[ANL-Cooley] Read Integral cause Wrong Energy during selection

[TApplencourt]

If you wrote the AO and MO integrals after the SCF. And then read then in order to do an fci calculation this will give produce wrong value during the selection. The energy of the HF determinant is still correct.

This bug appear on Colley. I will investigate more in the following days.

How to reproduce

HF.xyz :

2
HF Geo: Experiment Mult: 1 symmetry: 10
F    0.0    0.0    0.0
H    0.0    0.0    0.9168

qp_create_ezfio_from_xyz HF.xyz -b 'cc-pvdz' -o HF.ezfio
qp_edit  HF.ezfio # Modify to wrote all the integral
qp_run SCF HF.ezfio
qp_run fci_zmq HF.ezfio # The PT2 will be wrong after the 2nd iteration. aka give not the same energy as is you recompute the integrals

PS : this is with the version 73de13320bf72983572d5e203ec46ec9d21f86ff

[TApplencourt]

I was not able to reproduce the problem with the last version of quantum_package (113f0757eee8f6f4ec58404fb2610a69368ca039).

We need to update the colley version and check it again.

[anbenali]

It is updated!

On Thu, Jun 29, 2017 at 9:03 AM, Thomas Applencourt < notifications@github.com> wrote:

I was not able to reproduce the problem with the last version of quantum_package (113f075 https://github.com/LCPQ/quantum_package/commit/113f0757eee8f6f4ec58404fb2610a69368ca039 ).

We need to update the colley version and check it again.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/LCPQ/quantum_package/issues/206#issuecomment-311976191, or mute the thread https://github.com/notifications/unsubscribe-auth/AXl6wwW3M4epZRdI8tdqlVEBwKr3imHuks5sI66zgaJpZM4OImOA .

--

Anouar Benali, PhD Leadership Computing Facility Argonne National Laboratory Building 240 Office - 2127 9700 S Cass Av., Argonne Il, 60439 (630) 252-0058

[TApplencourt]

Solved by sync the mmaped file