linux-kernel - Re: [PATCH v3 02/13] X86/nVMX: handle_vmptrld: Copy the VMCS12 directly from guest memory

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: <CALMp9eQfMOjwS2R-9d_9m8z2+XKdvg2FaL65q83DR-RhxLkf3g@mail.gmail.com>
Date:   Mon, 22 Oct 2018 14:32:52 -0700
From:   Jim Mattson <jmattson@...gle.com>
To:     KarimAllah Ahmed <karahmed@...zon.de>
Cc:     kvm list <kvm@...r.kernel.org>,
        LKML <linux-kernel@...r.kernel.org>,
        Paolo Bonzini <pbonzini@...hat.com>,
        Radim Krčmář <rkrcmar@...hat.com>
Subject: Re: [PATCH v3 02/13] X86/nVMX: handle_vmptrld: Copy the VMCS12
 directly from guest memory

On Sat, Oct 20, 2018 at 3:22 PM, KarimAllah Ahmed <karahmed@...zon.de> wrote:
> Copy the VMCS12 directly from guest memory instead of the map->copy->unmap
> sequence. This also avoids using kvm_vcpu_gpa_to_page() and kmap() which
> assumes that there is a "struct page" for guest memory.
>
> Signed-off-by: KarimAllah Ahmed <karahmed@...zon.de>
> ---
> v1 -> v2:
> - Massage commit message a bit.
> ---
>  arch/x86/kvm/vmx.c | 23 +++++++++--------------
>  1 file changed, 9 insertions(+), 14 deletions(-)
>
> diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
> index 358759a..bc45347 100644
> --- a/arch/x86/kvm/vmx.c
> +++ b/arch/x86/kvm/vmx.c
> @@ -8879,33 +8879,28 @@ static int handle_vmptrld(struct kvm_vcpu *vcpu)
>         }
>
>         if (vmx->nested.current_vmptr != vmptr) {
> -               struct vmcs12 *new_vmcs12;
> -               struct page *page;
> -               page = kvm_vcpu_gpa_to_page(vcpu, vmptr);
> -               if (is_error_page(page)) {
> +               struct vmcs12 *new_vmcs12 = (struct vmcs12 *)__get_free_page(GFP_KERNEL);
> +
> +               if (!new_vmcs12 ||
> +                   kvm_read_guest(vcpu->kvm, vmptr, new_vmcs12,
> +                                  sizeof(*new_vmcs12))) {
> +                       free_page((unsigned long)new_vmcs12);
>                         nested_vmx_failInvalid(vcpu);

I believe that this failure, like the one below, should be:

nested_vmx_failValid(vcpu, VMXERR_VMPTRLD_INCORRECT_VMCS_REVISION_ID);

First, if there is a valid current VMCS, then nested_vmx_failInvalid
is incorrect. Second, if there is no backing memory, then by PCI bus
error semantics, we should read all 1's, which isn't the correct VMCS
revision ID.

>                         return kvm_skip_emulated_instruction(vcpu);
>                 }
> -               new_vmcs12 = kmap(page);
> +
>                 if (new_vmcs12->hdr.revision_id != VMCS12_REVISION ||
>                     (new_vmcs12->hdr.shadow_vmcs &&
>                      !nested_cpu_has_vmx_shadow_vmcs(vcpu))) {
> -                       kunmap(page);
> -                       kvm_release_page_clean(page);
> +                       free_page((unsigned long)new_vmcs12);
>                         nested_vmx_failValid(vcpu,
>                                 VMXERR_VMPTRLD_INCORRECT_VMCS_REVISION_ID);
>                         return kvm_skip_emulated_instruction(vcpu);
>                 }
>
>                 nested_release_vmcs12(vmx);
> -               /*
> -                * Load VMCS12 from guest memory since it is not already
> -                * cached.
> -                */
>                 memcpy(vmx->nested.cached_vmcs12, new_vmcs12, VMCS12_SIZE);
> -               kunmap(page);
> -               kvm_release_page_clean(page);
> -
> +               free_page((unsigned long)new_vmcs12);
>                 set_current_vmptr(vmx, vmptr);
>         }
>
> --
> 2.7.4
>