Block layer patches

-----BEGIN PGP SIGNATURE-----
 Version: GnuPG v2.0.22 (GNU/Linux)
 
 iQIcBAABAgAGBQJXNIcBAAoJEH8JsnLIjy/WiUQP/Rzfo8pe7TWA2InxdcDOPsx4
 2/tHHJdVkffnNX5rdBvc0mOUZNJxej0NtJu2e63BB+ydYju//xw8gruKU7TR+Nd3
 nPNSsqk80prK3RNgWu7qymBvIkHDDcDQhlp48HKq+dxrfConXtHmoXapGsqc0S47
 xu03oC6WzSIyLf7TLytcjUmEprQSaCGOwsb/XaHAWL750fFAGcdy/K5PWBpUv6DN
 T0jZ3u4UneE1jeabRmqAwjgDJXC9l6riH9fP/ZtYhgNlNj84zlMXajUHSULhGknP
 cTGjwwg9tOvhcjTdhdRmWlvG1m0T77ZX3icfZLhcTdb/Uz68NXVqs8P25IGV9McD
 DPrb3T/M8JUoqLXJxIpxUm2Levof5v0dUF1PHmN5bT7pshcqv/1J7v8Fdtf9l9mp
 zI0+FK1TZ102C0H2F7AWYZSlo2EfNUSd02QQx6MbfDokDIlIxY+EgP1/Es5XlkqC
 wc7HrJvq+uix2zXw9bn9Vg9p/nDuxlRx+ppRRarNNRonaqTrx/1qAaas4bsqc9Gz
 H6gxw7BHybm0TZFdHqAdIonpesecYw6yWUXT/mQehbfphsmQmu/d2HvF2C9uUm4X
 O0JduBlKOTm2hMcg5qL6Gko8WaQIctdCJH/1Onts92cZnm8Vr/9zcmMgwGoCd7sE
 +t6Yg0jqpTUJwhZhIuCw
 =NbjJ
 -----END PGP SIGNATURE-----

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

Block layer patches

# gpg: Signature made Thu 12 May 2016 14:37:05 BST using RSA key ID C88F2FD6
# gpg: Good signature from "Kevin Wolf <kwolf@redhat.com>"

* remotes/kevin/tags/for-upstream: (69 commits)
  qemu-iotests: iotests: fail hard if not run via "check"
  block: enable testing of LUKS driver with block I/O tests
  block: add support for encryption secrets in block I/O tests
  block: add support for --image-opts in block I/O tests
  qemu-io: Add 'write -z -u' to test MAY_UNMAP flag
  qemu-io: Add 'write -f' to test FUA flag
  qemu-io: Allow unaligned access by default
  qemu-io: Use bool for command line flags
  qemu-io: Make 'open' subcommand more like command line
  qemu-io: Add missing option documentation
  qmp: add monitor command to add/remove a child
  quorum: implement bdrv_add_child() and bdrv_del_child()
  Add new block driver interface to add/delete a BDS's child
  qemu-img: check block status of backing file when converting.
  iotests: fix the redirection order in 083
  block: Inactivate all children
  block: Drop superfluous invalidating bs->file from drivers
  block: Invalidate all children
  nbd: Simplify client FUA handling
  block: Honor BDRV_REQ_FUA during write_zeroes
  ...

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

block.c

@@ -218,8 +218,6 @@ void bdrv_get_full_backing_filename(BlockDriverState *bs, char *dest, size_t sz,
 
 void bdrv_register(BlockDriver *bdrv)
 {
-    bdrv_setup_io_funcs(bdrv);
-
     QLIST_INSERT_HEAD(&bdrv_drivers, bdrv, list);
 }
 
@@ -1176,10 +1174,10 @@ BdrvChild *bdrv_root_attach_child(BlockDriverState *child_bs,
     return child;
 }
 
-static BdrvChild *bdrv_attach_child(BlockDriverState *parent_bs,
+BdrvChild *bdrv_attach_child(BlockDriverState *parent_bs,
-                                    BlockDriverState *child_bs,
+                             BlockDriverState *child_bs,
-                                    const char *child_name,
+                             const char *child_name,
-                                    const BdrvChildRole *child_role)
+                             const BdrvChildRole *child_role)
 {
     BdrvChild *child = bdrv_root_attach_child(child_bs, child_name, child_role);
     QLIST_INSERT_HEAD(&parent_bs->children, child, next);
@@ -2261,7 +2259,6 @@ static void swap_feature_fields(BlockDriverState *bs_top,
 
     assert(!bs_new->throttle_state);
     if (bs_top->throttle_state) {
-        assert(bs_top->io_limits_enabled);
         bdrv_io_limits_enable(bs_new, throttle_group_get_name(bs_top));
         bdrv_io_limits_disable(bs_top);
     }
@@ -3201,6 +3198,7 @@ void bdrv_init_with_whitelist(void)
 
 void bdrv_invalidate_cache(BlockDriverState *bs, Error **errp)
 {
+    BdrvChild *child;
     Error *local_err = NULL;
     int ret;
 
@@ -3215,13 +3213,20 @@ void bdrv_invalidate_cache(BlockDriverState *bs, Error **errp)
 
     if (bs->drv->bdrv_invalidate_cache) {
         bs->drv->bdrv_invalidate_cache(bs, &local_err);
-    } else if (bs->file) {
+        if (local_err) {
-        bdrv_invalidate_cache(bs->file->bs, &local_err);
+            bs->open_flags |= BDRV_O_INACTIVE;
+            error_propagate(errp, local_err);
+            return;
+        }
     }
-    if (local_err) {
+
-        bs->open_flags |= BDRV_O_INACTIVE;
+    QLIST_FOREACH(child, &bs->children, next) {
-        error_propagate(errp, local_err);
+        bdrv_invalidate_cache(child->bs, &local_err);
-        return;
+        if (local_err) {
+            bs->open_flags |= BDRV_O_INACTIVE;
+            error_propagate(errp, local_err);
+            return;
+        }
     }
 
     ret = refresh_total_sectors(bs, bs->total_sectors);
@@ -3250,38 +3255,63 @@ void bdrv_invalidate_cache_all(Error **errp)
     }
 }
 
-static int bdrv_inactivate(BlockDriverState *bs)
+static int bdrv_inactivate_recurse(BlockDriverState *bs,
+                                   bool setting_flag)
 {
+    BdrvChild *child;
     int ret;
 
-    if (bs->drv->bdrv_inactivate) {
+    if (!setting_flag && bs->drv->bdrv_inactivate) {
         ret = bs->drv->bdrv_inactivate(bs);
         if (ret < 0) {
             return ret;
         }
     }
 
-    bs->open_flags |= BDRV_O_INACTIVE;
+    QLIST_FOREACH(child, &bs->children, next) {
+        ret = bdrv_inactivate_recurse(child->bs, setting_flag);
+        if (ret < 0) {
+            return ret;
+        }
+    }
+
+    if (setting_flag) {
+        bs->open_flags |= BDRV_O_INACTIVE;
+    }
     return 0;
 }
 
 int bdrv_inactivate_all(void)
 {
     BlockDriverState *bs = NULL;
-    int ret;
+    int ret = 0;
+    int pass;
 
     while ((bs = bdrv_next(bs)) != NULL) {
-        AioContext *aio_context = bdrv_get_aio_context(bs);
+        aio_context_acquire(bdrv_get_aio_context(bs));
+    }
 
-        aio_context_acquire(aio_context);
+    /* We do two passes of inactivation. The first pass calls to drivers'
-        ret = bdrv_inactivate(bs);
+     * .bdrv_inactivate callbacks recursively so all cache is flushed to disk;
-        aio_context_release(aio_context);
+     * the second pass sets the BDRV_O_INACTIVE flag so that no further write
-        if (ret < 0) {
+     * is allowed. */
-            return ret;
+    for (pass = 0; pass < 2; pass++) {
+        bs = NULL;
+        while ((bs = bdrv_next(bs)) != NULL) {
+            ret = bdrv_inactivate_recurse(bs, pass);
+            if (ret < 0) {
+                goto out;
+            }
         }
     }
 
-    return 0;
+out:
+    bs = NULL;
+    while ((bs = bdrv_next(bs)) != NULL) {
+        aio_context_release(bdrv_get_aio_context(bs));
+    }
+
+    return ret;
 }
 
 /**************************************************************/
@@ -3981,3 +4011,52 @@ void bdrv_refresh_filename(BlockDriverState *bs)
         QDECREF(json);
     }
 }
+
+/*
+ * Hot add/remove a BDS's child. So the user can take a child offline when
+ * it is broken and take a new child online
+ */
+void bdrv_add_child(BlockDriverState *parent_bs, BlockDriverState *child_bs,
+                    Error **errp)
+{
+
+    if (!parent_bs->drv || !parent_bs->drv->bdrv_add_child) {
+        error_setg(errp, "The node %s does not support adding a child",
+                   bdrv_get_device_or_node_name(parent_bs));
+        return;
+    }
+
+    if (!QLIST_EMPTY(&child_bs->parents)) {
+        error_setg(errp, "The node %s already has a parent",
+                   child_bs->node_name);
+        return;
+    }
+
+    parent_bs->drv->bdrv_add_child(parent_bs, child_bs, errp);
+}
+
+void bdrv_del_child(BlockDriverState *parent_bs, BdrvChild *child, Error **errp)
+{
+    BdrvChild *tmp;
+
+    if (!parent_bs->drv || !parent_bs->drv->bdrv_del_child) {
+        error_setg(errp, "The node %s does not support removing a child",
+                   bdrv_get_device_or_node_name(parent_bs));
+        return;
+    }
+
+    QLIST_FOREACH(tmp, &parent_bs->children, next) {
+        if (tmp == child) {
+            break;
+        }
+    }
+
+    if (!tmp) {
+        error_setg(errp, "The node %s does not have a child named %s",
+                   bdrv_get_device_or_node_name(parent_bs),
+                   bdrv_get_device_or_node_name(child->bs));
+        return;
+    }
+
+    parent_bs->drv->bdrv_del_child(parent_bs, child, errp);
+}

block/block-backend.c

@@ -1,7 +1,7 @@
 /*
  * QEMU Block backends
  *
- * Copyright (C) 2014 Red Hat, Inc.
+ * Copyright (C) 2014-2016 Red Hat, Inc.
  *
  * Authors:
  *  Markus Armbruster <armbru@redhat.com>,
@@ -692,7 +692,7 @@ static int coroutine_fn blk_co_preadv(BlockBackend *blk, int64_t offset,
         return ret;
     }
 
-    return bdrv_co_do_preadv(blk_bs(blk), offset, bytes, qiov, flags);
+    return bdrv_co_preadv(blk_bs(blk), offset, bytes, qiov, flags);
 }
 
 static int coroutine_fn blk_co_pwritev(BlockBackend *blk, int64_t offset,
@@ -710,7 +710,7 @@ static int coroutine_fn blk_co_pwritev(BlockBackend *blk, int64_t offset,
         flags |= BDRV_REQ_FUA;
     }
 
-    return bdrv_co_do_pwritev(blk_bs(blk), offset, bytes, qiov, flags);
+    return bdrv_co_pwritev(blk_bs(blk), offset, bytes, qiov, flags);
 }
 
 typedef struct BlkRwCo {
@@ -772,55 +772,28 @@ static int blk_prw(BlockBackend *blk, int64_t offset, uint8_t *buf,
     return rwco.ret;
 }
 
-static int blk_rw(BlockBackend *blk, int64_t sector_num, uint8_t *buf,
+int blk_pread_unthrottled(BlockBackend *blk, int64_t offset, uint8_t *buf,
-                  int nb_sectors, CoroutineEntry co_entry,
+                          int count)
-                  BdrvRequestFlags flags)
-{
-    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);
+    ret = blk_check_byte_request(blk, offset, count);
     if (ret < 0) {
         return ret;
     }
 
-    enabled = bs->io_limits_enabled;
+    bdrv_no_throttling_begin(bs);
-    bs->io_limits_enabled = false;
+    ret = blk_pread(blk, offset, buf, count);
-    ret = blk_read(blk, sector_num, buf, nb_sectors);
+    bdrv_no_throttling_end(bs);
-    bs->io_limits_enabled = enabled;
     return ret;
 }
 
-int blk_write(BlockBackend *blk, int64_t sector_num, const uint8_t *buf,
+int blk_write_zeroes(BlockBackend *blk, int64_t offset,
-              int nb_sectors)
+                     int count, BdrvRequestFlags flags)
 {
-    return blk_rw(blk, sector_num, (uint8_t*) buf, nb_sectors,
+    return blk_prw(blk, offset, NULL, count, blk_write_entry,
-                  blk_write_entry, 0);
+                   flags | BDRV_REQ_ZERO_WRITE);
-}
-
-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)
@@ -932,18 +905,12 @@ static void blk_aio_write_entry(void *opaque)
     blk_aio_complete(acb);
 }
 
-BlockAIOCB *blk_aio_write_zeroes(BlockBackend *blk, int64_t sector_num,
+BlockAIOCB *blk_aio_write_zeroes(BlockBackend *blk, int64_t offset,
-                                 int nb_sectors, BdrvRequestFlags flags,
+                                 int count, BdrvRequestFlags flags,
                                  BlockCompletionFunc *cb, void *opaque)
 {
-    if (nb_sectors < 0 || nb_sectors > BDRV_REQUEST_MAX_SECTORS) {
+    return blk_aio_prwv(blk, offset, count, NULL, blk_aio_write_entry,
-        return blk_abort_aio_request(blk, cb, opaque, -EINVAL);
+                        flags | BDRV_REQ_ZERO_WRITE, 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)
@@ -955,9 +922,11 @@ int blk_pread(BlockBackend *blk, int64_t offset, void *buf, int count)
     return count;
 }
 
-int blk_pwrite(BlockBackend *blk, int64_t offset, const void *buf, int count)
+int blk_pwrite(BlockBackend *blk, int64_t offset, const void *buf, int count,
+               BdrvRequestFlags flags)
 {
-    int ret = blk_prw(blk, offset, (void*) buf, count, blk_write_entry, 0);
+    int ret = blk_prw(blk, offset, (void *) buf, count, blk_write_entry,
+                      flags);
     if (ret < 0) {
         return ret;
     }
@@ -991,30 +960,20 @@ int64_t blk_nb_sectors(BlockBackend *blk)
     return bdrv_nb_sectors(blk_bs(blk));
 }
 
-BlockAIOCB *blk_aio_readv(BlockBackend *blk, int64_t sector_num,
+BlockAIOCB *blk_aio_preadv(BlockBackend *blk, int64_t offset,
-                          QEMUIOVector *iov, int nb_sectors,
+                           QEMUIOVector *qiov, BdrvRequestFlags flags,
-                          BlockCompletionFunc *cb, void *opaque)
-{
-    if (nb_sectors < 0 || nb_sectors > BDRV_REQUEST_MAX_SECTORS) {
-        return blk_abort_aio_request(blk, cb, opaque, -EINVAL);
-    }
-
-    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)
 {
-    if (nb_sectors < 0 || nb_sectors > BDRV_REQUEST_MAX_SECTORS) {
+    return blk_aio_prwv(blk, offset, qiov->size, qiov,
-        return blk_abort_aio_request(blk, cb, opaque, -EINVAL);
+                        blk_aio_read_entry, flags, cb, opaque);
-    }
+}
 
-    assert(nb_sectors << BDRV_SECTOR_BITS == iov->size);
+BlockAIOCB *blk_aio_pwritev(BlockBackend *blk, int64_t offset,
-    return blk_aio_prwv(blk, sector_num << BDRV_SECTOR_BITS, iov->size, iov,
+                            QEMUIOVector *qiov, BdrvRequestFlags flags,
-                        blk_aio_write_entry, 0, cb, opaque);
+                            BlockCompletionFunc *cb, void *opaque)
+{
+    return blk_aio_prwv(blk, offset, qiov->size, qiov,
+                        blk_aio_write_entry, flags, cb, opaque);
 }
 
 BlockAIOCB *blk_aio_flush(BlockBackend *blk,
@@ -1444,15 +1403,10 @@ void *blk_aio_get(const AIOCBInfo *aiocb_info, BlockBackend *blk,
     return qemu_aio_get(aiocb_info, blk_bs(blk), cb, opaque);
 }
 
-int coroutine_fn blk_co_write_zeroes(BlockBackend *blk, int64_t sector_num,
+int coroutine_fn blk_co_write_zeroes(BlockBackend *blk, int64_t offset,
-                                     int nb_sectors, BdrvRequestFlags flags)
+                                     int count, BdrvRequestFlags flags)
 {
-    if (nb_sectors < 0 || nb_sectors > BDRV_REQUEST_MAX_SECTORS) {
+    return blk_co_pwritev(blk, offset, count, NULL,
-        return -EINVAL;
-    }
-
-    return blk_co_pwritev(blk, sector_num << BDRV_SECTOR_BITS,
-                          nb_sectors << BDRV_SECTOR_BITS, NULL,
                           flags | BDRV_REQ_ZERO_WRITE);
 }
 

block/bochs.c

@@ -104,6 +104,7 @@ static int bochs_open(BlockDriverState *bs, QDict *options, int flags,
     int ret;
 
     bs->read_only = 1; // no write support yet
+    bs->request_alignment = BDRV_SECTOR_SIZE; /* No sub-sector I/O supported */
 
     ret = bdrv_pread(bs->file->bs, 0, &bochs, sizeof(bochs));
     if (ret < 0) {
@@ -221,38 +222,52 @@ static int64_t seek_to_sector(BlockDriverState *bs, int64_t sector_num)
     return bitmap_offset + (512 * (s->bitmap_blocks + extent_offset));
 }
 
-static int bochs_read(BlockDriverState *bs, int64_t sector_num,
+static int coroutine_fn
-                    uint8_t *buf, int nb_sectors)
+bochs_co_preadv(BlockDriverState *bs, uint64_t offset, uint64_t bytes,
+                QEMUIOVector *qiov, int flags)
 {
+    BDRVBochsState *s = bs->opaque;
+    uint64_t sector_num = offset >> BDRV_SECTOR_BITS;
+    int nb_sectors = bytes >> BDRV_SECTOR_BITS;
+    uint64_t bytes_done = 0;
+    QEMUIOVector local_qiov;
     int ret;
 
+    assert((offset & (BDRV_SECTOR_SIZE - 1)) == 0);
+    assert((bytes & (BDRV_SECTOR_SIZE - 1)) == 0);
+
+    qemu_iovec_init(&local_qiov, qiov->niov);
+    qemu_co_mutex_lock(&s->lock);
+
     while (nb_sectors > 0) {
         int64_t block_offset = seek_to_sector(bs, sector_num);
         if (block_offset < 0) {
-            return block_offset;
+            ret = block_offset;
-        } else if (block_offset > 0) {
+            goto fail;
-            ret = bdrv_pread(bs->file->bs, block_offset, buf, 512);
+        }
+
+        qemu_iovec_reset(&local_qiov);
+        qemu_iovec_concat(&local_qiov, qiov, bytes_done, 512);
+
+        if (block_offset > 0) {
+            ret = bdrv_co_preadv(bs->file->bs, block_offset, 512,
+                                 &local_qiov, 0);
             if (ret < 0) {
-                return ret;
+                goto fail;
             }
         } else {
-            memset(buf, 0, 512);
+            qemu_iovec_memset(&local_qiov, 0, 0, 512);
         }
         nb_sectors--;
         sector_num++;
-        buf += 512;
+        bytes_done += 512;
     }
-    return 0;
-}
 
-static coroutine_fn int bochs_co_read(BlockDriverState *bs, int64_t sector_num,
+    ret = 0;
-                                      uint8_t *buf, int nb_sectors)
+fail:
-{
-    int ret;
-    BDRVBochsState *s = bs->opaque;
-    qemu_co_mutex_lock(&s->lock);
-    ret = bochs_read(bs, sector_num, buf, nb_sectors);
     qemu_co_mutex_unlock(&s->lock);
+    qemu_iovec_destroy(&local_qiov);
+
     return ret;
 }
 
@@ -267,7 +282,7 @@ static BlockDriver bdrv_bochs = {
     .instance_size	= sizeof(BDRVBochsState),
     .bdrv_probe		= bochs_probe,
     .bdrv_open		= bochs_open,
-    .bdrv_read          = bochs_co_read,
+    .bdrv_co_preadv = bochs_co_preadv,
     .bdrv_close		= bochs_close,
 };
 

block/cloop.c

@@ -66,6 +66,7 @@ static int cloop_open(BlockDriverState *bs, QDict *options, int flags,
     int ret;
 
     bs->read_only = 1;
+    bs->request_alignment = BDRV_SECTOR_SIZE; /* No sub-sector I/O supported */
 
     /* read header */
     ret = bdrv_pread(bs->file->bs, 128, &s->block_size, 4);
@@ -229,33 +230,38 @@ static inline int cloop_read_block(BlockDriverState *bs, int block_num)
     return 0;
 }
 
-static int cloop_read(BlockDriverState *bs, int64_t sector_num,
+static int coroutine_fn
-                    uint8_t *buf, int nb_sectors)
+cloop_co_preadv(BlockDriverState *bs, uint64_t offset, uint64_t bytes,
+                QEMUIOVector *qiov, int flags)
 {
     BDRVCloopState *s = bs->opaque;
-    int i;
+    uint64_t sector_num = offset >> BDRV_SECTOR_BITS;
+    int nb_sectors = bytes >> BDRV_SECTOR_BITS;
+    int ret, i;
+
+    assert((offset & (BDRV_SECTOR_SIZE - 1)) == 0);
+    assert((bytes & (BDRV_SECTOR_SIZE - 1)) == 0);
+
+    qemu_co_mutex_lock(&s->lock);
 
     for (i = 0; i < nb_sectors; i++) {
+        void *data;
         uint32_t sector_offset_in_block =
             ((sector_num + i) % s->sectors_per_block),
             block_num = (sector_num + i) / s->sectors_per_block;
         if (cloop_read_block(bs, block_num) != 0) {
-            return -1;
+            ret = -EIO;
+            goto fail;
         }
-        memcpy(buf + i * 512,
-            s->uncompressed_block + sector_offset_in_block * 512, 512);
-    }
-    return 0;
-}
 
-static coroutine_fn int cloop_co_read(BlockDriverState *bs, int64_t sector_num,
+        data = s->uncompressed_block + sector_offset_in_block * 512;
-                                      uint8_t *buf, int nb_sectors)
+        qemu_iovec_from_buf(qiov, i * 512, data, 512);
-{
+    }
-    int ret;
+
-    BDRVCloopState *s = bs->opaque;
+    ret = 0;
-    qemu_co_mutex_lock(&s->lock);
+fail:
-    ret = cloop_read(bs, sector_num, buf, nb_sectors);
     qemu_co_mutex_unlock(&s->lock);
+
     return ret;
 }
 
@@ -273,7 +279,7 @@ static BlockDriver bdrv_cloop = {
     .instance_size  = sizeof(BDRVCloopState),
     .bdrv_probe     = cloop_probe,
     .bdrv_open      = cloop_open,
-    .bdrv_read      = cloop_co_read,
+    .bdrv_co_preadv = cloop_co_preadv,
     .bdrv_close     = cloop_close,
 };
 

block/crypto.c

@@ -91,7 +91,7 @@ static ssize_t block_crypto_write_func(QCryptoBlock *block,
     struct BlockCryptoCreateData *data = opaque;
     ssize_t ret;
 
-    ret = blk_pwrite(data->blk, offset, buf, buflen);
+    ret = blk_pwrite(data->blk, offset, buf, buflen, 0);
     if (ret < 0) {
         error_setg_errno(errp, -ret, "Could not write encryption header");
         return ret;

block/curl.c

@@ -36,10 +36,16 @@
 // #define DEBUG_VERBOSE
 
 #ifdef DEBUG_CURL
-#define DPRINTF(fmt, ...) do { printf(fmt, ## __VA_ARGS__); } while (0)
+#define DEBUG_CURL_PRINT 1
 #else
-#define DPRINTF(fmt, ...) do { } while (0)
+#define DEBUG_CURL_PRINT 0
 #endif
+#define DPRINTF(fmt, ...)                                            \
+    do {                                                             \
+        if (DEBUG_CURL_PRINT) {                                      \
+            fprintf(stderr, fmt, ## __VA_ARGS__);                    \
+        }                                                            \
+    } while (0)
 
 #if LIBCURL_VERSION_NUM >= 0x071000
 /* The multi interface timer callback was introduced in 7.16.0 */

block/dmg.c

@@ -440,6 +440,8 @@ static int dmg_open(BlockDriverState *bs, QDict *options, int flags,
     int ret;
 
     bs->read_only = 1;
+    bs->request_alignment = BDRV_SECTOR_SIZE; /* No sub-sector I/O supported */
+
     s->n_chunks = 0;
     s->offsets = s->lengths = s->sectors = s->sectorcounts = NULL;
     /* used by dmg_read_mish_block to keep track of the current I/O position */
@@ -659,38 +661,42 @@ static inline int dmg_read_chunk(BlockDriverState *bs, uint64_t sector_num)
     return 0;
 }
 
-static int dmg_read(BlockDriverState *bs, int64_t sector_num,
+static int coroutine_fn
-                    uint8_t *buf, int nb_sectors)
+dmg_co_preadv(BlockDriverState *bs, uint64_t offset, uint64_t bytes,
+              QEMUIOVector *qiov, int flags)
 {
     BDRVDMGState *s = bs->opaque;
-    int i;
+    uint64_t sector_num = offset >> BDRV_SECTOR_BITS;
+    int nb_sectors = bytes >> BDRV_SECTOR_BITS;
+    int ret, i;
+
+    assert((offset & (BDRV_SECTOR_SIZE - 1)) == 0);
+    assert((bytes & (BDRV_SECTOR_SIZE - 1)) == 0);
+
+    qemu_co_mutex_lock(&s->lock);
 
     for (i = 0; i < nb_sectors; i++) {
         uint32_t sector_offset_in_chunk;
+        void *data;
+
         if (dmg_read_chunk(bs, sector_num + i) != 0) {
-            return -1;
+            ret = -EIO;
+            goto fail;
         }
         /* Special case: current chunk is all zeroes. Do not perform a memcpy as
          * s->uncompressed_chunk may be too small to cover the large all-zeroes
          * section. dmg_read_chunk is called to find s->current_chunk */
         if (s->types[s->current_chunk] == 2) { /* all zeroes block entry */
-            memset(buf + i * 512, 0, 512);
+            qemu_iovec_memset(qiov, i * 512, 0, 512);
             continue;
         }
         sector_offset_in_chunk = sector_num + i - s->sectors[s->current_chunk];
-        memcpy(buf + i * 512,
+        data = s->uncompressed_chunk + sector_offset_in_chunk * 512;
-               s->uncompressed_chunk + sector_offset_in_chunk * 512, 512);
+        qemu_iovec_from_buf(qiov, i * 512, data, 512);
     }
-    return 0;
-}
 
-static coroutine_fn int dmg_co_read(BlockDriverState *bs, int64_t sector_num,
+    ret = 0;
-                                    uint8_t *buf, int nb_sectors)
+fail:
-{
-    int ret;
-    BDRVDMGState *s = bs->opaque;
-    qemu_co_mutex_lock(&s->lock);
-    ret = dmg_read(bs, sector_num, buf, nb_sectors);
     qemu_co_mutex_unlock(&s->lock);
     return ret;
 }
@@ -715,7 +721,7 @@ static BlockDriver bdrv_dmg = {
     .instance_size  = sizeof(BDRVDMGState),
     .bdrv_probe     = dmg_probe,
     .bdrv_open      = dmg_open,
-    .bdrv_read      = dmg_co_read,
+    .bdrv_co_preadv = dmg_co_preadv,
     .bdrv_close     = dmg_close,
 };
 

block/io.c

@@ -34,18 +34,6 @@
 
 #define NOT_DONE 0x7fffffff /* used while emulated sync operation in progress */
 
-static BlockAIOCB *bdrv_aio_readv_em(BlockDriverState *bs,
-        int64_t sector_num, QEMUIOVector *qiov, int nb_sectors,
-        BlockCompletionFunc *cb, void *opaque);
-static BlockAIOCB *bdrv_aio_writev_em(BlockDriverState *bs,
-        int64_t sector_num, QEMUIOVector *qiov, int nb_sectors,
-        BlockCompletionFunc *cb, void *opaque);
-static int coroutine_fn bdrv_co_readv_em(BlockDriverState *bs,
-                                         int64_t sector_num, int nb_sectors,
-                                         QEMUIOVector *iov);
-static int coroutine_fn bdrv_co_writev_em(BlockDriverState *bs,
-                                         int64_t sector_num, int nb_sectors,
-                                         QEMUIOVector *iov);
 static BlockAIOCB *bdrv_co_aio_rw_vector(BlockDriverState *bs,
                                          int64_t sector_num,
                                          QEMUIOVector *qiov,
@@ -62,48 +50,35 @@ static int coroutine_fn bdrv_co_do_write_zeroes(BlockDriverState *bs,
 void bdrv_set_io_limits(BlockDriverState *bs,
                         ThrottleConfig *cfg)
 {
-    int i;
-
     throttle_group_config(bs, cfg);
+}
 
-    for (i = 0; i < 2; i++) {
+void bdrv_no_throttling_begin(BlockDriverState *bs)
-        qemu_co_enter_next(&bs->throttled_reqs[i]);
+{
+    if (bs->io_limits_disabled++ == 0) {
+        throttle_group_restart_bs(bs);
     }
 }
 
-/* this function drain all the throttled IOs */
+void bdrv_no_throttling_end(BlockDriverState *bs)
-static bool bdrv_start_throttled_reqs(BlockDriverState *bs)
 {
-    bool drained = false;
+    assert(bs->io_limits_disabled);
-    bool enabled = bs->io_limits_enabled;
+    --bs->io_limits_disabled;
-    int i;
-
-    bs->io_limits_enabled = false;
-
-    for (i = 0; i < 2; i++) {
-        while (qemu_co_enter_next(&bs->throttled_reqs[i])) {
-            drained = true;
-        }
-    }
-
-    bs->io_limits_enabled = enabled;
-
-    return drained;
 }
 
 void bdrv_io_limits_disable(BlockDriverState *bs)
 {
-    bs->io_limits_enabled = false;
+    assert(bs->throttle_state);
-    bdrv_start_throttled_reqs(bs);
+    bdrv_no_throttling_begin(bs);
     throttle_group_unregister_bs(bs);
+    bdrv_no_throttling_end(bs);
 }
 
 /* should be called before bdrv_set_io_limits if a limit is set */
 void bdrv_io_limits_enable(BlockDriverState *bs, const char *group)
 {
-    assert(!bs->io_limits_enabled);
+    assert(!bs->throttle_state);
     throttle_group_register_bs(bs, group);
-    bs->io_limits_enabled = true;
 }
 
 void bdrv_io_limits_update_group(BlockDriverState *bs, const char *group)
@@ -123,24 +98,6 @@ void bdrv_io_limits_update_group(BlockDriverState *bs, const char *group)
     bdrv_io_limits_enable(bs, group);
 }
 
-void bdrv_setup_io_funcs(BlockDriver *bdrv)
-{
-    /* Block drivers without coroutine functions need emulation */
-    if (!bdrv->bdrv_co_readv) {
-        bdrv->bdrv_co_readv = bdrv_co_readv_em;
-        bdrv->bdrv_co_writev = bdrv_co_writev_em;
-
-        /* bdrv_co_readv_em()/brdv_co_writev_em() work in terms of aio, so if
-         * the block driver lacks aio we need to emulate that too.
-         */
-        if (!bdrv->bdrv_aio_readv) {
-            /* add AIO emulation layer */
-            bdrv->bdrv_aio_readv = bdrv_aio_readv_em;
-            bdrv->bdrv_aio_writev = bdrv_aio_writev_em;
-        }
-    }
-}
-
 void bdrv_refresh_limits(BlockDriverState *bs, Error **errp)
 {
     BlockDriver *drv = bs->drv;
@@ -260,18 +217,29 @@ typedef struct {
     bool done;
 } BdrvCoDrainData;
 
+static void bdrv_drain_poll(BlockDriverState *bs)
+{
+    bool busy = true;
+
+    while (busy) {
+        /* Keep iterating */
+        busy = bdrv_requests_pending(bs);
+        busy |= aio_poll(bdrv_get_aio_context(bs), busy);
+    }
+}
+
 static void bdrv_co_drain_bh_cb(void *opaque)
 {
     BdrvCoDrainData *data = opaque;
     Coroutine *co = data->co;
 
     qemu_bh_delete(data->bh);
-    bdrv_drain(data->bs);
+    bdrv_drain_poll(data->bs);
     data->done = true;
     qemu_coroutine_enter(co, NULL);
 }
 
-void coroutine_fn bdrv_co_drain(BlockDriverState *bs)
+static void coroutine_fn bdrv_co_yield_to_drain(BlockDriverState *bs)
 {
     BdrvCoDrainData data;
 
@@ -305,21 +273,28 @@ void coroutine_fn bdrv_co_drain(BlockDriverState *bs)
  * not depend on events in other AioContexts.  In that case, use
  * bdrv_drain_all() instead.
  */
+void coroutine_fn bdrv_co_drain(BlockDriverState *bs)
+{
+    bdrv_no_throttling_begin(bs);
+    bdrv_io_unplugged_begin(bs);
+    bdrv_drain_recurse(bs);
+    bdrv_co_yield_to_drain(bs);
+    bdrv_io_unplugged_end(bs);
+    bdrv_no_throttling_end(bs);
+}
+
 void bdrv_drain(BlockDriverState *bs)
 {
-    bool busy = true;
+    bdrv_no_throttling_begin(bs);
-
+    bdrv_io_unplugged_begin(bs);
     bdrv_drain_recurse(bs);
     if (qemu_in_coroutine()) {
-        bdrv_co_drain(bs);
+        bdrv_co_yield_to_drain(bs);
-        return;
+    } else {
-    }
+        bdrv_drain_poll(bs);
-    while (busy) {
-        /* Keep iterating */
-         bdrv_flush_io_queue(bs);
-         busy = bdrv_requests_pending(bs);
-         busy |= aio_poll(bdrv_get_aio_context(bs), busy);
     }
+    bdrv_io_unplugged_end(bs);
+    bdrv_no_throttling_end(bs);
 }
 
 /*
@@ -342,6 +317,8 @@ void bdrv_drain_all(void)
         if (bs->job) {
             block_job_pause(bs->job);
         }
+        bdrv_no_throttling_begin(bs);
+        bdrv_io_unplugged_begin(bs);
         bdrv_drain_recurse(bs);
         aio_context_release(aio_context);
 
@@ -366,7 +343,6 @@ void bdrv_drain_all(void)
             aio_context_acquire(aio_context);
             while ((bs = bdrv_next(bs))) {
                 if (aio_context == bdrv_get_aio_context(bs)) {
-                    bdrv_flush_io_queue(bs);
                     if (bdrv_requests_pending(bs)) {
                         busy = true;
                         aio_poll(aio_context, busy);
@@ -383,6 +359,8 @@ void bdrv_drain_all(void)
         AioContext *aio_context = bdrv_get_aio_context(bs);
 
         aio_context_acquire(aio_context);
+        bdrv_io_unplugged_end(bs);
+        bdrv_no_throttling_end(bs);
         if (bs->job) {
             block_job_resume(bs->job);
         }
@@ -581,13 +559,13 @@ static void coroutine_fn bdrv_rw_co_entry(void *opaque)
     RwCo *rwco = opaque;
 
     if (!rwco->is_write) {
-        rwco->ret = bdrv_co_do_preadv(rwco->bs, rwco->offset,
+        rwco->ret = bdrv_co_preadv(rwco->bs, rwco->offset,
-                                      rwco->qiov->size, rwco->qiov,
+                                   rwco->qiov->size, rwco->qiov,
-                                      rwco->flags);
+                                   rwco->flags);
     } else {
-        rwco->ret = bdrv_co_do_pwritev(rwco->bs, rwco->offset,
+        rwco->ret = bdrv_co_pwritev(rwco->bs, rwco->offset,
-                                       rwco->qiov->size, rwco->qiov,
+                                    rwco->qiov->size, rwco->qiov,
-                                       rwco->flags);
+                                    rwco->flags);
     }
 }
 
@@ -608,17 +586,6 @@ static int bdrv_prwv_co(BlockDriverState *bs, int64_t offset,
         .flags = flags,
     };
 
-    /**
-     * In sync call context, when the vcpu is blocked, this throttling timer
-     * will not fire; so the I/O throttling function has to be disabled here
-     * if it has been enabled.
-     */
-    if (bs->io_limits_enabled) {
-        fprintf(stderr, "Disabling I/O throttling on '%s' due "
-                        "to synchronous I/O.\n", bdrv_get_device_name(bs));
-        bdrv_io_limits_disable(bs);
-    }
-
     if (qemu_in_coroutine()) {
         /* Fast-path if already in coroutine context */
         bdrv_rw_co_entry(&rwco);
@@ -685,7 +652,8 @@ int bdrv_write_zeroes(BlockDriverState *bs, int64_t sector_num,
  * Completely zero out a block device with the help of bdrv_write_zeroes.
  * The operation is sped up by checking the block status and only writing
  * zeroes to the device if they currently do not return zeroes. Optional
- * flags are passed through to bdrv_write_zeroes (e.g. BDRV_REQ_MAY_UNMAP).
+ * flags are passed through to bdrv_write_zeroes (e.g. BDRV_REQ_MAY_UNMAP,
+ * BDRV_REQ_FUA).
  *
  * Returns < 0 on error, 0 on success. For error codes see bdrv_write().
  */
@@ -800,6 +768,109 @@ int bdrv_pwrite_sync(BlockDriverState *bs, int64_t offset,
     return 0;
 }
 
+typedef struct CoroutineIOCompletion {
+    Coroutine *coroutine;
+    int ret;
+} CoroutineIOCompletion;
+
+static void bdrv_co_io_em_complete(void *opaque, int ret)
+{
+    CoroutineIOCompletion *co = opaque;
+
+    co->ret = ret;
+    qemu_coroutine_enter(co->coroutine, NULL);
+}
+
+static int coroutine_fn bdrv_driver_preadv(BlockDriverState *bs,
+                                           uint64_t offset, uint64_t bytes,
+                                           QEMUIOVector *qiov, int flags)
+{
+    BlockDriver *drv = bs->drv;
+    int64_t sector_num;
+    unsigned int nb_sectors;
+
+    if (drv->bdrv_co_preadv) {
+        return drv->bdrv_co_preadv(bs, offset, bytes, qiov, flags);
+    }
+
+    sector_num = offset >> BDRV_SECTOR_BITS;
+    nb_sectors = bytes >> BDRV_SECTOR_BITS;
+
+    assert((offset & (BDRV_SECTOR_SIZE - 1)) == 0);
+    assert((bytes & (BDRV_SECTOR_SIZE - 1)) == 0);
+    assert((bytes >> BDRV_SECTOR_BITS) <= BDRV_REQUEST_MAX_SECTORS);
+
+    if (drv->bdrv_co_readv) {
+        return drv->bdrv_co_readv(bs, sector_num, nb_sectors, qiov);
+    } else {
+        BlockAIOCB *acb;
+        CoroutineIOCompletion co = {
+            .coroutine = qemu_coroutine_self(),
+        };
+
+        acb = bs->drv->bdrv_aio_readv(bs, sector_num, qiov, nb_sectors,
+                                      bdrv_co_io_em_complete, &co);
+        if (acb == NULL) {
+            return -EIO;
+        } else {
+            qemu_coroutine_yield();
+            return co.ret;
+        }
+    }
+}
+
+static int coroutine_fn bdrv_driver_pwritev(BlockDriverState *bs,
+                                            uint64_t offset, uint64_t bytes,
+                                            QEMUIOVector *qiov, int flags)
+{
+    BlockDriver *drv = bs->drv;
+    int64_t sector_num;
+    unsigned int nb_sectors;
+    int ret;
+
+    if (drv->bdrv_co_pwritev) {
+        ret = drv->bdrv_co_pwritev(bs, offset, bytes, qiov, flags);
+        goto emulate_flags;
+    }
+
+    sector_num = offset >> BDRV_SECTOR_BITS;
+    nb_sectors = bytes >> BDRV_SECTOR_BITS;
+
+    assert((offset & (BDRV_SECTOR_SIZE - 1)) == 0);
+    assert((bytes & (BDRV_SECTOR_SIZE - 1)) == 0);
+    assert((bytes >> BDRV_SECTOR_BITS) <= BDRV_REQUEST_MAX_SECTORS);
+
+    if (drv->bdrv_co_writev_flags) {
+        ret = drv->bdrv_co_writev_flags(bs, sector_num, nb_sectors, qiov,
+                                        flags & bs->supported_write_flags);
+        flags &= ~bs->supported_write_flags;
+    } else if (drv->bdrv_co_writev) {
+        assert(!bs->supported_write_flags);
+        ret = drv->bdrv_co_writev(bs, sector_num, nb_sectors, qiov);
+    } else {
+        BlockAIOCB *acb;
+        CoroutineIOCompletion co = {
+            .coroutine = qemu_coroutine_self(),
+        };
+
+        acb = bs->drv->bdrv_aio_writev(bs, sector_num, qiov, nb_sectors,
+                                       bdrv_co_io_em_complete, &co);
+        if (acb == NULL) {
+            ret = -EIO;
+        } else {
+            qemu_coroutine_yield();
+            ret = co.ret;
+        }
+    }
+
+emulate_flags:
+    if (ret == 0 && (flags & BDRV_REQ_FUA)) {
+        ret = bdrv_co_flush(bs);
+    }
+
+    return ret;
+}
+
 static int coroutine_fn bdrv_co_do_copy_on_readv(BlockDriverState *bs,
         int64_t sector_num, int nb_sectors, QEMUIOVector *qiov)
 {
@@ -836,8 +907,9 @@ static int coroutine_fn bdrv_co_do_copy_on_readv(BlockDriverState *bs,
 
     qemu_iovec_init_external(&bounce_qiov, &iov, 1);
 
-    ret = drv->bdrv_co_readv(bs, cluster_sector_num, cluster_nb_sectors,
+    ret = bdrv_driver_preadv(bs, cluster_sector_num * BDRV_SECTOR_SIZE,
-                             &bounce_qiov);
+                             cluster_nb_sectors * BDRV_SECTOR_SIZE,
+                             &bounce_qiov, 0);
     if (ret < 0) {
         goto err;
     }
@@ -850,8 +922,9 @@ static int coroutine_fn bdrv_co_do_copy_on_readv(BlockDriverState *bs,
         /* This does not change the data on the disk, it is not necessary
          * to flush even in cache=writethrough mode.
          */
-        ret = drv->bdrv_co_writev(bs, cluster_sector_num, cluster_nb_sectors,
+        ret = bdrv_driver_pwritev(bs, cluster_sector_num * BDRV_SECTOR_SIZE,
-                                  &bounce_qiov);
+                                  cluster_nb_sectors * BDRV_SECTOR_SIZE,
+                                  &bounce_qiov, 0);
     }
 
     if (ret < 0) {
@@ -880,7 +953,6 @@ static int coroutine_fn bdrv_aligned_preadv(BlockDriverState *bs,
     BdrvTrackedRequest *req, int64_t offset, unsigned int bytes,
     int64_t align, QEMUIOVector *qiov, int flags)
 {
-    BlockDriver *drv = bs->drv;
     int ret;
 
     int64_t sector_num = offset >> BDRV_SECTOR_BITS;
@@ -921,7 +993,7 @@ static int coroutine_fn bdrv_aligned_preadv(BlockDriverState *bs,
 
     /* Forward the request to the BlockDriver */
     if (!bs->zero_beyond_eof) {
-        ret = drv->bdrv_co_readv(bs, sector_num, nb_sectors, qiov);
+        ret = bdrv_driver_preadv(bs, offset, bytes, qiov, 0);
     } else {
         /* Read zeros after EOF */
         int64_t total_sectors, max_nb_sectors;
@@ -935,7 +1007,7 @@ static int coroutine_fn bdrv_aligned_preadv(BlockDriverState *bs,
         max_nb_sectors = ROUND_UP(MAX(0, total_sectors - sector_num),
                                   align >> BDRV_SECTOR_BITS);
         if (nb_sectors < max_nb_sectors) {
-            ret = drv->bdrv_co_readv(bs, sector_num, nb_sectors, qiov);
+            ret = bdrv_driver_preadv(bs, offset, bytes, qiov, 0);
         } else if (max_nb_sectors > 0) {
             QEMUIOVector local_qiov;
 
@@ -943,8 +1015,9 @@ static int coroutine_fn bdrv_aligned_preadv(BlockDriverState *bs,
             qemu_iovec_concat(&local_qiov, qiov, 0,
                               max_nb_sectors * BDRV_SECTOR_SIZE);
 
-            ret = drv->bdrv_co_readv(bs, sector_num, max_nb_sectors,
+            ret = bdrv_driver_preadv(bs, offset,
-                                     &local_qiov);
+                                     max_nb_sectors * BDRV_SECTOR_SIZE,
+                                     &local_qiov, 0);
 
             qemu_iovec_destroy(&local_qiov);
         } else {
@@ -967,7 +1040,7 @@ out:
 /*
  * Handle a read request in coroutine context
  */
-int coroutine_fn bdrv_co_do_preadv(BlockDriverState *bs,
+int coroutine_fn bdrv_co_preadv(BlockDriverState *bs,
     int64_t offset, unsigned int bytes, QEMUIOVector *qiov,
     BdrvRequestFlags flags)
 {
@@ -997,7 +1070,7 @@ int coroutine_fn bdrv_co_do_preadv(BlockDriverState *bs,
     }
 
     /* throttling disk I/O */
-    if (bs->io_limits_enabled) {
+    if (bs->throttle_state) {
         throttle_group_co_io_limits_intercept(bs, bytes, false);
     }
 
@@ -1049,8 +1122,8 @@ static int coroutine_fn bdrv_co_do_readv(BlockDriverState *bs,
         return -EINVAL;
     }
 
-    return bdrv_co_do_preadv(bs, sector_num << BDRV_SECTOR_BITS,
+    return bdrv_co_preadv(bs, sector_num << BDRV_SECTOR_BITS,
-                             nb_sectors << BDRV_SECTOR_BITS, qiov, flags);
+                          nb_sectors << BDRV_SECTOR_BITS, qiov, flags);
 }
 
 int coroutine_fn bdrv_co_readv(BlockDriverState *bs, int64_t sector_num,
@@ -1088,6 +1161,7 @@ static int coroutine_fn bdrv_co_do_write_zeroes(BlockDriverState *bs,
     QEMUIOVector qiov;
     struct iovec iov = {0};
     int ret = 0;
+    bool need_flush = false;
 
     int max_write_zeroes = MIN_NON_ZERO(bs->bl.max_write_zeroes,
                                         BDRV_REQUEST_MAX_SECTORS);
@@ -1120,13 +1194,29 @@ static int coroutine_fn bdrv_co_do_write_zeroes(BlockDriverState *bs,
         ret = -ENOTSUP;
         /* First try the efficient write zeroes operation */
         if (drv->bdrv_co_write_zeroes) {
-            ret = drv->bdrv_co_write_zeroes(bs, sector_num, num, flags);
+            ret = drv->bdrv_co_write_zeroes(bs, sector_num, num,
+                                            flags & bs->supported_zero_flags);
+            if (ret != -ENOTSUP && (flags & BDRV_REQ_FUA) &&
+                !(bs->supported_zero_flags & BDRV_REQ_FUA)) {
+                need_flush = true;
+            }
+        } else {
+            assert(!bs->supported_zero_flags);
         }
 
         if (ret == -ENOTSUP) {
             /* Fall back to bounce buffer if write zeroes is unsupported */
             int max_xfer_len = MIN_NON_ZERO(bs->bl.max_transfer_length,
                                             MAX_WRITE_ZEROES_BOUNCE_BUFFER);
+            BdrvRequestFlags write_flags = flags & ~BDRV_REQ_ZERO_WRITE;
+
+            if ((flags & BDRV_REQ_FUA) &&
+                !(bs->supported_write_flags & BDRV_REQ_FUA)) {
+                /* No need for bdrv_driver_pwrite() to do a fallback
+                 * flush on each chunk; use just one at the end */
+                write_flags &= ~BDRV_REQ_FUA;
+                need_flush = true;
+            }
             num = MIN(num, max_xfer_len);
             iov.iov_len = num * BDRV_SECTOR_SIZE;
             if (iov.iov_base == NULL) {
@@ -1139,7 +1229,9 @@ static int coroutine_fn bdrv_co_do_write_zeroes(BlockDriverState *bs,
             }
             qemu_iovec_init_external(&qiov, &iov, 1);
 
-            ret = drv->bdrv_co_writev(bs, sector_num, num, &qiov);
+            ret = bdrv_driver_pwritev(bs, sector_num * BDRV_SECTOR_SIZE,
+                                      num * BDRV_SECTOR_SIZE, &qiov,
+                                      write_flags);
 
             /* Keep bounce buffer around if it is big enough for all
              * all future requests.
@@ -1155,6 +1247,9 @@ static int coroutine_fn bdrv_co_do_write_zeroes(BlockDriverState *bs,
     }
 
 fail:
+    if (ret == 0 && need_flush) {
+        ret = bdrv_co_flush(bs);
+    }
     qemu_vfree(iov.iov_base);
     return ret;
 }
@@ -1199,23 +1294,12 @@ 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_write_zeroes(bs, sector_num, nb_sectors, flags);
-    } else if (drv->bdrv_co_writev_flags) {
-        bdrv_debug_event(bs, BLKDBG_PWRITEV);
-        ret = drv->bdrv_co_writev_flags(bs, sector_num, nb_sectors, qiov,
-                                        flags);
     } else {
-        assert(drv->supported_write_flags == 0);
         bdrv_debug_event(bs, BLKDBG_PWRITEV);
-        ret = drv->bdrv_co_writev(bs, sector_num, nb_sectors, qiov);
+        ret = bdrv_driver_pwritev(bs, offset, bytes, qiov, flags);
     }
     bdrv_debug_event(bs, BLKDBG_PWRITEV_DONE);
 
-    if (ret == 0 && (flags & BDRV_REQ_FUA) &&
-        !(drv->supported_write_flags & BDRV_REQ_FUA))
-    {
-        ret = bdrv_co_flush(bs);
-    }
-
     bdrv_set_dirty(bs, sector_num, nb_sectors);
 
     if (bs->wr_highest_offset < offset + bytes) {
@@ -1320,7 +1404,7 @@ fail:
 /*
  * Handle a write request in coroutine context
  */
-int coroutine_fn bdrv_co_do_pwritev(BlockDriverState *bs,
+int coroutine_fn bdrv_co_pwritev(BlockDriverState *bs,
     int64_t offset, unsigned int bytes, QEMUIOVector *qiov,
     BdrvRequestFlags flags)
 {
@@ -1347,7 +1431,7 @@ int coroutine_fn bdrv_co_do_pwritev(BlockDriverState *bs,
     }
 
     /* throttling disk I/O */
-    if (bs->io_limits_enabled) {
+    if (bs->throttle_state) {
         throttle_group_co_io_limits_intercept(bs, bytes, true);
     }
 
@@ -1455,8 +1539,8 @@ static int coroutine_fn bdrv_co_do_writev(BlockDriverState *bs,
         return -EINVAL;
     }
 
-    return bdrv_co_do_pwritev(bs, sector_num << BDRV_SECTOR_BITS,
+    return bdrv_co_pwritev(bs, sector_num << BDRV_SECTOR_BITS,
-                              nb_sectors << BDRV_SECTOR_BITS, qiov, flags);
+                           nb_sectors << BDRV_SECTOR_BITS, qiov, flags);
 }
 
 int coroutine_fn bdrv_co_writev(BlockDriverState *bs, int64_t sector_num,
@@ -2064,80 +2148,6 @@ void bdrv_aio_cancel_async(BlockAIOCB *acb)
 /**************************************************************/
 /* async block device emulation */
 
-typedef struct BlockAIOCBSync {
-    BlockAIOCB common;
-    QEMUBH *bh;
-    int ret;
-    /* vector translation state */
-    QEMUIOVector *qiov;
-    uint8_t *bounce;
-    int is_write;
-} BlockAIOCBSync;
-
-static const AIOCBInfo bdrv_em_aiocb_info = {
-    .aiocb_size         = sizeof(BlockAIOCBSync),
-};
-
-static void bdrv_aio_bh_cb(void *opaque)
-{
-    BlockAIOCBSync *acb = opaque;
-
-    if (!acb->is_write && acb->ret >= 0) {
-        qemu_iovec_from_buf(acb->qiov, 0, acb->bounce, acb->qiov->size);
-    }
-    qemu_vfree(acb->bounce);
-    acb->common.cb(acb->common.opaque, acb->ret);
-    qemu_bh_delete(acb->bh);
-    acb->bh = NULL;
-    qemu_aio_unref(acb);
-}
-
-static BlockAIOCB *bdrv_aio_rw_vector(BlockDriverState *bs,
-                                      int64_t sector_num,
-                                      QEMUIOVector *qiov,
-                                      int nb_sectors,
-                                      BlockCompletionFunc *cb,
-                                      void *opaque,
-                                      int is_write)
-
-{
-    BlockAIOCBSync *acb;
-
-    acb = qemu_aio_get(&bdrv_em_aiocb_info, bs, cb, opaque);
-    acb->is_write = is_write;
-    acb->qiov = qiov;
-    acb->bounce = qemu_try_blockalign(bs, qiov->size);
-    acb->bh = aio_bh_new(bdrv_get_aio_context(bs), bdrv_aio_bh_cb, acb);
-
-    if (acb->bounce == NULL) {
-        acb->ret = -ENOMEM;
-    } else if (is_write) {
-        qemu_iovec_to_buf(acb->qiov, 0, acb->bounce, qiov->size);
-        acb->ret = bs->drv->bdrv_write(bs, sector_num, acb->bounce, nb_sectors);
-    } else {
-        acb->ret = bs->drv->bdrv_read(bs, sector_num, acb->bounce, nb_sectors);
-    }
-
-    qemu_bh_schedule(acb->bh);
-
-    return &acb->common;
-}
-
-static BlockAIOCB *bdrv_aio_readv_em(BlockDriverState *bs,
-        int64_t sector_num, QEMUIOVector *qiov, int nb_sectors,
-        BlockCompletionFunc *cb, void *opaque)
-{
-    return bdrv_aio_rw_vector(bs, sector_num, qiov, nb_sectors, cb, opaque, 0);
-}
-
-static BlockAIOCB *bdrv_aio_writev_em(BlockDriverState *bs,
-        int64_t sector_num, QEMUIOVector *qiov, int nb_sectors,
-        BlockCompletionFunc *cb, void *opaque)
-{
-    return bdrv_aio_rw_vector(bs, sector_num, qiov, nb_sectors, cb, opaque, 1);
-}
-
-
 typedef struct BlockAIOCBCoroutine {
     BlockAIOCB common;
     BlockRequest req;
@@ -2314,59 +2324,6 @@ void qemu_aio_unref(void *p)
 /**************************************************************/
 /* Coroutine block device emulation */
 
-typedef struct CoroutineIOCompletion {
-    Coroutine *coroutine;
-    int ret;
-} CoroutineIOCompletion;
-
-static void bdrv_co_io_em_complete(void *opaque, int ret)
-{
-    CoroutineIOCompletion *co = opaque;
-
-    co->ret = ret;
-    qemu_coroutine_enter(co->coroutine, NULL);
-}
-
-static int coroutine_fn bdrv_co_io_em(BlockDriverState *bs, int64_t sector_num,
-                                      int nb_sectors, QEMUIOVector *iov,
-                                      bool is_write)
-{
-    CoroutineIOCompletion co = {
-        .coroutine = qemu_coroutine_self(),
-    };
-    BlockAIOCB *acb;
-
-    if (is_write) {
-        acb = bs->drv->bdrv_aio_writev(bs, sector_num, iov, nb_sectors,
-                                       bdrv_co_io_em_complete, &co);
-    } else {
-        acb = bs->drv->bdrv_aio_readv(bs, sector_num, iov, nb_sectors,
-                                      bdrv_co_io_em_complete, &co);
-    }
-
-    trace_bdrv_co_io_em(bs, sector_num, nb_sectors, is_write, acb);
-    if (!acb) {
-        return -EIO;
-    }
-    qemu_coroutine_yield();
-
-    return co.ret;
-}
-
-static int coroutine_fn bdrv_co_readv_em(BlockDriverState *bs,
-                                         int64_t sector_num, int nb_sectors,
-                                         QEMUIOVector *iov)
-{
-    return bdrv_co_io_em(bs, sector_num, nb_sectors, iov, false);
-}
-
-static int coroutine_fn bdrv_co_writev_em(BlockDriverState *bs,
-                                         int64_t sector_num, int nb_sectors,
-                                         QEMUIOVector *iov)
-{
-    return bdrv_co_io_em(bs, sector_num, nb_sectors, iov, true);
-}
-
 static void coroutine_fn bdrv_flush_co_entry(void *opaque)
 {
     RwCo *rwco = opaque;
@@ -2763,33 +2720,68 @@ void bdrv_add_before_write_notifier(BlockDriverState *bs,
 
 void bdrv_io_plug(BlockDriverState *bs)
 {
-    BlockDriver *drv = bs->drv;
+    BdrvChild *child;
-    if (drv && drv->bdrv_io_plug) {
+
-        drv->bdrv_io_plug(bs);
+    QLIST_FOREACH(child, &bs->children, next) {
-    } else if (bs->file) {
+        bdrv_io_plug(child->bs);
-        bdrv_io_plug(bs->file->bs);
+    }
+
+    if (bs->io_plugged++ == 0 && bs->io_plug_disabled == 0) {
+        BlockDriver *drv = bs->drv;
+        if (drv && drv->bdrv_io_plug) {
+            drv->bdrv_io_plug(bs);
+        }
     }
 }
 
 void bdrv_io_unplug(BlockDriverState *bs)
 {
-    BlockDriver *drv = bs->drv;
+    BdrvChild *child;
-    if (drv && drv->bdrv_io_unplug) {
+
-        drv->bdrv_io_unplug(bs);
+    assert(bs->io_plugged);
-    } else if (bs->file) {
+    if (--bs->io_plugged == 0 && bs->io_plug_disabled == 0) {
-        bdrv_io_unplug(bs->file->bs);
+        BlockDriver *drv = bs->drv;
+        if (drv && drv->bdrv_io_unplug) {
+            drv->bdrv_io_unplug(bs);
+        }
+    }
+
+    QLIST_FOREACH(child, &bs->children, next) {
+        bdrv_io_unplug(child->bs);
     }
 }
 
-void bdrv_flush_io_queue(BlockDriverState *bs)
+void bdrv_io_unplugged_begin(BlockDriverState *bs)
 {
-    BlockDriver *drv = bs->drv;
+    BdrvChild *child;
-    if (drv && drv->bdrv_flush_io_queue) {
+
-        drv->bdrv_flush_io_queue(bs);
+    if (bs->io_plug_disabled++ == 0 && bs->io_plugged > 0) {
-    } else if (bs->file) {
+        BlockDriver *drv = bs->drv;
-        bdrv_flush_io_queue(bs->file->bs);
+        if (drv && drv->bdrv_io_unplug) {
+            drv->bdrv_io_unplug(bs);
+        }
+    }
+
+    QLIST_FOREACH(child, &bs->children, next) {
+        bdrv_io_unplugged_begin(child->bs);
+    }
+}
+
+void bdrv_io_unplugged_end(BlockDriverState *bs)
+{
+    BdrvChild *child;
+
+    assert(bs->io_plug_disabled);
+    QLIST_FOREACH(child, &bs->children, next) {
+        bdrv_io_unplugged_end(child->bs);
+    }
+
+    if (--bs->io_plug_disabled == 0 && bs->io_plugged > 0) {
+        BlockDriver *drv = bs->drv;
+        if (drv && drv->bdrv_io_plug) {
+            drv->bdrv_io_plug(bs);
+        }
     }
-    bdrv_start_throttled_reqs(bs);
 }
 
 void bdrv_drained_begin(BlockDriverState *bs)

block/iscsi.c

@@ -456,8 +456,11 @@ iscsi_co_writev_flags(BlockDriverState *bs, int64_t sector_num, int nb_sectors,
     struct IscsiTask iTask;
     uint64_t lba;
     uint32_t num_sectors;
-    bool fua;
+    bool fua = flags & BDRV_REQ_FUA;
 
+    if (fua) {
+        assert(iscsilun->dpofua);
+    }
     if (!is_request_lun_aligned(sector_num, nb_sectors, iscsilun)) {
         return -EINVAL;
     }
@@ -472,7 +475,6 @@ iscsi_co_writev_flags(BlockDriverState *bs, int64_t sector_num, int nb_sectors,
     num_sectors = sector_qemu2lun(nb_sectors, iscsilun);
     iscsi_co_init_iscsitask(iscsilun, &iTask);
 retry:
-    fua = iscsilun->dpofua && (flags & BDRV_REQ_FUA);
     if (iscsilun->use_16_for_rw) {
         iTask.task = iscsi_write16_task(iscsilun->iscsi, iscsilun->lun, lba,
                                         NULL, num_sectors * iscsilun->block_size,
@@ -513,13 +515,6 @@ retry:
     return 0;
 }
 
-static int coroutine_fn
-iscsi_co_writev(BlockDriverState *bs, int64_t sector_num, int nb_sectors,
-                QEMUIOVector *iov)
-{
-    return iscsi_co_writev_flags(bs, sector_num, nb_sectors, iov, 0);
-}
-
 
 static bool iscsi_allocationmap_is_allocated(IscsiLun *iscsilun,
                                              int64_t sector_num, int nb_sectors)
@@ -1555,6 +1550,10 @@ static int iscsi_open(BlockDriverState *bs, QDict *options, int flags,
     task = NULL;
 
     iscsi_modesense_sync(iscsilun);
+    if (iscsilun->dpofua) {
+        bs->supported_write_flags = BDRV_REQ_FUA;
+    }
+    bs->supported_zero_flags = BDRV_REQ_MAY_UNMAP;
 
     /* Check the write protect flag of the LUN if we want to write */
     if (iscsilun->type == TYPE_DISK && (flags & BDRV_O_RDWR) &&
@@ -1847,9 +1846,7 @@ static BlockDriver bdrv_iscsi = {
     .bdrv_co_discard      = iscsi_co_discard,
     .bdrv_co_write_zeroes = iscsi_co_write_zeroes,
     .bdrv_co_readv         = iscsi_co_readv,
-    .bdrv_co_writev        = iscsi_co_writev,
     .bdrv_co_writev_flags  = iscsi_co_writev_flags,
-    .supported_write_flags = BDRV_REQ_FUA,
     .bdrv_co_flush_to_disk = iscsi_co_flush,
 
 #ifdef __linux__

block/linux-aio.c

@@ -30,7 +30,7 @@
 
 struct qemu_laiocb {
     BlockAIOCB common;
-    struct qemu_laio_state *ctx;
+    LinuxAioState *ctx;
     struct iocb iocb;
     ssize_t ret;
     size_t nbytes;
@@ -46,7 +46,7 @@ typedef struct {
     QSIMPLEQ_HEAD(, qemu_laiocb) pending;
 } LaioQueue;
 
-struct qemu_laio_state {
+struct LinuxAioState {
     io_context_t ctx;
     EventNotifier e;
 
@@ -60,7 +60,7 @@ struct qemu_laio_state {
     int event_max;
 };
 
-static void ioq_submit(struct qemu_laio_state *s);
+static void ioq_submit(LinuxAioState *s);
 
 static inline ssize_t io_event_ret(struct io_event *ev)
 {
@@ -70,8 +70,7 @@ static inline ssize_t io_event_ret(struct io_event *ev)
 /*
  * Completes an AIO request (calls the callback and frees the ACB).
  */
-static void qemu_laio_process_completion(struct qemu_laio_state *s,
+static void qemu_laio_process_completion(struct qemu_laiocb *laiocb)
-    struct qemu_laiocb *laiocb)
 {
     int ret;
 
@@ -99,7 +98,7 @@ static void qemu_laio_process_completion(struct qemu_laio_state *s,
  *
  * The function is somewhat tricky because it supports nested event loops, for
  * example when a request callback invokes aio_poll().  In order to do this,
- * the completion events array and index are kept in qemu_laio_state.  The BH
+ * the completion events array and index are kept in LinuxAioState.  The BH
  * reschedules itself as long as there are completions pending so it will
  * either be called again in a nested event loop or will be called after all
  * events have been completed.  When there are no events left to complete, the
@@ -107,7 +106,7 @@ static void qemu_laio_process_completion(struct qemu_laio_state *s,
  */
 static void qemu_laio_completion_bh(void *opaque)
 {
-    struct qemu_laio_state *s = opaque;
+    LinuxAioState *s = opaque;
 
     /* Fetch more completion events when empty */
     if (s->event_idx == s->event_max) {
@@ -136,7 +135,7 @@ static void qemu_laio_completion_bh(void *opaque)
         laiocb->ret = io_event_ret(&s->events[s->event_idx]);
         s->event_idx++;
 
-        qemu_laio_process_completion(s, laiocb);
+        qemu_laio_process_completion(laiocb);
     }
 
     if (!s->io_q.plugged && !QSIMPLEQ_EMPTY(&s->io_q.pending)) {
@@ -146,7 +145,7 @@ static void qemu_laio_completion_bh(void *opaque)
 
 static void qemu_laio_completion_cb(EventNotifier *e)
 {
-    struct qemu_laio_state *s = container_of(e, struct qemu_laio_state, e);
+    LinuxAioState *s = container_of(e, LinuxAioState, e);
 
     if (event_notifier_test_and_clear(&s->e)) {
         qemu_bh_schedule(s->completion_bh);
@@ -185,7 +184,7 @@ static void ioq_init(LaioQueue *io_q)
     io_q->blocked = false;
 }
 
-static void ioq_submit(struct qemu_laio_state *s)
+static void ioq_submit(LinuxAioState *s)
 {
     int ret, len;
     struct qemu_laiocb *aiocb;
@@ -216,33 +215,25 @@ static void ioq_submit(struct qemu_laio_state *s)
     s->io_q.blocked = (s->io_q.n > 0);
 }
 
-void laio_io_plug(BlockDriverState *bs, void *aio_ctx)
+void laio_io_plug(BlockDriverState *bs, LinuxAioState *s)
 {
-    struct qemu_laio_state *s = aio_ctx;
+    assert(!s->io_q.plugged);
-
+    s->io_q.plugged = 1;
-    s->io_q.plugged++;
 }
 
-void laio_io_unplug(BlockDriverState *bs, void *aio_ctx, bool unplug)
+void laio_io_unplug(BlockDriverState *bs, LinuxAioState *s)
 {
-    struct qemu_laio_state *s = aio_ctx;
+    assert(s->io_q.plugged);
-
+    s->io_q.plugged = 0;
-    assert(s->io_q.plugged > 0 || !unplug);
-
-    if (unplug && --s->io_q.plugged > 0) {
-        return;
-    }
-
     if (!s->io_q.blocked && !QSIMPLEQ_EMPTY(&s->io_q.pending)) {
         ioq_submit(s);
     }
 }
 
-BlockAIOCB *laio_submit(BlockDriverState *bs, void *aio_ctx, int fd,
+BlockAIOCB *laio_submit(BlockDriverState *bs, LinuxAioState *s, int fd,
         int64_t sector_num, QEMUIOVector *qiov, int nb_sectors,
         BlockCompletionFunc *cb, void *opaque, int type)
 {
-    struct qemu_laio_state *s = aio_ctx;
     struct qemu_laiocb *laiocb;
     struct iocb *iocbs;
     off_t offset = sector_num * 512;
@@ -284,26 +275,22 @@ out_free_aiocb:
     return NULL;
 }
 
-void laio_detach_aio_context(void *s_, AioContext *old_context)
+void laio_detach_aio_context(LinuxAioState *s, AioContext *old_context)
 {
-    struct qemu_laio_state *s = s_;
-
     aio_set_event_notifier(old_context, &s->e, false, NULL);
     qemu_bh_delete(s->completion_bh);
 }
 
-void laio_attach_aio_context(void *s_, AioContext *new_context)
+void laio_attach_aio_context(LinuxAioState *s, AioContext *new_context)
 {
-    struct qemu_laio_state *s = s_;
-
     s->completion_bh = aio_bh_new(new_context, qemu_laio_completion_bh, s);
     aio_set_event_notifier(new_context, &s->e, false,
                            qemu_laio_completion_cb);
 }
 
-void *laio_init(void)
+LinuxAioState *laio_init(void)
 {
-    struct qemu_laio_state *s;
+    LinuxAioState *s;
 
     s = g_malloc0(sizeof(*s));
     if (event_notifier_init(&s->e, false) < 0) {
@@ -325,10 +312,8 @@ out_free_state:
     return NULL;
 }
 
-void laio_cleanup(void *s_)
+void laio_cleanup(LinuxAioState *s)
 {
-    struct qemu_laio_state *s = s_;
-
     event_notifier_cleanup(&s->e);
 
     if (io_destroy(s->ctx) != 0) {

block/nbd-client.c

@@ -243,15 +243,15 @@ static int nbd_co_readv_1(BlockDriverState *bs, int64_t sector_num,
 
 static int nbd_co_writev_1(BlockDriverState *bs, int64_t sector_num,
                            int nb_sectors, QEMUIOVector *qiov,
-                           int offset, int *flags)
+                           int offset, int flags)
 {
     NbdClientSession *client = nbd_get_client_session(bs);
     struct nbd_request request = { .type = NBD_CMD_WRITE };
     struct nbd_reply reply;
     ssize_t ret;
 
-    if ((*flags & BDRV_REQ_FUA) && (client->nbdflags & NBD_FLAG_SEND_FUA)) {
+    if (flags & BDRV_REQ_FUA) {
-        *flags &= ~BDRV_REQ_FUA;
+        assert(client->nbdflags & NBD_FLAG_SEND_FUA);
         request.type |= NBD_CMD_FLAG_FUA;
     }
 
@@ -291,7 +291,7 @@ int nbd_client_co_readv(BlockDriverState *bs, int64_t sector_num,
 }
 
 int nbd_client_co_writev(BlockDriverState *bs, int64_t sector_num,
-                         int nb_sectors, QEMUIOVector *qiov, int *flags)
+                         int nb_sectors, QEMUIOVector *qiov, int flags)
 {
     int offset = 0;
     int ret;
@@ -414,6 +414,9 @@ int nbd_client_init(BlockDriverState *bs,
         logout("Failed to negotiate with the NBD server\n");
         return ret;
     }
+    if (client->nbdflags & NBD_FLAG_SEND_FUA) {
+        bs->supported_write_flags = BDRV_REQ_FUA;
+    }
 
     qemu_co_mutex_init(&client->send_mutex);
     qemu_co_mutex_init(&client->free_sema);

block/nbd-client.h

@@ -48,7 +48,7 @@ int nbd_client_co_discard(BlockDriverState *bs, int64_t sector_num,
                           int nb_sectors);
 int nbd_client_co_flush(BlockDriverState *bs);
 int nbd_client_co_writev(BlockDriverState *bs, int64_t sector_num,
-                         int nb_sectors, QEMUIOVector *qiov, int *flags);
+                         int nb_sectors, QEMUIOVector *qiov, int flags);
 int nbd_client_co_readv(BlockDriverState *bs, int64_t sector_num,
                         int nb_sectors, QEMUIOVector *qiov);
 

block/nbd.c

@@ -355,31 +355,6 @@ static int nbd_co_readv(BlockDriverState *bs, int64_t sector_num,
     return nbd_client_co_readv(bs, sector_num, nb_sectors, qiov);
 }
 
-static int nbd_co_writev_flags(BlockDriverState *bs, int64_t sector_num,
-                               int nb_sectors, QEMUIOVector *qiov, int flags)
-{
-    int ret;
-
-    ret = nbd_client_co_writev(bs, sector_num, nb_sectors, qiov, &flags);
-    if (ret < 0) {
-        return ret;
-    }
-
-    /* The flag wasn't sent to the server, so we need to emulate it with an
-     * explicit flush */
-    if (flags & BDRV_REQ_FUA) {
-        ret = nbd_client_co_flush(bs);
-    }
-
-    return ret;
-}
-
-static int nbd_co_writev(BlockDriverState *bs, int64_t sector_num,
-                         int nb_sectors, QEMUIOVector *qiov)
-{
-    return nbd_co_writev_flags(bs, sector_num, nb_sectors, qiov, 0);
-}
-
 static int nbd_co_flush(BlockDriverState *bs)
 {
     return nbd_client_co_flush(bs);
@@ -476,9 +451,7 @@ static BlockDriver bdrv_nbd = {
     .bdrv_parse_filename        = nbd_parse_filename,
     .bdrv_file_open             = nbd_open,
     .bdrv_co_readv              = nbd_co_readv,
-    .bdrv_co_writev             = nbd_co_writev,
+    .bdrv_co_writev_flags       = nbd_client_co_writev,
-    .bdrv_co_writev_flags       = nbd_co_writev_flags,
-    .supported_write_flags      = BDRV_REQ_FUA,
     .bdrv_close                 = nbd_close,
     .bdrv_co_flush_to_os        = nbd_co_flush,
     .bdrv_co_discard            = nbd_co_discard,
@@ -496,9 +469,7 @@ static BlockDriver bdrv_nbd_tcp = {
     .bdrv_parse_filename        = nbd_parse_filename,
     .bdrv_file_open             = nbd_open,
     .bdrv_co_readv              = nbd_co_readv,
-    .bdrv_co_writev             = nbd_co_writev,
+    .bdrv_co_writev_flags       = nbd_client_co_writev,
-    .bdrv_co_writev_flags       = nbd_co_writev_flags,
-    .supported_write_flags      = BDRV_REQ_FUA,
     .bdrv_close                 = nbd_close,
     .bdrv_co_flush_to_os        = nbd_co_flush,
     .bdrv_co_discard            = nbd_co_discard,
@@ -516,9 +487,7 @@ static BlockDriver bdrv_nbd_unix = {
     .bdrv_parse_filename        = nbd_parse_filename,
     .bdrv_file_open             = nbd_open,
     .bdrv_co_readv              = nbd_co_readv,
-    .bdrv_co_writev             = nbd_co_writev,
+    .bdrv_co_writev_flags       = nbd_client_co_writev,
-    .bdrv_co_writev_flags       = nbd_co_writev_flags,
-    .supported_write_flags      = BDRV_REQ_FUA,
     .bdrv_close                 = nbd_close,
     .bdrv_co_flush_to_os        = nbd_co_flush,
     .bdrv_co_discard            = nbd_co_discard,

block/parallels.c

@@ -512,11 +512,12 @@ static int parallels_create(const char *filename, QemuOpts *opts, Error **errp)
     memset(tmp, 0, sizeof(tmp));
     memcpy(tmp, &header, sizeof(header));
 
-    ret = blk_pwrite(file, 0, tmp, BDRV_SECTOR_SIZE);
+    ret = blk_pwrite(file, 0, tmp, BDRV_SECTOR_SIZE, 0);
     if (ret < 0) {
         goto exit;
     }
-    ret = blk_write_zeroes(file, 1, bat_sectors - 1, 0);
+    ret = blk_write_zeroes(file, BDRV_SECTOR_SIZE,
+                           (bat_sectors - 1) << BDRV_SECTOR_BITS, 0);
     if (ret < 0) {
         goto exit;
     }

block/qcow.c

@@ -853,14 +853,14 @@ static int qcow_create(const char *filename, QemuOpts *opts, Error **errp)
     }
 
     /* write all the data */
-    ret = blk_pwrite(qcow_blk, 0, &header, sizeof(header));
+    ret = blk_pwrite(qcow_blk, 0, &header, sizeof(header), 0);
     if (ret != sizeof(header)) {
         goto exit;
     }
 
     if (backing_file) {
         ret = blk_pwrite(qcow_blk, sizeof(header),
-            backing_file, backing_filename_len);
+                         backing_file, backing_filename_len, 0);
         if (ret != backing_filename_len) {
             goto exit;
         }
@@ -869,8 +869,8 @@ static int qcow_create(const char *filename, QemuOpts *opts, Error **errp)
     tmp = g_malloc0(BDRV_SECTOR_SIZE);
     for (i = 0; i < ((sizeof(uint64_t)*l1_size + BDRV_SECTOR_SIZE - 1)/
         BDRV_SECTOR_SIZE); i++) {
-        ret = blk_pwrite(qcow_blk, header_size +
+        ret = blk_pwrite(qcow_blk, header_size + BDRV_SECTOR_SIZE * i,
-            BDRV_SECTOR_SIZE*i, tmp, BDRV_SECTOR_SIZE);
+                         tmp, BDRV_SECTOR_SIZE, 0);
         if (ret != BDRV_SECTOR_SIZE) {
             g_free(tmp);
             goto exit;

block/qcow2.c

@@ -1757,13 +1757,6 @@ static void qcow2_invalidate_cache(BlockDriverState *bs, Error **errp)
 
     qcow2_close(bs);
 
-    bdrv_invalidate_cache(bs->file->bs, &local_err);
-    if (local_err) {
-        error_propagate(errp, local_err);
-        bs->drv = NULL;
-        return;
-    }
-
     memset(s, 0, sizeof(BDRVQcow2State));
     options = qdict_clone_shallow(bs->options);
 
@@ -2207,7 +2200,7 @@ static int qcow2_create2(const char *filename, int64_t total_size,
             cpu_to_be64(QCOW2_COMPAT_LAZY_REFCOUNTS);
     }
 
-    ret = blk_pwrite(blk, 0, header, cluster_size);
+    ret = blk_pwrite(blk, 0, header, cluster_size, 0);
     g_free(header);
     if (ret < 0) {
         error_setg_errno(errp, -ret, "Could not write qcow2 header");
@@ -2217,7 +2210,7 @@ static int qcow2_create2(const char *filename, int64_t total_size,
     /* Write a refcount table with one refcount block */
     refcount_table = g_malloc0(2 * cluster_size);
     refcount_table[0] = cpu_to_be64(2 * cluster_size);
-    ret = blk_pwrite(blk, cluster_size, refcount_table, 2 * cluster_size);
+    ret = blk_pwrite(blk, cluster_size, refcount_table, 2 * cluster_size, 0);
     g_free(refcount_table);
 
     if (ret < 0) {
@@ -2411,21 +2404,74 @@ finish:
     return ret;
 }
 
+
+static bool is_zero_cluster(BlockDriverState *bs, int64_t start)
+{
+    BDRVQcow2State *s = bs->opaque;
+    int nr;
+    BlockDriverState *file;
+    int64_t res = bdrv_get_block_status_above(bs, NULL, start,
+                                              s->cluster_sectors, &nr, &file);
+    return res >= 0 && ((res & BDRV_BLOCK_ZERO) || !(res & BDRV_BLOCK_DATA));
+}
+
+static bool is_zero_cluster_top_locked(BlockDriverState *bs, int64_t start)
+{
+    BDRVQcow2State *s = bs->opaque;
+    int nr = s->cluster_sectors;
+    uint64_t off;
+    int ret;
+
+    ret = qcow2_get_cluster_offset(bs, start << BDRV_SECTOR_BITS, &nr, &off);
+    return ret == QCOW2_CLUSTER_UNALLOCATED || ret == QCOW2_CLUSTER_ZERO;
+}
+
 static coroutine_fn int qcow2_co_write_zeroes(BlockDriverState *bs,
     int64_t sector_num, int nb_sectors, BdrvRequestFlags flags)
 {
     int ret;
     BDRVQcow2State *s = bs->opaque;
 
-    /* Emulate misaligned zero writes */
+    int head = sector_num % s->cluster_sectors;
-    if (sector_num % s->cluster_sectors || nb_sectors % s->cluster_sectors) {
+    int tail = (sector_num + nb_sectors) % s->cluster_sectors;
-        return -ENOTSUP;
+
+    if (head != 0 || tail != 0) {
+        int64_t cl_end = -1;
+
+        sector_num -= head;
+        nb_sectors += head;
+
+        if (tail != 0) {
+            nb_sectors += s->cluster_sectors - tail;
+        }
+
+        if (!is_zero_cluster(bs, sector_num)) {
+            return -ENOTSUP;
+        }
+
+        if (nb_sectors > s->cluster_sectors) {
+            /* Technically the request can cover 2 clusters, f.e. 4k write
+               at s->cluster_sectors - 2k offset. One of these cluster can
+               be zeroed, one unallocated */
+            cl_end = sector_num + nb_sectors - s->cluster_sectors;
+            if (!is_zero_cluster(bs, cl_end)) {
+                return -ENOTSUP;
+            }
+        }
+
+        qemu_co_mutex_lock(&s->lock);
+        /* We can have new write after previous check */
+        if (!is_zero_cluster_top_locked(bs, sector_num) ||
+                (cl_end > 0 && !is_zero_cluster_top_locked(bs, cl_end))) {
+            qemu_co_mutex_unlock(&s->lock);
+            return -ENOTSUP;
+        }
+    } else {
+        qemu_co_mutex_lock(&s->lock);
     }
 
     /* Whatever is left can use real zero clusters */
-    qemu_co_mutex_lock(&s->lock);
+    ret = qcow2_zero_clusters(bs, sector_num << BDRV_SECTOR_BITS, nb_sectors);
-    ret = qcow2_zero_clusters(bs, sector_num << BDRV_SECTOR_BITS,
-        nb_sectors);
     qemu_co_mutex_unlock(&s->lock);
 
     return ret;

block/qed.c

@@ -601,18 +601,18 @@ static int qed_create(const char *filename, uint32_t cluster_size,
     }
 
     qed_header_cpu_to_le(&header, &le_header);
-    ret = blk_pwrite(blk, 0, &le_header, sizeof(le_header));
+    ret = blk_pwrite(blk, 0, &le_header, sizeof(le_header), 0);
     if (ret < 0) {
         goto out;
     }
     ret = blk_pwrite(blk, sizeof(le_header), backing_file,
-                     header.backing_filename_size);
+                     header.backing_filename_size, 0);
     if (ret < 0) {
         goto out;
     }
 
     l1_table = g_malloc0(l1_size);
-    ret = blk_pwrite(blk, header.l1_table_offset, l1_table, l1_size);
+    ret = blk_pwrite(blk, header.l1_table_offset, l1_table, l1_size, 0);
     if (ret < 0) {
         goto out;
     }
@@ -1594,12 +1594,6 @@ static void bdrv_qed_invalidate_cache(BlockDriverState *bs, Error **errp)
 
     bdrv_qed_close(bs);
 
-    bdrv_invalidate_cache(bs->file->bs, &local_err);
-    if (local_err) {
-        error_propagate(errp, local_err);
-        return;
-    }
-
     memset(s, 0, sizeof(BDRVQEDState));
     ret = bdrv_qed_open(bs, NULL, bs->open_flags, &local_err);
     if (local_err) {

block/quorum.c

@@ -14,6 +14,7 @@
  */
 
 #include "qemu/osdep.h"
+#include "qemu/cutils.h"
 #include "block/block_int.h"
 #include "qapi/qmp/qbool.h"
 #include "qapi/qmp/qdict.h"
@@ -67,6 +68,9 @@ typedef struct QuorumVotes {
 typedef struct BDRVQuorumState {
     BdrvChild **children;  /* children BlockDriverStates */
     int num_children;      /* children count */
+    unsigned next_child_index;  /* the index of the next child that should
+                                 * be added
+                                 */
     int threshold;         /* if less than threshold children reads gave the
                             * same result a quorum error occurs.
                             */
@@ -747,21 +751,6 @@ static int64_t quorum_getlength(BlockDriverState *bs)
     return result;
 }
 
-static void quorum_invalidate_cache(BlockDriverState *bs, Error **errp)
-{
-    BDRVQuorumState *s = bs->opaque;
-    Error *local_err = NULL;
-    int i;
-
-    for (i = 0; i < s->num_children; i++) {
-        bdrv_invalidate_cache(s->children[i]->bs, &local_err);
-        if (local_err) {
-            error_propagate(errp, local_err);
-            return;
-        }
-    }
-}
-
 static coroutine_fn int quorum_co_flush(BlockDriverState *bs)
 {
     BDRVQuorumState *s = bs->opaque;
@@ -898,9 +887,9 @@ static int quorum_open(BlockDriverState *bs, QDict *options, int flags,
         ret = -EINVAL;
         goto exit;
     }
-    if (s->num_children < 2) {
+    if (s->num_children < 1) {
         error_setg(&local_err,
-                   "Number of provided children must be greater than 1");
+                   "Number of provided children must be 1 or more");
         ret = -EINVAL;
         goto exit;
     }
@@ -964,6 +953,7 @@ static int quorum_open(BlockDriverState *bs, QDict *options, int flags,
 
         opened[i] = true;
     }
+    s->next_child_index = s->num_children;
 
     g_free(opened);
     goto exit;
@@ -1020,6 +1010,72 @@ static void quorum_attach_aio_context(BlockDriverState *bs,
     }
 }
 
+static void quorum_add_child(BlockDriverState *bs, BlockDriverState *child_bs,
+                             Error **errp)
+{
+    BDRVQuorumState *s = bs->opaque;
+    BdrvChild *child;
+    char indexstr[32];
+    int ret;
+
+    assert(s->num_children <= INT_MAX / sizeof(BdrvChild *));
+    if (s->num_children == INT_MAX / sizeof(BdrvChild *) ||
+        s->next_child_index == UINT_MAX) {
+        error_setg(errp, "Too many children");
+        return;
+    }
+
+    ret = snprintf(indexstr, 32, "children.%u", s->next_child_index);
+    if (ret < 0 || ret >= 32) {
+        error_setg(errp, "cannot generate child name");
+        return;
+    }
+    s->next_child_index++;
+
+    bdrv_drained_begin(bs);
+
+    /* We can safely add the child now */
+    bdrv_ref(child_bs);
+    child = bdrv_attach_child(bs, child_bs, indexstr, &child_format);
+    s->children = g_renew(BdrvChild *, s->children, s->num_children + 1);
+    s->children[s->num_children++] = child;
+
+    bdrv_drained_end(bs);
+}
+
+static void quorum_del_child(BlockDriverState *bs, BdrvChild *child,
+                             Error **errp)
+{
+    BDRVQuorumState *s = bs->opaque;
+    int i;
+
+    for (i = 0; i < s->num_children; i++) {
+        if (s->children[i] == child) {
+            break;
+        }
+    }
+
+    /* we have checked it in bdrv_del_child() */
+    assert(i < s->num_children);
+
+    if (s->num_children <= s->threshold) {
+        error_setg(errp,
+            "The number of children cannot be lower than the vote threshold %d",
+            s->threshold);
+        return;
+    }
+
+    bdrv_drained_begin(bs);
+
+    /* We can safely remove this child now */
+    memmove(&s->children[i], &s->children[i + 1],
+            (s->num_children - i - 1) * sizeof(BdrvChild *));
+    s->children = g_renew(BdrvChild *, s->children, --s->num_children);
+    bdrv_unref_child(bs, child);
+
+    bdrv_drained_end(bs);
+}
+
 static void quorum_refresh_filename(BlockDriverState *bs, QDict *options)
 {
     BDRVQuorumState *s = bs->opaque;
@@ -1070,11 +1126,13 @@ static BlockDriver bdrv_quorum = {
 
     .bdrv_aio_readv                     = quorum_aio_readv,
     .bdrv_aio_writev                    = quorum_aio_writev,
-    .bdrv_invalidate_cache              = quorum_invalidate_cache,
 
     .bdrv_detach_aio_context            = quorum_detach_aio_context,
     .bdrv_attach_aio_context            = quorum_attach_aio_context,
 
+    .bdrv_add_child                     = quorum_add_child,
+    .bdrv_del_child                     = quorum_del_child,
+
     .is_filter                          = true,
     .bdrv_recurse_is_first_non_filter   = quorum_recurse_is_first_non_filter,
 };

block/raw-aio.h

@@ -35,15 +35,16 @@
 
 /* linux-aio.c - Linux native implementation */
 #ifdef CONFIG_LINUX_AIO
-void *laio_init(void);
+typedef struct LinuxAioState LinuxAioState;
-void laio_cleanup(void *s);
+LinuxAioState *laio_init(void);
-BlockAIOCB *laio_submit(BlockDriverState *bs, void *aio_ctx, int fd,
+void laio_cleanup(LinuxAioState *s);
+BlockAIOCB *laio_submit(BlockDriverState *bs, LinuxAioState *s, int fd,
         int64_t sector_num, QEMUIOVector *qiov, int nb_sectors,
         BlockCompletionFunc *cb, void *opaque, int type);
-void laio_detach_aio_context(void *s, AioContext *old_context);
+void laio_detach_aio_context(LinuxAioState *s, AioContext *old_context);
-void laio_attach_aio_context(void *s, AioContext *new_context);
+void laio_attach_aio_context(LinuxAioState *s, AioContext *new_context);
-void laio_io_plug(BlockDriverState *bs, void *aio_ctx);
+void laio_io_plug(BlockDriverState *bs, LinuxAioState *s);
-void laio_io_unplug(BlockDriverState *bs, void *aio_ctx, bool unplug);
+void laio_io_unplug(BlockDriverState *bs, LinuxAioState *s);
 #endif
 
 #ifdef _WIN32

block/raw-posix.c

@@ -139,7 +139,7 @@ typedef struct BDRVRawState {
 
 #ifdef CONFIG_LINUX_AIO
     int use_aio;
-    void *aio_ctx;
+    LinuxAioState *aio_ctx;
 #endif
 #ifdef CONFIG_XFS
     bool is_xfs:1;
@@ -398,7 +398,7 @@ static void raw_attach_aio_context(BlockDriverState *bs,
 }
 
 #ifdef CONFIG_LINUX_AIO
-static int raw_set_aio(void **aio_ctx, int *use_aio, int bdrv_flags)
+static int raw_set_aio(LinuxAioState **aio_ctx, int *use_aio, int bdrv_flags)
 {
     int ret = -1;
     assert(aio_ctx != NULL);
@@ -517,6 +517,7 @@ static int raw_open_common(BlockDriverState *bs, QDict *options,
 
     s->has_discard = true;
     s->has_write_zeroes = true;
+    bs->supported_zero_flags = BDRV_REQ_MAY_UNMAP;
     if ((bs->open_flags & BDRV_O_NOCACHE) != 0) {
         s->needs_alignment = true;
     }
@@ -1345,17 +1346,7 @@ static void raw_aio_unplug(BlockDriverState *bs)
 #ifdef CONFIG_LINUX_AIO
     BDRVRawState *s = bs->opaque;
     if (s->use_aio) {
-        laio_io_unplug(bs, s->aio_ctx, true);
+        laio_io_unplug(bs, s->aio_ctx);
-    }
-#endif
-}
-
-static void raw_aio_flush_io_queue(BlockDriverState *bs)
-{
-#ifdef CONFIG_LINUX_AIO
-    BDRVRawState *s = bs->opaque;
-    if (s->use_aio) {
-        laio_io_unplug(bs, s->aio_ctx, false);
     }
 #endif
 }
@@ -1949,7 +1940,6 @@ BlockDriver bdrv_file = {
     .bdrv_refresh_limits = raw_refresh_limits,
     .bdrv_io_plug = raw_aio_plug,
     .bdrv_io_unplug = raw_aio_unplug,
-    .bdrv_flush_io_queue = raw_aio_flush_io_queue,
 
     .bdrv_truncate = raw_truncate,
     .bdrv_getlength = raw_getlength,
@@ -2398,7 +2388,6 @@ static BlockDriver bdrv_host_device = {
     .bdrv_refresh_limits = raw_refresh_limits,
     .bdrv_io_plug = raw_aio_plug,
     .bdrv_io_unplug = raw_aio_unplug,
-    .bdrv_flush_io_queue = raw_aio_flush_io_queue,
 
     .bdrv_truncate      = raw_truncate,
     .bdrv_getlength	= raw_getlength,
@@ -2528,7 +2517,6 @@ static BlockDriver bdrv_host_cdrom = {
     .bdrv_refresh_limits = raw_refresh_limits,
     .bdrv_io_plug = raw_aio_plug,
     .bdrv_io_unplug = raw_aio_unplug,
-    .bdrv_flush_io_queue = raw_aio_flush_io_queue,
 
     .bdrv_truncate      = raw_truncate,
     .bdrv_getlength      = raw_getlength,
@@ -2664,7 +2652,6 @@ static BlockDriver bdrv_host_cdrom = {
     .bdrv_refresh_limits = raw_refresh_limits,
     .bdrv_io_plug = raw_aio_plug,
     .bdrv_io_unplug = raw_aio_unplug,
-    .bdrv_flush_io_queue = raw_aio_flush_io_queue,
 
     .bdrv_truncate      = raw_truncate,
     .bdrv_getlength      = raw_getlength,

block/raw_bsd.c

@@ -105,8 +105,8 @@ raw_co_writev_flags(BlockDriverState *bs, int64_t sector_num, int nb_sectors,
     }
 
     BLKDBG_EVENT(bs->file, BLKDBG_WRITE_AIO);
-    ret = bdrv_co_do_pwritev(bs->file->bs, sector_num * BDRV_SECTOR_SIZE,
+    ret = bdrv_co_pwritev(bs->file->bs, sector_num * BDRV_SECTOR_SIZE,
-                             nb_sectors * BDRV_SECTOR_SIZE, qiov, flags);
+                          nb_sectors * BDRV_SECTOR_SIZE, qiov, flags);
 
 fail:
     if (qiov == &local_qiov) {
@@ -116,13 +116,6 @@ fail:
     return ret;
 }
 
-static int coroutine_fn
-raw_co_writev(BlockDriverState *bs, int64_t sector_num, int nb_sectors,
-              QEMUIOVector *qiov)
-{
-    return raw_co_writev_flags(bs, sector_num, nb_sectors, qiov, 0);
-}
-
 static int64_t coroutine_fn raw_co_get_block_status(BlockDriverState *bs,
                                             int64_t sector_num,
                                             int nb_sectors, int *pnum,
@@ -211,6 +204,8 @@ static int raw_open(BlockDriverState *bs, QDict *options, int flags,
                     Error **errp)
 {
     bs->sg = bs->file->bs->sg;
+    bs->supported_write_flags = BDRV_REQ_FUA;
+    bs->supported_zero_flags = BDRV_REQ_FUA | BDRV_REQ_MAY_UNMAP;
 
     if (bs->probed && !bdrv_is_read_only(bs)) {
         fprintf(stderr,
@@ -256,9 +251,7 @@ BlockDriver bdrv_raw = {
     .bdrv_close           = &raw_close,
     .bdrv_create          = &raw_create,
     .bdrv_co_readv        = &raw_co_readv,
-    .bdrv_co_writev       = &raw_co_writev,
     .bdrv_co_writev_flags = &raw_co_writev_flags,
-    .supported_write_flags = BDRV_REQ_FUA,
     .bdrv_co_write_zeroes = &raw_co_write_zeroes,
     .bdrv_co_discard      = &raw_co_discard,
     .bdrv_co_get_block_status = &raw_co_get_block_status,

block/sheepdog.c

@@ -294,13 +294,16 @@ static inline size_t count_data_objs(const struct SheepdogInode *inode)
 
 #undef DPRINTF
 #ifdef DEBUG_SDOG
-#define DPRINTF(fmt, args...)                                       \
+#define DEBUG_SDOG_PRINT 1
-    do {                                                            \
-        fprintf(stdout, "%s %d: " fmt, __func__, __LINE__, ##args); \
-    } while (0)
 #else
-#define DPRINTF(fmt, args...)
+#define DEBUG_SDOG_PRINT 0
 #endif
+#define DPRINTF(fmt, args...)                                           \
+    do {                                                                \
+        if (DEBUG_SDOG_PRINT) {                                         \
+            fprintf(stderr, "%s %d: " fmt, __func__, __LINE__, ##args); \
+        }                                                               \
+    } while (0)
 
 typedef struct SheepdogAIOCB SheepdogAIOCB;
 
@@ -1678,7 +1681,7 @@ static int sd_prealloc(const char *filename, Error **errp)
         if (ret < 0) {
             goto out;
         }
-        ret = blk_pwrite(blk, idx * buf_size, buf, buf_size);
+        ret = blk_pwrite(blk, idx * buf_size, buf, buf_size, 0);
         if (ret < 0) {
             goto out;
         }

block/throttle-groups.c

@@ -219,6 +219,10 @@ static bool throttle_group_schedule_timer(BlockDriverState *bs,
     ThrottleGroup *tg = container_of(ts, ThrottleGroup, ts);
     bool must_wait;
 
+    if (bs->io_limits_disabled) {
+        return false;
+    }
+
     /* Check if any of the timers in this group is already armed */
     if (tg->any_timer_armed[is_write]) {
         return true;
@@ -313,6 +317,17 @@ void coroutine_fn throttle_group_co_io_limits_intercept(BlockDriverState *bs,
     qemu_mutex_unlock(&tg->lock);
 }
 
+void throttle_group_restart_bs(BlockDriverState *bs)
+{
+    int i;
+
+    for (i = 0; i < 2; i++) {
+        while (qemu_co_enter_next(&bs->throttled_reqs[i])) {
+            ;
+        }
+    }
+}
+
 /* Update the throttle configuration for a particular group. Similar
  * to throttle_config(), but guarantees atomicity within the
  * throttling group.
@@ -335,6 +350,9 @@ void throttle_group_config(BlockDriverState *bs, ThrottleConfig *cfg)
     }
     throttle_config(ts, tt, cfg);
     qemu_mutex_unlock(&tg->lock);
+
+    qemu_co_enter_next(&bs->throttled_reqs[0]);
+    qemu_co_enter_next(&bs->throttled_reqs[1]);
 }
 
 /* Get the throttle configuration from a particular group. Similar to

block/vdi.c

@@ -557,98 +557,109 @@ static int64_t coroutine_fn vdi_co_get_block_status(BlockDriverState *bs,
     return BDRV_BLOCK_DATA | BDRV_BLOCK_OFFSET_VALID | offset;
 }
 
-static int vdi_co_read(BlockDriverState *bs,
+static int coroutine_fn
-        int64_t sector_num, uint8_t *buf, int nb_sectors)
+vdi_co_preadv(BlockDriverState *bs, uint64_t offset, uint64_t bytes,
+              QEMUIOVector *qiov, int flags)
 {
     BDRVVdiState *s = bs->opaque;
+    QEMUIOVector local_qiov;
     uint32_t bmap_entry;
     uint32_t block_index;
-    uint32_t sector_in_block;
+    uint32_t offset_in_block;
-    uint32_t n_sectors;
+    uint32_t n_bytes;
+    uint64_t bytes_done = 0;
     int ret = 0;
 
     logout("\n");
 
-    while (ret >= 0 && nb_sectors > 0) {
+    qemu_iovec_init(&local_qiov, qiov->niov);
-        block_index = sector_num / s->block_sectors;
-        sector_in_block = sector_num % s->block_sectors;
-        n_sectors = s->block_sectors - sector_in_block;
-        if (n_sectors > nb_sectors) {
-            n_sectors = nb_sectors;
-        }
 
-        logout("will read %u sectors starting at sector %" PRIu64 "\n",
+    while (ret >= 0 && bytes > 0) {
-               n_sectors, sector_num);
+        block_index = offset / s->block_size;
+        offset_in_block = offset % s->block_size;
+        n_bytes = MIN(bytes, s->block_size - offset_in_block);
+
+        logout("will read %u bytes starting at offset %" PRIu64 "\n",
+               n_bytes, offset);
 
         /* prepare next AIO request */
         bmap_entry = le32_to_cpu(s->bmap[block_index]);
         if (!VDI_IS_ALLOCATED(bmap_entry)) {
             /* Block not allocated, return zeros, no need to wait. */
-            memset(buf, 0, n_sectors * SECTOR_SIZE);
+            qemu_iovec_memset(qiov, bytes_done, 0, n_bytes);
             ret = 0;
         } else {
-            uint64_t offset = s->header.offset_data / SECTOR_SIZE +
+            uint64_t data_offset = s->header.offset_data +
-                              (uint64_t)bmap_entry * s->block_sectors +
+                                   (uint64_t)bmap_entry * s->block_size +
-                              sector_in_block;
+                                   offset_in_block;
-            ret = bdrv_read(bs->file->bs, offset, buf, n_sectors);
-        }
-        logout("%u sectors read\n", n_sectors);
 
-        nb_sectors -= n_sectors;
+            qemu_iovec_reset(&local_qiov);
-        sector_num += n_sectors;
+            qemu_iovec_concat(&local_qiov, qiov, bytes_done, n_bytes);
-        buf += n_sectors * SECTOR_SIZE;
+
+            ret = bdrv_co_preadv(bs->file->bs, data_offset, n_bytes,
+                                 &local_qiov, 0);
+        }
+        logout("%u bytes read\n", n_bytes);
+
+        bytes -= n_bytes;
+        offset += n_bytes;
+        bytes_done += n_bytes;
     }
 
+    qemu_iovec_destroy(&local_qiov);
+
     return ret;
 }
 
-static int vdi_co_write(BlockDriverState *bs,
+static int coroutine_fn
-        int64_t sector_num, const uint8_t *buf, int nb_sectors)
+vdi_co_pwritev(BlockDriverState *bs, uint64_t offset, uint64_t bytes,
+               QEMUIOVector *qiov, int flags)
 {
     BDRVVdiState *s = bs->opaque;
+    QEMUIOVector local_qiov;
     uint32_t bmap_entry;
     uint32_t block_index;
-    uint32_t sector_in_block;
+    uint32_t offset_in_block;
-    uint32_t n_sectors;
+    uint32_t n_bytes;
     uint32_t bmap_first = VDI_UNALLOCATED;
     uint32_t bmap_last = VDI_UNALLOCATED;
     uint8_t *block = NULL;
+    uint64_t bytes_done = 0;
     int ret = 0;
 
     logout("\n");
 
-    while (ret >= 0 && nb_sectors > 0) {
+    qemu_iovec_init(&local_qiov, qiov->niov);
-        block_index = sector_num / s->block_sectors;
-        sector_in_block = sector_num % s->block_sectors;
-        n_sectors = s->block_sectors - sector_in_block;
-        if (n_sectors > nb_sectors) {
-            n_sectors = nb_sectors;
-        }
 
-        logout("will write %u sectors starting at sector %" PRIu64 "\n",
+    while (ret >= 0 && bytes > 0) {
-               n_sectors, sector_num);
+        block_index = offset / s->block_size;
+        offset_in_block = offset % s->block_size;
+        n_bytes = MIN(bytes, s->block_size - offset_in_block);
+
+        logout("will write %u bytes starting at offset %" PRIu64 "\n",
+               n_bytes, offset);
 
         /* prepare next AIO request */
         bmap_entry = le32_to_cpu(s->bmap[block_index]);
         if (!VDI_IS_ALLOCATED(bmap_entry)) {
             /* Allocate new block and write to it. */
-            uint64_t offset;
+            uint64_t data_offset;
             bmap_entry = s->header.blocks_allocated;
             s->bmap[block_index] = cpu_to_le32(bmap_entry);
             s->header.blocks_allocated++;
-            offset = s->header.offset_data / SECTOR_SIZE +
+            data_offset = s->header.offset_data +
-                     (uint64_t)bmap_entry * s->block_sectors;
+                          (uint64_t)bmap_entry * s->block_size;
             if (block == NULL) {
                 block = g_malloc(s->block_size);
                 bmap_first = block_index;
             }
             bmap_last = block_index;
             /* Copy data to be written to new block and zero unused parts. */
-            memset(block, 0, sector_in_block * SECTOR_SIZE);
+            memset(block, 0, offset_in_block);
-            memcpy(block + sector_in_block * SECTOR_SIZE,
+            qemu_iovec_to_buf(qiov, bytes_done, block + offset_in_block,
-                   buf, n_sectors * SECTOR_SIZE);
+                              n_bytes);
-            memset(block + (sector_in_block + n_sectors) * SECTOR_SIZE, 0,
+            memset(block + offset_in_block + n_bytes, 0,
-                   (s->block_sectors - n_sectors - sector_in_block) * SECTOR_SIZE);
+                   s->block_size - n_bytes - offset_in_block);
 
             /* Note that this coroutine does not yield anywhere from reading the
              * bmap entry until here, so in regards to all the coroutines trying
@@ -658,12 +669,12 @@ static int vdi_co_write(BlockDriverState *bs,
              * acquire the lock and thus the padded cluster is written before
              * the other coroutines can write to the affected area. */
             qemu_co_mutex_lock(&s->write_lock);
-            ret = bdrv_write(bs->file->bs, offset, block, s->block_sectors);
+            ret = bdrv_pwrite(bs->file->bs, data_offset, block, s->block_size);
             qemu_co_mutex_unlock(&s->write_lock);
         } else {
-            uint64_t offset = s->header.offset_data / SECTOR_SIZE +
+            uint64_t data_offset = s->header.offset_data +
-                              (uint64_t)bmap_entry * s->block_sectors +
+                                   (uint64_t)bmap_entry * s->block_size +
-                              sector_in_block;
+                                   offset_in_block;
             qemu_co_mutex_lock(&s->write_lock);
             /* This lock is only used to make sure the following write operation
              * is executed after the write issued by the coroutine allocating
@@ -674,16 +685,23 @@ static int vdi_co_write(BlockDriverState *bs,
              * that that write operation has returned (there may be other writes
              * in flight, but they do not concern this very operation). */
             qemu_co_mutex_unlock(&s->write_lock);
-            ret = bdrv_write(bs->file->bs, offset, buf, n_sectors);
+
+            qemu_iovec_reset(&local_qiov);
+            qemu_iovec_concat(&local_qiov, qiov, bytes_done, n_bytes);
+
+            ret = bdrv_co_pwritev(bs->file->bs, data_offset, n_bytes,
+                                  &local_qiov, 0);
         }
 
-        nb_sectors -= n_sectors;
+        bytes -= n_bytes;
-        sector_num += n_sectors;
+        offset += n_bytes;
-        buf += n_sectors * SECTOR_SIZE;
+        bytes_done += n_bytes;
 
-        logout("%u sectors written\n", n_sectors);
+        logout("%u bytes written\n", n_bytes);
     }
 
+    qemu_iovec_destroy(&local_qiov);
+
     logout("finished data write\n");
     if (ret < 0) {
         return ret;
@@ -694,6 +712,7 @@ static int vdi_co_write(BlockDriverState *bs,
         VdiHeader *header = (VdiHeader *) block;
         uint8_t *base;
         uint64_t offset;
+        uint32_t n_sectors;
 
         logout("now writing modified header\n");
         assert(VDI_IS_ALLOCATED(bmap_first));
@@ -808,7 +827,7 @@ static int vdi_create(const char *filename, QemuOpts *opts, Error **errp)
     vdi_header_print(&header);
 #endif
     vdi_header_to_le(&header);
-    ret = blk_pwrite(blk, offset, &header, sizeof(header));
+    ret = blk_pwrite(blk, offset, &header, sizeof(header), 0);
     if (ret < 0) {
         error_setg(errp, "Error writing header to %s", filename);
         goto exit;
@@ -829,7 +848,7 @@ static int vdi_create(const char *filename, QemuOpts *opts, Error **errp)
                 bmap[i] = VDI_UNALLOCATED;
             }
         }
-        ret = blk_pwrite(blk, offset, bmap, bmap_size);
+        ret = blk_pwrite(blk, offset, bmap, bmap_size, 0);
         if (ret < 0) {
             error_setg(errp, "Error writing bmap to %s", filename);
             goto exit;
@@ -903,9 +922,9 @@ static BlockDriver bdrv_vdi = {
     .bdrv_co_get_block_status = vdi_co_get_block_status,
     .bdrv_make_empty = vdi_make_empty,
 
-    .bdrv_read = vdi_co_read,
+    .bdrv_co_preadv     = vdi_co_preadv,
 #if defined(CONFIG_VDI_WRITE)
-    .bdrv_write = vdi_co_write,
+    .bdrv_co_pwritev    = vdi_co_pwritev,
 #endif
 
     .bdrv_get_info = vdi_get_info,

block/vhdx.c

@@ -1856,13 +1856,14 @@ static int vhdx_create(const char *filename, QemuOpts *opts, Error **errp)
     creator = g_utf8_to_utf16("QEMU v" QEMU_VERSION, -1, NULL,
                               &creator_items, NULL);
     signature = cpu_to_le64(VHDX_FILE_SIGNATURE);
-    ret = blk_pwrite(blk, VHDX_FILE_ID_OFFSET, &signature, sizeof(signature));
+    ret = blk_pwrite(blk, VHDX_FILE_ID_OFFSET, &signature, sizeof(signature),
+                     0);
     if (ret < 0) {
         goto delete_and_exit;
     }
     if (creator) {
         ret = blk_pwrite(blk, VHDX_FILE_ID_OFFSET + sizeof(signature),
-                         creator, creator_items * sizeof(gunichar2));
+                         creator, creator_items * sizeof(gunichar2), 0);
         if (ret < 0) {
             goto delete_and_exit;
         }

block/vmdk.c

@@ -1016,27 +1016,26 @@ static void vmdk_refresh_limits(BlockDriverState *bs, Error **errp)
  */
 static int get_whole_cluster(BlockDriverState *bs,
                              VmdkExtent *extent,
-                             uint64_t cluster_sector_num,
+                             uint64_t cluster_offset,
-                             uint64_t sector_num,
+                             uint64_t offset,
-                             uint64_t skip_start_sector,
+                             uint64_t skip_start_bytes,
-                             uint64_t skip_end_sector)
+                             uint64_t skip_end_bytes)
 {
     int ret = VMDK_OK;
     int64_t cluster_bytes;
     uint8_t *whole_grain;
 
     /* For COW, align request sector_num to cluster start */
-    sector_num = QEMU_ALIGN_DOWN(sector_num, extent->cluster_sectors);
     cluster_bytes = extent->cluster_sectors << BDRV_SECTOR_BITS;
+    offset = QEMU_ALIGN_DOWN(offset, cluster_bytes);
     whole_grain = qemu_blockalign(bs, cluster_bytes);
 
     if (!bs->backing) {
-        memset(whole_grain, 0,  skip_start_sector << BDRV_SECTOR_BITS);
+        memset(whole_grain, 0, skip_start_bytes);
-        memset(whole_grain + (skip_end_sector << BDRV_SECTOR_BITS), 0,
+        memset(whole_grain + skip_end_bytes, 0, cluster_bytes - skip_end_bytes);
-               cluster_bytes - (skip_end_sector << BDRV_SECTOR_BITS));
     }
 
-    assert(skip_end_sector <= extent->cluster_sectors);
+    assert(skip_end_bytes <= cluster_bytes);
     /* we will be here if it's first write on non-exist grain(cluster).
      * try to read from parent image, if exist */
     if (bs->backing && !vmdk_is_cid_valid(bs)) {
@@ -1045,42 +1044,43 @@ static int get_whole_cluster(BlockDriverState *bs,
     }
 
     /* Read backing data before skip range */
-    if (skip_start_sector > 0) {
+    if (skip_start_bytes > 0) {
         if (bs->backing) {
-            ret = bdrv_read(bs->backing->bs, sector_num,
+            ret = bdrv_pread(bs->backing->bs, offset, whole_grain,
-                            whole_grain, skip_start_sector);
+                             skip_start_bytes);
             if (ret < 0) {
                 ret = VMDK_ERROR;
                 goto exit;
             }
         }
-        ret = bdrv_write(extent->file->bs, cluster_sector_num, whole_grain,
+        ret = bdrv_pwrite(extent->file->bs, cluster_offset, whole_grain,
-                         skip_start_sector);
+                          skip_start_bytes);
         if (ret < 0) {
             ret = VMDK_ERROR;
             goto exit;
         }
     }
     /* Read backing data after skip range */
-    if (skip_end_sector < extent->cluster_sectors) {
+    if (skip_end_bytes < cluster_bytes) {
         if (bs->backing) {
-            ret = bdrv_read(bs->backing->bs, sector_num + skip_end_sector,
+            ret = bdrv_pread(bs->backing->bs, offset + skip_end_bytes,
-                            whole_grain + (skip_end_sector << BDRV_SECTOR_BITS),
+                             whole_grain + skip_end_bytes,
-                            extent->cluster_sectors - skip_end_sector);
+                             cluster_bytes - skip_end_bytes);
             if (ret < 0) {
                 ret = VMDK_ERROR;
                 goto exit;
             }
         }
-        ret = bdrv_write(extent->file->bs, cluster_sector_num + skip_end_sector,
+        ret = bdrv_pwrite(extent->file->bs, cluster_offset + skip_end_bytes,
-                         whole_grain + (skip_end_sector << BDRV_SECTOR_BITS),
+                          whole_grain + skip_end_bytes,
-                         extent->cluster_sectors - skip_end_sector);
+                          cluster_bytes - skip_end_bytes);
         if (ret < 0) {
             ret = VMDK_ERROR;
             goto exit;
         }
     }
 
+    ret = VMDK_OK;
 exit:
     qemu_vfree(whole_grain);
     return ret;
@@ -1142,8 +1142,8 @@ static int get_cluster_offset(BlockDriverState *bs,
                               uint64_t offset,
                               bool allocate,
                               uint64_t *cluster_offset,
-                              uint64_t skip_start_sector,
+                              uint64_t skip_start_bytes,
-                              uint64_t skip_end_sector)
+                              uint64_t skip_end_bytes)
 {
     unsigned int l1_index, l2_offset, l2_index;
     int min_index, i, j;
@@ -1230,10 +1230,8 @@ static int get_cluster_offset(BlockDriverState *bs,
          * This problem may occur because of insufficient space on host disk
          * or inappropriate VM shutdown.
          */
-        ret = get_whole_cluster(bs, extent,
+        ret = get_whole_cluster(bs, extent, cluster_sector * BDRV_SECTOR_SIZE,
-                                cluster_sector,
+                                offset, skip_start_bytes, skip_end_bytes);
-                                offset >> BDRV_SECTOR_BITS,
-                                skip_start_sector, skip_end_sector);
         if (ret) {
             return ret;
         }
@@ -1259,15 +1257,26 @@ static VmdkExtent *find_extent(BDRVVmdkState *s,
     return NULL;
 }
 
+static inline uint64_t vmdk_find_offset_in_cluster(VmdkExtent *extent,
+                                                   int64_t offset)
+{
+    uint64_t offset_in_cluster, extent_begin_offset, extent_relative_offset;
+    uint64_t cluster_size = extent->cluster_sectors * BDRV_SECTOR_SIZE;
+
+    extent_begin_offset =
+        (extent->end_sector - extent->sectors) * BDRV_SECTOR_SIZE;
+    extent_relative_offset = offset - extent_begin_offset;
+    offset_in_cluster = extent_relative_offset % cluster_size;
+
+    return offset_in_cluster;
+}
+
 static inline uint64_t vmdk_find_index_in_cluster(VmdkExtent *extent,
                                                   int64_t sector_num)
 {
-    uint64_t index_in_cluster, extent_begin_sector, extent_relative_sector_num;
+    uint64_t offset;
-
+    offset = vmdk_find_offset_in_cluster(extent, sector_num * BDRV_SECTOR_SIZE);
-    extent_begin_sector = extent->end_sector - extent->sectors;
+    return offset / BDRV_SECTOR_SIZE;
-    extent_relative_sector_num = sector_num - extent_begin_sector;
-    index_in_cluster = extent_relative_sector_num % extent->cluster_sectors;
-    return index_in_cluster;
 }
 
 static int64_t coroutine_fn vmdk_co_get_block_status(BlockDriverState *bs,
@@ -1319,38 +1328,57 @@ static int64_t coroutine_fn vmdk_co_get_block_status(BlockDriverState *bs,
 }
 
 static int vmdk_write_extent(VmdkExtent *extent, int64_t cluster_offset,
-                            int64_t offset_in_cluster, const uint8_t *buf,
+                            int64_t offset_in_cluster, QEMUIOVector *qiov,
-                            int nb_sectors, int64_t sector_num)
+                            uint64_t qiov_offset, uint64_t n_bytes,
+                            uint64_t offset)
 {
     int ret;
     VmdkGrainMarker *data = NULL;
     uLongf buf_len;
-    const uint8_t *write_buf = buf;
+    QEMUIOVector local_qiov;
-    int write_len = nb_sectors * 512;
+    struct iovec iov;
     int64_t write_offset;
     int64_t write_end_sector;
 
     if (extent->compressed) {
+        void *compressed_data;
+
         if (!extent->has_marker) {
             ret = -EINVAL;
             goto out;
         }
         buf_len = (extent->cluster_sectors << 9) * 2;
         data = g_malloc(buf_len + sizeof(VmdkGrainMarker));
-        if (compress(data->data, &buf_len, buf, nb_sectors << 9) != Z_OK ||
+
-                buf_len == 0) {
+        compressed_data = g_malloc(n_bytes);
+        qemu_iovec_to_buf(qiov, qiov_offset, compressed_data, n_bytes);
+        ret = compress(data->data, &buf_len, compressed_data, n_bytes);
+        g_free(compressed_data);
+
+        if (ret != Z_OK || buf_len == 0) {
             ret = -EINVAL;
             goto out;
         }
-        data->lba = sector_num;
-        data->size = buf_len;
-        write_buf = (uint8_t *)data;
-        write_len = buf_len + sizeof(VmdkGrainMarker);
-    }
-    write_offset = cluster_offset + offset_in_cluster,
-    ret = bdrv_pwrite(extent->file->bs, write_offset, write_buf, write_len);
 
-    write_end_sector = DIV_ROUND_UP(write_offset + write_len, BDRV_SECTOR_SIZE);
+        data->lba = offset >> BDRV_SECTOR_BITS;
+        data->size = buf_len;
+
+        n_bytes = buf_len + sizeof(VmdkGrainMarker);
+        iov = (struct iovec) {
+            .iov_base   = data,
+            .iov_len    = n_bytes,
+        };
+        qemu_iovec_init_external(&local_qiov, &iov, 1);
+    } else {
+        qemu_iovec_init(&local_qiov, qiov->niov);
+        qemu_iovec_concat(&local_qiov, qiov, qiov_offset, n_bytes);
+    }
+
+    write_offset = cluster_offset + offset_in_cluster,
+    ret = bdrv_co_pwritev(extent->file->bs, write_offset, n_bytes,
+                          &local_qiov, 0);
+
+    write_end_sector = DIV_ROUND_UP(write_offset + n_bytes, BDRV_SECTOR_SIZE);
 
     if (extent->compressed) {
         extent->next_cluster_sector = write_end_sector;
@@ -1359,19 +1387,21 @@ static int vmdk_write_extent(VmdkExtent *extent, int64_t cluster_offset,
                                           write_end_sector);
     }
 
-    if (ret != write_len) {
+    if (ret < 0) {
-        ret = ret < 0 ? ret : -EIO;
         goto out;
     }
     ret = 0;
  out:
     g_free(data);
+    if (!extent->compressed) {
+        qemu_iovec_destroy(&local_qiov);
+    }
     return ret;
 }
 
 static int vmdk_read_extent(VmdkExtent *extent, int64_t cluster_offset,
-                            int64_t offset_in_cluster, uint8_t *buf,
+                            int64_t offset_in_cluster, QEMUIOVector *qiov,
-                            int nb_sectors)
+                            int bytes)
 {
     int ret;
     int cluster_bytes, buf_bytes;
@@ -1383,14 +1413,13 @@ static int vmdk_read_extent(VmdkExtent *extent, int64_t cluster_offset,
 
 
     if (!extent->compressed) {
-        ret = bdrv_pread(extent->file->bs,
+        ret = bdrv_co_preadv(extent->file->bs,
-                          cluster_offset + offset_in_cluster,
+                             cluster_offset + offset_in_cluster, bytes,
-                          buf, nb_sectors * 512);
+                             qiov, 0);
-        if (ret == nb_sectors * 512) {
+        if (ret < 0) {
-            return 0;
+            return ret;
-        } else {
-            return -EIO;
         }
+        return 0;
     }
     cluster_bytes = extent->cluster_sectors * 512;
     /* Read two clusters in case GrainMarker + compressed data > one cluster */
@@ -1422,11 +1451,11 @@ static int vmdk_read_extent(VmdkExtent *extent, int64_t cluster_offset,
 
     }
     if (offset_in_cluster < 0 ||
-            offset_in_cluster + nb_sectors * 512 > buf_len) {
+            offset_in_cluster + bytes > buf_len) {
         ret = -EINVAL;
         goto out;
     }
-    memcpy(buf, uncomp_buf + offset_in_cluster, nb_sectors * 512);
+    qemu_iovec_from_buf(qiov, 0, uncomp_buf + offset_in_cluster, bytes);
     ret = 0;
 
  out:
@@ -1435,64 +1464,73 @@ static int vmdk_read_extent(VmdkExtent *extent, int64_t cluster_offset,
     return ret;
 }
 
-static int vmdk_read(BlockDriverState *bs, int64_t sector_num,
+static int coroutine_fn
-                    uint8_t *buf, int nb_sectors)
+vmdk_co_preadv(BlockDriverState *bs, uint64_t offset, uint64_t bytes,
+               QEMUIOVector *qiov, int flags)
 {
     BDRVVmdkState *s = bs->opaque;
     int ret;
-    uint64_t n, index_in_cluster;
+    uint64_t n_bytes, offset_in_cluster;
     VmdkExtent *extent = NULL;
+    QEMUIOVector local_qiov;
     uint64_t cluster_offset;
+    uint64_t bytes_done = 0;
 
-    while (nb_sectors > 0) {
+    qemu_iovec_init(&local_qiov, qiov->niov);
-        extent = find_extent(s, sector_num, extent);
+    qemu_co_mutex_lock(&s->lock);
+
+    while (bytes > 0) {
+        extent = find_extent(s, offset >> BDRV_SECTOR_BITS, extent);
         if (!extent) {
-            return -EIO;
+            ret = -EIO;
+            goto fail;
         }
         ret = get_cluster_offset(bs, extent, NULL,
-                                 sector_num << 9, false, &cluster_offset,
+                                 offset, false, &cluster_offset, 0, 0);
-                                 0, 0);
+        offset_in_cluster = vmdk_find_offset_in_cluster(extent, offset);
-        index_in_cluster = vmdk_find_index_in_cluster(extent, sector_num);
+
-        n = extent->cluster_sectors - index_in_cluster;
+        n_bytes = MIN(bytes, extent->cluster_sectors * BDRV_SECTOR_SIZE
-        if (n > nb_sectors) {
+                             - offset_in_cluster);
-            n = nb_sectors;
+
-        }
         if (ret != VMDK_OK) {
             /* if not allocated, try to read from parent image, if exist */
             if (bs->backing && ret != VMDK_ZEROED) {
                 if (!vmdk_is_cid_valid(bs)) {
-                    return -EINVAL;
+                    ret = -EINVAL;
+                    goto fail;
                 }
-                ret = bdrv_read(bs->backing->bs, sector_num, buf, n);
+
+                qemu_iovec_reset(&local_qiov);
+                qemu_iovec_concat(&local_qiov, qiov, bytes_done, n_bytes);
+
+                ret = bdrv_co_preadv(bs->backing->bs, offset, n_bytes,
+                                     &local_qiov, 0);
                 if (ret < 0) {
-                    return ret;
+                    goto fail;
                 }
             } else {
-                memset(buf, 0, 512 * n);
+                qemu_iovec_memset(qiov, bytes_done, 0, n_bytes);
             }
         } else {
-            ret = vmdk_read_extent(extent,
+            qemu_iovec_reset(&local_qiov);
-                            cluster_offset, index_in_cluster * 512,
+            qemu_iovec_concat(&local_qiov, qiov, bytes_done, n_bytes);
-                            buf, n);
+
+            ret = vmdk_read_extent(extent, cluster_offset, offset_in_cluster,
+                                   &local_qiov, n_bytes);
             if (ret) {
-                return ret;
+                goto fail;
             }
         }
-        nb_sectors -= n;
+        bytes -= n_bytes;
-        sector_num += n;
+        offset += n_bytes;
-        buf += n * 512;
+        bytes_done += n_bytes;
     }
-    return 0;
-}
 
-static coroutine_fn int vmdk_co_read(BlockDriverState *bs, int64_t sector_num,
+    ret = 0;
-                                     uint8_t *buf, int nb_sectors)
+fail:
-{
-    int ret;
-    BDRVVmdkState *s = bs->opaque;
-    qemu_co_mutex_lock(&s->lock);
-    ret = vmdk_read(bs, sector_num, buf, nb_sectors);
     qemu_co_mutex_unlock(&s->lock);
+    qemu_iovec_destroy(&local_qiov);
+
     return ret;
 }
 
@@ -1506,38 +1544,38 @@ static coroutine_fn int vmdk_co_read(BlockDriverState *bs, int64_t sector_num,
  *
  * Returns: error code with 0 for success.
  */
-static int vmdk_write(BlockDriverState *bs, int64_t sector_num,
+static int vmdk_pwritev(BlockDriverState *bs, uint64_t offset,
-                      const uint8_t *buf, int nb_sectors,
+                       uint64_t bytes, QEMUIOVector *qiov,
-                      bool zeroed, bool zero_dry_run)
+                       bool zeroed, bool zero_dry_run)
 {
     BDRVVmdkState *s = bs->opaque;
     VmdkExtent *extent = NULL;
     int ret;
-    int64_t index_in_cluster, n;
+    int64_t offset_in_cluster, n_bytes;
     uint64_t cluster_offset;
+    uint64_t bytes_done = 0;
     VmdkMetaData m_data;
 
-    if (sector_num > bs->total_sectors) {
+    if (DIV_ROUND_UP(offset, BDRV_SECTOR_SIZE) > bs->total_sectors) {
-        error_report("Wrong offset: sector_num=0x%" PRIx64
+        error_report("Wrong offset: offset=0x%" PRIx64
                      " total_sectors=0x%" PRIx64,
-                     sector_num, bs->total_sectors);
+                     offset, bs->total_sectors);
         return -EIO;
     }
 
-    while (nb_sectors > 0) {
+    while (bytes > 0) {
-        extent = find_extent(s, sector_num, extent);
+        extent = find_extent(s, offset >> BDRV_SECTOR_BITS, extent);
         if (!extent) {
             return -EIO;
         }
-        index_in_cluster = vmdk_find_index_in_cluster(extent, sector_num);
+        offset_in_cluster = vmdk_find_offset_in_cluster(extent, offset);
-        n = extent->cluster_sectors - index_in_cluster;
+        n_bytes = MIN(bytes, extent->cluster_sectors * BDRV_SECTOR_SIZE
-        if (n > nb_sectors) {
+                             - offset_in_cluster);
-            n = nb_sectors;
+
-        }
+        ret = get_cluster_offset(bs, extent, &m_data, offset,
-        ret = get_cluster_offset(bs, extent, &m_data, sector_num << 9,
                                  !(extent->compressed || zeroed),
-                                 &cluster_offset,
+                                 &cluster_offset, offset_in_cluster,
-                                 index_in_cluster, index_in_cluster + n);
+                                 offset_in_cluster + n_bytes);
         if (extent->compressed) {
             if (ret == VMDK_OK) {
                 /* Refuse write to allocated cluster for streamOptimized */
@@ -1546,7 +1584,7 @@ static int vmdk_write(BlockDriverState *bs, int64_t sector_num,
                 return -EIO;
             } else {
                 /* allocate */
-                ret = get_cluster_offset(bs, extent, &m_data, sector_num << 9,
+                ret = get_cluster_offset(bs, extent, &m_data, offset,
                                          true, &cluster_offset, 0, 0);
             }
         }
@@ -1556,9 +1594,9 @@ static int vmdk_write(BlockDriverState *bs, int64_t sector_num,
         if (zeroed) {
             /* Do zeroed write, buf is ignored */
             if (extent->has_zero_grain &&
-                    index_in_cluster == 0 &&
+                    offset_in_cluster == 0 &&
-                    n >= extent->cluster_sectors) {
+                    n_bytes >= extent->cluster_sectors * BDRV_SECTOR_SIZE) {
-                n = extent->cluster_sectors;
+                n_bytes = extent->cluster_sectors * BDRV_SECTOR_SIZE;
                 if (!zero_dry_run) {
                     /* update L2 tables */
                     if (vmdk_L2update(extent, &m_data, VMDK_GTE_ZEROED)
@@ -1570,9 +1608,8 @@ static int vmdk_write(BlockDriverState *bs, int64_t sector_num,
                 return -ENOTSUP;
             }
         } else {
-            ret = vmdk_write_extent(extent,
+            ret = vmdk_write_extent(extent, cluster_offset, offset_in_cluster,
-                            cluster_offset, index_in_cluster * 512,
+                                    qiov, bytes_done, n_bytes, offset);
-                            buf, n, sector_num);
             if (ret) {
                 return ret;
             }
@@ -1585,9 +1622,9 @@ static int vmdk_write(BlockDriverState *bs, int64_t sector_num,
                 }
             }
         }
-        nb_sectors -= n;
+        bytes -= n_bytes;
-        sector_num += n;
+        offset += n_bytes;
-        buf += n * 512;
+        bytes_done += n_bytes;
 
         /* update CID on the first write every time the virtual disk is
          * opened */
@@ -1602,25 +1639,65 @@ static int vmdk_write(BlockDriverState *bs, int64_t sector_num,
     return 0;
 }
 
-static coroutine_fn int vmdk_co_write(BlockDriverState *bs, int64_t sector_num,
+static int coroutine_fn
-                                      const uint8_t *buf, int nb_sectors)
+vmdk_co_pwritev(BlockDriverState *bs, uint64_t offset, uint64_t bytes,
+                QEMUIOVector *qiov, int flags)
 {
     int ret;
     BDRVVmdkState *s = bs->opaque;
     qemu_co_mutex_lock(&s->lock);
-    ret = vmdk_write(bs, sector_num, buf, nb_sectors, false, false);
+    ret = vmdk_pwritev(bs, offset, bytes, qiov, false, false);
     qemu_co_mutex_unlock(&s->lock);
     return ret;
 }
 
+typedef struct VmdkWriteCompressedCo {
+    BlockDriverState *bs;
+    int64_t sector_num;
+    const uint8_t *buf;
+    int nb_sectors;
+    int ret;
+} VmdkWriteCompressedCo;
+
+static void vmdk_co_write_compressed(void *opaque)
+{
+    VmdkWriteCompressedCo *co = opaque;
+    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) {
-        return vmdk_write(bs, sector_num, buf, nb_sectors, false, false);
+        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);
+        qemu_coroutine_enter(co, &data);
+        while (data.ret == -EINPROGRESS) {
+            aio_poll(aio_context, true);
+        }
+        return data.ret;
     } else {
         return -ENOTSUP;
     }
@@ -1633,12 +1710,15 @@ static int coroutine_fn vmdk_co_write_zeroes(BlockDriverState *bs,
 {
     int ret;
     BDRVVmdkState *s = bs->opaque;
+    uint64_t offset = sector_num * BDRV_SECTOR_SIZE;
+    uint64_t bytes = nb_sectors * BDRV_SECTOR_SIZE;
+
     qemu_co_mutex_lock(&s->lock);
     /* write zeroes could fail if sectors not aligned to cluster, test it with
      * dry_run == true before really updating image */
-    ret = vmdk_write(bs, sector_num, NULL, nb_sectors, true, true);
+    ret = vmdk_pwritev(bs, offset, bytes, NULL, true, true);
     if (!ret) {
-        ret = vmdk_write(bs, sector_num, NULL, nb_sectors, true, false);
+        ret = vmdk_pwritev(bs, offset, bytes, NULL, true, false);
     }
     qemu_co_mutex_unlock(&s->lock);
     return ret;
@@ -1728,12 +1808,12 @@ static int vmdk_create_extent(const char *filename, int64_t filesize,
     header.check_bytes[3] = 0xa;
 
     /* write all the data */
-    ret = blk_pwrite(blk, 0, &magic, sizeof(magic));
+    ret = blk_pwrite(blk, 0, &magic, sizeof(magic), 0);
     if (ret < 0) {
         error_setg(errp, QERR_IO_ERROR);
         goto exit;
     }
-    ret = blk_pwrite(blk, sizeof(magic), &header, sizeof(header));
+    ret = blk_pwrite(blk, sizeof(magic), &header, sizeof(header), 0);
     if (ret < 0) {
         error_setg(errp, QERR_IO_ERROR);
         goto exit;
@@ -1753,7 +1833,7 @@ static int vmdk_create_extent(const char *filename, int64_t filesize,
         gd_buf[i] = cpu_to_le32(tmp);
     }
     ret = blk_pwrite(blk, le64_to_cpu(header.rgd_offset) * BDRV_SECTOR_SIZE,
-                     gd_buf, gd_buf_size);
+                     gd_buf, gd_buf_size, 0);
     if (ret < 0) {
         error_setg(errp, QERR_IO_ERROR);
         goto exit;
@@ -1765,7 +1845,7 @@ static int vmdk_create_extent(const char *filename, int64_t filesize,
         gd_buf[i] = cpu_to_le32(tmp);
     }
     ret = blk_pwrite(blk, le64_to_cpu(header.gd_offset) * BDRV_SECTOR_SIZE,
-                     gd_buf, gd_buf_size);
+                     gd_buf, gd_buf_size, 0);
     if (ret < 0) {
         error_setg(errp, QERR_IO_ERROR);
         goto exit;
@@ -1829,8 +1909,8 @@ static int vmdk_create(const char *filename, QemuOpts *opts, Error **errp)
     int64_t total_size = 0, filesize;
     char *adapter_type = NULL;
     char *backing_file = NULL;
+    char *hw_version = NULL;
     char *fmt = NULL;
-    int flags = 0;
     int ret = 0;
     bool flat, split, compress;
     GString *ext_desc_lines;
@@ -1861,7 +1941,7 @@ static int vmdk_create(const char *filename, QemuOpts *opts, Error **errp)
         "# The Disk Data Base\n"
         "#DDB\n"
         "\n"
-        "ddb.virtualHWVersion = \"%d\"\n"
+        "ddb.virtualHWVersion = \"%s\"\n"
         "ddb.geometry.cylinders = \"%" PRId64 "\"\n"
         "ddb.geometry.heads = \"%" PRIu32 "\"\n"
         "ddb.geometry.sectors = \"63\"\n"
@@ -1878,8 +1958,20 @@ static int vmdk_create(const char *filename, QemuOpts *opts, Error **errp)
                           BDRV_SECTOR_SIZE);
     adapter_type = qemu_opt_get_del(opts, BLOCK_OPT_ADAPTER_TYPE);
     backing_file = qemu_opt_get_del(opts, BLOCK_OPT_BACKING_FILE);
+    hw_version = qemu_opt_get_del(opts, BLOCK_OPT_HWVERSION);
     if (qemu_opt_get_bool_del(opts, BLOCK_OPT_COMPAT6, false)) {
-        flags |= BLOCK_FLAG_COMPAT6;
+        if (strcmp(hw_version, "undefined")) {
+            error_setg(errp,
+                       "compat6 cannot be enabled with hwversion set");
+            ret = -EINVAL;
+            goto exit;
+        }
+        g_free(hw_version);
+        hw_version = g_strdup("6");
+    }
+    if (strcmp(hw_version, "undefined") == 0) {
+        g_free(hw_version);
+        hw_version = g_strdup("4");
     }
     fmt = qemu_opt_get_del(opts, BLOCK_OPT_SUBFMT);
     if (qemu_opt_get_bool_del(opts, BLOCK_OPT_ZEROED_GRAIN, false)) {
@@ -2001,7 +2093,7 @@ static int vmdk_create(const char *filename, QemuOpts *opts, Error **errp)
                            fmt,
                            parent_desc_line,
                            ext_desc_lines->str,
-                           (flags & BLOCK_FLAG_COMPAT6 ? 6 : 4),
+                           hw_version,
                            total_size /
                                (int64_t)(63 * number_heads * BDRV_SECTOR_SIZE),
                            number_heads,
@@ -2028,7 +2120,7 @@ static int vmdk_create(const char *filename, QemuOpts *opts, Error **errp)
 
     blk_set_allow_write_beyond_eof(new_blk, true);
 
-    ret = blk_pwrite(new_blk, desc_offset, desc, desc_len);
+    ret = blk_pwrite(new_blk, desc_offset, desc, desc_len, 0);
     if (ret < 0) {
         error_setg_errno(errp, -ret, "Could not write description");
         goto exit;
@@ -2047,6 +2139,7 @@ exit:
     }
     g_free(adapter_type);
     g_free(backing_file);
+    g_free(hw_version);
     g_free(fmt);
     g_free(desc);
     g_free(path);
@@ -2297,6 +2390,12 @@ static QemuOptsList vmdk_create_opts = {
             .help = "VMDK version 6 image",
             .def_value_str = "off"
         },
+        {
+            .name = BLOCK_OPT_HWVERSION,
+            .type = QEMU_OPT_STRING,
+            .help = "VMDK hardware version",
+            .def_value_str = "undefined"
+        },
         {
             .name = BLOCK_OPT_SUBFMT,
             .type = QEMU_OPT_STRING,
@@ -2321,8 +2420,8 @@ static BlockDriver bdrv_vmdk = {
     .bdrv_open                    = vmdk_open,
     .bdrv_check                   = vmdk_check,
     .bdrv_reopen_prepare          = vmdk_reopen_prepare,
-    .bdrv_read                    = vmdk_co_read,
+    .bdrv_co_preadv               = vmdk_co_preadv,
-    .bdrv_write                   = vmdk_co_write,
+    .bdrv_co_pwritev              = vmdk_co_pwritev,
     .bdrv_write_compressed        = vmdk_write_compressed,
     .bdrv_co_write_zeroes         = vmdk_co_write_zeroes,
     .bdrv_close                   = vmdk_close,

block/vpc.c

@@ -454,22 +454,21 @@ static int vpc_reopen_prepare(BDRVReopenState *state,
  * The parameter write must be 1 if the offset will be used for a write
  * operation (the block bitmaps is updated then), 0 otherwise.
  */
-static inline int64_t get_sector_offset(BlockDriverState *bs,
+static inline int64_t get_image_offset(BlockDriverState *bs, uint64_t offset,
-    int64_t sector_num, int write)
+                                       bool write)
 {
     BDRVVPCState *s = bs->opaque;
-    uint64_t offset = sector_num * 512;
     uint64_t bitmap_offset, block_offset;
-    uint32_t pagetable_index, pageentry_index;
+    uint32_t pagetable_index, offset_in_block;
 
     pagetable_index = offset / s->block_size;
-    pageentry_index = (offset % s->block_size) / 512;
+    offset_in_block = offset % s->block_size;
 
     if (pagetable_index >= s->max_table_entries || s->pagetable[pagetable_index] == 0xffffffff)
         return -1; /* not allocated */
 
     bitmap_offset = 512 * (uint64_t) s->pagetable[pagetable_index];
-    block_offset = bitmap_offset + s->bitmap_size + (512 * pageentry_index);
+    block_offset = bitmap_offset + s->bitmap_size + offset_in_block;
 
     /* We must ensure that we don't write to any sectors which are marked as
        unused in the bitmap. We get away with setting all bits in the block
@@ -487,6 +486,12 @@ static inline int64_t get_sector_offset(BlockDriverState *bs,
     return block_offset;
 }
 
+static inline int64_t get_sector_offset(BlockDriverState *bs,
+                                        int64_t sector_num, bool write)
+{
+    return get_image_offset(bs, sector_num * BDRV_SECTOR_SIZE, write);
+}
+
 /*
  * Writes the footer to the end of the image file. This is needed when the
  * file grows as it overwrites the old footer
@@ -513,7 +518,7 @@ static int rewrite_footer(BlockDriverState* bs)
  *
  * Returns the sectors' offset in the image file on success and < 0 on error
  */
-static int64_t alloc_block(BlockDriverState* bs, int64_t sector_num)
+static int64_t alloc_block(BlockDriverState* bs, int64_t offset)
 {
     BDRVVPCState *s = bs->opaque;
     int64_t bat_offset;
@@ -522,14 +527,13 @@ static int64_t alloc_block(BlockDriverState* bs, int64_t sector_num)
     uint8_t bitmap[s->bitmap_size];
 
     /* Check if sector_num is valid */
-    if ((sector_num < 0) || (sector_num > bs->total_sectors))
+    if ((offset < 0) || (offset > bs->total_sectors * BDRV_SECTOR_SIZE)) {
-        return -1;
+        return -EINVAL;
+    }
 
     /* Write entry into in-memory BAT */
-    index = (sector_num * 512) / s->block_size;
+    index = offset / s->block_size;
-    if (s->pagetable[index] != 0xFFFFFFFF)
+    assert(s->pagetable[index] == 0xFFFFFFFF);
-        return -1;
-
     s->pagetable[index] = s->free_data_block_offset / 512;
 
     /* Initialize the block's bitmap */
@@ -553,11 +557,11 @@ static int64_t alloc_block(BlockDriverState* bs, int64_t sector_num)
     if (ret < 0)
         goto fail;
 
-    return get_sector_offset(bs, sector_num, 0);
+    return get_image_offset(bs, offset, false);
 
 fail:
     s->free_data_block_offset -= (s->block_size + s->bitmap_size);
-    return -1;
+    return ret;
 }
 
 static int vpc_get_info(BlockDriverState *bs, BlockDriverInfo *bdi)
@@ -573,104 +577,105 @@ static int vpc_get_info(BlockDriverState *bs, BlockDriverInfo *bdi)
     return 0;
 }
 
-static int vpc_read(BlockDriverState *bs, int64_t sector_num,
+static int coroutine_fn
-                    uint8_t *buf, int nb_sectors)
+vpc_co_preadv(BlockDriverState *bs, uint64_t offset, uint64_t bytes,
+              QEMUIOVector *qiov, int flags)
 {
     BDRVVPCState *s = bs->opaque;
     int ret;
-    int64_t offset;
+    int64_t image_offset;
-    int64_t sectors, sectors_per_block;
+    int64_t n_bytes;
+    int64_t bytes_done = 0;
     VHDFooter *footer = (VHDFooter *) s->footer_buf;
+    QEMUIOVector local_qiov;
 
     if (be32_to_cpu(footer->type) == VHD_FIXED) {
-        return bdrv_read(bs->file->bs, sector_num, buf, nb_sectors);
+        return bdrv_co_preadv(bs->file->bs, offset, bytes, qiov, 0);
     }
-    while (nb_sectors > 0) {
-        offset = get_sector_offset(bs, sector_num, 0);
 
-        sectors_per_block = s->block_size >> BDRV_SECTOR_BITS;
+    qemu_co_mutex_lock(&s->lock);
-        sectors = sectors_per_block - (sector_num % sectors_per_block);
+    qemu_iovec_init(&local_qiov, qiov->niov);
-        if (sectors > nb_sectors) {
-            sectors = nb_sectors;
-        }
 
-        if (offset == -1) {
+    while (bytes > 0) {
-            memset(buf, 0, sectors * BDRV_SECTOR_SIZE);
+        image_offset = get_image_offset(bs, offset, false);
+        n_bytes = MIN(bytes, s->block_size - (offset % s->block_size));
+
+        if (image_offset == -1) {
+            qemu_iovec_memset(qiov, bytes_done, 0, n_bytes);
         } else {
-            ret = bdrv_pread(bs->file->bs, offset, buf,
+            qemu_iovec_reset(&local_qiov);
-                sectors * BDRV_SECTOR_SIZE);
+            qemu_iovec_concat(&local_qiov, qiov, bytes_done, n_bytes);
-            if (ret != sectors * BDRV_SECTOR_SIZE) {
+
-                return -1;
+            ret = bdrv_co_preadv(bs->file->bs, image_offset, n_bytes,
+                                 &local_qiov, 0);
+            if (ret < 0) {
+                goto fail;
             }
         }
 
-        nb_sectors -= sectors;
+        bytes -= n_bytes;
-        sector_num += sectors;
+        offset += n_bytes;
-        buf += sectors * BDRV_SECTOR_SIZE;
+        bytes_done += n_bytes;
     }
-    return 0;
-}
 
-static coroutine_fn int vpc_co_read(BlockDriverState *bs, int64_t sector_num,
+    ret = 0;
-                                    uint8_t *buf, int nb_sectors)
+fail:
-{
+    qemu_iovec_destroy(&local_qiov);
-    int ret;
-    BDRVVPCState *s = bs->opaque;
-    qemu_co_mutex_lock(&s->lock);
-    ret = vpc_read(bs, sector_num, buf, nb_sectors);
     qemu_co_mutex_unlock(&s->lock);
+
     return ret;
 }
 
-static int vpc_write(BlockDriverState *bs, int64_t sector_num,
+static int coroutine_fn
-    const uint8_t *buf, int nb_sectors)
+vpc_co_pwritev(BlockDriverState *bs, uint64_t offset, uint64_t bytes,
+               QEMUIOVector *qiov, int flags)
 {
     BDRVVPCState *s = bs->opaque;
-    int64_t offset;
+    int64_t image_offset;
-    int64_t sectors, sectors_per_block;
+    int64_t n_bytes;
+    int64_t bytes_done = 0;
     int ret;
     VHDFooter *footer =  (VHDFooter *) s->footer_buf;
+    QEMUIOVector local_qiov;
 
     if (be32_to_cpu(footer->type) == VHD_FIXED) {
-        return bdrv_write(bs->file->bs, sector_num, buf, nb_sectors);
+        return bdrv_co_pwritev(bs->file->bs, offset, bytes, qiov, 0);
-    }
-    while (nb_sectors > 0) {
-        offset = get_sector_offset(bs, sector_num, 1);
-
-        sectors_per_block = s->block_size >> BDRV_SECTOR_BITS;
-        sectors = sectors_per_block - (sector_num % sectors_per_block);
-        if (sectors > nb_sectors) {
-            sectors = nb_sectors;
-        }
-
-        if (offset == -1) {
-            offset = alloc_block(bs, sector_num);
-            if (offset < 0)
-                return -1;
-        }
-
-        ret = bdrv_pwrite(bs->file->bs, offset, buf,
-                          sectors * BDRV_SECTOR_SIZE);
-        if (ret != sectors * BDRV_SECTOR_SIZE) {
-            return -1;
-        }
-
-        nb_sectors -= sectors;
-        sector_num += sectors;
-        buf += sectors * BDRV_SECTOR_SIZE;
     }
 
-    return 0;
-}
-
-static coroutine_fn int vpc_co_write(BlockDriverState *bs, int64_t sector_num,
-                                     const uint8_t *buf, int nb_sectors)
-{
-    int ret;
-    BDRVVPCState *s = bs->opaque;
     qemu_co_mutex_lock(&s->lock);
-    ret = vpc_write(bs, sector_num, buf, nb_sectors);
+    qemu_iovec_init(&local_qiov, qiov->niov);
+
+    while (bytes > 0) {
+        image_offset = get_image_offset(bs, offset, true);
+        n_bytes = MIN(bytes, s->block_size - (offset % s->block_size));
+
+        if (image_offset == -1) {
+            image_offset = alloc_block(bs, offset);
+            if (image_offset < 0) {
+                ret = image_offset;
+                goto fail;
+            }
+        }
+
+        qemu_iovec_reset(&local_qiov);
+        qemu_iovec_concat(&local_qiov, qiov, bytes_done, n_bytes);
+
+        ret = bdrv_co_pwritev(bs->file->bs, image_offset, n_bytes,
+                              &local_qiov, 0);
+        if (ret < 0) {
+            goto fail;
+        }
+
+        bytes -= n_bytes;
+        offset += n_bytes;
+        bytes_done += n_bytes;
+    }
+
+    ret = 0;
+fail:
+    qemu_iovec_destroy(&local_qiov);
     qemu_co_mutex_unlock(&s->lock);
+
     return ret;
 }
 
@@ -783,13 +788,13 @@ static int create_dynamic_disk(BlockBackend *blk, uint8_t *buf,
     block_size = 0x200000;
     num_bat_entries = (total_sectors + block_size / 512) / (block_size / 512);
 
-    ret = blk_pwrite(blk, offset, buf, HEADER_SIZE);
+    ret = blk_pwrite(blk, offset, buf, HEADER_SIZE, 0);
     if (ret < 0) {
         goto fail;
     }
 
     offset = 1536 + ((num_bat_entries * 4 + 511) & ~511);
-    ret = blk_pwrite(blk, offset, buf, HEADER_SIZE);
+    ret = blk_pwrite(blk, offset, buf, HEADER_SIZE, 0);
     if (ret < 0) {
         goto fail;
     }
@@ -799,7 +804,7 @@ static int create_dynamic_disk(BlockBackend *blk, uint8_t *buf,
 
     memset(buf, 0xFF, 512);
     for (i = 0; i < (num_bat_entries * 4 + 511) / 512; i++) {
-        ret = blk_pwrite(blk, offset, buf, 512);
+        ret = blk_pwrite(blk, offset, buf, 512, 0);
         if (ret < 0) {
             goto fail;
         }
@@ -826,7 +831,7 @@ static int create_dynamic_disk(BlockBackend *blk, uint8_t *buf,
     /* Write the header */
     offset = 512;
 
-    ret = blk_pwrite(blk, offset, buf, 1024);
+    ret = blk_pwrite(blk, offset, buf, 1024, 0);
     if (ret < 0) {
         goto fail;
     }
@@ -848,7 +853,7 @@ static int create_fixed_disk(BlockBackend *blk, uint8_t *buf,
         return ret;
     }
 
-    ret = blk_pwrite(blk, total_size - HEADER_SIZE, buf, HEADER_SIZE);
+    ret = blk_pwrite(blk, total_size - HEADER_SIZE, buf, HEADER_SIZE, 0);
     if (ret < 0) {
         return ret;
     }
@@ -1056,8 +1061,8 @@ static BlockDriver bdrv_vpc = {
     .bdrv_reopen_prepare    = vpc_reopen_prepare,
     .bdrv_create            = vpc_create,
 
-    .bdrv_read                  = vpc_co_read,
+    .bdrv_co_preadv             = vpc_co_preadv,
-    .bdrv_write                 = vpc_co_write,
+    .bdrv_co_pwritev            = vpc_co_pwritev,
     .bdrv_co_get_block_status   = vpc_co_get_block_status,
 
     .bdrv_get_info          = vpc_get_info,

block/vvfat.c

@@ -1179,6 +1179,7 @@ static int vvfat_open(BlockDriverState *bs, QDict *options, int flags,
         bs->read_only = 0;
     }
 
+    bs->request_alignment = BDRV_SECTOR_SIZE; /* No sub-sector I/O supported */
     bs->total_sectors = cyls * heads * secs;
 
     if (init_directories(s, dirname, heads, secs, errp)) {
@@ -1421,14 +1422,31 @@ DLOG(fprintf(stderr, "sector %d not allocated\n", (int)sector_num));
     return 0;
 }
 
-static coroutine_fn int vvfat_co_read(BlockDriverState *bs, int64_t sector_num,
+static int coroutine_fn
-                                      uint8_t *buf, int nb_sectors)
+vvfat_co_preadv(BlockDriverState *bs, uint64_t offset, uint64_t bytes,
+                QEMUIOVector *qiov, int flags)
 {
     int ret;
     BDRVVVFATState *s = bs->opaque;
+    uint64_t sector_num = offset >> BDRV_SECTOR_BITS;
+    int nb_sectors = bytes >> BDRV_SECTOR_BITS;
+    void *buf;
+
+    assert((offset & (BDRV_SECTOR_SIZE - 1)) == 0);
+    assert((bytes & (BDRV_SECTOR_SIZE - 1)) == 0);
+
+    buf = g_try_malloc(bytes);
+    if (bytes && buf == NULL) {
+        return -ENOMEM;
+    }
+
     qemu_co_mutex_lock(&s->lock);
     ret = vvfat_read(bs, sector_num, buf, nb_sectors);
     qemu_co_mutex_unlock(&s->lock);
+
+    qemu_iovec_from_buf(qiov, 0, buf, bytes);
+    g_free(buf);
+
     return ret;
 }
 
@@ -2880,14 +2898,31 @@ DLOG(checkpoint());
     return 0;
 }
 
-static coroutine_fn int vvfat_co_write(BlockDriverState *bs, int64_t sector_num,
+static int coroutine_fn
-                                       const uint8_t *buf, int nb_sectors)
+vvfat_co_pwritev(BlockDriverState *bs, uint64_t offset, uint64_t bytes,
+                 QEMUIOVector *qiov, int flags)
 {
     int ret;
     BDRVVVFATState *s = bs->opaque;
+    uint64_t sector_num = offset >> BDRV_SECTOR_BITS;
+    int nb_sectors = bytes >> BDRV_SECTOR_BITS;
+    void *buf;
+
+    assert((offset & (BDRV_SECTOR_SIZE - 1)) == 0);
+    assert((bytes & (BDRV_SECTOR_SIZE - 1)) == 0);
+
+    buf = g_try_malloc(bytes);
+    if (bytes && buf == NULL) {
+        return -ENOMEM;
+    }
+    qemu_iovec_to_buf(qiov, 0, buf, bytes);
+
     qemu_co_mutex_lock(&s->lock);
     ret = vvfat_write(bs, sector_num, buf, nb_sectors);
     qemu_co_mutex_unlock(&s->lock);
+
+    g_free(buf);
+
     return ret;
 }
 
@@ -2904,8 +2939,10 @@ static int64_t coroutine_fn vvfat_co_get_block_status(BlockDriverState *bs,
     return BDRV_BLOCK_DATA;
 }
 
-static int write_target_commit(BlockDriverState *bs, int64_t sector_num,
+static int coroutine_fn
-	const uint8_t* buffer, int nb_sectors) {
+write_target_commit(BlockDriverState *bs, uint64_t offset, uint64_t bytes,
+                    QEMUIOVector *qiov, int flags)
+{
     BDRVVVFATState* s = *((BDRVVVFATState**) bs->opaque);
     return try_commit(s);
 }
@@ -2918,7 +2955,7 @@ static void write_target_close(BlockDriverState *bs) {
 
 static BlockDriver vvfat_write_target = {
     .format_name        = "vvfat_write_target",
-    .bdrv_write         = write_target_commit,
+    .bdrv_co_pwritev    = write_target_commit,
     .bdrv_close         = write_target_close,
 };
 
@@ -3014,8 +3051,8 @@ static BlockDriver bdrv_vvfat = {
     .bdrv_file_open         = vvfat_open,
     .bdrv_close             = vvfat_close,
 
-    .bdrv_read              = vvfat_co_read,
+    .bdrv_co_preadv         = vvfat_co_preadv,
-    .bdrv_write             = vvfat_co_write,
+    .bdrv_co_pwritev        = vvfat_co_pwritev,
     .bdrv_co_get_block_status = vvfat_co_get_block_status,
 };
 

blockdev.c

@@ -73,7 +73,7 @@ static int if_max_devs[IF_COUNT] = {
      * Do not change these numbers!  They govern how drive option
      * index maps to unit and bus.  That mapping is ABI.
      *
-     * All controllers used to imlement if=T drives need to support
+     * All controllers used to implement if=T drives need to support
      * if_max_devs[T] units, for any T with if_max_devs[T] != 0.
      * Otherwise, some index values map to "impossible" bus, unit
      * values.
@@ -4092,6 +4092,61 @@ out:
     aio_context_release(aio_context);
 }
 
+static BdrvChild *bdrv_find_child(BlockDriverState *parent_bs,
+                                  const char *child_name)
+{
+    BdrvChild *child;
+
+    QLIST_FOREACH(child, &parent_bs->children, next) {
+        if (strcmp(child->name, child_name) == 0) {
+            return child;
+        }
+    }
+
+    return NULL;
+}
+
+void qmp_x_blockdev_change(const char *parent, bool has_child,
+                           const char *child, bool has_node,
+                           const char *node, Error **errp)
+{
+    BlockDriverState *parent_bs, *new_bs = NULL;
+    BdrvChild *p_child;
+
+    parent_bs = bdrv_lookup_bs(parent, parent, errp);
+    if (!parent_bs) {
+        return;
+    }
+
+    if (has_child == has_node) {
+        if (has_child) {
+            error_setg(errp, "The parameters child and node are in conflict");
+        } else {
+            error_setg(errp, "Either child or node must be specified");
+        }
+        return;
+    }
+
+    if (has_child) {
+        p_child = bdrv_find_child(parent_bs, child);
+        if (!p_child) {
+            error_setg(errp, "Node '%s' does not have child '%s'",
+                       parent, child);
+            return;
+        }
+        bdrv_del_child(parent_bs, p_child, errp);
+    }
+
+    if (has_node) {
+        new_bs = bdrv_find_node(node);
+        if (!new_bs) {
+            error_setg(errp, "Node '%s' not found", node);
+            return;
+        }
+        bdrv_add_child(parent_bs, new_bs, errp);
+    }
+}
+
 BlockJobInfoList *qmp_query_block_jobs(Error **errp)
 {
     BlockJobInfoList *head = NULL, **p_next = &head;

dma-helpers.c

@@ -73,7 +73,7 @@ typedef struct {
     BlockBackend *blk;
     BlockAIOCB *acb;
     QEMUSGList *sg;
-    uint64_t sector_num;
+    uint64_t offset;
     DMADirection dir;
     int sg_cur_index;
     dma_addr_t sg_cur_byte;
@@ -130,7 +130,7 @@ static void dma_blk_cb(void *opaque, int ret)
     trace_dma_blk_cb(dbs, ret);
 
     dbs->acb = NULL;
-    dbs->sector_num += dbs->iov.size / 512;
+    dbs->offset += dbs->iov.size;
 
     if (dbs->sg_cur_index == dbs->sg->nsg || ret < 0) {
         dma_complete(dbs, ret);
@@ -164,8 +164,8 @@ static void dma_blk_cb(void *opaque, int ret)
         qemu_iovec_discard_back(&dbs->iov, dbs->iov.size & ~BDRV_SECTOR_MASK);
     }
 
-    dbs->acb = dbs->io_func(dbs->blk, dbs->sector_num, &dbs->iov,
+    dbs->acb = dbs->io_func(dbs->blk, dbs->offset, &dbs->iov, 0,
-                            dbs->iov.size / 512, dma_blk_cb, dbs);
+                            dma_blk_cb, dbs);
     assert(dbs->acb);
 }
 
@@ -203,7 +203,7 @@ BlockAIOCB *dma_blk_io(
     dbs->acb = NULL;
     dbs->blk = blk;
     dbs->sg = sg;
-    dbs->sector_num = sector_num;
+    dbs->offset = sector_num << BDRV_SECTOR_BITS;
     dbs->sg_cur_index = 0;
     dbs->sg_cur_byte = 0;
     dbs->dir = dir;
@@ -219,7 +219,7 @@ BlockAIOCB *dma_blk_read(BlockBackend *blk,
                          QEMUSGList *sg, uint64_t sector,
                          void (*cb)(void *opaque, int ret), void *opaque)
 {
-    return dma_blk_io(blk, sg, sector, blk_aio_readv, cb, opaque,
+    return dma_blk_io(blk, sg, sector, blk_aio_preadv, cb, opaque,
                       DMA_DIRECTION_FROM_DEVICE);
 }
 
@@ -227,7 +227,7 @@ BlockAIOCB *dma_blk_write(BlockBackend *blk,
                           QEMUSGList *sg, uint64_t sector,
                           void (*cb)(void *opaque, int ret), void *opaque)
 {
-    return dma_blk_io(blk, sg, sector, blk_aio_writev, cb, opaque,
+    return dma_blk_io(blk, sg, sector, blk_aio_pwritev, cb, opaque,
                       DMA_DIRECTION_TO_DEVICE);
 }
 

hw/block/fdc.c

@@ -223,6 +223,13 @@ static int fd_sector(FDrive *drv)
                           NUM_SIDES(drv));
 }
 
+/* Returns current position, in bytes, for given drive */
+static int fd_offset(FDrive *drv)
+{
+    g_assert(fd_sector(drv) < INT_MAX >> BDRV_SECTOR_BITS);
+    return fd_sector(drv) << BDRV_SECTOR_BITS;
+}
+
 /* Seek to a new position:
  * returns 0 if already on right track
  * returns 1 if track changed
@@ -1629,8 +1636,8 @@ static int fdctrl_transfer_handler (void *opaque, int nchan,
         if (fdctrl->data_dir != FD_DIR_WRITE ||
             len < FD_SECTOR_LEN || rel_pos != 0) {
             /* READ & SCAN commands and realign to a sector for WRITE */
-            if (blk_read(cur_drv->blk, fd_sector(cur_drv),
+            if (blk_pread(cur_drv->blk, fd_offset(cur_drv),
-                         fdctrl->fifo, 1) < 0) {
+                          fdctrl->fifo, BDRV_SECTOR_SIZE) < 0) {
                 FLOPPY_DPRINTF("Floppy: error getting sector %d\n",
                                fd_sector(cur_drv));
                 /* Sure, image size is too small... */
@@ -1657,8 +1664,8 @@ static int fdctrl_transfer_handler (void *opaque, int nchan,
 
             k->read_memory(fdctrl->dma, nchan, fdctrl->fifo + rel_pos,
                            fdctrl->data_pos, len);
-            if (blk_write(cur_drv->blk, fd_sector(cur_drv),
+            if (blk_pwrite(cur_drv->blk, fd_offset(cur_drv),
-                          fdctrl->fifo, 1) < 0) {
+                           fdctrl->fifo, BDRV_SECTOR_SIZE, 0) < 0) {
                 FLOPPY_DPRINTF("error writing sector %d\n",
                                fd_sector(cur_drv));
                 fdctrl_stop_transfer(fdctrl, FD_SR0_ABNTERM | FD_SR0_SEEK, 0x00, 0x00);
@@ -1741,7 +1748,8 @@ static uint32_t fdctrl_read_data(FDCtrl *fdctrl)
                                    fd_sector(cur_drv));
                     return 0;
                 }
-            if (blk_read(cur_drv->blk, fd_sector(cur_drv), fdctrl->fifo, 1)
+            if (blk_pread(cur_drv->blk, fd_offset(cur_drv), fdctrl->fifo,
+                          BDRV_SECTOR_SIZE)
                 < 0) {
                 FLOPPY_DPRINTF("error getting sector %d\n",
                                fd_sector(cur_drv));
@@ -1820,7 +1828,8 @@ static void fdctrl_format_sector(FDCtrl *fdctrl)
     }
     memset(fdctrl->fifo, 0, FD_SECTOR_LEN);
     if (cur_drv->blk == NULL ||
-        blk_write(cur_drv->blk, fd_sector(cur_drv), fdctrl->fifo, 1) < 0) {
+        blk_pwrite(cur_drv->blk, fd_offset(cur_drv), fdctrl->fifo,
+                   BDRV_SECTOR_SIZE, 0) < 0) {
         FLOPPY_DPRINTF("error formatting sector %d\n", fd_sector(cur_drv));
         fdctrl_stop_transfer(fdctrl, FD_SR0_ABNTERM | FD_SR0_SEEK, 0x00, 0x00);
     } else {
@@ -2243,8 +2252,8 @@ static void fdctrl_write_data(FDCtrl *fdctrl, uint32_t value)
         if (pos == FD_SECTOR_LEN - 1 ||
             fdctrl->data_pos == fdctrl->data_len) {
             cur_drv = get_cur_drv(fdctrl);
-            if (blk_write(cur_drv->blk, fd_sector(cur_drv), fdctrl->fifo, 1)
+            if (blk_pwrite(cur_drv->blk, fd_offset(cur_drv), fdctrl->fifo,
-                < 0) {
+                           BDRV_SECTOR_SIZE, 0) < 0) {
                 FLOPPY_DPRINTF("error writing sector %d\n",
                                fd_sector(cur_drv));
                 break;

hw/block/hd-geometry.c

@@ -66,7 +66,7 @@ static int guess_disk_lchs(BlockBackend *blk,
      * but also in async I/O mode. So the I/O throttling function has to
      * be disabled temporarily here, not permanently.
      */
-    if (blk_read_unthrottled(blk, 0, buf, 1) < 0) {
+    if (blk_pread_unthrottled(blk, 0, buf, BDRV_SECTOR_SIZE) < 0) {
         return -1;
     }
     /* test msdos magic */

hw/block/m25p80.c

@@ -358,25 +358,21 @@ static void blk_sync_complete(void *opaque, int ret)
 
 static void flash_sync_page(Flash *s, int page)
 {
-    int blk_sector, nb_sectors;
     QEMUIOVector iov;
 
     if (!s->blk || blk_is_read_only(s->blk)) {
         return;
     }
 
-    blk_sector = (page * s->pi->page_size) / BDRV_SECTOR_SIZE;
-    nb_sectors = DIV_ROUND_UP(s->pi->page_size, BDRV_SECTOR_SIZE);
     qemu_iovec_init(&iov, 1);
-    qemu_iovec_add(&iov, s->storage + blk_sector * BDRV_SECTOR_SIZE,
+    qemu_iovec_add(&iov, s->storage + page * s->pi->page_size,
-                   nb_sectors * BDRV_SECTOR_SIZE);
+                   s->pi->page_size);
-    blk_aio_writev(s->blk, blk_sector, &iov, nb_sectors, blk_sync_complete,
+    blk_aio_pwritev(s->blk, page * s->pi->page_size, &iov, 0,
-                   NULL);
+                    blk_sync_complete, NULL);
 }
 
 static inline void flash_sync_area(Flash *s, int64_t off, int64_t len)
 {
-    int64_t start, end, nb_sectors;
     QEMUIOVector iov;
 
     if (!s->blk || blk_is_read_only(s->blk)) {
@@ -384,13 +380,9 @@ static inline void flash_sync_area(Flash *s, int64_t off, int64_t len)
     }
 
     assert(!(len % BDRV_SECTOR_SIZE));
-    start = off / BDRV_SECTOR_SIZE;
-    end = (off + len) / BDRV_SECTOR_SIZE;
-    nb_sectors = end - start;
     qemu_iovec_init(&iov, 1);
-    qemu_iovec_add(&iov, s->storage + (start * BDRV_SECTOR_SIZE),
+    qemu_iovec_add(&iov, s->storage + off, len);
-                                        nb_sectors * BDRV_SECTOR_SIZE);
+    blk_aio_pwritev(s->blk, off, &iov, 0, blk_sync_complete, NULL);
-    blk_aio_writev(s->blk, start, &iov, nb_sectors, blk_sync_complete, NULL);
 }
 
 static void flash_erase(Flash *s, int offset, FlashCMD cmd)
@@ -907,8 +899,7 @@ static int m25p80_init(SSISlave *ss)
         s->storage = blk_blockalign(s->blk, s->size);
 
         /* FIXME: Move to late init */
-        if (blk_read(s->blk, 0, s->storage,
+        if (blk_pread(s->blk, 0, s->storage, s->size)) {
-                     DIV_ROUND_UP(s->size, BDRV_SECTOR_SIZE))) {
             fprintf(stderr, "Failed to initialize SPI flash!\n");
             return 1;
         }

hw/block/nand.c

@@ -663,7 +663,8 @@ static void glue(nand_blk_write_, PAGE_SIZE)(NANDFlashState *s)
         sector = SECTOR(s->addr);
         off = (s->addr & PAGE_MASK) + s->offset;
         soff = SECTOR_OFFSET(s->addr);
-        if (blk_read(s->blk, sector, iobuf, PAGE_SECTORS) < 0) {
+        if (blk_pread(s->blk, sector << BDRV_SECTOR_BITS, iobuf,
+                      PAGE_SECTORS << BDRV_SECTOR_BITS) < 0) {
             printf("%s: read error in sector %" PRIu64 "\n", __func__, sector);
             return;
         }
@@ -675,21 +676,24 @@ static void glue(nand_blk_write_, PAGE_SIZE)(NANDFlashState *s)
                             MIN(OOB_SIZE, off + s->iolen - PAGE_SIZE));
         }
 
-        if (blk_write(s->blk, sector, iobuf, PAGE_SECTORS) < 0) {
+        if (blk_pwrite(s->blk, sector << BDRV_SECTOR_BITS, iobuf,
+                       PAGE_SECTORS << BDRV_SECTOR_BITS, 0) < 0) {
             printf("%s: write error in sector %" PRIu64 "\n", __func__, sector);
         }
     } else {
         off = PAGE_START(s->addr) + (s->addr & PAGE_MASK) + s->offset;
         sector = off >> 9;
         soff = off & 0x1ff;
-        if (blk_read(s->blk, sector, iobuf, PAGE_SECTORS + 2) < 0) {
+        if (blk_pread(s->blk, sector << BDRV_SECTOR_BITS, iobuf,
+                      (PAGE_SECTORS + 2) << BDRV_SECTOR_BITS) < 0) {
             printf("%s: read error in sector %" PRIu64 "\n", __func__, sector);
             return;
         }
 
         mem_and(iobuf + soff, s->io, s->iolen);
 
-        if (blk_write(s->blk, sector, iobuf, PAGE_SECTORS + 2) < 0) {
+        if (blk_pwrite(s->blk, sector << BDRV_SECTOR_BITS, iobuf,
+                       (PAGE_SECTORS + 2) << BDRV_SECTOR_BITS, 0) < 0) {
             printf("%s: write error in sector %" PRIu64 "\n", __func__, sector);
         }
     }
@@ -716,17 +720,20 @@ static void glue(nand_blk_erase_, PAGE_SIZE)(NANDFlashState *s)
         i = SECTOR(addr);
         page = SECTOR(addr + (1 << (ADDR_SHIFT + s->erase_shift)));
         for (; i < page; i ++)
-            if (blk_write(s->blk, i, iobuf, 1) < 0) {
+            if (blk_pwrite(s->blk, i << BDRV_SECTOR_BITS, iobuf,
+                           BDRV_SECTOR_SIZE, 0) < 0) {
                 printf("%s: write error in sector %" PRIu64 "\n", __func__, i);
             }
     } else {
         addr = PAGE_START(addr);
         page = addr >> 9;
-        if (blk_read(s->blk, page, iobuf, 1) < 0) {
+        if (blk_pread(s->blk, page << BDRV_SECTOR_BITS, iobuf,
+                      BDRV_SECTOR_SIZE) < 0) {
             printf("%s: read error in sector %" PRIu64 "\n", __func__, page);
         }
         memset(iobuf + (addr & 0x1ff), 0xff, (~addr & 0x1ff) + 1);
-        if (blk_write(s->blk, page, iobuf, 1) < 0) {
+        if (blk_pwrite(s->blk, page << BDRV_SECTOR_BITS, iobuf,
+                       BDRV_SECTOR_SIZE, 0) < 0) {
             printf("%s: write error in sector %" PRIu64 "\n", __func__, page);
         }
 
@@ -734,18 +741,20 @@ static void glue(nand_blk_erase_, PAGE_SIZE)(NANDFlashState *s)
         i = (addr & ~0x1ff) + 0x200;
         for (addr += ((PAGE_SIZE + OOB_SIZE) << s->erase_shift) - 0x200;
                         i < addr; i += 0x200) {
-            if (blk_write(s->blk, i >> 9, iobuf, 1) < 0) {
+            if (blk_pwrite(s->blk, i, iobuf, BDRV_SECTOR_SIZE, 0) < 0) {
                 printf("%s: write error in sector %" PRIu64 "\n",
                        __func__, i >> 9);
             }
         }
 
         page = i >> 9;
-        if (blk_read(s->blk, page, iobuf, 1) < 0) {
+        if (blk_pread(s->blk, page << BDRV_SECTOR_BITS, iobuf,
+                      BDRV_SECTOR_SIZE) < 0) {
             printf("%s: read error in sector %" PRIu64 "\n", __func__, page);
         }
         memset(iobuf, 0xff, ((addr - 1) & 0x1ff) + 1);
-        if (blk_write(s->blk, page, iobuf, 1) < 0) {
+        if (blk_pwrite(s->blk, page << BDRV_SECTOR_BITS, iobuf,
+                       BDRV_SECTOR_SIZE, 0) < 0) {
             printf("%s: write error in sector %" PRIu64 "\n", __func__, page);
         }
     }
@@ -760,7 +769,8 @@ static void glue(nand_blk_load_, PAGE_SIZE)(NANDFlashState *s,
 
     if (s->blk) {
         if (s->mem_oob) {
-            if (blk_read(s->blk, SECTOR(addr), s->io, PAGE_SECTORS) < 0) {
+            if (blk_pread(s->blk, SECTOR(addr) << BDRV_SECTOR_BITS, s->io,
+                          PAGE_SECTORS << BDRV_SECTOR_BITS) < 0) {
                 printf("%s: read error in sector %" PRIu64 "\n",
                                 __func__, SECTOR(addr));
             }
@@ -769,8 +779,8 @@ static void glue(nand_blk_load_, PAGE_SIZE)(NANDFlashState *s,
                             OOB_SIZE);
             s->ioaddr = s->io + SECTOR_OFFSET(s->addr) + offset;
         } else {
-            if (blk_read(s->blk, PAGE_START(addr) >> 9,
+            if (blk_pread(s->blk, PAGE_START(addr), s->io,
-                         s->io, (PAGE_SECTORS + 2)) < 0) {
+                          (PAGE_SECTORS + 2) << BDRV_SECTOR_BITS) < 0) {
                 printf("%s: read error in sector %" PRIu64 "\n",
                                 __func__, PAGE_START(addr) >> 9);
             }

hw/block/onenand.c

@@ -224,7 +224,8 @@ static void onenand_reset(OneNANDState *s, int cold)
         /* Lock the whole flash */
         memset(s->blockwp, ONEN_LOCK_LOCKED, s->blocks);
 
-        if (s->blk_cur && blk_read(s->blk_cur, 0, s->boot[0], 8) < 0) {
+        if (s->blk_cur && blk_pread(s->blk_cur, 0, s->boot[0],
+                                    8 << BDRV_SECTOR_BITS) < 0) {
             hw_error("%s: Loading the BootRAM failed.\n", __func__);
         }
     }
@@ -240,8 +241,11 @@ static void onenand_system_reset(DeviceState *dev)
 static inline int onenand_load_main(OneNANDState *s, int sec, int secn,
                 void *dest)
 {
+    assert(UINT32_MAX >> BDRV_SECTOR_BITS > sec);
+    assert(UINT32_MAX >> BDRV_SECTOR_BITS > secn);
     if (s->blk_cur) {
-        return blk_read(s->blk_cur, sec, dest, secn) < 0;
+        return blk_pread(s->blk_cur, sec << BDRV_SECTOR_BITS, dest,
+                         secn << BDRV_SECTOR_BITS) < 0;
     } else if (sec + secn > s->secs_cur) {
         return 1;
     }
@@ -257,19 +261,22 @@ static inline int onenand_prog_main(OneNANDState *s, int sec, int secn,
     int result = 0;
 
     if (secn > 0) {
-        uint32_t size = (uint32_t)secn * 512;
+        uint32_t size = secn << BDRV_SECTOR_BITS;
+        uint32_t offset = sec << BDRV_SECTOR_BITS;
+        assert(UINT32_MAX >> BDRV_SECTOR_BITS > sec);
+        assert(UINT32_MAX >> BDRV_SECTOR_BITS > secn);
         const uint8_t *sp = (const uint8_t *)src;
         uint8_t *dp = 0;
         if (s->blk_cur) {
             dp = g_malloc(size);
-            if (!dp || blk_read(s->blk_cur, sec, dp, secn) < 0) {
+            if (!dp || blk_pread(s->blk_cur, offset, dp, size) < 0) {
                 result = 1;
             }
         } else {
             if (sec + secn > s->secs_cur) {
                 result = 1;
             } else {
-                dp = (uint8_t *)s->current + (sec << 9);
+                dp = (uint8_t *)s->current + offset;
             }
         }
         if (!result) {
@@ -278,7 +285,7 @@ static inline int onenand_prog_main(OneNANDState *s, int sec, int secn,
                 dp[i] &= sp[i];
             }
             if (s->blk_cur) {
-                result = blk_write(s->blk_cur, sec, dp, secn) < 0;
+                result = blk_pwrite(s->blk_cur, offset, dp, size, 0) < 0;
             }
         }
         if (dp && s->blk_cur) {
@@ -295,7 +302,8 @@ static inline int onenand_load_spare(OneNANDState *s, int sec, int secn,
     uint8_t buf[512];
 
     if (s->blk_cur) {
-        if (blk_read(s->blk_cur, s->secs_cur + (sec >> 5), buf, 1) < 0) {
+        uint32_t offset = (s->secs_cur + (sec >> 5)) << BDRV_SECTOR_BITS;
+        if (blk_pread(s->blk_cur, offset, buf, BDRV_SECTOR_SIZE) < 0) {
             return 1;
         }
         memcpy(dest, buf + ((sec & 31) << 4), secn << 4);
@@ -304,7 +312,7 @@ static inline int onenand_load_spare(OneNANDState *s, int sec, int secn,
     } else {
         memcpy(dest, s->current + (s->secs_cur << 9) + (sec << 4), secn << 4);
     }
- 
+
     return 0;
 }
 
@@ -315,10 +323,12 @@ static inline int onenand_prog_spare(OneNANDState *s, int sec, int secn,
     if (secn > 0) {
         const uint8_t *sp = (const uint8_t *)src;
         uint8_t *dp = 0, *dpp = 0;
+        uint32_t offset = (s->secs_cur + (sec >> 5)) << BDRV_SECTOR_BITS;
+        assert(UINT32_MAX >> BDRV_SECTOR_BITS > s->secs_cur + (sec >> 5));
         if (s->blk_cur) {
             dp = g_malloc(512);
             if (!dp
-                || blk_read(s->blk_cur, s->secs_cur + (sec >> 5), dp, 1) < 0) {
+                || blk_pread(s->blk_cur, offset, dp, BDRV_SECTOR_SIZE) < 0) {
                 result = 1;
             } else {
                 dpp = dp + ((sec & 31) << 4);
@@ -336,8 +346,8 @@ static inline int onenand_prog_spare(OneNANDState *s, int sec, int secn,
                 dpp[i] &= sp[i];
             }
             if (s->blk_cur) {
-                result = blk_write(s->blk_cur, s->secs_cur + (sec >> 5),
+                result = blk_pwrite(s->blk_cur, offset, dp,
-                                   dp, 1) < 0;
+                                    BDRV_SECTOR_SIZE, 0) < 0;
             }
         }
         g_free(dp);
@@ -355,14 +365,17 @@ static inline int onenand_erase(OneNANDState *s, int sec, int num)
     for (; num > 0; num--, sec++) {
         if (s->blk_cur) {
             int erasesec = s->secs_cur + (sec >> 5);
-            if (blk_write(s->blk_cur, sec, blankbuf, 1) < 0) {
+            if (blk_pwrite(s->blk_cur, sec << BDRV_SECTOR_BITS, blankbuf,
+                           BDRV_SECTOR_SIZE, 0) < 0) {
                 goto fail;
             }
-            if (blk_read(s->blk_cur, erasesec, tmpbuf, 1) < 0) {
+            if (blk_pread(s->blk_cur, erasesec << BDRV_SECTOR_BITS, tmpbuf,
+                          BDRV_SECTOR_SIZE) < 0) {
                 goto fail;
             }
             memcpy(tmpbuf + ((sec & 31) << 4), blankbuf, 1 << 4);
-            if (blk_write(s->blk_cur, erasesec, tmpbuf, 1) < 0) {
+            if (blk_pwrite(s->blk_cur, erasesec << BDRV_SECTOR_BITS, tmpbuf,
+                           BDRV_SECTOR_SIZE, 0) < 0) {
                 goto fail;
             }
         } else {

hw/block/pflash_cfi01.c

@@ -413,11 +413,11 @@ static void pflash_update(pflash_t *pfl, int offset,
     int offset_end;
     if (pfl->blk) {
         offset_end = offset + size;
-        /* round to sectors */
+        /* widen to sector boundaries */
-        offset = offset >> 9;
+        offset = QEMU_ALIGN_DOWN(offset, BDRV_SECTOR_SIZE);
-        offset_end = (offset_end + 511) >> 9;
+        offset_end = QEMU_ALIGN_UP(offset_end, BDRV_SECTOR_SIZE);
-        blk_write(pfl->blk, offset, pfl->storage + (offset << 9),
+        blk_pwrite(pfl->blk, offset, pfl->storage + offset,
-                  offset_end - offset);
+                   offset_end - offset, 0);
     }
 }
 
@@ -739,7 +739,7 @@ static void pflash_cfi01_realize(DeviceState *dev, Error **errp)
 
     if (pfl->blk) {
         /* read the initial flash content */
-        ret = blk_read(pfl->blk, 0, pfl->storage, total_len >> 9);
+        ret = blk_pread(pfl->blk, 0, pfl->storage, total_len);
 
         if (ret < 0) {
             vmstate_unregister_ram(&pfl->mem, DEVICE(pfl));

hw/block/pflash_cfi02.c

@@ -253,11 +253,11 @@ static void pflash_update(pflash_t *pfl, int offset,
     int offset_end;
     if (pfl->blk) {
         offset_end = offset + size;
-        /* round to sectors */
+        /* widen to sector boundaries */
-        offset = offset >> 9;
+        offset = QEMU_ALIGN_DOWN(offset, BDRV_SECTOR_SIZE);
-        offset_end = (offset_end + 511) >> 9;
+        offset_end = QEMU_ALIGN_UP(offset_end, BDRV_SECTOR_SIZE);
-        blk_write(pfl->blk, offset, pfl->storage + (offset << 9),
+        blk_pwrite(pfl->blk, offset, pfl->storage + offset,
-                  offset_end - offset);
+                   offset_end - offset, 0);
     }
 }
 
@@ -622,7 +622,7 @@ static void pflash_cfi02_realize(DeviceState *dev, Error **errp)
     pfl->chip_len = chip_len;
     if (pfl->blk) {
         /* read the initial flash content */
-        ret = blk_read(pfl->blk, 0, pfl->storage, chip_len >> 9);
+        ret = blk_pread(pfl->blk, 0, pfl->storage, chip_len);
         if (ret < 0) {
             vmstate_unregister_ram(&pfl->orig_mem, DEVICE(pfl));
             error_setg(errp, "failed to read the initial flash content");

hw/block/virtio-blk.c

@@ -322,7 +322,6 @@ static inline void submit_requests(BlockBackend *blk, MultiReqBuffer *mrb,
 {
     QEMUIOVector *qiov = &mrb->reqs[start]->qiov;
     int64_t sector_num = mrb->reqs[start]->sector_num;
-    int nb_sectors = mrb->reqs[start]->qiov.size / BDRV_SECTOR_SIZE;
     bool is_write = mrb->is_write;
 
     if (num_reqs > 1) {
@@ -331,7 +330,7 @@ static inline void submit_requests(BlockBackend *blk, MultiReqBuffer *mrb,
         int tmp_niov = qiov->niov;
 
         /* mrb->reqs[start]->qiov was initialized from external so we can't
-         * modifiy it here. We need to initialize it locally and then add the
+         * modify it here. We need to initialize it locally and then add the
          * external iovecs. */
         qemu_iovec_init(qiov, niov);
 
@@ -343,23 +342,22 @@ static inline void submit_requests(BlockBackend *blk, MultiReqBuffer *mrb,
             qemu_iovec_concat(qiov, &mrb->reqs[i]->qiov, 0,
                               mrb->reqs[i]->qiov.size);
             mrb->reqs[i - 1]->mr_next = mrb->reqs[i];
-            nb_sectors += mrb->reqs[i]->qiov.size / BDRV_SECTOR_SIZE;
         }
-        assert(nb_sectors == qiov->size / BDRV_SECTOR_SIZE);
 
-        trace_virtio_blk_submit_multireq(mrb, start, num_reqs, sector_num,
+        trace_virtio_blk_submit_multireq(mrb, start, num_reqs,
-                                         nb_sectors, is_write);
+                                         sector_num << BDRV_SECTOR_BITS,
+                                         qiov->size, is_write);
         block_acct_merge_done(blk_get_stats(blk),
                               is_write ? BLOCK_ACCT_WRITE : BLOCK_ACCT_READ,
                               num_reqs - 1);
     }
 
     if (is_write) {
-        blk_aio_writev(blk, sector_num, qiov, nb_sectors,
+        blk_aio_pwritev(blk, sector_num << BDRV_SECTOR_BITS, qiov, 0,
-                       virtio_blk_rw_complete, mrb->reqs[start]);
+                        virtio_blk_rw_complete, mrb->reqs[start]);
     } else {
-        blk_aio_readv(blk, sector_num, qiov, nb_sectors,
+        blk_aio_preadv(blk, sector_num << BDRV_SECTOR_BITS, qiov, 0,
-                      virtio_blk_rw_complete, mrb->reqs[start]);
+                       virtio_blk_rw_complete, mrb->reqs[start]);
     }
 }
 

hw/block/xen_disk.c

@@ -554,9 +554,8 @@ static int ioreq_runio_qemu_aio(struct ioreq *ioreq)
         block_acct_start(blk_get_stats(blkdev->blk), &ioreq->acct,
                          ioreq->v.size, BLOCK_ACCT_READ);
         ioreq->aio_inflight++;
-        blk_aio_readv(blkdev->blk, ioreq->start / BLOCK_SIZE,
+        blk_aio_preadv(blkdev->blk, ioreq->start, &ioreq->v, 0,
-                      &ioreq->v, ioreq->v.size / BLOCK_SIZE,
+                       qemu_aio_complete, ioreq);
-                      qemu_aio_complete, ioreq);
         break;
     case BLKIF_OP_WRITE:
     case BLKIF_OP_FLUSH_DISKCACHE:
@@ -569,9 +568,8 @@ static int ioreq_runio_qemu_aio(struct ioreq *ioreq)
                          ioreq->req.operation == BLKIF_OP_WRITE ?
                          BLOCK_ACCT_WRITE : BLOCK_ACCT_FLUSH);
         ioreq->aio_inflight++;
-        blk_aio_writev(blkdev->blk, ioreq->start / BLOCK_SIZE,
+        blk_aio_pwritev(blkdev->blk, ioreq->start, &ioreq->v, 0,
-                       &ioreq->v, ioreq->v.size / BLOCK_SIZE,
+                        qemu_aio_complete, ioreq);
-                       qemu_aio_complete, ioreq);
         break;
     case BLKIF_OP_DISCARD:
     {

hw/ide/atapi.c

@@ -28,6 +28,9 @@
 #include "hw/scsi/scsi.h"
 #include "sysemu/block-backend.h"
 
+#define ATAPI_SECTOR_BITS (2 + BDRV_SECTOR_BITS)
+#define ATAPI_SECTOR_SIZE (1 << ATAPI_SECTOR_BITS)
+
 static void ide_atapi_cmd_read_dma_cb(void *opaque, int ret);
 
 static void padstr8(uint8_t *buf, int buf_size, const char *src)
@@ -111,7 +114,7 @@ cd_read_sector_sync(IDEState *s)
 {
     int ret;
     block_acct_start(blk_get_stats(s->blk), &s->acct,
-                     4 * BDRV_SECTOR_SIZE, BLOCK_ACCT_READ);
+                     ATAPI_SECTOR_SIZE, BLOCK_ACCT_READ);
 
 #ifdef DEBUG_IDE_ATAPI
     printf("cd_read_sector_sync: lba=%d\n", s->lba);
@@ -119,12 +122,12 @@ cd_read_sector_sync(IDEState *s)
 
     switch (s->cd_sector_size) {
     case 2048:
-        ret = blk_read(s->blk, (int64_t)s->lba << 2,
+        ret = blk_pread(s->blk, (int64_t)s->lba << ATAPI_SECTOR_BITS,
-                       s->io_buffer, 4);
+                        s->io_buffer, ATAPI_SECTOR_SIZE);
         break;
     case 2352:
-        ret = blk_read(s->blk, (int64_t)s->lba << 2,
+        ret = blk_pread(s->blk, (int64_t)s->lba << ATAPI_SECTOR_BITS,
-                       s->io_buffer + 16, 4);
+                        s->io_buffer + 16, ATAPI_SECTOR_SIZE);
         if (ret >= 0) {
             cd_data_to_raw(s->io_buffer, s->lba);
         }
@@ -182,7 +185,7 @@ static int cd_read_sector(IDEState *s)
     s->iov.iov_base = (s->cd_sector_size == 2352) ?
                       s->io_buffer + 16 : s->io_buffer;
 
-    s->iov.iov_len = 4 * BDRV_SECTOR_SIZE;
+    s->iov.iov_len = ATAPI_SECTOR_SIZE;
     qemu_iovec_init_external(&s->qiov, &s->iov, 1);
 
 #ifdef DEBUG_IDE_ATAPI
@@ -190,7 +193,7 @@ static int cd_read_sector(IDEState *s)
 #endif
 
     block_acct_start(blk_get_stats(s->blk), &s->acct,
-                     4 * BDRV_SECTOR_SIZE, BLOCK_ACCT_READ);
+                     ATAPI_SECTOR_SIZE, BLOCK_ACCT_READ);
 
     ide_buffered_readv(s, (int64_t)s->lba << 2, &s->qiov, 4,
                        cd_read_sector_cb, s);
@@ -435,7 +438,7 @@ static void ide_atapi_cmd_read_dma_cb(void *opaque, int ret)
 #endif
 
     s->bus->dma->iov.iov_base = (void *)(s->io_buffer + data_offset);
-    s->bus->dma->iov.iov_len = n * 4 * 512;
+    s->bus->dma->iov.iov_len = n * ATAPI_SECTOR_SIZE;
     qemu_iovec_init_external(&s->bus->dma->qiov, &s->bus->dma->iov, 1);
 
     s->bus->dma->aiocb = ide_buffered_readv(s, (int64_t)s->lba << 2,

hw/ide/core.c

@@ -442,7 +442,7 @@ static void ide_issue_trim_cb(void *opaque, int ret)
 }
 
 BlockAIOCB *ide_issue_trim(BlockBackend *blk,
-        int64_t sector_num, QEMUIOVector *qiov, int nb_sectors,
+        int64_t offset, QEMUIOVector *qiov, BdrvRequestFlags flags,
         BlockCompletionFunc *cb, void *opaque)
 {
     TrimAIOCB *iocb;
@@ -616,8 +616,8 @@ BlockAIOCB *ide_buffered_readv(IDEState *s, int64_t sector_num,
     req->iov.iov_len = iov->size;
     qemu_iovec_init_external(&req->qiov, &req->iov, 1);
 
-    aioreq = blk_aio_readv(s->blk, sector_num, &req->qiov, nb_sectors,
+    aioreq = blk_aio_preadv(s->blk, sector_num << BDRV_SECTOR_BITS,
-                           ide_buffered_readv_cb, req);
+                            &req->qiov, 0, ide_buffered_readv_cb, req);
 
     QLIST_INSERT_HEAD(&s->buffered_requests, req, list);
     return aioreq;
@@ -1006,8 +1006,8 @@ static void ide_sector_write(IDEState *s)
 
     block_acct_start(blk_get_stats(s->blk), &s->acct,
                      n * BDRV_SECTOR_SIZE, BLOCK_ACCT_WRITE);
-    s->pio_aiocb = blk_aio_writev(s->blk, sector_num, &s->qiov, n,
+    s->pio_aiocb = blk_aio_pwritev(s->blk, sector_num << BDRV_SECTOR_BITS,
-                                  ide_sector_write_cb, s);
+                                   &s->qiov, 0, ide_sector_write_cb, s);
 }
 
 static void ide_flush_cb(void *opaque, int ret)

hw/ide/internal.h

@@ -614,7 +614,7 @@ void ide_transfer_start(IDEState *s, uint8_t *buf, int size,
 void ide_transfer_stop(IDEState *s);
 void ide_set_inactive(IDEState *s, bool more);
 BlockAIOCB *ide_issue_trim(BlockBackend *blk,
-        int64_t sector_num, QEMUIOVector *qiov, int nb_sectors,
+        int64_t offset, QEMUIOVector *qiov, BdrvRequestFlags flags,
         BlockCompletionFunc *cb, void *opaque);
 BlockAIOCB *ide_buffered_readv(IDEState *s, int64_t sector_num,
                                QEMUIOVector *iov, int nb_sectors,

hw/ide/macio.c

@@ -55,8 +55,8 @@ static const int debug_macio = 0;
 /*
  * Unaligned DMA read/write access functions required for OS X/Darwin which
  * don't perform DMA transactions on sector boundaries. These functions are
- * modelled on bdrv_co_do_preadv()/bdrv_co_do_pwritev() and so should be
+ * modelled on bdrv_co_preadv()/bdrv_co_pwritev() and so should be easy to
- * easy to remove if the unaligned block APIs are ever exposed.
+ * remove if the unaligned block APIs are ever exposed.
  */
 
 static void pmac_dma_read(BlockBackend *blk,
@@ -120,8 +120,7 @@ static void pmac_dma_read(BlockBackend *blk,
     MACIO_DPRINTF("--- Block read transfer - sector_num: %" PRIx64 "  "
                   "nsector: %x\n", (offset >> 9), (bytes >> 9));
 
-    s->bus->dma->aiocb = blk_aio_readv(blk, (offset >> 9), &io->iov,
+    s->bus->dma->aiocb = blk_aio_preadv(blk, offset, &io->iov, 0, cb, io);
-                             (bytes >> 9), cb, io);
 }
 
 static void pmac_dma_write(BlockBackend *blk,
@@ -205,8 +204,7 @@ static void pmac_dma_write(BlockBackend *blk,
     MACIO_DPRINTF("--- Block write transfer - sector_num: %" PRIx64 "  "
                   "nsector: %x\n", (offset >> 9), (bytes >> 9));
 
-    s->bus->dma->aiocb = blk_aio_writev(blk, (offset >> 9), &io->iov,
+    s->bus->dma->aiocb = blk_aio_pwritev(blk, offset, &io->iov, 0, cb, io);
-                             (bytes >> 9), cb, io);
 }
 
 static void pmac_dma_trim(BlockBackend *blk,
@@ -232,8 +230,7 @@ static void pmac_dma_trim(BlockBackend *blk,
     s->io_buffer_index += io->len;
     io->len = 0;
 
-    s->bus->dma->aiocb = ide_issue_trim(blk, (offset >> 9), &io->iov,
+    s->bus->dma->aiocb = ide_issue_trim(blk, offset, &io->iov, 0, cb, io);
-                             (bytes >> 9), cb, io);
 }
 
 static void pmac_ide_atapi_transfer_cb(void *opaque, int ret)

hw/nvram/spapr_nvram.c

@@ -124,7 +124,7 @@ static void rtas_nvram_store(PowerPCCPU *cpu, sPAPRMachineState *spapr,
 
     alen = len;
     if (nvram->blk) {
-        alen = blk_pwrite(nvram->blk, offset, membuf, len);
+        alen = blk_pwrite(nvram->blk, offset, membuf, len, 0);
     }
 
     assert(nvram->buf);
@@ -190,7 +190,7 @@ static int spapr_nvram_post_load(void *opaque, int version_id)
     sPAPRNVRAM *nvram = VIO_SPAPR_NVRAM(opaque);
 
     if (nvram->blk) {
-        int alen = blk_pwrite(nvram->blk, 0, nvram->buf, nvram->size);
+        int alen = blk_pwrite(nvram->blk, 0, nvram->buf, nvram->size, 0);
 
         if (alen < 0) {
             return alen;

hw/scsi/scsi-disk.c

@@ -108,7 +108,7 @@ static void scsi_check_condition(SCSIDiskReq *r, SCSISense sense)
     scsi_req_complete(&r->req, CHECK_CONDITION);
 }
 
-static uint32_t scsi_init_iovec(SCSIDiskReq *r, size_t size)
+static void scsi_init_iovec(SCSIDiskReq *r, size_t size)
 {
     SCSIDiskState *s = DO_UPCAST(SCSIDiskState, qdev, r->req.dev);
 
@@ -118,7 +118,6 @@ static uint32_t scsi_init_iovec(SCSIDiskReq *r, size_t size)
     }
     r->iov.iov_len = MIN(r->sector_count * 512, r->buflen);
     qemu_iovec_init_external(&r->qiov, &r->iov, 1);
-    return r->qiov.size / 512;
 }
 
 static void scsi_disk_save_request(QEMUFile *f, SCSIRequest *req)
@@ -316,7 +315,6 @@ done:
 static void scsi_do_read(SCSIDiskReq *r, int ret)
 {
     SCSIDiskState *s = DO_UPCAST(SCSIDiskState, qdev, r->req.dev);
-    uint32_t n;
 
     assert (r->req.aiocb == NULL);
 
@@ -340,11 +338,12 @@ static void scsi_do_read(SCSIDiskReq *r, int ret)
         r->req.aiocb = dma_blk_read(s->qdev.conf.blk, r->req.sg, r->sector,
                                     scsi_dma_complete, r);
     } else {
-        n = scsi_init_iovec(r, SCSI_DMA_BUF_SIZE);
+        scsi_init_iovec(r, SCSI_DMA_BUF_SIZE);
         block_acct_start(blk_get_stats(s->qdev.conf.blk), &r->acct,
-                         n * BDRV_SECTOR_SIZE, BLOCK_ACCT_READ);
+                         r->qiov.size, BLOCK_ACCT_READ);
-        r->req.aiocb = blk_aio_readv(s->qdev.conf.blk, r->sector, &r->qiov, n,
+        r->req.aiocb = blk_aio_preadv(s->qdev.conf.blk,
-                                     scsi_read_complete, r);
+                                      r->sector << BDRV_SECTOR_BITS, &r->qiov,
+                                      0, scsi_read_complete, r);
     }
 
 done:
@@ -504,7 +503,6 @@ static void scsi_write_data(SCSIRequest *req)
 {
     SCSIDiskReq *r = DO_UPCAST(SCSIDiskReq, req, req);
     SCSIDiskState *s = DO_UPCAST(SCSIDiskState, qdev, r->req.dev);
-    uint32_t n;
 
     /* No data transfer may already be in progress */
     assert(r->req.aiocb == NULL);
@@ -544,11 +542,11 @@ static void scsi_write_data(SCSIRequest *req)
         r->req.aiocb = dma_blk_write(s->qdev.conf.blk, r->req.sg, r->sector,
                                      scsi_dma_complete, r);
     } else {
-        n = r->qiov.size / 512;
         block_acct_start(blk_get_stats(s->qdev.conf.blk), &r->acct,
-                         n * BDRV_SECTOR_SIZE, BLOCK_ACCT_WRITE);
+                         r->qiov.size, BLOCK_ACCT_WRITE);
-        r->req.aiocb = blk_aio_writev(s->qdev.conf.blk, r->sector, &r->qiov, n,
+        r->req.aiocb = blk_aio_pwritev(s->qdev.conf.blk,
-                                      scsi_write_complete, r);
+                                       r->sector << BDRV_SECTOR_BITS, &r->qiov,
+                                       0, scsi_write_complete, r);
     }
 }
 
@@ -1730,13 +1728,13 @@ static void scsi_write_same_complete(void *opaque, int ret)
     if (data->iov.iov_len) {
         block_acct_start(blk_get_stats(s->qdev.conf.blk), &r->acct,
                          data->iov.iov_len, BLOCK_ACCT_WRITE);
-        /* blk_aio_write doesn't like the qiov size being different from
+        /* Reinitialize qiov, to handle unaligned WRITE SAME request
-         * nb_sectors, make sure they match.
+         * where final qiov may need smaller size */
-         */
         qemu_iovec_init_external(&data->qiov, &data->iov, 1);
-        r->req.aiocb = blk_aio_writev(s->qdev.conf.blk, data->sector,
+        r->req.aiocb = blk_aio_pwritev(s->qdev.conf.blk,
-                                      &data->qiov, data->iov.iov_len / 512,
+                                       data->sector << BDRV_SECTOR_BITS,
-                                      scsi_write_same_complete, data);
+                                       &data->qiov, 0,
+                                       scsi_write_same_complete, data);
         return;
     }
 
@@ -1781,8 +1779,8 @@ static void scsi_disk_emulate_write_same(SCSIDiskReq *r, uint8_t *inbuf)
                          nb_sectors * s->qdev.blocksize,
                         BLOCK_ACCT_WRITE);
         r->req.aiocb = blk_aio_write_zeroes(s->qdev.conf.blk,
-                                r->req.cmd.lba * (s->qdev.blocksize / 512),
+                                r->req.cmd.lba * s->qdev.blocksize,
-                                nb_sectors * (s->qdev.blocksize / 512),
+                                nb_sectors * s->qdev.blocksize,
                                 flags, scsi_aio_complete, r);
         return;
     }
@@ -1803,9 +1801,10 @@ static void scsi_disk_emulate_write_same(SCSIDiskReq *r, uint8_t *inbuf)
     scsi_req_ref(&r->req);
     block_acct_start(blk_get_stats(s->qdev.conf.blk), &r->acct,
                      data->iov.iov_len, BLOCK_ACCT_WRITE);
-    r->req.aiocb = blk_aio_writev(s->qdev.conf.blk, data->sector,
+    r->req.aiocb = blk_aio_pwritev(s->qdev.conf.blk,
-                                  &data->qiov, data->iov.iov_len / 512,
+                                   data->sector << BDRV_SECTOR_BITS,
-                                  scsi_write_same_complete, data);
+                                   &data->qiov, 0,
+                                   scsi_write_same_complete, data);
 }
 
 static void scsi_disk_emulate_write_data(SCSIRequest *req)

hw/sd/sd.c

@@ -123,7 +123,6 @@ struct SDState {
     qemu_irq readonly_cb;
     qemu_irq inserted_cb;
     BlockBackend *blk;
-    uint8_t *buf;
 
     bool enable;
 };
@@ -551,7 +550,7 @@ static const VMStateDescription sd_vmstate = {
         VMSTATE_UINT64(data_start, SDState),
         VMSTATE_UINT32(data_offset, SDState),
         VMSTATE_UINT8_ARRAY(data, SDState, 512),
-        VMSTATE_BUFFER_POINTER_UNSAFE(buf, SDState, 1, 512),
+        VMSTATE_UNUSED_V(1, 512),
         VMSTATE_BOOL(enable, SDState),
         VMSTATE_END_OF_LIST()
     },
@@ -1577,57 +1576,17 @@ send_response:
 
 static void sd_blk_read(SDState *sd, uint64_t addr, uint32_t len)
 {
-    uint64_t end = addr + len;
-
     DPRINTF("sd_blk_read: addr = 0x%08llx, len = %d\n",
             (unsigned long long) addr, len);
-    if (!sd->blk || blk_read(sd->blk, addr >> 9, sd->buf, 1) < 0) {
+    if (!sd->blk || blk_pread(sd->blk, addr, sd->data, len) < 0) {
         fprintf(stderr, "sd_blk_read: read error on host side\n");
-        return;
     }
-
-    if (end > (addr & ~511) + 512) {
-        memcpy(sd->data, sd->buf + (addr & 511), 512 - (addr & 511));
-
-        if (blk_read(sd->blk, end >> 9, sd->buf, 1) < 0) {
-            fprintf(stderr, "sd_blk_read: read error on host side\n");
-            return;
-        }
-        memcpy(sd->data + 512 - (addr & 511), sd->buf, end & 511);
-    } else
-        memcpy(sd->data, sd->buf + (addr & 511), len);
 }
 
 static void sd_blk_write(SDState *sd, uint64_t addr, uint32_t len)
 {
-    uint64_t end = addr + len;
+    if (!sd->blk || blk_pwrite(sd->blk, addr, sd->data, len, 0) < 0) {
-
+        fprintf(stderr, "sd_blk_write: write error on host side\n");
-    if ((addr & 511) || len < 512)
-        if (!sd->blk || blk_read(sd->blk, addr >> 9, sd->buf, 1) < 0) {
-            fprintf(stderr, "sd_blk_write: read error on host side\n");
-            return;
-        }
-
-    if (end > (addr & ~511) + 512) {
-        memcpy(sd->buf + (addr & 511), sd->data, 512 - (addr & 511));
-        if (blk_write(sd->blk, addr >> 9, sd->buf, 1) < 0) {
-            fprintf(stderr, "sd_blk_write: write error on host side\n");
-            return;
-        }
-
-        if (blk_read(sd->blk, end >> 9, sd->buf, 1) < 0) {
-            fprintf(stderr, "sd_blk_write: read error on host side\n");
-            return;
-        }
-        memcpy(sd->buf, sd->data + 512 - (addr & 511), end & 511);
-        if (blk_write(sd->blk, end >> 9, sd->buf, 1) < 0) {
-            fprintf(stderr, "sd_blk_write: write error on host side\n");
-        }
-    } else {
-        memcpy(sd->buf + (addr & 511), sd->data, len);
-        if (!sd->blk || blk_write(sd->blk, addr >> 9, sd->buf, 1) < 0) {
-            fprintf(stderr, "sd_blk_write: write error on host side\n");
-        }
     }
 }
 
@@ -1925,8 +1884,6 @@ static void sd_realize(DeviceState *dev, Error **errp)
         return;
     }
 
-    sd->buf = blk_blockalign(sd->blk, 512);
-
     if (sd->blk) {
         blk_set_dev_ops(sd->blk, &sd_block_ops, sd);
     }

include/block/block.h

@@ -476,6 +476,10 @@ void bdrv_disable_copy_on_read(BlockDriverState *bs);
 void bdrv_ref(BlockDriverState *bs);
 void bdrv_unref(BlockDriverState *bs);
 void bdrv_unref_child(BlockDriverState *parent, BdrvChild *child);
+BdrvChild *bdrv_attach_child(BlockDriverState *parent_bs,
+                             BlockDriverState *child_bs,
+                             const char *child_name,
+                             const BdrvChildRole *child_role);
 
 bool bdrv_op_is_blocked(BlockDriverState *bs, BlockOpType op, Error **errp);
 void bdrv_op_block(BlockDriverState *bs, BlockOpType op, Error *reason);
@@ -520,7 +524,8 @@ int bdrv_probe_geometry(BlockDriverState *bs, HDGeometry *geo);
 
 void bdrv_io_plug(BlockDriverState *bs);
 void bdrv_io_unplug(BlockDriverState *bs);
-void bdrv_flush_io_queue(BlockDriverState *bs);
+void bdrv_io_unplugged_begin(BlockDriverState *bs);
+void bdrv_io_unplugged_end(BlockDriverState *bs);
 
 /**
  * bdrv_drained_begin:
@@ -541,4 +546,8 @@ void bdrv_drained_begin(BlockDriverState *bs);
  */
 void bdrv_drained_end(BlockDriverState *bs);
 
+void bdrv_add_child(BlockDriverState *parent, BlockDriverState *child,
+                    Error **errp);
+void bdrv_del_child(BlockDriverState *parent, BdrvChild *child, Error **errp);
+
 #endif

include/block/block_int.h

@@ -38,12 +38,12 @@
 #include "qemu/throttle.h"
 
 #define BLOCK_FLAG_ENCRYPT          1
-#define BLOCK_FLAG_COMPAT6          4
 #define BLOCK_FLAG_LAZY_REFCOUNTS   8
 
 #define BLOCK_OPT_SIZE              "size"
 #define BLOCK_OPT_ENCRYPT           "encryption"
 #define BLOCK_OPT_COMPAT6           "compat6"
+#define BLOCK_OPT_HWVERSION         "hwversion"
 #define BLOCK_OPT_BACKING_FILE      "backing_file"
 #define BLOCK_OPT_BACKING_FMT       "backing_fmt"
 #define BLOCK_OPT_CLUSTER_SIZE      "cluster_size"
@@ -127,10 +127,6 @@ struct BlockDriver {
                      Error **errp);
     int (*bdrv_file_open)(BlockDriverState *bs, QDict *options, int flags,
                           Error **errp);
-    int (*bdrv_read)(BlockDriverState *bs, int64_t sector_num,
-                     uint8_t *buf, int nb_sectors);
-    int (*bdrv_write)(BlockDriverState *bs, int64_t sector_num,
-                      const uint8_t *buf, int nb_sectors);
     void (*bdrv_close)(BlockDriverState *bs);
     int (*bdrv_create)(const char *filename, QemuOpts *opts, Error **errp);
     int (*bdrv_set_key)(BlockDriverState *bs, const char *key);
@@ -153,18 +149,20 @@ struct BlockDriver {
 
     int coroutine_fn (*bdrv_co_readv)(BlockDriverState *bs,
         int64_t sector_num, int nb_sectors, QEMUIOVector *qiov);
+    int coroutine_fn (*bdrv_co_preadv)(BlockDriverState *bs,
+        uint64_t offset, uint64_t bytes, QEMUIOVector *qiov, int flags);
     int coroutine_fn (*bdrv_co_writev)(BlockDriverState *bs,
         int64_t sector_num, int nb_sectors, QEMUIOVector *qiov);
     int coroutine_fn (*bdrv_co_writev_flags)(BlockDriverState *bs,
         int64_t sector_num, int nb_sectors, QEMUIOVector *qiov, int flags);
-
+    int coroutine_fn (*bdrv_co_pwritev)(BlockDriverState *bs,
-    int supported_write_flags;
+        uint64_t offset, uint64_t bytes, QEMUIOVector *qiov, int flags);
 
     /*
      * Efficiently zero a region of the disk image.  Typically an image format
      * would use a compact metadata representation to implement this.  This
-     * function pointer may be NULL and .bdrv_co_writev() will be called
+     * function pointer may be NULL or return -ENOSUP and .bdrv_co_writev()
-     * instead.
+     * will be called instead.
      */
     int coroutine_fn (*bdrv_co_write_zeroes)(BlockDriverState *bs,
         int64_t sector_num, int nb_sectors, BdrvRequestFlags flags);
@@ -294,7 +292,6 @@ struct BlockDriver {
     /* io queue for linux-aio */
     void (*bdrv_io_plug)(BlockDriverState *bs);
     void (*bdrv_io_unplug)(BlockDriverState *bs);
-    void (*bdrv_flush_io_queue)(BlockDriverState *bs);
 
     /**
      * Try to get @bs's logical and physical block size.
@@ -317,6 +314,11 @@ struct BlockDriver {
      */
     void (*bdrv_drain)(BlockDriverState *bs);
 
+    void (*bdrv_add_child)(BlockDriverState *parent, BlockDriverState *child,
+                           Error **errp);
+    void (*bdrv_del_child)(BlockDriverState *parent, BdrvChild *child,
+                           Error **errp);
+
     QLIST_ENTRY(BlockDriver) list;
 };
 
@@ -424,10 +426,10 @@ struct BlockDriverState {
 
     /* I/O throttling.
      * throttle_state tells us if this BDS has I/O limits configured.
-     * io_limits_enabled tells us if they are currently being
+     * io_limits_disabled tells us if they are currently being enforced */
-     * enforced, but it can be temporarily set to false */
     CoQueue      throttled_reqs[2];
-    bool         io_limits_enabled;
+    unsigned int io_limits_disabled;
+
     /* The following fields are protected by the ThrottleGroup lock.
      * See the ThrottleGroup documentation for details. */
     ThrottleState *throttle_state;
@@ -446,6 +448,11 @@ struct BlockDriverState {
 
     /* Alignment requirement for offset/length of I/O requests */
     unsigned int request_alignment;
+    /* Flags honored during pwrite (so far: BDRV_REQ_FUA) */
+    unsigned int supported_write_flags;
+    /* Flags honored during write_zeroes (so far: BDRV_REQ_FUA,
+     * BDRV_REQ_MAY_UNMAP) */
+    unsigned int supported_zero_flags;
 
     /* the following member gives a name to every node on the bs graph. */
     char node_name[32];
@@ -484,6 +491,10 @@ struct BlockDriverState {
     uint64_t write_threshold_offset;
     NotifierWithReturn write_threshold_notifier;
 
+    /* counters for nested bdrv_io_plug and bdrv_io_unplugged_begin */
+    unsigned io_plugged;
+    unsigned io_plug_disabled;
+
     int quiesce_counter;
 };
 
@@ -517,10 +528,10 @@ extern BlockDriver bdrv_qcow2;
  */
 void bdrv_setup_io_funcs(BlockDriver *bdrv);
 
-int coroutine_fn bdrv_co_do_preadv(BlockDriverState *bs,
+int coroutine_fn bdrv_co_preadv(BlockDriverState *bs,
     int64_t offset, unsigned int bytes, QEMUIOVector *qiov,
     BdrvRequestFlags flags);
-int coroutine_fn bdrv_co_do_pwritev(BlockDriverState *bs,
+int coroutine_fn bdrv_co_pwritev(BlockDriverState *bs,
     int64_t offset, unsigned int bytes, QEMUIOVector *qiov,
     BdrvRequestFlags flags);
 
@@ -713,6 +724,9 @@ BdrvChild *bdrv_root_attach_child(BlockDriverState *child_bs,
                                   const BdrvChildRole *child_role);
 void bdrv_root_unref_child(BdrvChild *child);
 
+void bdrv_no_throttling_begin(BlockDriverState *bs);
+void bdrv_no_throttling_end(BlockDriverState *bs);
+
 void blk_dev_change_media_cb(BlockBackend *blk, bool load);
 bool blk_dev_has_removable_media(BlockBackend *blk);
 bool blk_dev_has_tray(BlockBackend *blk);

include/block/throttle-groups.h

@@ -38,6 +38,7 @@ void throttle_group_get_config(BlockDriverState *bs, ThrottleConfig *cfg);
 
 void throttle_group_register_bs(BlockDriverState *bs, const char *groupname);
 void throttle_group_unregister_bs(BlockDriverState *bs);
+void throttle_group_restart_bs(BlockDriverState *bs);
 
 void coroutine_fn throttle_group_co_io_limits_intercept(BlockDriverState *bs,
                                                         unsigned int bytes,

include/sysemu/block-backend.h

@@ -1,7 +1,7 @@
 /*
  * QEMU Block backends
  *
- * Copyright (C) 2014 Red Hat, Inc.
+ * Copyright (C) 2014-2016 Red Hat, Inc.
  *
  * Authors:
  *  Markus Armbruster <armbru@redhat.com>,
@@ -90,28 +90,25 @@ void blk_attach_dev_nofail(BlockBackend *blk, void *dev);
 void blk_detach_dev(BlockBackend *blk, void *dev);
 void *blk_get_attached_dev(BlockBackend *blk);
 void blk_set_dev_ops(BlockBackend *blk, const BlockDevOps *ops, void *opaque);
-int blk_read(BlockBackend *blk, int64_t sector_num, uint8_t *buf,
+int blk_pread_unthrottled(BlockBackend *blk, int64_t offset, uint8_t *buf,
-             int nb_sectors);
+                          int count);
-int blk_read_unthrottled(BlockBackend *blk, int64_t sector_num, uint8_t *buf,
+int blk_write_zeroes(BlockBackend *blk, int64_t offset,
-                         int nb_sectors);
+                     int count, BdrvRequestFlags flags);
-int blk_write(BlockBackend *blk, int64_t sector_num, const uint8_t *buf,
+BlockAIOCB *blk_aio_write_zeroes(BlockBackend *blk, int64_t offset,
-              int nb_sectors);
+                                 int count, BdrvRequestFlags flags,
-int blk_write_zeroes(BlockBackend *blk, int64_t sector_num,
-                     int nb_sectors, BdrvRequestFlags flags);
-BlockAIOCB *blk_aio_write_zeroes(BlockBackend *blk, int64_t sector_num,
-                                 int nb_sectors, BdrvRequestFlags flags,
                                  BlockCompletionFunc *cb, void *opaque);
 int blk_pread(BlockBackend *blk, int64_t offset, void *buf, int count);
-int blk_pwrite(BlockBackend *blk, int64_t offset, const void *buf, int count);
+int blk_pwrite(BlockBackend *blk, int64_t offset, const void *buf, int count,
+               BdrvRequestFlags flags);
 int64_t blk_getlength(BlockBackend *blk);
 void blk_get_geometry(BlockBackend *blk, uint64_t *nb_sectors_ptr);
 int64_t blk_nb_sectors(BlockBackend *blk);
-BlockAIOCB *blk_aio_readv(BlockBackend *blk, int64_t sector_num,
+BlockAIOCB *blk_aio_preadv(BlockBackend *blk, int64_t offset,
-                          QEMUIOVector *iov, int nb_sectors,
+                           QEMUIOVector *qiov, BdrvRequestFlags flags,
-                          BlockCompletionFunc *cb, void *opaque);
-BlockAIOCB *blk_aio_writev(BlockBackend *blk, int64_t sector_num,
-                           QEMUIOVector *iov, int nb_sectors,
                            BlockCompletionFunc *cb, void *opaque);
+BlockAIOCB *blk_aio_pwritev(BlockBackend *blk, int64_t offset,
+                            QEMUIOVector *qiov, BdrvRequestFlags flags,
+                            BlockCompletionFunc *cb, void *opaque);
 BlockAIOCB *blk_aio_flush(BlockBackend *blk,
                           BlockCompletionFunc *cb, void *opaque);
 BlockAIOCB *blk_aio_discard(BlockBackend *blk,
@@ -178,8 +175,8 @@ int blk_get_open_flags_from_root_state(BlockBackend *blk);
 
 void *blk_aio_get(const AIOCBInfo *aiocb_info, BlockBackend *blk,
                   BlockCompletionFunc *cb, void *opaque);
-int coroutine_fn blk_co_write_zeroes(BlockBackend *blk, int64_t sector_num,
+int coroutine_fn blk_co_write_zeroes(BlockBackend *blk, int64_t offset,
-                                     int nb_sectors, BdrvRequestFlags flags);
+                                     int count, BdrvRequestFlags flags);
 int blk_write_compressed(BlockBackend *blk, int64_t sector_num,
                          const uint8_t *buf, int nb_sectors);
 int blk_truncate(BlockBackend *blk, int64_t offset);

include/sysemu/dma.h

@@ -197,8 +197,8 @@ void qemu_sglist_add(QEMUSGList *qsg, dma_addr_t base, dma_addr_t len);
 void qemu_sglist_destroy(QEMUSGList *qsg);
 #endif
 
-typedef BlockAIOCB *DMAIOFunc(BlockBackend *blk, int64_t sector_num,
+typedef BlockAIOCB *DMAIOFunc(BlockBackend *blk, int64_t offset,
-                              QEMUIOVector *iov, int nb_sectors,
+                              QEMUIOVector *iov, BdrvRequestFlags flags,
                               BlockCompletionFunc *cb, void *opaque);
 
 BlockAIOCB *dma_blk_io(BlockBackend *blk,

nbd/server.c

@@ -1115,7 +1115,7 @@ static void nbd_trip(void *opaque)
         TRACE("Writing to device");
 
         ret = blk_pwrite(exp->blk, request.from + exp->dev_offset,
-                        req->data, request.len);
+                         req->data, request.len, 0);
         if (ret < 0) {
             LOG("writing to file failed");
             reply.error = -ret;

qapi/block-core.json

@@ -2556,3 +2556,35 @@
 ##
 { 'command': 'block-set-write-threshold',
   'data': { 'node-name': 'str', 'write-threshold': 'uint64' } }
+
+##
+# @x-blockdev-change
+#
+# Dynamically reconfigure the block driver state graph. It can be used
+# to add, remove, insert or replace a graph node. Currently only the
+# Quorum driver implements this feature to add or remove its child. This
+# is useful to fix a broken quorum child.
+#
+# If @node is specified, it will be inserted under @parent. @child
+# may not be specified in this case. If both @parent and @child are
+# specified but @node is not, @child will be detached from @parent.
+#
+# @parent: the id or name of the parent node.
+#
+# @child: #optional the name of a child under the given parent node.
+#
+# @node: #optional the name of the node that will be added.
+#
+# Note: this command is experimental, and its API is not stable. It
+# does not support all kinds of operations, all kinds of children, nor
+# all block drivers.
+#
+# Warning: The data in a new quorum child MUST be consistent with that of
+# the rest of the array.
+#
+# Since: 2.7
+##
+{ 'command': 'x-blockdev-change',
+  'data' : { 'parent': 'str',
+             '*child': 'str',
+             '*node': 'str' } }

qemu-doc.texi

@@ -693,6 +693,9 @@ Supported options:
 File name of a base image (see @option{create} subcommand).
 @item compat6
 Create a VMDK version 6 image (instead of version 4)
+@item hwversion
+Specify vmdk virtual hardware version. Compat6 flag cannot be enabled
+if hwversion is specified.
 @item subformat
 Specifies which VMDK subformat to use. Valid options are
 @code{monolithicSparse} (default),

qemu-img.c

@@ -1088,7 +1088,8 @@ static int check_empty_sectors(BlockBackend *blk, int64_t sect_num,
                                uint8_t *buffer, bool quiet)
 {
     int pnum, ret = 0;
-    ret = blk_read(blk, sect_num, buffer, sect_count);
+    ret = blk_pread(blk, sect_num << BDRV_SECTOR_BITS, buffer,
+                    sect_count << BDRV_SECTOR_BITS);
     if (ret < 0) {
         error_report("Error while reading offset %" PRId64 " of %s: %s",
                      sectors_to_bytes(sect_num), filename, strerror(-ret));
@@ -1301,7 +1302,8 @@ static int img_compare(int argc, char **argv)
             nb_sectors = MIN(pnum1, pnum2);
         } else if (allocated1 == allocated2) {
             if (allocated1) {
-                ret = blk_read(blk1, sector_num, buf1, nb_sectors);
+                ret = blk_pread(blk1, sector_num << BDRV_SECTOR_BITS, buf1,
+                                nb_sectors << BDRV_SECTOR_BITS);
                 if (ret < 0) {
                     error_report("Error while reading offset %" PRId64 " of %s:"
                                  " %s", sectors_to_bytes(sector_num), filename1,
@@ -1309,7 +1311,8 @@ static int img_compare(int argc, char **argv)
                     ret = 4;
                     goto out;
                 }
-                ret = blk_read(blk2, sector_num, buf2, nb_sectors);
+                ret = blk_pread(blk2, sector_num << BDRV_SECTOR_BITS, buf2,
+                                nb_sectors << BDRV_SECTOR_BITS);
                 if (ret < 0) {
                     error_report("Error while reading offset %" PRId64
                                  " of %s: %s", sectors_to_bytes(sector_num),
@@ -1472,10 +1475,21 @@ static int convert_iteration_sectors(ImgConvertState *s, int64_t sector_num)
         } else if (!s->target_has_backing) {
             /* Without a target backing file we must copy over the contents of
              * the backing file as well. */
-            /* TODO Check block status of the backing file chain to avoid
+            /* Check block status of the backing file chain to avoid
              * needlessly reading zeroes and limiting the iteration to the
              * buffer size */
-            s->status = BLK_DATA;
+            ret = bdrv_get_block_status_above(blk_bs(s->src[s->src_cur]), NULL,
+                                              sector_num - s->src_cur_offset,
+                                              n, &n, &file);
+            if (ret < 0) {
+                return ret;
+            }
+
+            if (ret & BDRV_BLOCK_ZERO) {
+                s->status = BLK_ZERO;
+            } else {
+                s->status = BLK_DATA;
+            }
         } else {
             s->status = BLK_BACKING_FILE;
         }
@@ -1522,7 +1536,9 @@ static int convert_read(ImgConvertState *s, int64_t sector_num, int nb_sectors,
         bs_sectors = s->src_sectors[s->src_cur];
 
         n = MIN(nb_sectors, bs_sectors - (sector_num - s->src_cur_offset));
-        ret = blk_read(blk, sector_num - s->src_cur_offset, buf, n);
+        ret = blk_pread(blk,
+                        (sector_num - s->src_cur_offset) << BDRV_SECTOR_BITS,
+                        buf, n << BDRV_SECTOR_BITS);
         if (ret < 0) {
             return ret;
         }
@@ -1577,7 +1593,8 @@ static int convert_write(ImgConvertState *s, int64_t sector_num, int nb_sectors,
             if (!s->min_sparse ||
                 is_allocated_sectors_min(buf, n, &n, s->min_sparse))
             {
-                ret = blk_write(s->target, sector_num, buf, n);
+                ret = blk_pwrite(s->target, sector_num << BDRV_SECTOR_BITS,
+                                 buf, n << BDRV_SECTOR_BITS, 0);
                 if (ret < 0) {
                     return ret;
                 }
@@ -1589,7 +1606,8 @@ static int convert_write(ImgConvertState *s, int64_t sector_num, int nb_sectors,
             if (s->has_zero_init) {
                 break;
             }
-            ret = blk_write_zeroes(s->target, sector_num, n, 0);
+            ret = blk_write_zeroes(s->target, sector_num << BDRV_SECTOR_BITS,
+                                   n << BDRV_SECTOR_BITS, 0);
             if (ret < 0) {
                 return ret;
             }
@@ -3023,7 +3041,8 @@ static int img_rebase(int argc, char **argv)
                     n = old_backing_num_sectors - sector;
                 }
 
-                ret = blk_read(blk_old_backing, sector, buf_old, n);
+                ret = blk_pread(blk_old_backing, sector << BDRV_SECTOR_BITS,
+                                buf_old, n << BDRV_SECTOR_BITS);
                 if (ret < 0) {
                     error_report("error while reading from old backing file");
                     goto out;
@@ -3037,7 +3056,8 @@ static int img_rebase(int argc, char **argv)
                     n = new_backing_num_sectors - sector;
                 }
 
-                ret = blk_read(blk_new_backing, sector, buf_new, n);
+                ret = blk_pread(blk_new_backing, sector << BDRV_SECTOR_BITS,
+                                buf_new, n << BDRV_SECTOR_BITS);
                 if (ret < 0) {
                     error_report("error while reading from new backing file");
                     goto out;
@@ -3053,8 +3073,10 @@ static int img_rebase(int argc, char **argv)
                 if (compare_sectors(buf_old + written * 512,
                     buf_new + written * 512, n - written, &pnum))
                 {
-                    ret = blk_write(blk, sector + written,
+                    ret = blk_pwrite(blk,
-                                    buf_old + written * 512, pnum);
+                                     (sector + written) << BDRV_SECTOR_BITS,
+                                     buf_old + written * 512,
+                                     pnum << BDRV_SECTOR_BITS, 0);
                     if (ret < 0) {
                         error_report("Error while writing to COW image: %s",
                             strerror(-ret));

qemu-io-cmds.c

@@ -1,7 +1,7 @@
 /*
  * Command line utility to exercise the QEMU I/O path.
  *
- * Copyright (C) 2009 Red Hat, Inc.
+ * Copyright (C) 2009-2016 Red Hat, Inc.
  * Copyright (c) 2003-2005 Silicon Graphics, Inc.
  *
  * This work is licensed under the terms of the GNU GPL, version 2 or later.
@@ -345,7 +345,7 @@ static void dump_buffer(const void *buffer, int64_t offset, int64_t len)
 }
 
 static void print_report(const char *op, struct timeval *t, int64_t offset,
-                         int64_t count, int64_t total, int cnt, int Cflag)
+                         int64_t count, int64_t total, int cnt, bool Cflag)
 {
     char s1[64], s2[64], ts[64];
 
@@ -395,12 +395,6 @@ create_iovec(BlockBackend *blk, QEMUIOVector *qiov, char **argv, int nr_iov,
             goto fail;
         }
 
-        if (len & 0x1ff) {
-            printf("length argument %" PRId64
-                   " is not sector aligned\n", len);
-            goto fail;
-        }
-
         sizes[i] = len;
         count += len;
     }
@@ -419,40 +413,6 @@ fail:
     return buf;
 }
 
-static int do_read(BlockBackend *blk, char *buf, int64_t offset, int64_t count,
-                   int64_t *total)
-{
-    int ret;
-
-    if (count >> 9 > INT_MAX) {
-        return -ERANGE;
-    }
-
-    ret = blk_read(blk, offset >> 9, (uint8_t *)buf, count >> 9);
-    if (ret < 0) {
-        return ret;
-    }
-    *total = count;
-    return 1;
-}
-
-static int do_write(BlockBackend *blk, char *buf, int64_t offset, int64_t count,
-                    int64_t *total)
-{
-    int ret;
-
-    if (count >> 9 > INT_MAX) {
-        return -ERANGE;
-    }
-
-    ret = blk_write(blk, offset >> 9, (uint8_t *)buf, count >> 9);
-    if (ret < 0) {
-        return ret;
-    }
-    *total = count;
-    return 1;
-}
-
 static int do_pread(BlockBackend *blk, char *buf, int64_t offset,
                     int64_t count, int64_t *total)
 {
@@ -468,13 +428,13 @@ static int do_pread(BlockBackend *blk, char *buf, int64_t offset,
 }
 
 static int do_pwrite(BlockBackend *blk, char *buf, int64_t offset,
-                     int64_t count, int64_t *total)
+                     int64_t count, int flags, int64_t *total)
 {
     if (count > INT_MAX) {
         return -ERANGE;
     }
 
-    *total = blk_pwrite(blk, offset, (uint8_t *)buf, count);
+    *total = blk_pwrite(blk, offset, (uint8_t *)buf, count, flags);
     if (*total < 0) {
         return *total;
     }
@@ -486,6 +446,7 @@ typedef struct {
     int64_t offset;
     int64_t count;
     int64_t *total;
+    int flags;
     int ret;
     bool done;
 } CoWriteZeroes;
@@ -494,8 +455,8 @@ static void coroutine_fn co_write_zeroes_entry(void *opaque)
 {
     CoWriteZeroes *data = opaque;
 
-    data->ret = blk_co_write_zeroes(data->blk, data->offset / BDRV_SECTOR_SIZE,
+    data->ret = blk_co_write_zeroes(data->blk, data->offset, data->count,
-                                    data->count / BDRV_SECTOR_SIZE, 0);
+                                    data->flags);
     data->done = true;
     if (data->ret < 0) {
         *data->total = data->ret;
@@ -506,7 +467,7 @@ static void coroutine_fn co_write_zeroes_entry(void *opaque)
 }
 
 static int do_co_write_zeroes(BlockBackend *blk, int64_t offset, int64_t count,
-                              int64_t *total)
+                              int flags, int64_t *total)
 {
     Coroutine *co;
     CoWriteZeroes data = {
@@ -514,6 +475,7 @@ static int do_co_write_zeroes(BlockBackend *blk, int64_t offset, int64_t count,
         .offset = offset,
         .count  = count,
         .total  = total,
+        .flags  = flags,
         .done   = false,
     };
 
@@ -589,8 +551,7 @@ static int do_aio_readv(BlockBackend *blk, QEMUIOVector *qiov,
 {
     int async_ret = NOT_DONE;
 
-    blk_aio_readv(blk, offset >> 9, qiov, qiov->size >> 9,
+    blk_aio_preadv(blk, offset, qiov, 0, aio_rw_done, &async_ret);
-                  aio_rw_done, &async_ret);
     while (async_ret == NOT_DONE) {
         main_loop_wait(false);
     }
@@ -600,12 +561,11 @@ static int do_aio_readv(BlockBackend *blk, QEMUIOVector *qiov,
 }
 
 static int do_aio_writev(BlockBackend *blk, QEMUIOVector *qiov,
-                         int64_t offset, int *total)
+                         int64_t offset, int flags, int *total)
 {
     int async_ret = NOT_DONE;
 
-    blk_aio_writev(blk, offset >> 9, qiov, qiov->size >> 9,
+    blk_aio_pwritev(blk, offset, qiov, flags, aio_rw_done, &async_ret);
-                   aio_rw_done, &async_ret);
     while (async_ret == NOT_DONE) {
         main_loop_wait(false);
     }
@@ -671,7 +631,7 @@ static void read_help(void)
 " -b, -- read from the VM state rather than the virtual disk\n"
 " -C, -- report statistics in a machine parsable format\n"
 " -l, -- length for pattern verification (only with -P)\n"
-" -p, -- use blk_pread to read the file\n"
+" -p, -- ignored for backwards compatibility\n"
 " -P, -- use a pattern to verify read data\n"
 " -q, -- quiet mode, do not show I/O statistics\n"
 " -s, -- start offset for pattern verification (only with -P)\n"
@@ -687,7 +647,7 @@ static const cmdinfo_t read_cmd = {
     .cfunc      = read_f,
     .argmin     = 2,
     .argmax     = -1,
-    .args       = "[-abCpqv] [-P pattern [-s off] [-l len]] off len",
+    .args       = "[-abCqv] [-P pattern [-s off] [-l len]] off len",
     .oneline    = "reads a number of bytes at a specified offset",
     .help       = read_help,
 };
@@ -695,8 +655,8 @@ static const cmdinfo_t read_cmd = {
 static int read_f(BlockBackend *blk, int argc, char **argv)
 {
     struct timeval t1, t2;
-    int Cflag = 0, pflag = 0, qflag = 0, vflag = 0;
+    bool Cflag = false, qflag = false, vflag = false;
-    int Pflag = 0, sflag = 0, lflag = 0, bflag = 0;
+    bool Pflag = false, sflag = false, lflag = false, bflag = false;
     int c, cnt;
     char *buf;
     int64_t offset;
@@ -709,13 +669,13 @@ static int read_f(BlockBackend *blk, int argc, char **argv)
     while ((c = getopt(argc, argv, "bCl:pP:qs:v")) != -1) {
         switch (c) {
         case 'b':
-            bflag = 1;
+            bflag = true;
             break;
         case 'C':
-            Cflag = 1;
+            Cflag = true;
             break;
         case 'l':
-            lflag = 1;
+            lflag = true;
             pattern_count = cvtnum(optarg);
             if (pattern_count < 0) {
                 print_cvtnum_err(pattern_count, optarg);
@@ -723,20 +683,20 @@ static int read_f(BlockBackend *blk, int argc, char **argv)
             }
             break;
         case 'p':
-            pflag = 1;
+            /* Ignored for backwards compatibility */
             break;
         case 'P':
-            Pflag = 1;
+            Pflag = true;
             pattern = parse_pattern(optarg);
             if (pattern < 0) {
                 return 0;
             }
             break;
         case 'q':
-            qflag = 1;
+            qflag = true;
             break;
         case 's':
-            sflag = 1;
+            sflag = true;
             pattern_offset = cvtnum(optarg);
             if (pattern_offset < 0) {
                 print_cvtnum_err(pattern_offset, optarg);
@@ -744,7 +704,7 @@ static int read_f(BlockBackend *blk, int argc, char **argv)
             }
             break;
         case 'v':
-            vflag = 1;
+            vflag = true;
             break;
         default:
             return qemuio_command_usage(&read_cmd);
@@ -755,11 +715,6 @@ static int read_f(BlockBackend *blk, int argc, char **argv)
         return qemuio_command_usage(&read_cmd);
     }
 
-    if (bflag && pflag) {
-        printf("-b and -p cannot be specified at the same time\n");
-        return 0;
-    }
-
     offset = cvtnum(argv[optind]);
     if (offset < 0) {
         print_cvtnum_err(offset, argv[optind]);
@@ -790,7 +745,7 @@ static int read_f(BlockBackend *blk, int argc, char **argv)
         return 0;
     }
 
-    if (!pflag) {
+    if (bflag) {
         if (offset & 0x1ff) {
             printf("offset %" PRId64 " is not sector aligned\n",
                    offset);
@@ -806,12 +761,10 @@ static int read_f(BlockBackend *blk, int argc, char **argv)
     buf = qemu_io_alloc(blk, count, 0xab);
 
     gettimeofday(&t1, NULL);
-    if (pflag) {
+    if (bflag) {
-        cnt = do_pread(blk, buf, offset, count, &total);
-    } else if (bflag) {
         cnt = do_load_vmstate(blk, buf, offset, count, &total);
     } else {
-        cnt = do_read(blk, buf, offset, count, &total);
+        cnt = do_pread(blk, buf, offset, count, &total);
     }
     gettimeofday(&t2, NULL);
 
@@ -875,7 +828,7 @@ static const cmdinfo_t readv_cmd = {
     .cfunc      = readv_f,
     .argmin     = 2,
     .argmax     = -1,
-    .args       = "[-Cqv] [-P pattern ] off len [len..]",
+    .args       = "[-Cqv] [-P pattern] off len [len..]",
     .oneline    = "reads a number of bytes at a specified offset",
     .help       = readv_help,
 };
@@ -883,7 +836,7 @@ static const cmdinfo_t readv_cmd = {
 static int readv_f(BlockBackend *blk, int argc, char **argv)
 {
     struct timeval t1, t2;
-    int Cflag = 0, qflag = 0, vflag = 0;
+    bool Cflag = false, qflag = false, vflag = false;
     int c, cnt;
     char *buf;
     int64_t offset;
@@ -892,25 +845,25 @@ static int readv_f(BlockBackend *blk, int argc, char **argv)
     int nr_iov;
     QEMUIOVector qiov;
     int pattern = 0;
-    int Pflag = 0;
+    bool Pflag = false;
 
     while ((c = getopt(argc, argv, "CP:qv")) != -1) {
         switch (c) {
         case 'C':
-            Cflag = 1;
+            Cflag = true;
             break;
         case 'P':
-            Pflag = 1;
+            Pflag = true;
             pattern = parse_pattern(optarg);
             if (pattern < 0) {
                 return 0;
             }
             break;
         case 'q':
-            qflag = 1;
+            qflag = true;
             break;
         case 'v':
-            vflag = 1;
+            vflag = true;
             break;
         default:
             return qemuio_command_usage(&readv_cmd);
@@ -929,12 +882,6 @@ static int readv_f(BlockBackend *blk, int argc, char **argv)
     }
     optind++;
 
-    if (offset & 0x1ff) {
-        printf("offset %" PRId64 " is not sector aligned\n",
-               offset);
-        return 0;
-    }
-
     nr_iov = argc - optind;
     buf = create_iovec(blk, &qiov, &argv[optind], nr_iov, 0xab);
     if (buf == NULL) {
@@ -991,10 +938,12 @@ static void write_help(void)
 " filled with a set pattern (0xcdcdcdcd).\n"
 " -b, -- write to the VM state rather than the virtual disk\n"
 " -c, -- write compressed data with blk_write_compressed\n"
-" -p, -- use blk_pwrite to write the file\n"
+" -f, -- use Force Unit Access semantics\n"
+" -p, -- ignored for backwards compatibility\n"
 " -P, -- use different pattern to fill file\n"
 " -C, -- report statistics in a machine parsable format\n"
 " -q, -- quiet mode, do not show I/O statistics\n"
+" -u, -- with -z, allow unmapping\n"
 " -z, -- write zeroes using blk_co_write_zeroes\n"
 "\n");
 }
@@ -1007,7 +956,7 @@ static const cmdinfo_t write_cmd = {
     .cfunc      = write_f,
     .argmin     = 2,
     .argmax     = -1,
-    .args       = "[-bcCpqz] [-P pattern ] off len",
+    .args       = "[-bcCfquz] [-P pattern] off len",
     .oneline    = "writes a number of bytes at a specified offset",
     .help       = write_help,
 };
@@ -1015,8 +964,9 @@ static const cmdinfo_t write_cmd = {
 static int write_f(BlockBackend *blk, int argc, char **argv)
 {
     struct timeval t1, t2;
-    int Cflag = 0, pflag = 0, qflag = 0, bflag = 0, Pflag = 0, zflag = 0;
+    bool Cflag = false, qflag = false, bflag = false;
-    int cflag = 0;
+    bool Pflag = false, zflag = false, cflag = false;
+    int flags = 0;
     int c, cnt;
     char *buf = NULL;
     int64_t offset;
@@ -1025,32 +975,38 @@ static int write_f(BlockBackend *blk, int argc, char **argv)
     int64_t total = 0;
     int pattern = 0xcd;
 
-    while ((c = getopt(argc, argv, "bcCpP:qz")) != -1) {
+    while ((c = getopt(argc, argv, "bcCfpP:quz")) != -1) {
         switch (c) {
         case 'b':
-            bflag = 1;
+            bflag = true;
             break;
         case 'c':
-            cflag = 1;
+            cflag = true;
             break;
         case 'C':
-            Cflag = 1;
+            Cflag = true;
+            break;
+        case 'f':
+            flags |= BDRV_REQ_FUA;
             break;
         case 'p':
-            pflag = 1;
+            /* Ignored for backwards compatibility */
             break;
         case 'P':
-            Pflag = 1;
+            Pflag = true;
             pattern = parse_pattern(optarg);
             if (pattern < 0) {
                 return 0;
             }
             break;
         case 'q':
-            qflag = 1;
+            qflag = true;
+            break;
+        case 'u':
+            flags |= BDRV_REQ_MAY_UNMAP;
             break;
         case 'z':
-            zflag = 1;
+            zflag = true;
             break;
         default:
             return qemuio_command_usage(&write_cmd);
@@ -1061,8 +1017,18 @@ static int write_f(BlockBackend *blk, int argc, char **argv)
         return qemuio_command_usage(&write_cmd);
     }
 
-    if (bflag + pflag + zflag > 1) {
+    if (bflag && zflag) {
-        printf("-b, -p, or -z cannot be specified at the same time\n");
+        printf("-b and -z cannot be specified at the same time\n");
+        return 0;
+    }
+
+    if ((flags & BDRV_REQ_FUA) && (bflag || cflag)) {
+        printf("-f and -b or -c cannot be specified at the same time\n");
+        return 0;
+    }
+
+    if ((flags & BDRV_REQ_MAY_UNMAP) && !zflag) {
+        printf("-u requires -z to be specified\n");
         return 0;
     }
 
@@ -1088,7 +1054,7 @@ static int write_f(BlockBackend *blk, int argc, char **argv)
         return 0;
     }
 
-    if (!pflag) {
+    if (bflag || cflag) {
         if (offset & 0x1ff) {
             printf("offset %" PRId64 " is not sector aligned\n",
                    offset);
@@ -1107,16 +1073,14 @@ static int write_f(BlockBackend *blk, int argc, char **argv)
     }
 
     gettimeofday(&t1, NULL);
-    if (pflag) {
+    if (bflag) {
-        cnt = do_pwrite(blk, buf, offset, count, &total);
-    } else if (bflag) {
         cnt = do_save_vmstate(blk, buf, offset, count, &total);
     } else if (zflag) {
-        cnt = do_co_write_zeroes(blk, offset, count, &total);
+        cnt = do_co_write_zeroes(blk, offset, count, flags, &total);
     } else if (cflag) {
         cnt = do_write_compressed(blk, buf, offset, count, &total);
     } else {
-        cnt = do_write(blk, buf, offset, count, &total);
+        cnt = do_pwrite(blk, buf, offset, count, flags, &total);
     }
     gettimeofday(&t2, NULL);
 
@@ -1155,6 +1119,7 @@ writev_help(void)
 " filled with a set pattern (0xcdcdcdcd).\n"
 " -P, -- use different pattern to fill file\n"
 " -C, -- report statistics in a machine parsable format\n"
+" -f, -- use Force Unit Access semantics\n"
 " -q, -- quiet mode, do not show I/O statistics\n"
 "\n");
 }
@@ -1166,7 +1131,7 @@ static const cmdinfo_t writev_cmd = {
     .cfunc      = writev_f,
     .argmin     = 2,
     .argmax     = -1,
-    .args       = "[-Cq] [-P pattern ] off len [len..]",
+    .args       = "[-Cfq] [-P pattern] off len [len..]",
     .oneline    = "writes a number of bytes at a specified offset",
     .help       = writev_help,
 };
@@ -1174,7 +1139,8 @@ static const cmdinfo_t writev_cmd = {
 static int writev_f(BlockBackend *blk, int argc, char **argv)
 {
     struct timeval t1, t2;
-    int Cflag = 0, qflag = 0;
+    bool Cflag = false, qflag = false;
+    int flags = 0;
     int c, cnt;
     char *buf;
     int64_t offset;
@@ -1187,10 +1153,13 @@ static int writev_f(BlockBackend *blk, int argc, char **argv)
     while ((c = getopt(argc, argv, "CqP:")) != -1) {
         switch (c) {
         case 'C':
-            Cflag = 1;
+            Cflag = true;
+            break;
+        case 'f':
+            flags |= BDRV_REQ_FUA;
             break;
         case 'q':
-            qflag = 1;
+            qflag = true;
             break;
         case 'P':
             pattern = parse_pattern(optarg);
@@ -1214,12 +1183,6 @@ static int writev_f(BlockBackend *blk, int argc, char **argv)
     }
     optind++;
 
-    if (offset & 0x1ff) {
-        printf("offset %" PRId64 " is not sector aligned\n",
-               offset);
-        return 0;
-    }
-
     nr_iov = argc - optind;
     buf = create_iovec(blk, &qiov, &argv[optind], nr_iov, pattern);
     if (buf == NULL) {
@@ -1227,7 +1190,7 @@ static int writev_f(BlockBackend *blk, int argc, char **argv)
     }
 
     gettimeofday(&t1, NULL);
-    cnt = do_aio_writev(blk, &qiov, offset, &total);
+    cnt = do_aio_writev(blk, &qiov, offset, flags, &total);
     gettimeofday(&t2, NULL);
 
     if (cnt < 0) {
@@ -1283,7 +1246,7 @@ static const cmdinfo_t multiwrite_cmd = {
 static int multiwrite_f(BlockBackend *blk, int argc, char **argv)
 {
     struct timeval t1, t2;
-    int Cflag = 0, qflag = 0;
+    bool Cflag = false, qflag = false;
     int c, cnt;
     char **buf;
     int64_t offset, first_offset = 0;
@@ -1299,10 +1262,10 @@ static int multiwrite_f(BlockBackend *blk, int argc, char **argv)
     while ((c = getopt(argc, argv, "CqP:")) != -1) {
         switch (c) {
         case 'C':
-            Cflag = 1;
+            Cflag = true;
             break;
         case 'q':
-            qflag = 1;
+            qflag = true;
             break;
         case 'P':
             pattern = parse_pattern(optarg);
@@ -1412,11 +1375,11 @@ struct aio_ctx {
     QEMUIOVector qiov;
     int64_t offset;
     char *buf;
-    int qflag;
+    bool qflag;
-    int vflag;
+    bool vflag;
-    int Cflag;
+    bool Cflag;
-    int Pflag;
+    bool Pflag;
-    int zflag;
+    bool zflag;
     BlockAcctCookie acct;
     int pattern;
     struct timeval t1;
@@ -1525,7 +1488,7 @@ static const cmdinfo_t aio_read_cmd = {
     .cfunc      = aio_read_f,
     .argmin     = 2,
     .argmax     = -1,
-    .args       = "[-Cqv] [-P pattern ] off len [len..]",
+    .args       = "[-Cqv] [-P pattern] off len [len..]",
     .oneline    = "asynchronously reads a number of bytes",
     .help       = aio_read_help,
 };
@@ -1539,10 +1502,10 @@ static int aio_read_f(BlockBackend *blk, int argc, char **argv)
     while ((c = getopt(argc, argv, "CP:qv")) != -1) {
         switch (c) {
         case 'C':
-            ctx->Cflag = 1;
+            ctx->Cflag = true;
             break;
         case 'P':
-            ctx->Pflag = 1;
+            ctx->Pflag = true;
             ctx->pattern = parse_pattern(optarg);
             if (ctx->pattern < 0) {
                 g_free(ctx);
@@ -1550,10 +1513,10 @@ static int aio_read_f(BlockBackend *blk, int argc, char **argv)
             }
             break;
         case 'q':
-            ctx->qflag = 1;
+            ctx->qflag = true;
             break;
         case 'v':
-            ctx->vflag = 1;
+            ctx->vflag = true;
             break;
         default:
             g_free(ctx);
@@ -1574,14 +1537,6 @@ static int aio_read_f(BlockBackend *blk, int argc, char **argv)
     }
     optind++;
 
-    if (ctx->offset & 0x1ff) {
-        printf("offset %" PRId64 " is not sector aligned\n",
-               ctx->offset);
-        block_acct_invalid(blk_get_stats(blk), BLOCK_ACCT_READ);
-        g_free(ctx);
-        return 0;
-    }
-
     nr_iov = argc - optind;
     ctx->buf = create_iovec(blk, &ctx->qiov, &argv[optind], nr_iov, 0xab);
     if (ctx->buf == NULL) {
@@ -1593,8 +1548,7 @@ static int aio_read_f(BlockBackend *blk, int argc, char **argv)
     gettimeofday(&ctx->t1, NULL);
     block_acct_start(blk_get_stats(blk), &ctx->acct, ctx->qiov.size,
                      BLOCK_ACCT_READ);
-    blk_aio_readv(blk, ctx->offset >> 9, &ctx->qiov,
+    blk_aio_preadv(blk, ctx->offset, &ctx->qiov, 0, aio_read_done, ctx);
-                  ctx->qiov.size >> 9, aio_read_done, ctx);
     return 0;
 }
 
@@ -1614,7 +1568,9 @@ static void aio_write_help(void)
 " used to ensure all outstanding aio requests have been completed.\n"
 " -P, -- use different pattern to fill file\n"
 " -C, -- report statistics in a machine parsable format\n"
+" -f, -- use Force Unit Access semantics\n"
 " -q, -- quiet mode, do not show I/O statistics\n"
+" -u, -- with -z, allow unmapping\n"
 " -z, -- write zeroes using blk_aio_write_zeroes\n"
 "\n");
 }
@@ -1626,7 +1582,7 @@ static const cmdinfo_t aio_write_cmd = {
     .cfunc      = aio_write_f,
     .argmin     = 2,
     .argmax     = -1,
-    .args       = "[-Cqz] [-P pattern ] off len [len..]",
+    .args       = "[-Cfquz] [-P pattern] off len [len..]",
     .oneline    = "asynchronously writes a number of bytes",
     .help       = aio_write_help,
 };
@@ -1636,15 +1592,22 @@ static int aio_write_f(BlockBackend *blk, int argc, char **argv)
     int nr_iov, c;
     int pattern = 0xcd;
     struct aio_ctx *ctx = g_new0(struct aio_ctx, 1);
+    int flags = 0;
 
     ctx->blk = blk;
-    while ((c = getopt(argc, argv, "CqP:z")) != -1) {
+    while ((c = getopt(argc, argv, "CfqP:z")) != -1) {
         switch (c) {
         case 'C':
-            ctx->Cflag = 1;
+            ctx->Cflag = true;
+            break;
+        case 'f':
+            flags |= BDRV_REQ_FUA;
             break;
         case 'q':
-            ctx->qflag = 1;
+            ctx->qflag = true;
+            break;
+        case 'u':
+            flags |= BDRV_REQ_MAY_UNMAP;
             break;
         case 'P':
             pattern = parse_pattern(optarg);
@@ -1654,7 +1617,7 @@ static int aio_write_f(BlockBackend *blk, int argc, char **argv)
             }
             break;
         case 'z':
-            ctx->zflag = 1;
+            ctx->zflag = true;
             break;
         default:
             g_free(ctx);
@@ -1673,6 +1636,11 @@ static int aio_write_f(BlockBackend *blk, int argc, char **argv)
         return 0;
     }
 
+    if ((flags & BDRV_REQ_MAY_UNMAP) && !ctx->zflag) {
+        printf("-u requires -z to be specified\n");
+        return 0;
+    }
+
     if (ctx->zflag && ctx->Pflag) {
         printf("-z and -P cannot be specified at the same time\n");
         g_free(ctx);
@@ -1687,24 +1655,17 @@ static int aio_write_f(BlockBackend *blk, int argc, char **argv)
     }
     optind++;
 
-    if (ctx->offset & 0x1ff) {
-        printf("offset %" PRId64 " is not sector aligned\n",
-               ctx->offset);
-        block_acct_invalid(blk_get_stats(blk), BLOCK_ACCT_WRITE);
-        g_free(ctx);
-        return 0;
-    }
-
     if (ctx->zflag) {
         int64_t count = cvtnum(argv[optind]);
         if (count < 0) {
             print_cvtnum_err(count, argv[optind]);
+            g_free(ctx);
             return 0;
         }
 
         ctx->qiov.size = count;
-        blk_aio_write_zeroes(blk, ctx->offset >> 9, count >> 9, 0,
+        blk_aio_write_zeroes(blk, ctx->offset, count, flags, aio_write_done,
-                             aio_write_done, ctx);
+                             ctx);
     } else {
         nr_iov = argc - optind;
         ctx->buf = create_iovec(blk, &ctx->qiov, &argv[optind], nr_iov,
@@ -1719,8 +1680,8 @@ static int aio_write_f(BlockBackend *blk, int argc, char **argv)
         block_acct_start(blk_get_stats(blk), &ctx->acct, ctx->qiov.size,
                          BLOCK_ACCT_WRITE);
 
-        blk_aio_writev(blk, ctx->offset >> 9, &ctx->qiov,
+        blk_aio_pwritev(blk, ctx->offset, &ctx->qiov, flags, aio_write_done,
-                       ctx->qiov.size >> 9, aio_write_done, ctx);
+                        ctx);
     }
     return 0;
 }
@@ -1884,17 +1845,17 @@ static const cmdinfo_t discard_cmd = {
 static int discard_f(BlockBackend *blk, int argc, char **argv)
 {
     struct timeval t1, t2;
-    int Cflag = 0, qflag = 0;
+    bool Cflag = false, qflag = false;
     int c, ret;
     int64_t offset, count;
 
     while ((c = getopt(argc, argv, "Cq")) != -1) {
         switch (c) {
         case 'C':
-            Cflag = 1;
+            Cflag = true;
             break;
         case 'q':
-            qflag = 1;
+            qflag = true;
             break;
         default:
             return qemuio_command_usage(&discard_cmd);

qemu-io.c

@@ -101,12 +101,15 @@ static void open_help(void)
 " opens a new file in the requested mode\n"
 "\n"
 " Example:\n"
-" 'open -Cn /tmp/data' - creates/opens data file read-write and uncached\n"
+" 'open -n -o driver=raw /tmp/data' - opens raw data file read-write, uncached\n"
 "\n"
 " Opens a file for subsequent use by all of the other qemu-io commands.\n"
 " -r, -- open file read-only\n"
 " -s, -- use snapshot file\n"
-" -n, -- disable host cache\n"
+" -n, -- disable host cache, short for -t none\n"
+" -k, -- use kernel AIO implementation (on Linux only)\n"
+" -t, -- use the given cache mode for the image\n"
+" -d, -- use the given discard mode for the image\n"
 " -o, -- options to be given to the block driver"
 "\n");
 }
@@ -120,7 +123,7 @@ static const cmdinfo_t open_cmd = {
     .argmin     = 1,
     .argmax     = -1,
     .flags      = CMD_NOFILE_OK,
-    .args       = "[-Crsn] [-o options] [path]",
+    .args       = "[-rsnk] [-t cache] [-d discard] [-o options] [path]",
     .oneline    = "open the file specified by path",
     .help       = open_help,
 };
@@ -137,14 +140,14 @@ static QemuOptsList empty_opts = {
 
 static int open_f(BlockBackend *blk, int argc, char **argv)
 {
-    int flags = 0;
+    int flags = BDRV_O_UNMAP;
     int readonly = 0;
     bool writethrough = true;
     int c;
     QemuOpts *qopts;
     QDict *opts;
 
-    while ((c = getopt(argc, argv, "snrgo:")) != -1) {
+    while ((c = getopt(argc, argv, "snro:kt:d:")) != -1) {
         switch (c) {
         case 's':
             flags |= BDRV_O_SNAPSHOT;
@@ -156,9 +159,27 @@ static int open_f(BlockBackend *blk, int argc, char **argv)
         case 'r':
             readonly = 1;
             break;
+        case 'k':
+            flags |= BDRV_O_NATIVE_AIO;
+            break;
+        case 't':
+            if (bdrv_parse_cache_mode(optarg, &flags, &writethrough) < 0) {
+                error_report("Invalid cache option: %s", optarg);
+                qemu_opts_reset(&empty_opts);
+                return 0;
+            }
+            break;
+        case 'd':
+            if (bdrv_parse_discard_flags(optarg, &flags) < 0) {
+                error_report("Invalid discard option: %s", optarg);
+                qemu_opts_reset(&empty_opts);
+                return 0;
+            }
+            break;
         case 'o':
             if (imageOpts) {
                 printf("--image-opts and 'open -o' are mutually exclusive\n");
+                qemu_opts_reset(&empty_opts);
                 return 0;
             }
             if (!qemu_opts_parse_noisily(&empty_opts, optarg, false)) {
@@ -216,20 +237,22 @@ static const cmdinfo_t quit_cmd = {
 static void usage(const char *name)
 {
     printf(
-"Usage: %s [-h] [-V] [-rsnm] [-f FMT] [-c STRING] ... [file]\n"
+"Usage: %s [OPTIONS]... [-c STRING]... [file]\n"
 "QEMU Disk exerciser\n"
 "\n"
 "  --object OBJECTDEF   define an object such as 'secret' for\n"
 "                       passwords and/or encryption keys\n"
+"  --image-opts         treat file as option string\n"
 "  -c, --cmd STRING     execute command with its arguments\n"
 "                       from the given string\n"
 "  -f, --format FMT     specifies the block driver to use\n"
 "  -r, --read-only      export read-only\n"
 "  -s, --snapshot       use snapshot file\n"
-"  -n, --nocache        disable host cache\n"
+"  -n, --nocache        disable host cache, short for -t none\n"
 "  -m, --misalign       misalign allocations for O_DIRECT\n"
 "  -k, --native-aio     use kernel AIO implementation (on Linux only)\n"
 "  -t, --cache=MODE     use the given cache mode for the image\n"
+"  -d, --discard=MODE   use the given discard mode for the image\n"
 "  -T, --trace FILE     enable trace events listed in the given file\n"
 "  -h, --help           display this help and exit\n"
 "  -V, --version        output version information and exit\n"
@@ -410,11 +433,10 @@ static QemuOptsList file_opts = {
 int main(int argc, char **argv)
 {
     int readonly = 0;
-    const char *sopt = "hVc:d:f:rsnmgkt:T:";
+    const char *sopt = "hVc:d:f:rsnmkt:T:";
     const struct option lopt[] = {
         { "help", no_argument, NULL, 'h' },
         { "version", no_argument, NULL, 'V' },
-        { "offset", required_argument, NULL, 'o' },
         { "cmd", required_argument, NULL, 'c' },
         { "format", required_argument, NULL, 'f' },
         { "read-only", no_argument, NULL, 'r' },

qemu-nbd.c

@@ -46,6 +46,8 @@
 #define QEMU_NBD_OPT_TLSCREDS      261
 #define QEMU_NBD_OPT_IMAGE_OPTS    262
 
+#define MBR_SIZE 512
+
 static NBDExport *exp;
 static bool newproto;
 static int verbose;
@@ -159,12 +161,13 @@ static int find_partition(BlockBackend *blk, int partition,
                           off_t *offset, off_t *size)
 {
     struct partition_record mbr[4];
-    uint8_t data[512];
+    uint8_t data[MBR_SIZE];
     int i;
     int ext_partnum = 4;
     int ret;
 
-    if ((ret = blk_read(blk, 0, data, 1)) < 0) {
+    ret = blk_pread(blk, 0, data, sizeof(data));
+    if (ret < 0) {
         error_report("error while reading: %s", strerror(-ret));
         exit(EXIT_FAILURE);
     }
@@ -182,10 +185,12 @@ static int find_partition(BlockBackend *blk, int partition,
 
         if (mbr[i].system == 0xF || mbr[i].system == 0x5) {
             struct partition_record ext[4];
-            uint8_t data1[512];
+            uint8_t data1[MBR_SIZE];
             int j;
 
-            if ((ret = blk_read(blk, mbr[i].start_sector_abs, data1, 1)) < 0) {
+            ret = blk_pread(blk, mbr[i].start_sector_abs * MBR_SIZE,
+                            data1, sizeof(data1));
+            if (ret < 0) {
                 error_report("error while reading: %s", strerror(-ret));
                 exit(EXIT_FAILURE);
             }

qmp-commands.hx

@@ -4395,6 +4395,59 @@ Example:
 
 <- { "return": {} }
 
+EQMP
+
+    {
+        .name       = "x-blockdev-change",
+        .args_type  = "parent:B,child:B?,node:B?",
+        .mhandler.cmd_new = qmp_marshal_x_blockdev_change,
+    },
+
+SQMP
+x-blockdev-change
+-----------------
+
+Dynamically reconfigure the block driver state graph. It can be used
+to add, remove, insert or replace a graph node. Currently only the
+Quorum driver implements this feature to add or remove its child. This
+is useful to fix a broken quorum child.
+
+If @node is specified, it will be inserted under @parent. @child
+may not be specified in this case. If both @parent and @child are
+specified but @node is not, @child will be detached from @parent.
+
+Arguments:
+- "parent": the id or name of the parent node (json-string)
+- "child": the name of a child under the given parent node (json-string, optional)
+- "node": the name of the node that will be added (json-string, optional)
+
+Note: this command is experimental, and not a stable API. It doesn't
+support all kinds of operations, all kinds of children, nor all block
+drivers.
+
+Warning: The data in a new quorum child MUST be consistent with that of
+the rest of the array.
+
+Example:
+
+Add a new node to a quorum
+-> { "execute": "blockdev-add",
+     "arguments": { "options": { "driver": "raw",
+                                 "node-name": "new_node",
+                                 "file": { "driver": "file",
+                                           "filename": "test.raw" } } } }
+<- { "return": {} }
+-> { "execute": "x-blockdev-change",
+     "arguments": { "parent": "disk1",
+                    "node": "new_node" } }
+<- { "return": {} }
+
+Delete a quorum's node
+-> { "execute": "x-blockdev-change",
+     "arguments": { "parent": "disk1",
+                    "child": "children.1" } }
+<- { "return": {} }
+
 EQMP
 
     {

tests/qemu-iotests/004

@@ -37,7 +37,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15
 . ./common.rc
 . ./common.filter
 
-_supported_fmt raw qcow qcow2 qed vdi vmdk vhdx
+_supported_fmt raw qcow qcow2 qed vdi vmdk vhdx luks
 _supported_proto generic
 _supported_os Linux
 

tests/qemu-iotests/012

@@ -43,13 +43,16 @@ _supported_fmt generic
 _supported_proto file
 _supported_os Linux
 
+# Remove once all tests are fixed to use TEST_IMG_FILE
+# correctly and common.rc sets it unconditionally
+test -z "$TEST_IMG_FILE" && TEST_IMG_FILE=$TEST_IMG
 
 size=128M
 _make_test_img $size
 
 echo
 echo "== mark image read-only"
-chmod a-w "$TEST_IMG"
+chmod a-w "$TEST_IMG_FILE"
 
 echo
 echo "== read from read-only image"

tests/qemu-iotests/039.out

@@ -12,9 +12,9 @@ Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728
 wrote 512/512 bytes at offset 0
 512 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 ./common.config: Killed                  ( if [ "${VALGRIND_QEMU}" == "y" ]; then
-    exec valgrind --log-file="${VALGRIND_LOGFILE}" --error-exitcode=99 "$QEMU_IO_PROG" $QEMU_IO_OPTIONS "$@";
+    exec valgrind --log-file="${VALGRIND_LOGFILE}" --error-exitcode=99 "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@";
 else
-    exec "$QEMU_IO_PROG" $QEMU_IO_OPTIONS "$@";
+    exec "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@";
 fi )
 incompatible_features     0x1
 ERROR cluster 5 refcount=0 reference=1
@@ -51,9 +51,9 @@ Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728
 wrote 512/512 bytes at offset 0
 512 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 ./common.config: Killed                  ( if [ "${VALGRIND_QEMU}" == "y" ]; then
-    exec valgrind --log-file="${VALGRIND_LOGFILE}" --error-exitcode=99 "$QEMU_IO_PROG" $QEMU_IO_OPTIONS "$@";
+    exec valgrind --log-file="${VALGRIND_LOGFILE}" --error-exitcode=99 "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@";
 else
-    exec "$QEMU_IO_PROG" $QEMU_IO_OPTIONS "$@";
+    exec "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@";
 fi )
 incompatible_features     0x1
 ERROR cluster 5 refcount=0 reference=1
@@ -69,9 +69,9 @@ Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728
 wrote 512/512 bytes at offset 0
 512 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 ./common.config: Killed                  ( if [ "${VALGRIND_QEMU}" == "y" ]; then
-    exec valgrind --log-file="${VALGRIND_LOGFILE}" --error-exitcode=99 "$QEMU_IO_PROG" $QEMU_IO_OPTIONS "$@";
+    exec valgrind --log-file="${VALGRIND_LOGFILE}" --error-exitcode=99 "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@";
 else
-    exec "$QEMU_IO_PROG" $QEMU_IO_OPTIONS "$@";
+    exec "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@";
 fi )
 incompatible_features     0x0
 No errors were found on the image.
@@ -92,9 +92,9 @@ Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728
 wrote 512/512 bytes at offset 0
 512 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 ./common.config: Killed                  ( if [ "${VALGRIND_QEMU}" == "y" ]; then
-    exec valgrind --log-file="${VALGRIND_LOGFILE}" --error-exitcode=99 "$QEMU_IO_PROG" $QEMU_IO_OPTIONS "$@";
+    exec valgrind --log-file="${VALGRIND_LOGFILE}" --error-exitcode=99 "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@";
 else
-    exec "$QEMU_IO_PROG" $QEMU_IO_OPTIONS "$@";
+    exec "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@";
 fi )
 incompatible_features     0x1
 ERROR cluster 5 refcount=0 reference=1
@@ -106,9 +106,9 @@ Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728
 wrote 512/512 bytes at offset 0
 512 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 ./common.config: Killed                  ( if [ "${VALGRIND_QEMU}" == "y" ]; then
-    exec valgrind --log-file="${VALGRIND_LOGFILE}" --error-exitcode=99 "$QEMU_IO_PROG" $QEMU_IO_OPTIONS "$@";
+    exec valgrind --log-file="${VALGRIND_LOGFILE}" --error-exitcode=99 "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@";
 else
-    exec "$QEMU_IO_PROG" $QEMU_IO_OPTIONS "$@";
+    exec "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@";
 fi )
 incompatible_features     0x0
 No errors were found on the image.

tests/qemu-iotests/048

@@ -31,13 +31,13 @@ _cleanup()
 {
     echo "Cleanup"
     _cleanup_test_img
-    rm "${TEST_IMG2}"
+    rm "${TEST_IMG_FILE2}"
 }
 trap "_cleanup; exit \$status" 0 1 2 3 15
 
 _compare()
 {
-    $QEMU_IMG compare "$@" "$TEST_IMG" "${TEST_IMG2}"
+    $QEMU_IMG compare $QEMU_IMG_EXTRA_ARGS "$@" "$TEST_IMG" "${TEST_IMG2}"
     echo $?
 }
 
@@ -46,25 +46,37 @@ _compare()
 . ./common.filter
 . ./common.pattern
 
-_supported_fmt raw qcow qcow2 qed
+_supported_fmt raw qcow qcow2 qed luks
 _supported_proto file
 _supported_os Linux
 
+# Remove once all tests are fixed to use TEST_IMG_FILE
+# correctly and common.rc sets it unconditionally
+test -z "$TEST_IMG_FILE" && TEST_IMG_FILE=$TEST_IMG
+
 # Setup test basic parameters
 TEST_IMG2=$TEST_IMG.2
+TEST_IMG_FILE2=$TEST_IMG_FILE.2
 CLUSTER_SIZE=4096
-size=1024M
+size=128M
 
 _make_test_img $size
 io_pattern write 524288 $CLUSTER_SIZE $CLUSTER_SIZE 4 45
 
 # Compare identical images
-cp "$TEST_IMG" "${TEST_IMG2}"
+cp "$TEST_IMG_FILE" "${TEST_IMG_FILE2}"
 _compare
 _compare -q
 
 # Compare images with different size
-$QEMU_IMG resize -f $IMGFMT "$TEST_IMG" +512M
+if [ "$IMGOPTSSYNTAX" = "true" ]; then
+    $QEMU_IMG resize $QEMU_IMG_EXTRA_ARGS "$TEST_IMG" +32M
+else
+    $QEMU_IMG resize -f $IMGFMT "$TEST_IMG" +32M
+fi
+# Ensure extended space is zero-initialized
+$QEMU_IO "$TEST_IMG" -c "write -z $size 32M" | _filter_qemu_io
+
 _compare
 _compare -s
 
@@ -77,7 +89,7 @@ _compare
 # Test unaligned case of mismatch offsets in allocated clusters
 _make_test_img $size
 io_pattern write 0 512 0 1 100
-cp "$TEST_IMG" "$TEST_IMG2"
+cp "$TEST_IMG_FILE" "$TEST_IMG_FILE2"
 io_pattern write 512 512 0 1 101
 _compare
 

tests/qemu-iotests/048.out

@@ -1,5 +1,5 @@
 QA output created by 048
-Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=1073741824
+Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728
 === IO: pattern 45
 wrote 4096/4096 bytes at offset 524288
 4 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
@@ -13,6 +13,8 @@ Images are identical.
 0
 0
 Image resized.
+wrote 33554432/33554432 bytes at offset 134217728
+32 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 Warning: Image size mismatch!
 Images are identical.
 0
@@ -28,7 +30,7 @@ wrote 4096/4096 bytes at offset 0
 4 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 Content mismatch at offset 0!
 1
-Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=1073741824
+Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728
 === IO: pattern 100
 wrote 512/512 bytes at offset 0
 512 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)

tests/qemu-iotests/052

@@ -47,6 +47,10 @@ _supported_cache_modes "writeback" "writethrough" "unsafe"
 size=128M
 _make_test_img $size
 
+echo
+echo "== initializing whole image =="
+$QEMU_IO -c "write -z 0 $size" "$TEST_IMG" | _filter_qemu_io
+
 echo
 echo "== reading whole image =="
 $QEMU_IO -s -c "read 0 $size" "$TEST_IMG" | _filter_qemu_io

tests/qemu-iotests/052.out

@@ -1,6 +1,10 @@
 QA output created by 052
 Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728
 
+== initializing whole image ==
+wrote 134217728/134217728 bytes at offset 0
+128 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+
 == reading whole image ==
 read 134217728/134217728 bytes at offset 0
 128 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)

tests/qemu-iotests/061.out

@@ -58,9 +58,9 @@ Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=67108864
 wrote 131072/131072 bytes at offset 0
 128 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 ./common.config: Killed                  ( if [ "${VALGRIND_QEMU}" == "y" ]; then
-    exec valgrind --log-file="${VALGRIND_LOGFILE}" --error-exitcode=99 "$QEMU_IO_PROG" $QEMU_IO_OPTIONS "$@";
+    exec valgrind --log-file="${VALGRIND_LOGFILE}" --error-exitcode=99 "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@";
 else
-    exec "$QEMU_IO_PROG" $QEMU_IO_OPTIONS "$@";
+    exec "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@";
 fi )
 magic                     0x514649fb
 version                   3
@@ -220,9 +220,9 @@ Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=67108864
 wrote 131072/131072 bytes at offset 0
 128 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 ./common.config: Killed                  ( if [ "${VALGRIND_QEMU}" == "y" ]; then
-    exec valgrind --log-file="${VALGRIND_LOGFILE}" --error-exitcode=99 "$QEMU_IO_PROG" $QEMU_IO_OPTIONS "$@";
+    exec valgrind --log-file="${VALGRIND_LOGFILE}" --error-exitcode=99 "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@";
 else
-    exec "$QEMU_IO_PROG" $QEMU_IO_OPTIONS "$@";
+    exec "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@";
 fi )
 magic                     0x514649fb
 version                   3

tests/qemu-iotests/083

@@ -43,7 +43,7 @@ choose_tcp_port() {
 
 wait_for_tcp_port() {
 	while ! (netstat --tcp --listening --numeric | \
-		 grep "$1.*0\\.0\\.0\\.0:\\*.*LISTEN") 2>&1 >/dev/null; do
+		 grep "$1.*0\\.0\\.0\\.0:\\*.*LISTEN") >/dev/null 2>&1; do
 		sleep 0.1
 	done
 }
@@ -70,7 +70,7 @@ EOF
 		nbd_url="nbd:127.0.0.1:$port:exportname=foo"
 	fi
 
-	$PYTHON nbd-fault-injector.py $extra_args "127.0.0.1:$port" "$TEST_DIR/nbd-fault-injector.conf" 2>&1 >/dev/null &
+	$PYTHON nbd-fault-injector.py $extra_args "127.0.0.1:$port" "$TEST_DIR/nbd-fault-injector.conf" >/dev/null 2>&1 &
 	wait_for_tcp_port "127\\.0\\.0\\.1:$port"
 	$QEMU_IO -c "read 0 512" "$nbd_url" 2>&1 | _filter_qemu_io | _filter_nbd
 

tests/qemu-iotests/100

@@ -47,6 +47,7 @@ size=128M
 echo
 echo "== Single request =="
 _make_test_img $size
+$QEMU_IO -c "write -z 0 8k" "$TEST_IMG" | _filter_qemu_io
 $QEMU_IO -c "multiwrite 0 4k" "$TEST_IMG" | _filter_qemu_io
 
 echo
@@ -59,6 +60,7 @@ _cleanup_test_img
 echo
 echo "== Sequential requests =="
 _make_test_img $size
+$QEMU_IO -c "write -z 0 12k" "$TEST_IMG" | _filter_qemu_io
 $QEMU_IO -c "multiwrite 0 4k ; 4k 4k" "$TEST_IMG" | _filter_qemu_io
 
 echo
@@ -72,6 +74,7 @@ _cleanup_test_img
 echo
 echo "== Superset overlapping requests =="
 _make_test_img $size
+$QEMU_IO -c "write -z 0 8k" "$TEST_IMG" | _filter_qemu_io
 $QEMU_IO -c "multiwrite 0 4k ; 1k 2k" "$TEST_IMG" | _filter_qemu_io
 
 echo
@@ -87,6 +90,7 @@ _cleanup_test_img
 echo
 echo "== Subset overlapping requests =="
 _make_test_img $size
+$QEMU_IO -c "write -z 0 8k" "$TEST_IMG" | _filter_qemu_io
 $QEMU_IO -c "multiwrite 1k 2k ; 0k 4k" "$TEST_IMG" | _filter_qemu_io
 
 echo
@@ -102,6 +106,7 @@ _cleanup_test_img
 echo
 echo "== Head overlapping requests =="
 _make_test_img $size
+$QEMU_IO -c "write -z 0 8k" "$TEST_IMG" | _filter_qemu_io
 $QEMU_IO -c "multiwrite 0k 2k ; 0k 4k" "$TEST_IMG" | _filter_qemu_io
 
 echo
@@ -116,6 +121,7 @@ _cleanup_test_img
 echo
 echo "== Tail overlapping requests =="
 _make_test_img $size
+$QEMU_IO -c "write -z 0 8k" "$TEST_IMG" | _filter_qemu_io
 $QEMU_IO -c "multiwrite 2k 2k ; 0k 4k" "$TEST_IMG" | _filter_qemu_io
 
 echo
@@ -130,6 +136,7 @@ _cleanup_test_img
 echo
 echo "== Disjoint requests =="
 _make_test_img $size
+$QEMU_IO -c "write -z 0 72k" "$TEST_IMG" | _filter_qemu_io
 $QEMU_IO -c "multiwrite 0 4k ; 64k 4k" "$TEST_IMG" | _filter_qemu_io
 
 echo

tests/qemu-iotests/100.out

@@ -2,6 +2,8 @@ QA output created by 100
 
 == Single request ==
 Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728
+wrote 8192/8192 bytes at offset 0
+8 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 wrote 4096/4096 bytes at offset 0
 4 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 
@@ -13,6 +15,8 @@ read 4096/4096 bytes at offset 4096
 
 == Sequential requests ==
 Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728
+wrote 12288/12288 bytes at offset 0
+12 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 wrote 8192/8192 bytes at offset 0
 8 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 
@@ -26,6 +30,8 @@ read 4096/4096 bytes at offset 8192
 
 == Superset overlapping requests ==
 Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728
+wrote 8192/8192 bytes at offset 0
+8 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 wrote 6144/6144 bytes at offset 0
 6 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 
@@ -39,6 +45,8 @@ read 4096/4096 bytes at offset 4096
 
 == Subset overlapping requests ==
 Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728
+wrote 8192/8192 bytes at offset 0
+8 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 wrote 6144/6144 bytes at offset 1024
 6 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 
@@ -52,6 +60,8 @@ read 4096/4096 bytes at offset 4096
 
 == Head overlapping requests ==
 Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728
+wrote 8192/8192 bytes at offset 0
+8 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 wrote 6144/6144 bytes at offset 0
 6 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 
@@ -63,6 +73,8 @@ read 4096/4096 bytes at offset 4096
 
 == Tail overlapping requests ==
 Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728
+wrote 8192/8192 bytes at offset 0
+8 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 wrote 6144/6144 bytes at offset 2048
 6 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 
@@ -74,6 +86,8 @@ read 4096/4096 bytes at offset 4096
 
 == Disjoint requests ==
 Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728
+wrote 73728/73728 bytes at offset 0
+72 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 wrote 8192/8192 bytes at offset 0
 8 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 

tests/qemu-iotests/137.out

@@ -32,9 +32,9 @@ Unsupported value 'blubb' for qcow2 option 'overlap-check'. Allowed are any of t
 wrote 512/512 bytes at offset 0
 512 bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 ./common.config: Killed                  ( if [ "${VALGRIND_QEMU}" == "y" ]; then
-    exec valgrind --log-file="${VALGRIND_LOGFILE}" --error-exitcode=99 "$QEMU_IO_PROG" $QEMU_IO_OPTIONS "$@";
+    exec valgrind --log-file="${VALGRIND_LOGFILE}" --error-exitcode=99 "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@";
 else
-    exec "$QEMU_IO_PROG" $QEMU_IO_OPTIONS "$@";
+    exec "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@";
 fi )
 incompatible_features     0x0
 Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=67108864

tests/qemu-iotests/common

@@ -53,6 +53,8 @@ export QEMU_IO_OPTIONS=""
 export CACHEMODE_IS_DEFAULT=true
 export QEMU_OPTIONS="-nodefaults"
 export VALGRIND_QEMU=
+export IMGKEYSECRET=
+export IMGOPTSSYNTAX=false
 
 for r
 do
@@ -207,6 +209,13 @@ testlist options
             xpand=false
             ;;
 
+        -luks)
+            IMGOPTSSYNTAX=true
+            IMGFMT=luks
+            IMGKEYSECRET=123456
+            xpand=false
+            ;;
+
         -qed)
             IMGFMT=qed
             xpand=false
@@ -399,7 +408,11 @@ BEGIN        { for (t='$start'; t<='$end'; t++) printf "%03d\n",t }' \
 done
 
 # Set qemu-io cache mode with $CACHEMODE we have
-QEMU_IO_OPTIONS="$QEMU_IO_OPTIONS -f $IMGFMT --cache $CACHEMODE"
+if [ "$IMGOPTSSYNTAX" = "true" ]; then
+    QEMU_IO_OPTIONS="$QEMU_IO_OPTIONS --cache $CACHEMODE"
+else
+    QEMU_IO_OPTIONS="$QEMU_IO_OPTIONS -f $IMGFMT --cache $CACHEMODE"
+fi
 
 # Set default options for qemu-img create -o if they were not specified
 _set_default_imgopts

tests/qemu-iotests/common.config

@@ -123,12 +123,19 @@ _qemu_img_wrapper()
 _qemu_io_wrapper()
 {
     local VALGRIND_LOGFILE="${TEST_DIR}"/$$.valgrind
+    local QEMU_IO_ARGS="$QEMU_IO_OPTIONS"
+    if [ "$IMGOPTSSYNTAX" = "true" ]; then
+        QEMU_IO_ARGS="--image-opts $QEMU_IO_ARGS"
+        if [ -n "$IMGKEYSECRET" ]; then
+            QEMU_IO_ARGS="--object secret,id=keysec0,data=$IMGKEYSECRET $QEMU_IO_ARGS"
+        fi
+    fi
     local RETVAL
     (
         if [ "${VALGRIND_QEMU}" == "y" ]; then
-            exec valgrind --log-file="${VALGRIND_LOGFILE}" --error-exitcode=99 "$QEMU_IO_PROG" $QEMU_IO_OPTIONS "$@"
+            exec valgrind --log-file="${VALGRIND_LOGFILE}" --error-exitcode=99 "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@"
         else
-            exec "$QEMU_IO_PROG" $QEMU_IO_OPTIONS "$@"
+            exec "$QEMU_IO_PROG" $QEMU_IO_ARGS "$@"
         fi
     )
     RETVAL=$?
@@ -154,6 +161,16 @@ export QEMU_IMG=_qemu_img_wrapper
 export QEMU_IO=_qemu_io_wrapper
 export QEMU_NBD=_qemu_nbd_wrapper
 
+QEMU_IMG_EXTRA_ARGS=
+if [ "$IMGOPTSSYNTAX" = "true" ]; then
+    QEMU_IMG_EXTRA_ARGS="--image-opts $QEMU_IMG_EXTRA_ARGS"
+    if [ -n "$IMGKEYSECRET" ]; then
+        QEMU_IMG_EXTRA_ARGS="--object secret,id=keysec0,data=$IMGKEYSECRET $QEMU_IMG_EXTRA_ARGS"
+    fi
+fi
+export QEMU_IMG_EXTRA_ARGS
+
+
 default_machine=$($QEMU -machine help | sed -n '/(default)/ s/ .*//p')
 default_alias_machine=$($QEMU -machine help | \
    sed -n "/(alias of $default_machine)/ { s/ .*//p; q; }")

tests/qemu-iotests/common.filter

@@ -92,12 +92,14 @@ _filter_img_create()
         -e "s# zeroed_grain=\\(on\\|off\\)##g" \
         -e "s# subformat='[^']*'##g" \
         -e "s# adapter_type='[^']*'##g" \
+        -e "s# hwversion=[^ ]*##g" \
         -e "s# lazy_refcounts=\\(on\\|off\\)##g" \
         -e "s# block_size=[0-9]\\+##g" \
         -e "s# block_state_zero=\\(on\\|off\\)##g" \
         -e "s# log_size=[0-9]\\+##g" \
         -e "s/archipelago:a/TEST_DIR\//g" \
-        -e "s# refcount_bits=[0-9]\\+##g"
+        -e "s# refcount_bits=[0-9]\\+##g" \
+        -e "s# key-secret=[a-zA-Z0-9]\\+##g"
 }
 
 _filter_img_info()
@@ -115,6 +117,7 @@ _filter_img_info()
         -e "/zeroed_grain: \\(on\\|off\\)/d" \
         -e "/subformat: '[^']*'/d" \
         -e "/adapter_type: '[^']*'/d" \
+        -e "/hwversion: '[^']*'/d" \
         -e "/lazy_refcounts: \\(on\\|off\\)/d" \
         -e "/block_size: [0-9]\\+/d" \
         -e "/block_state_zero: \\(on\\|off\\)/d" \

tests/qemu-iotests/common.rc

@@ -53,21 +53,45 @@ fi
 # make sure we have a standard umask
 umask 022
 
-if [ "$IMGPROTO" = "file" ]; then
+if [ "$IMGOPTSSYNTAX" = "true" ]; then
-    TEST_IMG=$TEST_DIR/t.$IMGFMT
+    DRIVER="driver=$IMGFMT"
-elif [ "$IMGPROTO" = "nbd" ]; then
+    if [ "$IMGFMT" = "luks" ]; then
-    TEST_IMG_FILE=$TEST_DIR/t.$IMGFMT
+        DRIVER="$DRIVER,key-secret=keysec0"
-    TEST_IMG="nbd:127.0.0.1:10810"
+    fi
-elif [ "$IMGPROTO" = "ssh" ]; then
+    if [ "$IMGPROTO" = "file" ]; then
-    TEST_IMG_FILE=$TEST_DIR/t.$IMGFMT
+        TEST_IMG_FILE=$TEST_DIR/t.$IMGFMT
-    TEST_IMG="ssh://127.0.0.1$TEST_IMG_FILE"
+        TEST_IMG="$DRIVER,file.filename=$TEST_DIR/t.$IMGFMT"
-elif [ "$IMGPROTO" = "nfs" ]; then
+    elif [ "$IMGPROTO" = "nbd" ]; then
-    TEST_DIR="nfs://127.0.0.1/$TEST_DIR"
+        TEST_IMG_FILE=$TEST_DIR/t.$IMGFMT
-    TEST_IMG=$TEST_DIR/t.$IMGFMT
+        TEST_IMG="$DRIVER,file.driver=nbd,file.host=127.0.0.1,file.port=10810"
-elif [ "$IMGPROTO" = "archipelago" ]; then
+    elif [ "$IMGPROTO" = "ssh" ]; then
-    TEST_IMG="archipelago:at.$IMGFMT"
+        TEST_IMG_FILE=$TEST_DIR/t.$IMGFMT
+        TEST_IMG="$DRIVER,file.driver=ssh,file.host=127.0.0.1,file.path=$TEST_IMG_FILE"
+    elif [ "$IMGPROTO" = "nfs" ]; then
+        TEST_DIR="$DRIVER,file.driver=nfs,file.filename=nfs://127.0.0.1/$TEST_DIR"
+        TEST_IMG=$TEST_DIR_OPTS/t.$IMGFMT
+    elif [ "$IMGPROTO" = "archipelago" ]; then
+        TEST_IMG="$DRIVER,file.driver=archipelago,file.volume=:at.$IMGFMT"
+    else
+        TEST_IMG="$DRIVER,file.driver=$IMGPROTO,file.filename=$TEST_DIR/t.$IMGFMT"
+    fi
 else
-    TEST_IMG=$IMGPROTO:$TEST_DIR/t.$IMGFMT
+    if [ "$IMGPROTO" = "file" ]; then
+        TEST_IMG=$TEST_DIR/t.$IMGFMT
+    elif [ "$IMGPROTO" = "nbd" ]; then
+        TEST_IMG_FILE=$TEST_DIR/t.$IMGFMT
+        TEST_IMG="nbd:127.0.0.1:10810"
+    elif [ "$IMGPROTO" = "ssh" ]; then
+        TEST_IMG_FILE=$TEST_DIR/t.$IMGFMT
+        TEST_IMG="ssh://127.0.0.1$TEST_IMG_FILE"
+    elif [ "$IMGPROTO" = "nfs" ]; then
+        TEST_DIR="nfs://127.0.0.1/$TEST_DIR"
+        TEST_IMG=$TEST_DIR/t.$IMGFMT
+    elif [ "$IMGPROTO" = "archipelago" ]; then
+        TEST_IMG="archipelago:at.$IMGFMT"
+    else
+        TEST_IMG=$IMGPROTO:$TEST_DIR/t.$IMGFMT
+    fi
 fi
 
 _optstr_add()
@@ -108,6 +132,7 @@ _make_test_img()
     local img_name=""
     local use_backing=0
     local backing_file=""
+    local object_options=""
 
     if [ -n "$TEST_IMG_FILE" ]; then
         img_name=$TEST_IMG_FILE
@@ -118,6 +143,10 @@ _make_test_img()
     if [ -n "$IMGOPTS" ]; then
         optstr=$(_optstr_add "$optstr" "$IMGOPTS")
     fi
+    if [ -n "$IMGKEYSECRET" ]; then
+        object_options="--object secret,id=keysec0,data=$IMGKEYSECRET"
+        optstr=$(_optstr_add "$optstr" "key-secret=keysec0")
+    fi
 
     if [ "$1" = "-b" ]; then
         use_backing=1
@@ -135,9 +164,9 @@ _make_test_img()
     # XXX(hch): have global image options?
     (
      if [ $use_backing = 1 ]; then
-        $QEMU_IMG create -f $IMGFMT $extra_img_options -b "$backing_file" "$img_name" $image_size 2>&1
+        $QEMU_IMG create $object_options -f $IMGFMT $extra_img_options -b "$backing_file" "$img_name" $image_size 2>&1
      else
-        $QEMU_IMG create -f $IMGFMT $extra_img_options "$img_name" $image_size 2>&1
+        $QEMU_IMG create $object_options -f $IMGFMT $extra_img_options "$img_name" $image_size 2>&1
      fi
     ) | _filter_img_create
 
@@ -199,7 +228,13 @@ _cleanup_test_img()
 
 _check_test_img()
 {
-    $QEMU_IMG check "$@" -f $IMGFMT "$TEST_IMG" 2>&1 | _filter_testdir | \
+    (
+        if [ "$IMGOPTSSYNTAX" = "true" ]; then
+            $QEMU_IMG check $QEMU_IMG_EXTRA_ARGS "$@" "$TEST_IMG" 2>&1
+        else
+            $QEMU_IMG check "$@" -f $IMGFMT "$TEST_IMG" 2>&1
+        fi
+    ) | _filter_testdir | \
         sed -e '/allocated.*fragmented.*compressed clusters/d' \
             -e 's/qemu-img: This image format does not support checks/No errors were found on the image./' \
             -e '/Image end offset: [0-9]\+/d'

tests/qemu-iotests/iotests.py

@@ -47,7 +47,7 @@ if os.environ.get('QEMU_OPTIONS'):
 
 imgfmt = os.environ.get('IMGFMT', 'raw')
 imgproto = os.environ.get('IMGPROTO', 'file')
-test_dir = os.environ.get('TEST_DIR', '/var/tmp')
+test_dir = os.environ.get('TEST_DIR')
 output_dir = os.environ.get('OUTPUT_DIR', '.')
 cachemode = os.environ.get('CACHEMODE')
 qemu_default_machine = os.environ.get('QEMU_DEFAULT_MACHINE')
@@ -461,6 +461,14 @@ def verify_quorum():
 def main(supported_fmts=[], supported_oses=['linux']):
     '''Run tests'''
 
+    # 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
+    # on.
+    if test_dir is None or qemu_default_machine is None:
+        sys.stderr.write('Please run this test via the "check" script\n')
+        sys.exit(os.EX_USAGE)
+
     debug = '-d' in sys.argv
     verbosity = 1
     verify_image_format(supported_fmts)

trace-events

@@ -74,7 +74,6 @@ bdrv_co_copy_on_readv(void *bs, int64_t sector_num, int nb_sector) "bs %p sector
 bdrv_co_readv_no_serialising(void *bs, int64_t sector_num, int nb_sector) "bs %p sector_num %"PRId64" nb_sectors %d"
 bdrv_co_writev(void *bs, int64_t sector_num, int nb_sector) "bs %p sector_num %"PRId64" nb_sectors %d"
 bdrv_co_write_zeroes(void *bs, int64_t sector_num, int nb_sector, int flags) "bs %p sector_num %"PRId64" nb_sectors %d flags %#x"
-bdrv_co_io_em(void *bs, int64_t sector_num, int nb_sectors, int is_write, void *acb) "bs %p sector_num %"PRId64" nb_sectors %d is_write %d acb %p"
 bdrv_co_do_copy_on_readv(void *bs, int64_t sector_num, int nb_sectors, int64_t cluster_sector_num, int cluster_nb_sectors) "bs %p sector_num %"PRId64" nb_sectors %d cluster_sector_num %"PRId64" cluster_nb_sectors %d"
 
 # block/stream.c
@@ -119,7 +118,7 @@ virtio_blk_req_complete(void *req, int status) "req %p status %d"
 virtio_blk_rw_complete(void *req, int ret) "req %p ret %d"
 virtio_blk_handle_write(void *req, uint64_t sector, size_t nsectors) "req %p sector %"PRIu64" nsectors %zu"
 virtio_blk_handle_read(void *req, uint64_t sector, size_t nsectors) "req %p sector %"PRIu64" nsectors %zu"
-virtio_blk_submit_multireq(void *mrb, int start, int num_reqs, uint64_t sector, size_t nsectors, bool is_write) "mrb %p start %d num_reqs %d sector %"PRIu64" nsectors %zu is_write %d"
+virtio_blk_submit_multireq(void *mrb, int start, int num_reqs, uint64_t offset, size_t size, bool is_write) "mrb %p start %d num_reqs %d offset %"PRIu64" size %zu is_write %d"
 
 # hw/block/dataplane/virtio-blk.c
 virtio_blk_data_plane_start(void *s) "dataplane %p"