wfp@dasys1.UUCP (William Phillips) (07/30/88)
I understand the sources to mg2a for MS-DOS are available. Have they been
posted to the Net (haven't seen them here!)? Where are they available (note:
ftp is not available here).
Thanks.
--
William Phillips {allegra,philabs,cmcl2}!phri\
Big Electric Cat Public Unix {bellcore,cmcl2}!cucard!dasys1!wfp
New York, NY, USA !!! JUST SAY "NO" TO OS/2 !!!wfp@dasys1.UUCP (William Phillips) (08/17/88)
After weeks of anticipation, I finally got the MS-DOS source for mg2a, and
gleefully compiled it.
Unfortunately, it doesn't work.
I compiled it first with Turbo C 1.5, which produced a few strange warnings
(mostly "possible use of <whatever> before definition" or words to that effect
-- and invariably incorrect), but produced an executable -- which bombed (had to
hit the big red switch to reboot).
Next, I tried Quick C 1.0, since I don't have MSC, which compiled without any
problems (after I put #ifndef MSC / #endif around the signal() function); it
also bombed, in exactly the same manner. This led me to believe there was a
bug in the source :-). I added a /Zr to check pointer usage at runtime, and
sure enough it hit that right away. I put in some printfs, and now have
narrowed the problem down to a problem with a line in the function modeline()
in display.c, namely:
n += vtputs(bp->b_modes[mode]->p_name);
...so far I haven't been able to pin down the problem precisely, but it's
there somewhere.
(The runtime error message was R6013 -- "Invalid far pointer reference"
according to the manual, though the message on the screen was slightly
different wording -- same meaning, however.)
Any suggestions?
--bill
--
William Phillips {allegra,philabs,cmcl2}!phri\
Big Electric Cat Public Unix {bellcore,cmcl2}!cucard!dasys1!wfp
New York, NY, USA !!! JUST SAY "NO" TO OS/2 !!!