qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Peter Xu <peterx@redhat.com>
To: qemu-devel@nongnu.org
Cc: Fabiano Rosas <farosas@suse.de>,
	David Hildenbrand <david@redhat.com>,
	peterx@redhat.com, Paolo Bonzini <pbonzini@redhat.com>,
	Juraj Marcin <jmarcin@redhat.com>
Subject: [PULL 31/36] migration: Move postcopy_ram_listen_thread() to postcopy-ram.c
Date: Mon,  3 Nov 2025 16:06:20 -0500	[thread overview]
Message-ID: <20251103210625.3689448-32-peterx@redhat.com> (raw)
In-Reply-To: <20251103210625.3689448-1-peterx@redhat.com>

From: Juraj Marcin <jmarcin@redhat.com>

This patch addresses a TODO about moving postcopy_ram_listen_thread() to
postcopy file.

Signed-off-by: Juraj Marcin <jmarcin@redhat.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Link: https://lore.kernel.org/r/20251103183301.3840862-4-jmarcin@redhat.com
Signed-off-by: Peter Xu <peterx@redhat.com>
---
 migration/postcopy-ram.h |   2 +
 migration/postcopy-ram.c | 105 ++++++++++++++++++++++++++++++++++++++
 migration/savevm.c       | 107 ---------------------------------------
 3 files changed, 107 insertions(+), 107 deletions(-)

diff --git a/migration/postcopy-ram.h b/migration/postcopy-ram.h
index ca19433b24..3e26db3e6b 100644
--- a/migration/postcopy-ram.h
+++ b/migration/postcopy-ram.h
@@ -199,4 +199,6 @@ bool postcopy_is_paused(MigrationStatus status);
 void mark_postcopy_blocktime_begin(uintptr_t addr, uint32_t ptid,
                                    RAMBlock *rb);
 
+void *postcopy_ram_listen_thread(void *opaque);
+
 #endif
diff --git a/migration/postcopy-ram.c b/migration/postcopy-ram.c
index 5471efb4f0..880b11f154 100644
--- a/migration/postcopy-ram.c
+++ b/migration/postcopy-ram.c
@@ -2077,3 +2077,108 @@ bool postcopy_is_paused(MigrationStatus status)
     return status == MIGRATION_STATUS_POSTCOPY_PAUSED ||
         status == MIGRATION_STATUS_POSTCOPY_RECOVER_SETUP;
 }
+
+/*
+ * Triggered by a postcopy_listen command; this thread takes over reading
+ * the input stream, leaving the main thread free to carry on loading the rest
+ * of the device state (from RAM).
+ */
+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();
+    Error *local_err = NULL;
+
+    object_ref(OBJECT(migr));
+
+    migrate_set_state(&mis->state, MIGRATION_STATUS_ACTIVE,
+                                   MIGRATION_STATUS_POSTCOPY_ACTIVE);
+    qemu_event_set(&mis->thread_sync_event);
+    trace_postcopy_ram_listen_thread_start();
+
+    rcu_register_thread();
+    /*
+     * Because we're a thread and not a coroutine we can't yield
+     * in qemu_file, and thus we must be blocking now.
+     */
+    qemu_file_set_blocking(f, true, &error_fatal);
+
+    /* TODO: sanity check that only postcopiable data will be loaded here */
+    load_res = qemu_loadvm_state_main(f, mis, &local_err);
+
+    /*
+     * This is tricky, but, mis->from_src_file can change after it
+     * returns, when postcopy recovery happened. In the future, we may
+     * want a wrapper for the QEMUFile handle.
+     */
+    f = mis->from_src_file;
+
+    /* And non-blocking again so we don't block in any cleanup */
+    qemu_file_set_blocking(f, false, &error_fatal);
+
+    trace_postcopy_ram_listen_thread_exit();
+    if (load_res < 0) {
+        qemu_file_set_error(f, load_res);
+        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: %s. 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, error_get_pretty(local_err));
+            g_clear_pointer(&local_err, error_free);
+            load_res = 0; /* prevent further exit() */
+        } else {
+            error_prepend(&local_err,
+                          "loadvm failed during postcopy: %d: ", load_res);
+            migrate_set_error(migr, local_err);
+            g_clear_pointer(&local_err, error_report_err);
+            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'
+         * state yet; wait for the end of the main thread.
+         */
+        qemu_event_wait(&mis->main_thread_load_event);
+    }
+    postcopy_ram_incoming_cleanup(mis);
+
+    if (load_res < 0) {
+        /*
+         * If something went wrong then we have a bad state so exit;
+         * depending how far we got it might be possible at this point
+         * to leave the guest running and fire MCEs for pages that never
+         * arrived as a desperate recovery step.
+         */
+        rcu_unregister_thread();
+        exit(EXIT_FAILURE);
+    }
+
+    migrate_set_state(&mis->state, MIGRATION_STATUS_POSTCOPY_ACTIVE,
+                                   MIGRATION_STATUS_COMPLETED);
+    /*
+     * If everything has worked fine, then the main thread has waited
+     * for us to start, and we're the last use of the mis.
+     * (If something broke then qemu will have to exit anyway since it's
+     * got a bad migration state).
+     */
+    bql_lock();
+    migration_incoming_state_destroy();
+    bql_unlock();
+
+    rcu_unregister_thread();
+    mis->have_listen_thread = false;
+    postcopy_state_set(POSTCOPY_INCOMING_END);
+
+    object_unref(OBJECT(migr));
+
+    return NULL;
+}
diff --git a/migration/savevm.c b/migration/savevm.c
index fa017378db..2f7ed0db64 100644
--- a/migration/savevm.c
+++ b/migration/savevm.c
@@ -2088,113 +2088,6 @@ static int loadvm_postcopy_ram_handle_discard(MigrationIncomingState *mis,
     return 0;
 }
 
-/*
- * Triggered by a postcopy_listen command; this thread takes over reading
- * the input stream, leaving the main thread free to carry on loading the rest
- * of the device state (from RAM).
- * (TODO:This could do with being in a postcopy file - but there again it's
- * just another input loop, not that postcopy specific)
- */
-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();
-    Error *local_err = NULL;
-
-    object_ref(OBJECT(migr));
-
-    migrate_set_state(&mis->state, MIGRATION_STATUS_ACTIVE,
-                                   MIGRATION_STATUS_POSTCOPY_ACTIVE);
-    qemu_event_set(&mis->thread_sync_event);
-    trace_postcopy_ram_listen_thread_start();
-
-    rcu_register_thread();
-    /*
-     * Because we're a thread and not a coroutine we can't yield
-     * in qemu_file, and thus we must be blocking now.
-     */
-    qemu_file_set_blocking(f, true, &error_fatal);
-
-    /* TODO: sanity check that only postcopiable data will be loaded here */
-    load_res = qemu_loadvm_state_main(f, mis, &local_err);
-
-    /*
-     * This is tricky, but, mis->from_src_file can change after it
-     * returns, when postcopy recovery happened. In the future, we may
-     * want a wrapper for the QEMUFile handle.
-     */
-    f = mis->from_src_file;
-
-    /* And non-blocking again so we don't block in any cleanup */
-    qemu_file_set_blocking(f, false, &error_fatal);
-
-    trace_postcopy_ram_listen_thread_exit();
-    if (load_res < 0) {
-        qemu_file_set_error(f, load_res);
-        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: %s. 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, error_get_pretty(local_err));
-            g_clear_pointer(&local_err, error_free);
-            load_res = 0; /* prevent further exit() */
-        } else {
-            error_prepend(&local_err,
-                          "loadvm failed during postcopy: %d: ", load_res);
-            migrate_set_error(migr, local_err);
-            g_clear_pointer(&local_err, error_report_err);
-            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'
-         * state yet; wait for the end of the main thread.
-         */
-        qemu_event_wait(&mis->main_thread_load_event);
-    }
-    postcopy_ram_incoming_cleanup(mis);
-
-    if (load_res < 0) {
-        /*
-         * If something went wrong then we have a bad state so exit;
-         * depending how far we got it might be possible at this point
-         * to leave the guest running and fire MCEs for pages that never
-         * arrived as a desperate recovery step.
-         */
-        rcu_unregister_thread();
-        exit(EXIT_FAILURE);
-    }
-
-    migrate_set_state(&mis->state, MIGRATION_STATUS_POSTCOPY_ACTIVE,
-                                   MIGRATION_STATUS_COMPLETED);
-    /*
-     * If everything has worked fine, then the main thread has waited
-     * for us to start, and we're the last use of the mis.
-     * (If something broke then qemu will have to exit anyway since it's
-     * got a bad migration state).
-     */
-    bql_lock();
-    migration_incoming_state_destroy();
-    bql_unlock();
-
-    rcu_unregister_thread();
-    mis->have_listen_thread = false;
-    postcopy_state_set(POSTCOPY_INCOMING_END);
-
-    object_unref(OBJECT(migr));
-
-    return NULL;
-}
-
 /* After this message we must be able to immediately receive postcopy data */
 static int loadvm_postcopy_handle_listen(MigrationIncomingState *mis,
                                          Error **errp)
-- 
2.50.1



  parent reply	other threads:[~2025-11-03 21:09 UTC|newest]

Thread overview: 39+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-11-03 21:05 [PULL 00/36] Staging patches Peter Xu
2025-11-03 21:05 ` [PULL 01/36] migration/savevm: Add a compatibility check for capabilities Peter Xu
2025-11-03 21:05 ` [PULL 02/36] MAINTAINERS: update cpr reviewers Peter Xu
2025-11-03 21:05 ` [PULL 03/36] migration/ram: fix docs of ram_handle_zero Peter Xu
2025-11-03 21:05 ` [PULL 04/36] migration: add FEATURE_SEEKABLE to QIOChannelBlock Peter Xu
2025-11-03 21:05 ` [PULL 05/36] migration: mapped-ram: handle zero pages Peter Xu
2025-11-03 21:05 ` [PULL 06/36] migration: Remove unused VMSTATE_UINTTL_EQUAL[_V]() macros Peter Xu
2025-11-03 21:05 ` [PULL 07/36] migration: Fix error leak in postcopy_ram_listen_thread() Peter Xu
2025-11-03 21:05 ` [PULL 08/36] migration/cpr: Fix coverity report in cpr_exec_persist_state() Peter Xu
2025-11-03 21:05 ` [PULL 09/36] migration/cpr: Fix UAF in cpr_exec_cb() when execvp() fails Peter Xu
2025-11-03 21:05 ` [PULL 10/36] migration/cpr: Avoid crashing QEMU when cpr-exec runs with no args Peter Xu
2025-11-03 21:06 ` [PULL 11/36] ram-block-attributes: fix interaction with hugetlb memory backends Peter Xu
2025-11-03 21:06 ` [PULL 12/36] ram-block-attributes: Unify the retrieval of the block size Peter Xu
2025-11-03 21:06 ` [PULL 13/36] migration/qmp: Update "resume" flag doc in "migrate" command Peter Xu
2025-11-05 12:27   ` Richard Henderson
2025-11-03 21:06 ` [PULL 14/36] migration/cpr: Document obscure usage of g_autofree when parse str Peter Xu
2025-11-03 21:06 ` [PULL 15/36] hostmem/shm: Allow shm memory backend serve as shared memory for coco-VMs Peter Xu
2025-11-03 21:06 ` [PULL 16/36] migration: Fix regression of passing error_fatal into vmstate_load_state() Peter Xu
2025-11-03 21:06 ` [PULL 17/36] migration: Don't free the reason after calling migrate_add_blocker Peter Xu
2025-11-03 21:06 ` [PULL 18/36] migration: Use unsigned instead of int for bit set of MigMode Peter Xu
2025-11-03 21:06 ` [PULL 19/36] migration: Use bitset of MigMode instead of variable arguments Peter Xu
2025-11-03 21:06 ` [PULL 20/36] migration: Put Error **errp parameter last Peter Xu
2025-11-03 21:06 ` [PULL 21/36] io: Add qio_channel_wait_cond() helper Peter Xu
2025-11-03 21:06 ` [PULL 22/36] migration: Properly wait on G_IO_IN when peeking messages Peter Xu
2025-11-03 21:06 ` [PULL 23/36] migration: vmstate_save_state_v(): fix error path Peter Xu
2025-11-03 21:06 ` [PULL 24/36] tmp_emulator: improve and fix use of errp Peter Xu
2025-11-03 21:06 ` [PULL 25/36] migration/vmstate: stop reporting error number for new _errp APIs Peter Xu
2025-11-03 21:06 ` [PULL 26/36] migration: vmsd errp handlers: return bool Peter Xu
2025-11-03 21:06 ` [PULL 27/36] scripts/vmstate-static-checker: Fix deprecation warnings with latest argparse Peter Xu
2025-11-03 21:06 ` [PULL 28/36] system/physmem: mark io_mem_unassigned lockless Peter Xu
2025-11-03 21:06 ` [PULL 29/36] migration: Flush migration channel after sending data of CMD_PACKAGED Peter Xu
2025-11-03 21:06 ` [PULL 30/36] migration: Do not try to start VM if disk activation fails Peter Xu
2025-11-03 21:06 ` Peter Xu [this message]
2025-11-03 21:06 ` [PULL 32/36] migration: Introduce postcopy incoming setup and cleanup functions Peter Xu
2025-11-03 21:06 ` [PULL 33/36] migration: Refactor all incoming cleanup info migration_incoming_destroy() Peter Xu
2025-11-03 21:06 ` [PULL 34/36] migration: Respect exit-on-error when migration fails before resuming Peter Xu
2025-11-03 21:06 ` [PULL 35/36] migration: Make postcopy listen thread joinable Peter Xu
2025-11-03 21:06 ` [PULL 36/36] migration: Introduce POSTCOPY_DEVICE state Peter Xu
2025-11-05  7:52 ` [PULL 00/36] Staging patches Richard Henderson

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20251103210625.3689448-32-peterx@redhat.com \
    --to=peterx@redhat.com \
    --cc=david@redhat.com \
    --cc=farosas@suse.de \
    --cc=jmarcin@redhat.com \
    --cc=pbonzini@redhat.com \
    --cc=qemu-devel@nongnu.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).