migration: do not wait for free thread

Instead of putting the main thread to sleep state to wait for
free compression thread, we can directly post it out as normal
page that reduces the latency and uses CPUs more efficiently

A parameter, compress-wait-thread, is introduced, it can be
enabled if the user really wants the old behavior

Reviewed-by: Peter Xu <peterx@redhat.com>
Signed-off-by: Xiao Guangrong <xiaoguangrong@tencent.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>

3 files changed, 48 insertions, 19 deletions

diff --git a/migration/migration.c b/migration/migration.c
index 4d76bee..4b316ec 100644
--- a/migration/migration.c
+++ b/migration/migration.c
@@ -673,6 +673,8 @@ MigrationParameters *qmp_query_migrate_parameters(Error **errp)
     params->compress_level = s->parameters.compress_level;
     params->has_compress_threads = true;
     params->compress_threads = s->parameters.compress_threads;
+    params->has_compress_wait_thread = true;
+    params->compress_wait_thread = s->parameters.compress_wait_thread;
     params->has_decompress_threads = true;
     params->decompress_threads = s->parameters.decompress_threads;
     params->has_cpu_throttle_initial = true;
@@ -1074,6 +1076,10 @@ static void migrate_params_test_apply(MigrateSetParameters *params,
         dest->compress_threads = params->compress_threads;
     }
 
+    if (params->has_compress_wait_thread) {
+        dest->compress_wait_thread = params->compress_wait_thread;
+    }
+
     if (params->has_decompress_threads) {
         dest->decompress_threads = params->decompress_threads;
     }
@@ -1142,6 +1148,10 @@ static void migrate_params_apply(MigrateSetParameters *params, Error **errp)
         s->parameters.compress_threads = params->compress_threads;
     }
 
+    if (params->has_compress_wait_thread) {
+        s->parameters.compress_wait_thread = params->compress_wait_thread;
+    }
+
     if (params->has_decompress_threads) {
         s->parameters.decompress_threads = params->decompress_threads;
     }
@@ -1890,6 +1900,15 @@ int migrate_compress_threads(void)
     return s->parameters.compress_threads;
 }
 
+int migrate_compress_wait_thread(void)
+{
+    MigrationState *s;
+
+    s = migrate_get_current();
+
+    return s->parameters.compress_wait_thread;
+}
+
 int migrate_decompress_threads(void)
 {
     MigrationState *s;
@@ -3151,6 +3170,8 @@ static Property migration_properties[] = {
     DEFINE_PROP_UINT8("x-compress-threads", MigrationState,
                       parameters.compress_threads,
                       DEFAULT_MIGRATE_COMPRESS_THREAD_COUNT),
+    DEFINE_PROP_BOOL("x-compress-wait-thread", MigrationState,
+                      parameters.compress_wait_thread, true),
     DEFINE_PROP_UINT8("x-decompress-threads", MigrationState,
                       parameters.decompress_threads,
                       DEFAULT_MIGRATE_DECOMPRESS_THREAD_COUNT),
diff --git a/migration/migration.h b/migration/migration.h
index eb0c04a..f7813f8 100644
--- a/migration/migration.h
+++ b/migration/migration.h
@@ -272,6 +272,7 @@ bool migrate_use_return_path(void);
 bool migrate_use_compression(void);
 int migrate_compress_level(void);
 int migrate_compress_threads(void);
+int migrate_compress_wait_thread(void);
 int migrate_decompress_threads(void);
 bool migrate_use_events(void);
 bool migrate_postcopy_blocktime(void);
diff --git a/migration/ram.c b/migration/ram.c
index 2f9e8bd..498834b 100644
--- a/migration/ram.c
+++ b/migration/ram.c
@@ -1896,30 +1896,34 @@ static int compress_page_with_multi_thread(RAMState *rs, RAMBlock *block,
                                            ram_addr_t offset)
 {
     int idx, thread_count, bytes_xmit = -1, pages = -1;
+    bool wait = migrate_compress_wait_thread();
 
     thread_count = migrate_compress_threads();
     qemu_mutex_lock(&comp_done_lock);
-    while (true) {
-        for (idx = 0; idx < thread_count; idx++) {
-            if (comp_param[idx].done) {
-                comp_param[idx].done = false;
-                bytes_xmit = qemu_put_qemu_file(rs->f, comp_param[idx].file);
-                qemu_mutex_lock(&comp_param[idx].mutex);
-                set_compress_params(&comp_param[idx], block, offset);
-                qemu_cond_signal(&comp_param[idx].cond);
-                qemu_mutex_unlock(&comp_param[idx].mutex);
-                pages = 1;
-                ram_counters.normal++;
-                ram_counters.transferred += bytes_xmit;
-                break;
-            }
-        }
-        if (pages > 0) {
+retry:
+    for (idx = 0; idx < thread_count; idx++) {
+        if (comp_param[idx].done) {
+            comp_param[idx].done = false;
+            bytes_xmit = qemu_put_qemu_file(rs->f, comp_param[idx].file);
+            qemu_mutex_lock(&comp_param[idx].mutex);
+            set_compress_params(&comp_param[idx], block, offset);
+            qemu_cond_signal(&comp_param[idx].cond);
+            qemu_mutex_unlock(&comp_param[idx].mutex);
+            pages = 1;
+            ram_counters.normal++;
+            ram_counters.transferred += bytes_xmit;
             break;
-        } else {
-            qemu_cond_wait(&comp_done_cond, &comp_done_lock);
         }
     }
+
+    /*
+     * wait for the free thread if the user specifies 'compress-wait-thread',
+     * otherwise we will post the page out in the main thread as normal page.
+     */
+    if (pages < 0 && wait) {
+        qemu_cond_wait(&comp_done_cond, &comp_done_lock);
+        goto retry;
+    }
     qemu_mutex_unlock(&comp_done_lock);
 
     return pages;
@@ -2233,7 +2237,10 @@ static int ram_save_target_page(RAMState *rs, PageSearchStatus *pss,
      * CPU resource.
      */
     if (block == rs->last_sent_block && save_page_use_compression(rs)) {
-        return compress_page_with_multi_thread(rs, block, offset);
+        res = compress_page_with_multi_thread(rs, block, offset);
+        if (res > 0) {
+            return res;
+        }
     } else if (migrate_use_multifd()) {
         return ram_save_multifd_page(rs, block, offset);
     }