Physics 329: Fortran Programming

• Most of the source code covered in class can be found on the phy329 account on einstein in ~phy329/f77/ex1, ~phy329/f77/ex2, ... (organized by lecture).
• IMPORTANT: You can refer to the on-line (postscript) version of the SGI Fortran 77 Language Reference Manual for full details of Fortran syntax.
• Lecture 1 (Sep 16)
  • Using Fortran and C in the Unix Environment (PS).
• Lecture 2 (Sep 18)
  • fdemo1.f: Program demonstrating some essential Fortran 77 language elements (PS).
  • fdemo1_output: Output from fdemo1 (PS).
  • Makefile: Note: Beware of cutting-and-pasting text to and from Makefiles. All lines which specify an action (shell command) for updating a target must begin with a TAB character.
• Lecture 3 (Sep 21)
  • Handout (PS) and slides (PS).
  • fdemo2.f: Program demonstrating Fortran 77 usage of character variables.
  • fdemo2_output: Output from fdemo2.
  • first100: Test file used in mysum and dvfrom examples. Session trace showing, among other things, how first_100 was created.
  • mysum.f: Reads real*8 values from stdin and writes sum on stdout. Less-commented mysum_s.f and sample output.
  • dvfrom.f: Subroutine which reads real*8 vector from file (or stdin). Testing routine tdvfrom.f and sample output.
  • dvto.f: Subroutine which writes real*8 vector to file (or stdout). Testing routine tdvto.f and sample output.
  • Makefile and output from make.
• Lecture 4 (Sep 23)
  • Please review the HTML notes concerning the use of Fortran and C in the (SGI) Unix Environment.
  • Handout (PS) and slides (PS).
  • trand.f: Demonstrates use of random number generator (rand). Program output.
  • tsavedata.f: Demonstrates use of SAVE and DATA statements. Program output.
  • tsub.f: Demonstrates essentials of argument passing in Fortran. Program output.
  • texternal.f: Demonstrates EXTERNAL statement and passing subprogram names as arguments to other subprograms. Program output.
  • tcommon.f: Demonstrates use of COMMON blocks for communicating variables between program units. Program output.
  • tcommon1.f: Demonstrates RECOMMMENDED implementation and use of common blocks using an 'include' file: coma.inc.
• Lecture 5 (Sep 25)
  • Handout (PS) and slides (PS).
  • Notes on Fortran 77 arrays: handout (PS) and slides (PS).
  • arraydemo.f Demonstrates a general technique for writing Fortran programs which define and manipulate multi-dimensional arrays whose bounds are determined at run time. Program output.
  • Sample output from nth, a filter (shell-script) which selects specified columns (separated by white space) from standard output and writes them to standard output.
• Lecture 6 (Sep 28)
  • Handout (PS) and slides (PS).
  • meps.f: Computes approximate real*8 machine epsilon. SGI (IEEE) output. Cray output and Cray Makefile.
  • tfpe.f: Illustrates SGI (IEEE) real*8 floating-point exceptional values. Normal output and output with FPE trapping enabled. Program output on Cray.
  • catprec.f: Illustrates catastrophic precision loss due to subtraction of nearly-equal floating point values. Program output.
• Lecture 7 (Sep 30)
  • Handout (PS) and slides (PS).
  • dmroutines.f: Two dimensional array (matrix) routines:
    • dmmmult: Matrix-matrix multiply (square matrices only)
    • dmfrom: Read matrix from file or stdin
    • dmto: Write matrix to file or stdout
  • tdm.f: Test driver for dmroutines.f. Program output on SGIs. Program output on Cray J90. Note that this last document also contains general instructions for 'porting' a Fortran code to the Cray.
  • Makefile for the application. Note that, provided the appropriate environment variables are set, this Makefile will work on both the SGIs and the Cray J90.