segmentation fault reading gadget snap

[jegarfa]

Hi, I would like to use ASOHF to produce halos from a gadget snapshot, however I am getting a Segmentation fault (core dumped) as soon as I run the code.

I changed the input parameters according to the instructions: GENERIC_READER=1, cosmology and Boxsize. I simlink the snapshot folder "simulations" and adapted the name of the input snapshot. I keep the default values for the rest of parameters.
The log file "asohf.out" is empty so I do not have a clue about what is happening.
have you had this issue before? do you know by chance what could be wrong (perhaps something in reading the snap)?

[dvallesp]

Hi Jorge,

A segmentation fault can be due to a broad variety of reasons, which my range from your hardware (e.g., not enough memory), to software (e.g., not enough virtual memory [one may need to manage the ulimits]), problems with the reading as you mention (I have not thoroughly tested nor optimized the GADGET reader, and as a matter of fact I have only used it and tested in on a couple of GADGET simulations, all of which used the unformatted binary output), or any other bug in the code.

However, since you say that nothing is written to the asohf.out file, and that the code crashes immediately, I would be tempted to say that is something related to the system (either insuficient memory, ulimits not set, etc.). You have an example of shell script to run ASOHF in the documentation, but it is assumed there that you have your stack size (ulimit -s) and your virtual memory size (ulimit -v) configured to a high enough value (as it is required by many codes!). Please check if this it what is causing your crashes.

If this does not solve your problem, please do not hesitate to contact me (preferentially by mail, you also have it in the documentation) with more information about your system, configuration, and how to reproduce your problem (especially if the data is public, I can try to reproduce the problem myself).