On 2013-12-20 22:04:05 +0100, Andres Freund wrote:
> Hi,
>
> On 2013-12-18 15:23:23 -0500, Robert Haas wrote:
> > It sounds like most people who have looked at this stuff are broadly
> > happy with it, so I'd like to push on toward commit soon, but it'd be
> > helpful, Andres, if you could review the comment additions to
> > shm-mq-v2.patch and see whether those address your concerns. If so,
> > I'll see about improving the overall comments for shm-toc-v1.patch as
> > well to clarify the points that seem to have caused a bit of
> > confusion; specific thoughts on what ought to be covered there, or any
> > other review, is most welcome.
>
> Some things:
One more thing:
static uint64
shm_mq_get_bytes_written(volatile shm_mq *mq, bool *detached)
{ uint64 v;
SpinLockAcquire(&mq->mq_mutex); v = mq->mq_bytes_written; *detached = mq->mq_detached;
SpinLockRelease(&mq->mq_mutex);
return mq->mq_bytes_written;
}
Note how you're returning mq->mq_bytes_written instead of v.
Greetings,
Andres Freund
-- Andres Freund http://www.2ndQuadrant.com/PostgreSQL Development, 24x7 Support, Training &
Services