Serializable implementation milestone: table SIREAD locks without correct lifespan - Mailing list pgsql-hackers

From Kevin Grittner
Subject Serializable implementation milestone: table SIREAD locks without correct lifespan
Date
Msg-id 4B572C78020000250002E8A3@gw.wicourts.gov
Whole thread Raw
List pgsql-hackers
Attached is a patch for the next milestone on the Serializable wiki
page: changing the table-level predicate locks to SIREAD locks
without worrying about lifespan.  (Implementing correct lifespan is
next.)

The result of not worrying about it is that they aren't cleaned up
at all, even when the transaction ends and the connection is closed.
In a way, that's not all bad, because neither of those events
*should* remove these locks; so there's nothing to *undo* for the
next step.

As before, this is "for the record" and not a request for commit or
official review.  If anyone looks at it out of interest in this
effort, any feedback is welcome.

Applies cleanly to head and passes regression tests.  ;-)

-Kevin

Attachment

pgsql-hackers by date:

Previous
From: Jeff Davis
Date:
Subject: Re: Listen / Notify - what to do when the queue is full
Next
From: Joachim Wieland
Date:
Subject: Re: Listen / Notify - what to do when the queue is full