Tim001 wrote:I tried out the FORTRAN G and H compilers with small test programs, and they worked fine.
That's flipping impressive for software that's 40(ish) years old!
Tim001 wrote:Note that for FORTRAN H I had to specify a region on the JOB card, since otherwise the binder aborts with an insufficient memory message. I don’t know what the default region size for jobs is on FanDeZhi, but I specified REGION=2048K and that worked O.K.
From an old FORTRAN programmer who used these compilers on a real S360/65 with MVT 40+ years ago, thanks for setting them up!
I'll go through the JCL, and will try to bring it more up-to-date, have already done so for some of them (IF ... THEN ... ENDIF logic), and in my EXEC library I have EJCLFMT, an edit macro to completely clean up JCL, feel free to give it a try (copy it to SYSFAN.USER.REXXLIB, you have RACF UPDATE access to this PDS) on personal copies of the procedures, standard invocation is just EJCLFMT while in edit, but invoke it as "EJCLFMT NOPROC" if the results are not good, it currently has some no-FanDeZhi hard-coded things in it that I need to change.
Off to Germany for a few days, back on Sunday, and probably little or no internet access during these days.