Re: Transaction-scope advisory locks

On Thu, Feb 10, 2011 at 08:36, Marko Tiikkaja
<marko.tiikkaja@cs.helsinki.fi> wrote:
>> One issue might be in pg_locks
> Robert suggested not doing this for 9.1, and I don't have anything against
> that.

Agreed.

> Updated patch attached.

Looks good to commit. I note a few minor issues for committer:

* Functions listed in "Table 9-62. Advisory Lock Functions" might need
  sorted in alphabetical order.

* We could extend LockReleaseAll() to have the 3rd mode
  instead of LockReleaseSession().  Existing behavior is:
| LockReleaseAll(LOCKMETHODID lockmethodid, bool allLocks)
|   allLocks == true: release all locks including session locks.
|   allLocks == false: release all non-session locks.

* Or, we might have one subroutine for LockReleaseSession() and
  LockReleaseCurrentOwner(). They have similar codes.

-- 
Itagaki Takahiro