pve-qemu-qoup/debian/patches/pve/0031-PVE-Backup-Add-dirty-bitmap-tracking-for-incremental.patch

From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
From: Stefan Reiter <s.reiter@proxmox.com>
Date: Mon, 29 Jun 2020 11:06:03 +0200
Subject: [PATCH] PVE-Backup: Add dirty-bitmap tracking for incremental backups

Uses QEMU's existing MIRROR_SYNC_MODE_BITMAP and a dirty-bitmap on top
of all backed-up drives. This will only execute the data-write callback
for any changed chunks, the PBS rust code will reuse chunks from the
previous index for everything it doesn't receive if reuse_index is true.

On error or cancellation, remove all dirty bitmaps to ensure
consistency.

Add PBS/incremental specific information to query backup info QMP and
HMP commands.

Only supported for PBS backups.

Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
Signed-off-by: Dietmar Maurer <dietmar@proxmox.com>
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
---
 block/monitor/block-hmp-cmds.c |   1 +
 monitor/hmp-cmds.c             |  45 ++++++++++----
 proxmox-backup-client.c        |   3 +-
 proxmox-backup-client.h        |   1 +
 pve-backup.c                   | 103 ++++++++++++++++++++++++++++++---
 qapi/block-core.json           |  12 +++-
 6 files changed, 142 insertions(+), 23 deletions(-)

diff --git a/block/monitor/block-hmp-cmds.c b/block/monitor/block-hmp-cmds.c
index 477044c54a..556af25861 100644
--- a/block/monitor/block-hmp-cmds.c
+++ b/block/monitor/block-hmp-cmds.c
@@ -1042,6 +1042,7 @@ void hmp_backup(Monitor *mon, const QDict *qdict)
         false, NULL, // PBS fingerprint
         false, NULL, // PBS backup-id
         false, 0, // PBS backup-time
+        false, false, // PBS incremental
         true, dir ? BACKUP_FORMAT_DIR : BACKUP_FORMAT_VMA,
         false, NULL, false, NULL, !!devlist,
         devlist, qdict_haskey(qdict, "speed"), speed, &error);
diff --git a/monitor/hmp-cmds.c b/monitor/hmp-cmds.c
index a40b25e906..670f783515 100644
--- a/monitor/hmp-cmds.c
+++ b/monitor/hmp-cmds.c
@@ -225,19 +225,42 @@ void hmp_info_backup(Monitor *mon, const QDict *qdict)
             monitor_printf(mon, "End time: %s", ctime(&info->end_time));
         }
 
-        int per = (info->has_total && info->total &&
-            info->has_transferred && info->transferred) ?
-            (info->transferred * 100)/info->total : 0;
-        int zero_per = (info->has_total && info->total &&
-                        info->has_zero_bytes && info->zero_bytes) ?
-            (info->zero_bytes * 100)/info->total : 0;
         monitor_printf(mon, "Backup file: %s\n", info->backup_file);
         monitor_printf(mon, "Backup uuid: %s\n", info->uuid);
-        monitor_printf(mon, "Total size: %zd\n", info->total);
-        monitor_printf(mon, "Transferred bytes: %zd (%d%%)\n",
-                       info->transferred, per);
-        monitor_printf(mon, "Zero bytes: %zd (%d%%)\n",
-                       info->zero_bytes, zero_per);
+
+        if (!(info->has_total && info->total))  {
+            // this should not happen normally
+            monitor_printf(mon, "Total size: %d\n", 0);
+        } else {
+            bool incremental = false;
+            size_t total_or_dirty = info->total;
+            if (info->has_transferred) {
+                if (info->has_dirty && info->dirty) {
+                     if (info->dirty < info->total) {
+                        total_or_dirty = info->dirty;
+                        incremental = true;
+                    }
+                }
+            }
+
+            int per = (info->transferred * 100)/total_or_dirty;
+
+            monitor_printf(mon, "Backup mode: %s\n", incremental ? "incremental" : "full");
+
+            int zero_per = (info->has_zero_bytes && info->zero_bytes) ?
+                (info->zero_bytes * 100)/info->total : 0;
+            monitor_printf(mon, "Total size: %zd\n", info->total);
+            monitor_printf(mon, "Transferred bytes: %zd (%d%%)\n",
+                           info->transferred, per);
+            monitor_printf(mon, "Zero bytes: %zd (%d%%)\n",
+                           info->zero_bytes, zero_per);
+
+            if (info->has_reused) {
+                int reused_per = (info->reused * 100)/total_or_dirty;
+                monitor_printf(mon, "Reused bytes: %zd (%d%%)\n",
+                               info->reused, reused_per);
+            }
+        }
     }
 
     qapi_free_BackupStatus(info);
diff --git a/proxmox-backup-client.c b/proxmox-backup-client.c
index a8f6653a81..4ce7bc0b5e 100644
--- a/proxmox-backup-client.c
+++ b/proxmox-backup-client.c
@@ -89,6 +89,7 @@ proxmox_backup_co_register_image(
     ProxmoxBackupHandle *pbs,
     const char *device_name,
     uint64_t size,
+    bool incremental,
     Error **errp)
 {
     Coroutine *co = qemu_coroutine_self();
@@ -98,7 +99,7 @@ proxmox_backup_co_register_image(
     int pbs_res = -1;
 
     proxmox_backup_register_image_async(
-        pbs, device_name, size ,proxmox_backup_schedule_wake, &waker, &pbs_res, &pbs_err);
+        pbs, device_name, size, incremental, proxmox_backup_schedule_wake, &waker, &pbs_res, &pbs_err);
     qemu_coroutine_yield();
     if (pbs_res < 0) {
         if (errp) error_setg(errp, "backup register image failed: %s", pbs_err ? pbs_err : "unknown error");
diff --git a/proxmox-backup-client.h b/proxmox-backup-client.h
index 1dda8b7d8f..8cbf645b2c 100644
--- a/proxmox-backup-client.h
+++ b/proxmox-backup-client.h
@@ -32,6 +32,7 @@ proxmox_backup_co_register_image(
     ProxmoxBackupHandle *pbs,
     const char *device_name,
     uint64_t size,
+    bool incremental,
     Error **errp);
 
 
diff --git a/pve-backup.c b/pve-backup.c
index 3d28975eaa..abd7062afe 100644
--- a/pve-backup.c
+++ b/pve-backup.c
@@ -28,6 +28,8 @@
  *
  */
 
+const char *PBS_BITMAP_NAME = "pbs-incremental-dirty-bitmap";
+
 static struct PVEBackupState {
     struct {
         // Everithing accessed from qmp_backup_query command is protected using lock
@@ -39,7 +41,9 @@ static struct PVEBackupState {
         uuid_t uuid;
         char uuid_str[37];
         size_t total;
+        size_t dirty;
         size_t transferred;
+        size_t reused;
         size_t zero_bytes;
     } stat;
     int64_t speed;
@@ -66,6 +70,7 @@ typedef struct PVEBackupDevInfo {
     uint8_t dev_id;
     bool completed;
     char targetfile[PATH_MAX];
+    BdrvDirtyBitmap *bitmap;
     BlockDriverState *target;
 } PVEBackupDevInfo;
 
@@ -107,11 +112,12 @@ static bool pvebackup_error_or_canceled(void)
     return error_or_canceled;
 }
 
-static void pvebackup_add_transfered_bytes(size_t transferred, size_t zero_bytes)
+static void pvebackup_add_transfered_bytes(size_t transferred, size_t zero_bytes, size_t reused)
 {
     qemu_mutex_lock(&backup_state.stat.lock);
     backup_state.stat.zero_bytes += zero_bytes;
     backup_state.stat.transferred += transferred;
+    backup_state.stat.reused += reused;
     qemu_mutex_unlock(&backup_state.stat.lock);
 }
 
@@ -150,7 +156,8 @@ pvebackup_co_dump_pbs_cb(
         pvebackup_propagate_error(local_err);
         return pbs_res;
     } else {
-        pvebackup_add_transfered_bytes(size, !buf ? size : 0);
+        size_t reused = (pbs_res == 0) ? size : 0;
+        pvebackup_add_transfered_bytes(size, !buf ? size : 0, reused);
     }
 
     return size;
@@ -210,11 +217,11 @@ pvebackup_co_dump_vma_cb(
         } else {
             if (remaining >= VMA_CLUSTER_SIZE) {
                 assert(ret == VMA_CLUSTER_SIZE);
-                pvebackup_add_transfered_bytes(VMA_CLUSTER_SIZE, zero_bytes);
+                pvebackup_add_transfered_bytes(VMA_CLUSTER_SIZE, zero_bytes, 0);
                 remaining -= VMA_CLUSTER_SIZE;
             } else {
                 assert(ret == remaining);
-                pvebackup_add_transfered_bytes(remaining, zero_bytes);
+                pvebackup_add_transfered_bytes(remaining, zero_bytes, 0);
                 remaining = 0;
             }
         }
@@ -250,6 +257,18 @@ static void coroutine_fn pvebackup_co_cleanup(void *unused)
             if (local_err != NULL) {
                 pvebackup_propagate_error(local_err);
             }
+        } else {
+            // on error or cancel we cannot ensure synchronization of dirty
+            // bitmaps with backup server, so remove all and do full backup next
+            GList *l = backup_state.di_list;
+            while (l) {
+                PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
+                l = g_list_next(l);
+
+                if (di->bitmap) {
+                    bdrv_release_dirty_bitmap(di->bitmap);
+                }
+            }
         }
 
         proxmox_backup_disconnect(backup_state.pbs);
@@ -305,6 +324,12 @@ static void pvebackup_complete_cb(void *opaque, int ret)
     // remove self from job queue
     backup_state.di_list = g_list_remove(backup_state.di_list, di);
 
+    if (di->bitmap && ret < 0) {
+        // on error or cancel we cannot ensure synchronization of dirty
+        // bitmaps with backup server, so remove all and do full backup next
+        bdrv_release_dirty_bitmap(di->bitmap);
+    }
+
     g_free(di);
 
     qemu_mutex_unlock(&backup_state.backup_mutex);
@@ -469,12 +494,18 @@ static bool create_backup_jobs(void) {
 
         assert(di->target != NULL);
 
+        MirrorSyncMode sync_mode = MIRROR_SYNC_MODE_FULL;
+        BitmapSyncMode bitmap_mode = BITMAP_SYNC_MODE_NEVER;
+        if (di->bitmap) {
+            sync_mode = MIRROR_SYNC_MODE_BITMAP;
+            bitmap_mode = BITMAP_SYNC_MODE_ON_SUCCESS;
+        }
         AioContext *aio_context = bdrv_get_aio_context(di->bs);
         aio_context_acquire(aio_context);
 
         BlockJob *job = backup_job_create(
-            NULL, di->bs, di->target, backup_state.speed, MIRROR_SYNC_MODE_FULL, NULL,
-            BITMAP_SYNC_MODE_NEVER, false, NULL, &perf, BLOCKDEV_ON_ERROR_REPORT, BLOCKDEV_ON_ERROR_REPORT,
+            NULL, di->bs, di->target, backup_state.speed, sync_mode, di->bitmap,
+            bitmap_mode, false, NULL, &perf, BLOCKDEV_ON_ERROR_REPORT, BLOCKDEV_ON_ERROR_REPORT,
             JOB_DEFAULT, pvebackup_complete_cb, di, NULL, &local_err);
 
         aio_context_release(aio_context);
@@ -525,6 +556,8 @@ typedef struct QmpBackupTask {
     const char *fingerprint;
     bool has_fingerprint;
     int64_t backup_time;
+    bool has_use_dirty_bitmap;
+    bool use_dirty_bitmap;
     bool has_format;
     BackupFormat format;
     bool has_config_file;
@@ -616,6 +649,7 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
     }
 
     size_t total = 0;
+    size_t dirty = 0;
 
     l = di_list;
     while (l) {
@@ -653,6 +687,8 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
         int dump_cb_block_size = PROXMOX_BACKUP_DEFAULT_CHUNK_SIZE; // Hardcoded (4M)
         firewall_name = "fw.conf";
 
+        bool use_dirty_bitmap = task->has_use_dirty_bitmap && task->use_dirty_bitmap;
+
         char *pbs_err = NULL;
         pbs = proxmox_backup_new(
             task->backup_file,
@@ -672,7 +708,8 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
             goto err;
         }
 
-        if (proxmox_backup_co_connect(pbs, task->errp) < 0)
+        int connect_result = proxmox_backup_co_connect(pbs, task->errp);
+        if (connect_result < 0)
             goto err;
 
         /* register all devices */
@@ -683,9 +720,40 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
 
             const char *devname = bdrv_get_device_name(di->bs);
 
-            int dev_id = proxmox_backup_co_register_image(pbs, devname, di->size, task->errp);
-            if (dev_id < 0)
+            BdrvDirtyBitmap *bitmap = bdrv_find_dirty_bitmap(di->bs, PBS_BITMAP_NAME);
+            bool expect_only_dirty = false;
+
+            if (use_dirty_bitmap) {
+                if (bitmap == NULL) {
+                    bitmap = bdrv_create_dirty_bitmap(di->bs, dump_cb_block_size, PBS_BITMAP_NAME, task->errp);
+                    if (!bitmap) {
+                        goto err;
+                    }
+                } else {
+                    expect_only_dirty = proxmox_backup_check_incremental(pbs, devname, di->size) != 0;
+                }
+
+                if (expect_only_dirty) {
+                    dirty += bdrv_get_dirty_count(bitmap);
+                } else {
+                    /* mark entire bitmap as dirty to make full backup */
+                    bdrv_set_dirty_bitmap(bitmap, 0, di->size);
+                    dirty += di->size;
+                }
+                di->bitmap = bitmap;
+            } else {
+                dirty += di->size;
+
+                /* after a full backup the old dirty bitmap is invalid anyway */
+                if (bitmap != NULL) {
+                    bdrv_release_dirty_bitmap(bitmap);
+                }
+            }
+
+            int dev_id = proxmox_backup_co_register_image(pbs, devname, di->size, expect_only_dirty, task->errp);
+            if (dev_id < 0) {
                 goto err;
+            }
 
             if (!(di->target = bdrv_backup_dump_create(dump_cb_block_size, di->size, pvebackup_co_dump_pbs_cb, di, task->errp))) {
                 goto err;
@@ -694,6 +762,8 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
             di->dev_id = dev_id;
         }
     } else if (format == BACKUP_FORMAT_VMA) {
+        dirty = total;
+
         vmaw = vma_writer_create(task->backup_file, uuid, &local_err);
         if (!vmaw) {
             if (local_err) {
@@ -721,6 +791,8 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
             }
         }
     } else if (format == BACKUP_FORMAT_DIR) {
+        dirty = total;
+
         if (mkdir(task->backup_file, 0640) != 0) {
             error_setg_errno(task->errp, errno, "can't create directory '%s'\n",
                              task->backup_file);
@@ -793,8 +865,10 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
     char *uuid_str = g_strdup(backup_state.stat.uuid_str);
 
     backup_state.stat.total = total;
+    backup_state.stat.dirty = dirty;
     backup_state.stat.transferred = 0;
     backup_state.stat.zero_bytes = 0;
+    backup_state.stat.reused = format == BACKUP_FORMAT_PBS && dirty >= total ? 0 : total - dirty;
 
     qemu_mutex_unlock(&backup_state.stat.lock);
 
@@ -818,6 +892,10 @@ err:
         PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
         l = g_list_next(l);
 
+        if (di->bitmap) {
+            bdrv_release_dirty_bitmap(di->bitmap);
+        }
+
         if (di->target) {
             bdrv_unref(di->target);
         }
@@ -859,6 +937,7 @@ UuidInfo *qmp_backup(
     bool has_fingerprint, const char *fingerprint,
     bool has_backup_id, const char *backup_id,
     bool has_backup_time, int64_t backup_time,
+    bool has_use_dirty_bitmap, bool use_dirty_bitmap,
     bool has_format, BackupFormat format,
     bool has_config_file, const char *config_file,
     bool has_firewall_file, const char *firewall_file,
@@ -877,6 +956,8 @@ UuidInfo *qmp_backup(
         .backup_id = backup_id,
         .has_backup_time = has_backup_time,
         .backup_time = backup_time,
+        .has_use_dirty_bitmap = has_use_dirty_bitmap,
+        .use_dirty_bitmap = use_dirty_bitmap,
         .has_format = has_format,
         .format = format,
         .has_config_file = has_config_file,
@@ -945,10 +1026,14 @@ BackupStatus *qmp_query_backup(Error **errp)
 
     info->has_total = true;
     info->total = backup_state.stat.total;
+    info->has_dirty = true;
+    info->dirty = backup_state.stat.dirty;
     info->has_zero_bytes = true;
     info->zero_bytes = backup_state.stat.zero_bytes;
     info->has_transferred = true;
     info->transferred = backup_state.stat.transferred;
+    info->has_reused = true;
+    info->reused = backup_state.stat.reused;
 
     qemu_mutex_unlock(&backup_state.stat.lock);
 
diff --git a/qapi/block-core.json b/qapi/block-core.json
index c3b6b93472..992e6c1e3f 100644
--- a/qapi/block-core.json
+++ b/qapi/block-core.json
@@ -753,8 +753,13 @@
 #
 # @total: total amount of bytes involved in the backup process
 #
+# @dirty: with incremental mode (PBS) this is the amount of bytes involved
+#         in the backup process which are marked dirty.
+#
 # @transferred: amount of bytes already backed up.
 #
+# @reused: amount of bytes reused due to deduplication.
+#
 # @zero-bytes: amount of 'zero' bytes detected.
 #
 # @start-time: time (epoch) when backup job started.
@@ -767,8 +772,8 @@
 #
 ##
 { 'struct': 'BackupStatus',
-  'data': {'*status': 'str', '*errmsg': 'str', '*total': 'int',
-           '*transferred': 'int', '*zero-bytes': 'int',
+  'data': {'*status': 'str', '*errmsg': 'str', '*total': 'int', '*dirty': 'int',
+           '*transferred': 'int', '*zero-bytes': 'int', '*reused': 'int',
            '*start-time': 'int', '*end-time': 'int',
            '*backup-file': 'str', '*uuid': 'str' } }
 
@@ -811,6 +816,8 @@
 #
 # @backup-time: backup timestamp (Unix epoch, required for format 'pbs')
 #
+# @use-dirty-bitmap: use dirty bitmap to detect incremental changes since last job (optional for format 'pbs')
+#
 # Returns: the uuid of the backup job
 #
 ##
@@ -821,6 +828,7 @@
                                     '*fingerprint': 'str',
                                     '*backup-id': 'str',
                                     '*backup-time': 'int',
+                                    '*use-dirty-bitmap': 'bool',
                                     '*format': 'BackupFormat',
                                     '*config-file': 'str',
                                     '*firewall-file': 'str',