WARNING - OLD ARCHIVES

This is an archived copy of the Xen.org mailing list, which we have preserved to ensure that existing links to archives are not broken. The live archive, which contains the latest emails, can be found at http://lists.xen.org/
   
 
 
Xen 
 
Home Products Support Community News
 
   
 

xen-devel

[Xen-devel] [PATCH]Don't unshadow current used shadow L2 table.

To: xen-devel <xen-devel@xxxxxxxxxxxxxxxxxxx>
Subject: [Xen-devel] [PATCH]Don't unshadow current used shadow L2 table.
From: "Ling, Xiaofeng" <xiaofeng.ling@xxxxxxxxx>
Date: Thu, 08 Dec 2005 16:02:42 +0800
Delivery-date: Thu, 08 Dec 2005 08:05:41 +0000
Envelope-to: www-data@xxxxxxxxxxxxxxxxxxx
List-help: <mailto:xen-devel-request@lists.xensource.com?subject=help>
List-id: Xen developer discussion <xen-devel.lists.xensource.com>
List-post: <mailto:xen-devel@lists.xensource.com>
List-subscribe: <http://lists.xensource.com/cgi-bin/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=subscribe>
List-unsubscribe: <http://lists.xensource.com/cgi-bin/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=unsubscribe>
Sender: xen-devel-bounces@xxxxxxxxxxxxxxxxxxx
User-agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.6) Gecko/20040510
# HG changeset patch
# User Xiaofeng Ling <xiaofeng.ling@xxxxxxxxx>
# Node ID cc6e19224d0b755d24358cd15d1a40f894d320b4
# Parent  b215584242c43415eccd008b90c812f3d04b398d
Don't unshadow when the shadow l2 is the current used shadow table.
Otherwise, when destorying the domain, if the current sl2 is unpinned,
the shadow page will be freed without accquire the shadow_lock.
This may also improve the performance due to avoiding uncessary
unshadow.

Signed-off-by: Xiaofeng Ling <xiaofeng.ling@xxxxxxxxx>
# HG changeset patch
# User Xiaofeng Ling <xiaofeng.ling@xxxxxxxxx>
# Node ID cc6e19224d0b755d24358cd15d1a40f894d320b4
# Parent  b215584242c43415eccd008b90c812f3d04b398d
Don't unshadow when the shadow l2 is the current used shadow table.
Otherwise, when destory the domain, if the current sl2 is unpinned,
the shadow page will be freed without accquire the shadow_lock.
This may also improve the performance due to avoiding uncessary
unshadow.

Signed-off-by: Xiaofeng Ling <xiaofeng.ling@xxxxxxxxx>

diff -r b215584242c4 xen/arch/x86/shadow.c
--- a/xen/arch/x86/shadow.c     Wed Dec  7 16:26:07 2005 +0000
+++ b/xen/arch/x86/shadow.c     Thu Dec  8 15:07:52 2005 +0800
@@ -1450,6 +1450,7 @@
     int changed;
     u32 min_max_shadow, min_max_snapshot;
     int min_shadow, max_shadow, min_snapshot, max_snapshot;
+    struct vcpu *v;
 
     ASSERT(shadow_lock_is_acquired(d));
 
@@ -1739,6 +1740,9 @@
 
         if ( unlikely(unshadow) )
         {
+            for_each_vcpu(d, v)
+                if(smfn == pagetable_get_pfn(v->arch.shadow_table))
+                    return need_flush;
             perfc_incrc(unshadow_l2_count);
             shadow_unpin(smfn);
 #if CONFIG_PAGING_LEVELS == 2
diff -r b215584242c4 xen/arch/x86/shadow32.c
--- a/xen/arch/x86/shadow32.c   Wed Dec  7 16:26:07 2005 +0000
+++ b/xen/arch/x86/shadow32.c   Thu Dec  8 15:07:52 2005 +0800
@@ -2326,6 +2326,7 @@
     int changed;
     u32 min_max_shadow, min_max_snapshot;
     int min_shadow, max_shadow, min_snapshot, max_snapshot;
+    struct vcpu *v;
 
     ASSERT(shadow_lock_is_acquired(d));
 
@@ -2527,6 +2528,9 @@
 
         if ( unlikely(unshadow) )
         {
+            for_each_vcpu(d, v)
+                if(smfn == pagetable_get_pfn(v->arch.shadow_table))
+                    return need_flush;
             perfc_incrc(unshadow_l2_count);
             shadow_unpin(smfn);
             if ( unlikely(shadow_mode_external(d)) )
_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel
<Prev in Thread] Current Thread [Next in Thread>
  • [Xen-devel] [PATCH]Don't unshadow current used shadow L2 table., Ling, Xiaofeng <=