On 07.08.19 16:46, Kevin Wolf wrote: > mirror_top_bs is currently implicitly drained through its connection to > the source or the target node. However, the drain section for target_bs > ends early after moving mirror_top_bs from src to target_bs, so that > requests can already be restarted while mirror_top_bs is still present > in the chain, but has dropped all permissions and therefore runs into an > assertion failure like this: > > qemu-system-x86_64: block/io.c:1634: bdrv_co_write_req_prepare: > Assertion `child->perm & BLK_PERM_WRITE' failed. > > Keep mirror_top_bs drained until all graph changes have completed. > > Signed-off-by: Kevin Wolf <kw...@redhat.com> > --- > block/mirror.c | 6 +++++- > 1 file changed, 5 insertions(+), 1 deletion(-) > > diff --git a/block/mirror.c b/block/mirror.c > index 9f5c59ece1..642d6570cc 100644 > --- a/block/mirror.c > +++ b/block/mirror.c > @@ -656,7 +656,10 @@ static int mirror_exit_common(Job *job) > s->target = NULL; > > /* We don't access the source any more. Dropping any WRITE/RESIZE is > - * required before it could become a backing file of target_bs. */ > + * required before it could become a backing file of target_bs. Not > having > + * these permissions any more means that we can't allow any new requests > on > + * mirror_top_bs from now on, so keep it drained. */
You’re lucky Patchew broke or it would have complained about multi-line comments needing /* and */ on separate lines. :-) I’m perfectly happy with this style, though: Reviewed-by: Max Reitz <mre...@redhat.com> > + bdrv_drained_begin(mirror_top_bs); > bs_opaque->stop = true; > bdrv_child_refresh_perms(mirror_top_bs, mirror_top_bs->backing, > &error_abort); > @@ -724,6 +727,7 @@ static int mirror_exit_common(Job *job) > bs_opaque->job = NULL; > > bdrv_drained_end(src); > + bdrv_drained_end(mirror_top_bs); > s->in_drain = false; > bdrv_unref(mirror_top_bs); > bdrv_unref(src); >
signature.asc
Description: OpenPGP digital signature