Commit 4cf077b5 authored by Kevin Wolf's avatar Kevin Wolf
Browse files

block: Remove aio_poll() in bdrv_drain_poll variants 



bdrv_drain_poll_top_level() was buggy because it didn't release the
AioContext lock of the node to be drained before calling aio_poll().
This way, callbacks called by aio_poll() would possibly take the lock a
second time and run into a deadlock with a nested AIO_WAIT_WHILE() call.

However, it turns out that the aio_poll() call isn't actually needed any
more. It was introduced in commit 91af091f, which is effectively
reverted by this patch. The cases it was supposed to fix are now covered
by bdrv_drain_poll(), which waits for block jobs to reach a quiescent
state.

Signed-off-by: default avatarKevin Wolf <kwolf@redhat.com>
Reviewed-by: default avatarFam Zheng <famz@redhat.com>
Reviewed-by: default avatarMax Reitz <mreitz@redhat.com>
parent b5a7a057

block/io.c

+0 −8
Original line number Diff line number Diff line
@@ -268,10 +268,6 @@ bool bdrv_drain_poll(BlockDriverState *bs, bool recursive, 
static bool bdrv_drain_poll_top_level(BlockDriverState *bs, bool recursive,

                                      BdrvChild *ignore_parent)

{

    /* Execute pending BHs first and check everything else only after the BHs

     * have executed. */

    while (aio_poll(bs->aio_context, false));



    return bdrv_drain_poll(bs, recursive, ignore_parent, false);

}
@@ -511,10 +507,6 @@ static bool bdrv_drain_all_poll(void) 
    BlockDriverState *bs = NULL;

    bool result = false;



    /* Execute pending BHs first (may modify the graph) and check everything

     * else only after the BHs have executed. */

    while (aio_poll(qemu_get_aio_context(), false));



    /* bdrv_drain_poll() can't make changes to the graph and we are holding the

     * main AioContext lock, so iterating bdrv_next_all_states() is safe. */

    while ((bs = bdrv_next_all_states(bs))) {