heechee@anyguay.acm.rpi.edu (Stephen Underwood) (07/31/89)
I've been trying since the source came out to compile 3.0 on an att 3b2 runing sysv rel 3.1. I've had to define STUPID, STUPID_CCP, and -DDUMB, and now I have all the source compiled and go to link I get ld fatal: fail to write symbol name resists_elec in string table for file nethack. I'm about readdy to though this machine out a window. (I didn't have near this much trouble on the Xenix 386 at work, or the sun 3/50) -Stephen Underwood -heechee@pawl.rpi.edu -usere3v4@mts.rpi.edu
paone@aramis.rutgers.edu (Phil Paone) (08/01/89)
Very simple solution, up your ulimit before the link and finish it as root, or up your systme's ulimit and do it yourself. -- Phil Paone attmail!ppaone !rutgers.edu!topaz.edu!ppaone paone@topaz.rutgers.edu "Dinna ya know a jailbreak when ya see it?"
mpl@cbnewsl.ATT.COM (michael.p.lindner) (08/02/89)
In article <6287@rpi.edu>, heechee@anyguay.acm.rpi.edu (Stephen Underwood) writes: > ld fatal: fail to write symbol name resists_elec in string table for file > nethack. > > -Stephen Underwood -heechee@pawl.rpi.edu -usere3v4@mts.rpi.edu Save your window. I compiled it without trouble on my 3B2 running 3.1. All I defined was STUPID_CPP, to get around a space limitation in the preprocessor. I understand that that limitation's been fixed in the next release of C. Anyway, the only time I've seen an error like that, it was because the compiler either ran out of temp file space, so the write failed (see the man page on how to change where the temp files are made), or because the ulimit was exceeded while writing the ourput file or a temp file (see the man page on ulimits). On my machine my ulimit is 2048 (1 megabyte). Lotsa Luck, Mike Lindner attunix!mpl AT&T Bell Laboratories 190 River Rd. Summit NJ 07901
rjd@occrsh.ATT.COM (Randy_Davis) (08/02/89)
In article <6287@rpi.edu> heechee@pawl.rpi.edu (Stephen Underwood) writes: |I've been trying since the source came out to compile 3.0 on an att |3b2 runing sysv rel 3.1. I've had to define STUPID, STUPID_CCP, and |-DDUMB, and now I have all the source compiled and go to link I |get |[...] |I'm about readdy to though this machine out a window. (I didn't have |near this much trouble on the Xenix 386 at work, or the sun 3/50) Though I am not real happy with the problems I have had compiling it on a 3B2 (System V release 3.2.1), I fail to see how you can totally blame the machine or the compiler - Remember it was written on a different machine with its own quirks that the authors were able to fix for you. As is noted in the documentation, the program has not even been tested by the authors on a 3B2. Yeah, but I wouldn't mind having a little discussion with the compiler writers for the 3B2 to find out exactly who is wrong... Randy Davis UUCP: ...(att!)ocrjd!randy, randy@ocrjd.att.com ...(att!)occrsh!rjd, rjd@occrsh.att.com