Mailing-List: contact cygwin-help AT sourceware DOT cygnus DOT com; run by ezmlm
List-Subscribe: <mailto:cygwin-subscribe AT sourceware DOT cygnus DOT com>
List-Archive: <http://sourceware.cygnus.com/ml/cygwin/>
List-Post: <mailto:cygwin AT sourceware DOT cygnus DOT com>
List-Help: <mailto:cygwin-help AT sourceware DOT cygnus DOT com>, <http://sourceware.cygnus.com/ml/#faqs>
Sender: cygwin-owner AT sourceware DOT cygnus DOT com
Delivered-To: mailing list cygwin AT sourceware DOT cygnus DOT com
X-Authentication-Warning: hp2.xraylith.wisc.edu: khan owned process doing -bs
Date: Fri, 4 Feb 2000 10:05:58 -0600 (CST)
From: Mumit Khan <khan AT NanoTech DOT Wisc DOT EDU>
To: M Dipperstein <mdipper AT cs DOT ucsb DOT edu>
cc: cygwin AT sourceware DOT cygnus DOT com
Subject: Re: Threads
In-Reply-To: <Pine.GSO.4.05.10002040619080.4230-100000@dinah>
Message-ID: <Pine.HPP.3.96.1000204100359.19647f-100000@hp2.xraylith.wisc.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII

On Fri, 4 Feb 2000, M Dipperstein wrote:

> There have been several postings to the list which discuss pthreads and
> the lack of "thread safety" in cygwin.  I was hoping somebody would
> clarify the statement that cygwin is not thread safe.

Cygwin is thread-safe by default as of 2000-01-05 or so snapshot (or
perhaps even a bit earlier). Last year, you had to build it yourself
to get the thread-safe code. It's still being worked on, so there are
bound to be problems, but seems to work ok.

> I'm developing a thread package from the QuickThreads primitives.  The
> final target is not cygwin, but it's a great tool for development along
> the way.  So far, there seem to be no problems using the QuickThreads
> primitives in code compiled with gcc under cygwin.

As long as it's not C++ using exception handling you should be fine. Until
we enhance the Cygwin's pthread interface, the C++ runtime is still single
threaded.

> Are the only concerns regarding thread safety related to non-reentrant code
> in the cygwin libraries?  If that's the only problem, I think I don't have
> to worry.  The thread package I'll be working with is cooperative, and
> will not allow for library routines to be reentered before they have
> returned.

Try it out and see what happens. We can try and fix the problems, but need
to know what those are first.

Regards,
Mumit



--
Want to unsubscribe from this list?
Send a message to cygwin-unsubscribe AT sourceware DOT cygnus DOT com