Experimental Physics and
Industrial Control System

Kazuro FURUKAWA <[email protected]> · Fri, 03 Oct 2008 08:21:54 +0900

Hello Andrew, hello Jeff, 

Thank you for your prompt processing.  And I've just realized that 
I missed your RC1 announcement.  Basically the same patch can work 
for RC1 as well.  

cap5 was enabled in RC1 but Tru64unix (osf-alpha) cannot co-op with 
the build script.  On Tru64unix, 

  perl is in /usr/bin/
  perl and xsubpp is in /usr/lib/perl-5.8.0/bin/
  (and I have perl and xsubpp in /usr/new/pkg/perl5.8/bin/)

I've tried several PATH orders, but somehow the error persists;

  /bin/sh: xsubpp: not found
or
  /bin/sh: /usr/ucb/xsubpp: not found

It seems that the typemap file location is recognized correctly. 
So I don't understand the behavior yet.  I'll look it in later.  

Regards. 

>>> On Thu, 02 Oct 2008 11:39:46 JST,  Andrew Johnson <[email protected]> wrote;
> Hi Kazuro,
> 
> Thanks for your report.
> 
> On Thursday 02 October 2008 08:46:53 Kazuro FURUKAWA wrote:
> >
> > As usual, osf-alpha needs the attached patch, that is almost the
> > same as that against 3.14.9.  I know it is dirty, but I don't know
> > any other solution.  And the resultant build and runtest logs are
> > placed at
> >   <URL:http://www-linac.kek.jp/epics/osf/>
> 
> Jeff Hill has already commented on your modifications to comBuf.h and I have 
> just committed the change to CONFIG.Common.osf-alpha.  As to your 
> implementation of epicsVsnprintf() we do not think this is necessary, since 
> osf-alpha is not the only target architecture that suffers from this 
> particular problem and .  I have modified the text in the Application 
> Developers Guide that describes these two functions to make it clear that the 
> return value is not guaranteed.  I may take out the return value tests from 
> the libCom/test/epicsStdioTest.c program too.
> 
> > Darwin-x86 showed error in the linkage stage of libCom shared lib,
> >   /usr/bin/c++ -o libCom.3.14.10pre1.dylib -dynamiclib ...
> ...
> 
> We already fixed the problem with the version number that affected both darwin 
> and WIN32, which was caused by my putting the "-pre1" part in the wrong 
> variable in CONFIG_BASE_VERSION.  We have taken steps to ensure that this 
> shouldn't happen in the future, and it shouldn't have been an issue in 
> the -RC1 release.
> 
> - Andrew
> -- 
> Talk is cheap. Show me the code. -- Linus Torvalds

-----
Kazuro FURUKAWA <[email protected]>
 Linac&KEKB,  High Energy Accelerator Research Organization (KEK), Japan

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System