Mailing-List: contact cygwin-help AT sourceware DOT cygnus DOT com; run by ezmlm
List-Subscribe: <mailto:cygwin-subscribe AT sources DOT redhat DOT com>
List-Archive: <http://sources.redhat.com/ml/cygwin/>
List-Post: <mailto:cygwin AT sources DOT redhat DOT com>
List-Help: <mailto:cygwin-help AT sources DOT redhat DOT com>, <http://sources.redhat.com/ml/#faqs>
Sender: cygwin-owner AT sources DOT redhat DOT com
Delivered-To: mailing list cygwin AT sources DOT redhat DOT com
Message-ID: <3AAAA33F.6EAAA325@ece.gatech.edu>
Date: Sat, 10 Mar 2001 16:57:19 -0500
From: Charles Wilson <cwilson AT ece DOT gatech DOT edu>
X-Mailer: Mozilla 4.75 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: cygwin AT sources DOT redhat DOT com
Subject: Re: windows showstopper re: cannot find entry symbol _WinMainCRTStartup
References: <3AAA1EB2 DOT 14C8F61 AT 666 DOT com> <3AAA9FDF DOT 579DC5 AT ece DOT gatech DOT edu>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Ben Wing wrote:
> 
> I do see the crt0 is the very first thing linked, and
> 
> /lib 540$ nm crt0.o
> 00000000 b .bss
> 00000000 d .data
> 00000000 t .text
> 00000000 D ___cygwin_crt0_bp
> 00000000 t ___gnu_compiled_c
>          U _cygwin_crt0
>          U _main
> 00000000 T _mainCRTStartup
> 00000000 t gcc2_compiled.
> 
> so this *may* be harmless, but I think it needs to be fixed, perhaps just with a
> flag that sets the entry point.
> 
> BUT ....  then again, maybe not.  There's also a crt1.0, like this:
> 
> /lib 542$ nm crt1.o
> 00000000 b .bss
> 00000000 d .data
> 00000000 t .text
>          U _ExitProcess AT 4
>          U _SetUnhandledExceptionFilter AT 4
> 00000230 T _WinMainCRTStartup
>          U __CRT_fmode
>          U __CRT_glob
>          U ___GetMainArgs
> 00000000 t ___gnu_compiled_c
> 000001b0 t ___mingw_CRTStartup
> 00000000 D __argc
> 00000004 D __argv
>          U __cexit
>          U __fileno
>          U __fpreset
> 000000d0 t __gnu_exception_handler AT 4
>          U __imp___environ_dll
>          U __imp___fmode_dll
>          U __imp___iob
> 00000030 t __mingw32_init_fmode
> 00000000 t __mingw32_init_mainargs
>          U __setmode
>          U _main
> 00000210 T _mainCRTStartup
>          U _signal
> 00000000 t gcc2_compiled.
> 
> Notice that this also has mainCRTStartup in it, as well as WinMainCRTStartup.
> My strong suspicion is that crt1.o is the correct CRT to be using for a
> windowed-subsystem XEmacs; crt0.o, with no references to WinMain, has to be for
> a console-subsystem XEmacs.  So, by using the wrong CRT initialization, we could
> easily be fucking ourselves in all sorts of ways.

Hmmm...I think that crt0.o is for cygwin builds, crt1.o is for builds
which use "cygwin-gcc -mno-cygwin" with a crt.dll runtime target, and
crt2.o is for builds which use "cygwin-gcc -mno-cygwin" with an
msvcrt.dll runtime target.

> 
> So once again, what changed to produce this?

If I'm right, cygwin's gcc recently changed so that the msvcrt runtime
was the default target for -mno-cygwin builds; but that would just mean
that crt1.o was replaced by crt2.o.  Since cygwin builds (-mwindows
notwithstanding) use crt0.o, this change would have no effect on cygwin
builds.  Hmm...

I've copied this message to the cygwin list, in case the gurus there
have any comment, or can confirm/deny my statement abouve about crtX.o.

--Chuck

--
Want to unsubscribe from this list?
Check out: http://cygwin.com/ml/#unsubscribe-simple