karl@ddsw1.MCS.COM (Karl Denninger) (08/04/89)
A few days ago I posted a complaint about nethack 3.0 dumping core on me while playing.... After lots of "compile/wait/sdb/wait/etc" (this sucker is HUGE - 1.7 MEG with the -g switch -- yikes!) I finally found the problem, and a fix. Xenix, if you use the "termlib" library, distinctly does not like the arguments to the "tputs()" call that are given. In fact, I couldn't find a set of argument type(s) that would not produce a core dump. Thus, if you are compiling for SCO Xenix V/386 2.3, set "Sysunix" at the top level, and in addition make sure you specify the terminal library as "curses", not "termlib". With these two changes (which are hinted at but not blatently stated) the game compiles and appears to run. I haven't managed to get past the third level yet though, so I have no idea how robust it is. It no longer dumps core the first time you do anything which launches a projectile. I like those (new) thrown knives and crossbow bolts by monsters -- they are NASTY! Several times I have been "hacking" away only to see "you are hit by a large black knife" -- which came from the doorway I am in front of! 3.0 looks like a MAJOR improvement to a killer game... thanks. -- Karl Denninger (karl@ddsw1.MCS.COM, <well-connected>!ddsw1!karl) Public Access Data Line: [+1 312 566-8911], Voice: [+1 312 566-8910] Macro Computer Solutions, Inc. "Quality Solutions at a Fair Price"
markt@polari.UUCP (Mark Tapper) (08/05/89)
In article <1989Aug4.143103.4518@ddsw1.MCS.COM>, karl@ddsw1.MCS.COM (Karl Denninger) writes: > > A few days ago I posted a complaint about nethack 3.0 dumping core on me > while playing.... > > After lots of "compile/wait/sdb/wait/etc" (this sucker is HUGE - 1.7 MEG > with the -g switch -- yikes!) I finally found the problem, and a fix. > > Xenix, if you use the "termlib" library, distinctly does not like the > arguments to the "tputs()" call that are given. In fact, I couldn't find a > set of argument type(s) that would not produce a core dump. I have come across this same problem in nethack 2.3 so I knew what to fix this time around, I just put a nap(50) in place of tputs() and it works fine. but I have found another bug that you will probably notice also. when yn_function calls sprintf it sometimes sends it something that causes sprint to coredump. I dont want to use sdb on it, it would take an hour to compile, so I think I will stick to adb and I will post here if I figure out what is causing yn_function() to core dump on a return. -- Mark Tapper, System Admin for PolarServ. markt@polari.uucp PolarServ 206/328-4945 (Data) Public access Xenix for Seattle/Tacoma