Commit 5b4cb385c1 (#7382) introduced a bug which had the effect
that RAND_add()/RAND_seed() failed for buffer sizes less than
32 bytes. The reason was that now the added random data was used
exlusively as entropy source for reseeding. When the random input
was too short or contained not enough entropy, the DRBG failed
without querying the available entropy sources.
This commit makes drbg_add() act smarter: it checks the entropy
requirements explicitely. If the random input fails this check,
it won't be added as entropy input, but only as additional data.
More precisely, the behaviour depends on whether an os entropy
source was configured (which is the default on most os):
- If an os entropy source is avaible then we declare the buffer
content as additional data by setting randomness to zero and
trigger a regular reseeding.
- If no os entropy source is available, a reseeding will fail
inevitably. So drbg_add() uses a trick to mix the buffer contents
into the DRBG state without forcing a reseeding: it generates a
dummy random byte, using the buffer content as additional data.
Related-to: #7449
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/7456)
(cherry picked from commit 8817215d5c)
How to add recipes
==================
For any test that you want to perform, you write a script located in
test/recipes/, named {nn}-test_{name}.t, where {nn} is a two digit number and
{name} is a unique name of your choice.
Please note that if a test involves a new testing executable, you will need to
do some additions in test/Makefile. More on this later.
Naming conventions
=================
A test executable is named test/{name}test.c
A test recipe is named test/recipes/{nn}-test_{name}.t, where {nn} is a two
digit number and {name} is a unique name of your choice.
The number {nn} is (somewhat loosely) grouped as follows:
00-04 sanity, internal and essential API tests
05-09 individual symmetric cipher algorithms
10-14 math (bignum)
15-19 individual asymmetric cipher algorithms
20-24 openssl commands (some otherwise not tested)
25-29 certificate forms, generation and verification
30-35 engine and evp
60-79 APIs
70 PACKET layer
80-89 "larger" protocols (CA, CMS, OCSP, SSL, TSA)
90-98 misc
99 most time consuming tests [such as test_fuzz]
A recipe that just runs a test executable
=========================================
A script that just runs a program looks like this:
#! /usr/bin/perl
use OpenSSL::Test::Simple;
simple_test("test_{name}", "{name}test", "{name}");
{name} is the unique name you have chosen for your test.
The second argument to `simple_test' is the test executable, and `simple_test'
expects it to be located in test/
For documentation on OpenSSL::Test::Simple, do
`perldoc util/perl/OpenSSL/Test/Simple.pm'.
A recipe that runs a more complex test
======================================
For more complex tests, you will need to read up on Test::More and
OpenSSL::Test. Test::More is normally preinstalled, do `man Test::More' for
documentation. For OpenSSL::Test, do `perldoc util/perl/OpenSSL/Test.pm'.
A script to start from could be this:
#! /usr/bin/perl
use strict;
use warnings;
use OpenSSL::Test;
setup("test_{name}");
plan tests => 2; # The number of tests being performed
ok(test1, "test1");
ok(test2, "test1");
sub test1
{
# test feature 1
}
sub test2
{
# test feature 2
}
Changes to test/build.info
==========================
Whenever a new test involves a new test executable you need to do the
following (at all times, replace {NAME} and {name} with the name of your
test):
* add {name} to the list of programs under PROGRAMS_NO_INST
* create a three line description of how to build the test, you will have
to modify the include paths and source files if you don't want to use the
basic test framework:
SOURCE[{name}]={name}.c
INCLUDE[{name}]=.. ../include
DEPEND[{name}]=../libcrypto libtestutil.a
Generic form of C test executables
==================================
#include "testutil.h"
static int my_test(void)
{
int testresult = 0; /* Assume the test will fail */
int observed;
observed = function(); /* Call the code under test */
if (!TEST_int_equal(observed, 2)) /* Check the result is correct */
goto end; /* Exit on failure - optional */
testresult = 1; /* Mark the test case a success */
end:
cleanup(); /* Any cleanup you require */
return testresult;
}
int setup_tests(void)
{
ADD_TEST(my_test); /* Add each test separately */
return 1; /* Indicate success */
}
You should use the TEST_xxx macros provided by testutil.h to test all failure
conditions. These macros produce an error message in a standard format if the
condition is not met (and nothing if the condition is met). Additional
information can be presented with the TEST_info macro that takes a printf
format string and arguments. TEST_error is useful for complicated conditions,
it also takes a printf format string and argument. In all cases the TEST_xxx
macros are guaranteed to evaluate their arguments exactly once. This means
that expressions with side effects are allowed as parameters. Thus,
if (!TEST_ptr(ptr = OPENSSL_malloc(..)))
works fine and can be used in place of:
ptr = OPENSSL_malloc(..);
if (!TEST_ptr(ptr))
The former produces a more meaningful message on failure than the latter.