On Mon, Sep 08, 2014 at 06:06:09PM +0200, Jakub Hrozek wrote:
On Wed, Aug 27, 2014 at 03:55:53PM +0200, Lukas Slebodnik wrote:
> On (27/08/14 14:24), Michal Židek wrote:
> >On 08/22/2014 05:38 PM, Jakub Hrozek wrote:
> >>On Thu, Aug 21, 2014 at 07:05:06PM +0200, Michal Židek wrote:
> >>>this is not very important patch, but suppresses
> >>>behaviour that is bothering some of us for quite
> >>>some time.
> >>>See patch description for details.
> >>This is not a solution, though, is it? Do we know why does the ordering
> >>change sometimes?
> >I am not sure if there is a real problem, maybe
> >the current version of the generated ifp files in git
> >were created by older version of the generator. If
> >so, then this is a solution. If not, than it at least
> >suppresses the noise.
> I tested few distributions: fedora-21-x86_64, fedora-20-i386,
> CentOS-6-x86_64, debian-testing-x86_64.
> After compilation, there was modified generated file
> src/responder/ifp/ifp_iface_generated.c, bue every time md5sum of generated
> file was the same.
> root@6c488ad30b66:/root/sssd# md5sum src/responder/ifp/ifp_iface_generated.c
> a2bab612f7bd33ac8ad8e6ce0cebdeec src/responder/ifp/ifp_iface_generated.c
> md5sum with Michal's patch is the same too.
> I don't know if Michal's patch fixes problem. It can just suppresses a
> We can try apply patch on master adn we will see in future if there will be the
> same problem.
After some testing, I concur. I wasn't able to reproduce the diff after
applying Michal's patch. It's possible we goofed something up when we
backported the IFP patches to 1.11.
So I'm ACKing the patch and I'll commit the fix. Chances are it's not
the real deal, but it's a marked improvement.
* master: 75c465e1f9e047a981b29634b77b50dc684c4c89