Im hesitant to merge this patch without further testing that it doesn't impact ipc behavior.
Regards -steve On Mon, 2009-03-30 at 20:28 +0200, Jim Meyering wrote: > From: Jim Meyering <[email protected]> > > * exec/coroipcs.c (pthread_ipc_consumer): Return NULL rather than > calling pthread_exit explicitly. When a pthread_create callback > function like this one returns NULL, it's equivalent to calling > pthread_exit(0). > Remove unreached call to pthread_exit after infloop. > --- > exec/coroipcs.c | 5 ++--- > 1 files changed, 2 insertions(+), 3 deletions(-) > > diff --git a/exec/coroipcs.c b/exec/coroipcs.c > index 4be798f..ab05bf7 100644 > --- a/exec/coroipcs.c > +++ b/exec/coroipcs.c > @@ -264,7 +264,7 @@ static void *pthread_ipc_consumer (void *conn) > struct sched_param sched_param; > > sched_param.sched_priority = api->sched_priority; > - res = pthread_setschedparam (conn_info->thread, SCHED_RR, > &sched_param); > + pthread_setschedparam (conn_info->thread, SCHED_RR, > &sched_param); > } > > for (;;) { > @@ -282,7 +282,7 @@ retry_semop: > } else > if ((res == -1) && (errno == EINVAL || errno == EIDRM)) { > coroipcs_refcount_dec (conn_info); > - pthread_exit (0); > + return NULL; > } > > coroipcs_refcount_inc (conn_info); > @@ -314,7 +314,6 @@ retry_semop: > api->sending_allowed_release > (conn_info->sending_allowed_private_data); > coroipcs_refcount_dec (conn); > } > - pthread_exit (0); > } > > static int _______________________________________________ Openais mailing list [email protected] https://lists.linux-foundation.org/mailman/listinfo/openais
