Hello, I am trying to do BSE calculation, on preceding step of the calculation, I faced a bug when I perform a low scaling, single shot GW calculation: G0W0R with ALGO=G0W0R ‘internal error in: GG_base.F at line: 204
internal error in GDES_SET: NRPLWV not dividable by cores 6032 52 313651
If you are not a developer, you should not encounter this problem. Please submit a bug report”
All input and output files are put in tar file.
Thank You
vasp.6.2.1: internal error in: GG_base.F at line: 204
Moderators: Global Moderator, Moderator
-
dinesh_acharya1
- Newbie
- Posts: 2
- Joined: Mon Dec 25, 2023 7:57 am
vasp.6.2.1: internal error in: GG_base.F at line: 204
You do not have the required permissions to view the files attached to this post.
-
michael_wolloch
- Global Moderator
- Posts: 58
- Joined: Tue Oct 17, 2023 10:17 am
Re: vasp.6.2.1: internal error in: GG_base.F at line: 204
Hi dinesh_acharya1,
thanks a lot for reporting a bug. With your input, I can also reproduce it in the current 6.4.3 version.
The actual problem is that you have set NCORE > 1, which is not supported for GW calculations. If you comment out that line in your INCAR, the BUG message will disappear.
Of course, the code should warn you about the issue and exit gracefully instead of producing this bug. It does print a warning if GW is started with NCORE > 1 if more than one k points are used. But it should do so even if you use the gamma-only version. We will fix this in the next release and I thank you again for bringing it to our attention.
As a side note, I noticed that your ENCUT is set to a value lower than the highest EMIN value in your POTCAR. This is probably not a good idea.
Cheers, Michael
thanks a lot for reporting a bug. With your input, I can also reproduce it in the current 6.4.3 version.
The actual problem is that you have set NCORE > 1, which is not supported for GW calculations. If you comment out that line in your INCAR, the BUG message will disappear.
Of course, the code should warn you about the issue and exit gracefully instead of producing this bug. It does print a warning if GW is started with NCORE > 1 if more than one k points are used. But it should do so even if you use the gamma-only version. We will fix this in the next release and I thank you again for bringing it to our attention.
As a side note, I noticed that your ENCUT is set to a value lower than the highest EMIN value in your POTCAR. This is probably not a good idea.
Cheers, Michael