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

Block layer patches:

- block: AioContext management, part 1
- qmp: forbid qmp_cont in RUN_STATE_FINISH_MIGRATE
- nvme: fix copy direction in DMA reads going to CMB
- file-posix: Fix block status for unaligned raw images with O_DIRECT
- file-posix: Fix xfs_write_zeroes() after EOF
- Documentation and iotests improvements

# gpg: Signature made Mon 20 May 2019 16:12:38 BST
# gpg:                using RSA key 7F09B272C88F2FD6
# gpg: Good signature from "Kevin Wolf <kwolf@redhat.com>" [full]
# Primary key fingerprint: DC3D EB15 9A9A F95D 3D74  56FE 7F09 B272 C88F 2FD6

* remotes/kevin/tags/for-upstream: (24 commits)
  iotests: Make 245 faster and more reliable
  iotests.py: Fix VM.run_job
  iotests.py: Let assert_qmp() accept an array
  block: Improve "Block node is read-only" message
  qemu-img.texi: Describe human-readable info output
  qemu-img.texi: Be specific about JSON object types
  iotests: Test unaligned raw images with O_DIRECT
  block/file-posix: Unaligned O_DIRECT block-status
  test-block-iothread: Test AioContext propagation for block jobs
  blockjob: Remove AioContext notifiers
  blockjob: Propagate AioContext change to all job nodes
  block: Add blk_set_allow_aio_context_change()
  block: Implement .(can_)set_aio_ctx for BlockBackend
  test-block-iothread: Test AioContext propagation through the tree
  block: Propagate AioContext change to parents
  block: Move recursion to bdrv_set_aio_context()
  block: Make bdrv_attach/detach_aio_context() static
  block: Add bdrv_try_set_aio_context()
  nvme: fix copy direction in DMA reads going to CMB
  iotest: fix 169: do not run qmp_cont in RUN_STATE_FINISH_MIGRATE
  ...

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

block.c

@@ -936,6 +936,20 @@ static int bdrv_child_cb_inactivate(BdrvChild *child)
     return 0;
 }
 
+static bool bdrv_child_cb_can_set_aio_ctx(BdrvChild *child, AioContext *ctx,
+                                          GSList **ignore, Error **errp)
+{
+    BlockDriverState *bs = child->opaque;
+    return bdrv_can_set_aio_context(bs, ctx, ignore, errp);
+}
+
+static void bdrv_child_cb_set_aio_ctx(BdrvChild *child, AioContext *ctx,
+                                      GSList **ignore)
+{
+    BlockDriverState *bs = child->opaque;
+    return bdrv_set_aio_context_ignore(bs, ctx, ignore);
+}
+
 /*
  * Returns the options and flags that a temporary snapshot should get, based on
  * the originally requested flags (the originally requested image will have
@@ -1003,6 +1017,8 @@ const BdrvChildRole child_file = {
     .attach          = bdrv_child_cb_attach,
     .detach          = bdrv_child_cb_detach,
     .inactivate      = bdrv_child_cb_inactivate,
+    .can_set_aio_ctx = bdrv_child_cb_can_set_aio_ctx,
+    .set_aio_ctx     = bdrv_child_cb_set_aio_ctx,
 };
 
 /*
@@ -1029,6 +1045,8 @@ const BdrvChildRole child_format = {
     .attach          = bdrv_child_cb_attach,
     .detach          = bdrv_child_cb_detach,
     .inactivate      = bdrv_child_cb_inactivate,
+    .can_set_aio_ctx = bdrv_child_cb_can_set_aio_ctx,
+    .set_aio_ctx     = bdrv_child_cb_set_aio_ctx,
 };
 
 static void bdrv_backing_attach(BdrvChild *c)
@@ -1152,6 +1170,8 @@ const BdrvChildRole child_backing = {
     .drained_end     = bdrv_child_cb_drained_end,
     .inactivate      = bdrv_child_cb_inactivate,
     .update_filename = bdrv_backing_update_filename,
+    .can_set_aio_ctx = bdrv_child_cb_can_set_aio_ctx,
+    .set_aio_ctx     = bdrv_child_cb_set_aio_ctx,
 };
 
 static int bdrv_open_flags(BlockDriverState *bs, int flags)
@@ -1689,6 +1709,8 @@ static int bdrv_child_check_perm(BdrvChild *c, BlockReopenQueue *q,
                                  GSList *ignore_children, Error **errp);
 static void bdrv_child_abort_perm_update(BdrvChild *c);
 static void bdrv_child_set_perm(BdrvChild *c, uint64_t perm, uint64_t shared);
+static void bdrv_get_cumulative_perm(BlockDriverState *bs, uint64_t *perm,
+                                     uint64_t *shared_perm);
 
 typedef struct BlockReopenQueueEntry {
      bool prepared;
@@ -1775,7 +1797,20 @@ static int bdrv_check_perm(BlockDriverState *bs, BlockReopenQueue *q,
     if ((cumulative_perms & (BLK_PERM_WRITE | BLK_PERM_WRITE_UNCHANGED)) &&
         !bdrv_is_writable_after_reopen(bs, q))
     {
-        error_setg(errp, "Block node is read-only");
+        if (!bdrv_is_writable_after_reopen(bs, NULL)) {
+            error_setg(errp, "Block node is read-only");
+        } else {
+            uint64_t current_perms, current_shared;
+            bdrv_get_cumulative_perm(bs, &current_perms, &current_shared);
+            if (current_perms & (BLK_PERM_WRITE | BLK_PERM_WRITE_UNCHANGED)) {
+                error_setg(errp, "Cannot make block node read-only, there is "
+                           "a writer on it");
+            } else {
+                error_setg(errp, "Cannot make block node read-only and create "
+                           "a writer on it");
+            }
+        }
+
         return -EPERM;
     }
 
@@ -5666,10 +5701,9 @@ static void bdrv_do_remove_aio_context_notifier(BdrvAioNotifier *ban)
     g_free(ban);
 }
 
-void bdrv_detach_aio_context(BlockDriverState *bs)
+static void bdrv_detach_aio_context(BlockDriverState *bs)
 {
     BdrvAioNotifier *baf, *baf_tmp;
-    BdrvChild *child;
 
     assert(!bs->walking_aio_notifiers);
     bs->walking_aio_notifiers = true;
@@ -5688,9 +5722,6 @@ void bdrv_detach_aio_context(BlockDriverState *bs)
     if (bs->drv && bs->drv->bdrv_detach_aio_context) {
         bs->drv->bdrv_detach_aio_context(bs);
     }
-    QLIST_FOREACH(child, &bs->children, next) {
-        bdrv_detach_aio_context(child->bs);
-    }
 
     if (bs->quiesce_counter) {
         aio_enable_external(bs->aio_context);
@@ -5698,11 +5729,10 @@ void bdrv_detach_aio_context(BlockDriverState *bs)
     bs->aio_context = NULL;
 }
 
-void bdrv_attach_aio_context(BlockDriverState *bs,
-                             AioContext *new_context)
+static void bdrv_attach_aio_context(BlockDriverState *bs,
+                                    AioContext *new_context)
 {
     BdrvAioNotifier *ban, *ban_tmp;
-    BdrvChild *child;
 
     if (bs->quiesce_counter) {
         aio_disable_external(new_context);
@@ -5710,9 +5740,6 @@ void bdrv_attach_aio_context(BlockDriverState *bs,
 
     bs->aio_context = new_context;
 
-    QLIST_FOREACH(child, &bs->children, next) {
-        bdrv_attach_aio_context(child->bs, new_context);
-    }
     if (bs->drv && bs->drv->bdrv_attach_aio_context) {
         bs->drv->bdrv_attach_aio_context(bs, new_context);
     }
@@ -5729,16 +5756,36 @@ void bdrv_attach_aio_context(BlockDriverState *bs,
     bs->walking_aio_notifiers = false;
 }
 
-/* The caller must own the AioContext lock for the old AioContext of bs, but it
- * must not own the AioContext lock for new_context (unless new_context is
- * the same as the current context of bs). */
-void bdrv_set_aio_context(BlockDriverState *bs, AioContext *new_context)
+/* @ignore will accumulate all visited BdrvChild object. The caller is
+ * responsible for freeing the list afterwards. */
+void bdrv_set_aio_context_ignore(BlockDriverState *bs,
+                                 AioContext *new_context, GSList **ignore)
 {
+    BdrvChild *child;
+
     if (bdrv_get_aio_context(bs) == new_context) {
         return;
     }
 
     bdrv_drained_begin(bs);
+
+    QLIST_FOREACH(child, &bs->children, next) {
+        if (g_slist_find(*ignore, child)) {
+            continue;
+        }
+        *ignore = g_slist_prepend(*ignore, child);
+        bdrv_set_aio_context_ignore(child->bs, new_context, ignore);
+    }
+    QLIST_FOREACH(child, &bs->parents, next_parent) {
+        if (g_slist_find(*ignore, child)) {
+            continue;
+        }
+        if (child->role->set_aio_ctx) {
+            *ignore = g_slist_prepend(*ignore, child);
+            child->role->set_aio_ctx(child, new_context, ignore);
+        }
+    }
+
     bdrv_detach_aio_context(bs);
 
     /* This function executes in the old AioContext so acquire the new one in
@@ -5750,6 +5797,101 @@ void bdrv_set_aio_context(BlockDriverState *bs, AioContext *new_context)
     aio_context_release(new_context);
 }
 
+/* The caller must own the AioContext lock for the old AioContext of bs, but it
+ * must not own the AioContext lock for new_context (unless new_context is
+ * the same as the current context of bs). */
+void bdrv_set_aio_context(BlockDriverState *bs, AioContext *new_context)
+{
+    GSList *ignore_list = NULL;
+    bdrv_set_aio_context_ignore(bs, new_context, &ignore_list);
+    g_slist_free(ignore_list);
+}
+
+static bool bdrv_parent_can_set_aio_context(BdrvChild *c, AioContext *ctx,
+                                            GSList **ignore, Error **errp)
+{
+    if (g_slist_find(*ignore, c)) {
+        return true;
+    }
+    *ignore = g_slist_prepend(*ignore, c);
+
+    /* A BdrvChildRole that doesn't handle AioContext changes cannot
+     * tolerate any AioContext changes */
+    if (!c->role->can_set_aio_ctx) {
+        char *user = bdrv_child_user_desc(c);
+        error_setg(errp, "Changing iothreads is not supported by %s", user);
+        g_free(user);
+        return false;
+    }
+    if (!c->role->can_set_aio_ctx(c, ctx, ignore, errp)) {
+        assert(!errp || *errp);
+        return false;
+    }
+    return true;
+}
+
+bool bdrv_child_can_set_aio_context(BdrvChild *c, AioContext *ctx,
+                                    GSList **ignore, Error **errp)
+{
+    if (g_slist_find(*ignore, c)) {
+        return true;
+    }
+    *ignore = g_slist_prepend(*ignore, c);
+    return bdrv_can_set_aio_context(c->bs, ctx, ignore, errp);
+}
+
+/* @ignore will accumulate all visited BdrvChild object. The caller is
+ * responsible for freeing the list afterwards. */
+bool bdrv_can_set_aio_context(BlockDriverState *bs, AioContext *ctx,
+                              GSList **ignore, Error **errp)
+{
+    BdrvChild *c;
+
+    if (bdrv_get_aio_context(bs) == ctx) {
+        return true;
+    }
+
+    QLIST_FOREACH(c, &bs->parents, next_parent) {
+        if (!bdrv_parent_can_set_aio_context(c, ctx, ignore, errp)) {
+            return false;
+        }
+    }
+    QLIST_FOREACH(c, &bs->children, next) {
+        if (!bdrv_child_can_set_aio_context(c, ctx, ignore, errp)) {
+            return false;
+        }
+    }
+
+    return true;
+}
+
+int bdrv_child_try_set_aio_context(BlockDriverState *bs, AioContext *ctx,
+                                   BdrvChild *ignore_child, Error **errp)
+{
+    GSList *ignore;
+    bool ret;
+
+    ignore = ignore_child ? g_slist_prepend(NULL, ignore_child) : NULL;
+    ret = bdrv_can_set_aio_context(bs, ctx, &ignore, errp);
+    g_slist_free(ignore);
+
+    if (!ret) {
+        return -EPERM;
+    }
+
+    ignore = ignore_child ? g_slist_prepend(NULL, ignore_child) : NULL;
+    bdrv_set_aio_context_ignore(bs, ctx, &ignore);
+    g_slist_free(ignore);
+
+    return 0;
+}
+
+int bdrv_try_set_aio_context(BlockDriverState *bs, AioContext *ctx,
+                             Error **errp)
+{
+    return bdrv_child_try_set_aio_context(bs, ctx, NULL, errp);
+}
+
 void bdrv_add_aio_context_notifier(BlockDriverState *bs,
         void (*attached_aio_context)(AioContext *new_context, void *opaque),
         void (*detach_aio_context)(void *opaque), void *opaque)

block/backup.c

@@ -300,13 +300,6 @@ static void backup_clean(Job *job)
     s->target = NULL;
 }
 
-static void backup_attached_aio_context(BlockJob *job, AioContext *aio_context)
-{
-    BackupBlockJob *s = container_of(job, BackupBlockJob, common);
-
-    blk_set_aio_context(s->target, aio_context);
-}
-
 void backup_do_checkpoint(BlockJob *job, Error **errp)
 {
     BackupBlockJob *backup_job = container_of(job, BackupBlockJob, common);
@@ -558,7 +551,6 @@ static const BlockJobDriver backup_job_driver = {
         .abort                  = backup_abort,
         .clean                  = backup_clean,
     },
-    .attached_aio_context   = backup_attached_aio_context,
     .drain                  = backup_drain,
 };
 

block/block-backend.c

@@ -71,6 +71,7 @@ struct BlockBackend {
     uint64_t shared_perm;
     bool disable_perm;
 
+    bool allow_aio_context_change;
     bool allow_write_beyond_eof;
 
     NotifierList remove_bs_notifiers, insert_bs_notifiers;
@@ -124,6 +125,11 @@ static void blk_root_drained_end(BdrvChild *child);
 static void blk_root_change_media(BdrvChild *child, bool load);
 static void blk_root_resize(BdrvChild *child);
 
+static bool blk_root_can_set_aio_ctx(BdrvChild *child, AioContext *ctx,
+                                     GSList **ignore, Error **errp);
+static void blk_root_set_aio_ctx(BdrvChild *child, AioContext *ctx,
+                                 GSList **ignore);
+
 static char *blk_root_get_parent_desc(BdrvChild *child)
 {
     BlockBackend *blk = child->opaque;
@@ -300,6 +306,9 @@ static const BdrvChildRole child_root = {
 
     .attach             = blk_root_attach,
     .detach             = blk_root_detach,
+
+    .can_set_aio_ctx    = blk_root_can_set_aio_ctx,
+    .set_aio_ctx        = blk_root_set_aio_ctx,
 };
 
 /*
@@ -1084,6 +1093,11 @@ void blk_set_allow_write_beyond_eof(BlockBackend *blk, bool allow)
     blk->allow_write_beyond_eof = allow;
 }
 
+void blk_set_allow_aio_context_change(BlockBackend *blk, bool allow)
+{
+    blk->allow_aio_context_change = allow;
+}
+
 static int blk_check_byte_request(BlockBackend *blk, int64_t offset,
                                   size_t size)
 {
@@ -1852,7 +1866,8 @@ static AioContext *blk_aiocb_get_aio_context(BlockAIOCB *acb)
     return blk_get_aio_context(blk_acb->blk);
 }
 
-void blk_set_aio_context(BlockBackend *blk, AioContext *new_context)
+static void blk_do_set_aio_context(BlockBackend *blk, AioContext *new_context,
+                                   bool update_root_node)
 {
     BlockDriverState *bs = blk_bs(blk);
     ThrottleGroupMember *tgm = &blk->public.throttle_group_member;
@@ -1864,10 +1879,46 @@ void blk_set_aio_context(BlockBackend *blk, AioContext *new_context)
             throttle_group_attach_aio_context(tgm, new_context);
             bdrv_drained_end(bs);
         }
-        bdrv_set_aio_context(bs, new_context);
+        if (update_root_node) {
+            GSList *ignore = g_slist_prepend(NULL, blk->root);
+            bdrv_set_aio_context_ignore(bs, new_context, &ignore);
+            g_slist_free(ignore);
+        }
     }
 }
 
+void blk_set_aio_context(BlockBackend *blk, AioContext *new_context)
+{
+    blk_do_set_aio_context(blk, new_context, true);
+}
+
+static bool blk_root_can_set_aio_ctx(BdrvChild *child, AioContext *ctx,
+                                     GSList **ignore, Error **errp)
+{
+    BlockBackend *blk = child->opaque;
+
+    if (blk->allow_aio_context_change) {
+        return true;
+    }
+
+    /* Only manually created BlockBackends that are not attached to anything
+     * can change their AioContext without updating their user. */
+    if (!blk->name || blk->dev) {
+        /* TODO Add BB name/QOM path */
+        error_setg(errp, "Cannot change iothread of active block backend");
+        return false;
+    }
+
+    return true;
+}
+
+static void blk_root_set_aio_ctx(BdrvChild *child, AioContext *ctx,
+                                 GSList **ignore)
+{
+    BlockBackend *blk = child->opaque;
+    blk_do_set_aio_context(blk, ctx, false);
+}
+
 void blk_add_aio_context_notifier(BlockBackend *blk,
         void (*attached_aio_context)(AioContext *new_context, void *opaque),
         void (*detach_aio_context)(void *opaque), void *opaque)

block/file-posix.c

@@ -1444,9 +1444,22 @@ out:
 #ifdef CONFIG_XFS
 static int xfs_write_zeroes(BDRVRawState *s, int64_t offset, uint64_t bytes)
 {
+    int64_t len;
     struct xfs_flock64 fl;
     int err;
 
+    len = lseek(s->fd, 0, SEEK_END);
+    if (len < 0) {
+        return -errno;
+    }
+
+    if (offset + bytes > len) {
+        /* XFS_IOC_ZERO_RANGE does not increase the file length */
+        if (ftruncate(s->fd, offset + bytes) < 0) {
+            return -errno;
+        }
+    }
+
     memset(&fl, 0, sizeof(fl));
     fl.l_whence = SEEK_SET;
     fl.l_start = offset;
@@ -2475,6 +2488,8 @@ static int coroutine_fn raw_co_block_status(BlockDriverState *bs,
     off_t data = 0, hole = 0;
     int ret;
 
+    assert(QEMU_IS_ALIGNED(offset | bytes, bs->bl.request_alignment));
+
     ret = fd_open(bs);
     if (ret < 0) {
         return ret;
@@ -2500,6 +2515,20 @@ static int coroutine_fn raw_co_block_status(BlockDriverState *bs,
         /* On a data extent, compute bytes to the end of the extent,
          * possibly including a partial sector at EOF. */
         *pnum = MIN(bytes, hole - offset);
+
+        /*
+         * We are not allowed to return partial sectors, though, so
+         * round up if necessary.
+         */
+        if (!QEMU_IS_ALIGNED(*pnum, bs->bl.request_alignment)) {
+            int64_t file_length = raw_getlength(bs);
+            if (file_length > 0) {
+                /* Ignore errors, this is just a safeguard */
+                assert(hole == file_length);
+            }
+            *pnum = ROUND_UP(*pnum, bs->bl.request_alignment);
+        }
+
         ret = BDRV_BLOCK_DATA;
     } else {
         /* On a hole, compute bytes to the beginning of the next extent.  */

block/io.c

@@ -769,7 +769,7 @@ static bool coroutine_fn wait_serialising_requests(BdrvTrackedRequest *self)
 static int bdrv_check_byte_request(BlockDriverState *bs, int64_t offset,
                                    size_t size)
 {
-    if (size > BDRV_REQUEST_MAX_SECTORS << BDRV_SECTOR_BITS) {
+    if (size > BDRV_REQUEST_MAX_BYTES) {
         return -EIO;
     }
 
@@ -1017,7 +1017,7 @@ static int coroutine_fn bdrv_driver_preadv(BlockDriverState *bs,
 
     assert((offset & (BDRV_SECTOR_SIZE - 1)) == 0);
     assert((bytes & (BDRV_SECTOR_SIZE - 1)) == 0);
-    assert((bytes >> BDRV_SECTOR_BITS) <= BDRV_REQUEST_MAX_SECTORS);
+    assert(bytes <= BDRV_REQUEST_MAX_BYTES);
     assert(drv->bdrv_co_readv);
 
     return drv->bdrv_co_readv(bs, sector_num, nb_sectors, qiov);
@@ -1070,7 +1070,7 @@ static int coroutine_fn bdrv_driver_pwritev(BlockDriverState *bs,
 
     assert((offset & (BDRV_SECTOR_SIZE - 1)) == 0);
     assert((bytes & (BDRV_SECTOR_SIZE - 1)) == 0);
-    assert((bytes >> BDRV_SECTOR_BITS) <= BDRV_REQUEST_MAX_SECTORS);
+    assert(bytes <= BDRV_REQUEST_MAX_BYTES);
 
     assert(drv->bdrv_co_writev);
     ret = drv->bdrv_co_writev(bs, sector_num, nb_sectors, qiov,

block/mirror.c

@@ -1142,13 +1142,6 @@ static bool mirror_drained_poll(BlockJob *job)
     return !!s->in_flight;
 }
 
-static void mirror_attached_aio_context(BlockJob *job, AioContext *new_context)
-{
-    MirrorBlockJob *s = container_of(job, MirrorBlockJob, common);
-
-    blk_set_aio_context(s->target, new_context);
-}
-
 static void mirror_drain(BlockJob *job)
 {
     MirrorBlockJob *s = container_of(job, MirrorBlockJob, common);
@@ -1178,7 +1171,6 @@ static const BlockJobDriver mirror_job_driver = {
         .complete               = mirror_complete,
     },
     .drained_poll           = mirror_drained_poll,
-    .attached_aio_context   = mirror_attached_aio_context,
     .drain                  = mirror_drain,
 };
 
@@ -1196,7 +1188,6 @@ static const BlockJobDriver commit_active_job_driver = {
         .complete               = mirror_complete,
     },
     .drained_poll           = mirror_drained_poll,
-    .attached_aio_context   = mirror_attached_aio_context,
     .drain                  = mirror_drain,
 };
 
@@ -1612,6 +1603,7 @@ static void mirror_start_job(const char *job_id, BlockDriverState *bs,
          * ensure that. */
         blk_set_force_allow_inactivate(s->target);
     }
+    blk_set_allow_aio_context_change(s->target, true);
 
     s->replaces = g_strdup(replaces);
     s->on_source_error = on_source_error;

block/qcow2-cluster.c

@@ -796,8 +796,9 @@ int qcow2_alloc_compressed_cluster_offset(BlockDriverState *bs,
         return cluster_offset;
     }
 
-    nb_csectors = ((cluster_offset + compressed_size - 1) >> 9) -
-                  (cluster_offset >> 9);
+    nb_csectors =
+        (cluster_offset + compressed_size - 1) / QCOW2_COMPRESSED_SECTOR_SIZE -
+        (cluster_offset / QCOW2_COMPRESSED_SECTOR_SIZE);
 
     cluster_offset |= QCOW_OFLAG_COMPRESSED |
                       ((uint64_t)nb_csectors << s->csize_shift);

block/qcow2-refcount.c

@@ -1172,12 +1172,11 @@ void qcow2_free_any_clusters(BlockDriverState *bs, uint64_t l2_entry,
     switch (ctype) {
     case QCOW2_CLUSTER_COMPRESSED:
         {
-            int nb_csectors;
-            nb_csectors = ((l2_entry >> s->csize_shift) &
-                           s->csize_mask) + 1;
-            qcow2_free_clusters(bs,
-                (l2_entry & s->cluster_offset_mask) & ~511,
-                nb_csectors * 512, type);
+            int64_t offset = (l2_entry & s->cluster_offset_mask)
+                & QCOW2_COMPRESSED_SECTOR_MASK;
+            int size = QCOW2_COMPRESSED_SECTOR_SIZE *
+                (((l2_entry >> s->csize_shift) & s->csize_mask) + 1);
+            qcow2_free_clusters(bs, offset, size, type);
         }
         break;
     case QCOW2_CLUSTER_NORMAL:
@@ -1317,9 +1316,12 @@ int qcow2_update_snapshot_refcount(BlockDriverState *bs,
                         nb_csectors = ((entry >> s->csize_shift) &
                                        s->csize_mask) + 1;
                         if (addend != 0) {
+                            uint64_t coffset = (entry & s->cluster_offset_mask)
+                                & QCOW2_COMPRESSED_SECTOR_MASK;
                             ret = update_refcount(
-                                bs, (entry & s->cluster_offset_mask) & ~511,
-                                nb_csectors * 512, abs(addend), addend < 0,
+                                bs, coffset,
+                                nb_csectors * QCOW2_COMPRESSED_SECTOR_SIZE,
+                                abs(addend), addend < 0,
                                 QCOW2_DISCARD_SNAPSHOT);
                             if (ret < 0) {
                                 goto fail;
@@ -1635,9 +1637,10 @@ static int check_refcounts_l2(BlockDriverState *bs, BdrvCheckResult *res,
             nb_csectors = ((l2_entry >> s->csize_shift) &
                            s->csize_mask) + 1;
             l2_entry &= s->cluster_offset_mask;
-            ret = qcow2_inc_refcounts_imrt(bs, res,
-                                           refcount_table, refcount_table_size,
-                                           l2_entry & ~511, nb_csectors * 512);
+            ret = qcow2_inc_refcounts_imrt(
+                bs, res, refcount_table, refcount_table_size,
+                l2_entry & QCOW2_COMPRESSED_SECTOR_MASK,
+                nb_csectors * QCOW2_COMPRESSED_SECTOR_SIZE);
             if (ret < 0) {
                 goto fail;
             }

block/qcow2.c

@@ -4187,7 +4187,8 @@ qcow2_co_preadv_compressed(BlockDriverState *bs,
 
     coffset = file_cluster_offset & s->cluster_offset_mask;
     nb_csectors = ((file_cluster_offset >> s->csize_shift) & s->csize_mask) + 1;
-    csize = nb_csectors * 512 - (coffset & 511);
+    csize = nb_csectors * QCOW2_COMPRESSED_SECTOR_SIZE -
+        (coffset & ~QCOW2_COMPRESSED_SECTOR_MASK);
 
     buf = g_try_malloc(csize);
     if (!buf) {

block/qcow2.h

@@ -74,6 +74,10 @@
 #define MIN_CLUSTER_BITS 9
 #define MAX_CLUSTER_BITS 21
 
+/* Defined in the qcow2 spec (compressed cluster descriptor) */
+#define QCOW2_COMPRESSED_SECTOR_SIZE 512U
+#define QCOW2_COMPRESSED_SECTOR_MASK (~(QCOW2_COMPRESSED_SECTOR_SIZE - 1))
+
 /* Must be at least 2 to cover COW */
 #define MIN_L2_CACHE_SIZE 2 /* cache entries */
 

blockjob.c

@@ -81,10 +81,6 @@ BlockJob *block_job_get(const char *id)
     }
 }
 
-static void block_job_attached_aio_context(AioContext *new_context,
-                                           void *opaque);
-static void block_job_detach_aio_context(void *opaque);
-
 void block_job_free(Job *job)
 {
     BlockJob *bjob = container_of(job, BlockJob, job);
@@ -92,28 +88,10 @@ void block_job_free(Job *job)
 
     bs->job = NULL;
     block_job_remove_all_bdrv(bjob);
-    blk_remove_aio_context_notifier(bjob->blk,
-                                    block_job_attached_aio_context,
-                                    block_job_detach_aio_context, bjob);
     blk_unref(bjob->blk);
     error_free(bjob->blocker);
 }
 
-static void block_job_attached_aio_context(AioContext *new_context,
-                                           void *opaque)
-{
-    BlockJob *job = opaque;
-    const JobDriver *drv = job->job.driver;
-    BlockJobDriver *bjdrv = container_of(drv, BlockJobDriver, job_driver);
-
-    job->job.aio_context = new_context;
-    if (bjdrv->attached_aio_context) {
-        bjdrv->attached_aio_context(job, new_context);
-    }
-
-    job_resume(&job->job);
-}
-
 void block_job_drain(Job *job)
 {
     BlockJob *bjob = container_of(job, BlockJob, job);
@@ -126,23 +104,6 @@ void block_job_drain(Job *job)
     }
 }
 
-static void block_job_detach_aio_context(void *opaque)
-{
-    BlockJob *job = opaque;
-
-    /* In case the job terminates during aio_poll()... */
-    job_ref(&job->job);
-
-    job_pause(&job->job);
-
-    while (!job->job.paused && !job_is_completed(&job->job)) {
-        job_drain(&job->job);
-    }
-
-    job->job.aio_context = NULL;
-    job_unref(&job->job);
-}
-
 static char *child_job_get_parent_desc(BdrvChild *c)
 {
     BlockJob *job = c->opaque;
@@ -183,11 +144,46 @@ static void child_job_drained_end(BdrvChild *c)
     job_resume(&job->job);
 }
 
+static bool child_job_can_set_aio_ctx(BdrvChild *c, AioContext *ctx,
+                                      GSList **ignore, Error **errp)
+{
+    BlockJob *job = c->opaque;
+    GSList *l;
+
+    for (l = job->nodes; l; l = l->next) {
+        BdrvChild *sibling = l->data;
+        if (!bdrv_child_can_set_aio_context(sibling, ctx, ignore, errp)) {
+            return false;
+        }
+    }
+    return true;
+}
+
+static void child_job_set_aio_ctx(BdrvChild *c, AioContext *ctx,
+                                  GSList **ignore)
+{
+    BlockJob *job = c->opaque;
+    GSList *l;
+
+    for (l = job->nodes; l; l = l->next) {
+        BdrvChild *sibling = l->data;
+        if (g_slist_find(*ignore, sibling)) {
+            continue;
+        }
+        *ignore = g_slist_prepend(*ignore, sibling);
+        bdrv_set_aio_context_ignore(sibling->bs, ctx, ignore);
+    }
+
+    job->job.aio_context = ctx;
+}
+
 static const BdrvChildRole child_job = {
     .get_parent_desc    = child_job_get_parent_desc,
     .drained_begin      = child_job_drained_begin,
     .drained_poll       = child_job_drained_poll,
     .drained_end        = child_job_drained_end,
+    .can_set_aio_ctx    = child_job_can_set_aio_ctx,
+    .set_aio_ctx        = child_job_set_aio_ctx,
     .stay_at_node       = true,
 };
 
@@ -438,8 +434,7 @@ void *block_job_create(const char *job_id, const BlockJobDriver *driver,
 
     bdrv_op_unblock(bs, BLOCK_OP_TYPE_DATAPLANE, job->blocker);
 
-    blk_add_aio_context_notifier(blk, block_job_attached_aio_context,
-                                 block_job_detach_aio_context, job);
+    blk_set_allow_aio_context_change(blk, true);
 
     /* Only set speed when necessary to avoid NotSupported error */
     if (speed != 0) {

hw/block/nvme.c

@@ -238,7 +238,7 @@ static uint16_t nvme_dma_read_prp(NvmeCtrl *n, uint8_t *ptr, uint32_t len,
         }
         qemu_sglist_destroy(&qsg);
     } else {
-        if (unlikely(qemu_iovec_to_buf(&iov, 0, ptr, len) != len)) {
+        if (unlikely(qemu_iovec_from_buf(&iov, 0, ptr, len) != len)) {
             trace_nvme_err_invalid_dma();
             status = NVME_INVALID_FIELD | NVME_DNR;
         }

include/block/block.h

@@ -586,6 +586,16 @@ void bdrv_coroutine_enter(BlockDriverState *bs, Coroutine *co);
  * This function must be called with iothread lock held.
  */
 void bdrv_set_aio_context(BlockDriverState *bs, AioContext *new_context);
+void bdrv_set_aio_context_ignore(BlockDriverState *bs,
+                                 AioContext *new_context, GSList **ignore);
+int bdrv_try_set_aio_context(BlockDriverState *bs, AioContext *ctx,
+                             Error **errp);
+int bdrv_child_try_set_aio_context(BlockDriverState *bs, AioContext *ctx,
+                                   BdrvChild *ignore_child, Error **errp);
+bool bdrv_child_can_set_aio_context(BdrvChild *c, AioContext *ctx,
+                                    GSList **ignore, Error **errp);
+bool bdrv_can_set_aio_context(BlockDriverState *bs, AioContext *ctx,
+                              GSList **ignore, Error **errp);
 int bdrv_probe_blocksizes(BlockDriverState *bs, BlockSizes *bsz);
 int bdrv_probe_geometry(BlockDriverState *bs, HDGeometry *geo);
 

include/block/block_int.h

@@ -691,6 +691,10 @@ struct BdrvChildRole {
      * can update its reference. */
     int (*update_filename)(BdrvChild *child, BlockDriverState *new_base,
                            const char *filename, Error **errp);
+
+    bool (*can_set_aio_ctx)(BdrvChild *child, AioContext *ctx,
+                            GSList **ignore, Error **errp);
+    void (*set_aio_ctx)(BdrvChild *child, AioContext *ctx, GSList **ignore);
 };
 
 extern const BdrvChildRole child_file;
@@ -962,27 +966,6 @@ void bdrv_parse_filename_strip_prefix(const char *filename, const char *prefix,
 void bdrv_add_before_write_notifier(BlockDriverState *bs,
                                     NotifierWithReturn *notifier);
 
-/**
- * bdrv_detach_aio_context:
- *
- * May be called from .bdrv_detach_aio_context() to detach children from the
- * current #AioContext.  This is only needed by block drivers that manage their
- * own children.  Both ->file and ->backing are automatically handled and
- * block drivers should not call this function on them explicitly.
- */
-void bdrv_detach_aio_context(BlockDriverState *bs);
-
-/**
- * bdrv_attach_aio_context:
- *
- * May be called from .bdrv_attach_aio_context() to attach children to the new
- * #AioContext.  This is only needed by block drivers that manage their own
- * children.  Both ->file and ->backing are automatically handled and block
- * drivers should not call this function on them explicitly.
- */
-void bdrv_attach_aio_context(BlockDriverState *bs,
-                             AioContext *new_context);
-
 /**
  * bdrv_add_aio_context_notifier:
  *

include/sysemu/block-backend.h

@@ -103,6 +103,7 @@ int blk_set_perm(BlockBackend *blk, uint64_t perm, uint64_t shared_perm,
 void blk_get_perm(BlockBackend *blk, uint64_t *perm, uint64_t *shared_perm);
 
 void blk_set_allow_write_beyond_eof(BlockBackend *blk, bool allow);
+void blk_set_allow_aio_context_change(BlockBackend *blk, bool allow);
 void blk_iostatus_enable(BlockBackend *blk);
 bool blk_iostatus_is_enabled(const BlockBackend *blk);
 BlockDeviceIoStatus blk_iostatus(const BlockBackend *blk);

qemu-img.texi

@@ -230,6 +230,7 @@ overridden with a pattern byte specified by @var{pattern}.
 
 Perform a consistency check on the disk image @var{filename}. The command can
 output in the format @var{ofmt} which is either @code{human} or @code{json}.
+The JSON output is an object of QAPI type @code{ImageCheck}.
 
 If @code{-r} is specified, qemu-img tries to repair any inconsistencies found
 during the check. @code{-r leaks} repairs only cluster leaks, whereas
@@ -406,8 +407,7 @@ The size syntax is similar to dd(1)'s size syntax.
 Give information about the disk image @var{filename}. Use it in
 particular to know the size reserved on disk which can be different
 from the displayed size. If VM snapshots are stored in the disk image,
-they are displayed too. The command can output in the format @var{ofmt}
-which is either @code{human} or @code{json}.
+they are displayed too.
 
 If a disk image has a backing file chain, information about each disk image in
 the chain can be recursively enumerated by using the option @code{--backing-chain}.
@@ -424,6 +424,51 @@ To enumerate information about each disk image in the above chain, starting from
 qemu-img info --backing-chain snap2.qcow2
 @end example
 
+The command can output in the format @var{ofmt} which is either @code{human} or
+@code{json}.  The JSON output is an object of QAPI type @code{ImageInfo}; with
+@code{--backing-chain}, it is an array of @code{ImageInfo} objects.
+
+@code{--output=human} reports the following information (for every image in the
+chain):
+@table @var
+@item image
+The image file name
+
+@item file format
+The image format
+
+@item virtual size
+The size of the guest disk
+
+@item disk size
+How much space the image file occupies on the host file system (may be shown as
+0 if this information is unavailable, e.g. because there is no file system)
+
+@item cluster_size
+Cluster size of the image format, if applicable
+
+@item encrypted
+Whether the image is encrypted (only present if so)
+
+@item cleanly shut down
+This is shown as @code{no} if the image is dirty and will have to be
+auto-repaired the next time it is opened in qemu.
+
+@item backing file
+The backing file name, if present
+
+@item backing file format
+The format of the backing file, if the image enforces it
+
+@item Snapshot list
+A list of all internal snapshots
+
+@item Format specific information
+Further information whose structure depends on the image format.  This section
+is a textual representation of the respective @code{ImageInfoSpecific*} QAPI
+object (e.g. @code{ImageInfoSpecificQCow2} for qcow2 images).
+@end table
+
 @item map [--object @var{objectdef}] [--image-opts] [-f @var{fmt}] [--output=@var{ofmt}] [-U] @var{filename}
 
 Dump the metadata of image @var{filename} and its backing file chain.
@@ -485,7 +530,8 @@ Calculate the file size required for a new image.  This information can be used
 to size logical volumes or SAN LUNs appropriately for the image that will be
 placed in them.  The values reported are guaranteed to be large enough to fit
 the image.  The command can output in the format @var{ofmt} which is either
-@code{human} or @code{json}.
+@code{human} or @code{json}.  The JSON output is an object of QAPI type
+@code{BlockMeasureInfo}.
 
 If the size @var{N} is given then act as if creating a new empty image file
 using @command{qemu-img create}.  If @var{filename} is given then act as if

qemu-io-cmds.c

@@ -538,7 +538,7 @@ static int do_write_compressed(BlockBackend *blk, char *buf, int64_t offset,
 {
     int ret;
 
-    if (bytes >> 9 > BDRV_REQUEST_MAX_SECTORS) {
+    if (bytes > BDRV_REQUEST_MAX_BYTES) {
         return -ERANGE;
     }
 
@@ -1781,10 +1781,9 @@ static int discard_f(BlockBackend *blk, int argc, char **argv)
     if (bytes < 0) {
         print_cvtnum_err(bytes, argv[optind]);
         return bytes;
-    } else if (bytes >> BDRV_SECTOR_BITS > BDRV_REQUEST_MAX_SECTORS) {
+    } else if (bytes > BDRV_REQUEST_MAX_BYTES) {
         printf("length cannot exceed %"PRIu64", given %s\n",
-               (uint64_t)BDRV_REQUEST_MAX_SECTORS << BDRV_SECTOR_BITS,
-               argv[optind]);
+               (uint64_t)BDRV_REQUEST_MAX_BYTES, argv[optind]);
         return -EINVAL;
     }
 

qmp.c

@@ -156,6 +156,9 @@ void qmp_cont(Error **errp)
         return;
     } else if (runstate_check(RUN_STATE_SUSPENDED)) {
         return;
+    } else if (runstate_check(RUN_STATE_FINISH_MIGRATE)) {
+        error_setg(errp, "Migration is not finalized yet");
+        return;
     }
 
     for (blk = blk_next(NULL); blk; blk = blk_next(blk)) {

tests/qemu-iotests/169

@@ -102,12 +102,17 @@ class TestDirtyBitmapMigration(iotests.QMPTestCase):
             event = self.vm_a.event_wait('MIGRATION')
             if event['data']['status'] == 'completed':
                 break
+        while True:
+            result = self.vm_a.qmp('query-status')
+            if (result['return']['status'] == 'postmigrate'):
+                break
 
         # test that bitmap is still here
         removed = (not migrate_bitmaps) and persistent
         self.check_bitmap(self.vm_a, False if removed else sha256)
 
-        self.vm_a.qmp('cont')
+        result = self.vm_a.qmp('cont')
+        self.assert_qmp(result, 'return', {})
 
         # test that bitmap is still here after invalidation
         self.check_bitmap(self.vm_a, sha256)

tests/qemu-iotests/221

@@ -1,6 +1,7 @@
 #!/usr/bin/env bash
 #
 # Test qemu-img vs. unaligned images
+# (See also 253, which is the O_DIRECT version)
 #
 # Copyright (C) 2018-2019 Red Hat, Inc.
 #
@@ -37,6 +38,9 @@ _supported_fmt raw
 _supported_proto file
 _supported_os Linux
 
+_default_cache_mode writeback
+_supported_cache_modes writeback writethrough unsafe
+
 echo
 echo "=== Check mapping of unaligned raw image ==="
 echo