Re: GiST index performance

Matthew Wakeling <matthew(at)flymine(dot)org> writes:
> The gistnext total doesn't seem to correspond to the amount I get by
> adding up all the individual lines in gistnest.

Hmm, hadn't you determined that some other stuff was being inlined into
gistnext? I'm not really sure how opannotate displays such cases, but
this might be an artifact of that.

> However, yes it does seem like fmgr.c accounts for a large proportion of
> samples. Also, I still seem to be getting mcount, even after recompiling
> without --enable-profiling.

You must still have some -pg code in there someplace. Maybe you didn't
recompile bioseg.so, or even psql? Remember the libc counts you are
looking at are for system-wide usage of libc.

regards, tom lane