[SSSD] [PATCH] krb5_child: fix value type and initialization

Jakub Hrozek jhrozek at redhat.com
Mon Mar 4 13:45:44 UTC 2013


On Mon, Mar 04, 2013 at 12:06:08PM +0100, Pavel Březina wrote:
> On 03/04/2013 11:46 AM, Ondrej Kos wrote:
> >On 03/04/2013 11:02 AM, Pavel Březina wrote:
> >>On 03/04/2013 10:55 AM, Ondrej Kos wrote:
> >>>On 03/04/2013 10:32 AM, Sumit Bose wrote:
> >>>>On Mon, Mar 04, 2013 at 10:12:05AM +0100, Ondrej Kos wrote:
> >>>>>I noticed the warning while testing another patch.
> >>>>>--
> >>>>>Ondrej Kos
> >>>>>Associate Software Engineer
> >>>>>Identity Management
> >>>>>Red Hat Czech
> >>>>>
> >>>>>phone: +420-532-294-558
> >>>>>cell:  +420-736-417-909
> >>>>>ext:   82-62558
> >>>>>loc:   1013 Brno 1 office
> >>>>>irc:   okos @ #sssd #brno
> >>>>
> >>>>> From f483339df40d1b69c0f19720b1802cfec4ede1c1 Mon Sep 17 00:00:00
> >>>>>2001
> >>>>>From: Ondrej Kos <okos at redhat.com>
> >>>>>Date: Mon, 4 Mar 2013 10:03:54 +0100
> >>>>>Subject: [PATCH] krb5_child: fix value type and initialization
> >>>>>
> >>>>>ret was defined as integer, instead of errno_t, and was uninitialized
> >>>>>---
> >>>>>  src/providers/krb5/krb5_child.c | 2 +-
> >>>>>  1 file changed, 1 insertion(+), 1 deletion(-)
> >>>>>
> >>>>>diff --git a/src/providers/krb5/krb5_child.c
> >>>>>b/src/providers/krb5/krb5_child.c
> >>>>>index
> >>>>>9182f2383dd7bcc686d0bcd4dfe696d990bfd150..431d21d1aed4fd93eb7af2b4f713e8affa4f0a4a
> >>>>>
> >>>>>
> >>>>>100644
> >>>>>--- a/src/providers/krb5/krb5_child.c
> >>>>>+++ b/src/providers/krb5/krb5_child.c
> >>>>>@@ -1899,7 +1899,7 @@ int main(int argc, const char *argv[])
> >>>>>      poptContext pc;
> >>>>>      int debug_fd = -1;
> >>>>>      int status;
> >>>>>-    int ret;
> >>>>>+    errno_t ret = EOK;
> >>>>
> >>>>If it is really necessary that a return code is initialized I would
> >>>>prefer to set it to an error value instead of EOK, to make sure an
> >>>>error
> >>>>is returned when something unexpected happens.
> >>>>
> >>>>In the case here ret is not set if debug_prg_name is NULL due to a
> >>>>failure of talloc_asprintf(). If you set ret to ENOMEM here, you do not
> >>>>need to initialize it and the compiler or Coverity can issue a warning
> >>>>in future if it wasn't set in a new code path.
> >>>>
> >>>>bye,
> >>>>Sumit
> >>>>>
> >>>>>      struct poptOption long_options[] = {
> >>>>>          POPT_AUTOHELP
> >>>>>--
> >>>>>1.8.1.2
> >>>>>
> >>>>
> >>>>>_______________________________________________
> >>>>>sssd-devel mailing list
> >>>>>sssd-devel at lists.fedorahosted.org
> >>>>>https://lists.fedorahosted.org/mailman/listinfo/sssd-devel
> >>>>
> >>>>_______________________________________________
> >>>>sssd-devel mailing list
> >>>>sssd-devel at lists.fedorahosted.org
> >>>>https://lists.fedorahosted.org/mailman/listinfo/sssd-devel
> >>>>
> >>>thanks, new patch attached
> >>
> >>Nack. See Jakub's comment.
> >
> >I haven't noticed it, and I kind of misunderstood what Sumit was
> >proposing, new patch attached.
> 
> Ack.

Pushed to master.



More information about the sssd-devel mailing list