Backport of gssproxy patches for Fedora 19

[Josh Boyer]

On Fri, Jun 07, 2013 at 08:24:41AM -0400, Simo Sorce wrote:
> On Fri, 2013-06-07 at 08:17 -0400, Josh Boyer wrote:
> > On Wed, Jun 05, 2013 at 01:09:39PM -0400, Simo Sorce wrote:
> > > On Tue, 2013-06-04 at 15:37 -0400, Josh Boyer wrote:
> > > > > > > I have a backport to 3.9.3 (which turned out to need a couple more
> > > > > > > patches in addition to the above) at
> > > > > > > 
> > > > > > > 	git://linux-nfs.org/~bfields/linux-topics.git 
> > > > > > 
> > > > > > Sorry, I forgot to give the branch: 3.9-gssproxy-backport
> > > > > 
> > > > > I'll grab them from there.  If I have troubles, I'll let you know.
> > > > 
> > > > Ok, they're applied in kernel-3.9.4-302.fc19.  Please test things out.
> > > 
> > > I am getting an oops, I will work with Bruce to find the cause and fix
> > > it.
> > 
> > Did you track this down?  We're probably going to be doing another build
> > to submit as an update today, and I'd rather not include known oopsing
> > patches.  If this doesn't have a fix yet, I'll temporarily remove the
> > patches for the update build.
> > 
> > josh
> 
> Bruce sent me 2 patches yesterday, but haven't got to test them yet.
> The code is not activated by default, so there is little harm in putting
> them in, the only thing that will fail is gssproxy, as the whole code
> path is enabled only if gssproxy is started at boot.

Did you get a chance to test them?  I'm doing an update today.

> But either way is fine, we are planning on verifying he fixes and make
> sure the bug is squashed today one way or another.

I'll go with disabling the patches unless you have fixes to apply.  The
functionality might not be used by default, but knowingly shipping code
that causes an oops when used isn't very responsible.

josh