On 2020-01-06 07:31, Michael Paquier wrote:
> On Sun, Dec 29, 2019 at 11:33:34AM +0100, Peter Eisentraut wrote:
>> Attached are two patches to remove these two symbols. STATUS_FOUND can be
>> replaced by a simple bool. STATUS_WAITING is replaced by a separate enum.
>
> Patch 0001 looks good to me, but I got to wonder why the check after
> waitMask in LockAcquireExtended() is not done directly in
> LockCheckConflicts().
You mean put he subsequent GrantLock() calls into LockCheckConflicts()?
That would technically save some duplicate code, but it seems weird,
because LockCheckConflicts() is notionally a read-only function that
shouldn't change state.
--
Peter Eisentraut http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services