Experimental Physics and
Industrial Control System

Tom Pelaia II <[email protected]> · Tue, 23 Oct 2007 08:30:45 -0400

Hi,
I've got a standard installer for Mac OS X (PPC) which installs EPICS
base, EDM, other common EPICS tools and an Objective-C framework for
EPICS. No building is necessary. However, you do have to install open
motif and libpng from Mac Ports http://www.macports.org/ yourself.
If you want the installer, you can download it at: http://ics-
web1.sns.ornl.gov/

and select "Software" and then "EPICS Framework".
The download includes a README file with instructions and an
explanation of what gets installed.
Since I made the switch to Intel Macs, the next installer will be for
Intel Macs only, unless universal binaries become available for open
motif. The last time I checked, Mac Ports still didn't have a
universal binary for open motif. That dependency is the only obstacle
for creating a universal binary.
Note that I have a newer version of the Objective-C EPICS framework
for Intel Macs which is more powerful than the one that gets
installed with the download. I haven't released it yet, because I
plan to make further significant changes after Leopard is released.
thanks,
tom

On Oct 23, 2007, at 12:36 AM, Bertrand H.J. Biritz wrote:
Sorry for the sudden silence. Last week was really busy for me.
I realized I had not followed Eric's suggestion fully of modifying
the Makefile by adding the Com. When I tried this I actually got
gcc from Apple to give me an error message.

Now I don't remember what the error message was exactly since in
the mean time I have had to return my PowerBook to Apple for repair
(it would freeze on me at least twice a day).
Once I get it back I will try once more - although the test might
be different since I am one of those fanboys who will install 10.5
immediately when it comes out on Friday :)
Thanks for all the help,
Bertrand
Quoting Eric Norum <[email protected]>:
Ah yes, it's coming back to me now.

That's why I had to add libCom to the beginning of the library
list --

to ensure that the constructors ran in the right order.

........did I mention that I really don't like C++......

On Oct 16, 2007, at 4:38 PM, Matthieu Bec wrote:

I had a vaguely similar issue on Linux. This thread on tech-talk
helped me resolve the problem:
http://www.aps.anl.gov/epics/tech-talk/2006/msg01325.php
Matthieu
Bertrand H.J. Biritz wrote:

Here is that output:

(gdb) run

Starting program: /usr/local/epics/extensions/bin/darwin-ppc/edm

Reading symbols for shared libraries ...++..+.++++...++++ done

Program received signal EXC_BAD_ACCESS, Could not access memory.

Reason: KERN_PROTECTION_FAILURE at address: 0x00000008

0x0021c7dc in epicsMutex::lock (this=0x1e6ecd0) at ../../../src/
libCom/osi/epicsMutex.cpp:118

118 epicsMutexOsdLock(pmutexNode->id);

(gdb) bt

#0 0x0021c7dc in epicsMutex::lock (this=0x1e6ecd0) at ../../../
src/libCom/osi/epicsMutex.cpp:118

#1 0x01e492f4 in ca_client_context::ca_client_context
(this=0x2400b20, enablePreemptiveCallback=false) at ../../../
include/epicsGuard.h:68

#2 0x01e2e69c in ca_context_create
(premptiveCallbackSelect=ca_disable_preemptive_callback) at ../
access.cpp:205

#3 0x018a23f0 in EPICS_PV_Factory::EPICS_PV_Factory
(this=0x2400930) at ../epics_pv_factory.cc:64

#4 0x018d525c in __static_initialization_and_destruction_0
(__initialize_p=31911120, __priority=37752208) at ../
pv_factory.cc:59

#5 0x8fe15608 in
__dyld__ZN16ImageLoaderMachO16doInitializationERKN11ImageLoader11Li
nkContextE ()

#6 0x8fe0ba68 in
__dyld__ZN11ImageLoader23recursiveInitializationERKNS_11LinkContext
E ()

#7 0x8fe0b9f8 in
__dyld__ZN11ImageLoader23recursiveInitializationERKNS_11LinkContext
E ()

#8 0x8fe0b9f8 in
__dyld__ZN11ImageLoader23recursiveInitializationERKNS_11LinkContext
E ()

#9 0x8fe0d5ec in
__dyld__ZN11ImageLoader15runInitializersERKNS_11LinkContextE ()

#10 0x8fe02cb4 in __dyld__ZN4dyld24initializeMainExecutableEv ()

#11 0x00002904 in _start ()

#12 0x000027d0 in start ()

On Oct 16, 2007, at 1:19 PM, Mark Rivers wrote:

Type "bt" at the gdb prompt to get a stack trace and see what
function

you are in when the access fault occurs.
Mark

--
Matthieu Bec              Gemini Observatory
Tel: +56 51 205785        c/o AURA, Casilla 603
Fax: +56 51 205650        La Serena, Chile

--
Eric Norum <[email protected]>
Advanced Photon Source
Argonne National Laboratory
(630) 252-4793

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System