block: Poll after drain on attaching a node

Commit dcf94a23b1 ('block: Don't poll in parent drain callbacks')
removed polling in bdrv_child_cb_drained_begin() on the grounds that the
original bdrv_drain() already will poll and BdrvChildRole.drained_begin
calls must not cause graph changes (and therefore must not call
aio_poll() or the recursion through the graph will break.

This reasoning is correct for calls through bdrv_do_drained_begin().
However, BdrvChildRole.drained_begin is also called when a node that is
already in a drained section (i.e. bdrv_do_drained_begin() has already
returned and therefore can't poll any more) is attached to a new parent.
In this case, we must explicitly poll to have all requests completed
before the drained new child can be attached to the parent.

In bdrv_replace_child_noperm(), we know that we're not inside the
recursion of bdrv_do_drained_begin() because graph changes are not
allowed there, and bdrv_replace_child_noperm() is a graph change. The
call of BdrvChildRole.drained_begin() must therefore be followed by a
BDRV_POLL_WHILE() that waits for the completion of requests.

Reported-by: Max Reitz <mreitz@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>

block.c

@@ -2066,7 +2066,7 @@ static void bdrv_replace_child_noperm(BdrvChild *child,
             }
             assert(num >= 0);
             for (i = 0; i < num; i++) {
-                child->role->drained_begin(child);
+                bdrv_parent_drained_begin_single(child, true);
             }
         }
 

block/io.c

@@ -52,9 +52,7 @@ void bdrv_parent_drained_begin(BlockDriverState *bs, BdrvChild *ignore,
         if (c == ignore || (ignore_bds_parents && c->role->parent_is_bds)) {
             continue;
         }
-        if (c->role->drained_begin) {
+        bdrv_parent_drained_begin_single(c, false);
-            c->role->drained_begin(c);
-        }
     }
 }
 
@@ -73,6 +71,14 @@ void bdrv_parent_drained_end(BlockDriverState *bs, BdrvChild *ignore,
     }
 }
 
+static bool bdrv_parent_drained_poll_single(BdrvChild *c)
+{
+    if (c->role->drained_poll) {
+        return c->role->drained_poll(c);
+    }
+    return false;
+}
+
 static bool bdrv_parent_drained_poll(BlockDriverState *bs, BdrvChild *ignore,
                                      bool ignore_bds_parents)
 {
@@ -83,14 +89,22 @@ static bool bdrv_parent_drained_poll(BlockDriverState *bs, BdrvChild *ignore,
         if (c == ignore || (ignore_bds_parents && c->role->parent_is_bds)) {
             continue;
         }
-        if (c->role->drained_poll) {
+        busy |= bdrv_parent_drained_poll_single(c);
-            busy |= c->role->drained_poll(c);
-        }
     }
 
     return busy;
 }
 
+void bdrv_parent_drained_begin_single(BdrvChild *c, bool poll)
+{
+    if (c->role->drained_begin) {
+        c->role->drained_begin(c);
+    }
+    if (poll) {
+        BDRV_POLL_WHILE(c->bs, bdrv_parent_drained_poll_single(c));
+    }
+}
+
 static void bdrv_merge_limits(BlockLimits *dst, const BlockLimits *src)
 {
     dst->opt_transfer = MAX(dst->opt_transfer, src->opt_transfer);

include/block/block.h

@@ -568,6 +568,14 @@ void bdrv_io_unplug(BlockDriverState *bs);
 void bdrv_parent_drained_begin(BlockDriverState *bs, BdrvChild *ignore,
                                bool ignore_bds_parents);
 
+/**
+ * bdrv_parent_drained_begin_single:
+ *
+ * Begin a quiesced section for the parent of @c. If @poll is true, wait for
+ * any pending activity to cease.
+ */
+void bdrv_parent_drained_begin_single(BdrvChild *c, bool poll);
+
 /**
  * bdrv_parent_drained_end:
  *

include/block/block_int.h

@@ -606,6 +606,9 @@ struct BdrvChildRole {
      * requests after returning from .drained_begin() until .drained_end() is
      * called.
      *
+     * These functions must not change the graph (and therefore also must not
+     * call aio_poll(), which could change the graph indirectly).
+     *
      * Note that this can be nested. If drained_begin() was called twice, new
      * I/O is allowed only after drained_end() was called twice, too.
      */