This is the mail archive of the gcc@gcc.gnu.org mailing list for the GCC project.

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]
Other format:	[Raw text]

Re: Unix Signals as OS Exceptions in C++

From: Andrew Pinski <pinskia at physics dot uc dot edu>
To: Luca Saiu <positron at gnu dot org>
Cc: Ian Lance Taylor <ian at wasabisystems dot com>, "Steven T. Hatton" <hattons at globalsymmetry dot com>, gcc at gcc dot gnu dot org
Date: Sun, 12 Sep 2004 10:52:31 -0700
Subject: Re: Unix Signals as OS Exceptions in C++
References: <200409112057.47137.hattons@globalsymmetry.com> <m3y8jfo9on.fsf@gossamer.airs.com> <4144644E.2070509@gnu.org>

On Sep 12, 2004, at 7:59 AM, Luca Saiu wrote:

[not completely related]

Ian Lance Taylor wrote:
It already works fine; just call throw from your signal handler.  This
programs works as expected on i386 GNU/Linux if you type ^C during the
sleep.
On my machines the exception is regularly thrown in the signal handler, but it can't be caught by catch, not even by catch(...); the program aborts. This happens on all my machines (GNU/Linux PowerPC and x86), with all the GCC versions I have installed (from 2.95 to 3.4.2). No twiddling helped, as far as I tried.

This seems a bug. Should I file a PR? Is it only me to see that?


If you are not using -fnon-call-exceptions then GCC will not know
that a non call can throw.

Thanks,
Andrew Pinski

References:
- Unix Signals as OS Exceptions in C++
  - From: Steven T. Hatton
- Re: Unix Signals as OS Exceptions in C++
  - From: Ian Lance Taylor
- Re: Unix Signals as OS Exceptions in C++
  - From: Luca Saiu

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]