[PATCH v3 2/9] mm: Place unscrubbed pages at the end of pagelist

[Boris Ostrovsky <boris.ostrovsky@oracle.com>]

. so that it's easy to find pages that need to be scrubbed (those pages are
now marked with _PGC_need_scrub bit).

Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
---
Changes in v3:
* Keep dirty bit per page, add dirty_head to page_info that indicates whether
  the buddy has dirty pages.
* Make page_list_add_scrub() set buddy's page order
* Data type adjustments (int -> unsigned)

 xen/common/page_alloc.c  |  119 +++++++++++++++++++++++++++++++++++++--------
 xen/include/asm-arm/mm.h |    6 ++
 xen/include/asm-x86/mm.h |    6 ++
 3 files changed, 110 insertions(+), 21 deletions(-)

diff --git a/xen/common/page_alloc.c b/xen/common/page_alloc.c
index 6fe55ee..9dcf6ee 100644
--- a/xen/common/page_alloc.c
+++ b/xen/common/page_alloc.c
@@ -383,6 +383,8 @@ typedef struct page_list_head heap_by_zone_and_order_t[NR_ZONES][MAX_ORDER+1];
 static heap_by_zone_and_order_t *_heap[MAX_NUMNODES];
 #define heap(node, zone, order) ((*_heap[node])[zone][order])
 
+static unsigned long node_need_scrub[MAX_NUMNODES];
+
 static unsigned long *avail[MAX_NUMNODES];
 static long total_avail_pages;
 
@@ -678,6 +680,20 @@ static void check_low_mem_virq(void)
     }
 }
 
+/* Pages that need scrub are added to tail, otherwise to head. */
+static void page_list_add_scrub(struct page_info *pg, unsigned int node,
+                                unsigned int zone, unsigned int order,
+                                bool need_scrub)
+{
+    PFN_ORDER(pg) = order;
+    pg->u.free.dirty_head = need_scrub;
+
+    if ( need_scrub )
+        page_list_add_tail(pg, &heap(node, zone, order));
+    else
+        page_list_add(pg, &heap(node, zone, order));
+}
+
 /* Allocate 2^@order contiguous pages. */
 static struct page_info *alloc_heap_pages(
     unsigned int zone_lo, unsigned int zone_hi,
@@ -802,7 +818,7 @@ static struct page_info *alloc_heap_pages(
     while ( j != order )
     {
         PFN_ORDER(pg) = --j;
-        page_list_add_tail(pg, &heap(node, zone, j));
+        page_list_add(pg, &heap(node, zone, j));
         pg += 1 << j;
     }
 
@@ -851,11 +867,14 @@ static int reserve_offlined_page(struct page_info *head)
     int zone = page_to_zone(head), i, head_order = PFN_ORDER(head), count = 0;
     struct page_info *cur_head;
     int cur_order;
+    bool need_scrub;
 
     ASSERT(spin_is_locked(&heap_lock));
 
     cur_head = head;
 
+    head->u.free.dirty_head = false;
+
     page_list_del(head, &heap(node, zone, head_order));
 
     while ( cur_head < (head + (1 << head_order)) )
@@ -892,8 +911,16 @@ static int reserve_offlined_page(struct page_info *head)
             {
             merge:
                 /* We don't consider merging outside the head_order. */
-                page_list_add_tail(cur_head, &heap(node, zone, cur_order));
-                PFN_ORDER(cur_head) = cur_order;
+
+                /* See if any of the pages need scrubbing. */
+                need_scrub = false;
+                for ( i = 0; i < (1 << cur_order); i++ )
+                    if ( test_bit(_PGC_need_scrub, &cur_head[i].count_info) )
+                    {
+                        need_scrub = true;
+                        break;
+                    }
+                page_list_add_scrub(cur_head, node, zone, cur_order, need_scrub);
                 cur_head += (1 << cur_order);
                 break;
             }
@@ -922,10 +949,13 @@ static int reserve_offlined_page(struct page_info *head)
 /* Returns new buddy head. */
 static struct page_info *
 merge_and_free_buddy(struct page_info *pg, unsigned int node,
-                     unsigned int zone, unsigned int order)
+                     unsigned int zone, unsigned int order,
+                     bool need_scrub)
 {
     ASSERT(spin_is_locked(&heap_lock));
 
+    pg->u.free.dirty_head = false;
+
     /* Merge chunks as far as possible. */
     while ( order < MAX_ORDER )
     {
@@ -944,6 +974,8 @@ merge_and_free_buddy(struct page_info *pg, unsigned int node,
             break;
 
         page_list_del(buddy, &heap(node, zone, order));
+        need_scrub |= buddy->u.free.dirty_head;
+        buddy->u.free.dirty_head = false;
 
         /* Adjust current buddy head if we merged backwards. */
         if ( buddy < pg )
@@ -952,15 +984,56 @@ merge_and_free_buddy(struct page_info *pg, unsigned int node,
         order++;
     }
 
-    PFN_ORDER(pg) = order;
-    page_list_add_tail(pg, &heap(node, zone, order));
+    page_list_add_scrub(pg, node, zone, order, need_scrub);
 
     return pg;
 }
 
+static void scrub_free_pages(unsigned int node)
+{
+    struct page_info *pg;
+    unsigned int zone, order;
+    unsigned long i;
+
+    ASSERT(spin_is_locked(&heap_lock));
+
+    if ( !node_need_scrub[node] )
+        return;
+
+    for ( zone = 0; zone < NR_ZONES; zone++ )
+    {
+        order = MAX_ORDER;
+        do {
+            while ( !page_list_empty(&heap(node, zone, order)) )
+            {
+                /* Unscrubbed pages are always at the end of the list. */
+                pg = page_list_last(&heap(node, zone, order));
+                if ( !pg->u.free.dirty_head )
+                    break;
+
+                for ( i = 0; i < (1UL << order); i++)
+                {
+                    if ( test_bit(_PGC_need_scrub, &pg[i].count_info) )
+                    {
+                        scrub_one_page(&pg[i]);
+                        pg[i].count_info &= ~PGC_need_scrub;
+                        node_need_scrub[node]--;
+                    }
+                }
+
+                page_list_del(pg, &heap(node, zone, order));
+                merge_and_free_buddy(pg, node, zone, order, false);
+
+                if ( node_need_scrub[node] == 0 )
+                    return;
+            }
+        } while ( order-- != 0 );
+    }
+}
+
 /* Free 2^@order set of pages. */
 static void free_heap_pages(
-    struct page_info *pg, unsigned int order)
+    struct page_info *pg, unsigned int order, bool need_scrub)
 {
     unsigned long mfn = page_to_mfn(pg);
     unsigned int i, node = phys_to_nid(page_to_maddr(pg)), tainted = 0;
@@ -1009,11 +1082,22 @@ static void free_heap_pages(
         midsize_alloc_zone_pages = max(
             midsize_alloc_zone_pages, total_avail_pages / MIDSIZE_ALLOC_FRAC);
 
-    pg = merge_and_free_buddy(pg, node, zone, order);
+    if ( need_scrub )
+    {
+        for ( i = 0; i < (1 << order); i++ )
+            pg[i].count_info |= PGC_need_scrub;
+
+        node_need_scrub[node] += (1UL << order);
+    }
+
+    pg = merge_and_free_buddy(pg, node, zone, order, need_scrub);
 
     if ( tainted )
         reserve_offlined_page(pg);
 
+    if ( need_scrub )
+        scrub_free_pages(node);
+
     spin_unlock(&heap_lock);
 }
 
@@ -1234,7 +1318,7 @@ unsigned int online_page(unsigned long mfn, uint32_t *status)
     spin_unlock(&heap_lock);
 
     if ( (y & PGC_state) == PGC_state_offlined )
-        free_heap_pages(pg, 0);
+        free_heap_pages(pg, 0, false);
 
     return ret;
 }
@@ -1303,7 +1387,7 @@ static void init_heap_pages(
             nr_pages -= n;
         }
 
-        free_heap_pages(pg+i, 0);
+        free_heap_pages(pg + i, 0, false);
     }
 }
 
@@ -1630,7 +1714,7 @@ void free_xenheap_pages(void *v, unsigned int order)
 
     memguard_guard_range(v, 1 << (order + PAGE_SHIFT));
 
-    free_heap_pages(virt_to_page(v), order);
+    free_heap_pages(virt_to_page(v), order, false);
 }
 
 #else
@@ -1684,12 +1768,9 @@ void free_xenheap_pages(void *v, unsigned int order)
     pg = virt_to_page(v);
 
     for ( i = 0; i < (1u << order); i++ )
-    {
-        scrub_one_page(&pg[i]);
         pg[i].count_info &= ~PGC_xen_heap;
-    }
 
-    free_heap_pages(pg, order);
+    free_heap_pages(pg, order, true);
 }
 
 #endif
@@ -1798,7 +1879,7 @@ struct page_info *alloc_domheap_pages(
     if ( d && !(memflags & MEMF_no_owner) &&
          assign_pages(d, pg, order, memflags) )
     {
-        free_heap_pages(pg, order);
+        free_heap_pages(pg, order, false);
         return NULL;
     }
     
@@ -1866,11 +1947,7 @@ void free_domheap_pages(struct page_info *pg, unsigned int order)
             scrub = 1;
         }
 
-        if ( unlikely(scrub) )
-            for ( i = 0; i < (1 << order); i++ )
-                scrub_one_page(&pg[i]);
-
-        free_heap_pages(pg, order);
+        free_heap_pages(pg, order, scrub);
     }
 
     if ( drop_dom_ref )
diff --git a/xen/include/asm-arm/mm.h b/xen/include/asm-arm/mm.h
index 0fef612..abc3f6b 100644
--- a/xen/include/asm-arm/mm.h
+++ b/xen/include/asm-arm/mm.h
@@ -45,6 +45,8 @@ struct page_info
         struct {
             /* Do TLBs need flushing for safety before next page use? */
             bool_t need_tlbflush;
+            /* Set on a buddy head if the buddy has unscrubbed pages. */
+            bool dirty_head;
         } free;
 
     } u;
@@ -115,6 +117,10 @@ struct page_info
 #define PGC_count_width   PG_shift(9)
 #define PGC_count_mask    ((1UL<<PGC_count_width)-1)
 
+/* Page needs to be scrubbed */
+#define _PGC_need_scrub   PG_shift(10)
+#define PGC_need_scrub    PG_mask(1, 10)
+
 extern unsigned long xenheap_mfn_start, xenheap_mfn_end;
 extern vaddr_t xenheap_virt_end;
 #ifdef CONFIG_ARM_64
diff --git a/xen/include/asm-x86/mm.h b/xen/include/asm-x86/mm.h
index 119d7de..5cf528a 100644
--- a/xen/include/asm-x86/mm.h
+++ b/xen/include/asm-x86/mm.h
@@ -89,6 +89,8 @@ struct page_info
         struct {
             /* Do TLBs need flushing for safety before next page use? */
             bool_t need_tlbflush;
+            /* Set on a buddy head if the buddy has unscrubbed pages. */
+            bool dirty_head;
         } free;
 
     } u;
@@ -233,6 +235,10 @@ struct page_info
 #define PGC_count_width   PG_shift(9)
 #define PGC_count_mask    ((1UL<<PGC_count_width)-1)
 
+/* Page needs to be scrubbed */
+#define _PGC_need_scrub   PG_shift(10)
+#define PGC_need_scrub    PG_mask(1, 10)
+
 struct spage_info
 {
        unsigned long type_info;
-- 
1.7.1


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel