Re: [Xen-devel] [PATCH v4 08/10] nEPT: handle invept instruction from L1 VMM

["Jan Beulich" <JBeulich@xxxxxxxx>]

>>> On 24.12.12 at 15:26, Xiantao Zhang <xiantao.zhang@xxxxxxxxx> wrote:
> --- a/xen/arch/x86/hvm/vmx/vvmx.c
> +++ b/xen/arch/x86/hvm/vmx/vvmx.c
> @@ -1390,6 +1390,42 @@ int nvmx_handle_vmwrite(struct cpu_user_regs *regs)
>      return X86EMUL_OKAY;
>  }
>  
> +int nvmx_handle_invept(struct cpu_user_regs *regs)
> +{
> +    struct vmx_inst_decoded decode;
> +    unsigned long eptp;
> +    u64 inv_type;
> +
> +    if ( decode_vmx_inst(regs, &decode, &eptp, 0) != X86EMUL_OKAY )
> +        return X86EMUL_EXCEPTION;

So in the overview you said you fixed this, but here it is again:
There are more than the two X86EMUL_* values referenced
above, and hence you can't imply that if it's not one, it's the
other.

> +
> +    inv_type = reg_read(regs, decode.reg2);
> +
> +    switch ( inv_type )

There doesn't appear to be a second use of inv_type, and hence
you can

    switch ( reg_read(regs, decode.reg2) )

and remove the local variable.

> +    {
> +    case INVEPT_SINGLE_CONTEXT:
> +    {
> +        struct p2m_domain *p2m = vcpu_nestedhvm(current).nv_p2m;
> +        if ( p2m )
> +        {
> +             p2m_flush(current, p2m);

And similarly you said you fixed all the white space issues.

Jan

> +            ept_sync_domain(p2m);
> +        }
> +        break;
> +    }
> +    case INVEPT_ALL_CONTEXT:
> +        p2m_flush_nestedp2m(current->domain);
> +        __invept(INVEPT_ALL_CONTEXT, 0, 0);
> +        break;
> +    default:
> +        vmreturn(regs, VMFAIL_INVALID);
> +        return X86EMUL_OKAY;
> +    }
> +    vmreturn(regs, VMSUCCEED);
> +    return X86EMUL_OKAY;
> +}
> +
> +
>  #define __emul_value(enable1, default1) \
>      ((enable1 | default1) << 32 | (default1))
>  



_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel