Re: profiling connection overhead

Robert Haas <robertmhaas(at)gmail(dot)com> writes:
> On Wed, Nov 24, 2010 at 2:10 AM, Heikki Linnakangas
> <heikki(dot)linnakangas(at)enterprisedb(dot)com> wrote:
>> Micro-optimizing that search for the non-zero value helps a little bit
>> (attached). Reduces the percentage shown by oprofile from about 16% to 12%
>> on my laptop.

That "micro-optimization" looks to me like your compiler leaves
something to be desired.

> The first optimization that occurred to me was "remove the loop
> altogether".

Or make it execute only in assert-enabled mode, perhaps.

This check had some use back in the bad old days, but the ResourceOwner
mechanism has probably removed a lot of the argument for it.

The counter-argument might be that failing to remove a buffer pin would
be disastrous; but I can't see that it'd be worse than failing to remove
an LWLock, and we have no belt-and-suspenders-too loop for those.

regards, tom lane