linux-kernel - Re: [PATCH] KVM: X86: Add missing KVM

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]

Date:   Fri, 5 Oct 2018 15:03:25 -0700
From:   Guenter Roeck <linux@...ck-us.net>
To:     Paolo Bonzini <pbonzini@...hat.com>
Cc:     Radim Krčmář <rkrcmar@...hat.com>,
        Thomas Gleixner <tglx@...utronix.de>,
        Ingo Molnar <mingo@...hat.com>, Borislav Petkov <bp@...en8.de>,
        x86@...nel.org, kvm@...r.kernel.org, linux-kernel@...r.kernel.org,
        Brijesh Singh <brijesh.singh@....com>,
        Borislav Petkov <bp@...e.de>
Subject: Re: [PATCH] KVM: X86: Add missing KVM_AMD dependency

On Fri, Oct 05, 2018 at 10:41:55PM +0200, Paolo Bonzini wrote:
> On 05/10/2018 20:46, Guenter Roeck wrote:
> > Analysis shows that commit 59414c9892208 ("KVM: SVM: Add support for
> > KVM_SEV_LAUNCH_START command") added a dependency of KVM_AMD on
> > CRYPTO_DEV_CCP_DD if CRYPTO_DEV_SP_PSP is enabled: If CRYPTO_DEV_CCP_DD
> > is built as module, KVM_AMD must be built as module as well.
> > 
> > Fixes: 59414c9892208 ("KVM: SVM: Add support for KVM_SEV_LAUNCH_START command")
> > Cc: Brijesh Singh <brijesh.singh@....com>
> > Cc: Borislav Petkov <bp@...e.de>
> > Signed-off-by: Guenter Roeck <linux@...ck-us.net>
> 
> This should be handled by
> 
> config KVM_AMD_SEV
>         def_bool y
>         bool "AMD Secure Encrypted Virtualization (SEV) support"
>         depends on KVM_AMD && X86_64
>         depends on CRYPTO_DEV_SP_PSP && !(KVM_AMD=y && CRYPTO_DEV_CCP_DD=m)
>         ---help---
>         Provides support for launching Encrypted VMs on AMD processors.
> 

Unfortunately it doesn't. It disables KVM_AMD_SEV, but that doesn't prevent
the calls.

> Maybe this works as well?  I haven't tested it yet:
> 

I am sure there are many possible solutions. I would personally prefer one
that enforces KVM_AMD=m with CRYPTO_DEV_CCP_DD=m, but that is just me.

Thanks,
Guenter

> diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c
> index 89c4c5aa15f1..55f10b17d044 100644
> --- a/arch/x86/kvm/svm.c
> +++ b/arch/x86/kvm/svm.c
> @@ -441,9 +441,13 @@ static inline bool svm_sev_enabled(void)
> 
>  static inline bool sev_guest(struct kvm *kvm)
>  {
> +#ifdef CONFIG_KVM_AMD_SEV
>  	struct kvm_sev_info *sev = &to_kvm_svm(kvm)->sev_info;
> 
>  	return sev->active;
> +#else
> +	return false;
> +#endif
>  }
> 
>  static inline int sev_get_asid(struct kvm *kvm)
> 
> Thanks,
> 
> Paolo