Mailing-List: contact cygwin-help AT cygwin DOT com; run by ezmlm
List-Subscribe: <mailto:cygwin-subscribe AT cygwin DOT com>
List-Archive: <http://sources.redhat.com/ml/cygwin/>
List-Post: <mailto:cygwin AT cygwin DOT com>
List-Help: <mailto:cygwin-help AT cygwin DOT com>, <http://sources.redhat.com/ml/#faqs>
Sender: cygwin-owner AT cygwin DOT com
Mail-Followup-To: cygwin AT cygwin DOT com
Delivered-To: mailing list cygwin AT cygwin DOT com
X-Info: This message was accepted for relay by
	smtp03.mrf.mail.rcn.net as the sender used SMTP authentication
X-Trace: UmFuZG9tSVaiFNEWI4d+dBvj/9mLEcZetaQrDIEZ8+gNV2SyuWORv3bEFBarYrNi
Message-ID: <3EBD344E.6010906@rfk.com>
Date: Sat, 10 May 2003 13:18:06 -0400
From: "Larry Hall (RFK Partners, Inc.)" <lhall AT rfk DOT com>
Reply-To: lhall AT rfk DOT com
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.3) Gecko/20030312
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: Dario Alcocer <alcocer AT helixdigital DOT com>
CC: cygwin AT cygwin DOT com
Subject: Re: cygipc (and PostgreSQL) XP problem resolved!
References: <20030506174725 DOT GE1652 AT tishler DOT net> <3EB84F52 DOT 3020608 AT ece DOT gatech DOT edu> <20030507133326 DOT GA1824 AT tishler DOT net> <3EB9A54B DOT 8060500 AT ece DOT gatech DOT edu> <20030508135217 DOT GD512 AT tishler DOT net> <3EBB22F5 DOT 4000801 AT ece DOT gatech DOT edu> <1052541657 DOT 1675 DOT 5 DOT camel AT localhost> <20030510072110 DOT B12261 AT ns DOT helixdigital DOT com> <20030510155401 DOT GI19367 AT cygbert DOT vinschen DOT de> <20030510095611 DOT A20781 AT ns DOT helixdigital DOT com>
In-Reply-To: <20030510095611.A20781@ns.helixdigital.com>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit

Dario Alcocer wrote:
> On Sat, May 10, 2003 at 05:54:01PM +0200, Corinna Vinschen wrote:
> 
>>On Sat, May 10, 2003 at 07:21:10AM -0700, Dario Alcocer wrote:
>>
>>>On Sat, May 10, 2003 at 02:40:57PM +1000, Robert Collins wrote:
>>>
>>>>Well, theres the API break coming up with cygwin1.dll - do it in sync
>>>>with that, IMO.
>>>
>>>Correct me if I'm wrong, but shouldn't cygwin1.dll be changed to
>>>cygwin2.dll if the API is changing?
>>
>>Only if it's not a backward compatible change.  Each new Cygwin version
>>did already add new functions so the API minor number has been bumped
>>up.
> 
> 
> But what if the data types used by the existing functions are changed?
> In other words, the old library has void FooFunc(off_t) where off_t is a
> 4-byte quantity, but the new library uses an 8-byte off_t?


This has been answered.  If the app/lib/DLL isn't recompiled, there's no
change.  Everything works as it always has.  If the app/lib/DLL is
recompiled, it gets the new 64 bit versions.  Then, changes may be
required to get all the dependencies back in sync.


> The reason I ask is that I've seen binaries compiled for an older
> cygwin1.dll run incorrectly or crash when dynamically linking against a
> new cygwin1.dll.
> 
> For instance, the PowerTV tool chain uses GNUPro 99, which has an
> ancient cygwin1.dll included. As long as I don't install the latest
> Cygwin stuff, everything is ok. However, as soon as I install the
> latest cygwin1.dll, the PowerTV gcc/binutils cease to function
> correctly. I suspect it has something to do with the Cygwin API and/or
> ABI changing, and since the Cygwin DLL filename is exactly the same, it
> loads the new library instead of using the old one.


Cygwin will bump the version number for the DLL (to cygwin2.dll) when
a backward incompatible change is made.  If you've had problems in the
past with the GNUPro toosl, you'd need to consult the provider of those
tools.  That's not a product of this project so this site doesn't support
it.  I can't speak to whether an ABI incompatible change has been made
w.r.t. the version of the Cygwin DLL that was distributed with that product.


> They way I've gotten around this issue is to binary edit the PowerTV
> tool chain and change references to the cygwin1.dll to refer to
> cygwin0.dll; this is a very ugly hack, and I don't recommend it. But it
> does serve to highlight the kinds of problems people might be running
> into in the field.


If people have run into this kind of problem using the Cygwin distribution,
it needs to be reported.  I haven't seen anything on this list that would
indicate that's the case though.


-- 
Larry Hall                              http://www.rfk.com
RFK Partners, Inc.                      (508) 893-9779 - RFK Office
838 Washington Street                   (508) 893-9889 - FAX
Holliston, MA 01746


--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Problem reports:       http://cygwin.com/problems.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/