security/nss/lib/freebl/mpi/doc/mpi-test.pod

Wed, 31 Dec 2014 06:09:35 +0100

author
Michael Schloh von Bennewitz <michael@schloh.com>
date
Wed, 31 Dec 2014 06:09:35 +0100
changeset 0
6474c204b198
permissions
-rw-r--r--

Cloned upstream origin tor-browser at tor-browser-31.3.0esr-4.5-1-build1
revision ID fc1c9ff7c1b2defdbc039f12214767608f46423f for hacking purpose.

michael@0 1 # This Source Code Form is subject to the terms of the Mozilla Public
michael@0 2 # License, v. 2.0. If a copy of the MPL was not distributed with this
michael@0 3 # file, You can obtain one at http://mozilla.org/MPL/2.0/.
michael@0 4
michael@0 5 =head1 NAME
michael@0 6
michael@0 7 mpi-test - automated test program for MPI library
michael@0 8
michael@0 9 =head1 SYNOPSIS
michael@0 10
michael@0 11 mpi-test <suite-name> [quiet]
michael@0 12 mpi-test list
michael@0 13 mpi-test help
michael@0 14
michael@0 15 =head1 DESCRIPTION
michael@0 16
michael@0 17 The B<mpi-test> program is a general unit test driver for the MPI
michael@0 18 library. It is used to verify that the library works as it is
michael@0 19 supposed to on your architecture. As with most such things, passing
michael@0 20 all the tests in B<mpi-test> does not guarantee the code is correct,
michael@0 21 but if any of them fail, there are certainly problems.
michael@0 22
michael@0 23 Each major function of the library can be tested individually. For a
michael@0 24 list of the test suites understood by B<mpi-test>, run it with the
michael@0 25 I<list> command line option:
michael@0 26
michael@0 27 mpi-test list
michael@0 28
michael@0 29 This will display a list of the available test suites and a brief
michael@0 30 synopsis of what each one does. For a brief overview of this
michael@0 31 document, run B<mpi-test> I<help>.
michael@0 32
michael@0 33 B<mpi-test> exits with a zero status if the selected test succeeds, or
michael@0 34 a nonzero status if it fails. If a I<suite-name> which is not
michael@0 35 understood by B<mpi-test> is given, a diagnostic is printed to the
michael@0 36 standard error, and the program exits with a result code of 2. If a
michael@0 37 test fails, the result code will be 1, and a diagnostic is ordinarily
michael@0 38 printed to the standard error. However, if the I<quiet> option is
michael@0 39 provided, these diagnostics will be suppressed.
michael@0 40
michael@0 41 =head1 RESTRICTIONS
michael@0 42
michael@0 43 Only a few canned test cases are provided. The solutions have been
michael@0 44 verified using the GNU bc(1) program, so bugs there may cause problems
michael@0 45 here; however, this is very unlikely, so if a test fails, it is almost
michael@0 46 certainly my fault, not bc(1)'s.
michael@0 47
michael@0 48 =head1 AUTHOR
michael@0 49
michael@0 50 Michael J. Fromberger <sting@linguist.dartmouth.edu>
michael@0 51 Thayer School of Engineering, Hanover, New Hampshire, USA

mercurial