Merge remote-tracking branch 'remotes/mjt/tags/trivial-patches-2014-02-02' into staging

[qemu.git] / migration.c

diff --git a/migration.c b/migration.c
index bfbc34544a30a5ab2959bb3fe3debbd4944cb076..7235c23ffe0ea90b5c5e54d6941e1f9ce504111d 100644 (file)
--- a/migration.c
+++ b/migration.c
@@ -14,6 +14,7 @@
  */
 
 #include "qemu-common.h"
  */
 
 #include "qemu-common.h"

+#include "qemu/main-loop.h"

 #include "migration/migration.h"
 #include "monitor/monitor.h"
 #include "migration/qemu-file.h"
 #include "migration/migration.h"
 #include "monitor/monitor.h"
 #include "migration/qemu-file.h"


@@ -36,8 +37,10 @@
 #endif
 
 enum {
 #endif
 
 enum {

-    MIG_STATE_ERROR,
+    MIG_STATE_ERROR = -1,
+    MIG_STATE_NONE,

     MIG_STATE_SETUP,
     MIG_STATE_SETUP,

+    MIG_STATE_CANCELLING,

     MIG_STATE_CANCELLED,
     MIG_STATE_ACTIVE,
     MIG_STATE_COMPLETED,
     MIG_STATE_CANCELLED,
     MIG_STATE_ACTIVE,
     MIG_STATE_COMPLETED,


@@ -63,9 +66,10 @@ static NotifierList migration_state_notifiers =
 MigrationState *migrate_get_current(void)
 {
     static MigrationState current_migration = {
 MigrationState *migrate_get_current(void)
 {
     static MigrationState current_migration = {

-        .state = MIG_STATE_SETUP,
+        .state = MIG_STATE_NONE,

         .bandwidth_limit = MAX_THROTTLE,
         .xbzrle_cache_size = DEFAULT_MIGRATE_CACHE_SIZE,
         .bandwidth_limit = MAX_THROTTLE,
         .xbzrle_cache_size = DEFAULT_MIGRATE_CACHE_SIZE,

+        .mbps = -1,

     };
 
     return &current_migration;
     };
 
     return &current_migration;


@@ -77,6 +81,10 @@ void qemu_start_incoming_migration(const char *uri, Error **errp)
 
     if (strstart(uri, "tcp:", &p))
         tcp_start_incoming_migration(p, errp);
 
     if (strstart(uri, "tcp:", &p))
         tcp_start_incoming_migration(p, errp);

+#ifdef CONFIG_RDMA
+    else if (strstart(uri, "x-rdma:", &p))
+        rdma_start_incoming_migration(p, errp);
+#endif

 #if !defined(WIN32)
     else if (strstart(uri, "exec:", &p))
         exec_start_incoming_migration(p, errp);
 #if !defined(WIN32)
     else if (strstart(uri, "exec:", &p))
         exec_start_incoming_migration(p, errp);


@@ -143,6 +151,7 @@ MigrationCapabilityStatusList *qmp_query_migrate_capabilities(Error **errp)
     MigrationState *s = migrate_get_current();
     int i;
 
     MigrationState *s = migrate_get_current();
     int i;
 

+    caps = NULL; /* silence compiler warning */

     for (i = 0; i < MIGRATION_CAPABILITY_MAX; i++) {
         if (head == NULL) {
             head = g_malloc0(sizeof(*caps));
     for (i = 0; i < MIGRATION_CAPABILITY_MAX; i++) {
         if (head == NULL) {
             head = g_malloc0(sizeof(*caps));


@@ -179,17 +188,25 @@ MigrationInfo *qmp_query_migrate(Error **errp)
     MigrationState *s = migrate_get_current();
 
     switch (s->state) {
     MigrationState *s = migrate_get_current();
 
     switch (s->state) {

-    case MIG_STATE_SETUP:
+    case MIG_STATE_NONE:

         /* no migration has happened ever */
         break;
         /* no migration has happened ever */
         break;

+    case MIG_STATE_SETUP:
+        info->has_status = true;
+        info->status = g_strdup("setup");
+        info->has_total_time = false;
+        break;

     case MIG_STATE_ACTIVE:
     case MIG_STATE_ACTIVE:

+    case MIG_STATE_CANCELLING:

         info->has_status = true;
         info->status = g_strdup("active");
         info->has_total_time = true;
         info->has_status = true;
         info->status = g_strdup("active");
         info->has_total_time = true;

-        info->total_time = qemu_get_clock_ms(rt_clock)
+        info->total_time = qemu_clock_get_ms(QEMU_CLOCK_REALTIME)

             - s->total_time;
         info->has_expected_downtime = true;
         info->expected_downtime = s->expected_downtime;
             - s->total_time;
         info->has_expected_downtime = true;
         info->expected_downtime = s->expected_downtime;

+        info->has_setup_time = true;
+        info->setup_time = s->setup_time;

 
         info->has_ram = true;
         info->ram = g_malloc0(sizeof(*info->ram));
 
         info->has_ram = true;
         info->ram = g_malloc0(sizeof(*info->ram));


@@ -201,6 +218,7 @@ MigrationInfo *qmp_query_migrate(Error **errp)
         info->ram->normal = norm_mig_pages_transferred();
         info->ram->normal_bytes = norm_mig_bytes_transferred();
         info->ram->dirty_pages_rate = s->dirty_pages_rate;
         info->ram->normal = norm_mig_pages_transferred();
         info->ram->normal_bytes = norm_mig_bytes_transferred();
         info->ram->dirty_pages_rate = s->dirty_pages_rate;

+        info->ram->mbps = s->mbps;

 
         if (blk_mig_active()) {
             info->has_disk = true;
 
         if (blk_mig_active()) {
             info->has_disk = true;


@@ -217,9 +235,12 @@ MigrationInfo *qmp_query_migrate(Error **errp)
 
         info->has_status = true;
         info->status = g_strdup("completed");
 
         info->has_status = true;
         info->status = g_strdup("completed");

+        info->has_total_time = true;

         info->total_time = s->total_time;
         info->has_downtime = true;
         info->downtime = s->downtime;
         info->total_time = s->total_time;
         info->has_downtime = true;
         info->downtime = s->downtime;

+        info->has_setup_time = true;
+        info->setup_time = s->setup_time;

 
         info->has_ram = true;
         info->ram = g_malloc0(sizeof(*info->ram));
 
         info->has_ram = true;
         info->ram = g_malloc0(sizeof(*info->ram));


@@ -230,6 +251,7 @@ MigrationInfo *qmp_query_migrate(Error **errp)
         info->ram->skipped = skipped_mig_pages_transferred();
         info->ram->normal = norm_mig_pages_transferred();
         info->ram->normal_bytes = norm_mig_bytes_transferred();
         info->ram->skipped = skipped_mig_pages_transferred();
         info->ram->normal = norm_mig_pages_transferred();
         info->ram->normal_bytes = norm_mig_bytes_transferred();

+        info->ram->mbps = s->mbps;

         break;
     case MIG_STATE_ERROR:
         info->has_status = true;
         break;
     case MIG_STATE_ERROR:
         info->has_status = true;


@@ -250,7 +272,7 @@ void qmp_migrate_set_capabilities(MigrationCapabilityStatusList *params,
     MigrationState *s = migrate_get_current();
     MigrationCapabilityStatusList *cap;
 
     MigrationState *s = migrate_get_current();
     MigrationCapabilityStatusList *cap;
 

-    if (s->state == MIG_STATE_ACTIVE) {
+    if (s->state == MIG_STATE_ACTIVE || s->state == MIG_STATE_SETUP) {

         error_set(errp, QERR_MIGRATION_ACTIVE);
         return;
     }
         error_set(errp, QERR_MIGRATION_ACTIVE);
         return;
     }


@@ -262,6 +284,13 @@ void qmp_migrate_set_capabilities(MigrationCapabilityStatusList *params,
 
 /* shared migration helpers */
 
 
 /* shared migration helpers */
 

+static void migrate_set_state(MigrationState *s, int old_state, int new_state)
+{
+    if (atomic_cmpxchg(&s->state, old_state, new_state) == new_state) {
+        trace_migrate_set_state(new_state);
+    }
+}
+

 static void migrate_fd_cleanup(void *opaque)
 {
     MigrationState *s = opaque;
 static void migrate_fd_cleanup(void *opaque)
 {
     MigrationState *s = opaque;


@@ -283,19 +312,14 @@ static void migrate_fd_cleanup(void *opaque)
 
     if (s->state != MIG_STATE_COMPLETED) {
         qemu_savevm_state_cancel();
 
     if (s->state != MIG_STATE_COMPLETED) {
         qemu_savevm_state_cancel();

+        if (s->state == MIG_STATE_CANCELLING) {
+            migrate_set_state(s, MIG_STATE_CANCELLING, MIG_STATE_CANCELLED);
+        }

     }
 
     notifier_list_notify(&migration_state_notifiers, s);
 }
 
     }
 
     notifier_list_notify(&migration_state_notifiers, s);
 }
 

-static void migrate_finish_set_state(MigrationState *s, int new_state)
-{
-    if (__sync_val_compare_and_swap(&s->state, MIG_STATE_ACTIVE,
-                                    new_state) == new_state) {
-        trace_migrate_set_state(new_state);
-    }
-}
-

 void migrate_fd_error(MigrationState *s)
 {
     DPRINTF("setting error state\n");
 void migrate_fd_error(MigrationState *s)
 {
     DPRINTF("setting error state\n");


@@ -307,9 +331,16 @@ void migrate_fd_error(MigrationState *s)
 
 static void migrate_fd_cancel(MigrationState *s)
 {
 
 static void migrate_fd_cancel(MigrationState *s)
 {

+    int old_state ;

     DPRINTF("cancelling migration\n");
 
     DPRINTF("cancelling migration\n");
 

-    migrate_finish_set_state(s, MIG_STATE_CANCELLED);
+    do {
+        old_state = s->state;
+        if (old_state != MIG_STATE_SETUP && old_state != MIG_STATE_ACTIVE) {
+            break;
+        }
+        migrate_set_state(s, old_state, MIG_STATE_CANCELLING);
+    } while (s->state != MIG_STATE_CANCELLING);

 }
 
 void add_migration_state_change_notifier(Notifier *notify)
 }
 
 void add_migration_state_change_notifier(Notifier *notify)


@@ -322,9 +353,9 @@ void remove_migration_state_change_notifier(Notifier *notify)
     notifier_remove(notify);
 }
 
     notifier_remove(notify);
 }
 

-bool migration_is_active(MigrationState *s)
+bool migration_in_setup(MigrationState *s)

 {
 {

-    return s->state == MIG_STATE_ACTIVE;
+    return s->state == MIG_STATE_SETUP;

 }
 
 bool migration_has_finished(MigrationState *s)
 }
 
 bool migration_has_finished(MigrationState *s)


@@ -349,7 +380,6 @@ static MigrationState *migrate_init(const MigrationParams *params)
            sizeof(enabled_capabilities));
 
     memset(s, 0, sizeof(*s));
            sizeof(enabled_capabilities));
 
     memset(s, 0, sizeof(*s));

-    s->bandwidth_limit = bandwidth_limit;

     s->params = *params;
     memcpy(s->enabled_capabilities, enabled_capabilities,
            sizeof(enabled_capabilities));
     s->params = *params;
     memcpy(s->enabled_capabilities, enabled_capabilities,
            sizeof(enabled_capabilities));


@@ -359,7 +389,7 @@ static MigrationState *migrate_init(const MigrationParams *params)
     s->state = MIG_STATE_SETUP;
     trace_migrate_set_state(MIG_STATE_SETUP);
 
     s->state = MIG_STATE_SETUP;
     trace_migrate_set_state(MIG_STATE_SETUP);
 

-    s->total_time = qemu_get_clock_ms(rt_clock);
+    s->total_time = qemu_clock_get_ms(QEMU_CLOCK_REALTIME);

     return s;
 }
 
     return s;
 }
 


@@ -384,10 +414,11 @@ void qmp_migrate(const char *uri, bool has_blk, bool blk,
     MigrationParams params;
     const char *p;
 
     MigrationParams params;
     const char *p;
 

-    params.blk = blk;
-    params.shared = inc;
+    params.blk = has_blk && blk;
+    params.shared = has_inc && inc;

 
 

-    if (s->state == MIG_STATE_ACTIVE) {
+    if (s->state == MIG_STATE_ACTIVE || s->state == MIG_STATE_SETUP ||
+        s->state == MIG_STATE_CANCELLING) {

         error_set(errp, QERR_MIGRATION_ACTIVE);
         return;
     }
         error_set(errp, QERR_MIGRATION_ACTIVE);
         return;
     }


@@ -405,6 +436,10 @@ void qmp_migrate(const char *uri, bool has_blk, bool blk,
 
     if (strstart(uri, "tcp:", &p)) {
         tcp_start_outgoing_migration(s, p, &local_err);
 
     if (strstart(uri, "tcp:", &p)) {
         tcp_start_outgoing_migration(s, p, &local_err);

+#ifdef CONFIG_RDMA
+    } else if (strstart(uri, "x-rdma:", &p)) {
+        rdma_start_outgoing_migration(s, p, &local_err);
+#endif

 #if !defined(WIN32)
     } else if (strstart(uri, "exec:", &p)) {
         exec_start_outgoing_migration(s, p, &local_err);
 #if !defined(WIN32)
     } else if (strstart(uri, "exec:", &p)) {
         exec_start_outgoing_migration(s, p, &local_err);


@@ -415,6 +450,7 @@ void qmp_migrate(const char *uri, bool has_blk, bool blk,
 #endif
     } else {
         error_set(errp, QERR_INVALID_PARAMETER_VALUE, "uri", "a valid migration protocol");
 #endif
     } else {
         error_set(errp, QERR_INVALID_PARAMETER_VALUE, "uri", "a valid migration protocol");

+        s->state = MIG_STATE_ERROR;

         return;
     }
 
         return;
     }
 


@@ -474,6 +510,33 @@ void qmp_migrate_set_downtime(double value, Error **errp)
     max_downtime = (uint64_t)value;
 }
 
     max_downtime = (uint64_t)value;
 }
 

+bool migrate_rdma_pin_all(void)
+{
+    MigrationState *s;
+
+    s = migrate_get_current();
+
+    return s->enabled_capabilities[MIGRATION_CAPABILITY_X_RDMA_PIN_ALL];
+}
+
+bool migrate_auto_converge(void)
+{
+    MigrationState *s;
+
+    s = migrate_get_current();
+
+    return s->enabled_capabilities[MIGRATION_CAPABILITY_AUTO_CONVERGE];
+}
+
+bool migrate_zero_blocks(void)
+{
+    MigrationState *s;
+
+    s = migrate_get_current();
+
+    return s->enabled_capabilities[MIGRATION_CAPABILITY_ZERO_BLOCKS];
+}
+

 int migrate_use_xbzrle(void)
 {
     MigrationState *s;
 int migrate_use_xbzrle(void)
 {
     MigrationState *s;


@@ -497,7 +560,8 @@ int64_t migrate_xbzrle_cache_size(void)
 static void *migration_thread(void *opaque)
 {
     MigrationState *s = opaque;
 static void *migration_thread(void *opaque)
 {
     MigrationState *s = opaque;

-    int64_t initial_time = qemu_get_clock_ms(rt_clock);
+    int64_t initial_time = qemu_clock_get_ms(QEMU_CLOCK_REALTIME);
+    int64_t setup_start = qemu_clock_get_ms(QEMU_CLOCK_HOST);

     int64_t initial_bytes = 0;
     int64_t max_size = 0;
     int64_t start_time = initial_time;
     int64_t initial_bytes = 0;
     int64_t max_size = 0;
     int64_t start_time = initial_time;


@@ -506,6 +570,11 @@ static void *migration_thread(void *opaque)
     DPRINTF("beginning savevm\n");
     qemu_savevm_state_begin(s->file, &s->params);
 
     DPRINTF("beginning savevm\n");
     qemu_savevm_state_begin(s->file, &s->params);
 

+    s->setup_time = qemu_clock_get_ms(QEMU_CLOCK_HOST) - setup_start;
+    migrate_set_state(s, MIG_STATE_SETUP, MIG_STATE_ACTIVE);
+
+    DPRINTF("setup complete\n");
+

     while (s->state == MIG_STATE_ACTIVE) {
         int64_t current_time;
         uint64_t pending_size;
     while (s->state == MIG_STATE_ACTIVE) {
         int64_t current_time;
         uint64_t pending_size;


@@ -513,37 +582,52 @@ static void *migration_thread(void *opaque)
         if (!qemu_file_rate_limit(s->file)) {
             DPRINTF("iterate\n");
             pending_size = qemu_savevm_state_pending(s->file, max_size);
         if (!qemu_file_rate_limit(s->file)) {
             DPRINTF("iterate\n");
             pending_size = qemu_savevm_state_pending(s->file, max_size);

-            DPRINTF("pending size %lu max %lu\n", pending_size, max_size);
+            DPRINTF("pending size %" PRIu64 " max %" PRIu64 "\n",
+                    pending_size, max_size);

             if (pending_size && pending_size >= max_size) {
                 qemu_savevm_state_iterate(s->file);
             } else {
             if (pending_size && pending_size >= max_size) {
                 qemu_savevm_state_iterate(s->file);
             } else {

+                int ret;
+

                 DPRINTF("done iterating\n");
                 qemu_mutex_lock_iothread();
                 DPRINTF("done iterating\n");
                 qemu_mutex_lock_iothread();

-                start_time = qemu_get_clock_ms(rt_clock);
+                start_time = qemu_clock_get_ms(QEMU_CLOCK_REALTIME);

                 qemu_system_wakeup_request(QEMU_WAKEUP_REASON_OTHER);
                 old_vm_running = runstate_is_running();
                 qemu_system_wakeup_request(QEMU_WAKEUP_REASON_OTHER);
                 old_vm_running = runstate_is_running();

-                vm_stop_force_state(RUN_STATE_FINISH_MIGRATE);
-                qemu_file_set_rate_limit(s->file, INT_MAX);
-                qemu_savevm_state_complete(s->file);
+
+                ret = vm_stop_force_state(RUN_STATE_FINISH_MIGRATE);
+                if (ret >= 0) {
+                    qemu_file_set_rate_limit(s->file, INT64_MAX);
+                    qemu_savevm_state_complete(s->file);
+                }

                 qemu_mutex_unlock_iothread();
                 qemu_mutex_unlock_iothread();

+
+                if (ret < 0) {
+                    migrate_set_state(s, MIG_STATE_ACTIVE, MIG_STATE_ERROR);
+                    break;
+                }
+

                 if (!qemu_file_get_error(s->file)) {
                 if (!qemu_file_get_error(s->file)) {

-                    migrate_finish_set_state(s, MIG_STATE_COMPLETED);
+                    migrate_set_state(s, MIG_STATE_ACTIVE, MIG_STATE_COMPLETED);

                     break;
                 }
             }
         }
 
         if (qemu_file_get_error(s->file)) {
                     break;
                 }
             }
         }
 
         if (qemu_file_get_error(s->file)) {

-            migrate_finish_set_state(s, MIG_STATE_ERROR);
+            migrate_set_state(s, MIG_STATE_ACTIVE, MIG_STATE_ERROR);

             break;
         }
             break;
         }

-        current_time = qemu_get_clock_ms(rt_clock);
+        current_time = qemu_clock_get_ms(QEMU_CLOCK_REALTIME);

         if (current_time >= initial_time + BUFFER_DELAY) {
             uint64_t transferred_bytes = qemu_ftell(s->file) - initial_bytes;
             uint64_t time_spent = current_time - initial_time;
             double bandwidth = transferred_bytes / time_spent;
             max_size = bandwidth * migrate_max_downtime() / 1000000;
 
         if (current_time >= initial_time + BUFFER_DELAY) {
             uint64_t transferred_bytes = qemu_ftell(s->file) - initial_bytes;
             uint64_t time_spent = current_time - initial_time;
             double bandwidth = transferred_bytes / time_spent;
             max_size = bandwidth * migrate_max_downtime() / 1000000;
 

+            s->mbps = time_spent ? (((double) transferred_bytes * 8.0) /
+                    ((double) time_spent / 1000.0)) / 1000.0 / 1000.0 : -1;
+

             DPRINTF("transferred %" PRIu64 " time_spent %" PRIu64
                     " bandwidth %g max_size %" PRId64 "\n",
                     transferred_bytes, time_spent, bandwidth, max_size);
             DPRINTF("transferred %" PRIu64 " time_spent %" PRIu64
                     " bandwidth %g max_size %" PRId64 "\n",
                     transferred_bytes, time_spent, bandwidth, max_size);


@@ -565,7 +649,7 @@ static void *migration_thread(void *opaque)
 
     qemu_mutex_lock_iothread();
     if (s->state == MIG_STATE_COMPLETED) {
 
     qemu_mutex_lock_iothread();
     if (s->state == MIG_STATE_COMPLETED) {

-        int64_t end_time = qemu_get_clock_ms(rt_clock);
+        int64_t end_time = qemu_clock_get_ms(QEMU_CLOCK_REALTIME);

         s->total_time = end_time - s->total_time;
         s->downtime = end_time - start_time;
         runstate_set(RUN_STATE_POSTMIGRATE);
         s->total_time = end_time - s->total_time;
         s->downtime = end_time - start_time;
         runstate_set(RUN_STATE_POSTMIGRATE);


@@ -582,8 +666,8 @@ static void *migration_thread(void *opaque)
 
 void migrate_fd_connect(MigrationState *s)
 {
 
 void migrate_fd_connect(MigrationState *s)
 {

-    s->state = MIG_STATE_ACTIVE;
-    trace_migrate_set_state(MIG_STATE_ACTIVE);
+    s->state = MIG_STATE_SETUP;
+    trace_migrate_set_state(MIG_STATE_SETUP);

 
     /* This is a best 1st approximation. ns to ms */
     s->expected_downtime = max_downtime/1000000;
 
     /* This is a best 1st approximation. ns to ms */
     s->expected_downtime = max_downtime/1000000;


@@ -592,7 +676,9 @@ void migrate_fd_connect(MigrationState *s)
     qemu_file_set_rate_limit(s->file,
                              s->bandwidth_limit / XFER_LIMIT_RATIO);
 
     qemu_file_set_rate_limit(s->file,
                              s->bandwidth_limit / XFER_LIMIT_RATIO);
 

+    /* Notify before starting migration thread */
+    notifier_list_notify(&migration_state_notifiers, s);
+

     qemu_thread_create(&s->thread, migration_thread, s,
                        QEMU_THREAD_JOINABLE);
     qemu_thread_create(&s->thread, migration_thread, s,
                        QEMU_THREAD_JOINABLE);

-    notifier_list_notify(&migration_state_notifiers, s);

 }
 }