🚀 go-pugleaf

In article <_:A6TTA@xds13.ferranti.com> peter@ficc.ferranti.com (Peter da Silva) writes:
>In article <15925@bfmny0.BFM.COM> tneff@bfmny0.BFM.COM (Tom Neff) writes:
>> We are zeroing in on something other than the original question.
>
>No, we're not. The problem at hand is doing x86 development on System V.
                                             ^^^
>It is not clear to me that the "modern" intel 86 tools (including the
>current 80286 and 80386 tool sets) are the best way.

The request was NOT this generic.  Our European friend has an 8086-based
numeric control target processor for which he *NEEDS* to generate OMF86
code.  Specifically.  I know this type of setup; I have one myself.
It's not a question of which kind of "x86" development setup is
theoretically nicer.  Intel's OMF86 based language tools exist in their
own universe.

>                                                      In fact, if I may
>be allowed a mild flame, the UNXUDI method is totally barmy, and we
>REALLY need native 80386 tools to get decent performance and functionality.

Oh, maybe.  As long as it works reliably!  I would like bigger symtabs.

>> Who
>> has seen how PLM86, for instance, behaves on the 286 platform?  I still
>> bet they don't use more than a meg.
>
>I'm sure they don't... but our copy of PLM86 from the good old days does.

If any version of PLM86 (not PLM286 or BND286, etc) uses more than a
meg, then I'd like to see a section of the listing file where it reports
more than 1024K available.  Quote it here?

-- 
Nobody wants justice.   /\ \/   Tom Neff
   -- Alan Dershowitz   /\ \/   tneff@bfmny0.BFM.COM