Welcome to ATALANTA <Version 2.0>!

ATALANTA is an automatic test pattern generator for stuck-at faults in combinational circuits. It employs the FAN algorithm for test pattern generation and the parallel pattern single fault propagation technique for fault simulation.

ATALANTA was developed in the Bradley Department of Electrical and Computer Engineering, Virginia Polytechnic Institute & State University (VPI&SU) and the copy right belongs to VPI&SU. The source code is released for teaching and research use only. Any publication in which ATALANTA was used to obtain the results should cite the reference given below. (Please do not say that a public domain tool was used in this research.)

This program, or any derivative thereof, may not be reproduced nor used for any commercial product without a written permission form from Prof. Dong S. Ha. For commercial use of ATALANTA or if any bugs found, please contact to

Prof. Dong S. Ha
Department of Electrical and Computer Eng.
Virginia Tech
Blacksburg, VA 24061

    Ph.: (540) 231-4942
    Fax: (540) 231-3362
    E-Mail: ha@vt.edu
    Web: http://www.ee.vt.edu/ha

REFERENCE: H. K. Lee and D. S. Ha, "On the Generation of Test Patterns for Combinational Circuits," Technical Report No. 12_93, Dep't of Electrical Eng., Virginia Polytechnic Institute and State University.

***************************** HISTORY **********************************

atalanta: version 1.0

    Original: H. K. Lee, 8/15/1990

atalanta: version 1.1

    Shuffling Compaction, Sarita Thakar 12/15/93
    Added learning, dynamic unique path sensitization, H. K. Lee, 4/30/1994
    Read the fault list file, H. K. Lee, 4/30/1994

atalanta: version 2.0           
    Removed the option -I, H. K. Lee, 6/30/1997
    Combined HOPE and atalanta, H. K. Lee, 6/30/1997
        (Added options: -H, -0, -1, -X, -R)
    Added diagnostic options, H. K. Lee, 6/30/1997
        (Added options: -A, -D, -Z, -N)

********************** Important Note **********************
*                                                          *
*  Atalanta reads circuits in ISCAS89 netlist format       *
*  rather than the original ISCAS85 netlist format.        *
*  This is done so, since ISCAS89 netlist format is more   *
*  flexible.                                               *
*                                                          *
************************************************************

----------------- Installation Procedure -----------------------------

I. To install atalanta, follow the procedures described below.

1. To install atalanta, make a bin directory under your home
   directory. Suppose that the home directory is ~cad and
   the source code of atalanta is under the directory
   ~cad/atalanta.

2. Go to the directory atalanta using
   "cd ~cad/atalanta".

3. To compile atalanta, type "make".
   An execution file "atalanta" will be created.

4. Copy or symbolically link atalanta to the directory ~cad/bin using
   "cp atalanta ~cad/bin" or
   "ln -s atalanta ~cad/bin".

II. Before using atalanta, each user should set the following two environment variables as described below.

1. Set your path (or PATH) environment variable to include
   ~cad/bin in your search path. This variable, in general,
   can be found in your ".cshrc" file. For example, the line

   set path = (... ~cad/bin ...)

   includes ~cad/bin to your search path.

2. Set the ATALANTA_MAN environment variable to indicate the
   directory atalanta, where the on-line manual of atalanta is 
   located, using the setenv statement as shown below.

   "setenv ATALANTA_MAN ~cad/atalanta"

   If you use atalanta frequently, add the above statement to your
   ".cshrc" file.

Run atalanta as described below.

----------------- User's Guide for ATALANTA --------------------------

NAME: atalanta --- Automatic Test Pattern Generator for stuck-at faults in combinational circuits

SYNOPSIS: atalanta [options] circuit_file [> outfile]

OPTIONS: Several options are available for atalanta. If an option is not specified, the default value is used.

-A Diagnostic mode Atalanta derives all test patterns for each fault. In this option, all unspecified inputs are left unknown, and fault simulation is not performed. (default: Normal test generation mode) -D n Diagnostic mode Atalanta derives n test patterns for each fault. In this option, all unspecified inputs are left unknown, and fault simulation is not performed. If both -A and -D option are specified, -D option is applied. (default: Normal test generation mode) -b n The number of maximum backtrackings for the FAN algorithm phase 1. (default: -b 10) -B n If -B n (n>0) option is specified, atalanta generates test patterns in two phases. In phase 1, static unique path sensitization is employed. If the test generation for a target fault is aborted in phase 1, the test generation is tried in phase 2. In phase 2, dynamic unique path sensitization is employed. If n=0, phase 2 is not performed. If n>0, phase 2 test generation is performed with the backtrack limit of n. (default: -B 0, i.e., phase 2 is not performed.) -f fn Faults are read from the file fn. This option is available only for ISCAS89 netlist format. (default: faults are generated internally.) -h f Displays the fault list format -h g Displays the on-line user's guide. -h n Displays an example netlist format. -h t Displays an example test pattern file. -h a Displays the entire on-line manual file. (default: no manual is displayed) -H HOPE, which is a parallel fault fault simulator, is employed for fault simulation. In this option, three logic values (0, 1 and X), instead of two logic values (0 and 1), are employed. Due to the embedding of the unknown logic value and the parallel fault fault simulation algorithm, the test generation time is slower than the default mode.) (default: FSIM, which is a parallel pattern fault simulator, is employed, and two logic values are used.) -l fn Log file is created. (default: no logfile is created) -L Static learning is performed. (default: no learning) -c n Atalanta compacts test patterns using two different methods: reverse order compaction and shuffling compaction. First, test patterns are applied in the reverse order and fault simulated (reverse order compaction). Second, test patterns are shuffled randomly and fault simulated (shuffling compaction). During the fault simulations, all the test patterns which do not detect a new fault are eliminated. The option -c n specifies the limit of shuffling compaction. If n>0, shuffling compaction is terminated if n consecutive shuffles do not drop a test pattern. If n=0, shuffling compaction is not included and compaction is done only by the reverse order fault simulation. (default: -c 2) -N Test compaction is not performed. (default: -c 2) -r n Random Pattern Testing (RPT) Session is included before deterministic test pattern generation session. The RPT session stops if any n consecutive packets of 32 random patterns do not detect any new fault. If n=0, the RPT session is not included. (default: -r 16) -s n Initial seed for the random number generator (random()). If n=0, the initial seed is the current time. (default: -s 0) -t fn Test patterns are put into the file "fn". (default: *.test for a circuit named *.bench) -u Atalanta prints out all aborted fauts in a file. The name of this file is <ckt>.ufaults. In default, all identified redundant faults are not included in the file, but you can include them using the option -v. Note that atalanta does not update a fault file if one already exists in the run directory. This fault list file can be directly read by atalanta or hope. (default: no file is created) -U fn The same as -u, but atalanta writes aborted faults to the given file name. (default: no file is created) -v Atalanta prints out all identified redundant faults as well as aborted fauts in a file. If -U fn option is not given, Atalanta prints out the faults in a default file, <ckt>.ufaults. (default: no file is created) -Z Atalanta derives one test pattern for each fault. In this option, no fault simulation is performed during the entire test generation (including random pattern test generation session, deterministic test generation session and test compaction session). All unspecified inputs are left unknown. -0, -1, -X, -R: During test generation, some inputs can be unspecified. Atalanta provides various options to set these unspecified inputs into a certain value. The options are described below. -0 Unspecified inputs are set to logic 0. -1 Unspecified inputs are set to logic 1. -R Unspecified inputs are set to logic 0 or logic 1 randomly. -X Unspecified inputs are left to unknown (X). (default: -R)

OUTPUTS: In default mode, one file is created. The summary of the test pattern generation is reported to the standard output and the test patterns are store

Atalanta

Install / Use

README