On Thu, Jul 15, 2010 at 11:50:15AM -0700, Steven Dake wrote: > shouldn't sync_aborted() only be called if sync_processing != 0?
That would be nice, but it is not acurate enough. sync_processing is set to 0 after syncv1 is done. Then syncv2 processing is done. If we get a config change after syncv1 is down, but before syncv2 is done then it won't get aborted. -Angus > > Regards > -steve > > > On 07/14/2010 08:56 PM, Angus Salkeld wrote: > >1) sync_callbacks.sync_abort can be null. > >2) sync_processing is set to 0 after syncv1 is done. > > Then syncv2 processing is down. If we get a config change > > after syncv1 is down, but before syncv2 is done then it won't > > get aborted. > > > >Signed-off-by: Angus Salkeld<[email protected]> > >--- > > exec/sync.c | 2 +- > > 1 files changed, 1 insertions(+), 1 deletions(-) > > > >diff --git a/exec/sync.c b/exec/sync.c > >index ce115a3..c4197a8 100644 > >--- a/exec/sync.c > >+++ b/exec/sync.c > >@@ -479,8 +479,8 @@ static void sync_confchg_fn ( > > memcpy (my_member_list, member_list, member_list_entries * sizeof > > (unsigned int)); > > my_member_list_entries = member_list_entries; > > > >+ sync_aborted (); > > if (sync_processing&& sync_callbacks.sync_abort != NULL) { > >- sync_aborted (); > > sync_callbacks.sync_abort (); > > sync_callbacks.sync_activate = NULL; > > } _______________________________________________ Openais mailing list [email protected] https://lists.linux-foundation.org/mailman/listinfo/openais
