Can CentOS be set to pure EBCDIC?

I need to directly manipulate files from a mainframe. Text would be encoded in EBCDIC. Data can contain binary and packed decimal fields. I know I can deal with the data. I am trying to avoid the conversion(s). (and the trouble with converting each field based on it being text or binary). I am familiar with mainframe, USS and CentOS systems. What I would like to do is flip CentOS to be pure EBCDIC. A search on the internet and StackExchange has revealed very little. I saw an article where Sammyrulez had "... succesfuly added EBCDIC support to a CentOS box by adding lang and encoding to /etc/i18n/sysconfig." It seems like there should be more to it than this. I plan to use COBOL via GCC or whatever product works best. (one hurdle at a time) Has anyone been through this already?