On Tue, May 22, 2012 at 11:36 PM, Ants Aasma <ants@cybertec.at> wrote:
> ... The free list
> itself is a bit trickier, but if it's still necessary/useful then
> SC->firstFreeBuffer and buf->freeNext are in effect a linked-list
> stack, there should plenty of tested lock free algorithms floating
> around for that. (btw. lastFreeBuffer looks like dead code, is that
> correct?)
Thinking about it a bit more, if the freelist is mostly empty, a
simpler alternative would be to make an unprotected read to check
SC->firstFreeBuffer and only acquire BufFreelistLock if there's
anything to pop. This would reduce the lock free parts to just
atomically incrementing a variable.
Ants Aasma
--
Cybertec Schönig & Schönig GmbH
Gröhrmühlgasse 26
A-2700 Wiener Neustadt
Web: http://www.postgresql-support.de