linux-kernel - Re: [PATCH v5 2/4] KVM: x86: report negative values from wrmsr to userspace

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20200921160812.GA23989@linux.intel.com>
Date:   Mon, 21 Sep 2020 09:08:12 -0700
From:   Sean Christopherson <sean.j.christopherson@...el.com>
To:     Maxim Levitsky <mlevitsk@...hat.com>
Cc:     kvm@...r.kernel.org, Vitaly Kuznetsov <vkuznets@...hat.com>,
        "H. Peter Anvin" <hpa@...or.com>, Joerg Roedel <joro@...tes.org>,
        Ingo Molnar <mingo@...hat.com>,
        "maintainer:X86 ARCHITECTURE (32-BIT AND 64-BIT)" <x86@...nel.org>,
        Wanpeng Li <wanpengli@...cent.com>,
        Borislav Petkov <bp@...en8.de>,
        Jim Mattson <jmattson@...gle.com>,
        linux-kernel@...r.kernel.org, Paolo Bonzini <pbonzini@...hat.com>,
        Thomas Gleixner <tglx@...utronix.de>
Subject: Re: [PATCH v5 2/4] KVM: x86: report negative values from wrmsr to
 userspace

On Mon, Sep 21, 2020 at 04:19:21PM +0300, Maxim Levitsky wrote:
> This will allow us to make some MSR writes fatal to the guest
> (e.g when out of memory condition occurs)
> 
> Signed-off-by: Maxim Levitsky <mlevitsk@...hat.com>
> ---
>  arch/x86/kvm/emulate.c | 7 +++++--
>  arch/x86/kvm/x86.c     | 5 +++--
>  2 files changed, 8 insertions(+), 4 deletions(-)
> 
> diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c
> index 1d450d7710d63..d855304f5a509 100644
> --- a/arch/x86/kvm/emulate.c
> +++ b/arch/x86/kvm/emulate.c
> @@ -3702,13 +3702,16 @@ static int em_dr_write(struct x86_emulate_ctxt *ctxt)
>  static int em_wrmsr(struct x86_emulate_ctxt *ctxt)
>  {
>  	u64 msr_data;
> +	int ret;
>  
>  	msr_data = (u32)reg_read(ctxt, VCPU_REGS_RAX)
>  		| ((u64)reg_read(ctxt, VCPU_REGS_RDX) << 32);
> -	if (ctxt->ops->set_msr(ctxt, reg_read(ctxt, VCPU_REGS_RCX), msr_data))
> +
> +	ret = ctxt->ops->set_msr(ctxt, reg_read(ctxt, VCPU_REGS_RCX), msr_data);
> +	if (ret > 0)
>  		return emulate_gp(ctxt, 0);
>  
> -	return X86EMUL_CONTINUE;
> +	return ret < 0 ? X86EMUL_UNHANDLEABLE : X86EMUL_CONTINUE;
>  }
>  
>  static int em_rdmsr(struct x86_emulate_ctxt *ctxt)
> diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
> index 063d70e736f7f..b6c67ab7c4f34 100644
> --- a/arch/x86/kvm/x86.c
> +++ b/arch/x86/kvm/x86.c
> @@ -1612,15 +1612,16 @@ int kvm_emulate_wrmsr(struct kvm_vcpu *vcpu)
>  {
>  	u32 ecx = kvm_rcx_read(vcpu);
>  	u64 data = kvm_read_edx_eax(vcpu);
> +	int ret = kvm_set_msr(vcpu, ecx, data);
>  
> -	if (kvm_set_msr(vcpu, ecx, data)) {
> +	if (ret > 0) {
>  		trace_kvm_msr_write_ex(ecx, data);
>  		kvm_inject_gp(vcpu, 0);
>  		return 1;
>  	}
>  
>  	trace_kvm_msr_write(ecx, data);

Tracing the access as non-faulting feels wrong.  The WRMSR has not completed,
e.g. if userspace cleanly handles -ENOMEM and restarts the guest, KVM would
trace the WRMSR twice.

What about:

	int ret = kvm_set_msr(vcpu, ecx, data);

	if (ret < 0)
		return ret;

	if (ret) {
		trace_kvm_msr_write_ex(ecx, data);
		kvm_inject_gp(vcpu, 0);
		return 1;
	}

	trace_kvm_msr_write(ecx, data);
	return kvm_skip_emulated_instruction(vcpu);

> -	return kvm_skip_emulated_instruction(vcpu);
> +	return ret < 0 ? ret : kvm_skip_emulated_instruction(vcpu);
>  }
>  EXPORT_SYMBOL_GPL(kvm_emulate_wrmsr);
>  
> -- 
> 2.26.2
>