On Tue, Mar 26, 2013 at 04:09:33PM +0100, Pascal Stumpf wrote:
> On Tue, 26 Mar 2013 15:47:45 +0100, Pascal Stumpf wrote:
> > So here's the final version of GCC 4.8.0. Tested by me on amd64, i386,
> > powerpc and by jsg some time ago on hppa. sparc64 hasn't been tested
> > yet, so it's disabled for now. Needless to say, it would be cool if
> > someone could try that out, as well as hppa again and maybe alpha. I
dies in stage2 on hppa:
/usr/ports/pobj/gcc-4.8.0/build-hppa/./prev-gcc/xg++
-B/usr/ports/pobj/gcc-4.8.0/build-hppa/./prev-gcc/
-B/usr/local/hppa-unknown-openbsd5.3/bin/ -nostdinc++
-B/usr/ports/pobj/gcc-4.8.0/build-hppa/prev-hppa-unknown-openbsd5.3/libstdc++-v3/src/.libs
-B/usr/ports/pobj/gcc-4.8.0/build-hppa/prev-hppa-unknown-openbsd5.3/libstdc++-v3/libsupc++/.libs
-I/usr/ports/pobj/gcc-4.8.0/build-hppa/prev-hppa-unknown-openbsd5.3/libstdc++-v3/include/hppa-unknown-openbsd5.3
-I/usr/ports/pobj/gcc-4.8.0/build-hppa/prev-hppa-unknown-openbsd5.3/libstdc++-v3/include
-I/usr/ports/pobj/gcc-4.8.0/gcc-4.8.0/libstdc++-v3/libsupc++
-L/usr/ports/pobj/gcc-4.8.0/build-hppa/prev-hppa-unknown-openbsd5.3/libstdc++-v3/src/.libs
-L/usr/ports/pobj/gcc-4.8.0/build-hppa/prev-hppa-unknown-openbsd5.3/libstdc++-v3/libsupc++/.libs
-g -O2 -gtoggle -DIN_GCC -fno-exceptions -fno-rtti
-fasynchronous-unwind-tables -W -Wall -Wno-narrowing -Wwrite-strings
-Wcast-qual -Wmissing-format-attribute -pedantic -Wno-long-long -Wno!
-variadic-macros -Wno-overlength-strings -DHAVE_CONFIG_H -DGENERATOR_FILE
-static-libstdc++ -static-libgcc -o build/gengenrtl \
build/gengenrtl.o build/errors.o .././libiberty/libiberty.a
build/gengenrtl > tmp-genrtl.h
Segmentation fault (core dumped)
gmake[3]: *** [s-genrtl-h] Error 139
gmake[3]: Leaving directory `/usr/ports/pobj/gcc-4.8.0/build-hppa/gcc'
gdb ./gengenrtl
GNU gdb 6.3
Copyright 2004 Free Software Foundation, Inc.
GDB is free software, covered by the GNU General Public License, and you are
welcome to change it and/or distribute copies of it under certain conditions.
Type "show copying" to see the conditions.
There is absolutely no warranty for GDB. Type "show warranty" for details.
This GDB was configured as "hppa-unknown-openbsd5.3"...
(gdb) run
Starting program: /usr/ports/pobj/gcc-4.8.0/build-hppa/gcc/build/gengenrtl
/* Generated automatically by gengenrtl from rtl.def. */
#ifndef GCC_GENRTL_H
#define GCC_GENRTL_H
#include "statistics.h"
static inline rtx
gen_rtx_fmt_0_stat (RTX_CODE code, enum machine_mode mode MEM_STAT_DECL)
{
rtx rt;
rt = rtx_alloc_stat (code PASS_MEM_STAT);
PUT_MODE (rt, mode);
X0EXP (rt, 0) = NULL_RTX;
return rt;
}
#define gen_rtx_fmt_0(c, m)\
gen_rtx_fmt_0_stat (c, m MEM_STAT_INFO)
static inline rtx
gen_rtx_fmt_ee_stat (RTX_CODE code, enum machine_mode mode,
_stat (RTX_CODE code, enum machine_mode modearg0,
arg1 MEM_STAT_DECL)
{
rtx rt;
rt = rtx_alloc_stat (code PASS_MEM_STAT);
PUT_MODE (rt, mode);
Program received signal SIGSEGV, Segmentation fault.
0x00001e50 in main ()
(gdb) bt
#0 0x00001e50 in main ()
> > haven't merged the mips64(el) backend patches yet, but I'll do that
> > together with bcallah@ as soon as binutils 2.17 is ready on those archs.
> >
> > -java works, and I've merged kurt@'s patches from 4.6. The resulting
> > gcj can compile classpath and pdftk. Also added support for -java on
> > powerpc.
> >
> > Disabled -faggressive-loop-optimizations. It breaks SPEC.
> >
> > The compiler is now written in C++. This means that adastrap must now
> > also contain the c++ part of GCC.
> >
> > PIC/no-PIC bullshit is no more (yay).
> >
> > I hope plists are correct. Updating this port without a working
> > update-plist is a real PITA.
>
> Aand of course I just found a small problem: move ${CONFIG]-e[gc]++ to
> PLIST-c++.
