[Ecls-list] SI:TOP-LEVEL, stack overflows

Philipp Marek philipp at marek.priv.at
Mon Aug 1 05:50:41 UTC 2011


>>    ;;;   gcc -o bin/ecl -L/home/.../build/ /home/.../build/eclinittsZavj.o
>> -L./
>> -Wl,--rpath,/usr/local/lib/ libecl.so -lpthread -ldl -lm sed -e
>> 's, at libdir\\@,/usr/local/lib/,' \
>>            -e 's, at includedir\\@,/usr/local/include/,' \
>>            -e 's,~A,/usr/local/lib/,' bin/ecl-config.pre > bin/ecl-config
>>    $ LD_LIBRARY_PATH=/home/.../build/ /home/.../build//bin/ecl -dir
>> /home/.../build/
>>     Wrong number of arguments passed to function #<compiled-function
>> SI:TOP-LEVEL>.
>>     No restarts available.
>>     Top level in: #<process TOP-LEVEL>.
> A few things: you did not reinstall ECL after building it and since it is
> using --rpath I doubt that the change in LD_LIBRARY_PATH had any effect.
> Besides this, did you completely your build directory after updating ECL and
> finally, why do you say that ECL did not load .eclrc at this point? My
> feeling is that all this is due to conflicting versions of the library and
> partial rebuilds, but I tend to be proven wrong.
Yes, you're right - I found this during my train ride this morning, it's the wrong .so
name (http://article.gmane.org/gmane.lisp.ecl.general/8417) again.

Sorry for the noise.


> In any case this does not explain the stack overflows during ECL's use of
> quicklisp libraries. Would it be possible for you to send me an excerpt of
> your .eclrc (by private email)so that I can try to reproduce the sequence of
> libraries you loaded?
Coming in separate mail.


Thank you for your help!


Regards,

Phil







More information about the ecl-devel mailing list