Tom Lane wrote:
> I wonder how trustworthy the measure of the visibilitymap_test call site
> as a consumer of cycles really is. I've frequently noticed that
> oprofile blames remarkably large fractions of the runtime on individual
> statements that appear to be quite trivial. I'm not sure if that
Are these statements perhaps near kernel calls?
-- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB
http://enterprisedb.com
+ It's impossible for everything to be true. +