<?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 Re: F90 Compiler Backend Problem in Operating System - Linux</title>
    <link>https://community.hpe.com/t5/operating-system-linux/f90-compiler-backend-problem/m-p/4182831#M91130</link>
    <description>&lt;!--!*#--&gt;It is defect in the compiler(backend). &lt;BR /&gt;-Chandra</description>
    <pubDate>Tue, 22 Apr 2008 04:19:02 GMT</pubDate>
    <dc:creator>Chandra Reddy</dc:creator>
    <dc:date>2008-04-22T04:19:02Z</dc:date>
    <item>
      <title>F90 Compiler Backend Problem</title>
      <link>https://community.hpe.com/t5/operating-system-linux/f90-compiler-backend-problem/m-p/4182829#M91128</link>
      <description>When recompiling a small block of FORTRAN 90 code that has not changed in a long time, the HP F90 compiler, v3.2.2, seems to parse the source code OK (no compile errors generated) but then generates a fatal "ASM_BACKEND" error during code generation.  Could someone explain just what the compiler is trying to tell me.  I've included the command line inputs and response below.  &lt;BR /&gt;&lt;BR /&gt;Also, is it possible to actually write source code that parses successful but fails to compile ?  Seems to me, that either the compiler front-end source code parser is ignoring a fatal source code error  or the back-end code generator is failing to assemble valid source code. Have I encountered a compiler bug ? &lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;Command line compile results:&lt;BR /&gt;&lt;BR /&gt;/opt/fortran90/bin/f90 -c initv.f &lt;BR /&gt;initv.f&lt;BR /&gt;   external subroutine INITV&lt;BR /&gt;&lt;BR /&gt;93 Lines Compiled&lt;BR /&gt;ASM_BACKEND fatal error: symbol `$.smcrt_@0000000000000044' already defined&lt;BR /&gt;f90: error 213: Errors detected.&lt;BR /&gt;&lt;BR /&gt;Compiler info:&lt;BR /&gt;&lt;BR /&gt;        HP F90 v3.2.2&lt;BR /&gt;        HP-UX f90 B.11.11.99.05 PHSS_37536 / B.11.23.37 PHSS_37537 / B.11.31.05 PHSS_37538 (Itanium(R)-based host and target) Mon Nov 26 16:39:03 2007 IST&lt;BR /&gt;&lt;BR /&gt;Machine &amp;amp; OS info:&lt;BR /&gt;&lt;BR /&gt;HP-UX sbhs905 B.11.23 U ia64 0368276839 unlimited-user license&lt;BR /&gt;model ia64 hp server rx4640&lt;BR /&gt;</description>
      <pubDate>Fri, 18 Apr 2008 16:15:49 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-linux/f90-compiler-backend-problem/m-p/4182829#M91128</guid>
      <dc:creator>david maynard_1</dc:creator>
      <dc:date>2008-04-18T16:15:49Z</dc:date>
    </item>
    <item>
      <title>Re: F90 Compiler Backend Problem</title>
      <link>https://community.hpe.com/t5/operating-system-linux/f90-compiler-backend-problem/m-p/4182830#M91129</link>
      <description>&amp;gt;Could someone explain just what the compiler is trying to tell me.&lt;BR /&gt;&lt;BR /&gt;The part of the backend that generates the object file is noticing that there are two symbols of the same name.  Typically this is a frontend bug.&lt;BR /&gt;&lt;BR /&gt;There was a recent optimizer bug like this when doing procedure splitting for nested functions but you aren't optimizing and using PBO.&lt;BR /&gt;&lt;BR /&gt;&amp;gt;is it possible to actually write source code that parses successful but fails to compile?&lt;BR /&gt;&lt;BR /&gt;Sure, that backend is awfully picky.  :-)&lt;BR /&gt;But the backend can have optimizer errors, especially since it is much more complicated than the frontend.&lt;BR /&gt;&lt;BR /&gt;&amp;gt;that either the compiler front-end source code parser is ignoring a fatal source code error&lt;BR /&gt;&lt;BR /&gt;Besides parsing, there is also semantic processing and code generation, this is where the error occur.&lt;BR /&gt;&lt;BR /&gt;&amp;gt;or the back-end code generator is failing to assemble valid source code.&lt;BR /&gt;&lt;BR /&gt;You're confused, there is no assembler here.  You're thinking of a foreign devil compiler.&lt;BR /&gt;&lt;BR /&gt;&amp;gt;Have I encountered a compiler bug?&lt;BR /&gt;&lt;BR /&gt;Most likely.  As you said, if it was your problem, the frontend should produce an error message.&lt;BR /&gt;You should contact the Response Center and report this bug.</description>
      <pubDate>Sat, 19 Apr 2008 01:03:48 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-linux/f90-compiler-backend-problem/m-p/4182830#M91129</guid>
      <dc:creator>Dennis Handly</dc:creator>
      <dc:date>2008-04-19T01:03:48Z</dc:date>
    </item>
    <item>
      <title>Re: F90 Compiler Backend Problem</title>
      <link>https://community.hpe.com/t5/operating-system-linux/f90-compiler-backend-problem/m-p/4182831#M91130</link>
      <description>&lt;!--!*#--&gt;It is defect in the compiler(backend). &lt;BR /&gt;-Chandra</description>
      <pubDate>Tue, 22 Apr 2008 04:19:02 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-linux/f90-compiler-backend-problem/m-p/4182831#M91130</guid>
      <dc:creator>Chandra Reddy</dc:creator>
      <dc:date>2008-04-22T04:19:02Z</dc:date>
    </item>
  </channel>
</rss>

