Re: [PATCH v2 03/12] x86/HVM: Hardware alternate p2m support detection.

[Andrew Cooper <andrew.cooper3@citrix.com>]

On 22/06/15 19:56, Ed White wrote:
> As implemented here, only supported on platforms with VMX HAP.
>
> By default this functionality is force-disabled, it can be enabled
> by specifying altp2m=1 on the Xen command line.
>
> Signed-off-by: Ed White <edmund.h.white@intel.com>
> ---
>  docs/misc/xen-command-line.markdown |  7 +++++++
>  xen/arch/x86/hvm/hvm.c              | 12 ++++++++++++
>  xen/arch/x86/hvm/vmx/vmx.c          |  1 +
>  xen/include/asm-x86/hvm/hvm.h       |  6 ++++++
>  4 files changed, 26 insertions(+)
>
> diff --git a/docs/misc/xen-command-line.markdown b/docs/misc/xen-command-line.markdown
> index aa684c0..3391c66 100644
> --- a/docs/misc/xen-command-line.markdown
> +++ b/docs/misc/xen-command-line.markdown
> @@ -139,6 +139,13 @@ mode during S3 resume.
>  > Default: `true`
>  
>  Permit Xen to use superpages when performing memory management.
> + 
> +### altp2m (Intel)
> +> `= <boolean>`
> +
> ++> Default: `false`
> +
> +Permit multiple copies of host p2m.
>  
>  ### apic
>  > `= bigsmp | default`
> diff --git a/xen/arch/x86/hvm/hvm.c b/xen/arch/x86/hvm/hvm.c
> index 65baa1b..af68d44 100644
> --- a/xen/arch/x86/hvm/hvm.c
> +++ b/xen/arch/x86/hvm/hvm.c
> @@ -94,6 +94,10 @@ bool_t opt_hvm_fep;
>  boolean_param("hvm_fep", opt_hvm_fep);
>  #endif
>  
> +/* Xen command-line option to enable altp2m */
> +static bool_t __initdata opt_altp2m_enabled = 0;
> +boolean_param("altp2m", opt_altp2m_enabled);
> +
>  static int cpu_callback(
>      struct notifier_block *nfb, unsigned long action, void *hcpu)
>  {
> @@ -160,6 +164,9 @@ static int __init hvm_enable(void)
>      if ( !fns->pvh_supported )
>          printk(XENLOG_INFO "HVM: PVH mode not supported on this platform\n");
>  
> +    if ( !opt_altp2m_enabled )
> +        hvm_funcs.altp2m_supported = 0;
> +
>      /*
>       * Allow direct access to the PC debug ports 0x80 and 0xed (they are
>       * often used for I/O delays, but the vmexits simply slow things down).
> @@ -6474,6 +6481,11 @@ enum hvm_intblk nhvm_interrupt_blocked(struct vcpu *v)
>      return hvm_funcs.nhvm_intr_blocked(v);
>  }
>  
> +bool_t hvm_altp2m_supported()
> +{
> +    return hvm_funcs.altp2m_supported;
> +}

I would put this as a static inline in hvm.h, as opposed to forcing a
call into a different translation unit to retrieve a global boolean.

Otherwise, Reviewed-by: Andrew Cooper <andrew.cooper3@citrix.com>