Re: Raising our compiler requirements for 9.6

On Wed, Aug 12, 2015 at 4:34 PM, Heikki Linnakangas <hlinnaka(at)iki(dot)fi> wrote:
> Andres didn't mention how big the performance benefit he saw with pgbench
> was, but I bet it was barely distinguishible from noise. But that's OK. In
> fact, there's no reason to believe this would make any difference to
> performance. The point is to make the code more readable, and it certainly
> achieves that.

I think that when Bruce macro-ized this ten years ago or whenever it
was, he got a significant performance benefit from it; otherwise I
don't think he would have done it. It may well be that the march of
time has improved compiler technology to the point where no benefit
remains, and that's fine. But just as with any other part of the
code, we shouldn't start with the premise that the existing code is
bad the way it is. If a careful analysis leads us to that conclusion,
that's fine.

In this particular case, I am more concerned about making sure that
people who have an opinion get a chance to air it than I am with the
outcome. I have no reason to believe that we shouldn't make this
change; I merely want to make sure that anyone who does have a concern
about this (or other changes in this area) has a chance to be heard
before we get too far down the path. Nobody's going to want to turn
back the clock once we do this.

--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company