[PATCH] Fix several memory and resource leaks

[Jeremy Allison]

On Fri, Aug 10, 2018 at 04:36:43PM +0200, Andreas Schneider wrote:
> On Friday, 10 August 2018 13:31:33 CEST Volker Lendecke wrote:
> > On Fri, Aug 10, 2018 at 09:56:05AM +0200, Andreas Schneider via samba-
> technical wrote:
> > > On Thursday, 9 August 2018 21:01:31 CEST Jeremy Allison wrote:
> > > > On Thu, Aug 09, 2018 at 11:16:08AM -0700, Jeremy Allison via
> > > > samba-technical> 
> > > wrote:
> > > > > Ah, Looks like Volker also had some good comments.
> > > > > 
> > > > > I won't push anything, I'll wait until you
> > > > > address both his and my comments (some of
> > > > > which were the same :-).
> > > > > 
> > > > > Feel free to push the ones that both Volker
> > > > > and I reviewed without comment though :-).
> > > > 
> > > > Just to be helpful, these are the ones that
> > > > both Volker and I +1'ed (although you might
> > > > want to look at his comments about re-arranging
> > > > around the 'continue's in the last patch.
> > > 
> > > Attached is an updated patchset. There are 4 patches with need to be
> > > reviewed again.
> > > 
> > > I've dropped one patch which was a false positive. I've reordered the
> > > patches that the already reviewed patchsets are at the beginning. When I
> > > modified a patch I removed the RB tags.
> > 
> > You didn't like my comment on rearranging the nss_init() code? As I
> > said, the code looks correct, but the flow is more confusing to me
> > than it needs to be.
> 
> I would keep the patch as it is, but would add an additional patch for master 
> which uses a talloc_strackframe(). This would mean we rewrite the code and 
> pass a talloc memory context to parse_nss_parm(). Then we just need one 
> TALLOC_FREE(frame) before we leave.

All the others RB+ me, and pushed. I'll let you fix
this one with Volker :-).

Thanks !

Jeremy.