In article <mack.726934159@redwood> firstname.lastname@example.org (Dan Mack) writes:
GNU's cpp failed for me as well. /usr/ccs/lib/cpp worked fine so I
was able to get as far as making the first library (Xau). Here's what
gcc -fstrength-reduce -fpcc-struct-return -c -O -I../../. AuWrite.c
rm -f libXau.a
ar clq libXau.a AuDispose.o AuFileName.o AuGetAddr.o AuGetBest.o AuLock.o AuRead.o AuUnlock.o AuWrite.o
ar: bad option `l'
usage: ar [-V] key[vcs] [posname] archive file [name] ...
where key is one of the following: r[uabi], m[abi], d, q, t, p, x
*** Error code 1
make: Fatal error: Command failed for target `libXau.a'
I guess I need to change how ar is used somewhere....
Well, mit/config/Imake.tmpl says that your ar command is determined by:
#if HasLargeTmp | SystemV4
#define ArCmd ar cq
#define ArCmd ar clq
So if you've got SystemV4 defined in sun.cf, you should be ok. Something
#define SystemV4 YES
should do it. That was a big part of my original problem (I didn't define
it as YES, I simply defined it). Luckily, I got mail from Danny Backx
<db.sunbim.be>, the author of the patches on export.lcs.mit.edu. He jogged
my memory about this, so thanks Danny!
I also heard back from Geert Jan de Groot <email@example.com> in the
form of a repost of a previous article from Dana Chee <firstname.lastname@example.org>,
both of whom I'm indebted to. Dana posted some additions to the patches
mentioned above as well as a fixed up sun.cf file. I'm trying these now.
Hope they work.
In any case, thanks to everyone. If it turns out that it still doesn't
compile, I'm still farther along than before!
-- =============================================================================== David J. Bianco Systems Dude & Part time Illusionist Old Dominion University <email@example.com> Dept of Computer Science, "BORGer King: We do it our way. Norfolk, VA 23529 Your way is irrelevant..." ===============================================================================
This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:07:22 CDT