On Mon, Apr 1, 2013 at 11:55:07AM -0500, Merlin Moncure wrote:
> > In fact, BufFreelistLock is really misnamed, because for the most
> > part, the "free list" as we implement is going to be empty. What the
> > BufFreelistLock is really doing is serializing the process of scanning
> > for a free buffer. I think THAT is the problem. If we could arrange
> > things so as to hold BufFreelistLock only for the amount of time
> > needed to remove a buffer from a freelist ... we'd probably buy
> > ourselves quite a bit.
>
> right. I'm imaging a buffer scan loop that looks something like
> (uncompiled, untested) this. "TryLockBufHdr" does a simple TAS
> without spin, returning the lock state (well, true if it acquired the
> lock). usage_count is specifically and deliberately adjusted without
> having a lock on the buffer header (this would require some careful
> testing and possible changes elsewhere):
TAS does a CPU 'lock' instruction which affects the cpu cache. Why not
just read the value with no lock?
-- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB
http://enterprisedb.com
+ It's impossible for everything to be true. +