x86 p2m: Fix comment regarding mfn_valid().

MMIO pages are not always above max_page (e.g., >=4GB system with MMIO
in the RAM hole below 4GB).

Signed-off-by: Keir Fraser <keir.fraser@citrix.com>

diff --git a/xen/arch/x86/mm/p2m.c b/xen/arch/x86/mm/p2m.c
index 7c0373effec5b2de5925b02af83a4bc22f675862..33bcc0dcea0ce8dc2c0c2706643087cbb1784bfe 100644 (file)
--- a/xen/arch/x86/mm/p2m.c
+++ b/xen/arch/x86/mm/p2m.c
@@ -2621,7 +2621,7 @@ clear_mmio_p2m_entry(struct p2m_domain *p2m, unsigned long gfn)
 
     mfn = gfn_to_mfn(p2m, gfn, &t);
 
-    /* Do not use mfn_valid() here as MMIO pages are always above max_page */
+    /* Do not use mfn_valid() here as it will usually fail for MMIO pages. */
     if ( (INVALID_MFN == mfn_x(mfn)) || (t != p2m_mmio_direct) )
     {
         gdprintk(XENLOG_ERR,