-r--r--r-- 6701 lib1305-20250407/doc/html/test.html raw
<html>
<head>
<meta http-equiv="content-type" content="text/html; charset=utf-8">
<meta name="viewport" content="width=device-width, initial-scale=1">
<style type="text/css">
html{overflow-y:scroll;background-color:#013050}
body{font-family:"Noto Sans","Droid Sans","DejaVu Sans","Arial",sans-serif;line-height:1.5}
tt,code{background-color:#f0f0f0;font-family:"Noto Sans Mono","Droid Sans Mono","DejaVu Sans Mono","Courier New",monospace,sans-serif;font-size:1em;}
pre{margin-left:3em}
p,ul,ol,blockquote,pre{font-size:1.0em;line-height:1.6}
li p{font-size:1.0em}
blockquote p{font-size:1.0em}
h1{font-size:1.5em}
h2{font-size:1.3em}
h3{font-size:1.0em}
h1 a{text-decoration:none}
table{border-collapse:collapse}
th,td{border:1px solid black}
table a{text-decoration:none}
table tr{font-size:1.0em;line-height:1.6em}
table tr{font-size:1.0em;line-height:1.5}
tbody tr:nth-child(20n+1){background-color:#f0ffff}
tbody tr:nth-child(20n+2){background-color:#f0ffff}
tbody tr:nth-child(20n+3){background-color:#f0ffff}
tbody tr:nth-child(20n+4){background-color:#f0ffff}
tbody tr:nth-child(20n+5){background-color:#f0ffff}
tbody tr:nth-child(20n+6){background-color:#f0ffff}
tbody tr:nth-child(20n+7){background-color:#f0ffff}
tbody tr:nth-child(20n+8){background-color:#f0ffff}
tbody tr:nth-child(20n+9){background-color:#f0ffff}
tbody tr:nth-child(20n+10){background-color:#f0ffff}
tbody tr:nth-child(20n+11){background-color:#fffff0}
tbody tr:nth-child(20n+12){background-color:#fffff0}
tbody tr:nth-child(20n+13){background-color:#fffff0}
tbody tr:nth-child(20n+14){background-color:#fffff0}
tbody tr:nth-child(20n+15){background-color:#fffff0}
tbody tr:nth-child(20n+16){background-color:#fffff0}
tbody tr:nth-child(20n+17){background-color:#fffff0}
tbody tr:nth-child(20n+18){background-color:#fffff0}
tbody tr:nth-child(20n+19){background-color:#fffff0}
tbody tr:nth-child(20n+20){background-color:#fffff0}
.headline{padding:0;font-weight:bold;font-size:1.0em;vertical-align:top;padding-bottom:0.5em;color:#ffffff;background-color:#013050}
.navt{display:block;box-sizing:border-box;-moz-box-sizing:border-box;-webkit-box-sizing:border-box;margin:0;padding:0;vertical-align:center;font-size:1.0em}
.here{background-color:#013050}
.here{color:#ffffff}
.away{background-color:#013050}
.away a{text-decoration:none;display:block;color:#ffffff}
.away a:hover,.away a:active{text-decoration:underline}
.main{padding:5px}
.main{background-color:#ffffff}
.pagetitle{font-size:1.4em;font-weight:bold}
@media only screen and (min-width:512px) {
.fixed{margin:0;padding:0;width:160px;height:100%;position:fixed;overflow:auto}
.main{margin-left:170px}
}
</style>
<title>
lib1305: Test</title>
</head>
<body>
<div class=fixed>
<div class=headline>
lib1305</div>
<div class="navt away"><a href=index.html>Intro</a>
</div><div class="navt away"><a href=download.html>Download</a>
</div><div class="navt away"><a href=install.html>Install</a>
</div><div class="navt here">Test
</div><div class="navt away"><a href=api.html>API</a>
</div><div class="navt away"><a href=security.html>Security</a>
</div><div class="navt away"><a href=speed.html>Speed</a>
</div><div class="navt away"><a href=internals.html>Internals</a>
</div><div class="navt away"><a href=people.html>People</a>
</div><div class="navt away"><a href=license.html>License</a>
</div></div>
<div class=main>
<div class=pagetitle>lib1305: Test</div>
<p>To run the full test suite
after compiling and installing lib1305,
run <code>lib1305-fulltest</code>.
This indicates success in two ways:
it prints <code>full tests succeeded</code> as its last line of output;
it exits 0.</p>
<p>Any change in the compiled library
(compiling for a different architecture, compiling with a different compiler, etc.) must be subjected to a new round of tests.
A compiled version of lib1305 that does not pass the full test suite is <strong>not supported</strong>.</p>
<p>One run of <code>lib1305-fulltest</code>
was observed to take 2 core-minutes on a 2.245GHz EPYC 7742 with overclocking disabled.
This test finished in 0.5 minutes of real time;
<code>lib1305-fulltest</code> includes some automatic parallelization.
To limit the number of threads used to 1,
run <code>env THREADS=1 lib1305-fulltest</code>.</p>
<p>lib1305 automatically selects
BMI2 implementations when it is running on an Intel/AMD CPU that supports BMI2,
while falling back to portable implementations otherwise.
Running <code>lib1305-fulltest</code> on an Intel/AMD CPU without BMI2
will say <code>CPU does not support implementation</code> for the BMI2 implementations
and will fail.
To test a compilation of lib1305 for Intel/AMD CPUs,
you have to pick an Intel/AMD CPU with BMI2 to run <code>lib1305-fulltest</code>.</p>
<p>The rest of this page says more about what is happening inside <code>lib1305-fulltest</code>.</p>
<h3><a name="conventional"></a>Conventional tests</h3>
<p>The workhorse inside <code>lib1305-fulltest</code>
is a separate <code>lib1305-test</code> program.</p>
<p>Simply calling <code>lib1305-test</code> without arguments
will run SUPERCOP-style tests that the subroutines in lib1305
produce the expected results for known inputs (including known randomness),
and will indicate success in two ways:
printing <code>all tests succeeded</code> as the last line of output,
and exiting 0.</p>
<p>For parallelism,
<code>lib1305-fulltest</code> calls <code>lib1305-test</code> many times,
using optional <code>lib1305-test</code> arguments to narrow which subroutines are being tested.</p>
<h3><a name="dataflow"></a>Data-flow tests</h3>
<p>Another way that <code>lib1305-fulltest</code>
runs <code>lib1305-test</code> is as follows,
running TIMECOP-style tests that branch conditions and array indices
are independent of secrets:</p>
<pre><code>env valgrind_multiplier=1 \
valgrind -q \
--error-exitcode=99 \
lib1305-test
</code></pre>
<p>This requires <code>valgrind</code> to be installed at test time.
The output will include a line <code>valgrind 1 declassify 1</code>
if the library was compiled with <code>--valgrind</code> (which is the only supported option),
or a line <code>valgrind 1 declassify 0 (expect false positives)</code> otherwise.</p>
<p>These data-flow tests
do not supersede the conventional tests.
The conventional tests run code directly on the CPU
and might catch issues hidden by the emulation in <code>valgrind</code>.
The conventional tests also include some memory tests that are disabled to improve the <code>valgrind</code> memory tests
but that are not necessarily superseded by the <code>valgrind</code> memory tests.</p><hr><font size=1><b>Version:</b>
This is version 2025.04.07 of the "Test" web page.
</font>
</div>
</body>
</html>