From mboxrd@z Thu Jan  1 00:00:00 1970
From: "Cihula, Joseph" <joseph.cihula@intel.com>
Subject: [RFC] fix xen_in_range()
Date: Wed, 22 Apr 2009 16:53:37 -0700
Message-ID: <4F65016F6CB04E49BFFA15D4F7B798D9988CF1A3@orsmsx506.amr.corp.intel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
Return-path: <xen-devel-bounces@lists.xensource.com>
Content-Language: en-US
List-Unsubscribe: <http://lists.xensource.com/mailman/listinfo/xen-devel>,
	<mailto:xen-devel-request@lists.xensource.com?subject=unsubscribe>
List-Post: <mailto:xen-devel@lists.xensource.com>
List-Help: <mailto:xen-devel-request@lists.xensource.com?subject=help>
List-Subscribe: <http://lists.xensource.com/mailman/listinfo/xen-devel>,
	<mailto:xen-devel-request@lists.xensource.com?subject=subscribe>
Sender: xen-devel-bounces@lists.xensource.com
Errors-To: xen-devel-bounces@lists.xensource.com
To: "xen-devel@lists.xensource.com" <xen-devel@lists.xensource.com>, Keir Fraser <keir.fraser@eu.citrix.com>
Cc: "Cui, Dexuan" <dexuan.cui@intel.com>, "Wang, Shane" <shane.wang@intel.com>, "Jiang, Yunhong" <yunhong.jiang@intel.com>, "Yang,
	Xiaowei" <xiaowei.yang@intel.com>, "Ke,
	Liping" <liping.ke@intel.com>, "Li, Xin" <xin.li@intel.com>
List-Id: xen-devel@lists.xenproject.org

The frametable check in xen_in_range() incorrectly compares virtual address=
es to physical addresses (i.e. the parameters).

Unfortunately, the frametable is only contiguous in the virtual address spa=
ce, so one can't simply take __pa() of its start and end.  And since it is =
quite large, iterating through each page to gets its phys addr adds a perce=
ptible delay when that check has to be done for each page of physical memor=
y (as is the case in the only caller, the VT-d routine that maps memory for=
 dom0).  But it also appears that we can't convert the phys addr arguments =
into their virt addrs to compare with the contiguous frametable range becau=
se they will convert to the DIRECTMAP va's instead.

So while I would prefer to find a way to keep the check against the frameta=
ble, I don't see any obvious way to do so.  So this RFC is to see if there =
are any ideas for how the test could be re-written.

If not, then the below patch should be applied to remove the frametable che=
ck altogether.

Signed-off-by:  Joseph Cihula <joseph.cihula@intel.com>

diff -r 655dc3bc1d8e xen/arch/x86/setup.c
--- a/xen/arch/x86/setup.c      Thu Apr 16 11:54:06 2009 +0100
+++ b/xen/arch/x86/setup.c      Wed Apr 22 15:23:46 2009 -0700
@@ -1119,7 +1119,7 @@ int xen_in_range(paddr_t start, paddr_t
     int i;
     static struct {
         paddr_t s, e;
-    } xen_regions[5];
+    } xen_regions[4];

     /* initialize first time */
     if ( !xen_regions[0].s )
@@ -1140,10 +1140,6 @@ int xen_in_range(paddr_t start, paddr_t
         /* bss + boot allocator bitmap */
         xen_regions[3].s =3D __pa(&__bss_start);
         xen_regions[3].e =3D allocator_bitmap_end;
-        /* frametable */
-        xen_regions[4].s =3D (unsigned long)frame_table;
-        xen_regions[4].e =3D (unsigned long)frame_table +
-                           PFN_UP(max_page * sizeof(*frame_table));
     }

     for ( i =3D 0; i < ARRAY_SIZE(xen_regions); i++ )