Block layer patches

- Managing inactive nodes (enables QSD migration with shared storage)
 - Fix swapped values for BLOCK_IO_ERROR 'device' and 'qom-path'
 - vpc: Read images exported from Azure correctly
 - scripts/qemu-gdb: Support coroutine dumps in coredumps
 - Minor cleanups
 -----BEGIN PGP SIGNATURE-----
 
 iQJFBAABCAAvFiEE3D3rFZqa+V09dFb+fwmycsiPL9YFAmek34IRHGt3b2xmQHJl
 ZGhhdC5jb20ACgkQfwmycsiPL9bDpxAAnTvwmdazAXG0g9GzqvrEB/+6rStjAsqE
 9MTWV4WxyN41d0RXxN8CYKb8CXSiTRyw6r3CSGNYEI2eShe9e934PriSkZm41HyX
 n9Yh5YxqGZqitzvPtx62Ii/1KG+PcjQbfHuK1p4+rlKa0yQ2eGlio1JIIrZrCkBZ
 ikZcQUrhIyD0XV8hTQ2+Ysa+ZN6itjnlTQIG3gS3m8f8WR7kyUXD8YFMQFJFyjVx
 NrAIpLnc/ln9+5PZR9tje8U7XEn2KCgI5pgGaQnrd0h0G1H4ig8ogzYYnKTLhjU/
 AmQpS8np8Tyg6S1UZTiekEq0VuAhThEQc5b3sGbmHWH/R2ABMStyf18oCBAkPzZ7
 s6h+3XzTKKY2Q5Q3ZG/ANkUJjTNBhdj1fcaARvbSWsqsuk5CWX/I3jzvgihFtCSs
 eGu+b/bLeW6P7hu4qPHBcgLHuB1Fc7Rd2t4BoIGM1wcO2CeC9DzUKOiIMZOEJIh0
 GGqCkEWDHgckDTakD4/vSqm0UDKt6FSlQC9ga/ILBY3IB5HpHoArY58selymy28i
 X7MgAvbjdsmNuUuXDZZOiObcFt3j8jlmwPJpPyzXPQIiPX1RXeBPRhVAEeZCKn6Z
 tfHr72SJdMeVOGXVTvOrJ2iW+4g03rPdmkDFCUhpOwo62RODq7ahvCIXsNf3nEFR
 rSB3T1M/8EM=
 =iQLP
 -----END PGP SIGNATURE-----

Merge tag 'for-upstream' of https://repo.or.cz/qemu/kevin into staging

Block layer patches

- Managing inactive nodes (enables QSD migration with shared storage)
- Fix swapped values for BLOCK_IO_ERROR 'device' and 'qom-path'
- vpc: Read images exported from Azure correctly
- scripts/qemu-gdb: Support coroutine dumps in coredumps
- Minor cleanups

# -----BEGIN PGP SIGNATURE-----
#
# iQJFBAABCAAvFiEE3D3rFZqa+V09dFb+fwmycsiPL9YFAmek34IRHGt3b2xmQHJl
# ZGhhdC5jb20ACgkQfwmycsiPL9bDpxAAnTvwmdazAXG0g9GzqvrEB/+6rStjAsqE
# 9MTWV4WxyN41d0RXxN8CYKb8CXSiTRyw6r3CSGNYEI2eShe9e934PriSkZm41HyX
# n9Yh5YxqGZqitzvPtx62Ii/1KG+PcjQbfHuK1p4+rlKa0yQ2eGlio1JIIrZrCkBZ
# ikZcQUrhIyD0XV8hTQ2+Ysa+ZN6itjnlTQIG3gS3m8f8WR7kyUXD8YFMQFJFyjVx
# NrAIpLnc/ln9+5PZR9tje8U7XEn2KCgI5pgGaQnrd0h0G1H4ig8ogzYYnKTLhjU/
# AmQpS8np8Tyg6S1UZTiekEq0VuAhThEQc5b3sGbmHWH/R2ABMStyf18oCBAkPzZ7
# s6h+3XzTKKY2Q5Q3ZG/ANkUJjTNBhdj1fcaARvbSWsqsuk5CWX/I3jzvgihFtCSs
# eGu+b/bLeW6P7hu4qPHBcgLHuB1Fc7Rd2t4BoIGM1wcO2CeC9DzUKOiIMZOEJIh0
# GGqCkEWDHgckDTakD4/vSqm0UDKt6FSlQC9ga/ILBY3IB5HpHoArY58selymy28i
# X7MgAvbjdsmNuUuXDZZOiObcFt3j8jlmwPJpPyzXPQIiPX1RXeBPRhVAEeZCKn6Z
# tfHr72SJdMeVOGXVTvOrJ2iW+4g03rPdmkDFCUhpOwo62RODq7ahvCIXsNf3nEFR
# rSB3T1M/8EM=
# =iQLP
# -----END PGP SIGNATURE-----
# gpg: Signature made Thu 06 Feb 2025 11:12:50 EST
# gpg:                using RSA key DC3DEB159A9AF95D3D7456FE7F09B272C88F2FD6
# gpg:                issuer "kwolf@redhat.com"
# gpg: Good signature from "Kevin Wolf <kwolf@redhat.com>" [full]
# Primary key fingerprint: DC3D EB15 9A9A F95D 3D74  56FE 7F09 B272 C88F 2FD6

* tag 'for-upstream' of https://repo.or.cz/qemu/kevin: (25 commits)
  block: remove unused BLOCK_OP_TYPE_DATAPLANE
  iotests: Add (NBD-based) tests for inactive nodes
  iotests: Add qsd-migrate case
  iotests: Add filter_qtest()
  nbd/server: Support inactive nodes
  block/export: Add option to allow export of inactive nodes
  block: Drain nodes before inactivating them
  block/export: Don't ignore image activation error in blk_exp_add()
  block: Support inactive nodes in blk_insert_bs()
  block: Add blockdev-set-active QMP command
  block: Add option to create inactive nodes
  block: Fix crash on block_resize on inactive node
  block: Don't attach inactive child to active node
  migration/block-active: Remove global active flag
  block: Inactivate external snapshot overlays when necessary
  block: Allow inactivating already inactive nodes
  block: Add 'active' field to BlockDeviceInfo
  block-backend: Fix argument order when calling 'qapi_event_send_block_io_error()'
  scripts/qemu-gdb: Support coroutine dumps in coredumps
  scripts/qemu-gdb: Simplify fs_base fetching for coroutines
  ...

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>

block.c

@@ -1573,6 +1573,10 @@ static void update_flags_from_options(int *flags, QemuOpts *opts)
     if (qemu_opt_get_bool_del(opts, BDRV_OPT_AUTO_READ_ONLY, false)) {
         *flags |= BDRV_O_AUTO_RDONLY;
     }
+
+    if (!qemu_opt_get_bool_del(opts, BDRV_OPT_ACTIVE, true)) {
+        *flags |= BDRV_O_INACTIVE;
+    }
 }
 
 static void update_options_from_flags(QDict *options, int flags)
@@ -1799,6 +1803,11 @@ QemuOptsList bdrv_runtime_opts = {
             .type = QEMU_OPT_BOOL,
             .help = "Ignore flush requests",
         },
+        {
+            .name = BDRV_OPT_ACTIVE,
+            .type = QEMU_OPT_BOOL,
+            .help = "Node is activated",
+        },
         {
             .name = BDRV_OPT_READ_ONLY,
             .type = QEMU_OPT_BOOL,
@@ -3077,6 +3086,13 @@ bdrv_attach_child_common(BlockDriverState *child_bs,
     assert(child_class->get_parent_desc);
     GLOBAL_STATE_CODE();
 
+    if (bdrv_is_inactive(child_bs) && (perm & ~BLK_PERM_CONSISTENT_READ)) {
+        g_autofree char *perm_names = bdrv_perm_names(perm);
+        error_setg(errp, "Permission '%s' unavailable on inactive node",
+                   perm_names);
+        return NULL;
+    }
+
     new_child = g_new(BdrvChild, 1);
     *new_child = (BdrvChild) {
         .bs             = NULL,
@@ -3183,6 +3199,11 @@ bdrv_attach_child_noperm(BlockDriverState *parent_bs,
                    child_bs->node_name, child_name, parent_bs->node_name);
         return NULL;
     }
+    if (bdrv_is_inactive(child_bs) && !bdrv_is_inactive(parent_bs)) {
+        error_setg(errp, "Inactive '%s' can't be a %s child of active '%s'",
+                   child_bs->node_name, child_name, parent_bs->node_name);
+        return NULL;
+    }
 
     bdrv_get_cumulative_perm(parent_bs, &perm, &shared_perm);
     bdrv_child_perm(parent_bs, child_bs, NULL, child_role, NULL,
@@ -6824,6 +6845,10 @@ void bdrv_init_with_whitelist(void)
     bdrv_init();
 }
 
+bool bdrv_is_inactive(BlockDriverState *bs) {
+    return bs->open_flags & BDRV_O_INACTIVE;
+}
+
 int bdrv_activate(BlockDriverState *bs, Error **errp)
 {
     BdrvChild *child, *parent;
@@ -6955,7 +6980,8 @@ bdrv_has_bds_parent(BlockDriverState *bs, bool only_active)
     return false;
 }
 
-static int GRAPH_RDLOCK bdrv_inactivate_recurse(BlockDriverState *bs)
+static int GRAPH_RDLOCK
+bdrv_inactivate_recurse(BlockDriverState *bs, bool top_level)
 {
     BdrvChild *child, *parent;
     int ret;
@@ -6973,7 +6999,14 @@ static int GRAPH_RDLOCK bdrv_inactivate_recurse(BlockDriverState *bs)
         return 0;
     }
 
-    assert(!(bs->open_flags & BDRV_O_INACTIVE));
+    /*
+     * Inactivating an already inactive node on user request is harmless, but if
+     * a child is already inactive before its parent, that's bad.
+     */
+    if (bs->open_flags & BDRV_O_INACTIVE) {
+        assert(top_level);
+        return 0;
+    }
 
     /* Inactivate this node */
     if (bs->drv->bdrv_inactivate) {
@@ -6999,7 +7032,9 @@ static int GRAPH_RDLOCK bdrv_inactivate_recurse(BlockDriverState *bs)
         return -EPERM;
     }
 
+    bdrv_drained_begin(bs);
     bs->open_flags |= BDRV_O_INACTIVE;
+    bdrv_drained_end(bs);
 
     /*
      * Update permissions, they may differ for inactive nodes.
@@ -7010,7 +7045,7 @@ static int GRAPH_RDLOCK bdrv_inactivate_recurse(BlockDriverState *bs)
 
     /* Recursively inactivate children */
     QLIST_FOREACH(child, &bs->children, next) {
-        ret = bdrv_inactivate_recurse(child->bs);
+        ret = bdrv_inactivate_recurse(child->bs, false);
         if (ret < 0) {
             return ret;
         }
@@ -7019,6 +7054,27 @@ static int GRAPH_RDLOCK bdrv_inactivate_recurse(BlockDriverState *bs)
     return 0;
 }
 
+int bdrv_inactivate(BlockDriverState *bs, Error **errp)
+{
+    int ret;
+
+    GLOBAL_STATE_CODE();
+    GRAPH_RDLOCK_GUARD_MAINLOOP();
+
+    if (bdrv_has_bds_parent(bs, true)) {
+        error_setg(errp, "Node has active parent node");
+        return -EPERM;
+    }
+
+    ret = bdrv_inactivate_recurse(bs, true);
+    if (ret < 0) {
+        error_setg_errno(errp, -ret, "Failed to inactivate node");
+        return ret;
+    }
+
+    return 0;
+}
+
 int bdrv_inactivate_all(void)
 {
     BlockDriverState *bs = NULL;
@@ -7035,7 +7091,7 @@ int bdrv_inactivate_all(void)
         if (bdrv_has_bds_parent(bs, false)) {
             continue;
         }
-        ret = bdrv_inactivate_recurse(bs);
+        ret = bdrv_inactivate_recurse(bs, true);
         if (ret < 0) {
             bdrv_next_cleanup(&it);
             break;

block/block-backend.c

@@ -253,7 +253,7 @@ static bool blk_can_inactivate(BlockBackend *blk)
      * guest.  For block job BBs that satisfy this, we can just allow
      * it.  This is the case for mirror job source, which is required
      * by libvirt non-shared block migration. */
-    if (!(blk->perm & (BLK_PERM_WRITE | BLK_PERM_WRITE_UNCHANGED))) {
+    if (!(blk->perm & ~BLK_PERM_CONSISTENT_READ)) {
         return true;
     }
 
@@ -900,14 +900,24 @@ void blk_remove_bs(BlockBackend *blk)
 int blk_insert_bs(BlockBackend *blk, BlockDriverState *bs, Error **errp)
 {
     ThrottleGroupMember *tgm = &blk->public.throttle_group_member;
+    uint64_t perm, shared_perm;
 
     GLOBAL_STATE_CODE();
     bdrv_ref(bs);
     bdrv_graph_wrlock();
+
+    if ((bs->open_flags & BDRV_O_INACTIVE) && blk_can_inactivate(blk)) {
+        blk->disable_perm = true;
+        perm = 0;
+        shared_perm = BLK_PERM_ALL;
+    } else {
+        perm = blk->perm;
+        shared_perm = blk->shared_perm;
+    }
+
     blk->root = bdrv_root_attach_child(bs, "root", &child_root,
                                        BDRV_CHILD_FILTERED | BDRV_CHILD_PRIMARY,
-                                       blk->perm, blk->shared_perm,
+                                       perm, shared_perm, blk, errp);
-                                       blk, errp);
     bdrv_graph_wrunlock();
     if (blk->root == NULL) {
         return -EPERM;
@@ -1019,6 +1029,10 @@ DeviceState *blk_get_attached_dev(BlockBackend *blk)
     return blk->dev;
 }
 
+/*
+ * The caller is responsible for releasing the value returned
+ * with g_free() after use.
+ */
 static char *blk_get_attached_dev_id_or_path(BlockBackend *blk, bool want_id)
 {
     DeviceState *dev = blk->dev;
@@ -1033,15 +1047,15 @@ static char *blk_get_attached_dev_id_or_path(BlockBackend *blk, bool want_id)
     return object_get_canonical_path(OBJECT(dev)) ?: g_strdup("");
 }
 
-/*
- * Return the qdev ID, or if no ID is assigned the QOM path, of the block
- * device attached to the BlockBackend.
- */
 char *blk_get_attached_dev_id(BlockBackend *blk)
 {
     return blk_get_attached_dev_id_or_path(blk, true);
 }
 
+/*
+ * The caller is responsible for releasing the value returned
+ * with g_free() after use.
+ */
 static char *blk_get_attached_dev_path(BlockBackend *blk)
 {
     return blk_get_attached_dev_id_or_path(blk, false);
@@ -2134,10 +2148,10 @@ static void send_qmp_error_event(BlockBackend *blk,
 {
     IoOperationType optype;
     BlockDriverState *bs = blk_bs(blk);
+    g_autofree char *path = blk_get_attached_dev_path(blk);
 
     optype = is_read ? IO_OPERATION_TYPE_READ : IO_OPERATION_TYPE_WRITE;
-    qapi_event_send_block_io_error(blk_name(blk),
+    qapi_event_send_block_io_error(path, blk_name(blk),
-                                   blk_get_attached_dev_path(blk),
                                    bs ? bdrv_get_node_name(bs) : NULL, optype,
                                    action, blk_iostatus_is_enabled(blk),
                                    error == ENOSPC, strerror(error));

block/export/export.c

@@ -75,6 +75,7 @@ static const BlockExportDriver *blk_exp_find_driver(BlockExportType type)
 BlockExport *blk_exp_add(BlockExportOptions *export, Error **errp)
 {
     bool fixed_iothread = export->has_fixed_iothread && export->fixed_iothread;
+    bool allow_inactive = export->has_allow_inactive && export->allow_inactive;
     const BlockExportDriver *drv;
     BlockExport *exp = NULL;
     BlockDriverState *bs;
@@ -138,14 +139,25 @@ BlockExport *blk_exp_add(BlockExportOptions *export, Error **errp)
         }
     }
 
-    /*
-     * Block exports are used for non-shared storage migration. Make sure
-     * that BDRV_O_INACTIVE is cleared and the image is ready for write
-     * access since the export could be available before migration handover.
-     * ctx was acquired in the caller.
-     */
     bdrv_graph_rdlock_main_loop();
-    bdrv_activate(bs, NULL);
+    if (allow_inactive) {
+        if (!drv->supports_inactive) {
+            error_setg(errp, "Export type does not support inactive exports");
+            bdrv_graph_rdunlock_main_loop();
+            goto fail;
+        }
+    } else {
+        /*
+         * Block exports are used for non-shared storage migration. Make sure
+         * that BDRV_O_INACTIVE is cleared and the image is ready for write
+         * access since the export could be available before migration handover.
+         */
+        ret = bdrv_activate(bs, errp);
+        if (ret < 0) {
+            bdrv_graph_rdunlock_main_loop();
+            goto fail;
+        }
+    }
     bdrv_graph_rdunlock_main_loop();
 
     perm = BLK_PERM_CONSISTENT_READ;
@@ -158,6 +170,9 @@ BlockExport *blk_exp_add(BlockExportOptions *export, Error **errp)
     if (!fixed_iothread) {
         blk_set_allow_aio_context_change(blk, true);
     }
+    if (allow_inactive) {
+        blk_set_force_allow_inactivate(blk);
+    }
 
     ret = blk_insert_bs(blk, bs, errp);
     if (ret < 0) {

block/monitor/block-hmp-cmds.c

@@ -630,11 +630,12 @@ static void print_block_info(Monitor *mon, BlockInfo *info,
     }
 
     if (inserted) {
-        monitor_printf(mon, ": %s (%s%s%s)\n",
+        monitor_printf(mon, ": %s (%s%s%s%s)\n",
                        inserted->file,
                        inserted->drv,
                        inserted->ro ? ", read-only" : "",
-                       inserted->encrypted ? ", encrypted" : "");
+                       inserted->encrypted ? ", encrypted" : "",
+                       inserted->active ? "" : ", inactive");
     } else {
         monitor_printf(mon, ": [not inserted]\n");
     }

block/qapi.c

@@ -63,6 +63,7 @@ BlockDeviceInfo *bdrv_block_device_info(BlockBackend *blk,
     info->file                   = g_strdup(bs->filename);
     info->ro                     = bdrv_is_read_only(bs);
     info->drv                    = g_strdup(bs->drv->format_name);
+    info->active                 = !bdrv_is_inactive(bs);
     info->encrypted              = bs->encrypted;
 
     info->cache = g_new(BlockdevCacheInfo, 1);

block/replication.c

@@ -576,7 +576,6 @@ static void replication_start(ReplicationState *rs, ReplicationMode mode,
             return;
         }
         bdrv_op_block_all(top_bs, s->blocker);
-        bdrv_op_unblock(top_bs, BLOCK_OP_TYPE_DATAPLANE, s->blocker);
 
         bdrv_graph_wrunlock();
 

block/vpc.c

@@ -216,6 +216,39 @@ static void vpc_parse_options(BlockDriverState *bs, QemuOpts *opts,
     }
 }
 
+/*
+ * Microsoft Virtual PC and Microsoft Hyper-V produce and read
+ * VHD image sizes differently.  VPC will rely on CHS geometry,
+ * while Hyper-V and disk2vhd use the size specified in the footer.
+ *
+ * We use a couple of approaches to try and determine the correct method:
+ * look at the Creator App field, and look for images that have CHS
+ * geometry that is the maximum value.
+ *
+ * If the CHS geometry is the maximum CHS geometry, then we assume that
+ * the size is the footer->current_size to avoid truncation.  Otherwise,
+ * we follow the table based on footer->creator_app:
+ *
+ *  Known creator apps:
+ *      'vpc '  :  CHS              Virtual PC (uses disk geometry)
+ *      'qemu'  :  CHS              QEMU (uses disk geometry)
+ *      'qem2'  :  current_size     QEMU (uses current_size)
+ *      'win '  :  current_size     Hyper-V
+ *      'd2v '  :  current_size     Disk2vhd
+ *      'tap\0' :  current_size     XenServer
+ *      'CTXS'  :  current_size     XenConverter
+ *      'wa\0\0':  current_size     Azure
+ *
+ *  The user can override the table values via drive options, however
+ *  even with an override we will still use current_size for images
+ *  that have CHS geometry of the maximum size.
+ */
+static bool vpc_ignore_current_size(VHDFooter *footer)
+{
+    return !strncmp(footer->creator_app, "vpc ", 4) ||
+           !strncmp(footer->creator_app, "qemu", 4);
+}
+
 static int vpc_open(BlockDriverState *bs, QDict *options, int flags,
                     Error **errp)
 {
@@ -304,36 +337,8 @@ static int vpc_open(BlockDriverState *bs, QDict *options, int flags,
     bs->total_sectors = (int64_t)
         be16_to_cpu(footer->cyls) * footer->heads * footer->secs_per_cyl;
 
-    /* Microsoft Virtual PC and Microsoft Hyper-V produce and read
+    /* Use CHS or current_size to determine the image size. */
-     * VHD image sizes differently.  VPC will rely on CHS geometry,
+    use_chs = vpc_ignore_current_size(footer) || s->force_use_chs;
-     * while Hyper-V and disk2vhd use the size specified in the footer.
-     *
-     * We use a couple of approaches to try and determine the correct method:
-     * look at the Creator App field, and look for images that have CHS
-     * geometry that is the maximum value.
-     *
-     * If the CHS geometry is the maximum CHS geometry, then we assume that
-     * the size is the footer->current_size to avoid truncation.  Otherwise,
-     * we follow the table based on footer->creator_app:
-     *
-     *  Known creator apps:
-     *      'vpc '  :  CHS              Virtual PC (uses disk geometry)
-     *      'qemu'  :  CHS              QEMU (uses disk geometry)
-     *      'qem2'  :  current_size     QEMU (uses current_size)
-     *      'win '  :  current_size     Hyper-V
-     *      'd2v '  :  current_size     Disk2vhd
-     *      'tap\0' :  current_size     XenServer
-     *      'CTXS'  :  current_size     XenConverter
-     *
-     *  The user can override the table values via drive options, however
-     *  even with an override we will still use current_size for images
-     *  that have CHS geometry of the maximum size.
-     */
-    use_chs = (!!strncmp(footer->creator_app, "win ", 4) &&
-               !!strncmp(footer->creator_app, "qem2", 4) &&
-               !!strncmp(footer->creator_app, "d2v ", 4) &&
-               !!strncmp(footer->creator_app, "CTXS", 4) &&
-               !!memcmp(footer->creator_app, "tap", 4)) || s->force_use_chs;
 
     if (!use_chs || bs->total_sectors == VHD_MAX_GEOMETRY || s->force_use_sz) {
         bs->total_sectors = be64_to_cpu(footer->current_size) /

blockdev.c

@@ -1497,6 +1497,22 @@ static void external_snapshot_action(TransactionAction *action,
         return;
     }
 
+    /*
+     * Older QEMU versions have allowed adding an active parent node to an
+     * inactive child node. This is unsafe in the general case, but there is an
+     * important use case, which is taking a VM snapshot with migration to file
+     * and then adding an external snapshot while the VM is still stopped and
+     * images are inactive. Requiring the user to explicitly create the overlay
+     * as inactive would break compatibility, so just do it automatically here
+     * to keep this working.
+     */
+    if (bdrv_is_inactive(state->old_bs) && !bdrv_is_inactive(state->new_bs)) {
+        ret = bdrv_inactivate(state->new_bs, errp);
+        if (ret < 0) {
+            return;
+        }
+    }
+
     ret = bdrv_append(state->new_bs, state->old_bs, errp);
     if (ret < 0) {
         return;
@@ -3455,6 +3471,38 @@ void qmp_blockdev_del(const char *node_name, Error **errp)
     bdrv_unref(bs);
 }
 
+void qmp_blockdev_set_active(const char *node_name, bool active, Error **errp)
+{
+    int ret;
+
+    GLOBAL_STATE_CODE();
+    GRAPH_RDLOCK_GUARD_MAINLOOP();
+
+    if (!node_name) {
+        if (active) {
+            bdrv_activate_all(errp);
+        } else {
+            ret = bdrv_inactivate_all();
+            if (ret < 0) {
+                error_setg_errno(errp, -ret, "Failed to inactivate all nodes");
+            }
+        }
+    } else {
+        BlockDriverState *bs = bdrv_find_node(node_name);
+        if (!bs) {
+            error_setg(errp, "Failed to find node with node-name='%s'",
+                       node_name);
+            return;
+        }
+
+        if (active) {
+            bdrv_activate(bs, errp);
+        } else {
+            bdrv_inactivate(bs, errp);
+        }
+    }
+}
+
 static BdrvChild * GRAPH_RDLOCK
 bdrv_find_child(BlockDriverState *parent_bs, const char *child_name)
 {

blockjob.c

@@ -539,8 +539,6 @@ void *block_job_create(const char *job_id, const BlockJobDriver *driver,
         goto fail;
     }
 
-    bdrv_op_unblock(bs, BLOCK_OP_TYPE_DATAPLANE, job->blocker);
-
     if (!block_job_set_speed(job, speed, errp)) {
         goto fail;
     }

hw/block/virtio-blk.c

@@ -1562,15 +1562,6 @@ static bool virtio_blk_vq_aio_context_init(VirtIOBlock *s, Error **errp)
             error_setg(errp, "ioeventfd is required for iothread");
             return false;
         }
-
-        /*
-         * If ioeventfd is (re-)enabled while the guest is running there could
-         * be block jobs that can conflict.
-         */
-        if (blk_op_is_blocked(conf->conf.blk, BLOCK_OP_TYPE_DATAPLANE, errp)) {
-            error_prepend(errp, "cannot start virtio-blk ioeventfd: ");
-            return false;
-        }
     }
 
     s->vq_aio_context = g_new(AioContext *, conf->num_queues);

hw/scsi/virtio-scsi.c

@@ -1065,9 +1065,6 @@ static void virtio_scsi_hotplug(HotplugHandler *hotplug_dev, DeviceState *dev,
     int ret;
 
     if (s->ctx && !s->dataplane_fenced) {
-        if (blk_op_is_blocked(sd->conf.blk, BLOCK_OP_TYPE_DATAPLANE, errp)) {
-            return;
-        }
         ret = blk_set_aio_context(sd->conf.blk, s->ctx, errp);
         if (ret < 0) {
             return;

include/block/block-common.h

@@ -257,6 +257,7 @@ typedef enum {
 #define BDRV_OPT_AUTO_READ_ONLY "auto-read-only"
 #define BDRV_OPT_DISCARD        "discard"
 #define BDRV_OPT_FORCE_SHARE    "force-share"
+#define BDRV_OPT_ACTIVE         "active"
 
 
 #define BDRV_SECTOR_BITS   9
@@ -355,7 +356,6 @@ typedef enum BlockOpType {
     BLOCK_OP_TYPE_CHANGE,
     BLOCK_OP_TYPE_COMMIT_SOURCE,
     BLOCK_OP_TYPE_COMMIT_TARGET,
-    BLOCK_OP_TYPE_DATAPLANE,
     BLOCK_OP_TYPE_DRIVE_DEL,
     BLOCK_OP_TYPE_EJECT,
     BLOCK_OP_TYPE_EXTERNAL_SNAPSHOT,

include/block/block-global-state.h

@@ -175,12 +175,18 @@ BlockDriverState * GRAPH_RDLOCK
 check_to_replace_node(BlockDriverState *parent_bs, const char *node_name,
                       Error **errp);
 
+
+bool GRAPH_RDLOCK bdrv_is_inactive(BlockDriverState *bs);
+
 int no_coroutine_fn GRAPH_RDLOCK
 bdrv_activate(BlockDriverState *bs, Error **errp);
 
 int coroutine_fn no_co_wrapper_bdrv_rdlock
 bdrv_co_activate(BlockDriverState *bs, Error **errp);
 
+int no_coroutine_fn
+bdrv_inactivate(BlockDriverState *bs, Error **errp);
+
 void bdrv_activate_all(Error **errp);
 int bdrv_inactivate_all(void);
 

include/block/export.h

@@ -29,6 +29,9 @@ typedef struct BlockExportDriver {
      */
     size_t instance_size;
 
+    /* True if the export type supports running on an inactive node */
+    bool supports_inactive;
+
     /* Creates and starts a new block export */
     int (*create)(BlockExport *, BlockExportOptions *, Error **);
 

include/system/block-backend-io.h

@@ -32,6 +32,13 @@ void blk_set_allow_aio_context_change(BlockBackend *blk, bool allow);
 void blk_set_disable_request_queuing(BlockBackend *blk, bool disable);
 bool blk_iostatus_is_enabled(const BlockBackend *blk);
 
+/*
+ * Return the qdev ID, or if no ID is assigned the QOM path,
+ * of the block device attached to the BlockBackend.
+ *
+ * The caller is responsible for releasing the value returned
+ * with g_free() after use.
+ */
 char *blk_get_attached_dev_id(BlockBackend *blk);
 
 BlockAIOCB *blk_aio_pwrite_zeroes(BlockBackend *blk, int64_t offset,

migration/block-active.c

@@ -12,51 +12,12 @@
 #include "qemu/error-report.h"
 #include "trace.h"
 
-/*
- * Migration-only cache to remember the block layer activation status.
- * Protected by BQL.
- *
- * We need this because..
- *
- * - Migration can fail after block devices are invalidated (during
- *   switchover phase).  When that happens, we need to be able to recover
- *   the block drive status by re-activating them.
- *
- * - Currently bdrv_inactivate_all() is not safe to be invoked on top of
- *   invalidated drives (even if bdrv_activate_all() is actually safe to be
- *   called any time!).  It means remembering this could help migration to
- *   make sure it won't invalidate twice in a row, crashing QEMU.  It can
- *   happen when we migrate a PAUSED VM from host1 to host2, then migrate
- *   again to host3 without starting it.  TODO: a cleaner solution is to
- *   allow safe invoke of bdrv_inactivate_all() at anytime, like
- *   bdrv_activate_all().
- *
- * For freshly started QEMU, the flag is initialized to TRUE reflecting the
- * scenario where QEMU owns block device ownerships.
- *
- * For incoming QEMU taking a migration stream, the flag is initialized to
- * FALSE reflecting that the incoming side doesn't own the block devices,
- * not until switchover happens.
- */
-static bool migration_block_active;
-
-/* Setup the disk activation status */
-void migration_block_active_setup(bool active)
-{
-    migration_block_active = active;
-}
-
 bool migration_block_activate(Error **errp)
 {
     ERRP_GUARD();
 
     assert(bql_locked());
 
-    if (migration_block_active) {
-        trace_migration_block_activation("active-skipped");
-        return true;
-    }
-
     trace_migration_block_activation("active");
 
     bdrv_activate_all(errp);
@@ -65,7 +26,6 @@ bool migration_block_activate(Error **errp)
         return false;
     }
 
-    migration_block_active = true;
     return true;
 }
 
@@ -75,11 +35,6 @@ bool migration_block_inactivate(void)
 
     assert(bql_locked());
 
-    if (!migration_block_active) {
-        trace_migration_block_activation("inactive-skipped");
-        return true;
-    }
-
     trace_migration_block_activation("inactive");
 
     ret = bdrv_inactivate_all();
@@ -89,6 +44,5 @@ bool migration_block_inactivate(void)
         return false;
     }
 
-    migration_block_active = false;
     return true;
 }

migration/migration.c

@@ -1895,12 +1895,6 @@ void qmp_migrate_incoming(const char *uri, bool has_channels,
         return;
     }
 
-    /*
-     * Newly setup incoming QEMU.  Mark the block active state to reflect
-     * that the src currently owns the disks.
-     */
-    migration_block_active_setup(false);
-
     once = false;
 }
 
@@ -3992,8 +3986,6 @@ static void migration_instance_init(Object *obj)
     ms->state = MIGRATION_STATUS_NONE;
     ms->mbps = -1;
     ms->pages_per_second = -1;
-    /* Freshly started QEMU owns all the block devices */
-    migration_block_active_setup(true);
     qemu_sem_init(&ms->pause_sem, 0);
     qemu_mutex_init(&ms->error_mutex);
 

migration/migration.h

@@ -554,7 +554,4 @@ void migration_bitmap_sync_precopy(bool last_stage);
 void dirty_bitmap_mig_init(void);
 bool should_send_vmdesc(void);
 
-/* migration/block-active.c */
-void migration_block_active_setup(bool active);
-
 #endif

nbd/server.c

@@ -2026,6 +2026,7 @@ static void nbd_export_delete(BlockExport *blk_exp)
 const BlockExportDriver blk_exp_nbd = {
     .type               = BLOCK_EXPORT_TYPE_NBD,
     .instance_size      = sizeof(NBDExport),
+    .supports_inactive  = true,
     .create             = nbd_export_create,
     .delete             = nbd_export_delete,
     .request_shutdown   = nbd_export_request_shutdown,
@@ -2920,6 +2921,22 @@ static coroutine_fn int nbd_handle_request(NBDClient *client,
     NBDExport *exp = client->exp;
     char *msg;
     size_t i;
+    bool inactive;
+
+    WITH_GRAPH_RDLOCK_GUARD() {
+        inactive = bdrv_is_inactive(blk_bs(exp->common.blk));
+        if (inactive) {
+            switch (request->type) {
+            case NBD_CMD_READ:
+                /* These commands are allowed on inactive nodes */
+                break;
+            default:
+                /* Return an error for the rest */
+                return nbd_send_generic_reply(client, request, -EPERM,
+                                              "export is inactive", errp);
+            }
+        }
+    }
 
     switch (request->type) {
     case NBD_CMD_CACHE:

qapi/block-core.json

@@ -486,6 +486,10 @@
 # @backing_file_depth: number of files in the backing file chain
 #     (since: 1.2)
 #
+# @active: true if the backend is active; typical cases for inactive backends
+#     are on the migration source instance after migration completes and on the
+#     destination before it completes. (since: 10.0)
+#
 # @encrypted: true if the backing device is encrypted
 #
 # @detect_zeroes: detect and optimize zero writes (Since 2.1)
@@ -556,7 +560,7 @@
 { 'struct': 'BlockDeviceInfo',
   'data': { 'file': 'str', '*node-name': 'str', 'ro': 'bool', 'drv': 'str',
             '*backing_file': 'str', 'backing_file_depth': 'int',
-            'encrypted': 'bool',
+            'active': 'bool', 'encrypted': 'bool',
             'detect_zeroes': 'BlockdevDetectZeroesOptions',
             'bps': 'int', 'bps_rd': 'int', 'bps_wr': 'int',
             'iops': 'int', 'iops_rd': 'int', 'iops_wr': 'int',
@@ -4679,6 +4683,11 @@
 #
 # @cache: cache-related options
 #
+# @active: whether the block node should be activated (default: true).
+#     Having inactive block nodes is useful primarily for migration because it
+#     allows opening an image on the destination while the source is still
+#     holding locks for it. (Since 10.0)
+#
 # @read-only: whether the block device should be read-only (default:
 #     false).  Note that some block drivers support only read-only
 #     access, either generally or in certain configurations.  In this
@@ -4705,6 +4714,7 @@
             '*node-name': 'str',
             '*discard': 'BlockdevDiscardOptions',
             '*cache': 'BlockdevCacheOptions',
+            '*active': 'bool',
             '*read-only': 'bool',
             '*auto-read-only': 'bool',
             '*force-share': 'bool',
@@ -4935,6 +4945,38 @@
 { 'command': 'blockdev-del', 'data': { 'node-name': 'str' },
   'allow-preconfig': true }
 
+##
+# @blockdev-set-active:
+#
+# Activate or inactivate a block device. Use this to manage the handover of
+# block devices on migration with qemu-storage-daemon.
+#
+# Activating a node automatically activates all of its child nodes first.
+# Inactivating a node automatically inactivates any of its child nodes that are
+# not in use by a still active node.
+#
+# @node-name: Name of the graph node to activate or inactivate. By default, all
+#     nodes are affected by the operation.
+#
+# @active: true if the nodes should be active when the command returns success,
+#     false if they should be inactive.
+#
+# Since: 10.0
+#
+# .. qmp-example::
+#
+#     -> { "execute": "blockdev-set-active",
+#          "arguments": {
+#               "node-name": "node0",
+#               "active": false
+#          }
+#        }
+#     <- { "return": {} }
+##
+{ 'command': 'blockdev-set-active',
+  'data': { '*node-name': 'str', 'active': 'bool' },
+  'allow-preconfig': true }
+
 ##
 # @BlockdevCreateOptionsFile:
 #

qapi/block-export.json

@@ -372,6 +372,13 @@
 #     cannot be moved to the iothread.  The default is false.
 #     (since: 5.2)
 #
+# @allow-inactive: If true, the export allows the exported node to be inactive.
+#     If it is created for an inactive block node, the node remains inactive. If
+#     the export type doesn't support running on an inactive node, an error is
+#     returned. If false, inactive block nodes are automatically activated before
+#     creating the export and trying to inactivate them later fails.
+#     (since: 10.0; default: false)
+#
 # Since: 4.2
 ##
 { 'union': 'BlockExportOptions',
@@ -381,7 +388,8 @@
             '*iothread': 'str',
             'node-name': 'str',
             '*writable': 'bool',
-            '*writethrough': 'bool' },
+            '*writethrough': 'bool',
+            '*allow-inactive': 'bool' },
   'discriminator': 'type',
   'data': {
       'nbd': 'BlockExportOptionsNbd',

scripts/qemu-gdb.py

@@ -45,3 +45,5 @@ coroutine.CoroutineBt()
 # Default to silently passing through SIGUSR1, because QEMU sends it
 # to itself a lot.
 gdb.execute('handle SIGUSR1 pass noprint nostop')
+# Always print full stack for python errors, easier to debug and report issues
+gdb.execute('set python print-stack full')

scripts/qemugdb/coroutine.py

@@ -13,28 +13,9 @@ import gdb
 
 VOID_PTR = gdb.lookup_type('void').pointer()
 
-def get_fs_base():
-    '''Fetch %fs base value using arch_prctl(ARCH_GET_FS).  This is
-       pthread_self().'''
-    # %rsp - 120 is scratch space according to the SystemV ABI
-    old = gdb.parse_and_eval('*(uint64_t*)($rsp - 120)')
-    gdb.execute('call (int)arch_prctl(0x1003, $rsp - 120)', False, True)
-    fs_base = gdb.parse_and_eval('*(uint64_t*)($rsp - 120)')
-    gdb.execute('set *(uint64_t*)($rsp - 120) = %s' % old, False, True)
-    return fs_base
-
 def pthread_self():
-    '''Fetch pthread_self() from the glibc start_thread function.'''
+    '''Fetch the base address of TLS.'''
-    f = gdb.newest_frame()
+    return gdb.parse_and_eval("$fs_base")
-    while f.name() != 'start_thread':
-        f = f.older()
-        if f is None:
-            return get_fs_base()
-
-    try:
-        return f.read_var("arg")
-    except ValueError:
-        return get_fs_base()
 
 def get_glibc_pointer_guard():
     '''Fetch glibc pointer guard value'''
@@ -65,9 +46,60 @@ def get_jmpbuf_regs(jmpbuf):
         'r15': jmpbuf[JB_R15],
         'rip': glibc_ptr_demangle(jmpbuf[JB_PC], pointer_guard) }
 
-def bt_jmpbuf(jmpbuf):
+def symbol_lookup(addr):
-    '''Backtrace a jmpbuf'''
+    # Example: "__clone3 + 44 in section .text of /lib64/libc.so.6"
-    regs = get_jmpbuf_regs(jmpbuf)
+    result = gdb.execute(f"info symbol {hex(addr)}", to_string=True).strip()
+    try:
+        if "+" in result:
+            (func, result) = result.split(" + ")
+            (offset, result) = result.split(" in ")
+        else:
+            offset = "0"
+            (func, result) = result.split(" in ")
+        func_str = f"{func}<+{offset}> ()"
+    except:
+        return f"??? ({result})"
+
+    # Example: Line 321 of "../util/coroutine-ucontext.c" starts at address
+    # 0x55cf3894d993 <qemu_coroutine_switch+99> and ends at 0x55cf3894d9ab
+    # <qemu_coroutine_switch+123>.
+    result = gdb.execute(f"info line *{hex(addr)}", to_string=True).strip()
+    if not result.startswith("Line "):
+        return func_str
+    result = result[5:]
+
+    try:
+        result = result.split(" starts ")[0]
+        (line, path) = result.split(" of ")
+        path = path.replace("\"", "")
+    except:
+        return func_str
+
+    return f"{func_str} at {path}:{line}"
+
+def dump_backtrace(regs):
+    '''
+    Backtrace dump with raw registers, mimic GDB command 'bt'.
+    '''
+    # Here only rbp and rip that matter..
+    rbp = regs['rbp']
+    rip = regs['rip']
+    i = 0
+
+    while rbp:
+        # For all return addresses on stack, we want to look up symbol/line
+        # on the CALL command, because the return address is the next
+        # instruction instead of the CALL.  Here -1 would work for any
+        # sized CALL instruction.
+        print(f"#{i}  {hex(rip)} in {symbol_lookup(rip if i == 0 else rip-1)}")
+        rip = gdb.parse_and_eval(f"*(uint64_t *)(uint64_t)({hex(rbp)} + 8)")
+        rbp = gdb.parse_and_eval(f"*(uint64_t *)(uint64_t)({hex(rbp)})")
+        i += 1
+
+def dump_backtrace_live(regs):
+    '''
+    Backtrace dump with gdb's 'bt' command, only usable in a live session.
+    '''
     old = dict()
 
     # remember current stack frame and select the topmost
@@ -88,6 +120,17 @@ def bt_jmpbuf(jmpbuf):
 
     selected_frame.select()
 
+def bt_jmpbuf(jmpbuf):
+    '''Backtrace a jmpbuf'''
+    regs = get_jmpbuf_regs(jmpbuf)
+    try:
+        # This reuses gdb's "bt" command, which can be slightly prettier
+        # but only works with live sessions.
+        dump_backtrace_live(regs)
+    except:
+        # If above doesn't work, fallback to poor man's unwind
+        dump_backtrace(regs)
+
 def co_cast(co):
     return co.cast(gdb.lookup_type('CoroutineUContext').pointer())
 
@@ -120,10 +163,15 @@ class CoroutineBt(gdb.Command):
 
         gdb.execute("bt")
 
-        if gdb.parse_and_eval("qemu_in_coroutine()") == False:
+        try:
-            return
+            # This only works with a live session
+            co_ptr = gdb.parse_and_eval("qemu_coroutine_self()")
+        except:
+            # Fallback to use hard-coded ucontext vars if it's coredump
+            co_ptr = gdb.parse_and_eval("co_tls_current")
 
-        co_ptr = gdb.parse_and_eval("qemu_coroutine_self()")
+        if co_ptr == False:
+            return
 
         while True:
             co = co_cast(co_ptr)

tests/qemu-iotests/041

@@ -1100,10 +1100,8 @@ class TestRepairQuorum(iotests.QMPTestCase):
 
         # Check the full error message now
         self.vm.shutdown()
-        log = self.vm.get_log()
+        log = iotests.filter_qtest(self.vm.get_log())
-        log = re.sub(r'^\[I \d+\.\d+\] OPENED\n', '', log)
         log = re.sub(r'^Formatting.*\n', '', log)
-        log = re.sub(r'\n\[I \+\d+\.\d+\] CLOSED\n?$', '', log)
         log = re.sub(r'^%s: ' % os.path.basename(iotests.qemu_prog), '', log)
 
         self.assertEqual(log,

tests/qemu-iotests/165

@@ -82,9 +82,7 @@ class TestPersistentDirtyBitmap(iotests.QMPTestCase):
         self.vm.shutdown()
 
         #catch 'Persistent bitmaps are lost' possible error
-        log = self.vm.get_log()
+        log = iotests.filter_qtest(self.vm.get_log())
-        log = re.sub(r'^\[I \d+\.\d+\] OPENED\n', '', log)
-        log = re.sub(r'\[I \+\d+\.\d+\] CLOSED\n?$', '', log)
         if log:
             print(log)
 

tests/qemu-iotests/184.out

@@ -26,6 +26,7 @@ Testing:
         {
             "iops_rd": 0,
             "detect_zeroes": "off",
+            "active": true,
             "image": {
                 "backing-image": {
                     "virtual-size": 1073741824,
@@ -59,6 +60,7 @@ Testing:
         {
             "iops_rd": 0,
             "detect_zeroes": "off",
+            "active": true,
             "image": {
                 "virtual-size": 1073741824,
                 "filename": "null-co://",

tests/qemu-iotests/191.out

@@ -114,6 +114,7 @@ wrote 65536/65536 bytes at offset 1048576
         {
             "iops_rd": 0,
             "detect_zeroes": "off",
+            "active": true,
             "image": {
                 "backing-image": {
                     "virtual-size": 67108864,
@@ -155,6 +156,7 @@ wrote 65536/65536 bytes at offset 1048576
         {
             "iops_rd": 0,
             "detect_zeroes": "off",
+            "active": true,
             "image": {
                 "virtual-size": 197120,
                 "filename": "TEST_DIR/t.IMGFMT.ovl2",
@@ -183,6 +185,7 @@ wrote 65536/65536 bytes at offset 1048576
         {
             "iops_rd": 0,
             "detect_zeroes": "off",
+            "active": true,
             "image": {
                 "backing-image": {
                     "virtual-size": 67108864,
@@ -224,6 +227,7 @@ wrote 65536/65536 bytes at offset 1048576
         {
             "iops_rd": 0,
             "detect_zeroes": "off",
+            "active": true,
             "image": {
                 "virtual-size": 197120,
                 "filename": "TEST_DIR/t.IMGFMT",
@@ -252,6 +256,7 @@ wrote 65536/65536 bytes at offset 1048576
         {
             "iops_rd": 0,
             "detect_zeroes": "off",
+            "active": true,
             "image": {
                 "backing-image": {
                     "virtual-size": 67108864,
@@ -293,6 +298,7 @@ wrote 65536/65536 bytes at offset 1048576
         {
             "iops_rd": 0,
             "detect_zeroes": "off",
+            "active": true,
             "image": {
                 "virtual-size": 393216,
                 "filename": "TEST_DIR/t.IMGFMT.mid",
@@ -321,6 +327,7 @@ wrote 65536/65536 bytes at offset 1048576
         {
             "iops_rd": 0,
             "detect_zeroes": "off",
+            "active": true,
             "image": {
                 "virtual-size": 67108864,
                 "filename": "TEST_DIR/t.IMGFMT.base",
@@ -350,6 +357,7 @@ wrote 65536/65536 bytes at offset 1048576
         {
             "iops_rd": 0,
             "detect_zeroes": "off",
+            "active": true,
             "image": {
                 "virtual-size": 393216,
                 "filename": "TEST_DIR/t.IMGFMT.base",
@@ -521,6 +529,7 @@ wrote 65536/65536 bytes at offset 1048576
         {
             "iops_rd": 0,
             "detect_zeroes": "off",
+            "active": true,
             "image": {
                 "backing-image": {
                     "virtual-size": 67108864,
@@ -562,6 +571,7 @@ wrote 65536/65536 bytes at offset 1048576
         {
             "iops_rd": 0,
             "detect_zeroes": "off",
+            "active": true,
             "image": {
                 "virtual-size": 197120,
                 "filename": "TEST_DIR/t.IMGFMT.ovl2",
@@ -590,6 +600,7 @@ wrote 65536/65536 bytes at offset 1048576
         {
             "iops_rd": 0,
             "detect_zeroes": "off",
+            "active": true,
             "image": {
                 "backing-image": {
                     "backing-image": {
@@ -642,6 +653,7 @@ wrote 65536/65536 bytes at offset 1048576
         {
             "iops_rd": 0,
             "detect_zeroes": "off",
+            "active": true,
             "image": {
                 "virtual-size": 197120,
                 "filename": "TEST_DIR/t.IMGFMT.ovl3",
@@ -670,6 +682,7 @@ wrote 65536/65536 bytes at offset 1048576
         {
             "iops_rd": 0,
             "detect_zeroes": "off",
+            "active": true,
             "image": {
                 "virtual-size": 67108864,
                 "filename": "TEST_DIR/t.IMGFMT.base",
@@ -699,6 +712,7 @@ wrote 65536/65536 bytes at offset 1048576
         {
             "iops_rd": 0,
             "detect_zeroes": "off",
+            "active": true,
             "image": {
                 "virtual-size": 393216,
                 "filename": "TEST_DIR/t.IMGFMT.base",
@@ -727,6 +741,7 @@ wrote 65536/65536 bytes at offset 1048576
         {
             "iops_rd": 0,
             "detect_zeroes": "off",
+            "active": true,
             "image": {
                 "backing-image": {
                     "virtual-size": 67108864,
@@ -768,6 +783,7 @@ wrote 65536/65536 bytes at offset 1048576
         {
             "iops_rd": 0,
             "detect_zeroes": "off",
+            "active": true,
             "image": {
                 "virtual-size": 197120,
                 "filename": "TEST_DIR/t.IMGFMT",

tests/qemu-iotests/273.out

@@ -23,6 +23,7 @@ Testing: -blockdev file,node-name=base,filename=TEST_DIR/t.IMGFMT.base -blockdev
         {
             "iops_rd": 0,
             "detect_zeroes": "off",
+            "active": true,
             "image": {
                 "backing-image": {
                     "backing-image": {
@@ -74,6 +75,7 @@ Testing: -blockdev file,node-name=base,filename=TEST_DIR/t.IMGFMT.base -blockdev
         {
             "iops_rd": 0,
             "detect_zeroes": "off",
+            "active": true,
             "image": {
                 "virtual-size": 197120,
                 "filename": "TEST_DIR/t.IMGFMT",
@@ -102,6 +104,7 @@ Testing: -blockdev file,node-name=base,filename=TEST_DIR/t.IMGFMT.base -blockdev
         {
             "iops_rd": 0,
             "detect_zeroes": "off",
+            "active": true,
             "image": {
                 "backing-image": {
                     "virtual-size": 197120,
@@ -142,6 +145,7 @@ Testing: -blockdev file,node-name=base,filename=TEST_DIR/t.IMGFMT.base -blockdev
         {
             "iops_rd": 0,
             "detect_zeroes": "off",
+            "active": true,
             "image": {
                 "virtual-size": 197120,
                 "filename": "TEST_DIR/t.IMGFMT.mid",
@@ -170,6 +174,7 @@ Testing: -blockdev file,node-name=base,filename=TEST_DIR/t.IMGFMT.base -blockdev
         {
             "iops_rd": 0,
             "detect_zeroes": "off",
+            "active": true,
             "image": {
                 "virtual-size": 197120,
                 "filename": "TEST_DIR/t.IMGFMT.base",

tests/qemu-iotests/iotests.py

@@ -701,6 +701,10 @@ def filter_qmp_imgfmt(qmsg):
 def filter_nbd_exports(output: str) -> str:
     return re.sub(r'((min|opt|max) block): [0-9]+', r'\1: XXX', output)
 
+def filter_qtest(output: str) -> str:
+    output = re.sub(r'^\[I \d+\.\d+\] OPENED\n', '', output)
+    output = re.sub(r'\n?\[I \+\d+\.\d+\] CLOSED\n?$', '', output)
+    return output
 
 Msg = TypeVar('Msg', Dict[str, Any], List[Any], str)
 
@@ -909,6 +913,10 @@ class VM(qtest.QEMUQtestMachine):
         self._args.append(addr)
         return self
 
+    def add_paused(self):
+        self._args.append('-S')
+        return self
+
     def hmp(self, command_line: str, use_log: bool = False) -> QMPMessage:
         cmd = 'human-monitor-command'
         kwargs: Dict[str, Any] = {'command-line': command_line}

tests/qemu-iotests/tests/copy-before-write

@@ -95,8 +95,7 @@ class TestCbwError(iotests.QMPTestCase):
 
         self.vm.shutdown()
         log = self.vm.get_log()
-        log = re.sub(r'^\[I \d+\.\d+\] OPENED\n', '', log)
+        log = iotests.filter_qtest(log)
-        log = re.sub(r'\[I \+\d+\.\d+\] CLOSED\n?$', '', log)
         log = iotests.filter_qemu_io(log)
         return log
 

tests/qemu-iotests/tests/inactive-node-nbd

@@ -0,0 +1,303 @@
+#!/usr/bin/env python3
+# group: rw quick
+#
+# Copyright (C) Red Hat, Inc.
+#
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 2 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program.  If not, see <http://www.gnu.org/licenses/>.
+#
+# Creator/Owner: Kevin Wolf <kwolf@redhat.com>
+
+import iotests
+
+from iotests import QemuIoInteractive
+from iotests import filter_qemu_io, filter_qtest, filter_qmp_testfiles
+
+iotests.script_initialize(supported_fmts=['generic'],
+                          supported_protocols=['file'],
+                          supported_platforms=['linux'])
+
+def get_export(node_name='disk-fmt', allow_inactive=None):
+    exp = {
+        'id': 'exp0',
+        'type': 'nbd',
+        'node-name': node_name,
+        'writable': True,
+    }
+
+    if allow_inactive is not None:
+        exp['allow-inactive'] = allow_inactive
+
+    return exp
+
+def node_is_active(_vm, node_name):
+    nodes = _vm.cmd('query-named-block-nodes', flat=True)
+    node = next(n for n in nodes if n['node-name'] == node_name)
+    return node['active']
+
+with iotests.FilePath('disk.img') as path, \
+     iotests.FilePath('snap.qcow2') as snap_path, \
+     iotests.FilePath('snap2.qcow2') as snap2_path, \
+     iotests.FilePath('target.img') as target_path, \
+     iotests.FilePath('nbd.sock', base_dir=iotests.sock_dir) as nbd_sock, \
+     iotests.VM() as vm:
+
+    img_size = '10M'
+
+    iotests.log('Preparing disk...')
+    iotests.qemu_img_create('-f', iotests.imgfmt, path, img_size)
+    iotests.qemu_img_create('-f', iotests.imgfmt, target_path, img_size)
+
+    iotests.qemu_img_create('-f', 'qcow2', '-b', path, '-F', iotests.imgfmt,
+                            snap_path)
+    iotests.qemu_img_create('-f', 'qcow2', '-b', snap_path, '-F', 'qcow2',
+                            snap2_path)
+
+    iotests.log('Launching VM...')
+    vm.add_blockdev(f'file,node-name=disk-file,filename={path}')
+    vm.add_blockdev(f'{iotests.imgfmt},file=disk-file,node-name=disk-fmt,'
+                     'active=off')
+    vm.add_blockdev(f'file,node-name=target-file,filename={target_path}')
+    vm.add_blockdev(f'{iotests.imgfmt},file=target-file,node-name=target-fmt')
+    vm.add_blockdev(f'file,node-name=snap-file,filename={snap_path}')
+    vm.add_blockdev(f'file,node-name=snap2-file,filename={snap2_path}')
+
+    # Actually running the VM activates all images
+    vm.add_paused()
+
+    vm.launch()
+    vm.qmp_log('nbd-server-start',
+                addr={'type': 'unix', 'data':{'path': nbd_sock}},
+                filters=[filter_qmp_testfiles])
+
+    iotests.log('\n=== Creating export of inactive node ===')
+
+    iotests.log('\nExports activate nodes without allow-inactive')
+    iotests.log('disk-fmt active: %s' % node_is_active(vm, 'disk-fmt'))
+    vm.qmp_log('block-export-add', **get_export())
+    iotests.log('disk-fmt active: %s' % node_is_active(vm, 'disk-fmt'))
+    vm.qmp_log('query-block-exports')
+    vm.qmp_log('block-export-del', id='exp0')
+    vm.event_wait('BLOCK_EXPORT_DELETED')
+    vm.qmp_log('query-block-exports')
+
+    iotests.log('\nExports activate nodes with allow-inactive=false')
+    vm.qmp_log('blockdev-set-active', node_name='disk-fmt', active=False)
+    iotests.log('disk-fmt active: %s' % node_is_active(vm, 'disk-fmt'))
+    vm.qmp_log('block-export-add', **get_export(allow_inactive=False))
+    iotests.log('disk-fmt active: %s' % node_is_active(vm, 'disk-fmt'))
+    vm.qmp_log('query-block-exports')
+    vm.qmp_log('block-export-del', id='exp0')
+    vm.event_wait('BLOCK_EXPORT_DELETED')
+    vm.qmp_log('query-block-exports')
+
+    iotests.log('\nExport leaves nodes inactive with allow-inactive=true')
+    vm.qmp_log('blockdev-set-active', node_name='disk-fmt', active=False)
+    iotests.log('disk-fmt active: %s' % node_is_active(vm, 'disk-fmt'))
+    vm.qmp_log('block-export-add', **get_export(allow_inactive=True))
+    iotests.log('disk-fmt active: %s' % node_is_active(vm, 'disk-fmt'))
+    vm.qmp_log('query-block-exports')
+    vm.qmp_log('block-export-del', id='exp0')
+    vm.event_wait('BLOCK_EXPORT_DELETED')
+    vm.qmp_log('query-block-exports')
+
+    iotests.log('\n=== Inactivating node with existing export ===')
+
+    iotests.log('\nInactivating nodes with an export fails without '
+                'allow-inactive')
+    vm.qmp_log('blockdev-set-active', node_name='disk-fmt', active=True)
+    vm.qmp_log('block-export-add', **get_export(node_name='disk-fmt'))
+    vm.qmp_log('blockdev-set-active', node_name='disk-fmt', active=False)
+    iotests.log('disk-fmt active: %s' % node_is_active(vm, 'disk-fmt'))
+    vm.qmp_log('query-block-exports')
+    vm.qmp_log('block-export-del', id='exp0')
+    vm.event_wait('BLOCK_EXPORT_DELETED')
+    vm.qmp_log('query-block-exports')
+
+    iotests.log('\nInactivating nodes with an export fails with '
+                'allow-inactive=false')
+    vm.qmp_log('blockdev-set-active', node_name='disk-fmt', active=True)
+    vm.qmp_log('block-export-add',
+               **get_export(node_name='disk-fmt', allow_inactive=False))
+    vm.qmp_log('blockdev-set-active', node_name='disk-fmt', active=False)
+    iotests.log('disk-fmt active: %s' % node_is_active(vm, 'disk-fmt'))
+    vm.qmp_log('query-block-exports')
+    vm.qmp_log('block-export-del', id='exp0')
+    vm.event_wait('BLOCK_EXPORT_DELETED')
+    vm.qmp_log('query-block-exports')
+
+    iotests.log('\nInactivating nodes with an export works with '
+                'allow-inactive=true')
+    vm.qmp_log('blockdev-set-active', node_name='disk-fmt', active=True)
+    vm.qmp_log('block-export-add',
+               **get_export(node_name='disk-fmt', allow_inactive=True))
+    vm.qmp_log('blockdev-set-active', node_name='disk-fmt', active=False)
+    iotests.log('disk-fmt active: %s' % node_is_active(vm, 'disk-fmt'))
+    vm.qmp_log('query-block-exports')
+    vm.qmp_log('block-export-del', id='exp0')
+    vm.event_wait('BLOCK_EXPORT_DELETED')
+    vm.qmp_log('query-block-exports')
+
+    iotests.log('\n=== Inactive nodes with parent ===')
+
+    iotests.log('\nInactivating nodes with an active parent fails')
+    vm.qmp_log('blockdev-set-active', node_name='disk-fmt', active=True)
+    vm.qmp_log('blockdev-set-active', node_name='disk-file', active=False)
+    iotests.log('disk-file active: %s' % node_is_active(vm, 'disk-file'))
+    iotests.log('disk-fmt active: %s' % node_is_active(vm, 'disk-fmt'))
+
+    iotests.log('\nInactivating nodes with an inactive parent works')
+    vm.qmp_log('blockdev-set-active', node_name='disk-fmt', active=False)
+    vm.qmp_log('blockdev-set-active', node_name='disk-file', active=False)
+    iotests.log('disk-file active: %s' % node_is_active(vm, 'disk-file'))
+    iotests.log('disk-fmt active: %s' % node_is_active(vm, 'disk-fmt'))
+
+    iotests.log('\nCreating active parent node with an inactive child fails')
+    vm.qmp_log('blockdev-add', driver='raw', file='disk-fmt',
+               node_name='disk-filter')
+    vm.qmp_log('blockdev-add', driver='raw', file='disk-fmt',
+               node_name='disk-filter', active=True)
+
+    iotests.log('\nCreating inactive parent node with an inactive child works')
+    vm.qmp_log('blockdev-add', driver='raw', file='disk-fmt',
+               node_name='disk-filter', active=False)
+    vm.qmp_log('blockdev-del', node_name='disk-filter')
+
+    iotests.log('\n=== Resizing an inactive node ===')
+    vm.qmp_log('block_resize', node_name='disk-fmt', size=16*1024*1024)
+
+    iotests.log('\n=== Taking a snapshot of an inactive node ===')
+
+    iotests.log('\nActive overlay over inactive backing file automatically '
+                'makes both inactive for compatibility')
+    vm.qmp_log('blockdev-add', driver='qcow2', node_name='snap-fmt',
+               file='snap-file', backing=None)
+    iotests.log('disk-fmt active: %s' % node_is_active(vm, 'disk-fmt'))
+    iotests.log('snap-fmt active: %s' % node_is_active(vm, 'snap-fmt'))
+    vm.qmp_log('blockdev-snapshot', node='disk-fmt', overlay='snap-fmt')
+    iotests.log('disk-fmt active: %s' % node_is_active(vm, 'disk-fmt'))
+    iotests.log('snap-fmt active: %s' % node_is_active(vm, 'snap-fmt'))
+    vm.qmp_log('blockdev-del', node_name='snap-fmt')
+
+    iotests.log('\nInactive overlay over inactive backing file just works')
+    vm.qmp_log('blockdev-add', driver='qcow2', node_name='snap-fmt',
+               file='snap-file', backing=None, active=False)
+    vm.qmp_log('blockdev-snapshot', node='disk-fmt', overlay='snap-fmt')
+
+    iotests.log('\n=== Block jobs with inactive nodes ===')
+
+    iotests.log('\nStreaming into an inactive node')
+    vm.qmp_log('block-stream', device='snap-fmt',
+               filters=[iotests.filter_qmp_generated_node_ids])
+
+    iotests.log('\nCommitting an inactive root node (active commit)')
+    vm.qmp_log('block-commit', job_id='job0', device='snap-fmt',
+               filters=[iotests.filter_qmp_generated_node_ids])
+
+    iotests.log('\nCommitting an inactive intermediate node to inactive base')
+    vm.qmp_log('blockdev-add', driver='qcow2', node_name='snap2-fmt',
+               file='snap2-file', backing='snap-fmt', active=False)
+
+    iotests.log('disk-fmt active: %s' % node_is_active(vm, 'disk-fmt'))
+    iotests.log('snap-fmt active: %s' % node_is_active(vm, 'snap-fmt'))
+    iotests.log('snap2-fmt active: %s' % node_is_active(vm, 'snap2-fmt'))
+
+    vm.qmp_log('block-commit', job_id='job0', device='snap2-fmt',
+               top_node='snap-fmt',
+               filters=[iotests.filter_qmp_generated_node_ids])
+
+    iotests.log('\nCommitting an inactive intermediate node to active base')
+    vm.qmp_log('blockdev-set-active', node_name='disk-fmt', active=True)
+    vm.qmp_log('block-commit', job_id='job0', device='snap2-fmt',
+               top_node='snap-fmt',
+               filters=[iotests.filter_qmp_generated_node_ids])
+
+    iotests.log('\nMirror from inactive source to active target')
+    vm.qmp_log('blockdev-mirror', job_id='job0', device='snap2-fmt',
+               target='target-fmt', sync='full',
+               filters=[iotests.filter_qmp_generated_node_ids])
+
+    iotests.log('\nMirror from active source to inactive target')
+
+    iotests.log('disk-fmt active: %s' % node_is_active(vm, 'disk-fmt'))
+    iotests.log('snap-fmt active: %s' % node_is_active(vm, 'snap-fmt'))
+    iotests.log('snap2-fmt active: %s' % node_is_active(vm, 'snap2-fmt'))
+    iotests.log('target-fmt active: %s' % node_is_active(vm, 'target-fmt'))
+
+    # Activating snap2-fmt recursively activates the whole backing chain
+    vm.qmp_log('blockdev-set-active', node_name='snap2-fmt', active=True)
+    vm.qmp_log('blockdev-set-active', node_name='target-fmt', active=False)
+
+    iotests.log('disk-fmt active: %s' % node_is_active(vm, 'disk-fmt'))
+    iotests.log('snap-fmt active: %s' % node_is_active(vm, 'snap-fmt'))
+    iotests.log('snap2-fmt active: %s' % node_is_active(vm, 'snap2-fmt'))
+    iotests.log('target-fmt active: %s' % node_is_active(vm, 'target-fmt'))
+
+    vm.qmp_log('blockdev-mirror', job_id='job0', device='snap2-fmt',
+               target='target-fmt', sync='full',
+               filters=[iotests.filter_qmp_generated_node_ids])
+
+    iotests.log('\nBackup from active source to inactive target')
+
+    vm.qmp_log('blockdev-backup', job_id='job0', device='snap2-fmt',
+               target='target-fmt', sync='full',
+               filters=[iotests.filter_qmp_generated_node_ids])
+
+    iotests.log('\nBackup from inactive source to active target')
+
+    # Inactivating snap2-fmt recursively inactivates the whole backing chain
+    vm.qmp_log('blockdev-set-active', node_name='snap2-fmt', active=False)
+    vm.qmp_log('blockdev-set-active', node_name='target-fmt', active=True)
+
+    iotests.log('disk-fmt active: %s' % node_is_active(vm, 'disk-fmt'))
+    iotests.log('snap-fmt active: %s' % node_is_active(vm, 'snap-fmt'))
+    iotests.log('snap2-fmt active: %s' % node_is_active(vm, 'snap2-fmt'))
+    iotests.log('target-fmt active: %s' % node_is_active(vm, 'target-fmt'))
+
+    vm.qmp_log('blockdev-backup', job_id='job0', device='snap2-fmt',
+               target='target-fmt', sync='full',
+               filters=[iotests.filter_qmp_generated_node_ids])
+
+    iotests.log('\n=== Accessing export on inactive node ===')
+
+    # Use the target node because it has the right image format and isn't the
+    # (read-only) backing file of a qcow2 node
+    vm.qmp_log('blockdev-set-active', node_name='target-fmt', active=False)
+    vm.qmp_log('block-export-add',
+               **get_export(node_name='target-fmt', allow_inactive=True))
+
+    # The read should succeed, everything else should fail gracefully
+    qemu_io = QemuIoInteractive('-f', 'raw',
+                                f'nbd+unix:///target-fmt?socket={nbd_sock}')
+    iotests.log(qemu_io.cmd('read 0 64k'), filters=[filter_qemu_io])
+    iotests.log(qemu_io.cmd('write 0 64k'), filters=[filter_qemu_io])
+    iotests.log(qemu_io.cmd('write -z 0 64k'), filters=[filter_qemu_io])
+    iotests.log(qemu_io.cmd('write -zu 0 64k'), filters=[filter_qemu_io])
+    iotests.log(qemu_io.cmd('discard 0 64k'), filters=[filter_qemu_io])
+    iotests.log(qemu_io.cmd('flush'), filters=[filter_qemu_io])
+    iotests.log(qemu_io.cmd('map'), filters=[filter_qemu_io])
+    qemu_io.close()
+
+    iotests.log('\n=== Resuming VM activates all images ===')
+    vm.qmp_log('cont')
+
+    iotests.log('disk-fmt active: %s' % node_is_active(vm, 'disk-fmt'))
+    iotests.log('snap-fmt active: %s' % node_is_active(vm, 'snap-fmt'))
+    iotests.log('snap2-fmt active: %s' % node_is_active(vm, 'snap2-fmt'))
+    iotests.log('target-fmt active: %s' % node_is_active(vm, 'target-fmt'))
+
+    iotests.log('\nShutting down...')
+    vm.shutdown()
+    log = vm.get_log()
+    if log:
+        iotests.log(log, [filter_qtest, filter_qemu_io])

tests/qemu-iotests/tests/inactive-node-nbd.out

@@ -0,0 +1,239 @@
+Preparing disk...
+Launching VM...
+{"execute": "nbd-server-start", "arguments": {"addr": {"data": {"path": "SOCK_DIR/PID-nbd.sock"}, "type": "unix"}}}
+{"return": {}}
+
+=== Creating export of inactive node ===
+
+Exports activate nodes without allow-inactive
+disk-fmt active: False
+{"execute": "block-export-add", "arguments": {"id": "exp0", "node-name": "disk-fmt", "type": "nbd", "writable": true}}
+{"return": {}}
+disk-fmt active: True
+{"execute": "query-block-exports", "arguments": {}}
+{"return": [{"id": "exp0", "node-name": "disk-fmt", "shutting-down": false, "type": "nbd"}]}
+{"execute": "block-export-del", "arguments": {"id": "exp0"}}
+{"return": {}}
+{"execute": "query-block-exports", "arguments": {}}
+{"return": []}
+
+Exports activate nodes with allow-inactive=false
+{"execute": "blockdev-set-active", "arguments": {"active": false, "node-name": "disk-fmt"}}
+{"return": {}}
+disk-fmt active: False
+{"execute": "block-export-add", "arguments": {"allow-inactive": false, "id": "exp0", "node-name": "disk-fmt", "type": "nbd", "writable": true}}
+{"return": {}}
+disk-fmt active: True
+{"execute": "query-block-exports", "arguments": {}}
+{"return": [{"id": "exp0", "node-name": "disk-fmt", "shutting-down": false, "type": "nbd"}]}
+{"execute": "block-export-del", "arguments": {"id": "exp0"}}
+{"return": {}}
+{"execute": "query-block-exports", "arguments": {}}
+{"return": []}
+
+Export leaves nodes inactive with allow-inactive=true
+{"execute": "blockdev-set-active", "arguments": {"active": false, "node-name": "disk-fmt"}}
+{"return": {}}
+disk-fmt active: False
+{"execute": "block-export-add", "arguments": {"allow-inactive": true, "id": "exp0", "node-name": "disk-fmt", "type": "nbd", "writable": true}}
+{"return": {}}
+disk-fmt active: False
+{"execute": "query-block-exports", "arguments": {}}
+{"return": [{"id": "exp0", "node-name": "disk-fmt", "shutting-down": false, "type": "nbd"}]}
+{"execute": "block-export-del", "arguments": {"id": "exp0"}}
+{"return": {}}
+{"execute": "query-block-exports", "arguments": {}}
+{"return": []}
+
+=== Inactivating node with existing export ===
+
+Inactivating nodes with an export fails without allow-inactive
+{"execute": "blockdev-set-active", "arguments": {"active": true, "node-name": "disk-fmt"}}
+{"return": {}}
+{"execute": "block-export-add", "arguments": {"id": "exp0", "node-name": "disk-fmt", "type": "nbd", "writable": true}}
+{"return": {}}
+{"execute": "blockdev-set-active", "arguments": {"active": false, "node-name": "disk-fmt"}}
+{"error": {"class": "GenericError", "desc": "Failed to inactivate node: Operation not permitted"}}
+disk-fmt active: True
+{"execute": "query-block-exports", "arguments": {}}
+{"return": [{"id": "exp0", "node-name": "disk-fmt", "shutting-down": false, "type": "nbd"}]}
+{"execute": "block-export-del", "arguments": {"id": "exp0"}}
+{"return": {}}
+{"execute": "query-block-exports", "arguments": {}}
+{"return": []}
+
+Inactivating nodes with an export fails with allow-inactive=false
+{"execute": "blockdev-set-active", "arguments": {"active": true, "node-name": "disk-fmt"}}
+{"return": {}}
+{"execute": "block-export-add", "arguments": {"allow-inactive": false, "id": "exp0", "node-name": "disk-fmt", "type": "nbd", "writable": true}}
+{"return": {}}
+{"execute": "blockdev-set-active", "arguments": {"active": false, "node-name": "disk-fmt"}}
+{"error": {"class": "GenericError", "desc": "Failed to inactivate node: Operation not permitted"}}
+disk-fmt active: True
+{"execute": "query-block-exports", "arguments": {}}
+{"return": [{"id": "exp0", "node-name": "disk-fmt", "shutting-down": false, "type": "nbd"}]}
+{"execute": "block-export-del", "arguments": {"id": "exp0"}}
+{"return": {}}
+{"execute": "query-block-exports", "arguments": {}}
+{"return": []}
+
+Inactivating nodes with an export works with allow-inactive=true
+{"execute": "blockdev-set-active", "arguments": {"active": true, "node-name": "disk-fmt"}}
+{"return": {}}
+{"execute": "block-export-add", "arguments": {"allow-inactive": true, "id": "exp0", "node-name": "disk-fmt", "type": "nbd", "writable": true}}
+{"return": {}}
+{"execute": "blockdev-set-active", "arguments": {"active": false, "node-name": "disk-fmt"}}
+{"return": {}}
+disk-fmt active: False
+{"execute": "query-block-exports", "arguments": {}}
+{"return": [{"id": "exp0", "node-name": "disk-fmt", "shutting-down": false, "type": "nbd"}]}
+{"execute": "block-export-del", "arguments": {"id": "exp0"}}
+{"return": {}}
+{"execute": "query-block-exports", "arguments": {}}
+{"return": []}
+
+=== Inactive nodes with parent ===
+
+Inactivating nodes with an active parent fails
+{"execute": "blockdev-set-active", "arguments": {"active": true, "node-name": "disk-fmt"}}
+{"return": {}}
+{"execute": "blockdev-set-active", "arguments": {"active": false, "node-name": "disk-file"}}
+{"error": {"class": "GenericError", "desc": "Node has active parent node"}}
+disk-file active: True
+disk-fmt active: True
+
+Inactivating nodes with an inactive parent works
+{"execute": "blockdev-set-active", "arguments": {"active": false, "node-name": "disk-fmt"}}
+{"return": {}}
+{"execute": "blockdev-set-active", "arguments": {"active": false, "node-name": "disk-file"}}
+{"return": {}}
+disk-file active: False
+disk-fmt active: False
+
+Creating active parent node with an inactive child fails
+{"execute": "blockdev-add", "arguments": {"driver": "raw", "file": "disk-fmt", "node-name": "disk-filter"}}
+{"error": {"class": "GenericError", "desc": "Inactive 'disk-fmt' can't be a file child of active 'disk-filter'"}}
+{"execute": "blockdev-add", "arguments": {"active": true, "driver": "raw", "file": "disk-fmt", "node-name": "disk-filter"}}
+{"error": {"class": "GenericError", "desc": "Inactive 'disk-fmt' can't be a file child of active 'disk-filter'"}}
+
+Creating inactive parent node with an inactive child works
+{"execute": "blockdev-add", "arguments": {"active": false, "driver": "raw", "file": "disk-fmt", "node-name": "disk-filter"}}
+{"return": {}}
+{"execute": "blockdev-del", "arguments": {"node-name": "disk-filter"}}
+{"return": {}}
+
+=== Resizing an inactive node ===
+{"execute": "block_resize", "arguments": {"node-name": "disk-fmt", "size": 16777216}}
+{"error": {"class": "GenericError", "desc": "Permission 'resize' unavailable on inactive node"}}
+
+=== Taking a snapshot of an inactive node ===
+
+Active overlay over inactive backing file automatically makes both inactive for compatibility
+{"execute": "blockdev-add", "arguments": {"backing": null, "driver": "qcow2", "file": "snap-file", "node-name": "snap-fmt"}}
+{"return": {}}
+disk-fmt active: False
+snap-fmt active: True
+{"execute": "blockdev-snapshot", "arguments": {"node": "disk-fmt", "overlay": "snap-fmt"}}
+{"return": {}}
+disk-fmt active: False
+snap-fmt active: False
+{"execute": "blockdev-del", "arguments": {"node-name": "snap-fmt"}}
+{"return": {}}
+
+Inactive overlay over inactive backing file just works
+{"execute": "blockdev-add", "arguments": {"active": false, "backing": null, "driver": "qcow2", "file": "snap-file", "node-name": "snap-fmt"}}
+{"return": {}}
+{"execute": "blockdev-snapshot", "arguments": {"node": "disk-fmt", "overlay": "snap-fmt"}}
+{"return": {}}
+
+=== Block jobs with inactive nodes ===
+
+Streaming into an inactive node
+{"execute": "block-stream", "arguments": {"device": "snap-fmt"}}
+{"error": {"class": "GenericError", "desc": "Could not create node: Inactive 'snap-fmt' can't be a file child of active 'NODE_NAME'"}}
+
+Committing an inactive root node (active commit)
+{"execute": "block-commit", "arguments": {"device": "snap-fmt", "job-id": "job0"}}
+{"error": {"class": "GenericError", "desc": "Inactive 'snap-fmt' can't be a backing child of active 'NODE_NAME'"}}
+
+Committing an inactive intermediate node to inactive base
+{"execute": "blockdev-add", "arguments": {"active": false, "backing": "snap-fmt", "driver": "qcow2", "file": "snap2-file", "node-name": "snap2-fmt"}}
+{"return": {}}
+disk-fmt active: False
+snap-fmt active: False
+snap2-fmt active: False
+{"execute": "block-commit", "arguments": {"device": "snap2-fmt", "job-id": "job0", "top-node": "snap-fmt"}}
+{"error": {"class": "GenericError", "desc": "Inactive 'snap-fmt' can't be a backing child of active 'NODE_NAME'"}}
+
+Committing an inactive intermediate node to active base
+{"execute": "blockdev-set-active", "arguments": {"active": true, "node-name": "disk-fmt"}}
+{"return": {}}
+{"execute": "block-commit", "arguments": {"device": "snap2-fmt", "job-id": "job0", "top-node": "snap-fmt"}}
+{"error": {"class": "GenericError", "desc": "Inactive 'snap-fmt' can't be a backing child of active 'NODE_NAME'"}}
+
+Mirror from inactive source to active target
+{"execute": "blockdev-mirror", "arguments": {"device": "snap2-fmt", "job-id": "job0", "sync": "full", "target": "target-fmt"}}
+{"error": {"class": "GenericError", "desc": "Inactive 'snap2-fmt' can't be a backing child of active 'NODE_NAME'"}}
+
+Mirror from active source to inactive target
+disk-fmt active: True
+snap-fmt active: False
+snap2-fmt active: False
+target-fmt active: True
+{"execute": "blockdev-set-active", "arguments": {"active": true, "node-name": "snap2-fmt"}}
+{"return": {}}
+{"execute": "blockdev-set-active", "arguments": {"active": false, "node-name": "target-fmt"}}
+{"return": {}}
+disk-fmt active: True
+snap-fmt active: True
+snap2-fmt active: True
+target-fmt active: False
+{"execute": "blockdev-mirror", "arguments": {"device": "snap2-fmt", "job-id": "job0", "sync": "full", "target": "target-fmt"}}
+{"error": {"class": "GenericError", "desc": "Permission 'write' unavailable on inactive node"}}
+
+Backup from active source to inactive target
+{"execute": "blockdev-backup", "arguments": {"device": "snap2-fmt", "job-id": "job0", "sync": "full", "target": "target-fmt"}}
+{"error": {"class": "GenericError", "desc": "Could not create node: Inactive 'target-fmt' can't be a target child of active 'NODE_NAME'"}}
+
+Backup from inactive source to active target
+{"execute": "blockdev-set-active", "arguments": {"active": false, "node-name": "snap2-fmt"}}
+{"return": {}}
+{"execute": "blockdev-set-active", "arguments": {"active": true, "node-name": "target-fmt"}}
+{"return": {}}
+disk-fmt active: False
+snap-fmt active: False
+snap2-fmt active: False
+target-fmt active: True
+{"execute": "blockdev-backup", "arguments": {"device": "snap2-fmt", "job-id": "job0", "sync": "full", "target": "target-fmt"}}
+{"error": {"class": "GenericError", "desc": "Could not create node: Inactive 'snap2-fmt' can't be a file child of active 'NODE_NAME'"}}
+
+=== Accessing export on inactive node ===
+{"execute": "blockdev-set-active", "arguments": {"active": false, "node-name": "target-fmt"}}
+{"return": {}}
+{"execute": "block-export-add", "arguments": {"allow-inactive": true, "id": "exp0", "node-name": "target-fmt", "type": "nbd", "writable": true}}
+{"return": {}}
+read 65536/65536 bytes at offset 0
+64 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+
+write failed: Operation not permitted
+
+write failed: Operation not permitted
+
+write failed: Operation not permitted
+
+discard failed: Operation not permitted
+
+
+qemu-io: Failed to get allocation status: Operation not permitted
+
+
+=== Resuming VM activates all images ===
+{"execute": "cont", "arguments": {}}
+{"return": {}}
+disk-fmt active: True
+snap-fmt active: True
+snap2-fmt active: True
+target-fmt active: True
+
+Shutting down...
+

tests/qemu-iotests/tests/migrate-bitmaps-test

@@ -122,11 +122,10 @@ class TestDirtyBitmapMigration(iotests.QMPTestCase):
 
         # catch 'Could not reopen qcow2 layer: Bitmap already exists'
         # possible error
-        log = self.vm_a.get_log()
+        log = iotests.filter_qtest(self.vm_a.get_log())
-        log = re.sub(r'^\[I \d+\.\d+\] OPENED\n', '', log)
+        log = re.sub(r'^(wrote .* bytes at offset .*\n'
-        log = re.sub(r'^(wrote .* bytes at offset .*\n.*KiB.*ops.*sec.*\n){3}',
+                     r'.*KiB.*ops.*sec.*\n?){3}',
                      '', log)
-        log = re.sub(r'\[I \+\d+\.\d+\] CLOSED\n?$', '', log)
         self.assertEqual(log, '')
 
         # test that bitmap is still persistent

tests/qemu-iotests/tests/qsd-migrate

@@ -0,0 +1,140 @@
+#!/usr/bin/env python3
+# group: rw quick
+#
+# Copyright (C) Red Hat, Inc.
+#
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 2 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program.  If not, see <http://www.gnu.org/licenses/>.
+#
+# Creator/Owner: Kevin Wolf <kwolf@redhat.com>
+
+import iotests
+
+from iotests import filter_qemu_io, filter_qtest
+
+iotests.script_initialize(supported_fmts=['generic'],
+                          supported_protocols=['file'],
+                          supported_platforms=['linux'])
+
+with iotests.FilePath('disk.img') as path, \
+     iotests.FilePath('nbd-src.sock', base_dir=iotests.sock_dir) as nbd_src, \
+     iotests.FilePath('nbd-dst.sock', base_dir=iotests.sock_dir) as nbd_dst, \
+     iotests.FilePath('migrate.sock', base_dir=iotests.sock_dir) as mig_sock, \
+     iotests.VM(path_suffix="-src") as vm_src, \
+     iotests.VM(path_suffix="-dst") as vm_dst:
+
+    img_size = '10M'
+
+    iotests.log('Preparing disk...')
+    iotests.qemu_img_create('-f', iotests.imgfmt, path, img_size)
+
+    iotests.log('Launching source QSD...')
+    qsd_src = iotests.QemuStorageDaemon(
+        '--blockdev', f'file,node-name=disk-file,filename={path}',
+        '--blockdev', f'{iotests.imgfmt},file=disk-file,node-name=disk-fmt',
+        '--nbd-server', f'addr.type=unix,addr.path={nbd_src}',
+        '--export', 'nbd,id=exp0,node-name=disk-fmt,writable=true,'
+                    'allow-inactive=true',
+        qmp=True,
+    )
+
+    iotests.log('Launching source VM...')
+    vm_src.add_args('-blockdev', f'nbd,node-name=disk,server.type=unix,'
+                                 f'server.path={nbd_src},export=disk-fmt')
+    vm_src.add_args('-device', 'virtio-blk,drive=disk,id=virtio0')
+    vm_src.launch()
+
+    iotests.log('Launching destination QSD...')
+    qsd_dst = iotests.QemuStorageDaemon(
+        '--blockdev', f'file,node-name=disk-file,filename={path},active=off',
+        '--blockdev', f'{iotests.imgfmt},file=disk-file,node-name=disk-fmt,'
+                      f'active=off',
+        '--nbd-server', f'addr.type=unix,addr.path={nbd_dst}',
+        '--export', 'nbd,id=exp0,node-name=disk-fmt,writable=true,'
+                    'allow-inactive=true',
+        qmp=True,
+        instance_id='b',
+    )
+
+    iotests.log('Launching destination VM...')
+    vm_dst.add_args('-blockdev', f'nbd,node-name=disk,server.type=unix,'
+                                 f'server.path={nbd_dst},export=disk-fmt')
+    vm_dst.add_args('-device', 'virtio-blk,drive=disk,id=virtio0')
+    vm_dst.add_args('-incoming', f'unix:{mig_sock}')
+    vm_dst.launch()
+
+    iotests.log('\nTest I/O on the source')
+    vm_src.hmp_qemu_io('virtio0/virtio-backend', 'write -P 0x11 0 4k',
+                       use_log=True, qdev=True)
+    vm_src.hmp_qemu_io('virtio0/virtio-backend', 'read -P 0x11 0 4k',
+                       use_log=True, qdev=True)
+
+    iotests.log('\nStarting migration...')
+
+    mig_caps = [
+        {'capability': 'events', 'state': True},
+        {'capability': 'pause-before-switchover', 'state': True},
+    ]
+    vm_src.qmp_log('migrate-set-capabilities', capabilities=mig_caps)
+    vm_dst.qmp_log('migrate-set-capabilities', capabilities=mig_caps)
+    vm_src.qmp_log('migrate', uri=f'unix:{mig_sock}',
+                   filters=[iotests.filter_qmp_testfiles])
+
+    vm_src.event_wait('MIGRATION',
+                      match={'data': {'status': 'pre-switchover'}})
+
+    iotests.log('\nPre-switchover: Reconfigure QSD instances')
+
+    iotests.log(qsd_src.qmp('blockdev-set-active', {'active': False}))
+
+    # Reading is okay from both sides while the image is inactive. Note that
+    # the destination may have stale data until it activates the image, though.
+    vm_src.hmp_qemu_io('virtio0/virtio-backend', 'read -P 0x11 0 4k',
+                       use_log=True, qdev=True)
+    vm_dst.hmp_qemu_io('virtio0/virtio-backend', 'read 0 4k',
+                       use_log=True, qdev=True)
+
+    iotests.log(qsd_dst.qmp('blockdev-set-active', {'active': True}))
+
+    iotests.log('\nCompleting migration...')
+
+    vm_src.qmp_log('migrate-continue', state='pre-switchover')
+    vm_dst.event_wait('MIGRATION', match={'data': {'status': 'completed'}})
+
+    iotests.log('\nTest I/O on the destination')
+
+    # Now the destination must see what the source wrote
+    vm_dst.hmp_qemu_io('virtio0/virtio-backend', 'read -P 0x11 0 4k',
+                       use_log=True, qdev=True)
+
+    # And be able to overwrite it
+    vm_dst.hmp_qemu_io('virtio0/virtio-backend', 'write -P 0x22 0 4k',
+                       use_log=True, qdev=True)
+    vm_dst.hmp_qemu_io('virtio0/virtio-backend', 'read -P 0x22 0 4k',
+                       use_log=True, qdev=True)
+
+    iotests.log('\nDone')
+
+    vm_src.shutdown()
+    iotests.log('\n--- vm_src log ---')
+    log = vm_src.get_log()
+    if log:
+        iotests.log(log, [filter_qtest, filter_qemu_io])
+    qsd_src.stop()
+
+    vm_dst.shutdown()
+    iotests.log('\n--- vm_dst log ---')
+    log = vm_dst.get_log()
+    if log:
+        iotests.log(log, [filter_qtest, filter_qemu_io])
+    qsd_dst.stop()

tests/qemu-iotests/tests/qsd-migrate.out

@@ -0,0 +1,59 @@
+Preparing disk...
+Launching source QSD...
+Launching source VM...
+Launching destination QSD...
+Launching destination VM...
+
+Test I/O on the source
+{"execute": "human-monitor-command", "arguments": {"command-line": "qemu-io -d virtio0/virtio-backend \"write -P 0x11 0 4k\""}}
+{"return": ""}
+{"execute": "human-monitor-command", "arguments": {"command-line": "qemu-io -d virtio0/virtio-backend \"read -P 0x11 0 4k\""}}
+{"return": ""}
+
+Starting migration...
+{"execute": "migrate-set-capabilities", "arguments": {"capabilities": [{"capability": "events", "state": true}, {"capability": "pause-before-switchover", "state": true}]}}
+{"return": {}}
+{"execute": "migrate-set-capabilities", "arguments": {"capabilities": [{"capability": "events", "state": true}, {"capability": "pause-before-switchover", "state": true}]}}
+{"return": {}}
+{"execute": "migrate", "arguments": {"uri": "unix:SOCK_DIR/PID-migrate.sock"}}
+{"return": {}}
+
+Pre-switchover: Reconfigure QSD instances
+{"return": {}}
+{"execute": "human-monitor-command", "arguments": {"command-line": "qemu-io -d virtio0/virtio-backend \"read -P 0x11 0 4k\""}}
+{"return": ""}
+{"execute": "human-monitor-command", "arguments": {"command-line": "qemu-io -d virtio0/virtio-backend \"read 0 4k\""}}
+{"return": ""}
+{"return": {}}
+
+Completing migration...
+{"execute": "migrate-continue", "arguments": {"state": "pre-switchover"}}
+{"return": {}}
+
+Test I/O on the destination
+{"execute": "human-monitor-command", "arguments": {"command-line": "qemu-io -d virtio0/virtio-backend \"read -P 0x11 0 4k\""}}
+{"return": ""}
+{"execute": "human-monitor-command", "arguments": {"command-line": "qemu-io -d virtio0/virtio-backend \"write -P 0x22 0 4k\""}}
+{"return": ""}
+{"execute": "human-monitor-command", "arguments": {"command-line": "qemu-io -d virtio0/virtio-backend \"read -P 0x22 0 4k\""}}
+{"return": ""}
+
+Done
+
+--- vm_src log ---
+wrote 4096/4096 bytes at offset 0
+4 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+read 4096/4096 bytes at offset 0
+4 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+read 4096/4096 bytes at offset 0
+4 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+
+--- vm_dst log ---
+read 4096/4096 bytes at offset 0
+4 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+read 4096/4096 bytes at offset 0
+4 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+wrote 4096/4096 bytes at offset 0
+4 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+read 4096/4096 bytes at offset 0
+4 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)