Port-amiga archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: netbsd-3-0 build failure



On Thu, Feb 16, 2006 at 11:18:00AM +0100, Gunther Nikl wrote:
> On Fri, Feb 10, 2006 at 12:41:59AM +0100, Anders Lindgren wrote:
> > #   compile  GENERIC/if_x25subr.o
> > /usr/build/tools.amiga/bin/m68k--netbsdelf-gcc -pipe -m68060 -Wa,-m68030 
> > -Wa,-m68851 -msoft-float -ffreestanding
> > -pipe -Werror -Wall -Wno-main -Wno-format-zero-length -Wpointer-arith 
> > -Wmissing-prototypes -Wstrict-prototypes -Wno-sign-compare 
> > -fno-zero-initialized-in-bss -Damiga -DFPCOPROC -I. -I/usr/src/sys/arch 
> > -I/usr/src/sys -nostdinc
> > -DDRACO -DM68060 -DM68040 -DM68030 -DM68020 -DLKM -DMAXUSERS=8 -D_KERNEL 
> > -D_KERNEL_OPT -I/usr/src/sys/dist/ipf -c 
> > /usr/src/sys/netccitt/if_x25subr.c
> > /usr/src/sys/netccitt/if_x25subr.c: In function `x25_ifinput':
> > /usr/src/sys/netccitt/if_x25subr.c:218: error: inconsistent operand 
> > constraints in an `asm'
> 
> [snip]
> 
> > ..in /usr/src/sys/arch/amiga/include/mtpr.h, meaning the culprit should be 
> > in /usr/src/sys/arch/m68k/include/asm_single.h:43:
> > 
> > #define single_inst_bset_b(var, bit)    \
> >         __asm __volatile ("orb %1,%0"   \
> >                 : "=m" (var)            \
> >                 : "di" ((u_char)bit), "0" (var))
> > 
> > ..and this is where I stop, since I don't speak m68k assembly. :-)
> > 
> > Ideas? The file asm_single.h doesn't appear to have changed since at least
> > 2005-12-24 according to cvs log...
> 
> Maybe this was unused before. Or the compiler has changed. Which GCC version
> are you using? According to the compiler the asm constraints are invalid.
> Unfortunately I can't tell you what exactly the error is :-/

Uhm.... the compiler that should be used is the same as for the official
builds, which succeeded; part of the build.sh is building a cross-compiler
of exactly the same version independent of the host machine.

And no - setsoftnet() etc. have been in use for years!

[Note: Cross-posted to port-m68k and tech-toolchain, followup-to port-m68k,
 which is a more appropriate list]

Regards,
        -is



Home | Main Index | Thread Index | Old Index