[omniORB] memory allocation/free under Win32
Francois Laflamme
francois.laflamme@voxco.com
Thu, 6 Jul 2000 07:27:56 -0400
This message is in MIME format. Since your mail reader does not understand
this format, some or all of this message may not be legible.
------_=_NextPart_001_01BFE73D.3BA120E0
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Hi Jonathan,
I had the same kind of problem with omni 2.8.=20
It seem that you use the wrong library with the wrong switchs.=20
For the debug version of your project you should use:
CL /MDd (Debug MultiThread DLL) and link with omni*rtd.lib=20
For the release version use:
CL /MD (MultiThread DLL) and link with omni*_rt.lib
Also make sure that you pass the program arguments to the orb and =
boa/poa.=20
ex: CORBA::ORB_var Orb =3D CORBA::ORB_init(argc,argv,"omniORB2");
Hope that will help,
Fran=E7ois Laflamme=20
VOXCO
1134, Ste-Catherine Ouest=20
Suite 600=20
Montr=E9al, Qc=20
H3B 1H4=20
EMail:=A0 mailto:francois.laflamme@voxco.com
Tel.:=A0=A0 514-861-9255=20
Fax:=A0=A0=A0 514-861-9209=20
> -----Original Message-----
> From: Jonathan Clark [mailto:jc@jonathanclark.com]
> Sent: 6 juillet, 2000 06:38
> To: omniorb-list@uk.research.att.com
> Subject: [omniORB] memory allocation/free under Win32
>=20
>=20
> I've had some problems with memory allocation and
> deallocation under Win32. The problem is that all corba=20
> free() type operations cause a crash when compiled/linked the wrong=20
> way.=20
>=20
> I fixed the problem previously by playing with cl.exe and link.exe
> options, but recently I've moved to a new build process (and started
> using omniorb 3.0) and the problem has resurfaced. I'm not sure what =
> I'm doing wrong. Can you shed any light on this? I'm using cl /MD=20
> (also tried /MT) to compile and linking with omni's *_rt.lib (also=20
> tried the debug versions *_rtd.lib).
>=20
> Thanks,
>=20
> Jonathan
>=20
> -----Original Message-----
> From: David Scott [mailto:djs@uk.research.att.com]
> Sent: Thursday, July 06, 2000 2:23 AM
> To: Jonathan Clark
> Subject: RE: [omniORB] omniidl 3.0 request and questions
>=20
>=20
> On Thu, 6 Jul 2000, Jonathan Clark wrote:
>=20
> > ah ha. thanks! :)
> >
> No problem :)
>=20
> Thanks for mentioning this- it's really difficult to spot=20
> where things are
> a little strange/ unclear/ unintuitive when you're so close=20
> to something!
>=20
> David
>=20
> --=20
> Dave Scott, Research Engineer, AT&T Laboratories Cambridge UK
>=20
>=20
------_=_NextPart_001_01BFE73D.3BA120E0
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
<HTML>
<HEAD>
<META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; =
charset=3Diso-8859-1">
<META NAME=3D"Generator" CONTENT=3D"MS Exchange Server version =
5.5.2652.35">
<TITLE>RE: [omniORB] memory allocation/free under Win32</TITLE>
</HEAD>
<BODY>
<P><FONT SIZE=3D2>Hi Jonathan,</FONT>
</P>
<P><FONT SIZE=3D2>I had the same kind of problem with omni 2.8. </FONT>
<BR><FONT SIZE=3D2>It seem that you use the wrong library with the =
wrong switchs. </FONT>
</P>
<P><FONT SIZE=3D2>For the debug version of your project you should =
use:</FONT>
<BR><FONT SIZE=3D2>CL /MDd (Debug MultiThread DLL) and link with =
omni*rtd.lib </FONT>
</P>
<P><FONT SIZE=3D2>For the release version use:</FONT>
<BR><FONT SIZE=3D2>CL /MD (MultiThread DLL) and link with =
omni*_rt.lib</FONT>
</P>
<P><FONT SIZE=3D2>Also make sure that you pass the program arguments to =
the orb and boa/poa. </FONT>
<BR><FONT SIZE=3D2>ex: CORBA::ORB_var Orb =3D =
CORBA::ORB_init(argc,argv,"omniORB2");</FONT>
</P>
<P><FONT SIZE=3D2>Hope that will help,</FONT>
</P>
<P><FONT SIZE=3D2>Fran=E7ois Laflamme </FONT>
<BR><FONT SIZE=3D2>VOXCO</FONT>
</P>
<P><FONT SIZE=3D2>1134, Ste-Catherine Ouest </FONT>
<BR><FONT SIZE=3D2>Suite 600 </FONT>
<BR><FONT SIZE=3D2>Montr=E9al, Qc </FONT>
<BR><FONT SIZE=3D2>H3B 1H4 </FONT>
</P>
<P><FONT SIZE=3D2>EMail:=A0 <A =
HREF=3D"mailto:francois.laflamme@voxco.com">mailto:francois.laflamme@vox=
co.com</A></FONT>
<BR><FONT SIZE=3D2>Tel.:=A0=A0 514-861-9255 </FONT>
<BR><FONT SIZE=3D2>Fax:=A0=A0=A0 514-861-9209 </FONT>
</P>
<BR>
<BR>
<BR>
<P><FONT SIZE=3D2>> -----Original Message-----</FONT>
<BR><FONT SIZE=3D2>> From: Jonathan Clark [<A =
HREF=3D"mailto:jc@jonathanclark.com">mailto:jc@jonathanclark.com</A>]</F=
ONT>
<BR><FONT SIZE=3D2>> Sent: 6 juillet, 2000 06:38</FONT>
<BR><FONT SIZE=3D2>> To: omniorb-list@uk.research.att.com</FONT>
<BR><FONT SIZE=3D2>> Subject: [omniORB] memory allocation/free under =
Win32</FONT>
<BR><FONT SIZE=3D2>> </FONT>
<BR><FONT SIZE=3D2>> </FONT>
<BR><FONT SIZE=3D2>> I've had some problems with memory allocation =
and</FONT>
<BR><FONT SIZE=3D2>> deallocation under Win32. The problem is =
that all corba </FONT>
<BR><FONT SIZE=3D2>> free() type operations cause a crash when =
compiled/linked the wrong </FONT>
<BR><FONT SIZE=3D2>> way. </FONT>
<BR><FONT SIZE=3D2>> </FONT>
<BR><FONT SIZE=3D2>> I fixed the problem previously by playing with =
cl.exe and link.exe</FONT>
<BR><FONT SIZE=3D2>> options, but recently I've moved to a new build =
process (and started</FONT>
<BR><FONT SIZE=3D2>> using omniorb 3.0) and the problem has =
resurfaced. I'm not sure what </FONT>
<BR><FONT SIZE=3D2>> I'm doing wrong. Can you shed any light =
on this? I'm using cl /MD </FONT>
<BR><FONT SIZE=3D2>> (also tried /MT) to compile and linking with =
omni's *_rt.lib (also </FONT>
<BR><FONT SIZE=3D2>> tried the debug versions *_rtd.lib).</FONT>
<BR><FONT SIZE=3D2>> </FONT>
<BR><FONT SIZE=3D2>> Thanks,</FONT>
<BR><FONT SIZE=3D2>> </FONT>
<BR><FONT SIZE=3D2>> Jonathan</FONT>
<BR><FONT SIZE=3D2>> </FONT>
<BR><FONT SIZE=3D2>> -----Original Message-----</FONT>
<BR><FONT SIZE=3D2>> From: David Scott [<A =
HREF=3D"mailto:djs@uk.research.att.com">mailto:djs@uk.research.att.com</=
A>]</FONT>
<BR><FONT SIZE=3D2>> Sent: Thursday, July 06, 2000 2:23 AM</FONT>
<BR><FONT SIZE=3D2>> To: Jonathan Clark</FONT>
<BR><FONT SIZE=3D2>> Subject: RE: [omniORB] omniidl 3.0 request and =
questions</FONT>
<BR><FONT SIZE=3D2>> </FONT>
<BR><FONT SIZE=3D2>> </FONT>
<BR><FONT SIZE=3D2>> On Thu, 6 Jul 2000, Jonathan Clark =
wrote:</FONT>
<BR><FONT SIZE=3D2>> </FONT>
<BR><FONT SIZE=3D2>> > ah ha. thanks! :)</FONT>
<BR><FONT SIZE=3D2>> ></FONT>
<BR><FONT SIZE=3D2>> No problem :)</FONT>
<BR><FONT SIZE=3D2>> </FONT>
<BR><FONT SIZE=3D2>> Thanks for mentioning this- it's really =
difficult to spot </FONT>
<BR><FONT SIZE=3D2>> where things are</FONT>
<BR><FONT SIZE=3D2>> a little strange/ unclear/ unintuitive when =
you're so close </FONT>
<BR><FONT SIZE=3D2>> to something!</FONT>
<BR><FONT SIZE=3D2>> </FONT>
<BR><FONT SIZE=3D2>> David</FONT>
<BR><FONT SIZE=3D2>> </FONT>
<BR><FONT SIZE=3D2>> -- </FONT>
<BR><FONT SIZE=3D2>> Dave Scott, Research Engineer, =
AT&T Laboratories Cambridge UK</FONT>
<BR><FONT SIZE=3D2>> </FONT>
<BR><FONT SIZE=3D2>> </FONT>
</P>
</BODY>
</HTML>
------_=_NextPart_001_01BFE73D.3BA120E0--