Li Benchmark Specifications

  1.  GENERAL
  1.1.  Classification
      Li is a CPU intensive integer benchmark  written  in  C.  The
      benchmark performs minimal I/O.
  1.2  Description
      Li is a Lisp interpreter written in C. We measure the time Li
      takes  to  solve  the  9-queens  problem [ref. Winston & Horn
      Chapter 11]. The input file to the interpreter contains  Lisp
      code that defines the 9-queens problem.
  1.3  Source/Author
      This benchmark has been developed at Sun Microsystems and  is
      based  on XLISP 1.6 written by David Michael Betz. XLISP is a
      small implementation of Lisp  with  object-oriented  program-
      ming.
  1.4  Version/Date
      Benchmark Version 1.0, Date 11/22/88.
  1.5  Other Information
      We have successfully run this benchmark on all Sun  machines,
      Micro  VAX  II,  SGI IRIS-4D/80 GT, DECstation 3100, and MIPS
      M2000.

  2.  PERFORMANCE
  2.1.  Metrics
      The metric used is the  elapsed  (real)  time  as  output  by
      /bin/time.   On  a  Sun-4/280  the  elapsed  time is about 675
      seconds.
  2.2.  Elapsed Time
      As above. The reference time (to 3 sig. fig. ) is 6210 seconds.
  2.3.  Reports
      The  output  of  /bin/time  is  directed  to  a  file  called
      "time.out".   The output of the XLISP interpreter is directed
      to "result.out" under normal circumstances.  We recommend  that
      at least 5 runs be made.
  2.4  Additional Performance Considerations
      Variations of the elapsed time between  runs  should  not  be
      more  than  5%  under  normal  circumstances.  This benchmark
      should be run on an idle machine or one in  single-user  mode
      since elapsed time is sensitive to other activities on the
      machine.








				    2


  3.  SOFTWARE
  3.1  Language
      The XLISP interpreter is written in C.
  3.2  Operating System
      The benchmark runs on a variety of UNIX-based  systems.  Some
      support  files for MS-DOS, etc. are provided by the author of
      XLISP but have not been tried out yet.
  3.3  Portability
      The benchmark has been run on both 4.2 BSD and System V based
      machines.   No OS dependencies (e.g. BSD or System V specific
      include files) are known to exist so  porting  it  should  be
      easy.
  3.4  Vectorizability/Multi-Processor Issues
      Non Vectorizable.
  3.5  Miscellaneous Software
      None needed for execution of the benchmark.
  3.6  Known Bugs
      None.
  3.7  Additional Software Considerations
      None.

  4.  HARDWARE
  4.1  Memory
      The performance of this benchmark is not memory dependent. It
      has been run on 4 MB systems.
  4.2  Disks
      No special requirements.
  4.3  Communication
      None.
  4.4  Special Hardware
      None.
  4.5  Additional Hardware Considerations
      None













				     3


  5.  OPERATIONAL
  5.1  Disk Space
      Total disk space required is < 4MB.  Size  of  executable  is
      about 150 KB.
  5.2  Installation
      Retrieve "li" directory using shar  or  tar  (or  appropriate
      utilities). No special installation procedures required.
  5.3  Execution
      Compile at the highest level of optimization  possible.   The
      script executes the following command:

	 % /bin/time xlisp li-input.lsp > result.out 2> time.out

  5.4  Correctness Verification
      Correctness validation is performed as follows:

	 % diff result.ref/result.out result/result.out

  5.5  Additional Operational Considerations
      Since the amount of program output is small (< 20 KB).
  5.6 Run
      The input file to be used is call "li-input.lsp".   In  the
      normal  mode  of  execution the program output is directed to
      "result.out". However, if the program  output is directed  to
      standard-output, the output would be as follows:

       % /bin/time xlisp li-input.lsp
      XLISP version 1.6, Copyright (c) 1985, by David Betz
      ; loading "li-input.lsp"
      ((1 1) (2 3) (3 6) (4 8) (5 2) (6 4) (7 9) (8 7) (9 5))
      ((1 1) (2 3) (3 7) (4 2) (5 8) (6 5) (7 9) (8 4) (9 6))
      ((1 9) (2 5) (3 8) (4 4) (5 1) (6 7) (7 2) (8 6) (9 3))
		----
		----
		----
      ((1 9) (2 6) (3 8) (4 2) (5 4) (6 1) (7 7) (8 5) (9 3))
      ((1 9) (2 7) (3 2) (4 4) (5 1) (6 8) (7 5) (8 3) (9 6))
      ((1 9) (2 7) (3 3) (4 8) (5 2) (6 5) (7 1) (8 6) (9 4))
      ((1 9) (2 7) (3 4) (4 2) (5 8) (6 6) (7 1) (8 3) (9 5))


      _6_7_5._4 _r_e_a_l       _6_7_0._1 _u_s_e_r         _5._3 _s_y_s