[C--] building on mac os x: a bit further

Zorg 421 <[email protected]> Tue, 24 Mar 2009 21:54:37 +0100
Newsgroups gmane.comp.lang.c--
Message-ID <[email protected]>
It turned out that the ./configure was setting the platform to x86-unknown...
now I've added a x86-darwin section anywhere it was needed.

but (:-)

...
river.cmo postexpander.cmo expander.cmo interpemit.cmo interpasm.cmo
peephole.cmo placevar.cmo ppc.cmo stack.cmo widen.cmo x86asm.cmo
x86call.cmo x86.cmo dummy.cmo lualink.cmo main.cmo zipncfg.cmo
cp qc-- ../bin/qc--
+ for i in '"$@"'
+ cd runtime
+ mk depend
mk: 'depend' is up to date
+ mk update
tmp=/tmp/cut$$
../bin/qc--  Options.swap=nil -S -o $tmp cut.c--
sed '/globalsig/d;/ref_to_global_area/d' $tmp > cut.s
rm -f $tmp
tmp=/tmp/thread$$
../bin/qc--  Options.swap=nil -S -o $tmp thread.c--
sed '/globalsig/d;/ref_to_global_area/d' $tmp > thread.s
rm -f $tmp
tmp=/tmp/yield$$
../bin/qc--  Options.swap=nil -S -o $tmp yield.c--
sed '/globalsig/d;/ref_to_global_area/d' $tmp > yield.s
rm -f $tmp
as -o cut.o cut.s
cut.s:2:Expected comma after segment-name
cut.s:3:Expected comma after segment-name
cut.s:3:Rest of line ignored. 1st junk character valued 67 (C).
cut.s:18:Expected comma after segment-name
mk: as -o cut.o cut.s  : exit status=exit(1), deleting 'cut.o'cut.o:
No such file or directory

+ exit 1
mk: ./mkdirs src runtime  : exit status=exit(1)


$ cat runtime/cut.s
.globl Cmm_CutTo
.section .data
/* reference to global-register signature */
.section .text
Cmm_CutTo:
	movl %esp,%eax
	movl $4,%ecx
	addl %ecx,%eax
	movl (%eax),%eax
	movl %esp,%ecx
	movl (%ecx),%ecx
.Linitialize_continuations_l3:
.Lproc_body_start_l2:
	movl $4,%ecx
	movl %eax,%edx
	addl %ecx,%edx
	movl (%edx),%ecx
	movl (%eax),%eax
	movl %ecx, %esp; jmp *%eax
.section .text

There is no need of those .section before .text and .data (reading the
apple doco), where may I tweak the templating system to not include
them?
Regards.
_______________________________________________
Cminusminus mailing list
[email protected]
https://www.eecs.harvard.edu/mailman/listinfo/cminusminus