Posted by
Jens Thiel-2 on
URL: http://quantlib.414.s1.nabble.com/exception-handling-on-Unix-tp2545p2554.html
Hi Xavier,
optimizations sometimes result in buggy code (that is why they are disabled
by default). Using higher -O levels, the compiler gets more and more
agressive with it assumptions on what can be "optimized". Do you have the
same errors without optimization? Have you tried to run the test-suite with
different optimization levels?
Jens.
-----Original Message-----
From:
[hidden email]
[mailto:
[hidden email]] On Behalf Of
[hidden email]
Sent: Wednesday, May 21, 2003 10:49 AM
To: Luigi Ballabio
Cc:
[hidden email]
Subject: Re: [Quantlib-users] exception handling on Unix
Hi Luigi,
I'm using the EuropeanOption.cpp example to deduce the implied volatility of
options. There's the try-catch in the source code as you explained me. Now
everything goes fine except when it is impossible to find a solution. the
Brent algorithm should give me an exception like:"Exception happend for
matrix point: root not bracketed: f[0.0001000000,4.0000000000] ->
[0.63761133978130147000,6.27375654274655490000]" (I see that on NT) but on
Unix I only see a core dump.
I'm wondering if it's not a bug with gcc with the option -O2 or -O3, I found
on internet that people reported the same problem with gcc (until
3.1) on Unix or Linux using this option to compile.
Xavier
Luigi Ballabio
<luigi.ballabio@fast To:
[hidden email],
[hidden email]
webnet.it> cc:
Subject: Re:
[Quantlib-users] exception handling on Unix
21/05/2003 10:34
At 09:45 AM 5/21/03 +0200,
[hidden email] wrote:
>Excuse me if this question is too basic but I'm trying to understand
>how
to
>handle Quantlib exceptions on Unix.
>I haven't changed any option to compile quantlib with gcc and when an
>exception is returned I have "core dump" instead of a beautiful error
>message.
Xavier,
I'm not sure I understand. What is it exactly that you're doing?
Are you linking the library to some application? And does the application
properly handle exceptions? I mean, if you just write:
int main() {
throw Error("Boo!");
return 0;
}
the program will just abort---you'll have to write:
int main() {
try {
throw Error("Boo!");
return 0;
} catch (std::exception& e) {
std::cerr << e.what() << std::endl;
return 1;
} catch (...) {
std::cerr << "unknown error" << std::endl;
return 2;
}
}
after which the program will tell you what was the exception. Also, I'm not
sure that even the above will work if some special kind of errors are raised
(such as division by zero). In that case, you'll have to find out when that
happens and add a check before the division...
Was this what you were asking, or am I off the mark?
Later,
Luigi
*************************************************************************
Ce message et toutes les pieces jointes (ci-apres le "message") sont
confidentiels et etablis a l'intention exclusive de ses destinataires. Toute
utilisation ou diffusion non autorisee est interdite. Tout message
electronique est susceptible d'alteration. La Fimat et ses filiales
declinent toute responsabilite au titre de ce message s'il a ete altere,
deforme ou falsifie.
********
This message and any attachments (the "message") are confidential and
intended solely for the addressees. Any unauthorised use or dissemination is
prohibited. E-mails are susceptible to alteration. Neither Fimat nor any of
its subsidiaries or affiliates shall be liable for the message if altered,
changed or falsified.
*************************************************************************
-------------------------------------------------------
This SF.net email is sponsored by: ObjectStore.
If flattening out C++ or Java code to make your application fit in a
relational database is painful, don't do it! Check out ObjectStore. Now part
of Progress Software.
http://www.objectstore.net/sourceforge_______________________________________________
Quantlib-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/quantlib-users