Date: Sat, 4 Jul 1998 18:18:20 -0400 (EDT)
Message-Id: <199807042218.SAA18906@delorie.com>
From: DJ Delorie <dj AT delorie DOT com>
To: ams AT ludd DOT luth DOT se
CC: djgpp-workers AT delorie DOT com
In-reply-to: <199807042204.AAA01315@sister.ludd.luth.se> (message from Martin
	Str|mberg on Sun, 5 Jul 1998 00:04:19 +0200 (MET DST))
Subject: Re: Trying to compile v2/alphas/980628/djlsr202.zip natively
Precedence: bulk


> 1. Where is libc/compat/stdio/vfscanf.c?

It's vfscanf.S, not vfscanf.c.

> 2. I'm using gcc281b.zip, so this happens:
> ld -s  ../../../lib/crt0.o symify.o ../../../lib/libdbg.a ../../../lib/libc.a -o ../../../bin/symify.exe ../../../lib/libgcc.a -T ../../../lib/djgpp.djl
> E:/DJGPP.202/BIN/ld.exe: cannot open ../../../lib/libgcc.a: No such file or directory (ENOENT)
> 
>    as libgcc.a lives in ..../../lib/gcc-lib/djgpp/2.81. How should this
>    really be setup to be portable?

We'd need to convert it to use $(CROSS_GCC) instead of ld, so that gcc
can locate these files.

> 3. After a while the making stops because it don't know how to make
>    makefile.oh. I corrected this by changing top level makefil.inc from:

Looks good to me but I'll check it after I apply it.

> 4. Later on this:
> makemake: scanning libemu for object files
> make ./../../lib/libemu.a
> make.exe[2]: `../../lib/libemu.a' is up to date.
> ./../../hostbin/dxegen.exe ./../../bin/emu387.dxe __emu_entry src/emu387.o id_emu.o src/emudummy.o -L../../lib -L../../lib/gcc-lib/djgpp/2.81 -lgcc -lc -lgcc
> ld -X -S -r -o dxe__tmp.o -Le:/djgpp.202/lib src/emu387.o id_emu.o src/emudummy.o -L../../lib -L../../lib/gcc-lib/djgpp/2.81 -lgcc -lc -lgcc  -T dxe.ld 
> Error: input file has more than one section; use -M for map
> make.exe[1]: *** [../../bin/emu387.dxe] Error 1
> 
>    Ughh! What to do now?

Last time I saw this, it happened because dxegen wasn't handling
endianness or structure sizes in a portable way, so the COFF headers
get mangled.