[Ecls-list] Error trying to build on ARM9

[Jeff Cunningham]

Juan Jose Garcia-Ripoll wrote:
> This is definitely an evidence that GMP failed to configure for your
> system. There is a workaround, which is to tell the library to use C
> instead of assembler. For instance, in my linux box I would type
>
> gmp_build="none-apple-darwin" ../src/configure  --prefix=$HOME
>
> By using "none" as a machine architecture, this is what it outputs at
> configuration time
> .....
> using ABI="long"
>       CC="gcc -fPIC"
>       CFLAGS="-O3"
>       CPPFLAGS="-DNO_ASM"
>       MPN_PATH="  generic"
>
> Given that there are no other issues, ECL should build as in any other
> platform, though computations with big integers will be a bit slower.
>
> Juanjo
>
>   

I tried your suggestion. It built for quite a few hours and was still 
going strong when I left it last night.  But this morning I found it in 
this state:

...
;;; Note: Replacing variable G7043 by its value
;;; Note: Replacing variable G7044 by its value
;;; Emitting code for EXTRACT-USER-FUNCTION-NAME.
;;; Calling the C compiler...
;;; Note: Invoking external command:
;;; gcc -I"/home/jcunningham/installed/ecl-0.9j/build/c"  -D_GNU_SOURCE 
-g -O2 -fPIC  -fstrict-aliasing -Dlinux 
-I"/home/jcunningham/installed/ecl-0.9j/src/c" -O 
"-I/home/jcunningham/installed/ecl-0.9j/build/" -w -c 
"/home/jcunningham/installed/ecl-0.9j/build/lsp/format.c" -o 
"/home/jcunningham/installed/ecl-0.9j/build/lsp/format.o"

LAMBDA: Too many arguments to function CONTINUE.
Broken at TOP-LEVEL.No restarts available.
Top level.
SI>

--Jeff