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] x86: adjust Dom0 initial memory allocation strategy

To: <xen-devel@xxxxxxxxxxxxxxxxxxx>
Subject: [Xen-devel] [PATCH] x86: adjust Dom0 initial memory allocation strategy
From: "Jan Beulich" <JBeulich@xxxxxxxxxx>
Date: Mon, 15 Mar 2010 08:58:51 +0000
Delivery-date: Mon, 15 Mar 2010 02:02:55 -0700
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/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=subscribe>
List-unsubscribe: <http://lists.xensource.com/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=unsubscribe>
Sender: xen-devel-bounces@xxxxxxxxxxxxxxxxxxx
Simply trying order-9 allocations until they won't succeed anymore
may consume unnecessarily much memory from the DMA zone (since the
page allocator will try to fulfill the request by using memory from
that zone when only lower order memory blocks are left in all other
zones). To avoid using DMA zone memory, make alloc_chunk() try to
allocate a second smaller chunk and use that one in favor of the
first one if it came from a higher addressed memory. This way, all
memory outside the DMA zone will be consumed before eating into that
zone.

Signed-off-by: Jan Beulich <jbeulich@xxxxxxxxxx>

--- 2010-03-09.orig/xen/arch/x86/domain_build.c 2010-03-12 12:25:18.000000000 
+0100
+++ 2010-03-09/xen/arch/x86/domain_build.c      2010-03-12 17:24:11.000000000 
+0100
@@ -127,7 +131,8 @@ static struct page_info * __init alloc_c
     struct domain *d, unsigned long max_pages)
 {
     struct page_info *page;
-    unsigned int order;
+    unsigned int order, free_order;
+
     /*
      * Allocate up to 2MB at a time: It prevents allocating very large chunks
      * from DMA pools before the >4GB pool is fully depleted.
@@ -140,6 +145,26 @@ static struct page_info * __init alloc_c
     while ( (page = alloc_domheap_pages(d, order, 0)) == NULL )
         if ( order-- == 0 )
             break;
+    /*
+     * Make a reasonable attempt at finding a smaller chunk at a higher
+     * address, to avoid allocating from low memory as much as possible.
+     */
+    for ( free_order = order; page && order--; )
+    {
+        struct page_info *pg2;
+
+        if ( d->tot_pages + (1 << order) > d->max_pages )
+            continue;
+        pg2 = alloc_domheap_pages(d, order, 0);
+        if ( pg2 > page )
+        {
+            free_domheap_pages(page, free_order);
+            page = pg2;
+            free_order = order;
+        }
+        else if ( pg2 )
+            free_domheap_pages(pg2, order);
+    }
     return page;
 }
 



Attachment: x86-dom0-alloc.patch
Description: Text document

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel
<Prev in Thread] Current Thread [Next in Thread>
  • [Xen-devel] [PATCH] x86: adjust Dom0 initial memory allocation strategy, Jan Beulich <=