<?xml version="1.0" encoding="UTF-8"?>
<rss xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:taxo="http://purl.org/rss/1.0/modules/taxonomy/" version="2.0">
  <channel>
    <title>topic Two C++ compilers on one cluster in Operating System - OpenVMS</title>
    <link>https://community.hpe.com/t5/operating-system-openvms/two-c-compilers-on-one-cluster/m-p/4073145#M38493</link>
    <description>ls.&lt;BR /&gt;&lt;BR /&gt;Is it possible to install more C++ compilers on a single (or cluster) system.&lt;BR /&gt;Here they want to install C++ V6.5 and C++ V7.1 on the same node.&lt;BR /&gt;&lt;BR /&gt;I adviced not to do, but some people are very cocky.&lt;BR /&gt;&lt;BR /&gt;Greetings,&lt;BR /&gt;&lt;BR /&gt;Piet Timmers</description>
    <pubDate>Thu, 20 Sep 2007 01:51:54 GMT</pubDate>
    <dc:creator>Piet Timmers_1</dc:creator>
    <dc:date>2007-09-20T01:51:54Z</dc:date>
    <item>
      <title>Two C++ compilers on one cluster</title>
      <link>https://community.hpe.com/t5/operating-system-openvms/two-c-compilers-on-one-cluster/m-p/4073145#M38493</link>
      <description>ls.&lt;BR /&gt;&lt;BR /&gt;Is it possible to install more C++ compilers on a single (or cluster) system.&lt;BR /&gt;Here they want to install C++ V6.5 and C++ V7.1 on the same node.&lt;BR /&gt;&lt;BR /&gt;I adviced not to do, but some people are very cocky.&lt;BR /&gt;&lt;BR /&gt;Greetings,&lt;BR /&gt;&lt;BR /&gt;Piet Timmers</description>
      <pubDate>Thu, 20 Sep 2007 01:51:54 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-openvms/two-c-compilers-on-one-cluster/m-p/4073145#M38493</guid>
      <dc:creator>Piet Timmers_1</dc:creator>
      <dc:date>2007-09-20T01:51:54Z</dc:date>
    </item>
    <item>
      <title>Re: Two C++ compilers on one cluster</title>
      <link>https://community.hpe.com/t5/operating-system-openvms/two-c-compilers-on-one-cluster/m-p/4073146#M38494</link>
      <description>As fas as I know, you can even install multiple versions on one system, a procedure sys$system:cxx$show_version shows the installed versions, cxx$set_version activates one version for you.&lt;BR /&gt;&lt;BR /&gt;The CXX installation kit asks if you want to retain an older version.&lt;BR /&gt;&lt;BR /&gt;regards Kalle</description>
      <pubDate>Thu, 20 Sep 2007 03:38:03 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-openvms/two-c-compilers-on-one-cluster/m-p/4073146#M38494</guid>
      <dc:creator>Karl Rohwedder</dc:creator>
      <dc:date>2007-09-20T03:38:03Z</dc:date>
    </item>
    <item>
      <title>Re: Two C++ compilers on one cluster</title>
      <link>https://community.hpe.com/t5/operating-system-openvms/two-c-compilers-on-one-cluster/m-p/4073147#M38495</link>
      <description>Installing compilers is not "daily business" so I have to rely on memory.&lt;BR /&gt;&lt;BR /&gt;Compilers are, AFAIK, directly installed on [VMS$COMMON] and subsequent directories (SYSEXE, SYSLIB). In that respect, the answer is NO, unless, in a cluster nvironment, run systems from a local system disk. That means: each node has it's own installation of C++, and that can be different versions.&lt;BR /&gt;&lt;BR /&gt;in the past, I have tried installing compilers on a different location, and in that case, directory [VMS$COMMON] and subsequent directories will be created, on that location. To get this running, you'll need to redefine a number of system logicals to include these locations. However, OpenVMS activity is very much depending on these logicals so the slightest mastake will render the system ununsable, or worst case: inaccessable.&lt;BR /&gt;I found it _might_ work for compilation but it did severely interfere with linking the program (unresolved references) that did not occutr after teh compiler was installed "properly"&lt;BR /&gt;&lt;BR /&gt;So the answer is: it MIGHT be possible but it is VERY VERY UNWISE to do so.&lt;BR /&gt;&lt;BR /&gt;(Suggestion: take a (small) machine to run the separate compiler. You may thiink of  Alpha emulation on a PC as well.</description>
      <pubDate>Thu, 20 Sep 2007 03:40:18 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-openvms/two-c-compilers-on-one-cluster/m-p/4073147#M38495</guid>
      <dc:creator>Willem Grooters</dc:creator>
      <dc:date>2007-09-20T03:40:18Z</dc:date>
    </item>
    <item>
      <title>Re: Two C++ compilers on one cluster</title>
      <link>https://community.hpe.com/t5/operating-system-openvms/two-c-compilers-on-one-cluster/m-p/4073148#M38496</link>
      <description>Just had a look into the CXX-installguide, see section 2.1:&lt;BR /&gt;&lt;BR /&gt;excerpt:&lt;BR /&gt;&lt;BR /&gt;       2.1 Installing and Using Multiple Compiler Versions&lt;BR /&gt;&lt;BR /&gt;              C++ Version 7.1 provides support for installing and using&lt;BR /&gt;              multiple versions of the compiler on the same node.&lt;BR /&gt;...&lt;BR /&gt;&lt;BR /&gt;regards kalle</description>
      <pubDate>Thu, 20 Sep 2007 05:04:42 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-openvms/two-c-compilers-on-one-cluster/m-p/4073148#M38496</guid>
      <dc:creator>Karl Rohwedder</dc:creator>
      <dc:date>2007-09-20T05:04:42Z</dc:date>
    </item>
    <item>
      <title>Re: Two C++ compilers on one cluster</title>
      <link>https://community.hpe.com/t5/operating-system-openvms/two-c-compilers-on-one-cluster/m-p/4073149#M38497</link>
      <description>Piet,&lt;BR /&gt;AFAIK, having 2 versions of the C++ compiler sitting next to one another is supported (we have several development systems that have the 6.5 and the 7.x variant).&lt;BR /&gt;Karl is correct about the command procedures.&lt;BR /&gt;The installation procedure of the (newer) C++ compiler also asks which version is to be the default.&lt;BR /&gt;Regards,&lt;BR /&gt;Kris (aka Qkcl)</description>
      <pubDate>Thu, 20 Sep 2007 07:52:07 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-openvms/two-c-compilers-on-one-cluster/m-p/4073149#M38497</guid>
      <dc:creator>Kris Clippeleyr</dc:creator>
      <dc:date>2007-09-20T07:52:07Z</dc:date>
    </item>
    <item>
      <title>Re: Two C++ compilers on one cluster</title>
      <link>https://community.hpe.com/t5/operating-system-openvms/two-c-compilers-on-one-cluster/m-p/4073150#M38498</link>
      <description>Piet,&lt;BR /&gt;&lt;BR /&gt;  Compilers are unprivileged usermode programs that really just process a text file. You can have as many versions of a compiler as you can fit in storage, and since they tend to be relatively small that's rarely an issue. I recommend you keep copies of ALL versions of all compilers.&lt;BR /&gt;&lt;BR /&gt;  If you happen to find a problem with a particular compiler, you can easily revert to an older compiler. Remember that it is NOT necessary to have all modules linked into a simple program compiled with the same version of compiler.&lt;BR /&gt;&lt;BR /&gt;  Some compiler kits include multiversion support. C and Pascal certainly do with SHOW_VERSION and SET_VERSION command procedures, but really all they do is flip a few logical names around. When installing a new compiler, say NO to the question about purging files replaced. You may also want to rename the compiler image to include the version number, just to make things clear.&lt;BR /&gt;&lt;BR /&gt;  Another thing you can do is pull the compiler images out of a kit into your own private directory and define logical names to use a newer version of a compiler without actually installing it. This means if you have a problem with your current compiler, you can quickly check if it's fixed in a later version without making it system wide. Again, this is non-privileged, usermode, so if you have just one "cocky" person who wants a different compiler from everyone else, you can give them the one they want.</description>
      <pubDate>Thu, 20 Sep 2007 17:12:59 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-openvms/two-c-compilers-on-one-cluster/m-p/4073150#M38498</guid>
      <dc:creator>John Gillings</dc:creator>
      <dc:date>2007-09-20T17:12:59Z</dc:date>
    </item>
    <item>
      <title>Re: Two C++ compilers on one cluster</title>
      <link>https://community.hpe.com/t5/operating-system-openvms/two-c-compilers-on-one-cluster/m-p/4073151#M38499</link>
      <description>While Linus has had some incendiary remarks about C++ of late, these folks really seem to like an extra helping of complexity by instantiating the compiler itself.  But yes, you can implement that -- with HP support -- with recent C and C++ compilers.&lt;BR /&gt;&lt;BR /&gt;An organization I'm familiar with keeps large numbers of compilers on hand and has done so for many years -- one or more sets of tools for each software release is kept on-hand -- and the requisite compiler build  is activated via its CLD and logical name when work on the release is required.  This long before formal support in the compilers arrived.&lt;BR /&gt;&lt;BR /&gt;As for rolling your own, this is an approach likely wholly unsupported by HP, but usually easy enough to figure out.  You'll often need to include compiler-provided libraries and related and ancillary pieces in the package, of course.&lt;BR /&gt;&lt;BR /&gt;</description>
      <pubDate>Thu, 20 Sep 2007 18:10:09 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-openvms/two-c-compilers-on-one-cluster/m-p/4073151#M38499</guid>
      <dc:creator>Hoff</dc:creator>
      <dc:date>2007-09-20T18:10:09Z</dc:date>
    </item>
    <item>
      <title>Re: Two C++ compilers on one cluster</title>
      <link>https://community.hpe.com/t5/operating-system-openvms/two-c-compilers-on-one-cluster/m-p/4073152#M38500</link>
      <description>Piet,&lt;BR /&gt;&lt;BR /&gt;Your profile:&lt;BR /&gt;&amp;gt;&amp;gt;&amp;gt;&lt;BR /&gt;I have assigned points to 66 of 103 answers of my questions&lt;BR /&gt;&amp;lt;&amp;lt;&amp;lt;&lt;BR /&gt;&lt;BR /&gt;Is het geen tijd om &lt;BR /&gt;&lt;A href="http://forums1.itrc.hp.com/service/forums/pageList.do?userId=BR815919&amp;amp;listType=unassigned&amp;amp;forumId=1" target="_blank"&gt;http://forums1.itrc.hp.com/service/forums/pageList.do?userId=BR815919&amp;amp;listType=unassigned&amp;amp;forumId=1&lt;/A&gt;&lt;BR /&gt;eens langs te lopen, en de "unassigned"s te verhelpen.&lt;BR /&gt;En ook 0 (nul) is een mogelijk assign...&lt;BR /&gt;&lt;BR /&gt;Groetjes.&lt;BR /&gt;&lt;BR /&gt;Proost.&lt;BR /&gt;&lt;BR /&gt;Have one on me.&lt;BR /&gt;&lt;BR /&gt;jpe</description>
      <pubDate>Fri, 21 Sep 2007 03:14:12 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-openvms/two-c-compilers-on-one-cluster/m-p/4073152#M38500</guid>
      <dc:creator>Jan van den Ende</dc:creator>
      <dc:date>2007-09-21T03:14:12Z</dc:date>
    </item>
    <item>
      <title>Re: Two C++ compilers on one cluster</title>
      <link>https://community.hpe.com/t5/operating-system-openvms/two-c-compilers-on-one-cluster/m-p/4073153#M38501</link>
      <description>Never too old to learn :)&lt;BR /&gt;&lt;BR /&gt;The point that I ran into (no C++ nor Pascal) was just seeting these logicals. If the packages contains a SET_VERSION to take care of that - that's fine. Without knowing what logicals to floip, it's more troublesome - and it took me far to much time to figure it all out.&lt;BR /&gt;&lt;BR /&gt;</description>
      <pubDate>Sat, 22 Sep 2007 15:21:56 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-openvms/two-c-compilers-on-one-cluster/m-p/4073153#M38501</guid>
      <dc:creator>Willem Grooters</dc:creator>
      <dc:date>2007-09-22T15:21:56Z</dc:date>
    </item>
  </channel>
</rss>

