diff options
Diffstat (limited to 'security/nss/lib/freebl/mpi/doc/mpi-test.pod')
-rw-r--r-- | security/nss/lib/freebl/mpi/doc/mpi-test.pod | 51 |
1 files changed, 51 insertions, 0 deletions
diff --git a/security/nss/lib/freebl/mpi/doc/mpi-test.pod b/security/nss/lib/freebl/mpi/doc/mpi-test.pod new file mode 100644 index 000000000..b05f866e5 --- /dev/null +++ b/security/nss/lib/freebl/mpi/doc/mpi-test.pod @@ -0,0 +1,51 @@ +# This Source Code Form is subject to the terms of the Mozilla Public +# License, v. 2.0. If a copy of the MPL was not distributed with this +# file, You can obtain one at http://mozilla.org/MPL/2.0/. + +=head1 NAME + + mpi-test - automated test program for MPI library + +=head1 SYNOPSIS + + mpi-test <suite-name> [quiet] + mpi-test list + mpi-test help + +=head1 DESCRIPTION + +The B<mpi-test> program is a general unit test driver for the MPI +library. It is used to verify that the library works as it is +supposed to on your architecture. As with most such things, passing +all the tests in B<mpi-test> does not guarantee the code is correct, +but if any of them fail, there are certainly problems. + +Each major function of the library can be tested individually. For a +list of the test suites understood by B<mpi-test>, run it with the +I<list> command line option: + + mpi-test list + +This will display a list of the available test suites and a brief +synopsis of what each one does. For a brief overview of this +document, run B<mpi-test> I<help>. + +B<mpi-test> exits with a zero status if the selected test succeeds, or +a nonzero status if it fails. If a I<suite-name> which is not +understood by B<mpi-test> is given, a diagnostic is printed to the +standard error, and the program exits with a result code of 2. If a +test fails, the result code will be 1, and a diagnostic is ordinarily +printed to the standard error. However, if the I<quiet> option is +provided, these diagnostics will be suppressed. + +=head1 RESTRICTIONS + +Only a few canned test cases are provided. The solutions have been +verified using the GNU bc(1) program, so bugs there may cause problems +here; however, this is very unlikely, so if a test fails, it is almost +certainly my fault, not bc(1)'s. + +=head1 AUTHOR + + Michael J. Fromberger <sting@linguist.dartmouth.edu> + Thayer School of Engineering, Hanover, New Hampshire, USA |