Experimental Physics and
Industrial Control System

Vladis Korobov <[email protected]> · Thu, 30 Jan 2003 18:12:11 +0100 (MET)

On Thu, 30 Jan 2003, Jie Chen wrote:

> Vladis Korobov wrote:
>
> >---------- Forwarded message ----------
> >Date: Wed, 29 Jan 2003 15:39:35 +0100 (MET)
> >From: Vladis Korobov <[email protected]>
> >To: [email protected]
> >Subject: Can't build the CMLOG under Tornado 2.0 for mv162, ppc etc.
> >
> >Hi all,
> >
> >I have problem to compile CMLOG stuff (namely, cmlogClientD) under Tornado 2.0. for
> >any targets: mv162, ppc, etc.
> >Although, under Tornado I it was built OK.
> >
> >Under Tonado I compilation looks like:
> >
> >/usr/wind/host/sun4-solaris2/bin/c++68k -I. -O -I/usr/wind/target.68k/h -DCPU=MC68040
> >-DCPU_FAMILY=MC680X0 -fstrength-reduce -fforce-mem -finline-functions -fno-builtin
> >-nostdinc -ansi -pipe -Dvxworks -fno-for-scope -I. -I../data -I../Thread -I../Common
> >-I../Network -I/u/mks/slava/EpicsR3.13.1/extensions/CMLOG-2_1_B1/C++/SACE-4.0
> >-c cmlogClientD.cc -o .obj/mv162/cmlogClientD.o
> >cmlogClientD.cc: In function `int cmlogClientD(int, int)':
> >cmlogClientD.cc:104: warning: ANSI C++ prohibits conversion from `(class cpThreadManager *)'
> >to `(...)'
> >rm -f cmlogClientD
> >/usr/wind/host/sun4-solaris2/bin/ld68k -r -o cmlogClientD .obj/mv162/cmlogMsg.o
> >.obj/mv162/cmlogUtil.o .obj/mv162/cmlogNetUtil.o .obj/mv162/cmlogSlist.o
> >.obj/mv162/cmlogSvcLocaterRep.o .obj/mv162/cmlogServerLocater.o
> >.obj/mv162/cmlogClntCxt.o .obj/mv162/cmlogUpIO.o .obj/mv162/cmlogServerHB.o
> >.obj/mv162/cmlogUxToServer.o .obj/mv162/cmlogClntDSigHandler.o
> >.obj/mv162/cmlogClientD.o -L../lib/mv162 -ldata
> >-L/u/mks/slava/EpicsR3.13.1/extensions/CMLOG-2_1_B1/C++/SACE-4.0/lib/mv162 -lSACE
> >if test ! -d ../bin/mv162; then mkdir -p ../bin/mv162; else echo; fi
> >
> >mv cmlogClientD ../bin/mv162
> >gmake[2]: Leaving directory `/import/epicsf2/group/mks/slava/EpicsR3.13.1/
> >extensions/CMLOG-2_1_B1/ClientD'
> >gmake[1]: Leaving directory `/import/epicsf2/group/mks/slava/EpicsR3.13.1/
> >extensions/CMLOG-2_1_B1/ClientD'
> >gmake[1]: Entering directory `/import/epicsf2/group/mks/slava/EpicsR3.13.1/
> >extensions/CMLOG-2_1_B1/Client'
> >---------------------------------------------------------------------------
> >
> >Under Tornado 2.0 it looks like:
> >
> >/import/epicsf2/u2/tornado_2.0/host/sun4-solaris2/bin/c++68k -I. -O
> >-I/import/epicsf2/u2/tornado_2.0/target.m68k/h -DCPU=MC68040 -DCPU_FAMILY=MC680X0
> >-fstrength-reduce -fforce-mem -finline-functions -fno-builtin -nostdinc -ansi -pipe
> >-Dvxworks -fno-for-scope -I. -I../data -I../Thread -I../Common -I../Network
> >-I/u/mks/slava/EpicsR3.13.7/extensions/CMLOG-2_1_B1/C++/SACE-4.0 -c cmlogClientD.cc
> >-o .obj/mv162/cmlogClientD.o
> >cmlogClientD.cc: In function `cmlogClientD':
> >cmlogClientD.cc:104: warning: ANSI C++ prohibits conversion from `(class cpThreadManager *)'
> >to `(...)'
> >cmlogClientD.cc: In function `realCmlogClientD':
> >cmlogClientD.cc:277: no matching function for call to `cpThread::add_cleanup_handler (void ()(ACE_Reactor *), void *)'
> >../Thread/cpThread.h:350: candidates are: cpThread::add_cleanup_handler(void (*)(void *), void *)
> >gmake[2]: *** [.obj/mv162/cmlogClientD.o] Error 1
> >gmake[2]: Leaving directory `/import/epicsf2/group/mks/slava/EpicsR3.13.7/
> >extensions/CMLOG-2_1_B1/ClientD'
> >gmake[1]: *** [mv162] Error 2
> >gmake[1]: Leaving directory `/import/epicsf2/group/mks/slava/EpicsR3.13.7/
> >extensions/CMLOG-2_1_B1/ClientD'
> >gmake[1]: Entering directory `/import/epicsf2/group/mks/slava/EpicsR3.13.7/
> >extensions/CMLOG-2_1_B1/Client'
> >-------------------------------------------------------------------------------
> >
> >If anybody knows where is the problem, please, help me. Thank you in advance.
> >Regards,
> >V.Korobov.
> >
> >
> >
> >
> >
> >
> Hi, Mr. Korobov:
>
>     The Tornado 2.0 has new version of gcc which requires strict
> function prototype. In order to make the cmlogClientD.cc to compile, you
> only need to change
>
> static void
> cmlogClientDCleanup (ACE_Reactor* reactor)
> {
>  reactor->close ();
> }
>
>
> to the following
>
> static void
> cmlogClientDCleanup (void* reactor)
> {
>  ((ACE_Reactor *)reactor)->close ();
> }
>

I've done this change and the problem was solved. Thank You.

> The newer version of cmlog has this fixed. The new version of cmlog can
> be obtained at
> http://www.jlab.org/cdev/cmlog.html

We'll get the newer version of CMLOG from this site.

Regards,
V:Korobov

>
> --
> #####################################################
> Jie Chen
> Thomas Jefferson National Accelerator Facility
> 12000, Jefferson Ave.
> Newport News, VA 23606
>
> (757)269-5046 (office) (757)269-6248 (fax)
> [email protected]
> #####################################################
>
>
>
>

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System