Separate migration bitmap

This patch creates a migration bitmap, which is periodically kept in
sync with the qemu bitmap. A separate copy of the dirty bitmap for the
migration limits the amount of concurrent access to the qemu bitmap
from iothread and migration thread (which requires taking the big
lock).

We use the qemu bitmap type.  We have to "undo" the dirty_pages
counting optimization on the general dirty bitmap and do the counting
optimization with the migration local bitmap.

Signed-off-by: Umesh Deshpande <udeshpan@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>

3 files changed, 43 insertions, 31 deletions

diff --git a/arch_init.c b/arch_init.c
index cec3ead8de..d918844028 100644
--- a/arch_init.c
+++ b/arch_init.c
@@ -31,6 +31,8 @@
 #include "config.h"
 #include "monitor.h"
 #include "sysemu.h"
+#include "bitops.h"
+#include "bitmap.h"
 #include "arch_init.h"
 #include "audio/audio.h"
 #include "hw/pc.h"
@@ -331,39 +333,57 @@ static int save_xbzrle_page(QEMUFile *f, uint8_t *current_data,
 
 static RAMBlock *last_block;
 static ram_addr_t last_offset;
+static unsigned long *migration_bitmap;
+static uint64_t migration_dirty_pages;
 
 static inline bool migration_bitmap_test_and_reset_dirty(MemoryRegion *mr,
                                                          ram_addr_t offset)
 {
-    bool ret = memory_region_get_dirty(mr, offset, TARGET_PAGE_SIZE,
-                                       DIRTY_MEMORY_MIGRATION);
+    bool ret;
+    int nr = (mr->ram_addr + offset) >> TARGET_PAGE_BITS;
+
+    ret = test_and_clear_bit(nr, migration_bitmap);
 
     if (ret) {
-        memory_region_reset_dirty(mr, offset, TARGET_PAGE_SIZE,
-                                  DIRTY_MEMORY_MIGRATION);
+        migration_dirty_pages--;
     }
     return ret;
 }
 
-static inline void migration_bitmap_set_dirty(MemoryRegion *mr, int length)
+static inline bool migration_bitmap_set_dirty(MemoryRegion *mr,
+                                              ram_addr_t offset)
 {
-    ram_addr_t addr;
+    bool ret;
+    int nr = (mr->ram_addr + offset) >> TARGET_PAGE_BITS;
 
-    for (addr = 0; addr < length; addr += TARGET_PAGE_SIZE) {
-        if (!memory_region_get_dirty(mr, addr, TARGET_PAGE_SIZE,
-                                     DIRTY_MEMORY_MIGRATION)) {
-            memory_region_set_dirty(mr, addr, TARGET_PAGE_SIZE);
-        }
+    ret = test_and_set_bit(nr, migration_bitmap);
+
+    if (!ret) {
+        migration_dirty_pages++;
     }
+    return ret;
 }
 
 static void migration_bitmap_sync(void)
 {
-    uint64_t num_dirty_pages_init = ram_list.dirty_pages;
+    RAMBlock *block;
+    ram_addr_t addr;
+    uint64_t num_dirty_pages_init = migration_dirty_pages;
 
     trace_migration_bitmap_sync_start();
     memory_global_sync_dirty_bitmap(get_system_memory());
-    trace_migration_bitmap_sync_end(ram_list.dirty_pages
+
+    QLIST_FOREACH(block, &ram_list.blocks, next) {
+        for (addr = 0; addr < block->length; addr += TARGET_PAGE_SIZE) {
+            if (memory_region_get_dirty(block->mr, addr, TARGET_PAGE_SIZE,
+                                        DIRTY_MEMORY_MIGRATION)) {
+                migration_bitmap_set_dirty(block->mr, addr);
+            }
+        }
+        memory_region_reset_dirty(block->mr, 0, block->length,
+                                  DIRTY_MEMORY_MIGRATION);
+    }
+    trace_migration_bitmap_sync_end(migration_dirty_pages
                                     - num_dirty_pages_init);
 }
 
@@ -442,7 +462,7 @@ static uint64_t bytes_transferred;
 
 static ram_addr_t ram_save_remaining(void)
 {
-    return ram_list.dirty_pages;
+    return migration_dirty_pages;
 }
 
 uint64_t ram_bytes_remaining(void)
@@ -527,6 +547,11 @@ static void reset_ram_globals(void)
 static int ram_save_setup(QEMUFile *f, void *opaque)
 {
     RAMBlock *block;
+    int64_t ram_pages = last_ram_offset() >> TARGET_PAGE_BITS;
+
+    migration_bitmap = bitmap_new(ram_pages);
+    bitmap_set(migration_bitmap, 1, ram_pages);
+    migration_dirty_pages = ram_pages;
 
     bytes_transferred = 0;
     reset_ram_globals();
@@ -544,13 +569,8 @@ static int ram_save_setup(QEMUFile *f, void *opaque)
         acct_clear();
     }
 
-    /* Make sure all dirty bits are set */
-    QLIST_FOREACH(block, &ram_list.blocks, next) {
-        migration_bitmap_set_dirty(block->mr, block->length);
-    }
-
     memory_global_dirty_log_start();
-    memory_global_sync_dirty_bitmap(get_system_memory());
+    migration_bitmap_sync();
 
     qemu_put_be64(f, ram_bytes_total() | RAM_SAVE_FLAG_MEM_SIZE);
 
@@ -655,6 +675,9 @@ static int ram_save_complete(QEMUFile *f, void *opaque)
 
     qemu_put_be64(f, RAM_SAVE_FLAG_EOS);
 
+    g_free(migration_bitmap);
+    migration_bitmap = NULL;
+
     return 0;
 }
 
diff --git a/cpu-all.h b/cpu-all.h
index 287b0003ea..6aa7e58cb1 100644
--- a/cpu-all.h
+++ b/cpu-all.h
@@ -500,7 +500,6 @@ typedef struct RAMBlock {
 typedef struct RAMList {
     uint8_t *phys_dirty;
     QLIST_HEAD(, RAMBlock) blocks;
-    uint64_t dirty_pages;
 } RAMList;
 extern RAMList ram_list;
 
diff --git a/exec-obsolete.h b/exec-obsolete.h
index 286e2f75e1..6d35d4b83e 100644
--- a/exec-obsolete.h
+++ b/exec-obsolete.h
@@ -75,11 +75,6 @@ static inline int cpu_physical_memory_get_dirty(ram_addr_t start,
 static inline int cpu_physical_memory_set_dirty_flags(ram_addr_t addr,
                                                       int dirty_flags)
 {
-    if ((dirty_flags & MIGRATION_DIRTY_FLAG) &&
-        !cpu_physical_memory_get_dirty(addr, TARGET_PAGE_SIZE,
-                                       MIGRATION_DIRTY_FLAG)) {
-        ram_list.dirty_pages++;
-    }
     return ram_list.phys_dirty[addr >> TARGET_PAGE_BITS] |= dirty_flags;
 }
 
@@ -93,11 +88,6 @@ static inline int cpu_physical_memory_clear_dirty_flags(ram_addr_t addr,
 {
     int mask = ~dirty_flags;
 
-    if ((dirty_flags & MIGRATION_DIRTY_FLAG) &&
-        cpu_physical_memory_get_dirty(addr, TARGET_PAGE_SIZE,
-                                      MIGRATION_DIRTY_FLAG)) {
-        ram_list.dirty_pages--;
-    }
     return ram_list.phys_dirty[addr >> TARGET_PAGE_BITS] &= mask;
 }