2nd Stage of xGCC -- failure

[Yves Rutschle]

> thanks for your answer. I am downloading the toolchains now.  But,
> supposed I want to create my own xgcc for ARM, do you have 
> any hint, how I can circumvent the pefore mentioned problem of 

No, I never managed to compile GCC for C++ either...

> In file included from
> /home/ruf/Cross/arm-linux/lib/gcc-lib/arm-linux/2.95.3/../../.
> ./../arm-linux/sys-include/math.h:350,
>                  from 
> /home/ruf/Cross/src/gcc-2.95.3/libstdc++/cmath:7,
>                  from
> /home/ruf/Cross/src/gcc-2.95.3/libstdc++/cmathi.cc:7:
> /home/ruf/Cross/arm-linux/lib/gcc-lib/arm-linux/2.95.3/../../.
> ./../arm-linux/sys-include/bits/mathinline.h:
> In function `double atan2(double, double)':
> /home/ruf/Cross/arm-linux/lib/gcc-lib/arm-linux/2.95.3/../../.
> ./../arm-linux/sys-include/bits/mathinline.h:475:
> unknown register name `st(1)' in `asm'
> 
> which expands to 
> 
> #define __llrint_code \
>   long long int __llrintres;                  \
>   __asm__ __volatile__                    \
>     ("fistpll %0"                   \
>      : "=m" (__llrintres) : "t" (__x) : "st");              \
>   return __llrintres
> __MATH_INLINE long long int

Now that you mention it, this doesn't look like ARM assembly
at all. And, I don't have any mathinline.h file in my ARM Glibc,
too (the only arch that have it are alpha, i386, m68k,
powerPC and sparc). I suspect you have a config mismatch somewhere
and the wrong headers are used.

Y.


------
Want more information?  See the CrossGCC FAQ, http://www.objsw.com/CrossGCC/
Want to unsubscribe? Send a note to crossgcc-unsubscribe@sourceware.cygnus.com