On Thu, Dec 15, 2016 at 10:39 PM, Richard Guy Briggs <rgb(a)redhat.com> wrote:
On 2016-12-15 22:12, Steve Grubb wrote:
> On Thursday, December 15, 2016 7:50:48 PM EST Paul Moore wrote:
> > On Thu, Dec 15, 2016 at 7:22 PM, Steve Grubb <sgrubb(a)redhat.com> wrote:
> > > I'm planning to replace all the config change logging with the
> > > audit_log_task_simple function I sent so that we have everything. Can we
> > > go ahead and pull that in so that we can start using it?
> >
> > There needs to be more than one user of the function to make it
> > worthwhile; so far that function has only been proposed with a single
> > user. Propose it with multiple users and we can look at it seriously.
>
> That's because I have several unrelated patches that use it. Do you want me to
> send all of them at once? There's going to be at least 5 users of the
> function. Possibly more. I want it to be the default for all future events
> added because it concisely gives the necessary information for well-formed
> events.
I'd send the audit_log_task_simple() patch alone, then send each feature
that uses it in a separate patch set. Failing that, send it as a
separate patch in the first patch set to make it available for all, then
follow it with more separate patchsets for other events.
There is a chicken and egg problem here.
The extremely safe way to do this would be to submit the unrelated
patches first, each written to *not* make use of your new
consolidation function, then submit a single patch which adds the
function and integrates it with the others.
This approach also has the advantage that you are able to submit fixes
as you run across then and not have to wait for everything. I know
you already have at least one patch ready to, you just need to remove
the references to audit_log_task_simple.
--
paul moore
www.paul-moore.com