diff options
| author | Kevin Wolf <kwolf@redhat.com> | 2023-11-15 18:20:10 +0100 |
|---|---|---|
| committer | Kevin Wolf <kwolf@redhat.com> | 2023-11-21 12:45:21 +0100 |
| commit | 6bc0bcc89f847839cf3d459a55290dda8801d9d3 (patch) | |
| tree | 8e37ecf19e54f557f80666bb69d64f8df6a0803d /block/replication.c | |
| parent | bb092d6d8f53b9a5ce76e3f5a5a66e6a0475295f (diff) | |
block: Fix deadlocks in bdrv_graph_wrunlock()
bdrv_graph_wrunlock() calls aio_poll(), which may run callbacks that
have a nested event loop. Nested event loops can depend on other
iothreads making progress, so in order to allow them to make progress it
must not hold the AioContext lock of another thread while calling
aio_poll().
This introduces a @bs parameter to bdrv_graph_wrunlock() whose
AioContext is temporarily dropped (which matches bdrv_graph_wrlock()),
and a bdrv_graph_wrunlock_ctx() that can be used if the BlockDriverState
doesn't necessarily exist any more when unlocking.
This also requires a change to bdrv_schedule_unref(), which was relying
on the incorrectly taken lock. It needs to take the lock itself now.
While this is a separate bug, it can't be fixed a separate patch because
otherwise the intermediate state would either deadlock or try to release
a lock that we don't even hold.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Message-ID: <20231115172012.112727-3-kwolf@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
[kwolf: Fixed up bdrv_schedule_unref()]
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Diffstat (limited to 'block/replication.c')
| -rw-r--r-- | block/replication.c | 10 |
1 files changed, 5 insertions, 5 deletions
diff --git a/block/replication.c b/block/replication.c index 43e259444b..5ded5f1ca9 100644 --- a/block/replication.c +++ b/block/replication.c @@ -568,7 +568,7 @@ static void replication_start(ReplicationState *rs, ReplicationMode mode, &local_err); if (local_err) { error_propagate(errp, local_err); - bdrv_graph_wrunlock(); + bdrv_graph_wrunlock(bs); aio_context_release(aio_context); return; } @@ -579,7 +579,7 @@ static void replication_start(ReplicationState *rs, ReplicationMode mode, BDRV_CHILD_DATA, &local_err); if (local_err) { error_propagate(errp, local_err); - bdrv_graph_wrunlock(); + bdrv_graph_wrunlock(bs); aio_context_release(aio_context); return; } @@ -592,7 +592,7 @@ static void replication_start(ReplicationState *rs, ReplicationMode mode, if (!top_bs || !bdrv_is_root_node(top_bs) || !check_top_bs(top_bs, bs)) { error_setg(errp, "No top_bs or it is invalid"); - bdrv_graph_wrunlock(); + bdrv_graph_wrunlock(bs); reopen_backing_file(bs, false, NULL); aio_context_release(aio_context); return; @@ -600,7 +600,7 @@ static void replication_start(ReplicationState *rs, ReplicationMode mode, bdrv_op_block_all(top_bs, s->blocker); bdrv_op_unblock(top_bs, BLOCK_OP_TYPE_DATAPLANE, s->blocker); - bdrv_graph_wrunlock(); + bdrv_graph_wrunlock(bs); s->backup_job = backup_job_create( NULL, s->secondary_disk->bs, s->hidden_disk->bs, @@ -696,7 +696,7 @@ static void replication_done(void *opaque, int ret) s->secondary_disk = NULL; bdrv_unref_child(bs, s->hidden_disk); s->hidden_disk = NULL; - bdrv_graph_wrunlock(); + bdrv_graph_wrunlock(NULL); s->error = 0; } else { |