1.1 --- /dev/null Thu Jan 01 00:00:00 1970 +0000
1.2 +++ b/security/nss/lib/freebl/mpi/doc/mpi-test.pod Wed Dec 31 06:09:35 2014 +0100
1.3 @@ -0,0 +1,51 @@
1.4 +# This Source Code Form is subject to the terms of the Mozilla Public
1.5 +# License, v. 2.0. If a copy of the MPL was not distributed with this
1.6 +# file, You can obtain one at http://mozilla.org/MPL/2.0/.
1.7 +
1.8 +=head1 NAME
1.9 +
1.10 + mpi-test - automated test program for MPI library
1.11 +
1.12 +=head1 SYNOPSIS
1.13 +
1.14 + mpi-test <suite-name> [quiet]
1.15 + mpi-test list
1.16 + mpi-test help
1.17 +
1.18 +=head1 DESCRIPTION
1.19 +
1.20 +The B<mpi-test> program is a general unit test driver for the MPI
1.21 +library. It is used to verify that the library works as it is
1.22 +supposed to on your architecture. As with most such things, passing
1.23 +all the tests in B<mpi-test> does not guarantee the code is correct,
1.24 +but if any of them fail, there are certainly problems.
1.25 +
1.26 +Each major function of the library can be tested individually. For a
1.27 +list of the test suites understood by B<mpi-test>, run it with the
1.28 +I<list> command line option:
1.29 +
1.30 + mpi-test list
1.31 +
1.32 +This will display a list of the available test suites and a brief
1.33 +synopsis of what each one does. For a brief overview of this
1.34 +document, run B<mpi-test> I<help>.
1.35 +
1.36 +B<mpi-test> exits with a zero status if the selected test succeeds, or
1.37 +a nonzero status if it fails. If a I<suite-name> which is not
1.38 +understood by B<mpi-test> is given, a diagnostic is printed to the
1.39 +standard error, and the program exits with a result code of 2. If a
1.40 +test fails, the result code will be 1, and a diagnostic is ordinarily
1.41 +printed to the standard error. However, if the I<quiet> option is
1.42 +provided, these diagnostics will be suppressed.
1.43 +
1.44 +=head1 RESTRICTIONS
1.45 +
1.46 +Only a few canned test cases are provided. The solutions have been
1.47 +verified using the GNU bc(1) program, so bugs there may cause problems
1.48 +here; however, this is very unlikely, so if a test fails, it is almost
1.49 +certainly my fault, not bc(1)'s.
1.50 +
1.51 +=head1 AUTHOR
1.52 +
1.53 + Michael J. Fromberger <sting@linguist.dartmouth.edu>
1.54 + Thayer School of Engineering, Hanover, New Hampshire, USA
