On 11 Oct 2009 05:34:17 -0700, in bit.listserv.ibm-main you wrote: >2009-10-09 Farley, Peter x23353 <[email protected]>: >> I have yet to see >> COBOL code generation that takes account of potential superscalar >> pipeline interruptions. Why is that? > >A perhaps more interesting related question is why IBM's languages, >including COBOL, seem to continue to have their own code generation, >rather than using a GCC-like scheme of language-specific front end, >followed by common code generation and optimization back (and middle, >to use GCC terminology) ends. They've already got them all running >under LE at run time.
The code base and generation philosophy for COBOL dates back to around 1989 or 1990 (COBOL II). This a sophisticated code generator and would be expensive to replace. If the design can't be easily modified to handle either the 2002 standard or 64 addressing (to the extent that C/C++ handles it), then it might be worth considering. The other reason would be if it would allow compile time optimization of mixed language code. I would not want to give up optimizations that take advantage of the knowledge of the COBOL code. > >Is this just typical IBM NIH syndrome, or something else? > >Tony H. > ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
