[PATCH] x86 shadow: avoid race between resync and fast path

[Gianluca Guida <gianluca.guida@eu.citrix.com>]

[-- Attachment #1: Type: text/plain, Size: 532 bytes --]

Hello,

I think that using fast path and OOS is not incompatible, it was just 
buggy at the moment. This patch should fix it. I'll remove the offending 
feature (fast path gnp) anyway, but since we are still interested in 
fast path mmio and there could still be a incredibly rare corruption 
(*), it is better to fix the mechanism.

Frank, can you check that with this patch only the bug goes away?

Thanks,
Gianluca

Fix lock-less race between resync and fast path.

Signed-off-by: Gianluca Guida <gianluca.guida@eu.citrix.com>



[-- Attachment #2: fix-fastpath-resync-race.patch --]
[-- Type: text/x-diff, Size: 3105 bytes --]

diff -r 7397608bce87 xen/arch/x86/mm/shadow/multi.c
--- a/xen/arch/x86/mm/shadow/multi.c	Mon Jun 29 15:50:32 2009 +0100
+++ b/xen/arch/x86/mm/shadow/multi.c	Fri Jul 03 13:53:35 2009 +0100
@@ -2975,6 +2975,30 @@
 #if (SHADOW_OPTIMIZATIONS & SHOPT_FAST_FAULT_PATH)
     if ( (regs->error_code & PFEC_reserved_bit) )
     {
+#if (SHADOW_OPTIMIZATIONS & SHOPT_OUT_OF_SYNC) 
+        /* First, need to check that this isn't an out-of-sync
+         * shadow l1e.  If it is, we fall back to the slow path, which
+         * will sync it up again. */
+        {
+            shadow_l2e_t sl2e;
+            mfn_t gl1mfn;
+            if ( (__copy_from_user(&sl2e,
+                                   (sh_linear_l2_table(v)
+                                    + shadow_l2_linear_offset(va)),
+                                   sizeof(sl2e)) != 0)
+                 || !(shadow_l2e_get_flags(sl2e) & _PAGE_PRESENT)
+                 || !mfn_valid(gl1mfn = _mfn(mfn_to_page(
+                                  shadow_l2e_get_mfn(sl2e))->v.sh.back))
+                 || unlikely(mfn_is_out_of_sync(gl1mfn)) )
+            {
+                /* Hit the slow path as if there had been no 
+                 * shadow entry at all, and let it tidy up */
+                ASSERT(regs->error_code & PFEC_page_present);
+                regs->error_code ^= (PFEC_reserved_bit|PFEC_page_present);
+                goto page_fault_slow_path;
+            }
+        }
+#endif /* SHOPT_OUT_OF_SYNC */
         /* The only reasons for reserved bits to be set in shadow entries 
          * are the two "magic" shadow_l1e entries. */
         if ( likely((__copy_from_user(&sl1e, 
@@ -2983,30 +3007,6 @@
                                       sizeof(sl1e)) == 0)
                     && sh_l1e_is_magic(sl1e)) )
         {
-#if (SHADOW_OPTIMIZATIONS & SHOPT_OUT_OF_SYNC) 
-             /* First, need to check that this isn't an out-of-sync
-              * shadow l1e.  If it is, we fall back to the slow path, which
-              * will sync it up again. */
-            {
-                shadow_l2e_t sl2e;
-                mfn_t gl1mfn;
-               if ( (__copy_from_user(&sl2e,
-                                       (sh_linear_l2_table(v)
-                                        + shadow_l2_linear_offset(va)),
-                                       sizeof(sl2e)) != 0)
-                     || !(shadow_l2e_get_flags(sl2e) & _PAGE_PRESENT)
-                     || !mfn_valid(gl1mfn = _mfn(mfn_to_page(
-                                      shadow_l2e_get_mfn(sl2e))->v.sh.back))
-                     || unlikely(mfn_is_out_of_sync(gl1mfn)) )
-               {
-                   /* Hit the slow path as if there had been no 
-                    * shadow entry at all, and let it tidy up */
-                   ASSERT(regs->error_code & PFEC_page_present);
-                   regs->error_code ^= (PFEC_reserved_bit|PFEC_page_present);
-                   goto page_fault_slow_path;
-               }
-            }
-#endif /* SHOPT_OUT_OF_SYNC */
 
             if ( sh_l1e_is_gnp(sl1e) )
             {

[-- Attachment #3: Type: text/plain, Size: 138 bytes --]

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel