[Ecls-list] Frequent stackoverflow errors

Juan Jose Garcia-Ripoll juanjose.garciaripoll at googlemail.com
Fri Feb 25 15:11:07 UTC 2011


On Fri, Feb 25, 2011 at 11:44 AM, Marko Kocić <marko.kocic at gmail.com> wrote:

> Is there a way I can help to debug this stack overflow problem to find
> a root cause?
>

I am a bit puzzled. Is this very recent? Unfortunately the compiler farm was
off-line for most of the period I have been abroad and I could not verify
the progress of the updates I have been doing.

Regarding how to debug this, ECL has to be ran from inside GDB with a
breakpoint at the function that signals these errors, ecl_cs_overflow. When
the stack is overflown, ECL invokes a function (ecl_cs_overflow) which
simply jumps to some inner CATCH_ALL point (typically one set up in the
top-level). No debugger is started because it does not make sense -- it
could not run with the exhausted stack and we can not start a new stack
without confusing the garbage collector.

Juanjo

-- 
Instituto de Física Fundamental, CSIC
c/ Serrano, 113b, Madrid 28006 (Spain)
http://juanjose.garciaripoll.googlepages.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.common-lisp.net/pipermail/ecl-devel/attachments/20110225/e312cb2d/attachment.html>


More information about the ecl-devel mailing list