paulgboul...@aim.com (Paul Gilmartin) writes: > Where did you put NUCON? IIRC, there's an EXTRN for NUCON, which > implies the possibility of relocating it and using an ADCON as a > base. But IIRC also there's so much hardcoded "USING NUCON,R0" > that relocation would be futile. Aren't the PSWs also mapped in > NUCON? Or does the CMS nucleus much care about that?
there is the approach that has simulation of CMS system functions ... as opposed to running the CMS kernel ... these would be akin to CMS simulation of os/360 system functions for running os/360 applications. There was joke in the late 70s that the 64K bytes of os/360 simulation code in cms was much more cost/effective than the MVS 8mbyte os/360 simulation ... and as been alluded to in previous posts, there were very large, major internal MVS applications that started pushing system boundaries and had to be ported to CMS (as way of getting around 7mbyte application restriction and other limits). In a couple cases, there was 12kbytes of additional simulation written to include support for os/360 features not already supported by CMS. note that the simulation of vm370 (diag) functions in MVS was method used for getting vm370 tcp/ip support runing in MVS environment ... for original MVS tcp/ip product. I know there have been lots of comments about the efficiency of that product ... but the base wasn't much better in vm370. However, early on, I added the rfc1044 support to the base product and in some tuning tests at cray research, got approx 500 times improvement in the number of instructions executed per byte moved. misc. past posts mentioning rfc1044 support http://www.garlic.com/~lynn/subnetwork.html#rfc1044 -- virtualization experience starting Jan1968, online at home since Mar1970 ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
