[PATCH v2 3/5] x86: don't override INVALID_M2P_ENTRY with SHARED_M2P_ENTRY

[Jan Beulich <jbeulich@xxxxxxxx>]

While in most cases code ahead of the invocation of set_gpfn_from_mfn()
deals with shared pages, at least in set_typed_p2m_entry() I can't spot
such handling (it's entirely possible there's code missing there). Let's
try to play safe and add an extra check.

Signed-off-by: Jan Beulich <jbeulich@xxxxxxxx>
---
v2: Re-base over set_gpfn_from_mfn() conversion to function.

--- a/xen/arch/x86/x86_64/mm.c
+++ b/xen/arch/x86/x86_64/mm.c
@@ -1485,12 +1485,19 @@ destroy_frametable:
 
 void set_gpfn_from_mfn(unsigned long mfn, unsigned long pfn)
 {
-    const struct domain *d = page_get_owner(mfn_to_page(_mfn(mfn)));
-    unsigned long entry = (d && (d == dom_cow)) ? SHARED_M2P_ENTRY : pfn;
+    unsigned long entry = pfn;
 
     if ( unlikely(!machine_to_phys_mapping_valid) )
         return;
 
+    if ( entry != INVALID_M2P_ENTRY )
+    {
+        const struct domain *d = page_get_owner(mfn_to_page(_mfn(mfn)));
+
+        if ( d && (d == dom_cow) )
+            entry = SHARED_M2P_ENTRY;
+    }
+
     if ( opt_pv32 &&
          mfn < (RDWR_COMPAT_MPT_VIRT_END - RDWR_COMPAT_MPT_VIRT_START) / 4 )
         compat_machine_to_phys_mapping[mfn] = entry;