Merge remote-tracking branch 'quintela/migration-next-20121017' into staging

* quintela/migration-next-20121017: (41 commits)
  cpus: create qemu_in_vcpu_thread()
  savevm: make qemu_file_put_notify() return errors
  savevm: un-export qemu_file_set_error()
  block-migration: handle errors with the return codes correctly
  block-migration:  Switch meaning of return value
  block-migration: make flush_blks() return errors
  buffered_file: buffered_put_buffer() don't need to set last_error
  savevm: Only qemu_fflush() can generate errors
  savevm: make qemu_fill_buffer() be consistent
  savevm: unexport qemu_ftell()
  savevm: unfold qemu_fclose_internal()
  savevm: make qemu_fflush() return an error code
  savevm: Remove qemu_fseek()
  virtio-net: use qemu_get_buffer() in a temp buffer
  savevm: unexport qemu_fflush
  migration: make migrate_fd_wait_for_unfreeze() return errors
  buffered_file: make buffered_flush return the error code
  buffered_file: callers of buffered_flush() already check for errors
  buffered_file: We can access directly to bandwidth_limit
  buffered_file: unfold migrate_fd_close
  ...

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"
@@ -45,6 +47,7 @@
 #include "hw/pcspk.h"
 #include "qemu/page_cache.h"
 #include "qmp-commands.h"
+#include "trace.h"
 
 #ifdef DEBUG_ARCH_INIT
 #define DPRINTF(fmt, ...) \
@@ -330,6 +333,78 @@ 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;
+    int nr = (mr->ram_addr + offset) >> TARGET_PAGE_BITS;
+
+    ret = test_and_clear_bit(nr, migration_bitmap);
+
+    if (ret) {
+        migration_dirty_pages--;
+    }
+    return ret;
+}
+
+static inline bool migration_bitmap_set_dirty(MemoryRegion *mr,
+                                              ram_addr_t offset)
+{
+    bool ret;
+    int nr = (mr->ram_addr + offset) >> TARGET_PAGE_BITS;
+
+    ret = test_and_set_bit(nr, migration_bitmap);
+
+    if (!ret) {
+        migration_dirty_pages++;
+    }
+    return ret;
+}
+
+static void migration_bitmap_sync(void)
+{
+    RAMBlock *block;
+    ram_addr_t addr;
+    uint64_t num_dirty_pages_init = migration_dirty_pages;
+    MigrationState *s = migrate_get_current();
+    static int64_t start_time;
+    static int64_t num_dirty_pages_period;
+    int64_t end_time;
+
+    if (!start_time) {
+        start_time = qemu_get_clock_ms(rt_clock);
+    }
+
+    trace_migration_bitmap_sync_start();
+    memory_global_sync_dirty_bitmap(get_system_memory());
+
+    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);
+    num_dirty_pages_period += migration_dirty_pages - num_dirty_pages_init;
+    end_time = qemu_get_clock_ms(rt_clock);
+
+    /* more than 1 second = 1000 millisecons */
+    if (end_time > start_time + 1000) {
+        s->dirty_pages_rate = num_dirty_pages_period * 1000
+            / (end_time - start_time);
+        start_time = end_time;
+        num_dirty_pages_period = 0;
+    }
+}
+
 
 /*
  * ram_save_block: Writes a page of memory to the stream f
@@ -352,14 +427,10 @@ static int ram_save_block(QEMUFile *f, bool last_stage)
 
     do {
         mr = block->mr;
-        if (memory_region_get_dirty(mr, offset, TARGET_PAGE_SIZE,
+        if (migration_bitmap_test_and_reset_dirty(mr, offset)) {
-                                    DIRTY_MEMORY_MIGRATION)) {
             uint8_t *p;
             int cont = (block == last_block) ? RAM_SAVE_FLAG_CONTINUE : 0;
 
-            memory_region_reset_dirty(mr, offset, TARGET_PAGE_SIZE,
-                                      DIRTY_MEMORY_MIGRATION);
-
             p = memory_region_get_ram_ptr(mr) + offset;
 
             if (is_dup_page(p)) {
@@ -409,7 +480,7 @@ 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)
@@ -481,17 +552,27 @@ static void ram_migration_cancel(void *opaque)
     migration_end();
 }
 
+
+static void reset_ram_globals(void)
+{
+    last_block = NULL;
+    last_offset = 0;
+    sort_ram_list();
+}
+
 #define MAX_WAIT 50 /* ms, half buffered_file limit */
 
 static int ram_save_setup(QEMUFile *f, void *opaque)
 {
-    ram_addr_t addr;
     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;
-    last_block = NULL;
+    reset_ram_globals();
-    last_offset = 0;
-    sort_ram_list();
 
     if (migrate_use_xbzrle()) {
         XBZRLE.cache = cache_init(migrate_xbzrle_cache_size() /
@@ -506,17 +587,8 @@ 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) {
-        for (addr = 0; addr < block->length; addr += TARGET_PAGE_SIZE) {
-            if (!memory_region_get_dirty(block->mr, addr, TARGET_PAGE_SIZE,
-                                         DIRTY_MEMORY_MIGRATION)) {
-                memory_region_set_dirty(block->mr, addr, TARGET_PAGE_SIZE);
-            }
-        }
-    }
-
     memory_global_dirty_log_start();
+    migration_bitmap_sync();
 
     qemu_put_be64(f, ram_bytes_total() | RAM_SAVE_FLAG_MEM_SIZE);
 
@@ -537,7 +609,8 @@ static int ram_save_iterate(QEMUFile *f, void *opaque)
     double bwidth = 0;
     int ret;
     int i;
-    uint64_t expected_time;
+    uint64_t expected_downtime;
+    MigrationState *s = migrate_get_current();
 
     bytes_transferred_last = bytes_transferred;
     bwidth = qemu_get_clock_ns(rt_clock);
@@ -576,31 +649,32 @@ static int ram_save_iterate(QEMUFile *f, void *opaque)
     bwidth = qemu_get_clock_ns(rt_clock) - bwidth;
     bwidth = (bytes_transferred - bytes_transferred_last) / bwidth;
 
-    /* if we haven't transferred anything this round, force expected_time to a
+    /* if we haven't transferred anything this round, force
-     * a very high value, but without crashing */
+     * expected_downtime to a very high value, but without
+     * crashing */
     if (bwidth == 0) {
         bwidth = 0.000001;
     }
 
     qemu_put_be64(f, RAM_SAVE_FLAG_EOS);
 
-    expected_time = ram_save_remaining() * TARGET_PAGE_SIZE / bwidth;
+    expected_downtime = ram_save_remaining() * TARGET_PAGE_SIZE / bwidth;
+    DPRINTF("ram_save_live: expected(%" PRIu64 ") <= max(" PRIu64 ")?\n",
+            expected_downtime, migrate_max_downtime());
 
-    DPRINTF("ram_save_live: expected(%" PRIu64 ") <= max(%" PRIu64 ")?\n",
+    if (expected_downtime <= migrate_max_downtime()) {
-            expected_time, migrate_max_downtime());
+        migration_bitmap_sync();
+        expected_downtime = ram_save_remaining() * TARGET_PAGE_SIZE / bwidth;
+        s->expected_downtime = expected_downtime / 1000000; /* ns -> ms */
 
-    if (expected_time <= migrate_max_downtime()) {
+        return expected_downtime <= migrate_max_downtime();
-        memory_global_sync_dirty_bitmap(get_system_memory());
-        expected_time = ram_save_remaining() * TARGET_PAGE_SIZE / bwidth;
-
-        return expected_time <= migrate_max_downtime();
     }
     return 0;
 }
 
 static int ram_save_complete(QEMUFile *f, void *opaque)
 {
-    memory_global_sync_dirty_bitmap(get_system_memory());
+    migration_bitmap_sync();
 
     /* try transferring iterative blocks of memory */
 
@@ -619,6 +693,9 @@ 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;
 }
 

block-migration.c

@@ -423,20 +423,23 @@ static int mig_save_device_dirty(QEMUFile *f, BlkMigDevState *bmds,
 
 error:
     DPRINTF("Error reading sector %" PRId64 "\n", sector);
-    qemu_file_set_error(f, ret);
     g_free(blk->buf);
     g_free(blk);
-    return 0;
+    return ret;
 }
 
+/* return value:
+ * 0: too much data for max_downtime
+ * 1: few enough data for max_downtime
+*/
 static int blk_mig_save_dirty_block(QEMUFile *f, int is_async)
 {
     BlkMigDevState *bmds;
-    int ret = 0;
+    int ret = 1;
 
     QSIMPLEQ_FOREACH(bmds, &block_mig_state.bmds_list, entry) {
-        if (mig_save_device_dirty(f, bmds, is_async) == 0) {
+        ret = mig_save_device_dirty(f, bmds, is_async);
-            ret = 1;
+        if (ret <= 0) {
             break;
         }
     }
@@ -444,9 +447,10 @@ static int blk_mig_save_dirty_block(QEMUFile *f, int is_async)
     return ret;
 }
 
-static void flush_blks(QEMUFile* f)
+static int flush_blks(QEMUFile *f)
 {
     BlkMigBlock *blk;
+    int ret = 0;
 
     DPRINTF("%s Enter submitted %d read_done %d transferred %d\n",
             __FUNCTION__, block_mig_state.submitted, block_mig_state.read_done,
@@ -457,7 +461,7 @@ static void flush_blks(QEMUFile* f)
             break;
         }
         if (blk->ret < 0) {
-            qemu_file_set_error(f, blk->ret);
+            ret = blk->ret;
             break;
         }
         blk_send(f, blk);
@@ -474,6 +478,7 @@ static void flush_blks(QEMUFile* f)
     DPRINTF("%s Exit submitted %d read_done %d transferred %d\n", __FUNCTION__,
             block_mig_state.submitted, block_mig_state.read_done,
             block_mig_state.transferred);
+    return ret;
 }
 
 static int64_t get_remaining_dirty(void)
@@ -555,9 +560,7 @@ static int block_save_setup(QEMUFile *f, void *opaque)
     /* start track dirty blocks */
     set_dirty_tracking(1);
 
-    flush_blks(f);
+    ret = flush_blks(f);
-
-    ret = qemu_file_get_error(f);
     if (ret) {
         blk_mig_cleanup();
         return ret;
@@ -577,9 +580,7 @@ static int block_save_iterate(QEMUFile *f, void *opaque)
     DPRINTF("Enter save live iterate submitted %d transferred %d\n",
             block_mig_state.submitted, block_mig_state.transferred);
 
-    flush_blks(f);
+    ret = flush_blks(f);
-
-    ret = qemu_file_get_error(f);
     if (ret) {
         blk_mig_cleanup();
         return ret;
@@ -598,16 +599,19 @@ static int block_save_iterate(QEMUFile *f, void *opaque)
                 block_mig_state.bulk_completed = 1;
             }
         } else {
-            if (blk_mig_save_dirty_block(f, 1) == 0) {
+            ret = blk_mig_save_dirty_block(f, 1);
+            if (ret != 0) {
                 /* no more dirty blocks */
                 break;
             }
         }
     }
+    if (ret) {
+        blk_mig_cleanup();
+        return ret;
+    }
 
-    flush_blks(f);
+    ret = flush_blks(f);
-
-    ret = qemu_file_get_error(f);
     if (ret) {
         blk_mig_cleanup();
         return ret;
@@ -625,9 +629,7 @@ static int block_save_complete(QEMUFile *f, void *opaque)
     DPRINTF("Enter save live complete submitted %d transferred %d\n",
             block_mig_state.submitted, block_mig_state.transferred);
 
-    flush_blks(f);
+    ret = flush_blks(f);
-
-    ret = qemu_file_get_error(f);
     if (ret) {
         blk_mig_cleanup();
         return ret;
@@ -639,18 +641,16 @@ static int block_save_complete(QEMUFile *f, void *opaque)
        all async read completed */
     assert(block_mig_state.submitted == 0);
 
-    while (blk_mig_save_dirty_block(f, 0) != 0) {
+    do {
-        /* Do nothing */
+        ret = blk_mig_save_dirty_block(f, 0);
-    }
+    } while (ret == 0);
+
     blk_mig_cleanup();
-
-    /* report completion */
-    qemu_put_be64(f, (100 << BDRV_SECTOR_BITS) | BLK_MIG_FLAG_PROGRESS);
-
-    ret = qemu_file_get_error(f);
     if (ret) {
         return ret;
     }
+    /* report completion */
+    qemu_put_be64(f, (100 << BDRV_SECTOR_BITS) | BLK_MIG_FLAG_PROGRESS);
 
     DPRINTF("Block migration completed\n");
 

buffered_file.c

@@ -23,11 +23,7 @@
 
 typedef struct QEMUFileBuffered
 {
-    BufferedPutFunc *put_buffer;
+    MigrationState *migration_state;
-    BufferedPutReadyFunc *put_ready;
-    BufferedWaitForUnfreezeFunc *wait_for_unfreeze;
-    BufferedCloseFunc *close;
-    void *opaque;
     QEMUFile *file;
     int freeze_output;
     size_t bytes_xfer;
@@ -50,70 +46,60 @@ static void buffered_append(QEMUFileBuffered *s,
                             const uint8_t *buf, size_t size)
 {
     if (size > (s->buffer_capacity - s->buffer_size)) {
-        void *tmp;
-
         DPRINTF("increasing buffer capacity from %zu by %zu\n",
                 s->buffer_capacity, size + 1024);
 
         s->buffer_capacity += size + 1024;
 
-        tmp = g_realloc(s->buffer, s->buffer_capacity);
+        s->buffer = g_realloc(s->buffer, s->buffer_capacity);
-        if (tmp == NULL) {
-            fprintf(stderr, "qemu file buffer expansion failed\n");
-            exit(1);
-        }
-
-        s->buffer = tmp;
     }
 
     memcpy(s->buffer + s->buffer_size, buf, size);
     s->buffer_size += size;
 }
 
-static void buffered_flush(QEMUFileBuffered *s)
+static ssize_t buffered_flush(QEMUFileBuffered *s)
 {
     size_t offset = 0;
-    int error;
+    ssize_t ret = 0;
-
-    error = qemu_file_get_error(s->file);
-    if (error != 0) {
-        DPRINTF("flush when error, bailing: %s\n", strerror(-error));
-        return;
-    }
 
     DPRINTF("flushing %zu byte(s) of data\n", s->buffer_size);
 
-    while (offset < s->buffer_size) {
+    while (s->bytes_xfer < s->xfer_limit && offset < s->buffer_size) {
-        ssize_t ret;
 
-        ret = s->put_buffer(s->opaque, s->buffer + offset,
+        ret = migrate_fd_put_buffer(s->migration_state, s->buffer + offset,
-                            s->buffer_size - offset);
+                                    s->buffer_size - offset);
         if (ret == -EAGAIN) {
             DPRINTF("backend not ready, freezing\n");
+            ret = 0;
             s->freeze_output = 1;
             break;
         }
 
         if (ret <= 0) {
             DPRINTF("error flushing data, %zd\n", ret);
-            qemu_file_set_error(s->file, ret);
             break;
         } else {
             DPRINTF("flushed %zd byte(s)\n", ret);
             offset += ret;
+            s->bytes_xfer += ret;
         }
     }
 
     DPRINTF("flushed %zu of %zu byte(s)\n", offset, s->buffer_size);
     memmove(s->buffer, s->buffer + offset, s->buffer_size - offset);
     s->buffer_size -= offset;
+
+    if (ret < 0) {
+        return ret;
+    }
+    return offset;
 }
 
 static int buffered_put_buffer(void *opaque, const uint8_t *buf, int64_t pos, int size)
 {
     QEMUFileBuffered *s = opaque;
-    int offset = 0, error;
+    ssize_t error;
-    ssize_t ret;
 
     DPRINTF("putting %d bytes at %" PRId64 "\n", size, pos);
 
@@ -126,65 +112,54 @@ static int buffered_put_buffer(void *opaque, const uint8_t *buf, int64_t pos, in
     DPRINTF("unfreezing output\n");
     s->freeze_output = 0;
 
-    buffered_flush(s);
+    if (size > 0) {
-
+        DPRINTF("buffering %d bytes\n", size - offset);
-    while (!s->freeze_output && offset < size) {
+        buffered_append(s, buf, size);
-        if (s->bytes_xfer > s->xfer_limit) {
-            DPRINTF("transfer limit exceeded when putting\n");
-            break;
-        }
-
-        ret = s->put_buffer(s->opaque, buf + offset, size - offset);
-        if (ret == -EAGAIN) {
-            DPRINTF("backend not ready, freezing\n");
-            s->freeze_output = 1;
-            break;
-        }
-
-        if (ret <= 0) {
-            DPRINTF("error putting\n");
-            qemu_file_set_error(s->file, ret);
-            offset = -EINVAL;
-            break;
-        }
-
-        DPRINTF("put %zd byte(s)\n", ret);
-        offset += ret;
-        s->bytes_xfer += ret;
     }
 
-    if (offset >= 0) {
+    error = buffered_flush(s);
-        DPRINTF("buffering %d bytes\n", size - offset);
+    if (error < 0) {
-        buffered_append(s, buf + offset, size - offset);
+        DPRINTF("buffered flush error. bailing: %s\n", strerror(-error));
-        offset = size;
+        return error;
     }
 
     if (pos == 0 && size == 0) {
         DPRINTF("file is ready\n");
-        if (s->bytes_xfer <= s->xfer_limit) {
+        if (!s->freeze_output && s->bytes_xfer < s->xfer_limit) {
             DPRINTF("notifying client\n");
-            s->put_ready(s->opaque);
+            migrate_fd_put_ready(s->migration_state);
         }
     }
 
-    return offset;
+    return size;
 }
 
 static int buffered_close(void *opaque)
 {
     QEMUFileBuffered *s = opaque;
-    int ret;
+    ssize_t ret = 0;
+    int ret2;
 
     DPRINTF("closing\n");
 
+    s->xfer_limit = INT_MAX;
     while (!qemu_file_get_error(s->file) && s->buffer_size) {
-        buffered_flush(s);
+        ret = buffered_flush(s);
-        if (s->freeze_output)
+        if (ret < 0) {
-            s->wait_for_unfreeze(s->opaque);
+            break;
+        }
+        if (s->freeze_output) {
+            ret = migrate_fd_wait_for_unfreeze(s->migration_state);
+            if (ret < 0) {
+                break;
+            }
+        }
     }
 
-    ret = s->close(s->opaque);
+    ret2 = migrate_fd_close(s->migration_state);
-
+    if (ret >= 0) {
+        ret = ret2;
+    }
     qemu_del_timer(s->timer);
     qemu_free_timer(s->timer);
     g_free(s->buffer);
@@ -256,29 +231,17 @@ static void buffered_rate_tick(void *opaque)
 
     s->bytes_xfer = 0;
 
-    buffered_flush(s);
+    buffered_put_buffer(s, NULL, 0, 0);
-
-    /* Add some checks around this */
-    s->put_ready(s->opaque);
 }
 
-QEMUFile *qemu_fopen_ops_buffered(void *opaque,
+QEMUFile *qemu_fopen_ops_buffered(MigrationState *migration_state)
-                                  size_t bytes_per_sec,
-                                  BufferedPutFunc *put_buffer,
-                                  BufferedPutReadyFunc *put_ready,
-                                  BufferedWaitForUnfreezeFunc *wait_for_unfreeze,
-                                  BufferedCloseFunc *close)
 {
     QEMUFileBuffered *s;
 
     s = g_malloc0(sizeof(*s));
 
-    s->opaque = opaque;
+    s->migration_state = migration_state;
-    s->xfer_limit = bytes_per_sec / 10;
+    s->xfer_limit = migration_state->bandwidth_limit / 10;
-    s->put_buffer = put_buffer;
-    s->put_ready = put_ready;
-    s->wait_for_unfreeze = wait_for_unfreeze;
-    s->close = close;
 
     s->file = qemu_fopen_ops(s, buffered_put_buffer, NULL,
                              buffered_close, buffered_rate_limit,

buffered_file.h

@@ -15,16 +15,8 @@
 #define QEMU_BUFFERED_FILE_H
 
 #include "hw/hw.h"
+#include "migration.h"
 
-typedef ssize_t (BufferedPutFunc)(void *opaque, const void *data, size_t size);
+QEMUFile *qemu_fopen_ops_buffered(MigrationState *migration_state);
-typedef void (BufferedPutReadyFunc)(void *opaque);
-typedef void (BufferedWaitForUnfreezeFunc)(void *opaque);
-typedef int (BufferedCloseFunc)(void *opaque);
-
-QEMUFile *qemu_fopen_ops_buffered(void *opaque, size_t xfer_limit,
-                                  BufferedPutFunc *put_buffer,
-                                  BufferedPutReadyFunc *put_ready,
-                                  BufferedWaitForUnfreezeFunc *wait_for_unfreeze,
-                                  BufferedCloseFunc *close);
 
 #endif

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;
 
@@ -518,6 +517,7 @@ extern int mem_prealloc;
 #define TLB_MMIO        (1 << 5)
 
 void dump_exec_info(FILE *f, fprintf_function cpu_fprintf);
+ram_addr_t last_ram_offset(void);
 #endif /* !CONFIG_USER_ONLY */
 
 int cpu_memory_rw_debug(CPUArchState *env, target_ulong addr,

cpus.c

@@ -898,6 +898,11 @@ int qemu_cpu_is_self(void *_env)
     return qemu_thread_is_self(cpu->thread);
 }
 
+static bool qemu_in_vcpu_thread(void)
+{
+    return cpu_single_env && qemu_cpu_is_self(cpu_single_env);
+}
+
 void qemu_mutex_lock_iothread(void)
 {
     if (!tcg_enabled()) {
@@ -943,7 +948,7 @@ void pause_all_vcpus(void)
         penv = penv->next_cpu;
     }
 
-    if (!qemu_thread_is_self(&io_thread)) {
+    if (qemu_in_vcpu_thread()) {
         cpu_stop_current();
         if (!kvm_enabled()) {
             while (penv) {
@@ -1060,7 +1065,7 @@ void cpu_stop_current(void)
 
 void vm_stop(RunState state)
 {
-    if (!qemu_thread_is_self(&io_thread)) {
+    if (qemu_in_vcpu_thread()) {
         qemu_system_vmstop_request(state);
         /*
          * FIXME: should not return to device code in case

exec.c

@@ -2449,7 +2449,7 @@ static ram_addr_t find_ram_offset(ram_addr_t size)
     return offset;
 }
 
-static ram_addr_t last_ram_offset(void)
+ram_addr_t last_ram_offset(void)
 {
     RAMBlock *block;
     ram_addr_t last = 0;

hmp.c

@@ -152,6 +152,14 @@ void hmp_info_migrate(Monitor *mon)
         monitor_printf(mon, "Migration status: %s\n", info->status);
         monitor_printf(mon, "total time: %" PRIu64 " milliseconds\n",
                        info->total_time);
+        if (info->has_expected_downtime) {
+            monitor_printf(mon, "expected downtime: %" PRIu64 " milliseconds\n",
+                           info->expected_downtime);
+        }
+        if (info->has_downtime) {
+            monitor_printf(mon, "downtime: %" PRIu64 " milliseconds\n",
+                           info->downtime);
+        }
     }
 
     if (info->has_ram) {
@@ -167,6 +175,10 @@ void hmp_info_migrate(Monitor *mon)
                        info->ram->normal);
         monitor_printf(mon, "normal bytes: %" PRIu64 " kbytes\n",
                        info->ram->normal_bytes >> 10);
+        if (info->ram->dirty_pages_rate) {
+            monitor_printf(mon, "dirty pages rate: %" PRIu64 " pages\n",
+                           info->ram->dirty_pages_rate);
+        }
     }
 
     if (info->has_disk) {

hw/virtio-net.c

@@ -921,7 +921,9 @@ static int virtio_net_load(QEMUFile *f, void *opaque, int version_id)
             qemu_get_buffer(f, n->mac_table.macs,
                             n->mac_table.in_use * ETH_ALEN);
         } else if (n->mac_table.in_use) {
-            qemu_fseek(f, n->mac_table.in_use * ETH_ALEN, SEEK_CUR);
+            uint8_t *buf = g_malloc0(n->mac_table.in_use);
+            qemu_get_buffer(f, buf, n->mac_table.in_use * ETH_ALEN);
+            g_free(buf);
             n->mac_table.multi_overflow = n->mac_table.uni_overflow = 1;
             n->mac_table.in_use = 0;
         }

memory-internal.h

@@ -90,11 +90,6 @@ 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;
 }
 
@@ -108,11 +103,6 @@ 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;
 }
 

migration.c

@@ -53,7 +53,7 @@ static NotifierList migration_state_notifiers =
    migrations at once.  For now we don't need to add
    dynamic creation of migration */
 
-static MigrationState *migrate_get_current(void)
+MigrationState *migrate_get_current(void)
 {
     static MigrationState current_migration = {
         .state = MIG_STATE_SETUP,
@@ -169,6 +169,8 @@ MigrationInfo *qmp_query_migrate(Error **errp)
         info->has_total_time = true;
         info->total_time = qemu_get_clock_ms(rt_clock)
             - s->total_time;
+        info->has_expected_downtime = true;
+        info->expected_downtime = s->expected_downtime;
 
         info->has_ram = true;
         info->ram = g_malloc0(sizeof(*info->ram));
@@ -178,6 +180,8 @@ MigrationInfo *qmp_query_migrate(Error **errp)
         info->ram->duplicate = dup_mig_pages_transferred();
         info->ram->normal = norm_mig_pages_transferred();
         info->ram->normal_bytes = norm_mig_bytes_transferred();
+        info->ram->dirty_pages_rate = s->dirty_pages_rate;
+
 
         if (blk_mig_active()) {
             info->has_disk = true;
@@ -195,6 +199,8 @@ MigrationInfo *qmp_query_migrate(Error **errp)
         info->has_status = true;
         info->status = g_strdup("completed");
         info->total_time = s->total_time;
+        info->has_downtime = true;
+        info->downtime = s->downtime;
 
         info->has_ram = true;
         info->ram = g_malloc0(sizeof(*info->ram));
@@ -281,18 +287,18 @@ static void migrate_fd_completed(MigrationState *s)
 static void migrate_fd_put_notify(void *opaque)
 {
     MigrationState *s = opaque;
+    int ret;
 
     qemu_set_fd_handler2(s->fd, NULL, NULL, NULL, NULL);
-    qemu_file_put_notify(s->file);
+    ret = qemu_file_put_notify(s->file);
-    if (s->file && qemu_file_get_error(s->file)) {
+    if (ret) {
         migrate_fd_error(s);
     }
 }
 
-static ssize_t migrate_fd_put_buffer(void *opaque, const void *data,
+ssize_t migrate_fd_put_buffer(MigrationState *s, const void *data,
-                                     size_t size)
+                              size_t size)
 {
-    MigrationState *s = opaque;
     ssize_t ret;
 
     if (s->state != MIG_STATE_ACTIVE) {
@@ -313,9 +319,8 @@ static ssize_t migrate_fd_put_buffer(void *opaque, const void *data,
     return ret;
 }
 
-static void migrate_fd_put_ready(void *opaque)
+void migrate_fd_put_ready(MigrationState *s)
 {
-    MigrationState *s = opaque;
     int ret;
 
     if (s->state != MIG_STATE_ACTIVE) {
@@ -329,8 +334,10 @@ static void migrate_fd_put_ready(void *opaque)
         migrate_fd_error(s);
     } else if (ret == 1) {
         int old_vm_running = runstate_is_running();
+        int64_t start_time, end_time;
 
         DPRINTF("done iterating\n");
+        start_time = qemu_get_clock_ms(rt_clock);
         qemu_system_wakeup_request(QEMU_WAKEUP_REASON_OTHER);
         vm_stop_force_state(RUN_STATE_FINISH_MIGRATE);
 
@@ -339,7 +346,9 @@ static void migrate_fd_put_ready(void *opaque)
         } else {
             migrate_fd_completed(s);
         }
-        s->total_time = qemu_get_clock_ms(rt_clock) - s->total_time;
+        end_time = qemu_get_clock_ms(rt_clock);
+        s->total_time = end_time - s->total_time;
+        s->downtime = end_time - start_time;
         if (s->state != MIG_STATE_COMPLETED) {
             if (old_vm_running) {
                 vm_start();
@@ -362,14 +371,13 @@ static void migrate_fd_cancel(MigrationState *s)
     migrate_fd_cleanup(s);
 }
 
-static void migrate_fd_wait_for_unfreeze(void *opaque)
+int migrate_fd_wait_for_unfreeze(MigrationState *s)
 {
-    MigrationState *s = opaque;
     int ret;
 
     DPRINTF("wait for unfreeze\n");
     if (s->state != MIG_STATE_ACTIVE)
-        return;
+        return -EINVAL;
 
     do {
         fd_set wfds;
@@ -381,14 +389,13 @@ static void migrate_fd_wait_for_unfreeze(void *opaque)
     } while (ret == -1 && (s->get_error(s)) == EINTR);
 
     if (ret == -1) {
-        qemu_file_set_error(s->file, -s->get_error(s));
+        return -s->get_error(s);
     }
+    return 0;
 }
 
-static int migrate_fd_close(void *opaque)
+int migrate_fd_close(MigrationState *s)
 {
-    MigrationState *s = opaque;
-
     qemu_set_fd_handler2(s->fd, NULL, NULL, NULL, NULL);
     return s->close(s);
 }
@@ -424,12 +431,7 @@ void migrate_fd_connect(MigrationState *s)
     int ret;
 
     s->state = MIG_STATE_ACTIVE;
-    s->file = qemu_fopen_ops_buffered(s,
+    s->file = qemu_fopen_ops_buffered(s);
-                                      s->bandwidth_limit,
-                                      migrate_fd_put_buffer,
-                                      migrate_fd_put_ready,
-                                      migrate_fd_wait_for_unfreeze,
-                                      migrate_fd_close);
 
     DPRINTF("beginning savevm\n");
     ret = qemu_savevm_state_begin(s->file, &s->params);

migration.h

@@ -40,6 +40,9 @@ struct MigrationState
     void *opaque;
     MigrationParams params;
     int64_t total_time;
+    int64_t downtime;
+    int64_t expected_downtime;
+    int64_t dirty_pages_rate;
     bool enabled_capabilities[MIGRATION_CAPABILITY_MAX];
     int64_t xbzrle_cache_size;
 };
@@ -75,11 +78,18 @@ void migrate_fd_error(MigrationState *s);
 
 void migrate_fd_connect(MigrationState *s);
 
+ssize_t migrate_fd_put_buffer(MigrationState *s, const void *data,
+                              size_t size);
+void migrate_fd_put_ready(MigrationState *s);
+int migrate_fd_wait_for_unfreeze(MigrationState *s);
+int migrate_fd_close(MigrationState *s);
+
 void add_migration_state_change_notifier(Notifier *notify);
 void remove_migration_state_change_notifier(Notifier *notify);
 bool migration_is_active(MigrationState *);
 bool migration_has_finished(MigrationState *);
 bool migration_has_failed(MigrationState *);
+MigrationState *migrate_get_current(void);
 
 uint64_t ram_bytes_remaining(void);
 uint64_t ram_bytes_transferred(void);

qapi-schema.json

@@ -383,13 +383,17 @@
 #
 # @normal : number of normal pages (since 1.2)
 #
-# @normal-bytes : number of normal bytes sent (since 1.2)
+# @normal-bytes: number of normal bytes sent (since 1.2)
+#
+# @dirty-pages-rate: number of pages dirtied by second by the
+#        guest (since 1.3)
 #
 # Since: 0.14.0
 ##
 { 'type': 'MigrationStats',
   'data': {'transferred': 'int', 'remaining': 'int', 'total': 'int' ,
-           'duplicate': 'int', 'normal': 'int', 'normal-bytes': 'int' } }
+           'duplicate': 'int', 'normal': 'int', 'normal-bytes': 'int',
+           'dirty-pages-rate' : 'int' } }
 
 ##
 # @XBZRLECacheStats
@@ -438,13 +442,23 @@
 #        If migration has ended, it returns the total migration
 #        time. (since 1.2)
 #
+# @downtime: #optional only present when migration finishes correctly
+#        total downtime in milliseconds for the guest.
+#        (since 1.3)
+#
+# @expected-downtime: #optional only present while migration is active
+#        expected downtime in milliseconds for the guest in last walk
+#        of the dirty bitmap. (since 1.3)
+#
 # Since: 0.14.0
 ##
 { 'type': 'MigrationInfo',
   'data': {'*status': 'str', '*ram': 'MigrationStats',
            '*disk': 'MigrationStats',
            '*xbzrle-cache': 'XBZRLECacheStats',
-           '*total-time': 'int'} }
+           '*total-time': 'int',
+           '*expected-downtime': 'int',
+           '*downtime': 'int'} }
 
 ##
 # @query-migrate

qemu-file.h

@@ -71,7 +71,6 @@ QEMUFile *qemu_fopen_socket(int fd);
 QEMUFile *qemu_popen(FILE *popen_file, const char *mode);
 QEMUFile *qemu_popen_cmd(const char *command, const char *mode);
 int qemu_stdio_fd(QEMUFile *f);
-void qemu_fflush(QEMUFile *f);
 int qemu_fclose(QEMUFile *f);
 void qemu_put_buffer(QEMUFile *f, const uint8_t *buf, int size);
 void qemu_put_byte(QEMUFile *f, int v);
@@ -104,12 +103,11 @@ int qemu_file_rate_limit(QEMUFile *f);
 int64_t qemu_file_set_rate_limit(QEMUFile *f, int64_t new_rate);
 int64_t qemu_file_get_rate_limit(QEMUFile *f);
 int qemu_file_get_error(QEMUFile *f);
-void qemu_file_set_error(QEMUFile *f, int error);
 
 /* Try to send any outstanding data.  This function is useful when output is
  * halted due to rate limiting or EAGAIN errors occur as it can be used to
  * resume output. */
-void qemu_file_put_notify(QEMUFile *f);
+int qemu_file_put_notify(QEMUFile *f);
 
 static inline void qemu_put_be64s(QEMUFile *f, const uint64_t *pv)
 {
@@ -231,8 +229,4 @@ static inline void qemu_get_sbe64s(QEMUFile *f, int64_t *pv)
 {
     qemu_get_be64s(f, (uint64_t *)pv);
 }
-
-int64_t qemu_ftell(QEMUFile *f);
-int64_t qemu_fseek(QEMUFile *f, int64_t pos, int whence);
-
 #endif

qmp-commands.hx

@@ -2304,6 +2304,11 @@ The main json-object contains the following:
 - "total-time": total amount of ms since migration started.  If
                 migration has ended, it returns the total migration
 		 time (json-int)
+- "downtime": only present when migration has finished correctly
+              total amount in ms for downtime that happened (json-int)
+- "expected-downtime": only present while migration is active
+                total amount in ms for downtime that was calculated on
+		the last bitmap round (json-int)
 - "ram": only present if "status" is "active", it is a json-object with the
   following RAM information (in bytes):
          - "transferred": amount transferred (json-int)
@@ -2341,6 +2346,7 @@ Examples:
           "remaining":123,
           "total":246,
           "total-time":12345,
+          "downtime":12345,
           "duplicate":123,
           "normal":123,
           "normal-bytes":123456
@@ -2364,6 +2370,7 @@ Examples:
             "remaining":123,
             "total":246,
             "total-time":12345,
+            "expected-downtime":12345,
             "duplicate":123,
             "normal":123,
             "normal-bytes":123456
@@ -2382,6 +2389,7 @@ Examples:
             "remaining":1053304,
             "transferred":3720,
             "total-time":12345,
+            "expected-downtime":12345,
             "duplicate":123,
             "normal":123,
             "normal-bytes":123456
@@ -2406,6 +2414,7 @@ Examples:
             "remaining":1053304,
             "transferred":3720,
             "total-time":12345,
+            "expected-downtime":12345,
             "duplicate":10,
             "normal":3333,
             "normal-bytes":3412992

savevm.c

@@ -440,42 +440,29 @@ int qemu_file_get_error(QEMUFile *f)
     return f->last_error;
 }
 
-void qemu_file_set_error(QEMUFile *f, int ret)
+static void qemu_file_set_error(QEMUFile *f, int ret)
 {
     f->last_error = ret;
 }
 
-/** Sets last_error conditionally
- *
- * Sets last_error only if ret is negative _and_ no error
- * was set before.
- */
-static void qemu_file_set_if_error(QEMUFile *f, int ret)
-{
-    if (ret < 0 && !f->last_error) {
-        qemu_file_set_error(f, ret);
-    }
-}
-
 /** Flushes QEMUFile buffer
  *
- * In case of error, last_error is set.
  */
-void qemu_fflush(QEMUFile *f)
+static int qemu_fflush(QEMUFile *f)
 {
+    int ret = 0;
+
     if (!f->put_buffer)
-        return;
+        return 0;
 
     if (f->is_write && f->buf_index > 0) {
-        int len;
+        ret = f->put_buffer(f->opaque, f->buf, f->buf_offset, f->buf_index);
-
+        if (ret >= 0) {
-        len = f->put_buffer(f->opaque, f->buf, f->buf_offset, f->buf_index);
-        if (len > 0)
             f->buf_offset += f->buf_index;
-        else
+        }
-            qemu_file_set_error(f, -EINVAL);
         f->buf_index = 0;
     }
+    return ret;
 }
 
 static void qemu_fill_buffer(QEMUFile *f)
@@ -502,27 +489,11 @@ static void qemu_fill_buffer(QEMUFile *f)
         f->buf_size += len;
         f->buf_offset += len;
     } else if (len == 0) {
-        f->last_error = -EIO;
+        qemu_file_set_error(f, -EIO);
     } else if (len != -EAGAIN)
         qemu_file_set_error(f, len);
 }
 
-/** Calls close function and set last_error if needed
- *
- * Internal function. qemu_fflush() must be called before this.
- *
- * Returns f->close() return value, or 0 if close function is not set.
- */
-static int qemu_fclose_internal(QEMUFile *f)
-{
-    int ret = 0;
-    if (f->close) {
-        ret = f->close(f->opaque);
-        qemu_file_set_if_error(f, ret);
-    }
-    return ret;
-}
-
 /** Closes the file
  *
  * Returns negative error value if any error happened on previous operations or
@@ -534,8 +505,14 @@ static int qemu_fclose_internal(QEMUFile *f)
 int qemu_fclose(QEMUFile *f)
 {
     int ret;
-    qemu_fflush(f);
+    ret = qemu_fflush(f);
-    ret = qemu_fclose_internal(f);
+
+    if (f->close) {
+        int ret2 = f->close(f->opaque);
+        if (ret >= 0) {
+            ret = ret2;
+        }
+    }
     /* If any error was spotted before closing, we should report it
      * instead of the close() return value.
      */
@@ -546,22 +523,26 @@ int qemu_fclose(QEMUFile *f)
     return ret;
 }
 
-void qemu_file_put_notify(QEMUFile *f)
+int qemu_file_put_notify(QEMUFile *f)
 {
-    f->put_buffer(f->opaque, NULL, 0, 0);
+    return f->put_buffer(f->opaque, NULL, 0, 0);
 }
 
 void qemu_put_buffer(QEMUFile *f, const uint8_t *buf, int size)
 {
     int l;
 
-    if (!f->last_error && f->is_write == 0 && f->buf_index > 0) {
+    if (f->last_error) {
+        return;
+    }
+
+    if (f->is_write == 0 && f->buf_index > 0) {
         fprintf(stderr,
                 "Attempted to write to buffer while read buffer is not empty\n");
         abort();
     }
 
-    while (!f->last_error && size > 0) {
+    while (size > 0) {
         l = IO_BUF_SIZE - f->buf_index;
         if (l > size)
             l = size;
@@ -570,14 +551,23 @@ void qemu_put_buffer(QEMUFile *f, const uint8_t *buf, int size)
         f->buf_index += l;
         buf += l;
         size -= l;
-        if (f->buf_index >= IO_BUF_SIZE)
+        if (f->buf_index >= IO_BUF_SIZE) {
-            qemu_fflush(f);
+            int ret = qemu_fflush(f);
+            if (ret < 0) {
+                qemu_file_set_error(f, ret);
+                break;
+            }
+        }
     }
 }
 
 void qemu_put_byte(QEMUFile *f, int v)
 {
-    if (!f->last_error && f->is_write == 0 && f->buf_index > 0) {
+    if (f->last_error) {
+        return;
+    }
+
+    if (f->is_write == 0 && f->buf_index > 0) {
         fprintf(stderr,
                 "Attempted to write to buffer while read buffer is not empty\n");
         abort();
@@ -585,8 +575,12 @@ void qemu_put_byte(QEMUFile *f, int v)
 
     f->buf[f->buf_index++] = v;
     f->is_write = 1;
-    if (f->buf_index >= IO_BUF_SIZE)
+    if (f->buf_index >= IO_BUF_SIZE) {
-        qemu_fflush(f);
+        int ret = qemu_fflush(f);
+        if (ret < 0) {
+            qemu_file_set_error(f, ret);
+        }
+    }
 }
 
 static void qemu_file_skip(QEMUFile *f, int size)
@@ -671,32 +665,11 @@ int qemu_get_byte(QEMUFile *f)
     return result;
 }
 
-int64_t qemu_ftell(QEMUFile *f)
+static int64_t qemu_ftell(QEMUFile *f)
 {
     return f->buf_offset - f->buf_size + f->buf_index;
 }
 
-int64_t qemu_fseek(QEMUFile *f, int64_t pos, int whence)
-{
-    if (whence == SEEK_SET) {
-        /* nothing to do */
-    } else if (whence == SEEK_CUR) {
-        pos += qemu_ftell(f);
-    } else {
-        /* SEEK_END not supported */
-        return -1;
-    }
-    if (f->put_buffer) {
-        qemu_fflush(f);
-        f->buf_offset = pos;
-    } else {
-        f->buf_offset = pos;
-        f->buf_index = 0;
-        f->buf_size = 0;
-    }
-    return pos;
-}
-
 int qemu_file_rate_limit(QEMUFile *f)
 {
     if (f->rate_limit)

trace-events

@@ -921,6 +921,10 @@ ppm_save(const char *filename, void *display_surface) "%s surface=%p"
 savevm_section_start(void) ""
 savevm_section_end(unsigned int section_id) "section_id %u"
 
+# arch_init.c
+migration_bitmap_sync_start(void) ""
+migration_bitmap_sync_end(uint64_t dirty_pages) "dirty_pages %" PRIu64""
+
 # hw/qxl.c
 disable qxl_interface_set_mm_time(int qid, uint32_t mm_time) "%d %d"
 disable qxl_io_write_vga(int qid, const char *mode, uint32_t addr, uint32_t val) "%d %s addr=%u val=%u"