[IA64] fix domVTi boot with maxmem

We found the following bug.
http://bugzilla.xensource.com/bugzilla/show_bug.cgi?id=800

I change assigned page number by assign_domain_page()
from d->max_pages to d->tot_pages in vmx_build_physmap_table()
because domVTi don't support maxmem.

Signed-off-by: Akio Takebe <takebe_akio@jp.fujitsu.com>

diff --git a/xen/arch/ia64/vmx/vmx_init.c b/xen/arch/ia64/vmx/vmx_init.c
index 9d8fbe8ec8c4b58c7ea3963e91dd436d58f0d423..46634f011933f63e677b63a46d7a153d166c25f8 100644 (file)
--- a/xen/arch/ia64/vmx/vmx_init.c
+++ b/xen/arch/ia64/vmx/vmx_init.c
@@ -364,7 +364,10 @@ static const io_range_t io_ranges[] = {
 
 /* Reseve 1 page for shared I/O ,1 page for xenstore and 1 page for buffer I/O.  */
 #define VMX_SYS_PAGES  (3 + (GFW_SIZE >> PAGE_SHIFT))
-#define VMX_CONFIG_PAGES(d) ((d)->max_pages - VMX_SYS_PAGES)
+/* If we support maxmem for domVTi, we should change from tot_page to max_pages.
+ * #define VMX_CONFIG_PAGES(d) ((d)->max_pages - VMX_SYS_PAGES)
+ */
+#define VMX_CONFIG_PAGES(d) ((d)->tot_pages - VMX_SYS_PAGES)
 
 static void vmx_build_physmap_table(struct domain *d)
 {