Merge tag 'migration-staging-pull-request' of https://gitlab.com/peterx/qemu into staging

Migration pull

- William's fix on hwpoison migration which used to crash QEMU
- Peter's multifd cleanup + bugfix + optimizations
- Avihai's fix on multifd crash over non-socket channels
- Fabiano's multifd thread-race fix
- Peter's CI fix series

# -----BEGIN PGP SIGNATURE-----
#
# iIgEABYKADAWIQS5GE3CDMRX2s990ak7X8zN86vXBgUCZcREtRIccGV0ZXJ4QHJl
# ZGhhdC5jb20ACgkQO1/MzfOr1wacrwEAl2aeQkh51h/e+OKX7MG4/4Y6Edf6Oz7o
# IJLk/cyrUFQA/2exo2lOdv5zHNOJKwAYj8HYDraezrC/MK1eED4Wji0M
# =k53l
# -----END PGP SIGNATURE-----
# gpg: Signature made Thu 08 Feb 2024 03:04:21 GMT
# gpg:                using EDDSA key B9184DC20CC457DACF7DD1A93B5FCCCDF3ABD706
# gpg:                issuer "peterx@redhat.com"
# gpg: Good signature from "Peter Xu <xzpeter@gmail.com>" [marginal]
# gpg:                 aka "Peter Xu <peterx@redhat.com>" [marginal]
# gpg: WARNING: This key is not certified with sufficiently trusted signatures!
# gpg:          It is not certain that the signature belongs to the owner.
# Primary key fingerprint: B918 4DC2 0CC4 57DA CF7D  D1A9 3B5F CCCD F3AB D706

* tag 'migration-staging-pull-request' of https://gitlab.com/peterx/qemu: (34 commits)
  ci: Update comment for migration-compat-aarch64
  ci: Remove tag dependency for build-previous-qemu
  tests/migration-test: Stick with gicv3 in aarch64 test
  migration/multifd: Add a synchronization point for channel creation
  migration/multifd: Unify multifd and TLS connection paths
  migration/multifd: Move multifd_send_setup into migration thread
  migration/multifd: Move multifd_send_setup error handling in to the function
  migration/multifd: Remove p->running
  migration/multifd: Join the TLS thread
  migration: Fix logic of channels and transport compatibility check
  migration/multifd: Optimize sender side to be lockless
  migration/multifd: Fix MultiFDSendParams.packet_num race
  migration/multifd: Stick with send/recv on function names
  migration/multifd: Cleanup multifd_load_cleanup()
  migration/multifd: Cleanup multifd_save_cleanup()
  migration/multifd: Rewrite multifd_queue_page()
  migration/multifd: Change retval of multifd_send_pages()
  migration/multifd: Change retval of multifd_queue_page()
  migration/multifd: Split multifd_send_terminate_threads()
  migration/multifd: Forbid spurious wakeups
  ...

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>

1 files changed, 28 insertions, 20 deletions

diff --git a/migration/migration.c b/migration/migration.c
index d5f705ceef..ab21de2cad 100644
--- a/migration/migration.c
+++ b/migration/migration.c
@@ -67,6 +67,7 @@
 #include "options.h"
 #include "sysemu/dirtylimit.h"
 #include "qemu/sockets.h"
+#include "sysemu/kvm.h"
 
 static NotifierList migration_state_notifiers =
     NOTIFIER_LIST_INITIALIZER(migration_state_notifiers);
@@ -128,11 +129,17 @@ static bool migration_needs_multiple_sockets(void)
     return migrate_multifd() || migrate_postcopy_preempt();
 }
 
-static bool transport_supports_multi_channels(SocketAddress *saddr)
+static bool transport_supports_multi_channels(MigrationAddress *addr)
 {
-    return saddr->type == SOCKET_ADDRESS_TYPE_INET ||
-           saddr->type == SOCKET_ADDRESS_TYPE_UNIX ||
-           saddr->type == SOCKET_ADDRESS_TYPE_VSOCK;
+    if (addr->transport == MIGRATION_ADDRESS_TYPE_SOCKET) {
+        SocketAddress *saddr = &addr->u.socket;
+
+        return saddr->type == SOCKET_ADDRESS_TYPE_INET ||
+               saddr->type == SOCKET_ADDRESS_TYPE_UNIX ||
+               saddr->type == SOCKET_ADDRESS_TYPE_VSOCK;
+    }
+
+    return false;
 }
 
 static bool
@@ -140,8 +147,7 @@ migration_channels_and_transport_compatible(MigrationAddress *addr,
                                             Error **errp)
 {
     if (migration_needs_multiple_sockets() &&
-        (addr->transport == MIGRATION_ADDRESS_TYPE_SOCKET) &&
-        !transport_supports_multi_channels(&addr->u.socket)) {
+        !transport_supports_multi_channels(addr)) {
         error_setg(errp, "Migration requires multi-channel URIs (e.g. tcp)");
         return false;
     }
@@ -311,7 +317,7 @@ void migration_incoming_state_destroy(void)
 {
     struct MigrationIncomingState *mis = migration_incoming_get_current();
 
-    multifd_load_cleanup();
+    multifd_recv_cleanup();
     compress_threads_load_cleanup();
 
     if (mis->to_src_file) {
@@ -662,7 +668,7 @@ static void process_incoming_migration_bh(void *opaque)
 
     trace_vmstate_downtime_checkpoint("dst-precopy-bh-announced");
 
-    multifd_load_shutdown();
+    multifd_recv_shutdown();
 
     dirty_bitmap_mig_before_vm_start();
 
@@ -759,7 +765,7 @@ fail:
                       MIGRATION_STATUS_FAILED);
     qemu_fclose(mis->from_src_file);
 
-    multifd_load_cleanup();
+    multifd_recv_cleanup();
     compress_threads_load_cleanup();
 
     exit(EXIT_FAILURE);
@@ -885,7 +891,7 @@ void migration_ioc_process_incoming(QIOChannel *ioc, Error **errp)
         default_channel = !mis->from_src_file;
     }
 
-    if (multifd_load_setup(errp) != 0) {
+    if (multifd_recv_setup(errp) != 0) {
         return;
     }
 
@@ -1331,7 +1337,7 @@ static void migrate_fd_cleanup(MigrationState *s)
         }
         bql_lock();
 
-        multifd_save_cleanup();
+        multifd_send_shutdown();
         qemu_mutex_lock(&s->qemu_file_lock);
         tmp = s->to_dst_file;
         s->to_dst_file = NULL;
@@ -1906,6 +1912,12 @@ static bool migrate_prepare(MigrationState *s, bool blk, bool blk_inc,
         return false;
     }
 
+    if (kvm_hwpoisoned_mem()) {
+        error_setg(errp, "Can't migrate this vm with hardware poisoned memory, "
+                   "please reboot the vm and try again");
+        return false;
+    }
+
     if (migration_is_blocked(errp)) {
         return false;
     }
@@ -3315,6 +3327,10 @@ static void *migration_thread(void *opaque)
     object_ref(OBJECT(s));
     update_iteration_initial_status(s);
 
+    if (!multifd_send_setup()) {
+        goto out;
+    }
+
     bql_lock();
     qemu_savevm_state_header(s->to_dst_file);
     bql_unlock();
@@ -3386,6 +3402,7 @@ static void *migration_thread(void *opaque)
         urgent = migration_rate_limit();
     }
 
+out:
     trace_migration_thread_after_loop();
     migration_iteration_finish(s);
     object_unref(OBJECT(s));
@@ -3623,15 +3640,6 @@ void migrate_fd_connect(MigrationState *s, Error *error_in)
         return;
     }
 
-    if (multifd_save_setup(&local_err) != 0) {
-        migrate_set_error(s, local_err);
-        error_report_err(local_err);
-        migrate_set_state(&s->state, MIGRATION_STATUS_SETUP,
-                          MIGRATION_STATUS_FAILED);
-        migrate_fd_cleanup(s);
-        return;
-    }
-
     if (migrate_background_snapshot()) {
         qemu_thread_create(&s->thread, "bg_snapshot",
                 bg_migration_thread, s, QEMU_THREAD_JOINABLE);