Re: [Chicken-users] Interrupt handling overhaul

[Alan Post]

On Tue, Nov 29, 2011 at 09:22:32AM +0100, Christian Kellermann wrote:
> Dear chicken fans,
> 
> I have just pushed felix's proposed change to chicken's interrupt
> handling to the master branch. This change might affect ALL chicken
> code that uses units posix, tcp and as well as interpreted code
> (since csi installs now its own SIGINT handler).
> 
> I have been running this code for three weeks now without problems
> for my projects. Also I have not found a way to break the signal
> code. As I am merely human, I therefore apply to you, my adventurous
> spirits, to unleash your hacking and testing skills and try to break
> the signal handling code in master.
> 
> My thanks go to Felix, John, Alan and Jerry for the insightful
> discussions (and of course the patch). For reference the thread can
> be found in the chicken-hackers archive, where usually these kind
> of technical core hacking issues and patches are discussed:
> https://lists.gnu.org/archive/html/chicken-hackers/2011-10/msg00127.html
> 
> Thanks for flying CHICKEN wings.
> 
> Kind regards,
> 
> Christian
> 

I've been using this patch for some time while running an
application that handles I/O and child processes.  It has
been working well for me--though my program forks based on
keyboard input: it isn't pushing the limits of this code.

If you are running an application that:

 1) handles a variety of signals
 2) handles one or more signals at high frequency

We could really use your help testing.  This might be a
relatively rare case--making it all the more important to
find any uses in the wild and ensure they work.

Thank you all!

-Alan
-- 
.i ma'a lo bradi cu penmi gi'e du