signals and fhandlers

Robert Collins
Fri Mar 30 16:28:00 GMT 2001

----- Original Message -----
From: "Christopher Faylor" <>
To: "egor duda" <>
Sent: Saturday, March 31, 2001 1:44 AM
Subject: Re: signals and fhandlers

> On Fri, Mar 30, 2001 at 04:48:23PM +1000, Robert Collins wrote:
> >Chris's comment on inherited handles was that non-cygwin applications
> >don't know what to do _anyway_ with inherited handles. (Chris: maybe
> >got this wrong? ). Anyway the worst case is that the writers will not
> >know that the readers have all died and won't return EPIPE. This can
> >happen with things are killed via task manager anyway. (That's why
> >there's a warning!).
> That's the worst case that Egor was mentioning.  If we can develop
> a scenario that allows the right thing to happen when a program takes
> a non-standard exit then we should pursue this at all costs.  I have
> always tried to do this in all of my cygwin development.  I have, so
> far, avoided relying on cygwin cleanups for proper operations.

Referencing all
process handles are closed on a TerminateProcess() call. I'll see if I
can build detection of readers around some handle rather than a counter
as I have today. That'll give an accurate count even if the process is

> >Also, the pipes will have to be in the cygwin shared memory area, and
> >Chris indicated he didn't want any more data in there (I was going to
> >try anon pipes, when I posted my question and got that answer)..
> I just said that you shouldn't use the existing shared memory areas.
> can create your own small shared memory areas.

That's what I took it to mean. In fact I am using named objects for my


More information about the Cygwin-developers mailing list