On Mon, Jun 19, 2017 at 10:49 AM, Roland Haas rhaas@illinois.edu wrote:

Present: Steve, Roland, Ian, Roberto, Christian, Bill Gabella, Frank, Gabrielle, Eliu, Qian, Vassili

ET Release:

• still have outstanding problems before the release or even before large scale testing
• ML testsuites still failing when computing constraints

** need to do this before the release ** Peter, who looked into this, cannot work on it this week ** need a volunteer: Frank will approach persons in private, specifically persons funded by the ET grant

• SphericalHarminicGen

** two failing tests (Frank sees SEGFAULTs and HDF5 errors) ** Roland will make time to fix it

• MPI fails to build if needed

** Roland suggests to check if OS is linux in perl script then add librt and libutil as a workaround. Frank agrees. All agree, will test this.

I disagree. It's too close before the release to change the Perl script. Instead, we should recommend setting the respective options manually, as we are already doing on many systems. Since there is a work-around (that we've been using for years), this is not a critical fix.

-erik

• GRHydro

** still fails with NaN ** on Roland's plate

• want these updates until Wednesday
• test on machines, Frank request feedback for results on personal workstations
• name for release, pick one from: https://docs.einsteintoolkit.org/et-docs/Release_Details

Yours, Roland

-- My email is as private as my paper mail. I therefore support encrypting and signing email messages. Get my PGP key from http://pgp.mit.edu .

---

Users mailing list Users@einsteintoolkit.org http://lists.einsteintoolkit.org/mailman/listinfo/users