[Xen-devel] [PATCH 2/6] x86/shadow: remove pointless loops over all vCPU-s

["Jan Beulich" <JBeulich@xxxxxxxx>]

The vCPU count can be had more directly.

Signed-off-by: Jan Beulich <jbeulich@xxxxxxxx>
---
In the sh_make_shadow() case the question is whether it really was
intended to count all vCPU-s, rather than e.g. only all initialized
ones. I guess the problem would be the phase before the guest
actually starts secondary processors, but that could perhaps be
covered by using ->max_vcpus if otherwise 1 would result.

--- a/xen/arch/x86/mm/shadow/common.c
+++ b/xen/arch/x86/mm/shadow/common.c
@@ -1214,13 +1214,7 @@ const u8 sh_type_to_size[] = {
  * worth to make sure we never return zero. */
 static unsigned int shadow_min_acceptable_pages(struct domain *d)
 {
-    u32 vcpu_count = 1;
-    struct vcpu *v;
-
-    for_each_vcpu(d, v)
-        vcpu_count++;
-
-    return (vcpu_count * 128);
+    return (d->max_vcpus + 1) * 128;
 }
 
 /* Dispatcher function: call the per-mode function that will unhook the
--- a/xen/arch/x86/mm/shadow/multi.c
+++ b/xen/arch/x86/mm/shadow/multi.c
@@ -1476,16 +1476,14 @@ sh_make_shadow(struct vcpu *v, mfn_t gmf
          * pinning l3es.  This is not very quick but it doesn't happen
          * very often. */
         struct page_info *sp, *t;
-        struct vcpu *v2;
-        int l4count = 0, vcpus = 0;
+        unsigned int l4count = 0;
+
         page_list_for_each(sp, &d->arch.paging.shadow.pinned_shadows)
         {
             if ( sp->u.sh.type == SH_type_l4_64_shadow )
                 l4count++;
         }
-        for_each_vcpu ( d, v2 )
-            vcpus++;
-        if ( l4count > 2 * vcpus )
+        if ( l4count > 2 * d->max_vcpus )
         {
             /* Unpin all the pinned l3 tables, and don't pin any more. */
             page_list_for_each_safe(sp, t, 
&d->arch.paging.shadow.pinned_shadows)




_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxxx
https://lists.xenproject.org/mailman/listinfo/xen-devel