From mboxrd@z Thu Jan  1 00:00:00 1970
From: Radim =?utf-8?B?S3LEjW3DocWZ?= <rkrcmar@redhat.com>
Subject: Re: [PATCH v2 06/13] KVM: x86: save/load state on SMM switch
Date: Wed, 3 Jun 2015 21:02:53 +0200
Message-ID: <20150603190253.GA13488@potion.brq.redhat.com>
References: <1432746314-50196-1-git-send-email-pbonzini@redhat.com>
 <1432746314-50196-7-git-send-email-pbonzini@redhat.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Cc: linux-kernel@vger.kernel.org, kvm@vger.kernel.org,
	guangrong.xiao@linux.intel.com, bdas@redhat.com
To: Paolo Bonzini <pbonzini@redhat.com>
Return-path: <linux-kernel-owner@vger.kernel.org>
Content-Disposition: inline
In-Reply-To: <1432746314-50196-7-git-send-email-pbonzini@redhat.com>
Sender: linux-kernel-owner@vger.kernel.org
List-Id: kvm.vger.kernel.org

2015-05-27 19:05+0200, Paolo Bonzini:
> The big ugly one.  This patch adds support for switching in and out of
> system management mode, respectively upon receiving KVM_REQ_SMI and upon
> executing a RSM instruction.  Both 32- and 64-bit formats are supported
> for the SMM state save area.
> 
> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
> ---
> 	RFC->v1: shift access rights left by 8 for 32-bit format
> 		 move tracepoint to kvm_set_hflags
> 		 fix NMI handling
> ---
> diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
>  static void process_smi(struct kvm_vcpu *vcpu)
>  {
> +	struct kvm_segment cs, ds;
> +	char buf[512];
> +	u32 cr0;
> +	int r;
> +
>  	if (is_smm(vcpu)) {
>  		vcpu->arch.smi_pending = true;
>  		return;
>  	}
>  
> -	printk_once(KERN_DEBUG "Ignoring guest SMI\n");
> +	trace_kvm_enter_smm(vcpu->vcpu_id, vcpu->arch.smbase, true);
> +	vcpu->arch.hflags |= HF_SMM_MASK;
> +	memset(buf, 0, 512);
> +	if (guest_cpuid_has_longmode(vcpu))
> +		process_smi_save_state_64(vcpu, buf);
> +	else
> +		process_smi_save_state_32(vcpu, buf);
> +
> +	r = kvm_write_guest(vcpu->kvm, vcpu->arch.smbase + 0xfe00, buf, sizeof(buf));

The state is saved in SMRAM, but we are accessing it using the non-SMM
address space ... how did it pass testing?
(Restore is using SMM address space, so I'm guessing that the mapping
 from QEMU wasn't really utilizing two separate address spaces.)

> +	if (r < 0)
> +		return;

And if we fail to write it, is there other option than throwing an error
to userspace?  (Unset HF_SMM_MASK and pretend that nothing happened
doesn't find much support in docs.)

Thanks.