On Wed, Jan 25, 2012 at 11:11 AM, Patrick Ohly <patrick.ohly(a)intel.com> wrote:
On Mi, 2012-01-25 at 02:32 +0100, Chris Kühl wrote:
> This was done yesterday (or 2 days ago now). It my be good to give a
> little more detailed update at this point.
Thanks, that's very useful.
> With the above changes, the initialization of and communication with
> the helper binary seems to work well. I'm continuing to work my way
> through the tests. Right now I'm dealing with getting autosync to work
> properly. This is requiring some additional communication between the
> helper and server binaries as well as reworking some code I changed
> when removing the priority queues in the server.
I wonder whether we should just leave it broken for now. I find the
whole autosync implementation unintuitive and already planned to rewrite
it from scratch when adding "sync when changes are detected". The
problem is that I don't know yet when I will have time for that. If I
don't find the time soon, it would block merging your work because
regressions in the master branch are not acceptable.
Ok, I'll keep that in mind. Looking at the AutoSync stuff is leading
me to fix other related issues (shutdown) so I'll see how far this