[Ecls-list] ECL on iPhone
Red Daly
reddaly at gmail.com
Sun Sep 14 02:04:10 UTC 2008
Initiliazed *ALIEN-DECLARATIONS* @ index 1445
Initiliazed TRAP-FPE @ index 1449
My latest problem, after building ECL without all the extensions you
mentioned, comes when ECL tries to look up the value of the symbol
*ALIEN-DECLARATIONS* using ECL_SYM("*ALIEN-DECLARATIONS*",1449).
Unfortunately, *ALIEN-DECLARATIONS* seems to be stored at 1445 in the
cl_symbols array. For some reason these indexes appear to be different
during the compilation phase and the run-time phase. any advice?
Here is the function causing the problems:
static cl_object L2create_type_name(cl_object V1)
{ VT3 VLEX3 CLSR3 STCK3
cl_object value0;
{
TTL:
if((ecl_memql(V1,ecl_symbol_value(ECL_SYM("*ALIEN-DECLARATIONS*",1449))))==Cnil){
goto L29;}
value0=cl_error(2,VV[3],V1) /*
ERROR */;
return value0;
L29:;
value0=Cnil; NVALUES=1;
return value0;
}}
-Red
On Wed, Sep 10, 2008 at 11:32 PM, Juan Jose Garcia-Ripoll <
juanjose.garciaripoll at googlemail.com> wrote:
> On Wed, Sep 10, 2008 at 9:39 AM, Red Daly <reddaly at gmail.com> wrote:
>
>> (gdb) print fname->symbol.name->base_string.self
>> $6 = (unsigned char *) 0x12e390 "CDECL"
>>
>
> That is a symbol that appears in FFI calls and definitions. Could you try
> building ECL without the compiler. It does not make sense for you to install
> the compiler in the iPhone when it is not going to have gcc around. I would
> also deactivate all default extensions from the following list, at least
> until you get a running image
>
> --with-tcp include socket interface (default=YES)
> --with-serve-event include serve-event module (default=YES)
> --with-asdf include ASDF building facility (default=YES if
> shared library support is on)
> --with-defsystem include DEFSYSTEM building facility (default=YES
> if
> shared library support is on)
> --with-cmp build in the compiler (default=YES)
> --with-rt include MIT-RT testing environment (default=YES)
> --with-profile include CMUCL's simple profiler (default=YES if
> Boehm-Weiser garbage collector used)
>
> Juanjo
>
> --
> Instituto de Física Fundamental
> CSIC, Serrano, 113, Madrid 28040 (Spain)
> http://juanjose.garciaripoll.googlepages.com
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.common-lisp.net/pipermail/ecl-devel/attachments/20080913/c28f9372/attachment.html>
More information about the ecl-devel
mailing list