Experimental Physics and
Industrial Control System

Hello,

 

When running my st.cmd, I get a bad interpreter error for my ioc’s executable:

../../bin/linux-x86_64/<PROD_IOC>^M: bad interpreter: No such file or directory

 

This was a working IOC I’m trying to port to a new host, at one point I even got it to work on the new host ($(TOP): make distclean, make, cd iocboot/ioc<name>, ./st.cmd) using CRLF settings on git, but ended up reverting back and can’t re-create this solution.

 

Prevention is probably the best method to deal with this, by ensuring all editors are using LF/UNIX settings. But now that I’m in this state, I’m unsure what the best course of action is.. I can’t seem to tell what file is corrupting my binary output. How does the EPICS community handle fixing CRLF/LF eols, or even know where to find the conflict? Is it even expected that my freshly output binary executable would be corrupted due to (…any?) ‘CRLF files’ having existed under TOP at time of make?

 

For reference, what I thought made it work before was using LF only for files matching: *.c, *.h, *.db, and *.substitutions.

                                                                                                                                                                                       

Any and all help, or direction is very much appreciated.

Thank you!

-Dan