
> +
> +	/* If imm is non-zero, it's not defined */
> +	if (kvm_vcpu_hvc_get_imm(vcpu)) {
> +		kvm_inject_undefined(vcpu);
> +		return 1;
> +	}
> +
> +	/*
> +	 * If imm is zero, it's a psci call.

That's only a necessary, but not sufficient requirement.  So we should
say, it may be a PSCI call or we check if it's a PSCI call...



> +	 * Note that on ARMv8.3, even if EL3 is not implemented, SMC executed
> +	 * at Non-secure EL1 is trapped to EL2 if HCR_EL2.TSC==1, rather than
> +	 * being treated as UNDEFINED.
> +	 */
> +	ret = kvm_psci_call(vcpu);
> +	if (ret < 0) {
> +		kvm_inject_undefined(vcpu);
> +		return 1;
> +	}
> +	kvm_skip_instr(vcpu, kvm_vcpu_trap_il_is32bit(vcpu));
> +
> +	return ret;
>  }
>  
>  /*
> -- 
> 1.9.1
> 

Thanks,
-Christoffer

Back to linux.kernel | Previous | NextPrevious in thread | Find similar | Unroll thread


Thread

[RFC PATCH v2 22/38] KVM: arm64: Handle PSCI call via smc from the guest Jintack Lim <jintack.lim@linaro.org> - 2017-07-18 19:10 +0200
  Re: [RFC PATCH v2 22/38] KVM: arm64: Handle PSCI call via smc from  the guest Christoffer Dall <cdall@linaro.org> - 2017-07-30 22:10 +0200

csiph-web