lmb@sat.uucp (Larry Blair) (02/02/91)
I have spent far too much time trying to get gcc up on my HP-UX 7.0 system. The first problem, which has nothing to do with gcc, is that I can't build ranlib. I'm getting an undefined for _psignal. Where is this supposed to come from? My current stumbling block is that gnu-ld doesn't like /lib/crt0.o. I had been told that HP-UX used the BSD a.out format, but this would seem to indicate otherwise. I tried to apply the robotussin program, on the chance that it is in COFF, but I can't get robotussin.c to compile. What next? And what else do I need to look out for? -- Larry Blair apple!sat!lmb lmb%sat.uucp@apple.com
cedman@golem.ps.uci.edu (Carl Edman) (02/02/91)
In article <1991Feb01.193535.15868@sat.uucp> lmb@sat.uucp (Larry Blair) writes:
I have spent far too much time trying to get gcc up on my HP-UX 7.0 system.
The first problem, which has nothing to do with gcc, is that I can't build
ranlib. I'm getting an undefined for _psignal. Where is this supposed to
come from?
Sorry. I have compiled gcc quite a few times on this system, and after
properly setting up the makefile all the rest is just a matter of
typeing 'make' and waiting an hour or two. Maybe you should check your
makefile.
My current stumbling block is that gnu-ld doesn't like /lib/crt0.o. I had
been told that HP-UX used the BSD a.out format, but this would seem to
indicate otherwise. I tried to apply the robotussin program, on the chance
that it is in COFF, but I can't get robotussin.c to compile.
This problem is very common. OK, here is the explanation again. You
probably configured your gcc to output gnu format object/library
files. Then you have to use the gnu linker, stripper, assembler, nm,
a.s.o. To be able to use old libary files you have to convert them and
store them somewhere else (e.g. /usr/local/lib/gnu). To do this you
use either 'hpxt' on each file or the 'mkhplib' script to convert all
library files.
What next? And what else do I need to look out for?
I think that should fix most of your problems. If not, mail me.
Carl Edman
Theoretical Physicist,N.: A physicist whose | Send mail
existence is postulated, to make the numbers | to
balance but who is never actually observed | cedman@golem.ps.uci.edu
in the laboratory. | edmanc@uciph0.ps.uci.edu
* Did you know that you can have a typo in your signature, send 100s of *
* mails and 100s of world-wide posts, and nobody will ever tell you ? *