Hi Zheyong
In order to avoid problems with file sizes, I will include the files in a
comment on gitlab.
With regards
Joakim
________________________________
From: gpumd-bounce@xxxxxxxxxxxxx <gpumd-bounce@xxxxxxxxxxxxx> on behalf of
Bruce Fan <brucenju@xxxxxxxxx>
Sent: Monday, January 13, 2020 10:54:03 AM
To: gpumd@xxxxxxxxxxxxx <gpumd@xxxxxxxxxxxxx>
Subject: [gpumd] Re: Calculations stop unexpectedly without error message.
From you log file, I can only infer that the code encounters problems for the
inputs from the run.in<http://run.in> file, but I am not clear where it is. Is
it convenient for you to send me your input files? Perhaps only the second
order force constant (should be a small file?) data to check the exact place of
the error?
Zheyong
On Mon, Jan 13, 2020 at 11:33 AM Bruce Fan
<brucenju@xxxxxxxxx<mailto:brucenju@xxxxxxxxx>> wrote:
But FCP is only in the developing version... I will have a closer look at your
problem...
Zheyong
Joakim Brorsson <joabro@xxxxxxxxxxx<mailto:joabro@xxxxxxxxxxx>> 于 2020年1月13日周一
11:29写道：
Hi
Yes, I can run the examples without experiencing any issues, which is why I am
a bit confused. Perhaps I should try to install the latest official release
instead and see if the same problem occurs or not.
With regards
Joakim
________________________________
From: gpumd-bounce@xxxxxxxxxxxxx<mailto:gpumd-bounce@xxxxxxxxxxxxx>
<gpumd-bounce@xxxxxxxxxxxxx<mailto:gpumd-bounce@xxxxxxxxxxxxx>> on behalf of
Bruce Fan <brucenju@xxxxxxxxx<mailto:brucenju@xxxxxxxxx>>
Sent: Monday, January 13, 2020 10:20:47 AM
To: gpumd@xxxxxxxxxxxxx<mailto:gpumd@xxxxxxxxxxxxx>
<gpumd@xxxxxxxxxxxxx<mailto:gpumd@xxxxxxxxxxxxx>>
Subject: [gpumd] Re: Calculations stop unexpectedly without error message.
Hi Joakim, Oh, sounds like a bug, but have you tried the examples? Do they run
correctly?
Zheyong
Joakim Brorsson <joabro@xxxxxxxxxxx<mailto:joabro@xxxxxxxxxxx>> 于 2020年1月13日周一
11:09写道：
Hi
After cloning the latest version of the GPUMD gitlab repository, my
calculations does not seem to get past the "initializing positions and related
parameters" stage. Specfically the last lines in the log file are the following:
---------------------------------------------------------------
Finished initializing positions and related parameters.
---------------------------------------------------------------
Even so, there are no error messages and as can be seen from the
run.in<http://run.in> file I have actually defined a run step:
potential ols_original_model_1_2x2x2_orders_2-3_tol1e-08.txt 0 1 2
velocity 100
ensemble nvt_lan 100 100 200
time_step 2
dump_position 1000 format netcdf
compute 0 1000 1 temperature potential force
run 100000
Does anyone have any idea what the problem might be? Do you think the problem
is related to the (1) run script, (2) potential, (3) GPUMD compilation, (4)
graphics cards or (5) computer system? By the way I am using a potential file
created from a Hiphive force constant potential.
With regards
Joakim
