On 3/9/18 19:36, Tom Lane wrote: > I happened to notice that if you look at > https://coverage.postgresql.org/src/backend/optimizer/util/predtest.c.gcov.html > it claims that the two occurrences of > elog(ERROR, "predicate_classify returned a bogus value"); > at lines 492 and 803 are reached. > > This would be quite astonishing if true, but it isn't true, since > the regression tests aren't reporting any such failure. > > Needless to say, this puts quite a damper on my faith in the coverage > reports. Maybe that machine needs a software update?
I can't reproduce this locally, so yeah, maybe we should check whether the toolchain could be updated.
Is it specifically "lcov" we are considering here? If so, we are on version 1.13, and there is no newer version packaged for debian or from what I can tell even released?