From mboxrd@z Thu Jan 1 00:00:00 1970 From: Bart Van Assche Subject: Re: [PATCH] Don't exit() from library Date: Tue, 11 Mar 2014 09:45:14 +0100 Message-ID: <531ECD1A.8050807@acm.org> References: <1394518211-8245-1-git-send-email-rrs@debian.org> Reply-To: device-mapper development Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <1394518211-8245-1-git-send-email-rrs@debian.org> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: dm-devel-bounces@redhat.com Errors-To: dm-devel-bounces@redhat.com To: device-mapper development , christophe.varoqui@opensvc.com Cc: Ritesh Raj Sarraf List-Id: dm-devel.ids On 03/11/14 07:10, Ritesh Raj Sarraf wrote: > Also call pthread_attr_destroy() on thread creation failure > > Signed-off-by: Ritesh Raj Sarraf > --- > libmpathpersist/mpath_persist.c | 11 +++++++---- > 1 file changed, 7 insertions(+), 4 deletions(-) > > diff --git a/libmpathpersist/mpath_persist.c b/libmpathpersist/mpath_persist.c > index bd30125..b74ce4e 100644 > --- a/libmpathpersist/mpath_persist.c > +++ b/libmpathpersist/mpath_persist.c > @@ -585,11 +585,14 @@ int send_prout_activepath(char * dev, int rq_servact, int rq_scope, > rc = pthread_create(&thread, &attr, mpath_prout_pthread_fn, (void *)(¶m)); > if (rc){ > condlog (3, "%s: failed to create thread %d", dev, rc); > - exit(-1); > + /* Destroy the thread attribute since we failed to create */ > + pthread_attr_destroy(&attr); > + } > + else { > + /* Free attribute and wait for the other threads */ > + pthread_attr_destroy(&attr); > + rc = pthread_join(thread, NULL); > } > - /* Free attribute and wait for the other threads */ > - pthread_attr_destroy(&attr); > - rc = pthread_join(thread, NULL); > > return (param.status); > } Hello Ritesh, Maybe there is something I'm not aware of, but is there any reason why the pthread_attr_destroy() call occurs twice in this patch instead of invoking it before the "if (rc) {" statement ? From the pthread_attr_destroy() man page: "Destroying a thread attributes object has no effect on threads that were created using that object". Thanks, Bart.