Re: Error in egcs on solaris

[dvv at dvv dot ru (Dima Volodin)]

It's a bug in g++ - the compiler shouldn't have let you get away with
this cast:

 x.Set(&x, (Method) &X::bug);
           ^^^^^^^^^^^^^^^^

Dima


On Fri, 14 May 1999 16:25:08 +0200 (MET DST), you wrote:

>
>Does anybody have any comments to this?
>
>Best regards,
>--------------------------------------------------------------------------
> Klavs Wismann             -------------
> Odegaard A/S             |   /\        |    telephone:  +45 35 26 60 11     
> 1, Kroghsgade            |__/  \  /\___|      telefax:  +45 35 26 50 18
> 2100 Copenhagen East     |      \/     |       e-mail:      kw@oedan.dk
> Denmark                   -------------     home page: www.oedegaard.com
>--------------------------------------------------------------------------
>
>---------- Forwarded message ----------
>Date: Fri, 7 May 1999 10:58:54 +0200 (MET DST)
>From: Klavs Wismann <kw@oedan.dk>
>To: egcs-bugs@egcs.cygnus.com
>Subject: Error in egcs on solaris
>
>
>I'm using egcs-1.1.2 on solaris 2.6, Sun Ultra 5.
>
>I have a problem executing this small program test.C. The output should
>be:
>before2
>before
>bug
>after
>after2
>(This is produced with same program test.C and compiler on AIX 4.2.1)
>
>but solaris produces:
>before2
>before
>bug
>after2
>
>Somehow the 'Method bug' method does not return to Callback::Call.
>
>As I said earlier the same code and compiler works fine on AIX.
>
>Can anyone tell what is wrong and hopefully provide a fix?!?!?!
>
>
>Best regards,
>--------------------------------------------------------------------------
> Klavs Wismann             -------------
> Odegaard A/S             |   /\        |    telephone:  +45 35 26 60 11     
> 1, Kroghsgade            |__/  \  /\___|      telefax:  +45 35 26 50 18
> 2100 Copenhagen East     |      \/     |       e-mail:      kw@oedan.dk
> Denmark                   -------------     home page: www.oedegaard.com
>--------------------------------------------------------------------------