gcc and g++ for IRIX 3.3.1 again
Knobi der Rechnerschrat
XBR2D96D at DDATHD21.BITNET
Fri Nov 2 05:32:23 AEST 1990
Hello,
I've finally solved my stdarg problem posted yesterday to the list.
The changes in stdarg.h are appended to this mail.
Now I've tried to generate g++-1.37.1. I've got the 1.37.0 distribution
from prep.ai.mit.edu and applied the 1.37.1 changes to the files (I did
it by hand. Is ther a way to do them automagically?). Then I folowwed the
instructions in the README file, changed the Makefile and configed the
files for 'iris'. The compilation process completed successfully (I only
had to apply "my" patches to tree.c and gcc.c again). After that I tried
to compile libg++ and that also seemed to be ok. Now I have problems:
when trying to build the tests in libg++, ld fails and complains about
unresolved externals __DTOR__ and __CTOR__. As I don't have gnu-as and
gnu-ld for the mips architecture (btw. are there "official" mips ports
of gnu-as and gnu-ld ?) I've defined FASCIST_ASSEMBLER for the g++ build.
The problem still persists. What am I doing wrong?
Regards
Martin Knoblauch
TH-Darmstadt
Physical Chemistry 1
Petersenstrasse 20
D-6100 Darmstadt, FRG
BITNET: <XBR2D96D at DDATHD21>
-------------------- diffs for stdarg.h -------------------------------
*** stdarg.oh Wed Oct 31 17:23:48 1990
--- stdarg.h Thu Nov 1 19:17:01 1990
***************
*** 2,5 ****
--- 2,10 ----
#define _STDARG_H
+ /* This is for IRIX 3.3.1 , M. Knoblauch */
+ #ifdef sgi
+ #define _VA_LIST_
+ #endif
+
typedef char *va_list;
***************
*** 10,14 ****
(((sizeof (TYPE) + sizeof (int) - 1) / sizeof (int)) * sizeof (int))
! #ifndef __sparc__
#define va_start(AP, LASTARG) \
(AP = ((char *) &(LASTARG) + __va_rounded_size (LASTARG)))
--- 15,19 ----
(((sizeof (TYPE) + sizeof (int) - 1) / sizeof (int)) * sizeof (int))
! #if !defined(__sparc__) && !defined(__mips__)
#define va_start(AP, LASTARG) \
(AP = ((char *) &(LASTARG) + __va_rounded_size (LASTARG)))
---------------------------end of diffs----------------------------------------
More information about the Comp.sys.sgi
mailing list