No results of G0W0

[sdwang]

Dear:
I performed GW calculation used v-3.2.2-r.616, when the process finished, there is no quasiparticle energy o.qp in my work directory,and it also does not exist in report.
My procedure is:
>yambo -i
>yambo -N
>yambo -g n -p p
>yambo -N
and I have attatched my log and report.
I hope you can help me.Thanks!
SDwang

[Daniele Varsano]

Dear Shudong,
it could be a memory issues, as well as other problem.
Please, post the input file, and eventually any standard
output and standard error of the failed run, as with the
file you sent it is not easy to understand what's going wrong.

Cheers,

Daniele

[sdwang]

I post my input for qp calculation and r_setup for init. If it run well, the "Game over ..." will appear in log, but in my one, there is no such words and the process terminated with no report error at all .
sdwang

[Daniele Varsano]

Dear Shudong,
I can't see nothing wrong. You can check if it is a memory issues:
try to run again your input file lowering the FFTGvecs variable.

Best,
Daniele

[sdwang]

When I run this with no revised input file under v-3.2.1-r.44,it run well .But the v-3.2.2-r.6 is not well.
sdwang

[andrea marini]

Dear Shudong Wang, with these little informations I cannot really trace back where the problem is. I have tried to use your input using the latest revision (6) without finding any problem. So the problem may be due to several reasons (compiler, system you're using, bug ...). We do need much more information:

1. The code does not reach the end, but it stops immediately after the PPA loop. What is exactly the error message ? Seg. fault, nothing ?
2. Can you reproduce the error on a much smaller system that we can try to reproduce ?
3. If you re-run the code now (with ndb.pp done) does the code stops again ?
4. What compiler, optimization are you using ?

Sorry but without more info is practically impossible for use to understand where the problem is.

[sdwang]

Thank you for your reply.
The fault is Seg. fault. Does this relate to compiler ? I compile r.6 the same system with v-3.2.1.
sdwang

[andrea marini]

Without reproducing the error there is no way I can help you. Please try to reproduce the error on a smaller system and give us the PW/abinit input.