linux-kernel - Re: [PATCH v5 2/7] nSVM: introduce smv->nested.save to cache save area fields

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] [day] [month] [year] [list]

Message-ID: <556d68697f5cd30e32307b1c56ae51b42f87cd4f.camel@redhat.com>
Date:   Fri, 12 Nov 2021 09:43:47 +0200
From:   Maxim Levitsky <mlevitsk@...hat.com>
To:     Paolo Bonzini <pbonzini@...hat.com>,
        Emanuele Giuseppe Esposito <eesposit@...hat.com>,
        kvm@...r.kernel.org
Cc:     Sean Christopherson <seanjc@...gle.com>,
        Vitaly Kuznetsov <vkuznets@...hat.com>,
        Wanpeng Li <wanpengli@...cent.com>,
        Jim Mattson <jmattson@...gle.com>,
        Joerg Roedel <joro@...tes.org>,
        Thomas Gleixner <tglx@...utronix.de>,
        Ingo Molnar <mingo@...hat.com>, Borislav Petkov <bp@...en8.de>,
        x86@...nel.org, "H. Peter Anvin" <hpa@...or.com>,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH v5 2/7] nSVM: introduce smv->nested.save to cache save
 area fields

On Thu, 2021-11-11 at 14:52 +0100, Paolo Bonzini wrote:
> On 11/3/21 15:05, Emanuele Giuseppe Esposito wrote:
> > Note that in svm_set_nested_state() we want to cache the L2
> > save state only if we are in normal non guest mode, because
> > otherwise it is not touched.
> 
> I think that call to nested_copy_vmcb_save_to_cache is not necessary at 
> all, because svm->nested.save is not used afterwards and is not valid 
> after VMRUN.

Yes, but since setting nested state is absolutely not performance critical,
having it do the same thing as normal VMRUN is always better.

Best regards,
	Maxim Levitsky

> 
> The relevant checks have already been done before:
> 
>          if (!(vcpu->arch.efer & EFER_SVME)) {
>                  /* GIF=1 and no guest mode are required if SVME=0.  */
>                  if (kvm_state->flags != KVM_STATE_NESTED_GIF_SET)
>                          return -EINVAL;
>          }
> 
> 	...
> 
>          /*
>           * Processor state contains L2 state.  Check that it is
>           * valid for guest mode (see nested_vmcb_check_save).
>           */
>          cr0 = kvm_read_cr0(vcpu);
>          if (((cr0 & X86_CR0_CD) == 0) && (cr0 & X86_CR0_NW))
>                  goto out_free;
> 
> (and all other checks are done by KVM_SET_SREGS, KVM_SET_DEBUGREGS etc.)
> 
> Paolo
>