codemanagement.html | searchcode

/docs/codemanagement.html

https://bitbucket.org/alexei-matveev/petsc-debian-pkg
HTML | 248 lines | 208 code | 40 blank | 0 comment | 0 complexity | a37bcd4f57abd317f5fcb5732cbad99f MD5 | raw file


 <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
 <html>
  <head> <link rel="canonical" href="http://www.mcs.anl.gov/petsc/petsc-current/docs/codemanagement.html" />
    <meta http-equiv="content-type" content="text/html;charset=utf-8">
    <title>Documentation: Code Management</title>
  </head>
  <body bgcolor="#ffffff">
   <div id="version" align=right><b>petsc-3.3-p3 2012-08-29</b></div>

    <h1>Documentation: Code Management</h1>

    

    <div id="main">

      <p>
        In this file we list some of the techniques that may be used to increase
        one's efficiency when developing PETSc application codes. We have learned
        to use these techniques ourselves, and they have improved our efficiency
        tremendously.
      </p>

      <h2>Editing and Compiling</h2>

      <p>
        The biggest time sink in code development is generally the cycle of
        EDIT-COMPILE-LINK-RUN. We often see users working in a single window
        with a cycle such as:
      </p>

      <ul>
        <li>Edit a file with <code>emacs</code>.</li>
        <li>Exit <code>emacs</code>.</li>
        <li>Run <code>make</code> and see some error messages.</li>

        <li>
          Start <code>emacs</code> and try to fix the errors; often starting
          <code>emacs</code> hides the error messages so that users cannot
          remember all of them and thus do not fix all the compiler errors.
        </li>

        <li>Run <code>make</code> generating a bunch of object (.o) files.</li>

        <li>
          Link the executable (which also removes the .o files).
          Users may delete the .o files because they anticipate compiling
          the next time on a different machine that uses a different
          compiler.
        </li>

        <li>Run the executable.</li>
        <li>Detect some error condition and restart the cycle.</li>
      </ul>

      <p>
        In addition, during this process the user often moves manually
        among different directories for editing, compiling, and running.
      </p>

      <h2>Several easy ways to improve the cycle:</h2>

      <ul>
        <li>
          <code>emacs</code> has a feature to allow the user to compile using make
          and have the editor automatically jump to the line of source
          code where the compiler detects an error, even when not
          currently editing the erroneous file.
        </li>

        <li>
          The etags feature of <code>emacs</code> enables one to search quickly
          through a group of user-defined source files (and/or PETSc
          source files) regardless of the directory in which they are located.
          Etags usage tips (where <code>M-</code> denotes the "meta" key in <code>emacs</code>):
          <ul>
            <li>
              To access the predefined PETSc tags in <code>emacs</code>, type
              <code>M-x visit-tags-table</code> and specify the file
              <code>${PETSC_DIR}/TAGS</code>
            </li>
            <li>
              To move to where a PETSc function is defined, enter <code>M-.</code>
              This is useful for viewing manpages and locating source code.
            </li>
            <li>
              To search for a string and move to the first
              occurrence, use <code>M-x tags-search</code> Then, to locate later
              occurrences, type <code>M-,</code> In particular, this is useful for
              quickly locating examples of usage of certain PETSc routines.
            </li>
            <li>
              Another quite helpful command is <code>M-x tags-query-replace</code>
              which replaces occurrences of a string hroughout all tagged
              files (can retain any entries if desired); this enables quick and
              painless revisions of source code, particularly when routines are
              spread throughout many files and directories.
            </li>
          </ul>
        </li>

        <li>
          Also, <code>emacs</code> easily enables:
          <ul>
            <li>editing files that reside in any directory and retaining one's place within each of them</li>

            <li>searching for files in any directory as one attempts to load them into the editor</li>
          </ul>
        </li>
      </ul>

      <p>
        Admittedly, it takes a while to get the hang of using <code>emacs</code>.
        But we've found that the increase in our programming efficiency has
        been well worth the time spent learning to use this editor. We
        seriously believe that we could not have developed PETSc as it is today
        if we had not switched to <code>emacs</code>.
      </p>

      <p>
        <code>vi</code> and <code>Eclipse</code> also provide tools for quickly finding things
        and searching in PETSc source code, see the Emacs, VI/Vim and Eclipse
        sections of the <a href="http://www.mcs.anl.gov/petsc/petsc-dev/docs/manual.pdf">users manual</a>.
        <a href="http://gnu.org/s/global">GNU Global</a> is a richer alternative to
        etags, see the user's manual for more details.<br>
      </p>

      <h2>Debugging</h2>

      <p>
        Most code development for PETSc codes should be done on one processor;
        hence, using a standard debugger such as dbx, gdb, xdbx, etc. is fine.
        For debugging parallel runs we suggest <b>Totalview</b> if it
        is available on your machine. Otherwise, you can run each process in
        a separate debugger; this is not the same as using a parallel debugger,
        but in most cases it is not so bad. The PETSc run-time options
        <code>-start_in_debugger</code> [-display xdisplay:0] will open
        separate windows and debuggers for each process. You should debug using
        the debugging versions of the libraries (run ./configure with the
        additional option --with-debugging (the default)).
      </p>

      <p>
        It really pays to learn how to use a debugger; you will end up
        writing more interesting and far more ambitious codes once it
        is easy for you to track down problems in the codes.
      </p>

      <h2>Other suggestions</h2>

      <ul>
        <li>
          Choose consistent and obvious names for variables and functions.
          (Short variable names may be faster to type, but by using longer
          names you don't have to remember what they represent since it is
          clear from the name.)
        </li>

        <li>Use informative comments.</li>
        <li>Leave space in the code to make it readable.</li>

        <li>
          Line things up in the code for readability. Remember that any code
          written for an application will be visited <b>over and
            over</b> again, so spending an extra 20 percent of effort on
          it the first time will make each of those visits faster and more
          efficient.
        </li>

        <li>
          Realize that you <b>will</b> have to debug your code.
          <b>No one</b> writes perfect code, so always write code
          that may be debugged and learn how to use a debugger. In most cases
          using the debugger to track down problems is much faster than using
          print statements.
        </li>

        <li>
          <b>Never</b> hardwire a large problem size into your code.
          Instead, allow a command line option to run a small problem. We've
          seen people developing codes who have to wait 15 minutes after
          starting a run to reach the crashing point; this is an inefficient
          way of developing code.
        </li>

        <li>
          Develop your code on the simplest machine to which you have access.
          We have accounts on a variety of large parallel machines, but we
          write and initially test all our code on laptops or workstations
          because the user interface is friendlier, and the turn-around time
          for compiling and running is much faster than for the parallel
          machines. We use the parallel machines only for large jobs. Since
          PETSc code is completely portable, switching to a parallel machine
          from our laptop/workstation development environment simply means
          logging into another machine -- there are no code or makefile
          changes.
        </li>

        <li>
          Never develop code directly on a multi-node computing system; your
          productivity will be much lower than if you developed on
          a well-organized workstation.
        </li>

        <li>
          Keep your machines' operating systems and compilers up-to-date (or
          force your systems people to do this :-). You should always work with
          whatever tools are currently the best available. It may seem that you
          are saving time by not spending the time upgrading your system, but,
          in fact, your loss in efficiency by sticking with an outdated system
          is probably larger than then the time required to keep it up-to-date.
        </li>
      </ul>

      <h2>Fortran notes</h2>

      <p>
        Since Fortran77 does not provide type checking of routine input/output
        parameters, we find that many errors encountered within PETSc Fortran77
        programs result from accidentally using incorrect calling sequences.
        Thus, if your Fortran program is crashing, one of the first things to
        check is that all subroutines are being called with the correct arguments.
      </p>

      <p>
        When passing floating point numbers into Fortran subroutines, always
        make sure you have them marked as double precision (e.g., pass in 10.d0
        instead of 10.0). Otherwise, the compiler interprets the input as
        a single precision number, which can cause crashes or other mysterious
        problems. Make sure to declare all variables (do not use the implicit
        feature of Fortran). In fact, we <b>highly</b> recommend
        using the <b>implicit none</b> option at the begining of each
        Fortran subroutine you write.
      </p>

      <p>
        Fortran 90 provides the ability to do some type checking of subroutine
        calls by using Fortran 90 interfaces or modules. PETSc now provides
        interfaces and modules for Fortran 90; see the
        <a href="manualpages/Sys/UsingFortran.html">manual page</a>
        for how to use them.
      </p>

    </div>

    
  </body>
</html>