[FLASH-USERS] Two questions from a newbie, thanks!
Di Sheng
disheng222 at gmail.com
Fri Feb 6 11:04:49 EST 2015
Hi Po-Yu,
Thanks a lot for your help!
As for the first question, after adding "+hdf5typeio" (i.e., ./setup -auto
Cellular +hdf5typeio parallelIO=False), I still got the same errors
(Assertion `dataset >= 0' failed). In fact, when I use "./setup -auto
Cellular +ug", the errors will disappear and the execution goes well. There
must be some reason for it. I mean, should I adopt some "-with-unit=...."
for Cellular or some more settings for setup ?
As for the second question, I didn't set the group in the flash.par. Could
you tell me how to specify the group frequency?
Did I miss any other parameters?
ps. My objective is just to run this application normally, so as to
evaluate my own MPI middleware. That's it. I'll be happy as long as it can
be run, no matter what kind of settings/parameters are used.
Many thanks again!
Best
Sheng
http://www.mcs.anl.gov/~shdi
2015-02-06 9:45 GMT-06:00 pchang <pchang at lle.rochester.edu>:
> Hi Sheng,
> For your first question, you may try adding "+hdf5typeio" in your setup.
> For your second question, did you specify the radiation/opacity group
> frequency in the flash.par and xe-100grp-lte.imx? Are the specified the
> same?
>
> Po-Yu
>
>
> --
> Po-Yu Chang
>
> Postdoctoral Associate
> Department of Mechanical Engineering,
> Fusion Science Center for Extreme States of Matter,
> Laboratory for Laser Energetics
> University of Rochester
> 250 East River Road
> Rochester, New York 14623
> Phone: (585) 273-5179
> FAX: (585) 275-5960
> pchang at lle.rochester.edu
>
>
>
>
> On 2/5/2015 6:08 PM, Di Sheng wrote:
>
>> Hi everyone,
>> I am a newbie of FLASH. I ran into the following problems which already
>> bothered me for a long while. Could any one answer any one of them? many
>> thanks!
>>
>> 1. How to deal with "Assertion `dataset >= 0' failed"?
>>
>> "./setup -auto Cellular" success, and "make" is also successful. However,
>> when I run "mpirun -np 1 flash4", it shows the following errors:
>>
>> HDF5-DIAG: Error detected in HDF5 (1.8.14) thread 0:
>> #000: H5Ddeprec.c line 178 in H5Dcreate1(): not a location ID
>> major: Invalid arguments to routine
>> minor: Inappropriate type
>> #001: H5Gloc.c line 253 in H5G_loc(): invalid object ID
>> major: Invalid arguments to routine
>> minor: Bad value
>> flash4: io_h5create_dataset.c:57: io_h5create_dataset: Assertion `dataset
>> >= 0' failed.
>>
>> Program received signal SIGABRT: Process abort signal.
>> .......
>>
>> The same situation goes to the "Jeans" application (SimulationMain/Jeans).
>>
>> 2. As for MGDInfinite, I setup it as follows:
>> ./setup -auto MGDInfinite -1d +hdf5typeio +mtmmmt +mgd +uhd3t
>> -with-unit=physics/materialProperties/Opacity/OpacityMain/Constant
>> species=xe mgd_meshgroups=4 parallelIO=FALSE
>>
>> Then, I add the following lines to the flash.par:
>> sim_xeMassFrac = 1
>>
>> eos_xeEosType = "eos_tab"
>> eos_xeSubType = "ionmix4"
>> eos_xeTableFile = "xe-100grp-lte.imx"
>> However, when I run "mpirun -np 1 flash4", I got errors as follows:
>>
>> [eos_tabBrowseIonmix4Tables] IONMIX4 file found: xe-100grp-lte.imx
>> in eos_inittabulated, tableName = xe-100grp-lte.imx
>> in eos_inittabulated, groupName = -none-
>> At line 158 of file eos_tabReadIonmix4Tables.F90 (unit = 7, file =
>> 'xe-100grp-lte.imx')
>> Fortran runtime error: Bad value during integer read
>>
>> ------------
>> Best
>> Sheng
>>
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://flash.rochester.edu/pipermail/flash-users/attachments/20150206/159fc25e/attachment-0001.htm>
More information about the flash-users
mailing list