These changes are the raw update to qemu-2.6.

Collission happened in the following patches:

migration: do cleanup operation after completion(738df5b9)
Bug fix.(1750c932f86)
kvmclock: add a new function to update env->tsc.(b52baab2)

The code provided by the patches was already in the upstreamed
version.

Change-Id: I3cc11841a6a76ae20887b2e245710199e1ea7f9a
Signed-off-by: José Pekkarinen <jose.pekkarinen@nokia.com>

1 files changed, 902 insertions, 187 deletions

diff --git a/qemu/block/block-backend.c b/qemu/block/block-backend.c
index aee8a1202..16c9d5e0f 100644
--- a/qemu/block/block-backend.c
+++ b/qemu/block/block-backend.c
@@ -10,74 +10,105 @@
  * or later.  See the COPYING.LIB file in the top-level directory.
  */
 
+#include "qemu/osdep.h"
 #include "sysemu/block-backend.h"
 #include "block/block_int.h"
+#include "block/blockjob.h"
+#include "block/throttle-groups.h"
 #include "sysemu/blockdev.h"
+#include "sysemu/sysemu.h"
 #include "qapi-event.h"
+#include "qemu/id.h"
 
 /* Number of coroutines to reserve per attached device model */
 #define COROUTINE_POOL_RESERVATION 64
 
+#define NOT_DONE 0x7fffffff /* used while emulated sync operation in progress */
+
+static AioContext *blk_aiocb_get_aio_context(BlockAIOCB *acb);
+
 struct BlockBackend {
     char *name;
     int refcnt;
-    BlockDriverState *bs;
+    BdrvChild *root;
     DriveInfo *legacy_dinfo;    /* null unless created by drive_new() */
-    QTAILQ_ENTRY(BlockBackend) link; /* for blk_backends */
+    QTAILQ_ENTRY(BlockBackend) link;         /* for block_backends */
+    QTAILQ_ENTRY(BlockBackend) monitor_link; /* for monitor_block_backends */
 
     void *dev;                  /* attached device model, if any */
     /* TODO change to DeviceState when all users are qdevified */
     const BlockDevOps *dev_ops;
     void *dev_opaque;
+
+    /* the block size for which the guest device expects atomicity */
+    int guest_block_size;
+
+    /* If the BDS tree is removed, some of its options are stored here (which
+     * can be used to restore those options in the new BDS on insert) */
+    BlockBackendRootState root_state;
+
+    bool enable_write_cache;
+
+    /* I/O stats (display with "info blockstats"). */
+    BlockAcctStats stats;
+
+    BlockdevOnError on_read_error, on_write_error;
+    bool iostatus_enabled;
+    BlockDeviceIoStatus iostatus;
+
+    bool allow_write_beyond_eof;
+
+    NotifierList remove_bs_notifiers, insert_bs_notifiers;
 };
 
 typedef struct BlockBackendAIOCB {
     BlockAIOCB common;
     QEMUBH *bh;
+    BlockBackend *blk;
     int ret;
 } BlockBackendAIOCB;
 
 static const AIOCBInfo block_backend_aiocb_info = {
+    .get_aio_context = blk_aiocb_get_aio_context,
     .aiocb_size = sizeof(BlockBackendAIOCB),
 };
 
 static void drive_info_del(DriveInfo *dinfo);
 
-/* All the BlockBackends (except for hidden ones) */
-static QTAILQ_HEAD(, BlockBackend) blk_backends =
-    QTAILQ_HEAD_INITIALIZER(blk_backends);
+/* All BlockBackends */
+static QTAILQ_HEAD(, BlockBackend) block_backends =
+    QTAILQ_HEAD_INITIALIZER(block_backends);
+
+/* All BlockBackends referenced by the monitor and which are iterated through by
+ * blk_next() */
+static QTAILQ_HEAD(, BlockBackend) monitor_block_backends =
+    QTAILQ_HEAD_INITIALIZER(monitor_block_backends);
+
+static void blk_root_inherit_options(int *child_flags, QDict *child_options,
+                                     int parent_flags, QDict *parent_options)
+{
+    /* We're not supposed to call this function for root nodes */
+    abort();
+}
+
+static const BdrvChildRole child_root = {
+    .inherit_options = blk_root_inherit_options,
+};
 
 /*
- * Create a new BlockBackend with @name, with a reference count of one.
- * @name must not be null or empty.
- * Fail if a BlockBackend with this name already exists.
+ * Create a new BlockBackend with a reference count of one.
  * Store an error through @errp on failure, unless it's null.
  * Return the new BlockBackend on success, null on failure.
  */
-BlockBackend *blk_new(const char *name, Error **errp)
+BlockBackend *blk_new(Error **errp)
 {
     BlockBackend *blk;
 
-    assert(name && name[0]);
-    if (!id_wellformed(name)) {
-        error_setg(errp, "Invalid device name");
-        return NULL;
-    }
-    if (blk_by_name(name)) {
-        error_setg(errp, "Device with id '%s' already exists", name);
-        return NULL;
-    }
-    if (bdrv_find_node(name)) {
-        error_setg(errp,
-                   "Device name '%s' conflicts with an existing node name",
-                   name);
-        return NULL;
-    }
-
     blk = g_new0(BlockBackend, 1);
-    blk->name = g_strdup(name);
     blk->refcnt = 1;
-    QTAILQ_INSERT_TAIL(&blk_backends, blk, link);
+    notifier_list_init(&blk->remove_bs_notifiers);
+    notifier_list_init(&blk->insert_bs_notifiers);
+    QTAILQ_INSERT_TAIL(&block_backends, blk, link);
     return blk;
 }
 
@@ -85,18 +116,18 @@ BlockBackend *blk_new(const char *name, Error **errp)
  * Create a new BlockBackend with a new BlockDriverState attached.
  * Otherwise just like blk_new(), which see.
  */
-BlockBackend *blk_new_with_bs(const char *name, Error **errp)
+BlockBackend *blk_new_with_bs(Error **errp)
 {
     BlockBackend *blk;
     BlockDriverState *bs;
 
-    blk = blk_new(name, errp);
+    blk = blk_new(errp);
     if (!blk) {
         return NULL;
     }
 
     bs = bdrv_new_root();
-    blk->bs = bs;
+    blk->root = bdrv_root_attach_child(bs, "root", &child_root);
     bs->blk = blk;
     return blk;
 }
@@ -113,44 +144,46 @@ BlockBackend *blk_new_with_bs(const char *name, Error **errp)
  * though, so callers of this function have to be able to specify @filename and
  * @flags.
  */
-BlockBackend *blk_new_open(const char *name, const char *filename,
-                           const char *reference, QDict *options, int flags,
-                           Error **errp)
+BlockBackend *blk_new_open(const char *filename, const char *reference,
+                           QDict *options, int flags, Error **errp)
 {
     BlockBackend *blk;
     int ret;
 
-    blk = blk_new_with_bs(name, errp);
+    blk = blk_new_with_bs(errp);
     if (!blk) {
         QDECREF(options);
         return NULL;
     }
 
-    ret = bdrv_open(&blk->bs, filename, reference, options, flags, NULL, errp);
+    ret = bdrv_open(&blk->root->bs, filename, reference, options, flags, errp);
     if (ret < 0) {
         blk_unref(blk);
         return NULL;
     }
 
+    blk_set_enable_write_cache(blk, true);
+
     return blk;
 }
 
 static void blk_delete(BlockBackend *blk)
 {
     assert(!blk->refcnt);
+    assert(!blk->name);
     assert(!blk->dev);
-    if (blk->bs) {
-        assert(blk->bs->blk == blk);
-        blk->bs->blk = NULL;
-        bdrv_unref(blk->bs);
-        blk->bs = NULL;
+    if (blk->root) {
+        blk_remove_bs(blk);
     }
-    /* Avoid double-remove after blk_hide_on_behalf_of_hmp_drive_del() */
-    if (blk->name[0]) {
-        QTAILQ_REMOVE(&blk_backends, blk, link);
+    assert(QLIST_EMPTY(&blk->remove_bs_notifiers.notifiers));
+    assert(QLIST_EMPTY(&blk->insert_bs_notifiers.notifiers));
+    if (blk->root_state.throttle_state) {
+        g_free(blk->root_state.throttle_group);
+        throttle_group_unref(blk->root_state.throttle_state);
     }
-    g_free(blk->name);
+    QTAILQ_REMOVE(&block_backends, blk, link);
     drive_info_del(blk->legacy_dinfo);
+    block_acct_cleanup(&blk->stats);
     g_free(blk);
 }
 
@@ -164,6 +197,11 @@ static void drive_info_del(DriveInfo *dinfo)
     g_free(dinfo);
 }
 
+int blk_get_refcnt(BlockBackend *blk)
+{
+    return blk ? blk->refcnt : 0;
+}
+
 /*
  * Increment @blk's reference count.
  * @blk must not be null.
@@ -189,7 +227,32 @@ void blk_unref(BlockBackend *blk)
 }
 
 /*
- * Return the BlockBackend after @blk.
+ * Behaves similarly to blk_next() but iterates over all BlockBackends, even the
+ * ones which are hidden (i.e. are not referenced by the monitor).
+ */
+static BlockBackend *blk_all_next(BlockBackend *blk)
+{
+    return blk ? QTAILQ_NEXT(blk, link)
+               : QTAILQ_FIRST(&block_backends);
+}
+
+void blk_remove_all_bs(void)
+{
+    BlockBackend *blk = NULL;
+
+    while ((blk = blk_all_next(blk)) != NULL) {
+        AioContext *ctx = blk_get_aio_context(blk);
+
+        aio_context_acquire(ctx);
+        if (blk->root) {
+            blk_remove_bs(blk);
+        }
+        aio_context_release(ctx);
+    }
+}
+
+/*
+ * Return the monitor-owned BlockBackend after @blk.
  * If @blk is null, return the first one.
  * Else, return @blk's next sibling, which may be null.
  *
@@ -200,17 +263,91 @@ void blk_unref(BlockBackend *blk)
  */
 BlockBackend *blk_next(BlockBackend *blk)
 {
-    return blk ? QTAILQ_NEXT(blk, link) : QTAILQ_FIRST(&blk_backends);
+    return blk ? QTAILQ_NEXT(blk, monitor_link)
+               : QTAILQ_FIRST(&monitor_block_backends);
+}
+
+/*
+ * Iterates over all BlockDriverStates which are attached to a BlockBackend.
+ * This function is for use by bdrv_next().
+ *
+ * @bs must be NULL or a BDS that is attached to a BB.
+ */
+BlockDriverState *blk_next_root_bs(BlockDriverState *bs)
+{
+    BlockBackend *blk;
+
+    if (bs) {
+        assert(bs->blk);
+        blk = bs->blk;
+    } else {
+        blk = NULL;
+    }
+
+    do {
+        blk = blk_all_next(blk);
+    } while (blk && !blk->root);
+
+    return blk ? blk->root->bs : NULL;
+}
+
+/*
+ * Add a BlockBackend into the list of backends referenced by the monitor, with
+ * the given @name acting as the handle for the monitor.
+ * Strictly for use by blockdev.c.
+ *
+ * @name must not be null or empty.
+ *
+ * Returns true on success and false on failure. In the latter case, an Error
+ * object is returned through @errp.
+ */
+bool monitor_add_blk(BlockBackend *blk, const char *name, Error **errp)
+{
+    assert(!blk->name);
+    assert(name && name[0]);
+
+    if (!id_wellformed(name)) {
+        error_setg(errp, "Invalid device name");
+        return false;
+    }
+    if (blk_by_name(name)) {
+        error_setg(errp, "Device with id '%s' already exists", name);
+        return false;
+    }
+    if (bdrv_find_node(name)) {
+        error_setg(errp,
+                   "Device name '%s' conflicts with an existing node name",
+                   name);
+        return false;
+    }
+
+    blk->name = g_strdup(name);
+    QTAILQ_INSERT_TAIL(&monitor_block_backends, blk, monitor_link);
+    return true;
+}
+
+/*
+ * Remove a BlockBackend from the list of backends referenced by the monitor.
+ * Strictly for use by blockdev.c.
+ */
+void monitor_remove_blk(BlockBackend *blk)
+{
+    if (!blk->name) {
+        return;
+    }
+
+    QTAILQ_REMOVE(&monitor_block_backends, blk, monitor_link);
+    g_free(blk->name);
+    blk->name = NULL;
 }
 
 /*
  * Return @blk's name, a non-null string.
- * Wart: the name is empty iff @blk has been hidden with
- * blk_hide_on_behalf_of_hmp_drive_del().
+ * Returns an empty string iff @blk is not referenced by the monitor.
  */
 const char *blk_name(BlockBackend *blk)
 {
-    return blk->name;
+    return blk->name ?: "";
 }
 
 /*
@@ -219,10 +356,10 @@ const char *blk_name(BlockBackend *blk)
  */
 BlockBackend *blk_by_name(const char *name)
 {
-    BlockBackend *blk;
+    BlockBackend *blk = NULL;
 
     assert(name);
-    QTAILQ_FOREACH(blk, &blk_backends, link) {
+    while ((blk = blk_next(blk)) != NULL) {
         if (!strcmp(name, blk->name)) {
             return blk;
         }
@@ -235,7 +372,7 @@ BlockBackend *blk_by_name(const char *name)
  */
 BlockDriverState *blk_bs(BlockBackend *blk)
 {
-    return blk->bs;
+    return blk->root ? blk->root->bs : NULL;
 }
 
 /*
@@ -263,9 +400,9 @@ DriveInfo *blk_set_legacy_dinfo(BlockBackend *blk, DriveInfo *dinfo)
  */
 BlockBackend *blk_by_legacy_dinfo(DriveInfo *dinfo)
 {
-    BlockBackend *blk;
+    BlockBackend *blk = NULL;
 
-    QTAILQ_FOREACH(blk, &blk_backends, link) {
+    while ((blk = blk_next(blk)) != NULL) {
         if (blk->legacy_dinfo == dinfo) {
             return blk;
         }
@@ -274,21 +411,32 @@ BlockBackend *blk_by_legacy_dinfo(DriveInfo *dinfo)
 }
 
 /*
- * Hide @blk.
- * @blk must not have been hidden already.
- * Make attached BlockDriverState, if any, anonymous.
- * Once hidden, @blk is invisible to all functions that don't receive
- * it as argument.  For example, blk_by_name() won't return it.
- * Strictly for use by do_drive_del().
- * TODO get rid of it!
+ * Disassociates the currently associated BlockDriverState from @blk.
  */
-void blk_hide_on_behalf_of_hmp_drive_del(BlockBackend *blk)
+void blk_remove_bs(BlockBackend *blk)
 {
-    QTAILQ_REMOVE(&blk_backends, blk, link);
-    blk->name[0] = 0;
-    if (blk->bs) {
-        bdrv_make_anon(blk->bs);
-    }
+    assert(blk->root->bs->blk == blk);
+
+    notifier_list_notify(&blk->remove_bs_notifiers, blk);
+
+    blk_update_root_state(blk);
+
+    blk->root->bs->blk = NULL;
+    bdrv_root_unref_child(blk->root);
+    blk->root = NULL;
+}
+
+/*
+ * Associates a new BlockDriverState with @blk.
+ */
+void blk_insert_bs(BlockBackend *blk, BlockDriverState *bs)
+{
+    assert(!blk->root && !bs->blk);
+    bdrv_ref(bs);
+    blk->root = bdrv_root_attach_child(bs, "root", &child_root);
+    bs->blk = blk;
+
+    notifier_list_notify(&blk->insert_bs_notifiers, blk);
 }
 
 /*
@@ -303,7 +451,7 @@ int blk_attach_dev(BlockBackend *blk, void *dev)
     }
     blk_ref(blk);
     blk->dev = dev;
-    bdrv_iostatus_reset(blk->bs);
+    blk_iostatus_reset(blk);
     return 0;
 }
 
@@ -330,7 +478,7 @@ void blk_detach_dev(BlockBackend *blk, void *dev)
     blk->dev = NULL;
     blk->dev_ops = NULL;
     blk->dev_opaque = NULL;
-    bdrv_set_guest_block_size(blk->bs, 512);
+    blk->guest_block_size = 512;
     blk_unref(blk);
 }
 
@@ -364,18 +512,15 @@ void blk_set_dev_ops(BlockBackend *blk, const BlockDevOps *ops,
 void blk_dev_change_media_cb(BlockBackend *blk, bool load)
 {
     if (blk->dev_ops && blk->dev_ops->change_media_cb) {
-        bool tray_was_closed = !blk_dev_is_tray_open(blk);
+        bool tray_was_open, tray_is_open;
 
+        tray_was_open = blk_dev_is_tray_open(blk);
         blk->dev_ops->change_media_cb(blk->dev_opaque, load);
-        if (tray_was_closed) {
-            /* tray open */
-            qapi_event_send_device_tray_moved(blk_name(blk),
-                                              true, &error_abort);
-        }
-        if (load) {
-            /* tray close */
-            qapi_event_send_device_tray_moved(blk_name(blk),
-                                              false, &error_abort);
+        tray_is_open = blk_dev_is_tray_open(blk);
+
+        if (tray_was_open != tray_is_open) {
+            qapi_event_send_device_tray_moved(blk_name(blk), tray_is_open,
+                                              &error_abort);
         }
     }
 }
@@ -390,6 +535,14 @@ bool blk_dev_has_removable_media(BlockBackend *blk)
 }
 
 /*
+ * Does @blk's attached device model have a tray?
+ */
+bool blk_dev_has_tray(BlockBackend *blk)
+{
+    return blk->dev_ops && blk->dev_ops->is_tray_open;
+}
+
+/*
  * Notify @blk's attached device model of a media eject request.
  * If @force is true, the medium is about to be yanked out forcefully.
  */
@@ -405,7 +558,7 @@ void blk_dev_eject_request(BlockBackend *blk, bool force)
  */
 bool blk_dev_is_tray_open(BlockBackend *blk)
 {
-    if (blk->dev_ops && blk->dev_ops->is_tray_open) {
+    if (blk_dev_has_tray(blk)) {
         return blk->dev_ops->is_tray_open(blk->dev_opaque);
     }
     return false;
@@ -435,7 +588,53 @@ void blk_dev_resize_cb(BlockBackend *blk)
 
 void blk_iostatus_enable(BlockBackend *blk)
 {
-    bdrv_iostatus_enable(blk->bs);
+    blk->iostatus_enabled = true;
+    blk->iostatus = BLOCK_DEVICE_IO_STATUS_OK;
+}
+
+/* The I/O status is only enabled if the drive explicitly
+ * enables it _and_ the VM is configured to stop on errors */
+bool blk_iostatus_is_enabled(const BlockBackend *blk)
+{
+    return (blk->iostatus_enabled &&
+           (blk->on_write_error == BLOCKDEV_ON_ERROR_ENOSPC ||
+            blk->on_write_error == BLOCKDEV_ON_ERROR_STOP   ||
+            blk->on_read_error == BLOCKDEV_ON_ERROR_STOP));
+}
+
+BlockDeviceIoStatus blk_iostatus(const BlockBackend *blk)
+{
+    return blk->iostatus;
+}
+
+void blk_iostatus_disable(BlockBackend *blk)
+{
+    blk->iostatus_enabled = false;
+}
+
+void blk_iostatus_reset(BlockBackend *blk)
+{
+    if (blk_iostatus_is_enabled(blk)) {
+        BlockDriverState *bs = blk_bs(blk);
+        blk->iostatus = BLOCK_DEVICE_IO_STATUS_OK;
+        if (bs && bs->job) {
+            block_job_iostatus_reset(bs->job);
+        }
+    }
+}
+
+void blk_iostatus_set_err(BlockBackend *blk, int error)
+{
+    assert(blk_iostatus_is_enabled(blk));
+    if (blk->iostatus == BLOCK_DEVICE_IO_STATUS_OK) {
+        blk->iostatus = error == ENOSPC ? BLOCK_DEVICE_IO_STATUS_NOSPACE :
+                                          BLOCK_DEVICE_IO_STATUS_FAILED;
+    }
+}
+
+void blk_set_allow_write_beyond_eof(BlockBackend *blk, bool allow)
+{
+    blk->allow_write_beyond_eof = allow;
 }
 
 static int blk_check_byte_request(BlockBackend *blk, int64_t offset,
@@ -447,21 +646,23 @@ static int blk_check_byte_request(BlockBackend *blk, int64_t offset,
         return -EIO;
     }
 
-    if (!blk_is_inserted(blk)) {
+    if (!blk_is_available(blk)) {
         return -ENOMEDIUM;
     }
 
-    len = blk_getlength(blk);
-    if (len < 0) {
-        return len;
-    }
-
     if (offset < 0) {
         return -EIO;
     }
 
-    if (offset > len || len - offset < size) {
-        return -EIO;
+    if (!blk->allow_write_beyond_eof) {
+        len = blk_getlength(blk);
+        if (len < 0) {
+            return len;
+        }
+
+        if (offset > len || len - offset < size) {
+            return -EIO;
+        }
     }
 
     return 0;
@@ -482,48 +683,144 @@ static int blk_check_request(BlockBackend *blk, int64_t sector_num,
                                   nb_sectors * BDRV_SECTOR_SIZE);
 }
 
-int blk_read(BlockBackend *blk, int64_t sector_num, uint8_t *buf,
-             int nb_sectors)
+static int coroutine_fn blk_co_preadv(BlockBackend *blk, int64_t offset,
+                                      unsigned int bytes, QEMUIOVector *qiov,
+                                      BdrvRequestFlags flags)
 {
-    int ret = blk_check_request(blk, sector_num, nb_sectors);
+    int ret = blk_check_byte_request(blk, offset, bytes);
     if (ret < 0) {
         return ret;
     }
 
-    return bdrv_read(blk->bs, sector_num, buf, nb_sectors);
+    return bdrv_co_do_preadv(blk_bs(blk), offset, bytes, qiov, flags);
 }
 
-int blk_read_unthrottled(BlockBackend *blk, int64_t sector_num, uint8_t *buf,
-                         int nb_sectors)
+static int coroutine_fn blk_co_pwritev(BlockBackend *blk, int64_t offset,
+                                      unsigned int bytes, QEMUIOVector *qiov,
+                                      BdrvRequestFlags flags)
 {
-    int ret = blk_check_request(blk, sector_num, nb_sectors);
+    int ret;
+
+    ret = blk_check_byte_request(blk, offset, bytes);
     if (ret < 0) {
         return ret;
     }
 
-    return bdrv_read_unthrottled(blk->bs, sector_num, buf, nb_sectors);
+    if (!blk->enable_write_cache) {
+        flags |= BDRV_REQ_FUA;
+    }
+
+    return bdrv_co_do_pwritev(blk_bs(blk), offset, bytes, qiov, flags);
 }
 
-int blk_write(BlockBackend *blk, int64_t sector_num, const uint8_t *buf,
-              int nb_sectors)
+typedef struct BlkRwCo {
+    BlockBackend *blk;
+    int64_t offset;
+    QEMUIOVector *qiov;
+    int ret;
+    BdrvRequestFlags flags;
+} BlkRwCo;
+
+static void blk_read_entry(void *opaque)
 {
-    int ret = blk_check_request(blk, sector_num, nb_sectors);
-    if (ret < 0) {
-        return ret;
+    BlkRwCo *rwco = opaque;
+
+    rwco->ret = blk_co_preadv(rwco->blk, rwco->offset, rwco->qiov->size,
+                              rwco->qiov, rwco->flags);
+}
+
+static void blk_write_entry(void *opaque)
+{
+    BlkRwCo *rwco = opaque;
+
+    rwco->ret = blk_co_pwritev(rwco->blk, rwco->offset, rwco->qiov->size,
+                               rwco->qiov, rwco->flags);
+}
+
+static int blk_prw(BlockBackend *blk, int64_t offset, uint8_t *buf,
+                   int64_t bytes, CoroutineEntry co_entry,
+                   BdrvRequestFlags flags)
+{
+    AioContext *aio_context;
+    QEMUIOVector qiov;
+    struct iovec iov;
+    Coroutine *co;
+    BlkRwCo rwco;
+
+    iov = (struct iovec) {
+        .iov_base = buf,
+        .iov_len = bytes,
+    };
+    qemu_iovec_init_external(&qiov, &iov, 1);
+
+    rwco = (BlkRwCo) {
+        .blk    = blk,
+        .offset = offset,
+        .qiov   = &qiov,
+        .flags  = flags,
+        .ret    = NOT_DONE,
+    };
+
+    co = qemu_coroutine_create(co_entry);
+    qemu_coroutine_enter(co, &rwco);
+
+    aio_context = blk_get_aio_context(blk);
+    while (rwco.ret == NOT_DONE) {
+        aio_poll(aio_context, true);
     }
 
-    return bdrv_write(blk->bs, sector_num, buf, nb_sectors);
+    return rwco.ret;
 }
 
-int blk_write_zeroes(BlockBackend *blk, int64_t sector_num,
-                     int nb_sectors, BdrvRequestFlags flags)
+static int blk_rw(BlockBackend *blk, int64_t sector_num, uint8_t *buf,
+                  int nb_sectors, CoroutineEntry co_entry,
+                  BdrvRequestFlags flags)
 {
-    int ret = blk_check_request(blk, sector_num, nb_sectors);
+    if (nb_sectors < 0 || nb_sectors > BDRV_REQUEST_MAX_SECTORS) {
+        return -EINVAL;
+    }
+
+    return blk_prw(blk, sector_num << BDRV_SECTOR_BITS, buf,
+                   nb_sectors << BDRV_SECTOR_BITS, co_entry, flags);
+}
+
+int blk_read(BlockBackend *blk, int64_t sector_num, uint8_t *buf,
+             int nb_sectors)
+{
+    return blk_rw(blk, sector_num, buf, nb_sectors, blk_read_entry, 0);
+}
+
+int blk_read_unthrottled(BlockBackend *blk, int64_t sector_num, uint8_t *buf,
+                         int nb_sectors)
+{
+    BlockDriverState *bs = blk_bs(blk);
+    bool enabled;
+    int ret;
+
+    ret = blk_check_request(blk, sector_num, nb_sectors);
     if (ret < 0) {
         return ret;
     }
 
-    return bdrv_write_zeroes(blk->bs, sector_num, nb_sectors, flags);
+    enabled = bs->io_limits_enabled;
+    bs->io_limits_enabled = false;
+    ret = blk_read(blk, sector_num, buf, nb_sectors);
+    bs->io_limits_enabled = enabled;
+    return ret;
+}
+
+int blk_write(BlockBackend *blk, int64_t sector_num, const uint8_t *buf,
+              int nb_sectors)
+{
+    return blk_rw(blk, sector_num, (uint8_t*) buf, nb_sectors,
+                  blk_write_entry, 0);
+}
+
+int blk_write_zeroes(BlockBackend *blk, int64_t sector_num,
+                     int nb_sectors, BdrvRequestFlags flags)
+{
+    return blk_rw(blk, sector_num, NULL, nb_sectors, blk_write_entry,
+                  flags | BDRV_REQ_ZERO_WRITE);
 }
 
 static void error_callback_bh(void *opaque)
@@ -534,13 +831,15 @@ static void error_callback_bh(void *opaque)
     qemu_aio_unref(acb);
 }
 
-static BlockAIOCB *abort_aio_request(BlockBackend *blk, BlockCompletionFunc *cb,
-                                     void *opaque, int ret)
+BlockAIOCB *blk_abort_aio_request(BlockBackend *blk,
+                                  BlockCompletionFunc *cb,
+                                  void *opaque, int ret)
 {
     struct BlockBackendAIOCB *acb;
     QEMUBH *bh;
 
     acb = blk_aio_get(&block_backend_aiocb_info, blk, cb, opaque);
+    acb->blk = blk;
     acb->ret = ret;
 
     bh = aio_bh_new(blk_get_aio_context(blk), error_callback_bh, acb);
@@ -550,82 +849,182 @@ static BlockAIOCB *abort_aio_request(BlockBackend *blk, BlockCompletionFunc *cb,
     return &acb->common;
 }
 
+typedef struct BlkAioEmAIOCB {
+    BlockAIOCB common;
+    BlkRwCo rwco;
+    int bytes;
+    bool has_returned;
+    QEMUBH* bh;
+} BlkAioEmAIOCB;
+
+static const AIOCBInfo blk_aio_em_aiocb_info = {
+    .aiocb_size         = sizeof(BlkAioEmAIOCB),
+};
+
+static void blk_aio_complete(BlkAioEmAIOCB *acb)
+{
+    if (acb->bh) {
+        assert(acb->has_returned);
+        qemu_bh_delete(acb->bh);
+    }
+    if (acb->has_returned) {
+        acb->common.cb(acb->common.opaque, acb->rwco.ret);
+        qemu_aio_unref(acb);
+    }
+}
+
+static void blk_aio_complete_bh(void *opaque)
+{
+    blk_aio_complete(opaque);
+}
+
+static BlockAIOCB *blk_aio_prwv(BlockBackend *blk, int64_t offset, int bytes,
+                                QEMUIOVector *qiov, CoroutineEntry co_entry,
+                                BdrvRequestFlags flags,
+                                BlockCompletionFunc *cb, void *opaque)
+{
+    BlkAioEmAIOCB *acb;
+    Coroutine *co;
+
+    acb = blk_aio_get(&blk_aio_em_aiocb_info, blk, cb, opaque);
+    acb->rwco = (BlkRwCo) {
+        .blk    = blk,
+        .offset = offset,
+        .qiov   = qiov,
+        .flags  = flags,
+        .ret    = NOT_DONE,
+    };
+    acb->bytes = bytes;
+    acb->bh = NULL;
+    acb->has_returned = false;
+
+    co = qemu_coroutine_create(co_entry);
+    qemu_coroutine_enter(co, acb);
+
+    acb->has_returned = true;
+    if (acb->rwco.ret != NOT_DONE) {
+        acb->bh = aio_bh_new(blk_get_aio_context(blk), blk_aio_complete_bh, acb);
+        qemu_bh_schedule(acb->bh);
+    }
+
+    return &acb->common;
+}
+
+static void blk_aio_read_entry(void *opaque)
+{
+    BlkAioEmAIOCB *acb = opaque;
+    BlkRwCo *rwco = &acb->rwco;
+
+    assert(rwco->qiov->size == acb->bytes);
+    rwco->ret = blk_co_preadv(rwco->blk, rwco->offset, acb->bytes,
+                              rwco->qiov, rwco->flags);
+    blk_aio_complete(acb);
+}
+
+static void blk_aio_write_entry(void *opaque)
+{
+    BlkAioEmAIOCB *acb = opaque;
+    BlkRwCo *rwco = &acb->rwco;
+
+    assert(!rwco->qiov || rwco->qiov->size == acb->bytes);
+    rwco->ret = blk_co_pwritev(rwco->blk, rwco->offset, acb->bytes,
+                               rwco->qiov, rwco->flags);
+    blk_aio_complete(acb);
+}
+
 BlockAIOCB *blk_aio_write_zeroes(BlockBackend *blk, int64_t sector_num,
                                  int nb_sectors, BdrvRequestFlags flags,
                                  BlockCompletionFunc *cb, void *opaque)
 {
-    int ret = blk_check_request(blk, sector_num, nb_sectors);
-    if (ret < 0) {
-        return abort_aio_request(blk, cb, opaque, ret);
+    if (nb_sectors < 0 || nb_sectors > BDRV_REQUEST_MAX_SECTORS) {
+        return blk_abort_aio_request(blk, cb, opaque, -EINVAL);
     }
 
-    return bdrv_aio_write_zeroes(blk->bs, sector_num, nb_sectors, flags,
-                                 cb, opaque);
+    return blk_aio_prwv(blk, sector_num << BDRV_SECTOR_BITS,
+                        nb_sectors << BDRV_SECTOR_BITS, NULL,
+                        blk_aio_write_entry, flags | BDRV_REQ_ZERO_WRITE,
+                        cb, opaque);
 }
 
 int blk_pread(BlockBackend *blk, int64_t offset, void *buf, int count)
 {
-    int ret = blk_check_byte_request(blk, offset, count);
+    int ret = blk_prw(blk, offset, buf, count, blk_read_entry, 0);
     if (ret < 0) {
         return ret;
     }
-
-    return bdrv_pread(blk->bs, offset, buf, count);
+    return count;
 }
 
 int blk_pwrite(BlockBackend *blk, int64_t offset, const void *buf, int count)
 {
-    int ret = blk_check_byte_request(blk, offset, count);
+    int ret = blk_prw(blk, offset, (void*) buf, count, blk_write_entry, 0);
     if (ret < 0) {
         return ret;
     }
-
-    return bdrv_pwrite(blk->bs, offset, buf, count);
+    return count;
 }
 
 int64_t blk_getlength(BlockBackend *blk)
 {
-    return bdrv_getlength(blk->bs);
+    if (!blk_is_available(blk)) {
+        return -ENOMEDIUM;
+    }
+
+    return bdrv_getlength(blk_bs(blk));
 }
 
 void blk_get_geometry(BlockBackend *blk, uint64_t *nb_sectors_ptr)
 {
-    bdrv_get_geometry(blk->bs, nb_sectors_ptr);
+    if (!blk_bs(blk)) {
+        *nb_sectors_ptr = 0;
+    } else {
+        bdrv_get_geometry(blk_bs(blk), nb_sectors_ptr);
+    }
 }
 
 int64_t blk_nb_sectors(BlockBackend *blk)
 {
-    return bdrv_nb_sectors(blk->bs);
+    if (!blk_is_available(blk)) {
+        return -ENOMEDIUM;
+    }
+
+    return bdrv_nb_sectors(blk_bs(blk));
 }
 
 BlockAIOCB *blk_aio_readv(BlockBackend *blk, int64_t sector_num,
                           QEMUIOVector *iov, int nb_sectors,
                           BlockCompletionFunc *cb, void *opaque)
 {
-    int ret = blk_check_request(blk, sector_num, nb_sectors);
-    if (ret < 0) {
-        return abort_aio_request(blk, cb, opaque, ret);
+    if (nb_sectors < 0 || nb_sectors > BDRV_REQUEST_MAX_SECTORS) {
+        return blk_abort_aio_request(blk, cb, opaque, -EINVAL);
     }
 
-    return bdrv_aio_readv(blk->bs, sector_num, iov, nb_sectors, cb, opaque);
+    assert(nb_sectors << BDRV_SECTOR_BITS == iov->size);
+    return blk_aio_prwv(blk, sector_num << BDRV_SECTOR_BITS, iov->size, iov,
+                        blk_aio_read_entry, 0, cb, opaque);
 }
 
 BlockAIOCB *blk_aio_writev(BlockBackend *blk, int64_t sector_num,
                            QEMUIOVector *iov, int nb_sectors,
                            BlockCompletionFunc *cb, void *opaque)
 {
-    int ret = blk_check_request(blk, sector_num, nb_sectors);
-    if (ret < 0) {
-        return abort_aio_request(blk, cb, opaque, ret);
+    if (nb_sectors < 0 || nb_sectors > BDRV_REQUEST_MAX_SECTORS) {
+        return blk_abort_aio_request(blk, cb, opaque, -EINVAL);
     }
 
-    return bdrv_aio_writev(blk->bs, sector_num, iov, nb_sectors, cb, opaque);
+    assert(nb_sectors << BDRV_SECTOR_BITS == iov->size);
+    return blk_aio_prwv(blk, sector_num << BDRV_SECTOR_BITS, iov->size, iov,
+                        blk_aio_write_entry, 0, cb, opaque);
 }
 
 BlockAIOCB *blk_aio_flush(BlockBackend *blk,
                           BlockCompletionFunc *cb, void *opaque)
 {
-    return bdrv_aio_flush(blk->bs, cb, opaque);
+    if (!blk_is_available(blk)) {
+        return blk_abort_aio_request(blk, cb, opaque, -ENOMEDIUM);
+    }
+
+    return bdrv_aio_flush(blk_bs(blk), cb, opaque);
 }
 
 BlockAIOCB *blk_aio_discard(BlockBackend *blk,
@@ -634,10 +1033,10 @@ BlockAIOCB *blk_aio_discard(BlockBackend *blk,
 {
     int ret = blk_check_request(blk, sector_num, nb_sectors);
     if (ret < 0) {
-        return abort_aio_request(blk, cb, opaque, ret);
+        return blk_abort_aio_request(blk, cb, opaque, ret);
     }
 
-    return bdrv_aio_discard(blk->bs, sector_num, nb_sectors, cb, opaque);
+    return bdrv_aio_discard(blk_bs(blk), sector_num, nb_sectors, cb, opaque);
 }
 
 void blk_aio_cancel(BlockAIOCB *acb)
@@ -661,18 +1060,26 @@ int blk_aio_multiwrite(BlockBackend *blk, BlockRequest *reqs, int num_reqs)
         }
     }
 
-    return bdrv_aio_multiwrite(blk->bs, reqs, num_reqs);
+    return bdrv_aio_multiwrite(blk_bs(blk), reqs, num_reqs);
 }
 
 int blk_ioctl(BlockBackend *blk, unsigned long int req, void *buf)
 {
-    return bdrv_ioctl(blk->bs, req, buf);
+    if (!blk_is_available(blk)) {
+        return -ENOMEDIUM;
+    }
+
+    return bdrv_ioctl(blk_bs(blk), req, buf);
 }
 
 BlockAIOCB *blk_aio_ioctl(BlockBackend *blk, unsigned long int req, void *buf,
                           BlockCompletionFunc *cb, void *opaque)
 {
-    return bdrv_aio_ioctl(blk->bs, req, buf, cb, opaque);
+    if (!blk_is_available(blk)) {
+        return blk_abort_aio_request(blk, cb, opaque, -ENOMEDIUM);
+    }
+
+    return bdrv_aio_ioctl(blk_bs(blk), req, buf, cb, opaque);
 }
 
 int blk_co_discard(BlockBackend *blk, int64_t sector_num, int nb_sectors)
@@ -682,27 +1089,32 @@ int blk_co_discard(BlockBackend *blk, int64_t sector_num, int nb_sectors)
         return ret;
     }
 
-    return bdrv_co_discard(blk->bs, sector_num, nb_sectors);
+    return bdrv_co_discard(blk_bs(blk), sector_num, nb_sectors);
 }
 
 int blk_co_flush(BlockBackend *blk)
 {
-    return bdrv_co_flush(blk->bs);
+    if (!blk_is_available(blk)) {
+        return -ENOMEDIUM;
+    }
+
+    return bdrv_co_flush(blk_bs(blk));
 }
 
 int blk_flush(BlockBackend *blk)
 {
-    return bdrv_flush(blk->bs);
-}
+    if (!blk_is_available(blk)) {
+        return -ENOMEDIUM;
+    }
 
-int blk_flush_all(void)
-{
-    return bdrv_flush_all();
+    return bdrv_flush(blk_bs(blk));
 }
 
 void blk_drain(BlockBackend *blk)
 {
-    bdrv_drain(blk->bs);
+    if (blk_bs(blk)) {
+        bdrv_drain(blk_bs(blk));
+    }
 }
 
 void blk_drain_all(void)
@@ -710,119 +1122,273 @@ void blk_drain_all(void)
     bdrv_drain_all();
 }
 
+void blk_set_on_error(BlockBackend *blk, BlockdevOnError on_read_error,
+                      BlockdevOnError on_write_error)
+{
+    blk->on_read_error = on_read_error;
+    blk->on_write_error = on_write_error;
+}
+
 BlockdevOnError blk_get_on_error(BlockBackend *blk, bool is_read)
 {
-    return bdrv_get_on_error(blk->bs, is_read);
+    return is_read ? blk->on_read_error : blk->on_write_error;
 }
 
 BlockErrorAction blk_get_error_action(BlockBackend *blk, bool is_read,
                                       int error)
 {
-    return bdrv_get_error_action(blk->bs, is_read, error);
+    BlockdevOnError on_err = blk_get_on_error(blk, is_read);
+
+    switch (on_err) {
+    case BLOCKDEV_ON_ERROR_ENOSPC:
+        return (error == ENOSPC) ?
+               BLOCK_ERROR_ACTION_STOP : BLOCK_ERROR_ACTION_REPORT;
+    case BLOCKDEV_ON_ERROR_STOP:
+        return BLOCK_ERROR_ACTION_STOP;
+    case BLOCKDEV_ON_ERROR_REPORT:
+        return BLOCK_ERROR_ACTION_REPORT;
+    case BLOCKDEV_ON_ERROR_IGNORE:
+        return BLOCK_ERROR_ACTION_IGNORE;
+    default:
+        abort();
+    }
+}
+
+static void send_qmp_error_event(BlockBackend *blk,
+                                 BlockErrorAction action,
+                                 bool is_read, int error)
+{
+    IoOperationType optype;
+
+    optype = is_read ? IO_OPERATION_TYPE_READ : IO_OPERATION_TYPE_WRITE;
+    qapi_event_send_block_io_error(blk_name(blk), optype, action,
+                                   blk_iostatus_is_enabled(blk),
+                                   error == ENOSPC, strerror(error),
+                                   &error_abort);
 }
 
+/* This is done by device models because, while the block layer knows
+ * about the error, it does not know whether an operation comes from
+ * the device or the block layer (from a job, for example).
+ */
 void blk_error_action(BlockBackend *blk, BlockErrorAction action,
                       bool is_read, int error)
 {
-    bdrv_error_action(blk->bs, action, is_read, error);
+    assert(error >= 0);
+
+    if (action == BLOCK_ERROR_ACTION_STOP) {
+        /* First set the iostatus, so that "info block" returns an iostatus
+         * that matches the events raised so far (an additional error iostatus
+         * is fine, but not a lost one).
+         */
+        blk_iostatus_set_err(blk, error);
+
+        /* Then raise the request to stop the VM and the event.
+         * qemu_system_vmstop_request_prepare has two effects.  First,
+         * it ensures that the STOP event always comes after the
+         * BLOCK_IO_ERROR event.  Second, it ensures that even if management
+         * can observe the STOP event and do a "cont" before the STOP
+         * event is issued, the VM will not stop.  In this case, vm_start()
+         * also ensures that the STOP/RESUME pair of events is emitted.
+         */
+        qemu_system_vmstop_request_prepare();
+        send_qmp_error_event(blk, action, is_read, error);
+        qemu_system_vmstop_request(RUN_STATE_IO_ERROR);
+    } else {
+        send_qmp_error_event(blk, action, is_read, error);
+    }
 }
 
 int blk_is_read_only(BlockBackend *blk)
 {
-    return bdrv_is_read_only(blk->bs);
+    BlockDriverState *bs = blk_bs(blk);
+
+    if (bs) {
+        return bdrv_is_read_only(bs);
+    } else {
+        return blk->root_state.read_only;
+    }
 }
 
 int blk_is_sg(BlockBackend *blk)
 {
-    return bdrv_is_sg(blk->bs);
+    BlockDriverState *bs = blk_bs(blk);
+
+    if (!bs) {
+        return 0;
+    }
+
+    return bdrv_is_sg(bs);
 }
 
 int blk_enable_write_cache(BlockBackend *blk)
 {
-    return bdrv_enable_write_cache(blk->bs);
+    return blk->enable_write_cache;
 }
 
 void blk_set_enable_write_cache(BlockBackend *blk, bool wce)
 {
-    bdrv_set_enable_write_cache(blk->bs, wce);
+    blk->enable_write_cache = wce;
 }
 
 void blk_invalidate_cache(BlockBackend *blk, Error **errp)
 {
-    bdrv_invalidate_cache(blk->bs, errp);
+    BlockDriverState *bs = blk_bs(blk);
+
+    if (!bs) {
+        error_setg(errp, "Device '%s' has no medium", blk->name);
+        return;
+    }
+
+    bdrv_invalidate_cache(bs, errp);
+}
+
+bool blk_is_inserted(BlockBackend *blk)
+{
+    BlockDriverState *bs = blk_bs(blk);
+
+    return bs && bdrv_is_inserted(bs);
 }
 
-int blk_is_inserted(BlockBackend *blk)
+bool blk_is_available(BlockBackend *blk)
 {
-    return bdrv_is_inserted(blk->bs);
+    return blk_is_inserted(blk) && !blk_dev_is_tray_open(blk);
 }
 
 void blk_lock_medium(BlockBackend *blk, bool locked)
 {
-    bdrv_lock_medium(blk->bs, locked);
+    BlockDriverState *bs = blk_bs(blk);
+
+    if (bs) {
+        bdrv_lock_medium(bs, locked);
+    }
 }
 
 void blk_eject(BlockBackend *blk, bool eject_flag)
 {
-    bdrv_eject(blk->bs, eject_flag);
+    BlockDriverState *bs = blk_bs(blk);
+
+    if (bs) {
+        bdrv_eject(bs, eject_flag);
+    }
 }
 
 int blk_get_flags(BlockBackend *blk)
 {
-    return bdrv_get_flags(blk->bs);
+    BlockDriverState *bs = blk_bs(blk);
+
+    if (bs) {
+        return bdrv_get_flags(bs);
+    } else {
+        return blk->root_state.open_flags;
+    }
 }
 
 int blk_get_max_transfer_length(BlockBackend *blk)
 {
-    return blk->bs->bl.max_transfer_length;
+    BlockDriverState *bs = blk_bs(blk);
+
+    if (bs) {
+        return bs->bl.max_transfer_length;
+    } else {
+        return 0;
+    }
+}
+
+int blk_get_max_iov(BlockBackend *blk)
+{
+    return blk->root->bs->bl.max_iov;
 }
 
 void blk_set_guest_block_size(BlockBackend *blk, int align)
 {
-    bdrv_set_guest_block_size(blk->bs, align);
+    blk->guest_block_size = align;
+}
+
+void *blk_try_blockalign(BlockBackend *blk, size_t size)
+{
+    return qemu_try_blockalign(blk ? blk_bs(blk) : NULL, size);
 }
 
 void *blk_blockalign(BlockBackend *blk, size_t size)
 {
-    return qemu_blockalign(blk ? blk->bs : NULL, size);
+    return qemu_blockalign(blk ? blk_bs(blk) : NULL, size);
 }
 
 bool blk_op_is_blocked(BlockBackend *blk, BlockOpType op, Error **errp)
 {
-    return bdrv_op_is_blocked(blk->bs, op, errp);
+    BlockDriverState *bs = blk_bs(blk);
+
+    if (!bs) {
+        return false;
+    }
+
+    return bdrv_op_is_blocked(bs, op, errp);
 }
 
 void blk_op_unblock(BlockBackend *blk, BlockOpType op, Error *reason)
 {
-    bdrv_op_unblock(blk->bs, op, reason);
+    BlockDriverState *bs = blk_bs(blk);
+
+    if (bs) {
+        bdrv_op_unblock(bs, op, reason);
+    }
 }
 
 void blk_op_block_all(BlockBackend *blk, Error *reason)
 {
-    bdrv_op_block_all(blk->bs, reason);
+    BlockDriverState *bs = blk_bs(blk);
+
+    if (bs) {
+        bdrv_op_block_all(bs, reason);
+    }
 }
 
 void blk_op_unblock_all(BlockBackend *blk, Error *reason)
 {
-    bdrv_op_unblock_all(blk->bs, reason);
+    BlockDriverState *bs = blk_bs(blk);
+
+    if (bs) {
+        bdrv_op_unblock_all(bs, reason);
+    }
 }
 
 AioContext *blk_get_aio_context(BlockBackend *blk)
 {
-    return bdrv_get_aio_context(blk->bs);
+    BlockDriverState *bs = blk_bs(blk);
+
+    if (bs) {
+        return bdrv_get_aio_context(bs);
+    } else {
+        return qemu_get_aio_context();
+    }
+}
+
+static AioContext *blk_aiocb_get_aio_context(BlockAIOCB *acb)
+{
+    BlockBackendAIOCB *blk_acb = DO_UPCAST(BlockBackendAIOCB, common, acb);
+    return blk_get_aio_context(blk_acb->blk);
 }
 
 void blk_set_aio_context(BlockBackend *blk, AioContext *new_context)
 {
-    bdrv_set_aio_context(blk->bs, new_context);
+    BlockDriverState *bs = blk_bs(blk);
+
+    if (bs) {
+        bdrv_set_aio_context(bs, new_context);
+    }
 }
 
 void blk_add_aio_context_notifier(BlockBackend *blk,
         void (*attached_aio_context)(AioContext *new_context, void *opaque),
         void (*detach_aio_context)(void *opaque), void *opaque)
 {
-    bdrv_add_aio_context_notifier(blk->bs, attached_aio_context,
-                                  detach_aio_context, opaque);
+    BlockDriverState *bs = blk_bs(blk);
+
+    if (bs) {
+        bdrv_add_aio_context_notifier(bs, attached_aio_context,
+                                      detach_aio_context, opaque);
+    }
 }
 
 void blk_remove_aio_context_notifier(BlockBackend *blk,
@@ -831,28 +1397,45 @@ void blk_remove_aio_context_notifier(BlockBackend *blk,
                                      void (*detach_aio_context)(void *),
                                      void *opaque)
 {
-    bdrv_remove_aio_context_notifier(blk->bs, attached_aio_context,
-                                     detach_aio_context, opaque);
+    BlockDriverState *bs = blk_bs(blk);
+
+    if (bs) {
+        bdrv_remove_aio_context_notifier(bs, attached_aio_context,
+                                         detach_aio_context, opaque);
+    }
 }
 
-void blk_add_close_notifier(BlockBackend *blk, Notifier *notify)
+void blk_add_remove_bs_notifier(BlockBackend *blk, Notifier *notify)
 {
-    bdrv_add_close_notifier(blk->bs, notify);
+    notifier_list_add(&blk->remove_bs_notifiers, notify);
+}
+
+void blk_add_insert_bs_notifier(BlockBackend *blk, Notifier *notify)
+{
+    notifier_list_add(&blk->insert_bs_notifiers, notify);
 }
 
 void blk_io_plug(BlockBackend *blk)
 {
-    bdrv_io_plug(blk->bs);
+    BlockDriverState *bs = blk_bs(blk);
+
+    if (bs) {
+        bdrv_io_plug(bs);
+    }
 }
 
 void blk_io_unplug(BlockBackend *blk)
 {
-    bdrv_io_unplug(blk->bs);
+    BlockDriverState *bs = blk_bs(blk);
+
+    if (bs) {
+        bdrv_io_unplug(bs);
+    }
 }
 
 BlockAcctStats *blk_get_stats(BlockBackend *blk)
 {
-    return bdrv_get_stats(blk->bs);
+    return &blk->stats;
 }
 
 void *blk_aio_get(const AIOCBInfo *aiocb_info, BlockBackend *blk,
@@ -864,12 +1447,13 @@ void *blk_aio_get(const AIOCBInfo *aiocb_info, BlockBackend *blk,
 int coroutine_fn blk_co_write_zeroes(BlockBackend *blk, int64_t sector_num,
                                      int nb_sectors, BdrvRequestFlags flags)
 {
-    int ret = blk_check_request(blk, sector_num, nb_sectors);
-    if (ret < 0) {
-        return ret;
+    if (nb_sectors < 0 || nb_sectors > BDRV_REQUEST_MAX_SECTORS) {
+        return -EINVAL;
     }
 
-    return bdrv_co_write_zeroes(blk->bs, sector_num, nb_sectors, flags);
+    return blk_co_pwritev(blk, sector_num << BDRV_SECTOR_BITS,
+                          nb_sectors << BDRV_SECTOR_BITS, NULL,
+                          flags | BDRV_REQ_ZERO_WRITE);
 }
 
 int blk_write_compressed(BlockBackend *blk, int64_t sector_num,
@@ -880,12 +1464,16 @@ int blk_write_compressed(BlockBackend *blk, int64_t sector_num,
         return ret;
     }
 
-    return bdrv_write_compressed(blk->bs, sector_num, buf, nb_sectors);
+    return bdrv_write_compressed(blk_bs(blk), sector_num, buf, nb_sectors);
 }
 
 int blk_truncate(BlockBackend *blk, int64_t offset)
 {
-    return bdrv_truncate(blk->bs, offset);
+    if (!blk_is_available(blk)) {
+        return -ENOMEDIUM;
+    }
+
+    return bdrv_truncate(blk_bs(blk), offset);
 }
 
 int blk_discard(BlockBackend *blk, int64_t sector_num, int nb_sectors)
@@ -895,26 +1483,153 @@ int blk_discard(BlockBackend *blk, int64_t sector_num, int nb_sectors)
         return ret;
     }
 
-    return bdrv_discard(blk->bs, sector_num, nb_sectors);
+    return bdrv_discard(blk_bs(blk), sector_num, nb_sectors);
 }
 
 int blk_save_vmstate(BlockBackend *blk, const uint8_t *buf,
                      int64_t pos, int size)
 {
-    return bdrv_save_vmstate(blk->bs, buf, pos, size);
+    int ret;
+
+    if (!blk_is_available(blk)) {
+        return -ENOMEDIUM;
+    }
+
+    ret = bdrv_save_vmstate(blk_bs(blk), buf, pos, size);
+    if (ret < 0) {
+        return ret;
+    }
+
+    if (ret == size && !blk->enable_write_cache) {
+        ret = bdrv_flush(blk_bs(blk));
+    }
+
+    return ret < 0 ? ret : size;
 }
 
 int blk_load_vmstate(BlockBackend *blk, uint8_t *buf, int64_t pos, int size)
 {
-    return bdrv_load_vmstate(blk->bs, buf, pos, size);
+    if (!blk_is_available(blk)) {
+        return -ENOMEDIUM;
+    }
+
+    return bdrv_load_vmstate(blk_bs(blk), buf, pos, size);
 }
 
 int blk_probe_blocksizes(BlockBackend *blk, BlockSizes *bsz)
 {
-    return bdrv_probe_blocksizes(blk->bs, bsz);
+    if (!blk_is_available(blk)) {
+        return -ENOMEDIUM;
+    }
+
+    return bdrv_probe_blocksizes(blk_bs(blk), bsz);
 }
 
 int blk_probe_geometry(BlockBackend *blk, HDGeometry *geo)
 {
-    return bdrv_probe_geometry(blk->bs, geo);
+    if (!blk_is_available(blk)) {
+        return -ENOMEDIUM;
+    }
+
+    return bdrv_probe_geometry(blk_bs(blk), geo);
+}
+
+/*
+ * Updates the BlockBackendRootState object with data from the currently
+ * attached BlockDriverState.
+ */
+void blk_update_root_state(BlockBackend *blk)
+{
+    assert(blk->root);
+
+    blk->root_state.open_flags    = blk->root->bs->open_flags;
+    blk->root_state.read_only     = blk->root->bs->read_only;
+    blk->root_state.detect_zeroes = blk->root->bs->detect_zeroes;
+
+    if (blk->root_state.throttle_group) {
+        g_free(blk->root_state.throttle_group);
+        throttle_group_unref(blk->root_state.throttle_state);
+    }
+    if (blk->root->bs->throttle_state) {
+        const char *name = throttle_group_get_name(blk->root->bs);
+        blk->root_state.throttle_group = g_strdup(name);
+        blk->root_state.throttle_state = throttle_group_incref(name);
+    } else {
+        blk->root_state.throttle_group = NULL;
+        blk->root_state.throttle_state = NULL;
+    }
+}
+
+/*
+ * Applies the information in the root state to the given BlockDriverState. This
+ * does not include the flags which have to be specified for bdrv_open(), use
+ * blk_get_open_flags_from_root_state() to inquire them.
+ */
+void blk_apply_root_state(BlockBackend *blk, BlockDriverState *bs)
+{
+    bs->detect_zeroes = blk->root_state.detect_zeroes;
+    if (blk->root_state.throttle_group) {
+        bdrv_io_limits_enable(bs, blk->root_state.throttle_group);
+    }
+}
+
+/*
+ * Returns the flags to be used for bdrv_open() of a BlockDriverState which is
+ * supposed to inherit the root state.
+ */
+int blk_get_open_flags_from_root_state(BlockBackend *blk)
+{
+    int bs_flags;
+
+    bs_flags = blk->root_state.read_only ? 0 : BDRV_O_RDWR;
+    bs_flags |= blk->root_state.open_flags & ~BDRV_O_RDWR;
+
+    return bs_flags;
+}
+
+BlockBackendRootState *blk_get_root_state(BlockBackend *blk)
+{
+    return &blk->root_state;
+}
+
+int blk_commit_all(void)
+{
+    BlockBackend *blk = NULL;
+
+    while ((blk = blk_all_next(blk)) != NULL) {
+        AioContext *aio_context = blk_get_aio_context(blk);
+
+        aio_context_acquire(aio_context);
+        if (blk_is_inserted(blk) && blk->root->bs->backing) {
+            int ret = bdrv_commit(blk->root->bs);
+            if (ret < 0) {
+                aio_context_release(aio_context);
+                return ret;
+            }
+        }
+        aio_context_release(aio_context);
+    }
+    return 0;
+}
+
+int blk_flush_all(void)
+{
+    BlockBackend *blk = NULL;
+    int result = 0;
+
+    while ((blk = blk_all_next(blk)) != NULL) {
+        AioContext *aio_context = blk_get_aio_context(blk);
+        int ret;
+
+        aio_context_acquire(aio_context);
+        if (blk_is_inserted(blk)) {
+            ret = blk_flush(blk);
+            if (ret < 0 && !result) {
+                result = ret;
+            }
+        }
+        aio_context_release(aio_context);
+    }
+
+    return result;
 }