Merge tag 'migration-20230505-pull-request' of https://gitlab.com/jua…

…n.quintela/qemu into staging

Migration Pull request (20230505 edition)

In this series:
- fix block_bitmap_mapping (juan)
- RDMA cleanup (juan)
- qemu file cleanup (juan)

Please apply.

# -----BEGIN PGP SIGNATURE-----
#
# iQIzBAABCAAdFiEEGJn/jt6/WMzuA0uC9IfvGFhy1yMFAmRUUhUACgkQ9IfvGFhy
# 1yMxkRAAk1jfunLx/0lfN9R2W3IDwOOPrcOOOd6e7zKb7GzccObKPlqb/eQyvjCe
# FjOenJ8qgh90sW3m99ZBEla3BKekJlCB88olTjHX6FzWz7HUXMv0ip9Xx/Hg3KA/
# gW8AJhHDI+dKpevmKdhWc3sDG+TvMF6YE3hrSm5TtZ0XdHCH+McvwQm6w6O5/CN0
# HjSqV2veweNFctBRaDk9KNvQ5o9/1UYp98N3FjoRGAuxdTeilBZ2dcSmGXrRj789
# nKSCkNxEAjM0cawuo5EqZ5TCy/hFCqWA+W9hFcz63i9bZAwDu/KF7KyR62kKEh5q
# X1JNKqlyuVNutM3Pn8kbTausoWUxEUgbeGI7ID/iQYmP6V36AyyYlASFlLDwPyQs
# lgHdEAzPFmQlHMDior4TKE7+9ZtAE6/g5yYaIuGh04pwhlBzwJ/rgyi7Y5sP1Yqr
# Y5n+y6Ku7wj+gSswZK5iLQ3OFrJfwFQHIfHtW+22oR7oc9Vg0n+1Xsp1cHdJBrWu
# TzdWjX3KnSMTN9x40dJW/7wkt5+XCZcfPcTP/828dGmk0FN8TJMuCvfh79te8tUj
# TQ3NDjV4TO8jZBNB3p1ZZfMmwKHvDCBLRr0aj3MVZSvAcoHPlR6yjMENhsm4ERDA
# Xtsbqt3mIIq0fIvmQHgXDiUvy2FQw/w3Zhrdb9GbBFdYB/T+iFU=
# =79n1
# -----END PGP SIGNATURE-----
# gpg: Signature made Fri 05 May 2023 01:47:17 AM BST
# gpg:                using RSA key 1899FF8EDEBF58CCEE034B82F487EF185872D723
# gpg: Good signature from "Juan Quintela <quintela@redhat.com>" [undefined]
# gpg:                 aka "Juan Quintela <quintela@trasno.org>" [undefined]
# gpg: WARNING: This key is not certified with a trusted signature!
# gpg:          There is no indication that the signature belongs to the owner.
# Primary key fingerprint: 1899 FF8E DEBF 58CC EE03  4B82 F487 EF18 5872 D723

* tag 'migration-20230505-pull-request' of https://gitlab.com/juan.quintela/qemu:
  qemu-file: Make ram_control_save_page() use accessors for rate_limit
  qemu-file: Make total_transferred an uint64_t
  qemu-file: remove shutdown member
  qemu-file: No need to check for shutdown in qemu_file_rate_limit
  migration: qemu_file_total_transferred() function is monotonic
  migration: max_postcopy_bandwidth is a size parameter
  migration/rdma: Check for postcopy sooner
  migration/rdma: It makes no sense to recive that flag without RDMA
  migration/rdma: We can calculate the rioc from the QEMUFile
  migration/rdma: simplify ram_control_load_hook()
  migration: Make RAM_SAVE_FLAG_HOOK a normal case entry
  migration: Rename xbzrle_enabled xbzrle_started
  migration: Put zero_pages in alphabetical order
  migration: Document all migration_stats
  migration/rdma: Don't pass the QIOChannelRDMA as an opaque
  migration: Fix block_bitmap_mapping migration

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>

migration/block-dirty-bitmap.c

@@ -606,11 +606,9 @@ static int init_dirty_bitmap_migration(DBMSaveState *s)
     GHashTable *handled_by_blk = g_hash_table_new(NULL, NULL);
     BlockBackend *blk;
     GHashTable *alias_map = NULL;
-    const BitmapMigrationNodeAliasList *block_bitmap_mapping =
-        migrate_block_bitmap_mapping();
 
-    if (block_bitmap_mapping) {
-        alias_map = construct_alias_map(block_bitmap_mapping, true,
+    if (migrate_has_block_bitmap_mapping()) {
+        alias_map = construct_alias_map(migrate_block_bitmap_mapping(), true,
                                         &error_abort);
     }
 
@@ -1159,8 +1157,6 @@ static int dirty_bitmap_load_header(QEMUFile *f, DBMLoadState *s,
 static int dirty_bitmap_load(QEMUFile *f, void *opaque, int version_id)
 {
     GHashTable *alias_map = NULL;
-    const BitmapMigrationNodeAliasList *block_bitmap_mapping =
-        migrate_block_bitmap_mapping();
     DBMLoadState *s = &((DBMState *)opaque)->load;
     int ret = 0;
 
@@ -1172,9 +1168,9 @@ static int dirty_bitmap_load(QEMUFile *f, void *opaque, int version_id)
         return -EINVAL;
     }
 
-    if (block_bitmap_mapping) {
-        alias_map = construct_alias_map(block_bitmap_mapping,
-                                        false, &error_abort);
+    if (migrate_has_block_bitmap_mapping()) {
+        alias_map = construct_alias_map(migrate_block_bitmap_mapping(), false,
+                                        &error_abort);
     }
 
     do {

migration/block.c

@@ -747,8 +747,7 @@ static int block_save_setup(QEMUFile *f, void *opaque)
 static int block_save_iterate(QEMUFile *f, void *opaque)
 {
     int ret;
-    int64_t last_bytes = qemu_file_total_transferred(f);
-    int64_t delta_bytes;
+    uint64_t last_bytes = qemu_file_total_transferred(f);
 
     trace_migration_block_save("iterate", block_mig_state.submitted,
                                block_mig_state.transferred);
@@ -800,14 +799,8 @@ static int block_save_iterate(QEMUFile *f, void *opaque)
     }
 
     qemu_put_be64(f, BLK_MIG_FLAG_EOS);
-    delta_bytes = qemu_file_total_transferred(f) - last_bytes;
-    if (delta_bytes > 0) {
-        return 1;
-    } else if (delta_bytes < 0) {
-        return -1;
-    } else {
-        return 0;
-    }
+    uint64_t delta_bytes = qemu_file_total_transferred(f) - last_bytes;
+    return (delta_bytes > 0);
 }
 
 /* Called with iothread lock taken.  */

migration/migration-stats.h

@@ -22,18 +22,61 @@
  * one thread).
  */
 typedef struct {
+    /*
+     * Number of bytes that were dirty last time that we synced with
+     * the guest memory.  We use that to calculate the downtime.  As
+     * the remaining dirty amounts to what we know that is still dirty
+     * since last iteration, not counting what the guest has dirtied
+     * since we synchronized bitmaps.
+     */
     Stat64 dirty_bytes_last_sync;
+    /*
+     * Number of pages dirtied per second.
+     */
     Stat64 dirty_pages_rate;
+    /*
+     * Number of times we have synchronized guest bitmaps.
+     */
     Stat64 dirty_sync_count;
+    /*
+     * Number of times zero copy failed to send any page using zero
+     * copy.
+     */
     Stat64 dirty_sync_missed_zero_copy;
+    /*
+     * Number of bytes sent at migration completion stage while the
+     * guest is stopped.
+     */
     Stat64 downtime_bytes;
-    Stat64 zero_pages;
+    /*
+     * Number of bytes sent through multifd channels.
+     */
     Stat64 multifd_bytes;
+    /*
+     * Number of pages transferred that were not full of zeros.
+     */
     Stat64 normal_pages;
+    /*
+     * Number of bytes sent during postcopy.
+     */
     Stat64 postcopy_bytes;
+    /*
+     * Number of postcopy page faults that we have handled during
+     * postcopy stage.
+     */
     Stat64 postcopy_requests;
+    /*
+     * Number of bytes sent during precopy stage.
+     */
     Stat64 precopy_bytes;
+    /*
+     * Total number of bytes transferred.
+     */
     Stat64 transferred;
+    /*
+     * Number of pages transferred that were full of zeros.
+     */
+    Stat64 zero_pages;
 } MigrationAtomicStats;
 
 extern MigrationAtomicStats mig_stats;

migration/migration.c

@@ -2056,7 +2056,7 @@ static int postcopy_start(MigrationState *ms)
     QIOChannelBuffer *bioc;
     QEMUFile *fb;
     int64_t time_at_stop = qemu_clock_get_ms(QEMU_CLOCK_REALTIME);
-    int64_t bandwidth = migrate_max_postcopy_bandwidth();
+    uint64_t bandwidth = migrate_max_postcopy_bandwidth();
     bool restart_block = false;
     int cur_state = MIGRATION_STATUS_ACTIVE;
 
@@ -3176,7 +3176,7 @@ static void *bg_migration_thread(void *opaque)
 void migrate_fd_connect(MigrationState *s, Error *error_in)
 {
     Error *local_err = NULL;
-    int64_t rate_limit;
+    uint64_t rate_limit;
     bool resume = s->state == MIGRATION_STATUS_POSTCOPY_PAUSED;
 
     /*

migration/options.c

@@ -626,6 +626,13 @@ const BitmapMigrationNodeAliasList *migrate_block_bitmap_mapping(void)
     return s->parameters.block_bitmap_mapping;
 }
 
+bool migrate_has_block_bitmap_mapping(void)
+{
+    MigrationState *s = migrate_get_current();
+
+    return s->parameters.has_block_bitmap_mapping;
+}
+
 bool migrate_block_incremental(void)
 {
     MigrationState *s = migrate_get_current();
@@ -710,7 +717,7 @@ uint64_t migrate_max_bandwidth(void)
     return s->parameters.max_bandwidth;
 }
 
-int64_t migrate_max_postcopy_bandwidth(void)
+uint64_t migrate_max_postcopy_bandwidth(void)
 {
     MigrationState *s = migrate_get_current();
 

migration/options.h

@@ -71,6 +71,8 @@ bool migrate_cap_set(int cap, bool value, Error **errp);
 /* parameters */
 
 const BitmapMigrationNodeAliasList *migrate_block_bitmap_mapping(void);
+bool migrate_has_block_bitmap_mapping(void);
+
 bool migrate_block_incremental(void);
 uint32_t migrate_checkpoint_delay(void);
 int migrate_compress_level(void);
@@ -83,7 +85,7 @@ int migrate_decompress_threads(void);
 uint64_t migrate_downtime_limit(void);
 uint8_t migrate_max_cpu_throttle(void);
 uint64_t migrate_max_bandwidth(void);
-int64_t migrate_max_postcopy_bandwidth(void);
+uint64_t migrate_max_postcopy_bandwidth(void);
 int migrate_multifd_channels(void);
 MultiFDCompression migrate_multifd_compression(void);
 int migrate_multifd_zlib_level(void);

migration/qemu-file.c

@@ -51,7 +51,7 @@ struct QEMUFile {
     int64_t rate_limit_used;
 
     /* The sum of bytes transferred on the wire */
-    int64_t total_transferred;
+    uint64_t total_transferred;
 
     int buf_index;
     int buf_size; /* 0 when writing */
@@ -63,8 +63,6 @@ struct QEMUFile {
 
     int last_error;
     Error *last_error_obj;
-    /* has the file has been shutdown */
-    bool shutdown;
 };
 
 /*
@@ -78,8 +76,6 @@ int qemu_file_shutdown(QEMUFile *f)
 {
     int ret = 0;
 
-    f->shutdown = true;
-
     /*
      * We must set qemufile error before the real shutdown(), otherwise
      * there can be a race window where we thought IO all went though
@@ -294,7 +290,7 @@ void qemu_fflush(QEMUFile *f)
         return;
     }
 
-    if (f->shutdown) {
+    if (qemu_file_get_error(f)) {
         return;
     }
     if (f->iovcnt > 0) {
@@ -340,21 +336,11 @@ void ram_control_after_iterate(QEMUFile *f, uint64_t flags)
 
 void ram_control_load_hook(QEMUFile *f, uint64_t flags, void *data)
 {
-    int ret = -EINVAL;
-
     if (f->hooks && f->hooks->hook_ram_load) {
-        ret = f->hooks->hook_ram_load(f, flags, data);
+        int ret = f->hooks->hook_ram_load(f, flags, data);
         if (ret < 0) {
             qemu_file_set_error(f, ret);
         }
-    } else {
-        /*
-         * Hook is a hook specifically requested by the source sending a flag
-         * that expects there to be a hook on the destination.
-         */
-        if (flags == RAM_CONTROL_HOOK) {
-            qemu_file_set_error(f, ret);
-        }
     }
 }
 
@@ -366,7 +352,7 @@ size_t ram_control_save_page(QEMUFile *f, ram_addr_t block_offset,
         int ret = f->hooks->save_page(f, block_offset,
                                       offset, size, bytes_sent);
         if (ret != RAM_SAVE_CONTROL_NOT_SUPP) {
-            f->rate_limit_used += size;
+            qemu_file_acct_rate_limit(f, size);
         }
 
         if (ret != RAM_SAVE_CONTROL_DELAYED &&
@@ -407,7 +393,7 @@ static ssize_t coroutine_mixed_fn qemu_fill_buffer(QEMUFile *f)
     f->buf_index = 0;
     f->buf_size = pending;
 
-    if (f->shutdown) {
+    if (qemu_file_get_error(f)) {
         return 0;
     }
 
@@ -496,7 +482,7 @@ static int add_to_iovec(QEMUFile *f, const uint8_t *buf, size_t size,
     } else {
         if (f->iovcnt >= MAX_IOV_SIZE) {
             /* Should only happen if a previous fflush failed */
-            assert(f->shutdown || !qemu_file_is_writable(f));
+            assert(qemu_file_get_error(f) || !qemu_file_is_writable(f));
             return 1;
         }
         if (may_free) {
@@ -722,9 +708,9 @@ int coroutine_mixed_fn qemu_get_byte(QEMUFile *f)
     return result;
 }
 
-int64_t qemu_file_total_transferred_fast(QEMUFile *f)
+uint64_t qemu_file_total_transferred_fast(QEMUFile *f)
 {
-    int64_t ret = f->total_transferred;
+    uint64_t ret = f->total_transferred;
     int i;
 
     for (i = 0; i < f->iovcnt; i++) {
@@ -734,17 +720,14 @@ int64_t qemu_file_total_transferred_fast(QEMUFile *f)
     return ret;
 }
 
-int64_t qemu_file_total_transferred(QEMUFile *f)
+uint64_t qemu_file_total_transferred(QEMUFile *f)
 {
     qemu_fflush(f);
     return f->total_transferred;
 }
 
 int qemu_file_rate_limit(QEMUFile *f)
 {
-    if (f->shutdown) {
-        return 1;
-    }
     if (qemu_file_get_error(f)) {
         return 1;
     }

migration/qemu-file.h

@@ -83,7 +83,7 @@ int qemu_fclose(QEMUFile *f);
  *
  * Returns: the total bytes transferred
  */
-int64_t qemu_file_total_transferred(QEMUFile *f);
+uint64_t qemu_file_total_transferred(QEMUFile *f);
 
 /*
  * qemu_file_total_transferred_fast:
@@ -95,7 +95,7 @@ int64_t qemu_file_total_transferred(QEMUFile *f);
  *
  * Returns: the total bytes transferred and queued
  */
-int64_t qemu_file_total_transferred_fast(QEMUFile *f);
+uint64_t qemu_file_total_transferred_fast(QEMUFile *f);
 
 /*
  * put_buffer without copying the buffer.

migration/ram.c

@@ -388,8 +388,8 @@ struct RAMState {
     uint64_t xbzrle_pages_prev;
     /* Amount of xbzrle encoded bytes since the beginning of the period */
     uint64_t xbzrle_bytes_prev;
-    /* Start using XBZRLE (e.g., after the first round). */
-    bool xbzrle_enabled;
+    /* Are we really using XBZRLE (e.g., after the first round). */
+    bool xbzrle_started;
     /* Are we on the last stage of migration */
     bool last_stage;
     /* compression statistics since the beginning of the period */
@@ -1420,7 +1420,7 @@ static int ram_save_page(RAMState *rs, PageSearchStatus *pss)
     trace_ram_save_page(block->idstr, (uint64_t)offset, p);
 
     XBZRLE_cache_lock();
-    if (rs->xbzrle_enabled && !migration_in_postcopy()) {
+    if (rs->xbzrle_started && !migration_in_postcopy()) {
         pages = save_xbzrle_page(rs, pss, &p, current_addr,
                                  block, offset);
         if (!rs->last_stage) {
@@ -1636,7 +1636,7 @@ static int find_dirty_block(RAMState *rs, PageSearchStatus *pss)
             pss->complete_round = true;
             /* After the first round, enable XBZRLE. */
             if (migrate_xbzrle()) {
-                rs->xbzrle_enabled = true;
+                rs->xbzrle_started = true;
             }
         }
         /* Didn't find anything this time, but try again on the new block */
@@ -2288,7 +2288,7 @@ static bool save_page_use_compression(RAMState *rs)
      * using the data compression. In theory, xbzrle can do better than
      * compression.
      */
-    if (rs->xbzrle_enabled) {
+    if (rs->xbzrle_started) {
         return false;
     }
 
@@ -2357,7 +2357,7 @@ static int ram_save_target_page_legacy(RAMState *rs, PageSearchStatus *pss)
         /* Must let xbzrle know, otherwise a previous (now 0'd) cached
          * page would be stale
          */
-        if (rs->xbzrle_enabled) {
+        if (rs->xbzrle_started) {
             XBZRLE_cache_lock();
             xbzrle_cache_zero_page(rs, block->offset + offset);
             XBZRLE_cache_unlock();
@@ -2738,7 +2738,7 @@ static void ram_state_reset(RAMState *rs)
     rs->last_seen_block = NULL;
     rs->last_page = 0;
     rs->last_version = ram_list.version;
-    rs->xbzrle_enabled = false;
+    rs->xbzrle_started = false;
 }
 
 #define MAX_WAIT 50 /* ms, half buffered_file limit */
@@ -4445,14 +4445,12 @@ static int ram_load_precopy(QEMUFile *f)
                 multifd_recv_sync_main();
             }
             break;
+        case RAM_SAVE_FLAG_HOOK:
+            ram_control_load_hook(f, RAM_CONTROL_HOOK, NULL);
+            break;
         default:
-            if (flags & RAM_SAVE_FLAG_HOOK) {
-                ram_control_load_hook(f, RAM_CONTROL_HOOK, NULL);
-            } else {
-                error_report("Unknown combination of migration flags: 0x%x",
-                             flags);
-                ret = -EINVAL;
-            }
+            error_report("Unknown combination of migration flags: 0x%x", flags);
+            ret = -EINVAL;
         }
         if (!ret) {
             ret = qemu_file_get_error(f);