On Tue, Nov 10, 2020 at 1:15 PM Tom Lane <tgl@sss.pgh.pa.us> wrote:
>
> Tomas Vondra <tomas.vondra@enterprisedb.com> writes:
> > That's unlikely, I think. The regression tests are constructed so that
> > the estimates are stable. It's more likely this is some difference in
> > rounding behavior, for example.
>
> The reported delta is in the actual row count, not an estimate.
> How could that be subject to roundoff issues? And there's no
> delta in the Append's inputs, so this seems like it's a flat-out
> miss of a row count in EXPLAIN ANALYZE.
>
> > I wonder which msvc builds are used on
> > the machines that fail/pass the tests, and if the compiler flags are the
> > same.
>
> Yeah, this might be a fruitful way to figure out "what's different
> from the buildfarm".
>
> regards, tom lane
Hmm..anyway I can help here? I don't believe I am using any special
compile options. I am using VS 2019. The thing is, both systems I have
use the same build. I call msvcvars to set up the environment:
"%ProgramFiles(x86)%\Microsoft Visual
Studio\2019\Enterprise\VC\Auxiliary\Build\vcvarsall.bat" amd64
I also saw some recent changes have been made around these tests, so I
can try the latest too.
