On 2012-12-16 16:44:04 +0000, Simon Riggs wrote:
> On 13 December 2012 20:03, Andres Freund <andres@2ndquadrant.com> wrote:
>
> > Does anybody have an opinion on the attached patches? Especially 0001,
> > which contains the procarray changes?
> >
> > It moves a computation of the sort of:
> >
> > result -= vacuum_defer_cleanup_age;
> > if (!TransactionIdIsNormal(result))
> > result = FirstNormalTransactionId;
> >
> > inside ProcArrayLock. But I can't really imagine that to be relevant...
>
> I don't see why this is hard.
>
> Just make the lock acquisition/release conditional on another parameter.
>
> That way the only thing you'll be moving inside the lock is an if test
> on a constant boolean.
Thats not really cheaper. Two branches + additional parameter
passed/pushed vs one branch, one subtransaction, two assignments is a
close call.
As I don't think either really matters in the GetOldestXmin case, I
would be happy with that as well. If people prefer an additional
parameter + adjusting the few callsite vs. a separate function I will go
that way.
Greetings,
Andres Freund
--Andres Freund http://www.2ndQuadrant.com/PostgreSQL Development, 24x7 Support, Training &
Services