[qemu-xen staging] migration/savevm: don't worry if bitmap migration postcopy failed

[patchbot@xxxxxxx]

commit ee64722514fabcad2430982ade86180208f5be4f
Author:     Vladimir Sementsov-Ogievskiy <vsementsov@xxxxxxxxxxxxx>
AuthorDate: Mon Jul 27 22:42:32 2020 +0300
Commit:     Eric Blake <eblake@xxxxxxxxxx>
CommitDate: Mon Jul 27 15:41:34 2020 -0500

    migration/savevm: don't worry if bitmap migration postcopy failed
    
    First, if only bitmaps postcopy is enabled (and not ram postcopy)
    postcopy_pause_incoming crashes on an assertion
    assert(mis->to_src_file).
    
    And anyway, bitmaps postcopy is not prepared to be somehow recovered.
    The original idea instead is that if bitmaps postcopy failed, we just
    lose some bitmaps, which is not critical. So, on failure we just need
    to remove unfinished bitmaps and guest should continue execution on
    destination.
    
    Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@xxxxxxxxxxxxx>
    Reviewed-by: Dr. David Alan Gilbert <dgilbert@xxxxxxxxxx>
    Reviewed-by: Andrey Shinkevich <andrey.shinkevich@xxxxxxxxxxxxx>
    Reviewed-by: Eric Blake <eblake@xxxxxxxxxx>
    Message-Id: <20200727194236.19551-18-vsementsov@xxxxxxxxxxxxx>
    Signed-off-by: Eric Blake <eblake@xxxxxxxxxx>
---
 migration/savevm.c | 37 ++++++++++++++++++++++++++++++++-----
 1 file changed, 32 insertions(+), 5 deletions(-)

diff --git a/migration/savevm.c b/migration/savevm.c
index 45c9dd9d8a..a843d202b5 100644
--- a/migration/savevm.c
+++ b/migration/savevm.c
@@ -1813,6 +1813,9 @@ static void *postcopy_ram_listen_thread(void *opaque)
     MigrationIncomingState *mis = migration_incoming_get_current();
     QEMUFile *f = mis->from_src_file;
     int load_res;
+    MigrationState *migr = migrate_get_current();
+
+    object_ref(OBJECT(migr));
 
     migrate_set_state(&mis->state, MIGRATION_STATUS_ACTIVE,
                                    MIGRATION_STATUS_POSTCOPY_ACTIVE);
@@ -1839,11 +1842,24 @@ static void *postcopy_ram_listen_thread(void *opaque)
 
     trace_postcopy_ram_listen_thread_exit();
     if (load_res < 0) {
-        error_report("%s: loadvm failed: %d", __func__, load_res);
         qemu_file_set_error(f, load_res);
-        migrate_set_state(&mis->state, MIGRATION_STATUS_POSTCOPY_ACTIVE,
-                                       MIGRATION_STATUS_FAILED);
-    } else {
+        dirty_bitmap_mig_cancel_incoming();
+        if (postcopy_state_get() == POSTCOPY_INCOMING_RUNNING &&
+            !migrate_postcopy_ram() && migrate_dirty_bitmaps())
+        {
+            error_report("%s: loadvm failed during postcopy: %d. All states "
+                         "are migrated except dirty bitmaps. Some dirty "
+                         "bitmaps may be lost, and present migrated dirty "
+                         "bitmaps are correctly migrated and valid.",
+                         __func__, load_res);
+            load_res = 0; /* prevent further exit() */
+        } else {
+            error_report("%s: loadvm failed: %d", __func__, load_res);
+            migrate_set_state(&mis->state, MIGRATION_STATUS_POSTCOPY_ACTIVE,
+                                           MIGRATION_STATUS_FAILED);
+        }
+    }
+    if (load_res >= 0) {
         /*
          * This looks good, but it's possible that the device loading in the
          * main thread hasn't finished yet, and so we might not be in 'RUN'
@@ -1879,6 +1895,8 @@ static void *postcopy_ram_listen_thread(void *opaque)
     mis->have_listen_thread = false;
     postcopy_state_set(POSTCOPY_INCOMING_END);
 
+    object_unref(OBJECT(migr));
+
     return NULL;
 }
 
@@ -2437,6 +2455,8 @@ static bool 
postcopy_pause_incoming(MigrationIncomingState *mis)
 {
     trace_postcopy_pause_incoming();
 
+    assert(migrate_postcopy_ram());
+
     /* Clear the triggered bit to allow one recovery */
     mis->postcopy_recover_triggered = false;
 
@@ -2521,15 +2541,22 @@ out:
     if (ret < 0) {
         qemu_file_set_error(f, ret);
 
+        /* Cancel bitmaps incoming regardless of recovery */
+        dirty_bitmap_mig_cancel_incoming();
+
         /*
          * If we are during an active postcopy, then we pause instead
          * of bail out to at least keep the VM's dirty data.  Note
          * that POSTCOPY_INCOMING_LISTENING stage is still not enough,
          * during which we're still receiving device states and we
          * still haven't yet started the VM on destination.
+         *
+         * Only RAM postcopy supports recovery. Still, if RAM postcopy is
+         * enabled, canceled bitmaps postcopy will not affect RAM postcopy
+         * recovering.
          */
         if (postcopy_state_get() == POSTCOPY_INCOMING_RUNNING &&
-            postcopy_pause_incoming(mis)) {
+            migrate_postcopy_ram() && postcopy_pause_incoming(mis)) {
             /* Reset f to point to the newly created channel */
             f = mis->from_src_file;
             goto retry;
--
generated by git-patchbot for /home/xen/git/qemu-xen.git#staging