Re: Transaction-scope advisory locks

On 2011-01-28 10:12 AM +0200, I wrote:
> I still didn't
> address the issue with pg_advisory_unlock_all() releasing transaction
> scoped locks, but I'm going to.

.. and here's the patch.  I'm not too confident with the code I added to 
storage/lmgr/lock.c, but it seems to be working.

Earlier there was some discussion about adding regression tests for 
advisory locks.  However, I don't see where they would fit in our 
current .sql files and adding a new one just for a few tests didn't seem 
right.  Anyone have an idea where they should go or should I just add a 
new one?


Regards,
Marko Tiikkaja