From: gvaughan AT oranda DOT demon DOT co DOT uk (Gary V. Vaughan)
Subject: cygwin.dll signal bug [WAS Re: bash and '&']
29 Sep 1998 18:09:45 -0700
Message-ID: <3610CD3A.DC42841C.cygnus.gnu-win32@oranda.demon.co.uk>
References: <19980928113300 DOT 32747 DOT qmail AT findmail DOT com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
To: tjh AT quadstone DOT com
Cc: gnu-win32 AT cygnus DOT com

tjh AT quadstone DOT com wrote:
> 
> Apologies if this is slightly off-thread:

It's not...
 
> I think Cygwin32 is great (I use b19.1).
> But one thing that bugs me is that hitting
> ^C seems to interrupt all jobs started
> from bash, not just the current job.
> 
> What am I doing wrong?

Nothing.  It just seems to work this way =(O|  Under UNIX it would be a
sign that the signal masks in the signal handlers are not set up
correctly in the background jobs, and that they are waking up when their
controlling tty recieves a signal.  Since the problem seems to affect
*all* background jobs, it must be something to do with the bash
process...  I had a very quick look at the signal handling in my bash
sources, and it is identical to the working signal handling for my UNIX
build.

I guess this means that the cygnal handling [I'll let that typo stand!]
in the emulation layer is buggy... but then we are firmly in win32 land,
and I become confused very quickly =)O|

Are there plans to fix this?  I presume it is a known bug?

Cheers,
	Gary V. Vaughan
-
For help on using this list (especially unsubscribing), send a message to
"gnu-win32-request AT cygnus DOT com" with one line of text: "help".