Horton

Re: Reducing overhead of frequent table locks

Tom Lane <tgl@sss.pgh.pa.us>

View thread
From: Tom Lane <tgl@sss.pgh.pa.us>
To: Robert Haas <robertmhaas@gmail.com>
Cc: Noah Misch <noah@leadboat.com>, Alexey Klyukin <alexk@commandprompt.com>, pgsql-hackers@postgresql.org
Date: 2011-05-27T21:01:51Z
Lists: pgsql-hackers

Commits

Same data as JSON: GET /api/v1/messages/:b64id/commits the thread's linked commits as JSON, with link sources. API reference →

  1. Fix possible "tuple concurrently updated" error in ALTER TABLE.

Robert Haas <robertmhaas@gmail.com> writes:
> When a strong lock is taken or released, we have to increment or
> decrement strong_lock_counts[fasthashpartition].  Here's the question:
> is that atomic?  In other words, suppose that strong_lock_counts[42]
> starts out at 0, and two backends both do ++strong_lock_counts[42].
> Are we guaranteed to end up with "2" in that memory location or might
> we unluckily end up with "1"?  I think the latter is possible... and
> some guard is needed to make sure that doesn't happen.

There are "atomic increment" primitives on most/all multiprocessors,
although availing ourselves of them everywhere will take an amount of
work not unlike developing the spinlock primitives :-(.  You are dead
right that this is unsafe without that.

			regards, tom lane