Block layer patches

-----BEGIN PGP SIGNATURE-----
 Version: GnuPG v2.0.22 (GNU/Linux)
 
 iQIcBAABAgAGBQJXzpyJAAoJEH8JsnLIjy/WwToQAJ29bQ8dbVxybQtApZn0l3DH
 aBcguj822Vqa+KaxOLAfzkxmG5MurIvWzRQD1BvjxaprRykB+hDh4oAJCmVjfedP
 B28h24TUF+w8WIbpxf9weQFNpsT2Ire8ZySc0JZhpYMqxXCqy6NzDs98sjedDC0O
 jNbfic1L+yEpZumVE0Fzr4/YgPumt7wP0X42nb6G8R+VlChm3nweNCFF7hNQvTuB
 GNNbd9ckUS0BTcQazm04yRR/WzXW6uFqa00QeWsNGGd1mmZ0kUxiqxVgx/fuBMrL
 yC4LxFit7eNRoeVqu/nu8GsG+2Ol5zsalfJKFcoWmpg8pygOayc5SXecRUZRw7tg
 3oB7ZijbrBUFlr4y6cNVCGPtRluQshpLGHlgo68ulEIlHprqECwgPIdoOPr0bs+v
 Gb8ho2Y+lrISPIsjYWK5UFSmZf0SIBGILZUSD3lzQ+oOHXGKbdPAaFvSUqXENHSN
 xjtMYjr5t+NjrNNd2Q+VUJPlimHGw5jAowjsQSTk3ndcvJYeIVs+AwLqNTKc3dY4
 Oxx1IZ2RngDC63PmZUgh2Bs8pwFg7HaZJejmtq5jY8eHJZM/QMkCJX9TSRCZsMRB
 n0GxfCYabX526h3Yo94d74s5xRnHaC+Lem8PU/VEGR3/dMn21jf/PI9e+/0BnBTC
 iET2gkU70c4lunWkFHMd
 =LZpU
 -----END PGP SIGNATURE-----

Merge remote-tracking branch 'remotes/kevin/tags/for-upstream' into staging

Block layer patches

# gpg: Signature made Tue 06 Sep 2016 11:38:01 BST
# gpg:                using RSA key 0x7F09B272C88F2FD6
# gpg: Good signature from "Kevin Wolf <kwolf@redhat.com>"
# Primary key fingerprint: DC3D EB15 9A9A F95D 3D74  56FE 7F09 B272 C88F 2FD6

* remotes/kevin/tags/for-upstream: (36 commits)
  block: Allow node name for 'qemu-io' HMP command
  qemu-iotests: Log QMP traffic in debug mode
  block jobs: Improve error message for missing job ID
  coroutine: Assert that no locks are held on termination
  coroutine: Let CoMutex remember who holds it
  qcow2: fix iovec size at qcow2_co_pwritev_compressed
  test-coroutine: Fix coroutine pool corruption
  qemu-iotests: add vmdk for test backup compression in 055
  qemu-iotests: test backup compression in 055
  blockdev-backup: added support for data compression
  drive-backup: added support for data compression
  block: simplify blockdev-backup
  block: simplify drive-backup
  block/io: turn on dirty_bitmaps for the compressed writes
  block: remove BlockDriver.bdrv_write_compressed
  qcow: cleanup qcow_co_pwritev_compressed to avoid the recursion
  qcow: add qcow_co_pwritev_compressed
  vmdk: add vmdk_co_pwritev_compressed
  qcow2: cleanup qcow2_co_pwritev_compressed to avoid the recursion
  qcow2: add qcow2_co_pwritev_compressed
  ...

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>

block.c

@@ -25,6 +25,7 @@
 #include "trace.h"
 #include "block/block_int.h"
 #include "block/blockjob.h"
+#include "block/nbd.h"
 #include "qemu/error-report.h"
 #include "qemu/module.h"
 #include "qapi/qmp/qerror.h"
@@ -2206,6 +2207,7 @@ static void bdrv_close(BlockDriverState *bs)
 void bdrv_close_all(void)
 {
     block_job_cancel_sync_all();
+    nbd_export_close_all();
 
     /* Drop references from requests still in flight, such as canceled block
      * jobs whose AIO context has not been polled yet */

block/backup.c

@@ -47,6 +47,7 @@ typedef struct BackupBlockJob {
     uint64_t sectors_read;
     unsigned long *done_bitmap;
     int64_t cluster_size;
+    bool compress;
     NotifierWithReturn before_write;
     QLIST_HEAD(, CowRequest) inflight_reqs;
 } BackupBlockJob;
@@ -154,7 +155,8 @@ static int coroutine_fn backup_do_cow(BackupBlockJob *job,
                                        bounce_qiov.size, BDRV_REQ_MAY_UNMAP);
         } else {
             ret = blk_co_pwritev(job->target, start * job->cluster_size,
-                                 bounce_qiov.size, &bounce_qiov, 0);
+                                 bounce_qiov.size, &bounce_qiov,
+                                 job->compress ? BDRV_REQ_WRITE_COMPRESSED : 0);
         }
         if (ret < 0) {
             trace_backup_do_cow_write_fail(job, start, ret);
@@ -477,6 +479,7 @@ static void coroutine_fn backup_run(void *opaque)
 void backup_start(const char *job_id, BlockDriverState *bs,
                   BlockDriverState *target, int64_t speed,
                   MirrorSyncMode sync_mode, BdrvDirtyBitmap *sync_bitmap,
+                  bool compress,
                   BlockdevOnError on_source_error,
                   BlockdevOnError on_target_error,
                   BlockCompletionFunc *cb, void *opaque,
@@ -507,6 +510,12 @@ void backup_start(const char *job_id, BlockDriverState *bs,
         return;
     }
 
+    if (compress && target->drv->bdrv_co_pwritev_compressed == NULL) {
+        error_setg(errp, "Compression is not supported for this drive %s",
+                   bdrv_get_device_name(target));
+        return;
+    }
+
     if (bdrv_op_is_blocked(bs, BLOCK_OP_TYPE_BACKUP_SOURCE, errp)) {
         return;
     }
@@ -555,6 +564,7 @@ void backup_start(const char *job_id, BlockDriverState *bs,
     job->sync_mode = sync_mode;
     job->sync_bitmap = sync_mode == MIRROR_SYNC_MODE_INCREMENTAL ?
                        sync_bitmap : NULL;
+    job->compress = compress;
 
     /* If there is no backing file on the target, we cannot rely on COW if our
      * backup cluster size is smaller than the target cluster size. Even for

block/block-backend.c

@@ -409,6 +409,22 @@ bool bdrv_has_blk(BlockDriverState *bs)
     return bdrv_first_blk(bs) != NULL;
 }
 
+/*
+ * Returns true if @bs has only BlockBackends as parents.
+ */
+bool bdrv_is_root_node(BlockDriverState *bs)
+{
+    BdrvChild *c;
+
+    QLIST_FOREACH(c, &bs->parents, next_parent) {
+        if (c->role != &child_root) {
+            return false;
+        }
+    }
+
+    return true;
+}
+
 /*
  * Return @blk's DriveInfo if any, else null.
  */
@@ -727,21 +743,6 @@ static int blk_check_byte_request(BlockBackend *blk, int64_t offset,
     return 0;
 }
 
-static int blk_check_request(BlockBackend *blk, int64_t sector_num,
-                             int nb_sectors)
-{
-    if (sector_num < 0 || sector_num > INT64_MAX / BDRV_SECTOR_SIZE) {
-        return -EIO;
-    }
-
-    if (nb_sectors < 0 || nb_sectors > INT_MAX / BDRV_SECTOR_SIZE) {
-        return -EIO;
-    }
-
-    return blk_check_byte_request(blk, sector_num * BDRV_SECTOR_SIZE,
-                                  nb_sectors * BDRV_SECTOR_SIZE);
-}
-
 int coroutine_fn blk_co_preadv(BlockBackend *blk, int64_t offset,
                                unsigned int bytes, QEMUIOVector *qiov,
                                BdrvRequestFlags flags)
@@ -1484,15 +1485,11 @@ int coroutine_fn blk_co_pwrite_zeroes(BlockBackend *blk, int64_t offset,
                           flags | BDRV_REQ_ZERO_WRITE);
 }
 
-int blk_write_compressed(BlockBackend *blk, int64_t sector_num,
+int blk_pwrite_compressed(BlockBackend *blk, int64_t offset, const void *buf,
-                         const uint8_t *buf, int nb_sectors)
+                          int count)
 {
-    int ret = blk_check_request(blk, sector_num, nb_sectors);
+    return blk_prw(blk, offset, (void *) buf, count, blk_write_entry,
-    if (ret < 0) {
+                   BDRV_REQ_WRITE_COMPRESSED);
-        return ret;
-    }
-
-    return bdrv_write_compressed(blk_bs(blk), sector_num, buf, nb_sectors);
 }
 
 int blk_truncate(BlockBackend *blk, int64_t offset)

block/io.c

@@ -540,17 +540,6 @@ static int bdrv_check_byte_request(BlockDriverState *bs, int64_t offset,
     return 0;
 }
 
-static int bdrv_check_request(BlockDriverState *bs, int64_t sector_num,
-                              int nb_sectors)
-{
-    if (nb_sectors < 0 || nb_sectors > BDRV_REQUEST_MAX_SECTORS) {
-        return -EIO;
-    }
-
-    return bdrv_check_byte_request(bs, sector_num * BDRV_SECTOR_SIZE,
-                                   nb_sectors * BDRV_SECTOR_SIZE);
-}
-
 typedef struct RwCo {
     BdrvChild *child;
     int64_t offset;
@@ -897,6 +886,19 @@ emulate_flags:
     return ret;
 }
 
+static int coroutine_fn
+bdrv_driver_pwritev_compressed(BlockDriverState *bs, uint64_t offset,
+                               uint64_t bytes, QEMUIOVector *qiov)
+{
+    BlockDriver *drv = bs->drv;
+
+    if (!drv->bdrv_co_pwritev_compressed) {
+        return -ENOTSUP;
+    }
+
+    return drv->bdrv_co_pwritev_compressed(bs, offset, bytes, qiov);
+}
+
 static int coroutine_fn bdrv_co_do_copy_on_readv(BlockDriverState *bs,
         int64_t offset, unsigned int bytes, QEMUIOVector *qiov)
 {
@@ -1315,6 +1317,8 @@ static int coroutine_fn bdrv_aligned_pwritev(BlockDriverState *bs,
     } else if (flags & BDRV_REQ_ZERO_WRITE) {
         bdrv_debug_event(bs, BLKDBG_PWRITEV_ZERO);
         ret = bdrv_co_do_pwrite_zeroes(bs, offset, bytes, flags);
+    } else if (flags & BDRV_REQ_WRITE_COMPRESSED) {
+        ret = bdrv_driver_pwritev_compressed(bs, offset, bytes, qiov);
     } else if (bytes <= max_transfer) {
         bdrv_debug_event(bs, BLKDBG_PWRITEV);
         ret = bdrv_driver_pwritev(bs, offset, bytes, qiov, flags);
@@ -1879,28 +1883,6 @@ int bdrv_is_allocated_above(BlockDriverState *top,
     return 0;
 }
 
-int bdrv_write_compressed(BlockDriverState *bs, int64_t sector_num,
-                          const uint8_t *buf, int nb_sectors)
-{
-    BlockDriver *drv = bs->drv;
-    int ret;
-
-    if (!drv) {
-        return -ENOMEDIUM;
-    }
-    if (!drv->bdrv_write_compressed) {
-        return -ENOTSUP;
-    }
-    ret = bdrv_check_request(bs, sector_num, nb_sectors);
-    if (ret < 0) {
-        return ret;
-    }
-
-    assert(QLIST_EMPTY(&bs->dirty_bitmaps));
-
-    return drv->bdrv_write_compressed(bs, sector_num, buf, nb_sectors);
-}
-
 typedef struct BdrvVmstateCo {
     BlockDriverState    *bs;
     QEMUIOVector        *qiov;

block/qcow.c

@@ -913,75 +913,32 @@ static int qcow_make_empty(BlockDriverState *bs)
     return 0;
 }
 
-typedef struct QcowWriteCo {
-    BlockDriverState *bs;
-    int64_t sector_num;
-    const uint8_t *buf;
-    int nb_sectors;
-    int ret;
-} QcowWriteCo;
-
-static void qcow_write_co_entry(void *opaque)
-{
-    QcowWriteCo *co = opaque;
-    QEMUIOVector qiov;
-
-    struct iovec iov = (struct iovec) {
-        .iov_base   = (uint8_t*) co->buf,
-        .iov_len    = co->nb_sectors * BDRV_SECTOR_SIZE,
-    };
-    qemu_iovec_init_external(&qiov, &iov, 1);
-
-    co->ret = qcow_co_writev(co->bs, co->sector_num, co->nb_sectors, &qiov);
-}
-
-/* Wrapper for non-coroutine contexts */
-static int qcow_write(BlockDriverState *bs, int64_t sector_num,
-                      const uint8_t *buf, int nb_sectors)
-{
-    Coroutine *co;
-    AioContext *aio_context = bdrv_get_aio_context(bs);
-    QcowWriteCo data = {
-        .bs         = bs,
-        .sector_num = sector_num,
-        .buf        = buf,
-        .nb_sectors = nb_sectors,
-        .ret        = -EINPROGRESS,
-    };
-    co = qemu_coroutine_create(qcow_write_co_entry, &data);
-    qemu_coroutine_enter(co);
-    while (data.ret == -EINPROGRESS) {
-        aio_poll(aio_context, true);
-    }
-    return data.ret;
-}
-
 /* XXX: put compressed sectors first, then all the cluster aligned
    tables to avoid losing bytes in alignment */
-static int qcow_write_compressed(BlockDriverState *bs, int64_t sector_num,
+static coroutine_fn int
-                                 const uint8_t *buf, int nb_sectors)
+qcow_co_pwritev_compressed(BlockDriverState *bs, uint64_t offset,
+                           uint64_t bytes, QEMUIOVector *qiov)
 {
     BDRVQcowState *s = bs->opaque;
+    QEMUIOVector hd_qiov;
+    struct iovec iov;
     z_stream strm;
     int ret, out_len;
-    uint8_t *out_buf;
+    uint8_t *buf, *out_buf;
     uint64_t cluster_offset;
 
-    if (nb_sectors != s->cluster_sectors) {
+    buf = qemu_blockalign(bs, s->cluster_size);
-        ret = -EINVAL;
+    if (bytes != s->cluster_size) {
-
+        if (bytes > s->cluster_size ||
-        /* Zero-pad last write if image size is not cluster aligned */
+            offset + bytes != bs->total_sectors << BDRV_SECTOR_BITS)
-        if (sector_num + nb_sectors == bs->total_sectors &&
+        {
-            nb_sectors < s->cluster_sectors) {
+            qemu_vfree(buf);
-            uint8_t *pad_buf = qemu_blockalign(bs, s->cluster_size);
+            return -EINVAL;
-            memset(pad_buf, 0, s->cluster_size);
-            memcpy(pad_buf, buf, nb_sectors * BDRV_SECTOR_SIZE);
-            ret = qcow_write_compressed(bs, sector_num,
-                                        pad_buf, s->cluster_sectors);
-            qemu_vfree(pad_buf);
         }
-        return ret;
+        /* Zero-pad last write if image size is not cluster aligned */
+        memset(buf + bytes, 0, s->cluster_size - bytes);
     }
+    qemu_iovec_to_buf(qiov, 0, buf, qiov->size);
 
     out_buf = g_malloc(s->cluster_size);
 
@@ -1012,27 +969,35 @@ static int qcow_write_compressed(BlockDriverState *bs, int64_t sector_num,
 
     if (ret != Z_STREAM_END || out_len >= s->cluster_size) {
         /* could not compress: write normal cluster */
-        ret = qcow_write(bs, sector_num, buf, s->cluster_sectors);
+        ret = qcow_co_writev(bs, offset >> BDRV_SECTOR_BITS,
-        if (ret < 0) {
+                             bytes >> BDRV_SECTOR_BITS, qiov);
-            goto fail;
-        }
-    } else {
-        cluster_offset = get_cluster_offset(bs, sector_num << 9, 2,
-                                            out_len, 0, 0);
-        if (cluster_offset == 0) {
-            ret = -EIO;
-            goto fail;
-        }
-
-        cluster_offset &= s->cluster_offset_mask;
-        ret = bdrv_pwrite(bs->file, cluster_offset, out_buf, out_len);
         if (ret < 0) {
             goto fail;
         }
+        goto success;
     }
+    qemu_co_mutex_lock(&s->lock);
+    cluster_offset = get_cluster_offset(bs, offset, 2, out_len, 0, 0);
+    qemu_co_mutex_unlock(&s->lock);
+    if (cluster_offset == 0) {
+        ret = -EIO;
+        goto fail;
+    }
+    cluster_offset &= s->cluster_offset_mask;
 
+    iov = (struct iovec) {
+        .iov_base   = out_buf,
+        .iov_len    = out_len,
+    };
+    qemu_iovec_init_external(&hd_qiov, &iov, 1);
+    ret = bdrv_co_pwritev(bs->file, cluster_offset, out_len, &hd_qiov, 0);
+    if (ret < 0) {
+        goto fail;
+    }
+success:
     ret = 0;
 fail:
+    qemu_vfree(buf);
     g_free(out_buf);
     return ret;
 }
@@ -1085,7 +1050,7 @@ static BlockDriver bdrv_qcow = {
 
     .bdrv_set_key           = qcow_set_key,
     .bdrv_make_empty        = qcow_make_empty,
-    .bdrv_write_compressed  = qcow_write_compressed,
+    .bdrv_co_pwritev_compressed = qcow_co_pwritev_compressed,
     .bdrv_get_info          = qcow_get_info,
 
     .create_opts            = &qcow_create_opts,

block/qcow2.c

@@ -2533,84 +2533,39 @@ static int qcow2_truncate(BlockDriverState *bs, int64_t offset)
     return 0;
 }
 
-typedef struct Qcow2WriteCo {
-    BlockDriverState *bs;
-    int64_t sector_num;
-    const uint8_t *buf;
-    int nb_sectors;
-    int ret;
-} Qcow2WriteCo;
-
-static void qcow2_write_co_entry(void *opaque)
-{
-    Qcow2WriteCo *co = opaque;
-    QEMUIOVector qiov;
-    uint64_t offset = co->sector_num * BDRV_SECTOR_SIZE;
-    uint64_t bytes = co->nb_sectors * BDRV_SECTOR_SIZE;
-
-    struct iovec iov = (struct iovec) {
-        .iov_base   = (uint8_t*) co->buf,
-        .iov_len    = bytes,
-    };
-    qemu_iovec_init_external(&qiov, &iov, 1);
-
-    co->ret = qcow2_co_pwritev(co->bs, offset, bytes, &qiov, 0);
-}
-
-/* Wrapper for non-coroutine contexts */
-static int qcow2_write(BlockDriverState *bs, int64_t sector_num,
-                       const uint8_t *buf, int nb_sectors)
-{
-    Coroutine *co;
-    AioContext *aio_context = bdrv_get_aio_context(bs);
-    Qcow2WriteCo data = {
-        .bs         = bs,
-        .sector_num = sector_num,
-        .buf        = buf,
-        .nb_sectors = nb_sectors,
-        .ret        = -EINPROGRESS,
-    };
-    co = qemu_coroutine_create(qcow2_write_co_entry, &data);
-    qemu_coroutine_enter(co);
-    while (data.ret == -EINPROGRESS) {
-        aio_poll(aio_context, true);
-    }
-    return data.ret;
-}
-
 /* XXX: put compressed sectors first, then all the cluster aligned
    tables to avoid losing bytes in alignment */
-static int qcow2_write_compressed(BlockDriverState *bs, int64_t sector_num,
+static coroutine_fn int
-                                  const uint8_t *buf, int nb_sectors)
+qcow2_co_pwritev_compressed(BlockDriverState *bs, uint64_t offset,
+                            uint64_t bytes, QEMUIOVector *qiov)
 {
     BDRVQcow2State *s = bs->opaque;
+    QEMUIOVector hd_qiov;
+    struct iovec iov;
     z_stream strm;
     int ret, out_len;
-    uint8_t *out_buf;
+    uint8_t *buf, *out_buf;
     uint64_t cluster_offset;
 
-    if (nb_sectors == 0) {
+    if (bytes == 0) {
         /* align end of file to a sector boundary to ease reading with
            sector based I/Os */
         cluster_offset = bdrv_getlength(bs->file->bs);
         return bdrv_truncate(bs->file->bs, cluster_offset);
     }
 
-    if (nb_sectors != s->cluster_sectors) {
+    buf = qemu_blockalign(bs, s->cluster_size);
-        ret = -EINVAL;
+    if (bytes != s->cluster_size) {
-
+        if (bytes > s->cluster_size ||
-        /* Zero-pad last write if image size is not cluster aligned */
+            offset + bytes != bs->total_sectors << BDRV_SECTOR_BITS)
-        if (sector_num + nb_sectors == bs->total_sectors &&
+        {
-            nb_sectors < s->cluster_sectors) {
+            qemu_vfree(buf);
-            uint8_t *pad_buf = qemu_blockalign(bs, s->cluster_size);
+            return -EINVAL;
-            memset(pad_buf, 0, s->cluster_size);
-            memcpy(pad_buf, buf, nb_sectors * BDRV_SECTOR_SIZE);
-            ret = qcow2_write_compressed(bs, sector_num,
-                                         pad_buf, s->cluster_sectors);
-            qemu_vfree(pad_buf);
         }
-        return ret;
+        /* Zero-pad last write if image size is not cluster aligned */
+        memset(buf + bytes, 0, s->cluster_size - bytes);
     }
+    qemu_iovec_to_buf(qiov, 0, buf, bytes);
 
     out_buf = g_malloc(s->cluster_size);
 
@@ -2641,33 +2596,44 @@ static int qcow2_write_compressed(BlockDriverState *bs, int64_t sector_num,
 
     if (ret != Z_STREAM_END || out_len >= s->cluster_size) {
         /* could not compress: write normal cluster */
-        ret = qcow2_write(bs, sector_num, buf, s->cluster_sectors);
+        ret = qcow2_co_pwritev(bs, offset, bytes, qiov, 0);
-        if (ret < 0) {
-            goto fail;
-        }
-    } else {
-        cluster_offset = qcow2_alloc_compressed_cluster_offset(bs,
-            sector_num << 9, out_len);
-        if (!cluster_offset) {
-            ret = -EIO;
-            goto fail;
-        }
-        cluster_offset &= s->cluster_offset_mask;
-
-        ret = qcow2_pre_write_overlap_check(bs, 0, cluster_offset, out_len);
-        if (ret < 0) {
-            goto fail;
-        }
-
-        BLKDBG_EVENT(bs->file, BLKDBG_WRITE_COMPRESSED);
-        ret = bdrv_pwrite(bs->file, cluster_offset, out_buf, out_len);
         if (ret < 0) {
             goto fail;
         }
+        goto success;
     }
 
+    qemu_co_mutex_lock(&s->lock);
+    cluster_offset =
+        qcow2_alloc_compressed_cluster_offset(bs, offset, out_len);
+    if (!cluster_offset) {
+        qemu_co_mutex_unlock(&s->lock);
+        ret = -EIO;
+        goto fail;
+    }
+    cluster_offset &= s->cluster_offset_mask;
+
+    ret = qcow2_pre_write_overlap_check(bs, 0, cluster_offset, out_len);
+    qemu_co_mutex_unlock(&s->lock);
+    if (ret < 0) {
+        goto fail;
+    }
+
+    iov = (struct iovec) {
+        .iov_base   = out_buf,
+        .iov_len    = out_len,
+    };
+    qemu_iovec_init_external(&hd_qiov, &iov, 1);
+
+    BLKDBG_EVENT(bs->file, BLKDBG_WRITE_COMPRESSED);
+    ret = bdrv_co_pwritev(bs->file, cluster_offset, out_len, &hd_qiov, 0);
+    if (ret < 0) {
+        goto fail;
+    }
+success:
     ret = 0;
 fail:
+    qemu_vfree(buf);
     g_free(out_buf);
     return ret;
 }
@@ -3412,7 +3378,7 @@ BlockDriver bdrv_qcow2 = {
     .bdrv_co_pwrite_zeroes  = qcow2_co_pwrite_zeroes,
     .bdrv_co_pdiscard       = qcow2_co_pdiscard,
     .bdrv_truncate          = qcow2_truncate,
-    .bdrv_write_compressed  = qcow2_write_compressed,
+    .bdrv_co_pwritev_compressed = qcow2_co_pwritev_compressed,
     .bdrv_make_empty        = qcow2_make_empty,
 
     .bdrv_snapshot_create   = qcow2_snapshot_create,

block/vmdk.c

@@ -1645,56 +1645,11 @@ vmdk_co_pwritev(BlockDriverState *bs, uint64_t offset, uint64_t bytes,
     return ret;
 }
 
-typedef struct VmdkWriteCompressedCo {
+static int coroutine_fn
-    BlockDriverState *bs;
+vmdk_co_pwritev_compressed(BlockDriverState *bs, uint64_t offset,
-    int64_t sector_num;
+                           uint64_t bytes, QEMUIOVector *qiov)
-    const uint8_t *buf;
-    int nb_sectors;
-    int ret;
-} VmdkWriteCompressedCo;
-
-static void vmdk_co_write_compressed(void *opaque)
 {
-    VmdkWriteCompressedCo *co = opaque;
+    return vmdk_co_pwritev(bs, offset, bytes, qiov, 0);
-    QEMUIOVector local_qiov;
-    uint64_t offset = co->sector_num * BDRV_SECTOR_SIZE;
-    uint64_t bytes = co->nb_sectors * BDRV_SECTOR_SIZE;
-
-    struct iovec iov = (struct iovec) {
-        .iov_base   = (uint8_t*) co->buf,
-        .iov_len    = bytes,
-    };
-    qemu_iovec_init_external(&local_qiov, &iov, 1);
-
-    co->ret = vmdk_pwritev(co->bs, offset, bytes, &local_qiov, false, false);
-}
-
-static int vmdk_write_compressed(BlockDriverState *bs,
-                                 int64_t sector_num,
-                                 const uint8_t *buf,
-                                 int nb_sectors)
-{
-    BDRVVmdkState *s = bs->opaque;
-
-    if (s->num_extents == 1 && s->extents[0].compressed) {
-        Coroutine *co;
-        AioContext *aio_context = bdrv_get_aio_context(bs);
-        VmdkWriteCompressedCo data = {
-            .bs         = bs,
-            .sector_num = sector_num,
-            .buf        = buf,
-            .nb_sectors = nb_sectors,
-            .ret        = -EINPROGRESS,
-        };
-        co = qemu_coroutine_create(vmdk_co_write_compressed, &data);
-        qemu_coroutine_enter(co);
-        while (data.ret == -EINPROGRESS) {
-            aio_poll(aio_context, true);
-        }
-        return data.ret;
-    } else {
-        return -ENOTSUP;
-    }
 }
 
 static int coroutine_fn vmdk_co_pwrite_zeroes(BlockDriverState *bs,
@@ -2393,7 +2348,7 @@ static BlockDriver bdrv_vmdk = {
     .bdrv_reopen_prepare          = vmdk_reopen_prepare,
     .bdrv_co_preadv               = vmdk_co_preadv,
     .bdrv_co_pwritev              = vmdk_co_pwritev,
-    .bdrv_write_compressed        = vmdk_write_compressed,
+    .bdrv_co_pwritev_compressed   = vmdk_co_pwritev_compressed,
     .bdrv_co_pwrite_zeroes        = vmdk_co_pwrite_zeroes,
     .bdrv_close                   = vmdk_close,
     .bdrv_create                  = vmdk_create,

blockdev-nbd.c

@@ -145,7 +145,8 @@ void qmp_nbd_server_start(SocketAddress *addr,
 void qmp_nbd_server_add(const char *device, bool has_writable, bool writable,
                         Error **errp)
 {
-    BlockBackend *blk;
+    BlockDriverState *bs = NULL;
+    BlockBackend *on_eject_blk;
     NBDExport *exp;
 
     if (!nbd_server) {
@@ -158,26 +159,22 @@ void qmp_nbd_server_add(const char *device, bool has_writable, bool writable,
         return;
     }
 
-    blk = blk_by_name(device);
+    on_eject_blk = blk_by_name(device);
-    if (!blk) {
+
-        error_set(errp, ERROR_CLASS_DEVICE_NOT_FOUND,
+    bs = bdrv_lookup_bs(device, device, errp);
-                  "Device '%s' not found", device);
+    if (!bs) {
-        return;
-    }
-    if (!blk_is_inserted(blk)) {
-        error_setg(errp, QERR_DEVICE_HAS_NO_MEDIUM, device);
         return;
     }
 
     if (!has_writable) {
         writable = false;
     }
-    if (blk_is_read_only(blk)) {
+    if (bdrv_is_read_only(bs)) {
         writable = false;
     }
 
-    exp = nbd_export_new(blk, 0, -1, writable ? 0 : NBD_FLAG_READ_ONLY, NULL,
+    exp = nbd_export_new(bs, 0, -1, writable ? 0 : NBD_FLAG_READ_ONLY,
-                         errp);
+                         NULL, false, on_eject_blk, errp);
     if (!exp) {
         return;
     }

blockdev.c

@@ -1174,6 +1174,28 @@ fail:
     return dinfo;
 }
 
+static BlockDriverState *qmp_get_root_bs(const char *name, Error **errp)
+{
+    BlockDriverState *bs;
+
+    bs = bdrv_lookup_bs(name, name, errp);
+    if (bs == NULL) {
+        return NULL;
+    }
+
+    if (!bdrv_is_root_node(bs)) {
+        error_setg(errp, "Need a root block node");
+        return NULL;
+    }
+
+    if (!bdrv_is_inserted(bs)) {
+        error_setg(errp, "Device has no medium");
+        return NULL;
+    }
+
+    return bs;
+}
+
 void hmp_commit(Monitor *mon, const QDict *qdict)
 {
     const char *device = qdict_get_str(qdict, "device");
@@ -1284,21 +1306,17 @@ SnapshotInfo *qmp_blockdev_snapshot_delete_internal_sync(const char *device,
                                                          Error **errp)
 {
     BlockDriverState *bs;
-    BlockBackend *blk;
     AioContext *aio_context;
     QEMUSnapshotInfo sn;
     Error *local_err = NULL;
     SnapshotInfo *info = NULL;
     int ret;
 
-    blk = blk_by_name(device);
+    bs = qmp_get_root_bs(device, errp);
-    if (!blk) {
+    if (!bs) {
-        error_set(errp, ERROR_CLASS_DEVICE_NOT_FOUND,
-                  "Device '%s' not found", device);
         return NULL;
     }
-
+    aio_context = bdrv_get_aio_context(bs);
-    aio_context = blk_get_aio_context(blk);
     aio_context_acquire(aio_context);
 
     if (!has_id) {
@@ -1314,12 +1332,6 @@ SnapshotInfo *qmp_blockdev_snapshot_delete_internal_sync(const char *device,
         goto out_aio_context;
     }
 
-    if (!blk_is_available(blk)) {
-        error_setg(errp, "Device '%s' has no medium", device);
-        goto out_aio_context;
-    }
-    bs = blk_bs(blk);
-
     if (bdrv_op_is_blocked(bs, BLOCK_OP_TYPE_INTERNAL_SNAPSHOT_DELETE, errp)) {
         goto out_aio_context;
     }
@@ -1499,7 +1511,6 @@ static void internal_snapshot_prepare(BlkActionState *common,
     Error *local_err = NULL;
     const char *device;
     const char *name;
-    BlockBackend *blk;
     BlockDriverState *bs;
     QEMUSnapshotInfo old_sn, *sn;
     bool ret;
@@ -1522,23 +1533,15 @@ static void internal_snapshot_prepare(BlkActionState *common,
         return;
     }
 
-    blk = blk_by_name(device);
+    bs = qmp_get_root_bs(device, errp);
-    if (!blk) {
+    if (!bs) {
-        error_set(errp, ERROR_CLASS_DEVICE_NOT_FOUND,
-                  "Device '%s' not found", device);
         return;
     }
 
     /* AioContext is released in .clean() */
-    state->aio_context = blk_get_aio_context(blk);
+    state->aio_context = bdrv_get_aio_context(bs);
     aio_context_acquire(state->aio_context);
 
-    if (!blk_is_available(blk)) {
-        error_setg(errp, QERR_DEVICE_HAS_NO_MEDIUM, device);
-        return;
-    }
-    bs = blk_bs(blk);
-
     state->bs = bs;
     bdrv_drained_begin(bs);
 
@@ -1838,56 +1841,31 @@ typedef struct DriveBackupState {
     BlockJob *job;
 } DriveBackupState;
 
-static void do_drive_backup(const char *job_id, const char *device,
+static void do_drive_backup(DriveBackup *backup, BlockJobTxn *txn,
-                            const char *target, bool has_format,
+                            Error **errp);
-                            const char *format, enum MirrorSyncMode sync,
-                            bool has_mode, enum NewImageMode mode,
-                            bool has_speed, int64_t speed,
-                            bool has_bitmap, const char *bitmap,
-                            bool has_on_source_error,
-                            BlockdevOnError on_source_error,
-                            bool has_on_target_error,
-                            BlockdevOnError on_target_error,
-                            BlockJobTxn *txn, Error **errp);
 
 static void drive_backup_prepare(BlkActionState *common, Error **errp)
 {
     DriveBackupState *state = DO_UPCAST(DriveBackupState, common, common);
-    BlockBackend *blk;
+    BlockDriverState *bs;
     DriveBackup *backup;
     Error *local_err = NULL;
 
     assert(common->action->type == TRANSACTION_ACTION_KIND_DRIVE_BACKUP);
     backup = common->action->u.drive_backup.data;
 
-    blk = blk_by_name(backup->device);
+    bs = qmp_get_root_bs(backup->device, errp);
-    if (!blk) {
+    if (!bs) {
-        error_set(errp, ERROR_CLASS_DEVICE_NOT_FOUND,
-                  "Device '%s' not found", backup->device);
-        return;
-    }
-
-    if (!blk_is_available(blk)) {
-        error_setg(errp, QERR_DEVICE_HAS_NO_MEDIUM, backup->device);
         return;
     }
 
     /* AioContext is released in .clean() */
-    state->aio_context = blk_get_aio_context(blk);
+    state->aio_context = bdrv_get_aio_context(bs);
     aio_context_acquire(state->aio_context);
-    bdrv_drained_begin(blk_bs(blk));
+    bdrv_drained_begin(bs);
-    state->bs = blk_bs(blk);
+    state->bs = bs;
 
-    do_drive_backup(backup->has_job_id ? backup->job_id : NULL,
+    do_drive_backup(backup, common->block_job_txn, &local_err);
-                    backup->device, backup->target,
-                    backup->has_format, backup->format,
-                    backup->sync,
-                    backup->has_mode, backup->mode,
-                    backup->has_speed, backup->speed,
-                    backup->has_bitmap, backup->bitmap,
-                    backup->has_on_source_error, backup->on_source_error,
-                    backup->has_on_target_error, backup->on_target_error,
-                    common->block_job_txn, &local_err);
     if (local_err) {
         error_propagate(errp, local_err);
         return;
@@ -1924,34 +1902,21 @@ typedef struct BlockdevBackupState {
     AioContext *aio_context;
 } BlockdevBackupState;
 
-static void do_blockdev_backup(const char *job_id, const char *device,
+static void do_blockdev_backup(BlockdevBackup *backup, BlockJobTxn *txn,
-                               const char *target, enum MirrorSyncMode sync,
+                               Error **errp);
-                               bool has_speed, int64_t speed,
-                               bool has_on_source_error,
-                               BlockdevOnError on_source_error,
-                               bool has_on_target_error,
-                               BlockdevOnError on_target_error,
-                               BlockJobTxn *txn, Error **errp);
 
 static void blockdev_backup_prepare(BlkActionState *common, Error **errp)
 {
     BlockdevBackupState *state = DO_UPCAST(BlockdevBackupState, common, common);
     BlockdevBackup *backup;
-    BlockBackend *blk;
+    BlockDriverState *bs, *target;
-    BlockDriverState *target;
     Error *local_err = NULL;
 
     assert(common->action->type == TRANSACTION_ACTION_KIND_BLOCKDEV_BACKUP);
     backup = common->action->u.blockdev_backup.data;
 
-    blk = blk_by_name(backup->device);
+    bs = qmp_get_root_bs(backup->device, errp);
-    if (!blk) {
+    if (!bs) {
-        error_setg(errp, "Device '%s' not found", backup->device);
-        return;
-    }
-
-    if (!blk_is_available(blk)) {
-        error_setg(errp, QERR_DEVICE_HAS_NO_MEDIUM, backup->device);
         return;
     }
 
@@ -1961,22 +1926,17 @@ static void blockdev_backup_prepare(BlkActionState *common, Error **errp)
     }
 
     /* AioContext is released in .clean() */
-    state->aio_context = blk_get_aio_context(blk);
+    state->aio_context = bdrv_get_aio_context(bs);
     if (state->aio_context != bdrv_get_aio_context(target)) {
         state->aio_context = NULL;
         error_setg(errp, "Backup between two IO threads is not implemented");
         return;
     }
     aio_context_acquire(state->aio_context);
-    state->bs = blk_bs(blk);
+    state->bs = bs;
     bdrv_drained_begin(state->bs);
 
-    do_blockdev_backup(backup->has_job_id ? backup->job_id : NULL,
+    do_blockdev_backup(backup, common->block_job_txn, &local_err);
-                       backup->device, backup->target, backup->sync,
-                       backup->has_speed, backup->speed,
-                       backup->has_on_source_error, backup->on_source_error,
-                       backup->has_on_target_error, backup->on_target_error,
-                       common->block_job_txn, &local_err);
     if (local_err) {
         error_propagate(errp, local_err);
         return;
@@ -2983,7 +2943,6 @@ void qmp_block_stream(bool has_job_id, const char *job_id, const char *device,
                       bool has_on_error, BlockdevOnError on_error,
                       Error **errp)
 {
-    BlockBackend *blk;
     BlockDriverState *bs;
     BlockDriverState *base_bs = NULL;
     AioContext *aio_context;
@@ -2994,22 +2953,14 @@ void qmp_block_stream(bool has_job_id, const char *job_id, const char *device,
         on_error = BLOCKDEV_ON_ERROR_REPORT;
     }
 
-    blk = blk_by_name(device);
+    bs = qmp_get_root_bs(device, errp);
-    if (!blk) {
+    if (!bs) {
-        error_set(errp, ERROR_CLASS_DEVICE_NOT_FOUND,
-                  "Device '%s' not found", device);
         return;
     }
 
-    aio_context = blk_get_aio_context(blk);
+    aio_context = bdrv_get_aio_context(bs);
     aio_context_acquire(aio_context);
 
-    if (!blk_is_available(blk)) {
-        error_setg(errp, "Device '%s' has no medium", device);
-        goto out;
-    }
-    bs = blk_bs(blk);
-
     if (bdrv_op_is_blocked(bs, BLOCK_OP_TYPE_STREAM, errp)) {
         goto out;
     }
@@ -3055,7 +3006,6 @@ void qmp_block_commit(bool has_job_id, const char *job_id, const char *device,
                       bool has_speed, int64_t speed,
                       Error **errp)
 {
-    BlockBackend *blk;
     BlockDriverState *bs;
     BlockDriverState *base_bs, *top_bs;
     AioContext *aio_context;
@@ -3074,22 +3024,22 @@ void qmp_block_commit(bool has_job_id, const char *job_id, const char *device,
      *  live commit feature versions; for this to work, we must make sure to
      *  perform the device lookup before any generic errors that may occur in a
      *  scenario in which all optional arguments are omitted. */
-    blk = blk_by_name(device);
+    bs = qmp_get_root_bs(device, &local_err);
-    if (!blk) {
+    if (!bs) {
-        error_set(errp, ERROR_CLASS_DEVICE_NOT_FOUND,
+        bs = bdrv_lookup_bs(device, device, NULL);
-                  "Device '%s' not found", device);
+        if (!bs) {
+            error_free(local_err);
+            error_set(errp, ERROR_CLASS_DEVICE_NOT_FOUND,
+                      "Device '%s' not found", device);
+        } else {
+            error_propagate(errp, local_err);
+        }
         return;
     }
 
-    aio_context = blk_get_aio_context(blk);
+    aio_context = bdrv_get_aio_context(bs);
     aio_context_acquire(aio_context);
 
-    if (!blk_is_available(blk)) {
-        error_setg(errp, "Device '%s' has no medium", device);
-        goto out;
-    }
-    bs = blk_bs(blk);
-
     if (bdrv_op_is_blocked(bs, BLOCK_OP_TYPE_COMMIT_SOURCE, errp)) {
         goto out;
     }
@@ -3155,19 +3105,8 @@ out:
     aio_context_release(aio_context);
 }
 
-static void do_drive_backup(const char *job_id, const char *device,
+static void do_drive_backup(DriveBackup *backup, BlockJobTxn *txn, Error **errp)
-                            const char *target, bool has_format,
-                            const char *format, enum MirrorSyncMode sync,
-                            bool has_mode, enum NewImageMode mode,
-                            bool has_speed, int64_t speed,
-                            bool has_bitmap, const char *bitmap,
-                            bool has_on_source_error,
-                            BlockdevOnError on_source_error,
-                            bool has_on_target_error,
-                            BlockdevOnError on_target_error,
-                            BlockJobTxn *txn, Error **errp)
 {
-    BlockBackend *blk;
     BlockDriverState *bs;
     BlockDriverState *target_bs;
     BlockDriverState *source = NULL;
@@ -3178,39 +3117,36 @@ static void do_drive_backup(const char *job_id, const char *device,
     int flags;
     int64_t size;
 
-    if (!has_speed) {
+    if (!backup->has_speed) {
-        speed = 0;
+        backup->speed = 0;
     }
-    if (!has_on_source_error) {
+    if (!backup->has_on_source_error) {
-        on_source_error = BLOCKDEV_ON_ERROR_REPORT;
+        backup->on_source_error = BLOCKDEV_ON_ERROR_REPORT;
     }
-    if (!has_on_target_error) {
+    if (!backup->has_on_target_error) {
-        on_target_error = BLOCKDEV_ON_ERROR_REPORT;
+        backup->on_target_error = BLOCKDEV_ON_ERROR_REPORT;
     }
-    if (!has_mode) {
+    if (!backup->has_mode) {
-        mode = NEW_IMAGE_MODE_ABSOLUTE_PATHS;
+        backup->mode = NEW_IMAGE_MODE_ABSOLUTE_PATHS;
+    }
+    if (!backup->has_job_id) {
+        backup->job_id = NULL;
+    }
+    if (!backup->has_compress) {
+        backup->compress = false;
     }
 
-    blk = blk_by_name(device);
+    bs = qmp_get_root_bs(backup->device, errp);
-    if (!blk) {
+    if (!bs) {
-        error_set(errp, ERROR_CLASS_DEVICE_NOT_FOUND,
-                  "Device '%s' not found", device);
         return;
     }
 
-    aio_context = blk_get_aio_context(blk);
+    aio_context = bdrv_get_aio_context(bs);
     aio_context_acquire(aio_context);
 
-    /* Although backup_run has this check too, we need to use bs->drv below, so
+    if (!backup->has_format) {
-     * do an early check redundantly. */
+        backup->format = backup->mode == NEW_IMAGE_MODE_EXISTING ?
-    if (!blk_is_available(blk)) {
+                         NULL : (char*) bs->drv->format_name;
-        error_setg(errp, QERR_DEVICE_HAS_NO_MEDIUM, device);
-        goto out;
-    }
-    bs = blk_bs(blk);
-
-    if (!has_format) {
-        format = mode == NEW_IMAGE_MODE_EXISTING ? NULL : bs->drv->format_name;
     }
 
     /* Early check to avoid creating target */
@@ -3222,13 +3158,13 @@ static void do_drive_backup(const char *job_id, const char *device,
 
     /* See if we have a backing HD we can use to create our new image
      * on top of. */
-    if (sync == MIRROR_SYNC_MODE_TOP) {
+    if (backup->sync == MIRROR_SYNC_MODE_TOP) {
         source = backing_bs(bs);
         if (!source) {
-            sync = MIRROR_SYNC_MODE_FULL;
+            backup->sync = MIRROR_SYNC_MODE_FULL;
         }
     }
-    if (sync == MIRROR_SYNC_MODE_NONE) {
+    if (backup->sync == MIRROR_SYNC_MODE_NONE) {
         source = bs;
     }
 
@@ -3238,14 +3174,14 @@ static void do_drive_backup(const char *job_id, const char *device,
         goto out;
     }
 
-    if (mode != NEW_IMAGE_MODE_EXISTING) {
+    if (backup->mode != NEW_IMAGE_MODE_EXISTING) {
-        assert(format);
+        assert(backup->format);
         if (source) {
-            bdrv_img_create(target, format, source->filename,
+            bdrv_img_create(backup->target, backup->format, source->filename,
                             source->drv->format_name, NULL,
                             size, flags, &local_err, false);
         } else {
-            bdrv_img_create(target, format, NULL, NULL, NULL,
+            bdrv_img_create(backup->target, backup->format, NULL, NULL, NULL,
                             size, flags, &local_err, false);
         }
     }
@@ -3255,30 +3191,30 @@ static void do_drive_backup(const char *job_id, const char *device,
         goto out;
     }
 
-    if (format) {
+    if (backup->format) {
         options = qdict_new();
-        qdict_put(options, "driver", qstring_from_str(format));
+        qdict_put(options, "driver", qstring_from_str(backup->format));
     }
 
-    target_bs = bdrv_open(target, NULL, options, flags, errp);
+    target_bs = bdrv_open(backup->target, NULL, options, flags, errp);
     if (!target_bs) {
         goto out;
     }
 
     bdrv_set_aio_context(target_bs, aio_context);
 
-    if (has_bitmap) {
+    if (backup->has_bitmap) {
-        bmap = bdrv_find_dirty_bitmap(bs, bitmap);
+        bmap = bdrv_find_dirty_bitmap(bs, backup->bitmap);
         if (!bmap) {
-            error_setg(errp, "Bitmap '%s' could not be found", bitmap);
+            error_setg(errp, "Bitmap '%s' could not be found", backup->bitmap);
             bdrv_unref(target_bs);
             goto out;
         }
     }
 
-    backup_start(job_id, bs, target_bs, speed, sync, bmap,
+    backup_start(backup->job_id, bs, target_bs, backup->speed, backup->sync,
-                 on_source_error, on_target_error,
+                 bmap, backup->compress, backup->on_source_error,
-                 block_job_cb, bs, txn, &local_err);
+                 backup->on_target_error, block_job_cb, bs, txn, &local_err);
     bdrv_unref(target_bs);
     if (local_err != NULL) {
         error_propagate(errp, local_err);
@@ -3289,24 +3225,9 @@ out:
     aio_context_release(aio_context);
 }
 
-void qmp_drive_backup(bool has_job_id, const char *job_id,
+void qmp_drive_backup(DriveBackup *arg, Error **errp)
-                      const char *device, const char *target,
-                      bool has_format, const char *format,
-                      enum MirrorSyncMode sync,
-                      bool has_mode, enum NewImageMode mode,
-                      bool has_speed, int64_t speed,
-                      bool has_bitmap, const char *bitmap,
-                      bool has_on_source_error, BlockdevOnError on_source_error,
-                      bool has_on_target_error, BlockdevOnError on_target_error,
-                      Error **errp)
 {
-    return do_drive_backup(has_job_id ? job_id : NULL, device, target,
+    return do_drive_backup(arg, NULL, errp);
-                           has_format, format, sync,
-                           has_mode, mode, has_speed, speed,
-                           has_bitmap, bitmap,
-                           has_on_source_error, on_source_error,
-                           has_on_target_error, on_target_error,
-                           NULL, errp);
 }
 
 BlockDeviceInfoList *qmp_query_named_block_nodes(Error **errp)
@@ -3314,47 +3235,38 @@ BlockDeviceInfoList *qmp_query_named_block_nodes(Error **errp)
     return bdrv_named_nodes_list(errp);
 }
 
-void do_blockdev_backup(const char *job_id, const char *device,
+void do_blockdev_backup(BlockdevBackup *backup, BlockJobTxn *txn, Error **errp)
-                        const char *target, enum MirrorSyncMode sync,
-                         bool has_speed, int64_t speed,
-                         bool has_on_source_error,
-                         BlockdevOnError on_source_error,
-                         bool has_on_target_error,
-                         BlockdevOnError on_target_error,
-                         BlockJobTxn *txn, Error **errp)
 {
-    BlockBackend *blk;
     BlockDriverState *bs;
     BlockDriverState *target_bs;
     Error *local_err = NULL;
     AioContext *aio_context;
 
-    if (!has_speed) {
+    if (!backup->has_speed) {
-        speed = 0;
+        backup->speed = 0;
     }
-    if (!has_on_source_error) {
+    if (!backup->has_on_source_error) {
-        on_source_error = BLOCKDEV_ON_ERROR_REPORT;
+        backup->on_source_error = BLOCKDEV_ON_ERROR_REPORT;
     }
-    if (!has_on_target_error) {
+    if (!backup->has_on_target_error) {
-        on_target_error = BLOCKDEV_ON_ERROR_REPORT;
+        backup->on_target_error = BLOCKDEV_ON_ERROR_REPORT;
+    }
+    if (!backup->has_job_id) {
+        backup->job_id = NULL;
+    }
+    if (!backup->has_compress) {
+        backup->compress = false;
     }
 
-    blk = blk_by_name(device);
+    bs = qmp_get_root_bs(backup->device, errp);
-    if (!blk) {
+    if (!bs) {
-        error_setg(errp, "Device '%s' not found", device);
         return;
     }
 
-    aio_context = blk_get_aio_context(blk);
+    aio_context = bdrv_get_aio_context(bs);
     aio_context_acquire(aio_context);
 
-    if (!blk_is_available(blk)) {
+    target_bs = bdrv_lookup_bs(backup->target, backup->target, errp);
-        error_setg(errp, "Device '%s' has no medium", device);
-        goto out;
-    }
-    bs = blk_bs(blk);
-
-    target_bs = bdrv_lookup_bs(target, target, errp);
     if (!target_bs) {
         goto out;
     }
@@ -3370,8 +3282,9 @@ void do_blockdev_backup(const char *job_id, const char *device,
             goto out;
         }
     }
-    backup_start(job_id, bs, target_bs, speed, sync, NULL, on_source_error,
+    backup_start(backup->job_id, bs, target_bs, backup->speed, backup->sync,
-                 on_target_error, block_job_cb, bs, txn, &local_err);
+                 NULL, backup->compress, backup->on_source_error,
+                 backup->on_target_error, block_job_cb, bs, txn, &local_err);
     if (local_err != NULL) {
         error_propagate(errp, local_err);
     }
@@ -3379,21 +3292,9 @@ out:
     aio_context_release(aio_context);
 }
 
-void qmp_blockdev_backup(bool has_job_id, const char *job_id,
+void qmp_blockdev_backup(BlockdevBackup *arg, Error **errp)
-                         const char *device, const char *target,
-                         enum MirrorSyncMode sync,
-                         bool has_speed, int64_t speed,
-                         bool has_on_source_error,
-                         BlockdevOnError on_source_error,
-                         bool has_on_target_error,
-                         BlockdevOnError on_target_error,
-                         Error **errp)
 {
-    do_blockdev_backup(has_job_id ? job_id : NULL, device, target,
+    do_blockdev_backup(arg, NULL, errp);
-                       sync, has_speed, speed,
-                       has_on_source_error, on_source_error,
-                       has_on_target_error, on_target_error,
-                       NULL, errp);
 }
 
 /* Parameter check and block job starting for drive mirroring.
@@ -3469,7 +3370,6 @@ static void blockdev_mirror_common(const char *job_id, BlockDriverState *bs,
 void qmp_drive_mirror(DriveMirror *arg, Error **errp)
 {
     BlockDriverState *bs;
-    BlockBackend *blk;
     BlockDriverState *source, *target_bs;
     AioContext *aio_context;
     BlockMirrorBackingMode backing_mode;
@@ -3479,21 +3379,14 @@ void qmp_drive_mirror(DriveMirror *arg, Error **errp)
     int64_t size;
     const char *format = arg->format;
 
-    blk = blk_by_name(arg->device);
+    bs = qmp_get_root_bs(arg->device, errp);
-    if (!blk) {
+    if (!bs) {
-        error_set(errp, ERROR_CLASS_DEVICE_NOT_FOUND,
-                  "Device '%s' not found", arg->device);
         return;
     }
 
-    aio_context = blk_get_aio_context(blk);
+    aio_context = bdrv_get_aio_context(bs);
     aio_context_acquire(aio_context);
 
-    if (!blk_is_available(blk)) {
-        error_setg(errp, QERR_DEVICE_HAS_NO_MEDIUM, arg->device);
-        goto out;
-    }
-    bs = blk_bs(blk);
     if (!arg->has_mode) {
         arg->mode = NEW_IMAGE_MODE_ABSOLUTE_PATHS;
     }
@@ -3630,21 +3523,13 @@ void qmp_blockdev_mirror(bool has_job_id, const char *job_id,
                          Error **errp)
 {
     BlockDriverState *bs;
-    BlockBackend *blk;
     BlockDriverState *target_bs;
     AioContext *aio_context;
     BlockMirrorBackingMode backing_mode = MIRROR_LEAVE_BACKING_CHAIN;
     Error *local_err = NULL;
 
-    blk = blk_by_name(device);
+    bs = qmp_get_root_bs(device, errp);
-    if (!blk) {
-        error_setg(errp, "Device '%s' not found", device);
-        return;
-    }
-    bs = blk_bs(blk);
-
     if (!bs) {
-        error_setg(errp, "Device '%s' has no media", device);
         return;
     }
 
@@ -3785,7 +3670,6 @@ void qmp_change_backing_file(const char *device,
                              const char *backing_file,
                              Error **errp)
 {
-    BlockBackend *blk;
     BlockDriverState *bs = NULL;
     AioContext *aio_context;
     BlockDriverState *image_bs = NULL;
@@ -3794,22 +3678,14 @@ void qmp_change_backing_file(const char *device,
     int open_flags;
     int ret;
 
-    blk = blk_by_name(device);
+    bs = qmp_get_root_bs(device, errp);
-    if (!blk) {
+    if (!bs) {
-        error_set(errp, ERROR_CLASS_DEVICE_NOT_FOUND,
-                  "Device '%s' not found", device);
         return;
     }
 
-    aio_context = blk_get_aio_context(blk);
+    aio_context = bdrv_get_aio_context(bs);
     aio_context_acquire(aio_context);
 
-    if (!blk_is_available(blk)) {
-        error_setg(errp, "Device '%s' has no medium", device);
-        goto out;
-    }
-    bs = blk_bs(blk);
-
     image_bs = bdrv_lookup_bs(NULL, image_node_name, &local_err);
     if (local_err) {
         error_propagate(errp, local_err);

blockjob.c

@@ -132,6 +132,10 @@ void *block_job_create(const char *job_id, const BlockJobDriver *driver,
 
     if (job_id == NULL) {
         job_id = bdrv_get_device_name(bs);
+        if (!*job_id) {
+            error_setg(errp, "An explicit job ID is required for this node");
+            return NULL;
+        }
     }
 
     if (!id_wellformed(job_id)) {

hmp-commands.hx

@@ -1182,8 +1182,8 @@ ETEXI
 
     {
         .name       = "drive_backup",
-        .args_type  = "reuse:-n,full:-f,device:B,target:s,format:s?",
+        .args_type  = "reuse:-n,full:-f,compress:-c,device:B,target:s,format:s?",
-        .params     = "[-n] [-f] device target [format]",
+        .params     = "[-n] [-f] [-c] device target [format]",
         .help       = "initiates a point-in-time\n\t\t\t"
                       "copy for a device. The device's contents are\n\t\t\t"
                       "copied to the new image file, excluding data that\n\t\t\t"
@@ -1191,7 +1191,9 @@ ETEXI
                       "The -n flag requests QEMU to reuse the image found\n\t\t\t"
                       "in new-image-file, instead of recreating it from scratch.\n\t\t\t"
                       "The -f flag requests QEMU to copy the whole disk,\n\t\t\t"
-                      "so that the result does not need a backing file.\n\t\t\t",
+                      "so that the result does not need a backing file.\n\t\t\t"
+                      "The -c flag requests QEMU to compress backup data\n\t\t\t"
+                      "(if the target format supports it).\n\t\t\t",
         .mhandler.cmd = hmp_drive_backup,
     },
 STEXI

hmp.c

@@ -1109,8 +1109,19 @@ void hmp_drive_backup(Monitor *mon, const QDict *qdict)
     const char *format = qdict_get_try_str(qdict, "format");
     bool reuse = qdict_get_try_bool(qdict, "reuse", false);
     bool full = qdict_get_try_bool(qdict, "full", false);
-    enum NewImageMode mode;
+    bool compress = qdict_get_try_bool(qdict, "compress", false);
     Error *err = NULL;
+    DriveBackup backup = {
+        .device = (char *)device,
+        .target = (char *)filename,
+        .has_format = !!format,
+        .format = (char *)format,
+        .sync = full ? MIRROR_SYNC_MODE_FULL : MIRROR_SYNC_MODE_TOP,
+        .has_mode = true,
+        .mode = reuse ? NEW_IMAGE_MODE_EXISTING : NEW_IMAGE_MODE_ABSOLUTE_PATHS,
+        .has_compress = !!compress,
+        .compress = compress,
+    };
 
     if (!filename) {
         error_setg(&err, QERR_MISSING_PARAMETER, "target");
@@ -1118,16 +1129,7 @@ void hmp_drive_backup(Monitor *mon, const QDict *qdict)
         return;
     }
 
-    if (reuse) {
+    qmp_drive_backup(&backup, &err);
-        mode = NEW_IMAGE_MODE_EXISTING;
-    } else {
-        mode = NEW_IMAGE_MODE_ABSOLUTE_PATHS;
-    }
-
-    qmp_drive_backup(false, NULL, device, filename, !!format, format,
-                     full ? MIRROR_SYNC_MODE_FULL : MIRROR_SYNC_MODE_TOP,
-                     true, mode, false, 0, false, NULL,
-                     false, 0, false, 0, &err);
     hmp_handle_error(mon, &err);
 }
 
@@ -1921,11 +1923,22 @@ void hmp_chardev_remove(Monitor *mon, const QDict *qdict)
 void hmp_qemu_io(Monitor *mon, const QDict *qdict)
 {
     BlockBackend *blk;
+    BlockBackend *local_blk = NULL;
     const char* device = qdict_get_str(qdict, "device");
     const char* command = qdict_get_str(qdict, "command");
     Error *err = NULL;
 
     blk = blk_by_name(device);
+    if (!blk) {
+        BlockDriverState *bs = bdrv_lookup_bs(NULL, device, &err);
+        if (bs) {
+            blk = local_blk = blk_new();
+            blk_insert_bs(blk, bs);
+        } else {
+            goto fail;
+        }
+    }
+
     if (blk) {
         AioContext *aio_context = blk_get_aio_context(blk);
         aio_context_acquire(aio_context);
@@ -1938,6 +1951,8 @@ void hmp_qemu_io(Monitor *mon, const QDict *qdict)
                   "Device '%s' not found", device);
     }
 
+fail:
+    blk_unref(local_blk);
     hmp_handle_error(mon, &err);
 }
 

hw/ide/qdev.c

@@ -75,10 +75,6 @@ static int ide_qdev_init(DeviceState *qdev)
     IDEDeviceClass *dc = IDE_DEVICE_GET_CLASS(dev);
     IDEBus *bus = DO_UPCAST(IDEBus, qbus, qdev->parent_bus);
 
-    if (!dev->conf.blk) {
-        error_report("No drive specified");
-        goto err;
-    }
     if (dev->unit == -1) {
         dev->unit = bus->master ? 1 : 0;
     }
@@ -158,6 +154,16 @@ static int ide_dev_initfn(IDEDevice *dev, IDEDriveKind kind)
     IDEState *s = bus->ifs + dev->unit;
     Error *err = NULL;
 
+    if (!dev->conf.blk) {
+        if (kind != IDE_CD) {
+            error_report("No drive specified");
+            return -1;
+        } else {
+            /* Anonymous BlockBackend for an empty drive */
+            dev->conf.blk = blk_new();
+        }
+    }
+
     if (dev->conf.discard_granularity == -1) {
         dev->conf.discard_granularity = 512;
     } else if (dev->conf.discard_granularity &&
@@ -257,7 +263,11 @@ static int ide_cd_initfn(IDEDevice *dev)
 
 static int ide_drive_initfn(IDEDevice *dev)
 {
-    DriveInfo *dinfo = blk_legacy_dinfo(dev->conf.blk);
+    DriveInfo *dinfo = NULL;
+
+    if (dev->conf.blk) {
+        dinfo = blk_legacy_dinfo(dev->conf.blk);
+    }
 
     return ide_dev_initfn(dev, dinfo && dinfo->media_cd ? IDE_CD : IDE_HD);
 }

hw/scsi/scsi-disk.c

@@ -2359,6 +2359,11 @@ static void scsi_hd_realize(SCSIDevice *dev, Error **errp)
 static void scsi_cd_realize(SCSIDevice *dev, Error **errp)
 {
     SCSIDiskState *s = DO_UPCAST(SCSIDiskState, qdev, dev);
+
+    if (!dev->conf.blk) {
+        dev->conf.blk = blk_new();
+    }
+
     s->qdev.blocksize = 2048;
     s->qdev.type = TYPE_ROM;
     s->features |= 1 << SCSI_DISK_F_REMOVABLE;

include/block/block.h

@@ -65,9 +65,10 @@ typedef enum {
     BDRV_REQ_MAY_UNMAP          = 0x4,
     BDRV_REQ_NO_SERIALISING     = 0x8,
     BDRV_REQ_FUA                = 0x10,
+    BDRV_REQ_WRITE_COMPRESSED   = 0x20,
 
     /* Mask of valid flags */
-    BDRV_REQ_MASK               = 0x1f,
+    BDRV_REQ_MASK               = 0x3f,
 } BdrvRequestFlags;
 
 typedef struct BlockSizes {
@@ -399,8 +400,6 @@ const char *bdrv_get_node_name(const BlockDriverState *bs);
 const char *bdrv_get_device_name(const BlockDriverState *bs);
 const char *bdrv_get_device_or_node_name(const BlockDriverState *bs);
 int bdrv_get_flags(BlockDriverState *bs);
-int bdrv_write_compressed(BlockDriverState *bs, int64_t sector_num,
-                          const uint8_t *buf, int nb_sectors);
 int bdrv_get_info(BlockDriverState *bs, BlockDriverInfo *bdi);
 ImageInfoSpecific *bdrv_get_specific_info(BlockDriverState *bs);
 void bdrv_round_sectors_to_clusters(BlockDriverState *bs,

include/block/block_int.h

@@ -204,8 +204,8 @@ struct BlockDriver {
     bool has_variable_length;
     int64_t (*bdrv_get_allocated_file_size)(BlockDriverState *bs);
 
-    int (*bdrv_write_compressed)(BlockDriverState *bs, int64_t sector_num,
+    int coroutine_fn (*bdrv_co_pwritev_compressed)(BlockDriverState *bs,
-                                 const uint8_t *buf, int nb_sectors);
+        uint64_t offset, uint64_t bytes, QEMUIOVector *qiov);
 
     int (*bdrv_snapshot_create)(BlockDriverState *bs,
                                 QEMUSnapshotInfo *sn_info);
@@ -765,6 +765,7 @@ void mirror_start(const char *job_id, BlockDriverState *bs,
 void backup_start(const char *job_id, BlockDriverState *bs,
                   BlockDriverState *target, int64_t speed,
                   MirrorSyncMode sync_mode, BdrvDirtyBitmap *sync_bitmap,
+                  bool compress,
                   BlockdevOnError on_source_error,
                   BlockdevOnError on_target_error,
                   BlockCompletionFunc *cb, void *opaque,

include/block/nbd.h

@@ -103,8 +103,9 @@ int nbd_disconnect(int fd);
 typedef struct NBDExport NBDExport;
 typedef struct NBDClient NBDClient;
 
-NBDExport *nbd_export_new(BlockBackend *blk, off_t dev_offset, off_t size,
+NBDExport *nbd_export_new(BlockDriverState *bs, off_t dev_offset, off_t size,
                           uint16_t nbdflags, void (*close)(NBDExport *),
+                          bool writethrough, BlockBackend *on_eject_blk,
                           Error **errp);
 void nbd_export_close(NBDExport *exp);
 void nbd_export_get(NBDExport *exp);

include/qemu/coroutine.h

@@ -143,6 +143,7 @@ bool qemu_co_queue_empty(CoQueue *queue);
  */
 typedef struct CoMutex {
     bool locked;
+    Coroutine *holder;
     CoQueue queue;
 } CoMutex;
 

include/qemu/coroutine_int.h

@@ -39,6 +39,7 @@ struct Coroutine {
     void *entry_arg;
     Coroutine *caller;
     QSLIST_ENTRY(Coroutine) pool_next;
+    size_t locks_held;
 
     /* Coroutines that should be woken up when we yield or terminate */
     QSIMPLEQ_HEAD(, Coroutine) co_queue_wakeup;

include/sysemu/block-backend.h

@@ -98,6 +98,7 @@ BlockDriverState *blk_bs(BlockBackend *blk);
 void blk_remove_bs(BlockBackend *blk);
 void blk_insert_bs(BlockBackend *blk, BlockDriverState *bs);
 bool bdrv_has_blk(BlockDriverState *bs);
+bool bdrv_is_root_node(BlockDriverState *bs);
 
 void blk_set_allow_write_beyond_eof(BlockBackend *blk, bool allow);
 void blk_iostatus_enable(BlockBackend *blk);
@@ -203,8 +204,8 @@ void *blk_aio_get(const AIOCBInfo *aiocb_info, BlockBackend *blk,
                   BlockCompletionFunc *cb, void *opaque);
 int coroutine_fn blk_co_pwrite_zeroes(BlockBackend *blk, int64_t offset,
                                       int count, BdrvRequestFlags flags);
-int blk_write_compressed(BlockBackend *blk, int64_t sector_num,
+int blk_pwrite_compressed(BlockBackend *blk, int64_t offset, const void *buf,
-                         const uint8_t *buf, int nb_sectors);
+                          int count);
 int blk_truncate(BlockBackend *blk, int64_t offset);
 int blk_pdiscard(BlockBackend *blk, int64_t offset, int count);
 int blk_save_vmstate(BlockBackend *blk, const uint8_t *buf,

nbd/server.c

@@ -69,6 +69,7 @@ struct NBDExport {
 
     AioContext *ctx;
 
+    BlockBackend *eject_notifier_blk;
     Notifier eject_notifier;
 };
 
@@ -807,11 +808,18 @@ static void nbd_eject_notifier(Notifier *n, void *data)
     nbd_export_close(exp);
 }
 
-NBDExport *nbd_export_new(BlockBackend *blk, off_t dev_offset, off_t size,
+NBDExport *nbd_export_new(BlockDriverState *bs, off_t dev_offset, off_t size,
                           uint16_t nbdflags, void (*close)(NBDExport *),
+                          bool writethrough, BlockBackend *on_eject_blk,
                           Error **errp)
 {
+    BlockBackend *blk;
     NBDExport *exp = g_malloc0(sizeof(NBDExport));
+
+    blk = blk_new();
+    blk_insert_bs(blk, bs);
+    blk_set_enable_write_cache(blk, !writethrough);
+
     exp->refcount = 1;
     QTAILQ_INIT(&exp->clients);
     exp->blk = blk;
@@ -827,11 +835,14 @@ NBDExport *nbd_export_new(BlockBackend *blk, off_t dev_offset, off_t size,
 
     exp->close = close;
     exp->ctx = blk_get_aio_context(blk);
-    blk_ref(blk);
     blk_add_aio_context_notifier(blk, blk_aio_attached, blk_aio_detach, exp);
 
-    exp->eject_notifier.notify = nbd_eject_notifier;
+    if (on_eject_blk) {
-    blk_add_remove_bs_notifier(blk, &exp->eject_notifier);
+        blk_ref(on_eject_blk);
+        exp->eject_notifier_blk = on_eject_blk;
+        exp->eject_notifier.notify = nbd_eject_notifier;
+        blk_add_remove_bs_notifier(on_eject_blk, &exp->eject_notifier);
+    }
 
     /*
      * NBD exports are used for non-shared storage migration.  Make sure
@@ -844,6 +855,7 @@ NBDExport *nbd_export_new(BlockBackend *blk, off_t dev_offset, off_t size,
     return exp;
 
 fail:
+    blk_unref(blk);
     g_free(exp);
     return NULL;
 }
@@ -914,7 +926,10 @@ void nbd_export_put(NBDExport *exp)
         }
 
         if (exp->blk) {
-            notifier_remove(&exp->eject_notifier);
+            if (exp->eject_notifier_blk) {
+                notifier_remove(&exp->eject_notifier);
+                blk_unref(exp->eject_notifier_blk);
+            }
             blk_remove_aio_context_notifier(exp->blk, blk_aio_attached,
                                             blk_aio_detach, exp);
             blk_unref(exp->blk);

qapi/block-core.json

@@ -876,7 +876,7 @@
 # @job-id: #optional identifier for the newly-created block job. If
 #          omitted, the device name will be used. (Since 2.7)
 #
-# @device: the name of the device which should be copied.
+# @device: the device name or node-name of a root node which should be copied.
 #
 # @target: the target of the new image. If the file exists, or if it
 #          is a device, the existing file/device will be used as the new
@@ -898,6 +898,9 @@
 #          Must be present if sync is "incremental", must NOT be present
 #          otherwise. (Since 2.4)
 #
+# @compress: #optional true to compress data, if the target format supports it.
+#            (default: false) (since 2.7)
+#
 # @on-source-error: #optional the action to take on an error on the source,
 #                   default 'report'.  'stop' and 'enospc' can only be used
 #                   if the block device supports io-status (see BlockInfo).
@@ -915,7 +918,7 @@
 { 'struct': 'DriveBackup',
   'data': { '*job-id': 'str', 'device': 'str', 'target': 'str',
             '*format': 'str', 'sync': 'MirrorSyncMode', '*mode': 'NewImageMode',
-            '*speed': 'int', '*bitmap': 'str',
+            '*speed': 'int', '*bitmap': 'str', '*compress': 'bool',
             '*on-source-error': 'BlockdevOnError',
             '*on-target-error': 'BlockdevOnError' } }
 
@@ -925,7 +928,7 @@
 # @job-id: #optional identifier for the newly-created block job. If
 #          omitted, the device name will be used. (Since 2.7)
 #
-# @device: the name of the device which should be copied.
+# @device: the device name or node-name of a root node which should be copied.
 #
 # @target: the device name or node-name of the backup target node.
 #
@@ -936,6 +939,9 @@
 # @speed: #optional the maximum speed, in bytes per second. The default is 0,
 #         for unlimited.
 #
+# @compress: #optional true to compress data, if the target format supports it.
+#            (default: false) (since 2.7)
+#
 # @on-source-error: #optional the action to take on an error on the source,
 #                   default 'report'.  'stop' and 'enospc' can only be used
 #                   if the block device supports io-status (see BlockInfo).
@@ -954,6 +960,7 @@
   'data': { '*job-id': 'str', 'device': 'str', 'target': 'str',
             'sync': 'MirrorSyncMode',
             '*speed': 'int',
+            '*compress': 'bool',
             '*on-source-error': 'BlockdevOnError',
             '*on-target-error': 'BlockdevOnError' } }
 
@@ -998,7 +1005,8 @@
 # @image-node-name: The name of the block driver state node of the
 #                   image to modify.
 #
-# @device:          The name of the device that owns image-node-name.
+# @device:          The device name or node-name of the root node that owns
+#                   image-node-name.
 #
 # @backing-file:    The string to write as the backing file.  This
 #                   string is not validated, so care should be taken
@@ -1020,7 +1028,7 @@
 # @job-id: #optional identifier for the newly-created block job. If
 #          omitted, the device name will be used. (Since 2.7)
 #
-# @device:  the name of the device
+# @device:  the device name or node-name of a root node
 #
 # @base:   #optional The file name of the backing image to write data into.
 #                    If not specified, this is the deepest backing image
@@ -1086,11 +1094,12 @@
 # For the arguments, see the documentation of DriveBackup.
 #
 # Returns: nothing on success
-#          If @device is not a valid block device, DeviceNotFound
+#          If @device is not a valid block device, GenericError
 #
 # Since 1.6
 ##
-{ 'command': 'drive-backup', 'data': 'DriveBackup' }
+{ 'command': 'drive-backup', 'boxed': true,
+  'data': 'DriveBackup' }
 
 ##
 # @blockdev-backup
@@ -1103,9 +1112,13 @@
 #
 # For the arguments, see the documentation of BlockdevBackup.
 #
+# Returns: nothing on success
+#          If @device is not a valid block device, DeviceNotFound
+#
 # Since 2.3
 ##
-{ 'command': 'blockdev-backup', 'data': 'BlockdevBackup' }
+{ 'command': 'blockdev-backup', 'boxed': true,
+  'data': 'BlockdevBackup' }
 
 
 ##
@@ -1127,7 +1140,7 @@
 # See DriveMirror for parameter descriptions
 #
 # Returns: nothing on success
-#          If @device is not a valid block device, DeviceNotFound
+#          If @device is not a valid block device, GenericError
 #
 # Since 1.3
 ##
@@ -1142,7 +1155,8 @@
 # @job-id: #optional identifier for the newly-created block job. If
 #          omitted, the device name will be used. (Since 2.7)
 #
-# @device:  the name of the device whose writes should be mirrored.
+# @device:  the device name or node-name of a root node whose writes should be
+#           mirrored.
 #
 # @target: the target of the new image. If the file exists, or if it
 #          is a device, the existing file/device will be used as the new
@@ -1277,7 +1291,8 @@
 # @job-id: #optional identifier for the newly-created block job. If
 #          omitted, the device name will be used. (Since 2.7)
 #
-# @device: the name of the device whose writes should be mirrored.
+# @device: The device name or node-name of a root node whose writes should be
+#          mirrored.
 #
 # @target: the id or node-name of the block device to mirror to. This mustn't be
 #          attached to guest.
@@ -1462,7 +1477,7 @@
 # @job-id: #optional identifier for the newly-created block job. If
 #          omitted, the device name will be used. (Since 2.7)
 #
-# @device: the device name
+# @device: the device name or node-name of a root node
 #
 # @base:   #optional the common backing file name
 #
@@ -1487,9 +1502,6 @@
 #            'stop' and 'enospc' can only be used if the block device
 #            supports io-status (see BlockInfo).  Since 1.3.
 #
-# Returns: Nothing on success
-#          If @device does not exist, DeviceNotFound
-#
 # Since: 1.1
 ##
 { 'command': 'block-stream',

qapi/block.json

@@ -58,7 +58,8 @@
 ##
 # @BlockdevSnapshotInternal
 #
-# @device: the name of the device to generate the snapshot from
+# @device: the device name or node-name of a root node to generate the snapshot
+#          from
 #
 # @name: the name of the internal snapshot to be created
 #
@@ -80,7 +81,7 @@
 # For the arguments, see the documentation of BlockdevSnapshotInternal.
 #
 # Returns: nothing on success
-#          If @device is not a valid block device, DeviceNotFound
+#          If @device is not a valid block device, GenericError
 #          If any snapshot matching @name exists, or @name is empty,
 #          GenericError
 #          If the format of the image used does not support it,
@@ -99,14 +100,15 @@
 # both. One of the name or id is required. Return SnapshotInfo for the
 # successfully deleted snapshot.
 #
-# @device: the name of the device to delete the snapshot from
+# @device: the device name or node-name of a root node to delete the snapshot
+#          from
 #
 # @id: optional the snapshot's ID to be deleted
 #
 # @name: optional the snapshot's name to be deleted
 #
 # Returns: SnapshotInfo on success
-#          If @device is not a valid block device, DeviceNotFound
+#          If @device is not a valid block device, GenericError
 #          If snapshot not found, GenericError
 #          If the format of the image used does not support it,
 #          BlockFormatFeatureNotSupported
@@ -159,9 +161,9 @@
 ##
 # @nbd-server-add:
 #
-# Export a device to QEMU's embedded NBD server.
+# Export a block node to QEMU's embedded NBD server.
 #
-# @device: Block device to be exported
+# @device: The device name or node name of the node to be exported
 #
 # @writable: Whether clients should be able to write to the device via the
 #     NBD connection (default false). #optional

qemu-img.c

@@ -1590,7 +1590,9 @@ static int convert_write(ImgConvertState *s, int64_t sector_num, int nb_sectors,
                     break;
                 }
 
-                ret = blk_write_compressed(s->target, sector_num, buf, n);
+                ret = blk_pwrite_compressed(s->target,
+                                            sector_num << BDRV_SECTOR_BITS,
+                                            buf, n << BDRV_SECTOR_BITS);
                 if (ret < 0) {
                     return ret;
                 }
@@ -1727,7 +1729,7 @@ static int convert_do_copy(ImgConvertState *s)
 
     if (s->compressed) {
         /* signal EOF to align */
-        ret = blk_write_compressed(s->target, 0, NULL, 0);
+        ret = blk_pwrite_compressed(s->target, 0, NULL, 0);
         if (ret < 0) {
             goto fail;
         }
@@ -2032,7 +2034,7 @@ static int img_convert(int argc, char **argv)
         const char *preallocation =
             qemu_opt_get(opts, BLOCK_OPT_PREALLOC);
 
-        if (!drv->bdrv_write_compressed) {
+        if (!drv->bdrv_co_pwritev_compressed) {
             error_report("Compression not supported for this file format");
             ret = -1;
             goto out;

qemu-io-cmds.c

@@ -504,7 +504,7 @@ static int do_write_compressed(BlockBackend *blk, char *buf, int64_t offset,
         return -ERANGE;
     }
 
-    ret = blk_write_compressed(blk, offset >> 9, (uint8_t *)buf, count >> 9);
+    ret = blk_pwrite_compressed(blk, offset, buf, count);
     if (ret < 0) {
         return ret;
     }

qemu-nbd.c

@@ -910,8 +910,8 @@ int main(int argc, char **argv)
         }
     }
 
-    exp = nbd_export_new(blk, dev_offset, fd_size, nbdflags, nbd_export_closed,
+    exp = nbd_export_new(bs, dev_offset, fd_size, nbdflags, nbd_export_closed,
-                         &local_err);
+                         writethrough, NULL, &local_err);
     if (!exp) {
         error_report_err(local_err);
         exit(EXIT_FAILURE);

qmp-commands.hx

@@ -1120,7 +1120,7 @@ Arguments:
 
 - "job-id": Identifier for the newly-created block job. If omitted,
             the device name will be used. (json-string, optional)
-- "device": The device's ID, must be unique (json-string)
+- "device": The device name or node-name of a root node (json-string)
 - "base": The file name of the backing image above which copying starts
           (json-string, optional)
 - "backing-file": The backing file string to write into the active layer. This
@@ -1166,7 +1166,7 @@ Arguments:
 
 - "job-id": Identifier for the newly-created block job. If omitted,
             the device name will be used. (json-string, optional)
-- "device": The device's ID, must be unique (json-string)
+- "device": The device name or node-name of a root node (json-string)
 - "base": The file name of the backing image to write data into.
           If not specified, this is the deepest backing image
           (json-string, optional)
@@ -1217,7 +1217,8 @@ EQMP
     {
         .name       = "drive-backup",
         .args_type  = "job-id:s?,sync:s,device:B,target:s,speed:i?,mode:s?,"
-                      "format:s?,bitmap:s?,on-source-error:s?,on-target-error:s?",
+                      "format:s?,bitmap:s?,compress:b?,"
+                      "on-source-error:s?,on-target-error:s?",
         .mhandler.cmd_new = qmp_marshal_drive_backup,
     },
 
@@ -1235,7 +1236,7 @@ Arguments:
 
 - "job-id": Identifier for the newly-created block job. If omitted,
             the device name will be used. (json-string, optional)
-- "device": the name of the device which should be copied.
+- "device": the device name or node-name of a root node which should be copied.
             (json-string)
 - "target": the target of the new image. If the file exists, or if it is a
             device, the existing file/device will be used as the new
@@ -1253,6 +1254,8 @@ Arguments:
 - "mode": whether and how QEMU should create a new image
           (NewImageMode, optional, default 'absolute-paths')
 - "speed": the maximum speed, in bytes per second (json-int, optional)
+- "compress": true to compress data, if the target format supports it.
+              (json-bool, optional, default false)
 - "on-source-error": the action to take on an error on the source, default
                      'report'.  'stop' and 'enospc' can only be used
                      if the block device supports io-status.
@@ -1272,7 +1275,7 @@ EQMP
 
     {
         .name       = "blockdev-backup",
-        .args_type  = "job-id:s?,sync:s,device:B,target:B,speed:i?,"
+        .args_type  = "job-id:s?,sync:s,device:B,target:B,speed:i?,compress:b?,"
                       "on-source-error:s?,on-target-error:s?",
         .mhandler.cmd_new = qmp_marshal_blockdev_backup,
     },
@@ -1288,7 +1291,7 @@ Arguments:
 
 - "job-id": Identifier for the newly-created block job. If omitted,
             the device name will be used. (json-string, optional)
-- "device": the name of the device which should be copied.
+- "device": the device name or node-name of a root node which should be copied.
             (json-string)
 - "target": the name of the backup target device. (json-string)
 - "sync": what parts of the disk image should be copied to the destination;
@@ -1296,6 +1299,8 @@ Arguments:
           sectors allocated in the topmost image, or "none" to only replicate
           new I/O (MirrorSyncMode).
 - "speed": the maximum speed, in bytes per second (json-int, optional)
+- "compress": true to compress data, if the target format supports it.
+              (json-bool, optional, default false)
 - "on-source-error": the action to take on an error on the source, default
                      'report'.  'stop' and 'enospc' can only be used
                      if the block device supports io-status.
@@ -1407,7 +1412,8 @@ actions array:
       - "mode": whether and how QEMU should create the snapshot file
         (NewImageMode, optional, default "absolute-paths")
       When "type" is "blockdev-snapshot-internal-sync":
-      - "device": device name to snapshot (json-string)
+      - "device": the device name or node-name of a root node to snapshot
+                  (json-string)
       - "name": name of the new snapshot (json-string)
 
 Example:
@@ -1608,7 +1614,8 @@ name already exists, the operation will fail.
 
 Arguments:
 
-- "device": device name to snapshot (json-string)
+- "device": the device name or node-name of a root node to snapshot
+            (json-string)
 - "name": name of the new snapshot (json-string)
 
 Example:
@@ -1639,7 +1646,7 @@ fail.
 
 Arguments:
 
-- "device": device name (json-string)
+- "device": the device name or node-name of a root node (json-string)
 - "id": ID of the snapshot (json-string, optional)
 - "name": name of the snapshot (json-string, optional)
 
@@ -1687,7 +1694,8 @@ Arguments:
 
 - "job-id": Identifier for the newly-created block job. If omitted,
             the device name will be used. (json-string, optional)
-- "device": device name to operate on (json-string)
+- "device": the device name or node-name of a root node whose writes should be
+            mirrored. (json-string)
 - "target": name of new image file (json-string)
 - "format": format of new image (json-string, optional)
 - "node-name": the name of the new block driver state in the node graph
@@ -1747,7 +1755,8 @@ Arguments:
 
 - "job-id": Identifier for the newly-created block job. If omitted,
             the device name will be used. (json-string, optional)
-- "device": device name to operate on (json-string)
+- "device": The device name or node-name of a root node whose writes should be
+            mirrored (json-string)
 - "target": device name to mirror to (json-string)
 - "replaces": the block driver node name to replace when finished
               (json-string, optional)
@@ -1803,7 +1812,8 @@ Arguments:
                         "device".
                         (json-string, optional)
 
-- "device":             The name of the device.
+- "device":             The device name or node-name of the root node that owns
+                        image-node-name.
                         (json-string)
 
 - "backing-file":       The string to write as the backing file.  This string is

tests/qemu-iotests/030

@@ -126,7 +126,7 @@ class TestSingleDrive(iotests.QMPTestCase):
 
     def test_device_not_found(self):
         result = self.vm.qmp('block-stream', device='nonexistent')
-        self.assert_qmp(result, 'error/class', 'DeviceNotFound')
+        self.assert_qmp(result, 'error/class', 'GenericError')
 
 
 class TestSmallerBackingFile(iotests.QMPTestCase):

tests/qemu-iotests/041

@@ -38,7 +38,6 @@ class TestSingleDrive(iotests.QMPTestCase):
     image_len = 1 * 1024 * 1024 # MB
     qmp_cmd = 'drive-mirror'
     qmp_target = target_img
-    not_found_error = 'DeviceNotFound'
 
     def setUp(self):
         iotests.create_image(backing_img, self.image_len)
@@ -176,7 +175,7 @@ class TestSingleDrive(iotests.QMPTestCase):
 
         result = self.vm.qmp(self.qmp_cmd, device='ide1-cd0', sync='full',
                              target=self.qmp_target)
-        self.assert_qmp(result, 'error/class', self.not_found_error)
+        self.assert_qmp(result, 'error/class', 'GenericError')
 
     def test_image_not_found(self):
         result = self.vm.qmp(self.qmp_cmd, device='drive0', sync='full',
@@ -186,12 +185,11 @@ class TestSingleDrive(iotests.QMPTestCase):
     def test_device_not_found(self):
         result = self.vm.qmp(self.qmp_cmd, device='nonexistent', sync='full',
                              target=self.qmp_target)
-        self.assert_qmp(result, 'error/class', self.not_found_error)
+        self.assert_qmp(result, 'error/class', 'GenericError')
 
 class TestSingleBlockdev(TestSingleDrive):
     qmp_cmd = 'blockdev-mirror'
     qmp_target = 'node1'
-    not_found_error = 'GenericError'
 
     def setUp(self):
         TestSingleDrive.setUp(self)
@@ -922,7 +920,7 @@ class TestRepairQuorum(iotests.QMPTestCase):
                              node_name='repair0',
                              replaces='img1',
                              target=quorum_repair_img, format=iotests.imgfmt)
-        self.assert_qmp(result, 'error/class', 'DeviceNotFound')
+        self.assert_qmp(result, 'error/class', 'GenericError')
 
     def test_wrong_sync_mode(self):
         if not self.has_quorum():

tests/qemu-iotests/055

@@ -134,10 +134,7 @@ class TestSingleDrive(iotests.QMPTestCase):
 
     def do_test_device_not_found(self, cmd, **args):
         result = self.vm.qmp(cmd, **args)
-        if cmd == 'drive-backup':
+        self.assert_qmp(result, 'error/class', 'GenericError')
-            self.assert_qmp(result, 'error/class', 'DeviceNotFound')
-        else:
-            self.assert_qmp(result, 'error/class', 'GenericError')
 
     def test_device_not_found(self):
         self.do_test_device_not_found('drive-backup', device='nonexistent',
@@ -371,7 +368,7 @@ class TestSingleTransaction(iotests.QMPTestCase):
                           'sync': 'full' },
             }
         ])
-        self.assert_qmp(result, 'error/class', 'DeviceNotFound')
+        self.assert_qmp(result, 'error/class', 'GenericError')
 
         result = self.vm.qmp('transaction', actions=[{
                 'type': 'blockdev-backup',
@@ -451,5 +448,123 @@ class TestSingleTransaction(iotests.QMPTestCase):
         self.assert_qmp(result, 'error/class', 'GenericError')
         self.assert_no_active_block_jobs()
 
+
+class TestDriveCompression(iotests.QMPTestCase):
+    image_len = 64 * 1024 * 1024 # MB
+    fmt_supports_compression = [{'type': 'qcow2', 'args': ()},
+                                {'type': 'vmdk', 'args': ('-o', 'subformat=streamOptimized')}]
+
+    def setUp(self):
+        # Write data to the image so we can compare later
+        qemu_img('create', '-f', iotests.imgfmt, test_img, str(TestDriveCompression.image_len))
+        qemu_io('-f', iotests.imgfmt, '-c', 'write -P0x11 0 64k', test_img)
+        qemu_io('-f', iotests.imgfmt, '-c', 'write -P0x00 64k 128k', test_img)
+        qemu_io('-f', iotests.imgfmt, '-c', 'write -P0x22 162k 32k', test_img)
+        qemu_io('-f', iotests.imgfmt, '-c', 'write -P0x33 67043328 64k', test_img)
+
+    def tearDown(self):
+        self.vm.shutdown()
+        os.remove(test_img)
+        os.remove(blockdev_target_img)
+        try:
+            os.remove(target_img)
+        except OSError:
+            pass
+
+    def do_prepare_drives(self, fmt, args):
+        self.vm = iotests.VM().add_drive(test_img)
+
+        qemu_img('create', '-f', fmt, blockdev_target_img,
+                 str(TestDriveCompression.image_len), *args)
+        self.vm.add_drive(blockdev_target_img, format=fmt)
+
+        self.vm.launch()
+
+    def do_test_compress_complete(self, cmd, format, **args):
+        self.do_prepare_drives(format['type'], format['args'])
+
+        self.assert_no_active_block_jobs()
+
+        result = self.vm.qmp(cmd, device='drive0', sync='full', compress=True, **args)
+        self.assert_qmp(result, 'return', {})
+
+        self.wait_until_completed()
+
+        self.vm.shutdown()
+        self.assertTrue(iotests.compare_images(test_img, blockdev_target_img,
+                                               iotests.imgfmt, format['type']),
+                        'target image does not match source after backup')
+
+    def test_complete_compress_drive_backup(self):
+        for format in TestDriveCompression.fmt_supports_compression:
+            self.do_test_compress_complete('drive-backup', format,
+                                           target=blockdev_target_img, mode='existing')
+
+    def test_complete_compress_blockdev_backup(self):
+        for format in TestDriveCompression.fmt_supports_compression:
+            self.do_test_compress_complete('blockdev-backup', format, target='drive1')
+
+    def do_test_compress_cancel(self, cmd, format, **args):
+        self.do_prepare_drives(format['type'], format['args'])
+
+        self.assert_no_active_block_jobs()
+
+        result = self.vm.qmp(cmd, device='drive0', sync='full', compress=True, **args)
+        self.assert_qmp(result, 'return', {})
+
+        event = self.cancel_and_wait()
+        self.assert_qmp(event, 'data/type', 'backup')
+
+        self.vm.shutdown()
+
+    def test_compress_cancel_drive_backup(self):
+        for format in TestDriveCompression.fmt_supports_compression:
+            self.do_test_compress_cancel('drive-backup', format,
+                                         target=blockdev_target_img, mode='existing')
+
+    def test_compress_cancel_blockdev_backup(self):
+       for format in TestDriveCompression.fmt_supports_compression:
+            self.do_test_compress_cancel('blockdev-backup', format, target='drive1')
+
+    def do_test_compress_pause(self, cmd, format, **args):
+        self.do_prepare_drives(format['type'], format['args'])
+
+        self.assert_no_active_block_jobs()
+
+        self.vm.pause_drive('drive0')
+        result = self.vm.qmp(cmd, device='drive0', sync='full', compress=True, **args)
+        self.assert_qmp(result, 'return', {})
+
+        result = self.vm.qmp('block-job-pause', device='drive0')
+        self.assert_qmp(result, 'return', {})
+
+        self.vm.resume_drive('drive0')
+        time.sleep(1)
+        result = self.vm.qmp('query-block-jobs')
+        offset = self.dictpath(result, 'return[0]/offset')
+
+        time.sleep(1)
+        result = self.vm.qmp('query-block-jobs')
+        self.assert_qmp(result, 'return[0]/offset', offset)
+
+        result = self.vm.qmp('block-job-resume', device='drive0')
+        self.assert_qmp(result, 'return', {})
+
+        self.wait_until_completed()
+
+        self.vm.shutdown()
+        self.assertTrue(iotests.compare_images(test_img, blockdev_target_img,
+                                               iotests.imgfmt, format['type']),
+                        'target image does not match source after backup')
+
+    def test_compress_pause_drive_backup(self):
+        for format in TestDriveCompression.fmt_supports_compression:
+            self.do_test_compress_pause('drive-backup', format,
+                                        target=blockdev_target_img, mode='existing')
+
+    def test_compress_pause_blockdev_backup(self):
+        for format in TestDriveCompression.fmt_supports_compression:
+            self.do_test_compress_pause('blockdev-backup', format, target='drive1')
+
 if __name__ == '__main__':
     iotests.main(supported_fmts=['raw', 'qcow2'])

tests/qemu-iotests/055.out

@@ -1,5 +1,5 @@
-........................
+..............................
 ----------------------------------------------------------------------
-Ran 24 tests
+Ran 30 tests
 
 OK

tests/qemu-iotests/057

@@ -182,7 +182,7 @@ class TestSingleTransaction(ImageSnapshotTestCase):
                               'name': 'a' },
                   }]
         result = self.vm.qmp('transaction', actions = actions)
-        self.assert_qmp(result, 'error/class', 'DeviceNotFound')
+        self.assert_qmp(result, 'error/class', 'GenericError')
 
     def test_error_exist(self):
         self.createSnapshotInTransaction(1)
@@ -239,7 +239,7 @@ class TestSnapshotDelete(ImageSnapshotTestCase):
         result = self.vm.qmp('blockdev-snapshot-delete-internal-sync',
                               device = 'drive_error',
                               id = '0')
-        self.assert_qmp(result, 'error/class', 'DeviceNotFound')
+        self.assert_qmp(result, 'error/class', 'GenericError')
 
     def test_error_no_id_and_name(self):
         result = self.vm.qmp('blockdev-snapshot-delete-internal-sync',

tests/qemu-iotests/iotests.py

@@ -50,6 +50,7 @@ cachemode = os.environ.get('CACHEMODE')
 qemu_default_machine = os.environ.get('QEMU_DEFAULT_MACHINE')
 
 socket_scm_helper = os.environ.get('SOCKET_SCM_HELPER', 'socket_scm_helper')
+debug = False
 
 def qemu_img(*args):
     '''Run qemu-img and return the exit code'''
@@ -86,10 +87,10 @@ def qemu_io(*args):
         sys.stderr.write('qemu-io received signal %i: %s\n' % (-exitcode, ' '.join(args)))
     return subp.communicate()[0]
 
-def compare_images(img1, img2):
+def compare_images(img1, img2, fmt1=imgfmt, fmt2=imgfmt):
     '''Return True if two image files are identical'''
-    return qemu_img('compare', '-f', imgfmt,
+    return qemu_img('compare', '-f', fmt1,
-                    '-F', imgfmt, img1, img2) == 0
+                    '-F', fmt2, img1, img2) == 0
 
 def create_image(name, size):
     '''Create a fully-allocated raw image with sector markers'''
@@ -134,6 +135,8 @@ class VM(qtest.QEMUQtestMachine):
     def __init__(self):
         super(VM, self).__init__(qemu_prog, qemu_opts, test_dir=test_dir,
                                  socket_scm_helper=socket_scm_helper)
+        if debug:
+            self._debug = True
         self._num_drives = 0
 
     def add_drive_raw(self, opts):
@@ -141,14 +144,14 @@ class VM(qtest.QEMUQtestMachine):
         self._args.append(opts)
         return self
 
-    def add_drive(self, path, opts='', interface='virtio'):
+    def add_drive(self, path, opts='', interface='virtio', format=imgfmt):
         '''Add a virtio-blk drive to the VM'''
         options = ['if=%s' % interface,
                    'id=drive%d' % self._num_drives]
 
         if path is not None:
             options.append('file=%s' % path)
-            options.append('format=%s' % imgfmt)
+            options.append('format=%s' % format)
             options.append('cache=%s' % cachemode)
 
         if opts:
@@ -318,6 +321,8 @@ def verify_quorum():
 def main(supported_fmts=[], supported_oses=['linux']):
     '''Run tests'''
 
+    global debug
+
     # We are using TEST_DIR and QEMU_DEFAULT_MACHINE as proxies to
     # indicate that we're not being run via "check". There may be
     # other things set up by "check" that individual test cases rely

tests/test-coroutine.c

@@ -139,13 +139,20 @@ static void test_co_queue(void)
 {
     Coroutine *c1;
     Coroutine *c2;
+    Coroutine tmp;
 
     c2 = qemu_coroutine_create(c2_fn, NULL);
     c1 = qemu_coroutine_create(c1_fn, c2);
 
     qemu_coroutine_enter(c1);
+
+    /* c1 shouldn't be used any more now; make sure we segfault if it is */
+    tmp = *c1;
     memset(c1, 0xff, sizeof(Coroutine));
     qemu_coroutine_enter(c2);
+
+    /* Must restore the coroutine now to avoid corrupted pool */
+    *c1 = tmp;
 }
 
 /*

util/qemu-coroutine-lock.c

@@ -129,6 +129,8 @@ void coroutine_fn qemu_co_mutex_lock(CoMutex *mutex)
     }
 
     mutex->locked = true;
+    mutex->holder = self;
+    self->locks_held++;
 
     trace_qemu_co_mutex_lock_return(mutex, self);
 }
@@ -140,9 +142,12 @@ void coroutine_fn qemu_co_mutex_unlock(CoMutex *mutex)
     trace_qemu_co_mutex_unlock_entry(mutex, self);
 
     assert(mutex->locked == true);
+    assert(mutex->holder == self);
     assert(qemu_in_coroutine());
 
     mutex->locked = false;
+    mutex->holder = NULL;
+    self->locks_held--;
     qemu_co_queue_next(&mutex->queue);
 
     trace_qemu_co_mutex_unlock_return(mutex, self);
@@ -156,14 +161,19 @@ void qemu_co_rwlock_init(CoRwlock *lock)
 
 void qemu_co_rwlock_rdlock(CoRwlock *lock)
 {
+    Coroutine *self = qemu_coroutine_self();
+
     while (lock->writer) {
         qemu_co_queue_wait(&lock->queue);
     }
     lock->reader++;
+    self->locks_held++;
 }
 
 void qemu_co_rwlock_unlock(CoRwlock *lock)
 {
+    Coroutine *self = qemu_coroutine_self();
+
     assert(qemu_in_coroutine());
     if (lock->writer) {
         lock->writer = false;
@@ -176,12 +186,16 @@ void qemu_co_rwlock_unlock(CoRwlock *lock)
             qemu_co_queue_next(&lock->queue);
         }
     }
+    self->locks_held--;
 }
 
 void qemu_co_rwlock_wrlock(CoRwlock *lock)
 {
+    Coroutine *self = qemu_coroutine_self();
+
     while (lock->writer || lock->reader) {
         qemu_co_queue_wait(&lock->queue);
     }
     lock->writer = true;
+    self->locks_held++;
 }

util/qemu-coroutine.c

@@ -122,6 +122,7 @@ void qemu_coroutine_enter(Coroutine *co)
     case COROUTINE_YIELD:
         return;
     case COROUTINE_TERMINATE:
+        assert(!co->locks_held);
         trace_qemu_coroutine_terminate(co);
         coroutine_delete(co);
         return;