tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
/*
|
|
|
|
|
* pkd_hello.c --
|
|
|
|
|
*
|
2017-08-17 10:18:20 +03:00
|
|
|
* (c) 2014, 2017 Jon Simons <jon@jonsimons.org>
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
*/
|
2017-10-28 15:31:37 +03:00
|
|
|
#include "config.h"
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
|
|
|
|
|
#include <setjmp.h> // for cmocka
|
|
|
|
|
#include <stdarg.h> // for cmocka
|
|
|
|
|
#include <stdio.h>
|
|
|
|
|
#include <unistd.h> // for cmocka
|
|
|
|
|
#include <cmocka.h>
|
|
|
|
|
|
|
|
|
|
#include "libssh/priv.h"
|
|
|
|
|
|
|
|
|
|
#include "pkd_client.h"
|
|
|
|
|
#include "pkd_daemon.h"
|
|
|
|
|
#include "pkd_keyutil.h"
|
|
|
|
|
#include "pkd_util.h"
|
|
|
|
|
|
|
|
|
|
#define DEFAULT_ITERATIONS 10
|
|
|
|
|
static struct pkd_daemon_args pkd_dargs;
|
|
|
|
|
|
|
|
|
|
#ifdef HAVE_ARGP_H
|
|
|
|
|
#include <argp.h>
|
|
|
|
|
#define PROGNAME "pkd_hello"
|
|
|
|
|
#define ARGP_PROGNAME "libssh " PROGNAME
|
2017-08-17 10:18:20 +03:00
|
|
|
const char *argp_program_version = ARGP_PROGNAME " 2017-07-12";
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
const char *argp_program_bug_address = "Jon Simons <jon@jonsimons.org>";
|
2017-08-17 10:18:20 +03:00
|
|
|
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
static char doc[] = \
|
|
|
|
|
"\nExample usage:\n\n"
|
|
|
|
|
" " PROGNAME "\n"
|
|
|
|
|
" Run all tests with default number of iterations.\n"
|
|
|
|
|
" " PROGNAME " --list\n"
|
|
|
|
|
" List available individual test names.\n"
|
|
|
|
|
" " PROGNAME " -i 1000 -t torture_pkd_rsa_ecdh_sha2_nistp256\n"
|
|
|
|
|
" Run only the torture_pkd_rsa_ecdh_sha2_nistp256 testcase 1000 times.\n"
|
2017-08-17 10:18:20 +03:00
|
|
|
" " PROGNAME " -i 1000 -m curve25519\n"
|
|
|
|
|
" Run all tests with the string 'curve25519' 1000 times.\n"
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
" " PROGNAME " -v -v -v -v -e -o\n"
|
|
|
|
|
" Run all tests with maximum libssh and pkd logging.\n"
|
|
|
|
|
;
|
|
|
|
|
|
|
|
|
|
static struct argp_option options[] = {
|
|
|
|
|
{ "stderr", 'e', NULL, 0,
|
|
|
|
|
"Emit pkd stderr messages", 0 },
|
|
|
|
|
{ "list", 'l', NULL, 0,
|
|
|
|
|
"List available individual test names", 0 },
|
|
|
|
|
{ "iterations", 'i', "number", 0,
|
|
|
|
|
"Run each test for the given number of iterations (default is 10)", 0 },
|
2017-08-17 10:18:20 +03:00
|
|
|
{ "match", 'm', "testmatch", 0,
|
|
|
|
|
"Run all tests with the given string", 0 },
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
{ "stdout", 'o', NULL, 0,
|
|
|
|
|
"Emit pkd stdout messages", 0 },
|
|
|
|
|
{ "test", 't', "testname", 0,
|
|
|
|
|
"Run tests matching the given testname", 0 },
|
|
|
|
|
{ "verbose", 'v', NULL, 0,
|
|
|
|
|
"Increase libssh verbosity (can be used multiple times)", 0 },
|
|
|
|
|
{ NULL, 0, NULL, 0,
|
|
|
|
|
NULL, 0 },
|
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
static error_t parse_opt(int key, char *arg, struct argp_state *state) {
|
|
|
|
|
(void) arg;
|
|
|
|
|
(void) state;
|
|
|
|
|
|
|
|
|
|
switch(key) {
|
|
|
|
|
case 'e':
|
|
|
|
|
pkd_dargs.opts.log_stderr = 1;
|
|
|
|
|
break;
|
|
|
|
|
case 'l':
|
|
|
|
|
pkd_dargs.opts.list = 1;
|
|
|
|
|
break;
|
|
|
|
|
case 'i':
|
|
|
|
|
pkd_dargs.opts.iterations = atoi(arg);
|
|
|
|
|
break;
|
2017-08-17 10:18:20 +03:00
|
|
|
case 'm':
|
|
|
|
|
pkd_dargs.opts.testmatch = arg;
|
|
|
|
|
break;
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
case 'o':
|
|
|
|
|
pkd_dargs.opts.log_stdout = 1;
|
|
|
|
|
break;
|
|
|
|
|
case 't':
|
|
|
|
|
pkd_dargs.opts.testname = arg;
|
|
|
|
|
break;
|
|
|
|
|
case 'v':
|
|
|
|
|
pkd_dargs.opts.libssh_log_level += 1;
|
|
|
|
|
break;
|
|
|
|
|
default:
|
|
|
|
|
return ARGP_ERR_UNKNOWN;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static struct argp parser = {
|
|
|
|
|
options,
|
|
|
|
|
parse_opt,
|
|
|
|
|
NULL,
|
|
|
|
|
doc,
|
|
|
|
|
NULL,
|
|
|
|
|
NULL,
|
|
|
|
|
NULL
|
|
|
|
|
};
|
|
|
|
|
#endif /* HAVE_ARGP_H */
|
|
|
|
|
|
|
|
|
|
static struct pkd_state *torture_pkd_setup(enum pkd_hostkey_type_e type,
|
|
|
|
|
const char *hostkeypath) {
|
|
|
|
|
int rc = 0;
|
|
|
|
|
|
|
|
|
|
pkd_dargs.type = type;
|
|
|
|
|
pkd_dargs.hostkeypath = hostkeypath;
|
|
|
|
|
|
|
|
|
|
rc = pkd_start(&pkd_dargs);
|
|
|
|
|
assert_int_equal(rc, 0);
|
|
|
|
|
|
|
|
|
|
return NULL;
|
|
|
|
|
}
|
|
|
|
|
|
2015-09-07 13:44:23 +03:00
|
|
|
static int torture_pkd_teardown(void **state) {
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
struct pkd_result result = { .ok = 0 };
|
|
|
|
|
|
|
|
|
|
(void) state;
|
|
|
|
|
|
|
|
|
|
pkd_stop(&result);
|
|
|
|
|
assert_int_equal(result.ok, 1);
|
2015-09-07 13:44:23 +03:00
|
|
|
|
|
|
|
|
return 0;
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* one setup for each server keytype ------------------------------------
|
|
|
|
|
*/
|
|
|
|
|
|
2015-09-07 13:44:23 +03:00
|
|
|
static int torture_pkd_setup_noop(void **state) {
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
*state = (void *) torture_pkd_setup(PKD_RSA, NULL /*path*/);
|
2015-09-07 13:44:23 +03:00
|
|
|
|
|
|
|
|
return 0;
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
}
|
|
|
|
|
|
2015-09-07 13:44:23 +03:00
|
|
|
static int torture_pkd_setup_rsa(void **state) {
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
setup_rsa_key();
|
|
|
|
|
*state = (void *) torture_pkd_setup(PKD_RSA, LIBSSH_RSA_TESTKEY);
|
2015-09-07 13:44:23 +03:00
|
|
|
|
|
|
|
|
return 0;
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
}
|
|
|
|
|
|
2017-12-28 13:10:43 +03:00
|
|
|
#ifdef HAVE_DSA
|
2015-09-07 13:44:23 +03:00
|
|
|
static int torture_pkd_setup_dsa(void **state) {
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
setup_dsa_key();
|
|
|
|
|
*state = (void *) torture_pkd_setup(PKD_DSA, LIBSSH_DSA_TESTKEY);
|
2015-09-07 13:44:23 +03:00
|
|
|
|
|
|
|
|
return 0;
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
}
|
2017-12-28 13:10:43 +03:00
|
|
|
#endif
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
|
2015-09-07 13:44:23 +03:00
|
|
|
static int torture_pkd_setup_ecdsa_256(void **state) {
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
setup_ecdsa_keys();
|
|
|
|
|
*state = (void *) torture_pkd_setup(PKD_ECDSA, LIBSSH_ECDSA_256_TESTKEY);
|
2015-09-07 13:44:23 +03:00
|
|
|
|
|
|
|
|
return 0;
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
}
|
|
|
|
|
|
2015-09-07 13:44:23 +03:00
|
|
|
static int torture_pkd_setup_ecdsa_384(void **state) {
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
setup_ecdsa_keys();
|
|
|
|
|
*state = (void *) torture_pkd_setup(PKD_ECDSA, LIBSSH_ECDSA_384_TESTKEY);
|
2015-09-07 13:44:23 +03:00
|
|
|
|
|
|
|
|
return 0;
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
}
|
|
|
|
|
|
2015-09-07 13:44:23 +03:00
|
|
|
static int torture_pkd_setup_ecdsa_521(void **state) {
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
setup_ecdsa_keys();
|
|
|
|
|
*state = (void *) torture_pkd_setup(PKD_ECDSA, LIBSSH_ECDSA_521_TESTKEY);
|
2015-09-07 13:44:23 +03:00
|
|
|
|
|
|
|
|
return 0;
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Test matrices: f(clientname, testname, ssh-command, setup-function, teardown-function).
|
|
|
|
|
*/
|
|
|
|
|
|
2017-12-28 13:10:43 +03:00
|
|
|
#ifdef HAVE_DSA
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
#define PKDTESTS_DEFAULT(f, client, cmd) \
|
|
|
|
|
/* Default passes by server key type. */ \
|
|
|
|
|
f(client, rsa_default, cmd, setup_rsa, teardown) \
|
|
|
|
|
f(client, dsa_default, cmd, setup_dsa, teardown) \
|
|
|
|
|
f(client, ecdsa_256_default, cmd, setup_ecdsa_256, teardown) \
|
|
|
|
|
f(client, ecdsa_384_default, cmd, setup_ecdsa_384, teardown) \
|
|
|
|
|
f(client, ecdsa_521_default, cmd, setup_ecdsa_521, teardown)
|
2017-12-28 13:10:43 +03:00
|
|
|
#else
|
|
|
|
|
#define PKDTESTS_DEFAULT(f, client, cmd) \
|
|
|
|
|
/* Default passes by server key type. */ \
|
|
|
|
|
f(client, rsa_default, cmd, setup_rsa, teardown) \
|
|
|
|
|
f(client, ecdsa_256_default, cmd, setup_ecdsa_256, teardown) \
|
|
|
|
|
f(client, ecdsa_384_default, cmd, setup_ecdsa_384, teardown) \
|
|
|
|
|
f(client, ecdsa_521_default, cmd, setup_ecdsa_521, teardown)
|
|
|
|
|
#endif
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
|
2017-12-28 13:10:43 +03:00
|
|
|
#ifdef HAVE_DSA
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
#define PKDTESTS_KEX(f, client, kexcmd) \
|
|
|
|
|
/* Kex algorithms. */ \
|
|
|
|
|
f(client, rsa_curve25519_sha256, kexcmd("curve25519-sha256@libssh.org"), setup_rsa, teardown) \
|
|
|
|
|
f(client, rsa_ecdh_sha2_nistp256, kexcmd("ecdh-sha2-nistp256 "), setup_rsa, teardown) \
|
2017-08-24 19:14:38 +03:00
|
|
|
f(client, rsa_ecdh_sha2_nistp384, kexcmd("ecdh-sha2-nistp384 "), setup_rsa, teardown) \
|
|
|
|
|
f(client, rsa_ecdh_sha2_nistp521, kexcmd("ecdh-sha2-nistp521 "), setup_rsa, teardown) \
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
f(client, rsa_diffie_hellman_group14_sha1, kexcmd("diffie-hellman-group14-sha1"), setup_rsa, teardown) \
|
|
|
|
|
f(client, rsa_diffie_hellman_group1_sha1, kexcmd("diffie-hellman-group1-sha1"), setup_rsa, teardown) \
|
|
|
|
|
f(client, dsa_curve25519_sha256, kexcmd("curve25519-sha256@libssh.org"), setup_dsa, teardown) \
|
|
|
|
|
f(client, dsa_ecdh_sha2_nistp256, kexcmd("ecdh-sha2-nistp256 "), setup_dsa, teardown) \
|
2017-08-24 19:14:38 +03:00
|
|
|
f(client, dsa_ecdh_sha2_nistp384, kexcmd("ecdh-sha2-nistp384 "), setup_dsa, teardown) \
|
|
|
|
|
f(client, dsa_ecdh_sha2_nistp521, kexcmd("ecdh-sha2-nistp521 "), setup_dsa, teardown) \
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
f(client, dsa_diffie_hellman_group14_sha1, kexcmd("diffie-hellman-group14-sha1"), setup_dsa, teardown) \
|
|
|
|
|
f(client, dsa_diffie_hellman_group1_sha1, kexcmd("diffie-hellman-group1-sha1"), setup_dsa, teardown) \
|
|
|
|
|
f(client, ecdsa_256_curve25519_sha256, kexcmd("curve25519-sha256@libssh.org"), setup_ecdsa_256, teardown) \
|
|
|
|
|
f(client, ecdsa_256_ecdh_sha2_nistp256, kexcmd("ecdh-sha2-nistp256 "), setup_ecdsa_256, teardown) \
|
2017-08-24 19:14:38 +03:00
|
|
|
f(client, ecdsa_256_ecdh_sha2_nistp384, kexcmd("ecdh-sha2-nistp384 "), setup_ecdsa_256, teardown) \
|
|
|
|
|
f(client, ecdsa_256_ecdh_sha2_nistp521, kexcmd("ecdh-sha2-nistp521 "), setup_ecdsa_256, teardown) \
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
f(client, ecdsa_256_diffie_hellman_group14_sha1, kexcmd("diffie-hellman-group14-sha1"), setup_ecdsa_256, teardown) \
|
|
|
|
|
f(client, ecdsa_256_diffie_hellman_group1_sha1, kexcmd("diffie-hellman-group1-sha1"), setup_ecdsa_256, teardown) \
|
|
|
|
|
f(client, ecdsa_384_curve25519_sha256, kexcmd("curve25519-sha256@libssh.org"), setup_ecdsa_384, teardown) \
|
|
|
|
|
f(client, ecdsa_384_ecdh_sha2_nistp256, kexcmd("ecdh-sha2-nistp256 "), setup_ecdsa_384, teardown) \
|
2017-08-24 19:14:38 +03:00
|
|
|
f(client, ecdsa_384_ecdh_sha2_nistp384, kexcmd("ecdh-sha2-nistp384 "), setup_ecdsa_384, teardown) \
|
|
|
|
|
f(client, ecdsa_384_ecdh_sha2_nistp521, kexcmd("ecdh-sha2-nistp521 "), setup_ecdsa_384, teardown) \
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
f(client, ecdsa_384_diffie_hellman_group14_sha1, kexcmd("diffie-hellman-group14-sha1"), setup_ecdsa_384, teardown) \
|
|
|
|
|
f(client, ecdsa_384_diffie_hellman_group1_sha1, kexcmd("diffie-hellman-group1-sha1"), setup_ecdsa_384, teardown) \
|
|
|
|
|
f(client, ecdsa_521_curve25519_sha256, kexcmd("curve25519-sha256@libssh.org"), setup_ecdsa_521, teardown) \
|
|
|
|
|
f(client, ecdsa_521_ecdh_sha2_nistp256, kexcmd("ecdh-sha2-nistp256 "), setup_ecdsa_521, teardown) \
|
2017-08-24 19:14:38 +03:00
|
|
|
f(client, ecdsa_521_ecdh_sha2_nistp384, kexcmd("ecdh-sha2-nistp384 "), setup_ecdsa_521, teardown) \
|
|
|
|
|
f(client, ecdsa_521_ecdh_sha2_nistp521, kexcmd("ecdh-sha2-nistp521 "), setup_ecdsa_521, teardown) \
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
f(client, ecdsa_521_diffie_hellman_group14_sha1, kexcmd("diffie-hellman-group14-sha1"), setup_ecdsa_521, teardown) \
|
|
|
|
|
f(client, ecdsa_521_diffie_hellman_group1_sha1, kexcmd("diffie-hellman-group1-sha1"), setup_ecdsa_521, teardown)
|
2017-12-28 13:10:43 +03:00
|
|
|
#else
|
|
|
|
|
#define PKDTESTS_KEX(f, client, kexcmd) \
|
|
|
|
|
/* Kex algorithms. */ \
|
|
|
|
|
f(client, rsa_curve25519_sha256, kexcmd("curve25519-sha256@libssh.org"), setup_rsa, teardown) \
|
|
|
|
|
f(client, rsa_ecdh_sha2_nistp256, kexcmd("ecdh-sha2-nistp256 "), setup_rsa, teardown) \
|
|
|
|
|
f(client, rsa_diffie_hellman_group14_sha1, kexcmd("diffie-hellman-group14-sha1"), setup_rsa, teardown) \
|
|
|
|
|
f(client, rsa_diffie_hellman_group1_sha1, kexcmd("diffie-hellman-group1-sha1"), setup_rsa, teardown) \
|
|
|
|
|
f(client, ecdsa_256_curve25519_sha256, kexcmd("curve25519-sha256@libssh.org"), setup_ecdsa_256, teardown) \
|
|
|
|
|
f(client, ecdsa_256_ecdh_sha2_nistp256, kexcmd("ecdh-sha2-nistp256 "), setup_ecdsa_256, teardown) \
|
|
|
|
|
f(client, ecdsa_256_diffie_hellman_group14_sha1, kexcmd("diffie-hellman-group14-sha1"), setup_ecdsa_256, teardown) \
|
|
|
|
|
f(client, ecdsa_256_diffie_hellman_group1_sha1, kexcmd("diffie-hellman-group1-sha1"), setup_ecdsa_256, teardown) \
|
|
|
|
|
f(client, ecdsa_384_curve25519_sha256, kexcmd("curve25519-sha256@libssh.org"), setup_ecdsa_384, teardown) \
|
|
|
|
|
f(client, ecdsa_384_ecdh_sha2_nistp256, kexcmd("ecdh-sha2-nistp256 "), setup_ecdsa_384, teardown) \
|
|
|
|
|
f(client, ecdsa_384_diffie_hellman_group14_sha1, kexcmd("diffie-hellman-group14-sha1"), setup_ecdsa_384, teardown) \
|
|
|
|
|
f(client, ecdsa_384_diffie_hellman_group1_sha1, kexcmd("diffie-hellman-group1-sha1"), setup_ecdsa_384, teardown) \
|
|
|
|
|
f(client, ecdsa_521_curve25519_sha256, kexcmd("curve25519-sha256@libssh.org"), setup_ecdsa_521, teardown) \
|
|
|
|
|
f(client, ecdsa_521_ecdh_sha2_nistp256, kexcmd("ecdh-sha2-nistp256 "), setup_ecdsa_521, teardown) \
|
|
|
|
|
f(client, ecdsa_521_diffie_hellman_group14_sha1, kexcmd("diffie-hellman-group14-sha1"), setup_ecdsa_521, teardown) \
|
|
|
|
|
f(client, ecdsa_521_diffie_hellman_group1_sha1, kexcmd("diffie-hellman-group1-sha1"), setup_ecdsa_521, teardown)
|
|
|
|
|
#endif
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
|
2017-12-28 13:10:43 +03:00
|
|
|
#ifdef HAVE_DSA
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
#define PKDTESTS_CIPHER(f, client, ciphercmd) \
|
|
|
|
|
/* Ciphers. */ \
|
|
|
|
|
f(client, rsa_3des_cbc, ciphercmd("3des-cbc"), setup_rsa, teardown) \
|
|
|
|
|
f(client, rsa_aes128_cbc, ciphercmd("aes128-cbc"), setup_rsa, teardown) \
|
|
|
|
|
f(client, rsa_aes128_ctr, ciphercmd("aes128-ctr"), setup_rsa, teardown) \
|
|
|
|
|
f(client, rsa_aes256_cbc, ciphercmd("aes256-cbc"), setup_rsa, teardown) \
|
|
|
|
|
f(client, rsa_aes256_ctr, ciphercmd("aes256-ctr"), setup_rsa, teardown) \
|
|
|
|
|
f(client, rsa_blowfish_cbc, ciphercmd("blowfish-cbc"), setup_rsa, teardown) \
|
|
|
|
|
f(client, dsa_3des_cbc, ciphercmd("3des-cbc"), setup_dsa, teardown) \
|
|
|
|
|
f(client, dsa_aes128_cbc, ciphercmd("aes128-cbc"), setup_dsa, teardown) \
|
|
|
|
|
f(client, dsa_aes128_ctr, ciphercmd("aes128-ctr"), setup_dsa, teardown) \
|
|
|
|
|
f(client, dsa_aes256_cbc, ciphercmd("aes256-cbc"), setup_dsa, teardown) \
|
|
|
|
|
f(client, dsa_aes256_ctr, ciphercmd("aes256-ctr"), setup_dsa, teardown) \
|
|
|
|
|
f(client, dsa_blowfish_cbc, ciphercmd("blowfish-cbc"), setup_dsa, teardown) \
|
|
|
|
|
f(client, ecdsa_256_3des_cbc, ciphercmd("3des-cbc"), setup_ecdsa_256, teardown) \
|
|
|
|
|
f(client, ecdsa_256_aes128_cbc, ciphercmd("aes128-cbc"), setup_ecdsa_256, teardown) \
|
|
|
|
|
f(client, ecdsa_256_aes128_ctr, ciphercmd("aes128-ctr"), setup_ecdsa_256, teardown) \
|
|
|
|
|
f(client, ecdsa_256_aes256_cbc, ciphercmd("aes256-cbc"), setup_ecdsa_256, teardown) \
|
|
|
|
|
f(client, ecdsa_256_aes256_ctr, ciphercmd("aes256-ctr"), setup_ecdsa_256, teardown) \
|
|
|
|
|
f(client, ecdsa_256_blowfish_cbc, ciphercmd("blowfish-cbc"), setup_ecdsa_256, teardown) \
|
|
|
|
|
f(client, ecdsa_384_3des_cbc, ciphercmd("3des-cbc"), setup_ecdsa_384, teardown) \
|
|
|
|
|
f(client, ecdsa_384_aes128_cbc, ciphercmd("aes128-cbc"), setup_ecdsa_384, teardown) \
|
|
|
|
|
f(client, ecdsa_384_aes128_ctr, ciphercmd("aes128-ctr"), setup_ecdsa_384, teardown) \
|
|
|
|
|
f(client, ecdsa_384_aes256_cbc, ciphercmd("aes256-cbc"), setup_ecdsa_384, teardown) \
|
|
|
|
|
f(client, ecdsa_384_aes256_ctr, ciphercmd("aes256-ctr"), setup_ecdsa_384, teardown) \
|
|
|
|
|
f(client, ecdsa_384_blowfish_cbc, ciphercmd("blowfish-cbc"), setup_ecdsa_384, teardown) \
|
|
|
|
|
f(client, ecdsa_521_3des_cbc, ciphercmd("3des-cbc"), setup_ecdsa_521, teardown) \
|
|
|
|
|
f(client, ecdsa_521_aes128_cbc, ciphercmd("aes128-cbc"), setup_ecdsa_521, teardown) \
|
|
|
|
|
f(client, ecdsa_521_aes128_ctr, ciphercmd("aes128-ctr"), setup_ecdsa_521, teardown) \
|
|
|
|
|
f(client, ecdsa_521_aes256_cbc, ciphercmd("aes256-cbc"), setup_ecdsa_521, teardown) \
|
|
|
|
|
f(client, ecdsa_521_aes256_ctr, ciphercmd("aes256-ctr"), setup_ecdsa_521, teardown) \
|
|
|
|
|
f(client, ecdsa_521_blowfish_cbc, ciphercmd("blowfish-cbc"), setup_ecdsa_521, teardown)
|
2017-12-28 13:10:43 +03:00
|
|
|
#else
|
|
|
|
|
#define PKDTESTS_CIPHER(f, client, ciphercmd) \
|
|
|
|
|
/* Ciphers. */ \
|
|
|
|
|
f(client, rsa_3des_cbc, ciphercmd("3des-cbc"), setup_rsa, teardown) \
|
|
|
|
|
f(client, rsa_aes128_cbc, ciphercmd("aes128-cbc"), setup_rsa, teardown) \
|
|
|
|
|
f(client, rsa_aes128_ctr, ciphercmd("aes128-ctr"), setup_rsa, teardown) \
|
|
|
|
|
f(client, rsa_aes256_cbc, ciphercmd("aes256-cbc"), setup_rsa, teardown) \
|
|
|
|
|
f(client, rsa_aes256_ctr, ciphercmd("aes256-ctr"), setup_rsa, teardown) \
|
|
|
|
|
f(client, rsa_blowfish_cbc, ciphercmd("blowfish-cbc"), setup_rsa, teardown) \
|
|
|
|
|
f(client, ecdsa_256_3des_cbc, ciphercmd("3des-cbc"), setup_ecdsa_256, teardown) \
|
|
|
|
|
f(client, ecdsa_256_aes128_cbc, ciphercmd("aes128-cbc"), setup_ecdsa_256, teardown) \
|
|
|
|
|
f(client, ecdsa_256_aes128_ctr, ciphercmd("aes128-ctr"), setup_ecdsa_256, teardown) \
|
|
|
|
|
f(client, ecdsa_256_aes256_cbc, ciphercmd("aes256-cbc"), setup_ecdsa_256, teardown) \
|
|
|
|
|
f(client, ecdsa_256_aes256_ctr, ciphercmd("aes256-ctr"), setup_ecdsa_256, teardown) \
|
|
|
|
|
f(client, ecdsa_256_blowfish_cbc, ciphercmd("blowfish-cbc"), setup_ecdsa_256, teardown) \
|
|
|
|
|
f(client, ecdsa_384_3des_cbc, ciphercmd("3des-cbc"), setup_ecdsa_384, teardown) \
|
|
|
|
|
f(client, ecdsa_384_aes128_cbc, ciphercmd("aes128-cbc"), setup_ecdsa_384, teardown) \
|
|
|
|
|
f(client, ecdsa_384_aes128_ctr, ciphercmd("aes128-ctr"), setup_ecdsa_384, teardown) \
|
|
|
|
|
f(client, ecdsa_384_aes256_cbc, ciphercmd("aes256-cbc"), setup_ecdsa_384, teardown) \
|
|
|
|
|
f(client, ecdsa_384_aes256_ctr, ciphercmd("aes256-ctr"), setup_ecdsa_384, teardown) \
|
|
|
|
|
f(client, ecdsa_384_blowfish_cbc, ciphercmd("blowfish-cbc"), setup_ecdsa_384, teardown) \
|
|
|
|
|
f(client, ecdsa_521_3des_cbc, ciphercmd("3des-cbc"), setup_ecdsa_521, teardown) \
|
|
|
|
|
f(client, ecdsa_521_aes128_cbc, ciphercmd("aes128-cbc"), setup_ecdsa_521, teardown) \
|
|
|
|
|
f(client, ecdsa_521_aes128_ctr, ciphercmd("aes128-ctr"), setup_ecdsa_521, teardown) \
|
|
|
|
|
f(client, ecdsa_521_aes256_cbc, ciphercmd("aes256-cbc"), setup_ecdsa_521, teardown) \
|
|
|
|
|
f(client, ecdsa_521_aes256_ctr, ciphercmd("aes256-ctr"), setup_ecdsa_521, teardown) \
|
|
|
|
|
f(client, ecdsa_521_blowfish_cbc, ciphercmd("blowfish-cbc"), setup_ecdsa_521, teardown)
|
|
|
|
|
#endif
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
|
2017-12-28 13:10:43 +03:00
|
|
|
#ifdef HAVE_DSA
|
2017-08-17 10:18:20 +03:00
|
|
|
#define PKDTESTS_CIPHER_OPENSSHONLY(f, client, ciphercmd) \
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
/* Ciphers. */ \
|
|
|
|
|
f(client, rsa_aes192_cbc, ciphercmd("aes192-cbc"), setup_rsa, teardown) \
|
|
|
|
|
f(client, rsa_aes192_ctr, ciphercmd("aes192-ctr"), setup_rsa, teardown) \
|
|
|
|
|
f(client, dsa_aes192_cbc, ciphercmd("aes192-cbc"), setup_dsa, teardown) \
|
|
|
|
|
f(client, dsa_aes192_ctr, ciphercmd("aes192-ctr"), setup_dsa, teardown) \
|
|
|
|
|
f(client, ecdsa_256_aes192_cbc, ciphercmd("aes192-cbc"), setup_ecdsa_256, teardown) \
|
|
|
|
|
f(client, ecdsa_256_aes192_ctr, ciphercmd("aes192-ctr"), setup_ecdsa_256, teardown) \
|
|
|
|
|
f(client, ecdsa_384_aes192_cbc, ciphercmd("aes192-cbc"), setup_ecdsa_384, teardown) \
|
|
|
|
|
f(client, ecdsa_384_aes192_ctr, ciphercmd("aes192-ctr"), setup_ecdsa_384, teardown) \
|
|
|
|
|
f(client, ecdsa_521_aes192_cbc, ciphercmd("aes192-cbc"), setup_ecdsa_521, teardown) \
|
|
|
|
|
f(client, ecdsa_521_aes192_ctr, ciphercmd("aes192-ctr"), setup_ecdsa_521, teardown)
|
2017-12-28 13:10:43 +03:00
|
|
|
#else
|
|
|
|
|
#define PKDTESTS_CIPHER_OPENSSHONLY(f, client, ciphercmd) \
|
|
|
|
|
/* Ciphers. */ \
|
|
|
|
|
f(client, rsa_aes192_cbc, ciphercmd("aes192-cbc"), setup_rsa, teardown) \
|
|
|
|
|
f(client, rsa_aes192_ctr, ciphercmd("aes192-ctr"), setup_rsa, teardown) \
|
|
|
|
|
f(client, ecdsa_256_aes192_cbc, ciphercmd("aes192-cbc"), setup_ecdsa_256, teardown) \
|
|
|
|
|
f(client, ecdsa_256_aes192_ctr, ciphercmd("aes192-ctr"), setup_ecdsa_256, teardown) \
|
|
|
|
|
f(client, ecdsa_384_aes192_cbc, ciphercmd("aes192-cbc"), setup_ecdsa_384, teardown) \
|
|
|
|
|
f(client, ecdsa_384_aes192_ctr, ciphercmd("aes192-ctr"), setup_ecdsa_384, teardown) \
|
|
|
|
|
f(client, ecdsa_521_aes192_cbc, ciphercmd("aes192-cbc"), setup_ecdsa_521, teardown) \
|
|
|
|
|
f(client, ecdsa_521_aes192_ctr, ciphercmd("aes192-ctr"), setup_ecdsa_521, teardown)
|
|
|
|
|
#endif
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
|
2017-12-28 13:10:43 +03:00
|
|
|
#ifdef HAVE_DSA
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
#define PKDTESTS_MAC(f, client, maccmd) \
|
|
|
|
|
/* MACs. */ \
|
|
|
|
|
f(client, rsa_hmac_sha1, maccmd("hmac-sha1"), setup_rsa, teardown) \
|
|
|
|
|
f(client, dsa_hmac_sha1, maccmd("hmac-sha1"), setup_dsa, teardown) \
|
|
|
|
|
f(client, ecdsa_256_hmac_sha1, maccmd("hmac-sha1"), setup_ecdsa_256, teardown) \
|
|
|
|
|
f(client, ecdsa_384_hmac_sha1, maccmd("hmac-sha1"), setup_ecdsa_384, teardown) \
|
|
|
|
|
f(client, ecdsa_521_hmac_sha1, maccmd("hmac-sha1"), setup_ecdsa_521, teardown) \
|
|
|
|
|
f(client, rsa_hmac_sha2_256, maccmd("hmac-sha2-256"), setup_rsa, teardown) \
|
|
|
|
|
f(client, dsa_hmac_sha2_256, maccmd("hmac-sha2-256"), setup_dsa, teardown) \
|
|
|
|
|
f(client, ecdsa_256_hmac_sha2_256, maccmd("hmac-sha2-256"), setup_ecdsa_256, teardown) \
|
|
|
|
|
f(client, ecdsa_384_hmac_sha2_256, maccmd("hmac-sha2-256"), setup_ecdsa_384, teardown) \
|
|
|
|
|
f(client, ecdsa_521_hmac_sha2_256, maccmd("hmac-sha2-256"), setup_ecdsa_521, teardown) \
|
|
|
|
|
f(client, rsa_hmac_sha2_512, maccmd("hmac-sha2-512"), setup_rsa, teardown) \
|
|
|
|
|
f(client, dsa_hmac_sha2_512, maccmd("hmac-sha2-512"), setup_dsa, teardown) \
|
|
|
|
|
f(client, ecdsa_256_hmac_sha2_512, maccmd("hmac-sha2-512"), setup_ecdsa_256, teardown) \
|
|
|
|
|
f(client, ecdsa_384_hmac_sha2_512, maccmd("hmac-sha2-512"), setup_ecdsa_384, teardown) \
|
|
|
|
|
f(client, ecdsa_521_hmac_sha2_512, maccmd("hmac-sha2-512"), setup_ecdsa_521, teardown)
|
2017-12-28 13:10:43 +03:00
|
|
|
#else
|
|
|
|
|
#define PKDTESTS_MAC(f, client, maccmd) \
|
|
|
|
|
/* MACs. */ \
|
|
|
|
|
f(client, rsa_hmac_sha1, maccmd("hmac-sha1"), setup_rsa, teardown) \
|
|
|
|
|
f(client, ecdsa_256_hmac_sha1, maccmd("hmac-sha1"), setup_ecdsa_256, teardown) \
|
|
|
|
|
f(client, ecdsa_384_hmac_sha1, maccmd("hmac-sha1"), setup_ecdsa_384, teardown) \
|
|
|
|
|
f(client, ecdsa_521_hmac_sha1, maccmd("hmac-sha1"), setup_ecdsa_521, teardown) \
|
|
|
|
|
f(client, rsa_hmac_sha2_256, maccmd("hmac-sha2-256"), setup_rsa, teardown) \
|
|
|
|
|
f(client, ecdsa_256_hmac_sha2_256, maccmd("hmac-sha2-256"), setup_ecdsa_256, teardown) \
|
|
|
|
|
f(client, ecdsa_384_hmac_sha2_256, maccmd("hmac-sha2-256"), setup_ecdsa_384, teardown) \
|
|
|
|
|
f(client, ecdsa_521_hmac_sha2_256, maccmd("hmac-sha2-256"), setup_ecdsa_521, teardown) \
|
|
|
|
|
f(client, rsa_hmac_sha2_512, maccmd("hmac-sha2-512"), setup_rsa, teardown) \
|
|
|
|
|
f(client, ecdsa_256_hmac_sha2_512, maccmd("hmac-sha2-512"), setup_ecdsa_256, teardown) \
|
|
|
|
|
f(client, ecdsa_384_hmac_sha2_512, maccmd("hmac-sha2-512"), setup_ecdsa_384, teardown) \
|
|
|
|
|
f(client, ecdsa_521_hmac_sha2_512, maccmd("hmac-sha2-512"), setup_ecdsa_521, teardown)
|
|
|
|
|
#endif
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
|
|
|
|
|
static void torture_pkd_client_noop(void **state) {
|
|
|
|
|
struct pkd_state *pstate = (struct pkd_state *) (*state);
|
|
|
|
|
(void) pstate;
|
|
|
|
|
return;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static void torture_pkd_runtest(const char *testname,
|
|
|
|
|
const char *testcmd)
|
|
|
|
|
{
|
|
|
|
|
int i, rc;
|
|
|
|
|
char logfile[1024] = { 0 };
|
|
|
|
|
int iterations =
|
|
|
|
|
(pkd_dargs.opts.iterations != 0) ? pkd_dargs.opts.iterations
|
|
|
|
|
: DEFAULT_ITERATIONS;
|
|
|
|
|
|
|
|
|
|
for (i = 0; i < iterations; i++) {
|
|
|
|
|
rc = system_checked(testcmd);
|
|
|
|
|
assert_int_equal(rc, 0);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/* Asserts did not trip: cleanup logs. */
|
|
|
|
|
snprintf(&logfile[0], sizeof(logfile), "%s.out", testname);
|
|
|
|
|
unlink(logfile);
|
|
|
|
|
snprintf(&logfile[0], sizeof(logfile), "%s.err", testname);
|
|
|
|
|
unlink(logfile);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Though each keytest function body is the same, separate functions are
|
|
|
|
|
* defined here to result in distinct output when running the tests.
|
|
|
|
|
*/
|
|
|
|
|
|
|
|
|
|
#define emit_keytest(client, testname, sshcmd, setup, teardown) \
|
|
|
|
|
static void torture_pkd_## client ## _ ## testname(void **state) { \
|
|
|
|
|
const char *tname = "torture_pkd_" #client "_" #testname; \
|
|
|
|
|
char testcmd[1024] = { 0 }; \
|
|
|
|
|
(void) state; \
|
|
|
|
|
snprintf(&testcmd[0], sizeof(testcmd), sshcmd, tname, tname); \
|
|
|
|
|
torture_pkd_runtest(tname, testcmd); \
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Actual test functions are emitted here.
|
|
|
|
|
*/
|
|
|
|
|
|
2017-12-28 13:10:43 +03:00
|
|
|
#ifdef HAVE_DSA
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
#define CLIENT_ID_FILE OPENSSH_DSA_TESTKEY
|
|
|
|
|
PKDTESTS_DEFAULT(emit_keytest, openssh_dsa, OPENSSH_CMD)
|
|
|
|
|
PKDTESTS_KEX(emit_keytest, openssh_dsa, OPENSSH_KEX_CMD)
|
|
|
|
|
PKDTESTS_CIPHER(emit_keytest, openssh_dsa, OPENSSH_CIPHER_CMD)
|
2017-08-17 10:18:20 +03:00
|
|
|
PKDTESTS_CIPHER_OPENSSHONLY(emit_keytest, openssh_dsa, OPENSSH_CIPHER_CMD)
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
PKDTESTS_MAC(emit_keytest, openssh_dsa, OPENSSH_MAC_CMD)
|
|
|
|
|
#undef CLIENT_ID_FILE
|
2017-12-28 13:10:43 +03:00
|
|
|
#endif
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
|
|
|
|
|
#define CLIENT_ID_FILE OPENSSH_RSA_TESTKEY
|
|
|
|
|
PKDTESTS_DEFAULT(emit_keytest, openssh_rsa, OPENSSH_CMD)
|
|
|
|
|
PKDTESTS_KEX(emit_keytest, openssh_rsa, OPENSSH_KEX_CMD)
|
|
|
|
|
PKDTESTS_CIPHER(emit_keytest, openssh_rsa, OPENSSH_CIPHER_CMD)
|
2017-08-17 10:18:20 +03:00
|
|
|
PKDTESTS_CIPHER_OPENSSHONLY(emit_keytest, openssh_rsa, OPENSSH_CIPHER_CMD)
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
PKDTESTS_MAC(emit_keytest, openssh_rsa, OPENSSH_MAC_CMD)
|
|
|
|
|
#undef CLIENT_ID_FILE
|
|
|
|
|
|
|
|
|
|
#define CLIENT_ID_FILE OPENSSH_ECDSA256_TESTKEY
|
|
|
|
|
PKDTESTS_DEFAULT(emit_keytest, openssh_e256, OPENSSH_CMD)
|
|
|
|
|
PKDTESTS_KEX(emit_keytest, openssh_e256, OPENSSH_KEX_CMD)
|
|
|
|
|
PKDTESTS_CIPHER(emit_keytest, openssh_e256, OPENSSH_CIPHER_CMD)
|
2017-08-17 10:18:20 +03:00
|
|
|
PKDTESTS_CIPHER_OPENSSHONLY(emit_keytest, openssh_e256, OPENSSH_CIPHER_CMD)
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
PKDTESTS_MAC(emit_keytest, openssh_e256, OPENSSH_MAC_CMD)
|
|
|
|
|
#undef CLIENT_ID_FILE
|
|
|
|
|
|
|
|
|
|
/* Could add these passes, too: */
|
|
|
|
|
//#define CLIENT_ID_FILE OPENSSH_ECDSA384_TESTKEY
|
|
|
|
|
//#define CLIENT_ID_FILE OPENSSH_ECDSA521_TESTKEY
|
|
|
|
|
|
|
|
|
|
#define CLIENT_ID_FILE OPENSSH_ED25519_TESTKEY
|
|
|
|
|
PKDTESTS_DEFAULT(emit_keytest, openssh_ed, OPENSSH_CMD)
|
|
|
|
|
PKDTESTS_KEX(emit_keytest, openssh_ed, OPENSSH_KEX_CMD)
|
|
|
|
|
PKDTESTS_CIPHER(emit_keytest, openssh_ed, OPENSSH_CIPHER_CMD)
|
2017-08-17 10:18:20 +03:00
|
|
|
PKDTESTS_CIPHER_OPENSSHONLY(emit_keytest, openssh_ed, OPENSSH_CIPHER_CMD)
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
PKDTESTS_MAC(emit_keytest, openssh_ed, OPENSSH_MAC_CMD)
|
|
|
|
|
#undef CLIENT_ID_FILE
|
|
|
|
|
|
|
|
|
|
#define CLIENT_ID_FILE DROPBEAR_RSA_TESTKEY
|
|
|
|
|
PKDTESTS_DEFAULT(emit_keytest, dropbear, DROPBEAR_CMD)
|
|
|
|
|
PKDTESTS_CIPHER(emit_keytest, dropbear, DROPBEAR_CIPHER_CMD)
|
|
|
|
|
PKDTESTS_MAC(emit_keytest, dropbear, DROPBEAR_MAC_CMD)
|
|
|
|
|
#undef CLIENT_ID_FILE
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Define an array of testname strings mapped to their associated
|
|
|
|
|
* test function. Enables running tests individually by name from
|
|
|
|
|
* the command line.
|
|
|
|
|
*/
|
|
|
|
|
|
|
|
|
|
#define emit_testmap(client, testname, sshcmd, setup, teardown) \
|
|
|
|
|
{ "torture_pkd_" #client "_" #testname, \
|
2017-08-17 10:18:20 +03:00
|
|
|
emit_unit_test(client, testname, sshcmd, setup, teardown) },
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
|
|
|
|
|
#define emit_unit_test(client, testname, sshcmd, setup, teardown) \
|
2015-09-07 13:44:23 +03:00
|
|
|
cmocka_unit_test_setup_teardown(torture_pkd_ ## client ## _ ## testname, \
|
|
|
|
|
torture_pkd_ ## setup, \
|
|
|
|
|
torture_pkd_ ## teardown)
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
|
|
|
|
|
#define emit_unit_test_comma(client, testname, sshcmd, setup, teardown) \
|
|
|
|
|
emit_unit_test(client, testname, sshcmd, setup, teardown),
|
|
|
|
|
|
|
|
|
|
struct {
|
|
|
|
|
const char *testname;
|
2017-08-17 10:18:20 +03:00
|
|
|
const struct CMUnitTest test;
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
} testmap[] = {
|
|
|
|
|
/* OpenSSH */
|
2017-12-28 13:10:43 +03:00
|
|
|
#ifdef HAVE_DSA
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
PKDTESTS_DEFAULT(emit_testmap, openssh_dsa, OPENSSH_CMD)
|
|
|
|
|
PKDTESTS_KEX(emit_testmap, openssh_dsa, OPENSSH_KEX_CMD)
|
|
|
|
|
PKDTESTS_CIPHER(emit_testmap, openssh_dsa, OPENSSH_CIPHER_CMD)
|
2017-08-17 10:18:20 +03:00
|
|
|
PKDTESTS_CIPHER_OPENSSHONLY(emit_testmap, openssh_dsa, OPENSSH_CIPHER_CMD)
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
PKDTESTS_MAC(emit_testmap, openssh_dsa, OPENSSH_MAC_CMD)
|
2017-12-28 13:10:43 +03:00
|
|
|
#endif
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
|
|
|
|
|
PKDTESTS_DEFAULT(emit_testmap, openssh_rsa, OPENSSH_CMD)
|
|
|
|
|
PKDTESTS_KEX(emit_testmap, openssh_rsa, OPENSSH_KEX_CMD)
|
|
|
|
|
PKDTESTS_CIPHER(emit_testmap, openssh_rsa, OPENSSH_CIPHER_CMD)
|
2017-08-17 10:18:20 +03:00
|
|
|
PKDTESTS_CIPHER_OPENSSHONLY(emit_testmap, openssh_rsa, OPENSSH_CIPHER_CMD)
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
PKDTESTS_MAC(emit_testmap, openssh_rsa, OPENSSH_MAC_CMD)
|
|
|
|
|
|
|
|
|
|
PKDTESTS_DEFAULT(emit_testmap, openssh_e256, OPENSSH_CMD)
|
|
|
|
|
PKDTESTS_KEX(emit_testmap, openssh_e256, OPENSSH_KEX_CMD)
|
|
|
|
|
PKDTESTS_CIPHER(emit_testmap, openssh_e256, OPENSSH_CIPHER_CMD)
|
2017-08-17 10:18:20 +03:00
|
|
|
PKDTESTS_CIPHER_OPENSSHONLY(emit_testmap, openssh_e256, OPENSSH_CIPHER_CMD)
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
PKDTESTS_MAC(emit_testmap, openssh_e256, OPENSSH_MAC_CMD)
|
|
|
|
|
|
|
|
|
|
PKDTESTS_DEFAULT(emit_testmap, openssh_ed, OPENSSH_CMD)
|
|
|
|
|
PKDTESTS_KEX(emit_testmap, openssh_ed, OPENSSH_KEX_CMD)
|
|
|
|
|
PKDTESTS_CIPHER(emit_testmap, openssh_ed, OPENSSH_CIPHER_CMD)
|
2017-08-17 10:18:20 +03:00
|
|
|
PKDTESTS_CIPHER_OPENSSHONLY(emit_testmap, openssh_ed, OPENSSH_CIPHER_CMD)
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
PKDTESTS_MAC(emit_testmap, openssh_ed, OPENSSH_MAC_CMD)
|
|
|
|
|
|
|
|
|
|
/* Dropbear */
|
|
|
|
|
PKDTESTS_DEFAULT(emit_testmap, dropbear, DROPBEAR_CMD)
|
|
|
|
|
PKDTESTS_CIPHER(emit_testmap, dropbear, DROPBEAR_CIPHER_CMD)
|
|
|
|
|
PKDTESTS_MAC(emit_testmap, dropbear, DROPBEAR_MAC_CMD)
|
|
|
|
|
|
|
|
|
|
/* Noop */
|
|
|
|
|
emit_testmap(client, noop, "", setup_noop, teardown)
|
|
|
|
|
|
|
|
|
|
/* NULL tail entry */
|
2017-08-17 10:18:20 +03:00
|
|
|
{ .testname = NULL,
|
|
|
|
|
.test = { .name = NULL,
|
|
|
|
|
.test_func = NULL,
|
|
|
|
|
.setup_func = NULL,
|
|
|
|
|
.teardown_func = NULL } }
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
};
|
|
|
|
|
|
|
|
|
|
static int pkd_run_tests(void) {
|
|
|
|
|
int rc = -1;
|
|
|
|
|
int tindex = 0;
|
|
|
|
|
|
2015-09-07 13:44:23 +03:00
|
|
|
const struct CMUnitTest openssh_tests[] = {
|
2017-12-28 13:10:43 +03:00
|
|
|
#ifdef HAVE_DSA
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
PKDTESTS_DEFAULT(emit_unit_test_comma, openssh_dsa, OPENSSH_CMD)
|
|
|
|
|
PKDTESTS_KEX(emit_unit_test_comma, openssh_dsa, OPENSSH_KEX_CMD)
|
|
|
|
|
PKDTESTS_CIPHER(emit_unit_test_comma, openssh_dsa, OPENSSH_CIPHER_CMD)
|
2017-08-17 10:18:20 +03:00
|
|
|
PKDTESTS_CIPHER_OPENSSHONLY(emit_unit_test_comma, openssh_dsa, OPENSSH_CIPHER_CMD)
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
PKDTESTS_MAC(emit_unit_test_comma, openssh_dsa, OPENSSH_MAC_CMD)
|
2017-12-28 13:10:43 +03:00
|
|
|
#endif
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
|
|
|
|
|
PKDTESTS_DEFAULT(emit_unit_test_comma, openssh_rsa, OPENSSH_CMD)
|
|
|
|
|
PKDTESTS_KEX(emit_unit_test_comma, openssh_rsa, OPENSSH_KEX_CMD)
|
|
|
|
|
PKDTESTS_CIPHER(emit_unit_test_comma, openssh_rsa, OPENSSH_CIPHER_CMD)
|
2017-08-17 10:18:20 +03:00
|
|
|
PKDTESTS_CIPHER_OPENSSHONLY(emit_unit_test_comma, openssh_rsa, OPENSSH_CIPHER_CMD)
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
PKDTESTS_MAC(emit_unit_test_comma, openssh_rsa, OPENSSH_MAC_CMD)
|
|
|
|
|
|
|
|
|
|
PKDTESTS_DEFAULT(emit_unit_test_comma, openssh_e256, OPENSSH_CMD)
|
|
|
|
|
PKDTESTS_KEX(emit_unit_test_comma, openssh_e256, OPENSSH_KEX_CMD)
|
|
|
|
|
PKDTESTS_CIPHER(emit_unit_test_comma, openssh_e256, OPENSSH_CIPHER_CMD)
|
2017-08-17 10:18:20 +03:00
|
|
|
PKDTESTS_CIPHER_OPENSSHONLY(emit_unit_test_comma, openssh_e256, OPENSSH_CIPHER_CMD)
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
PKDTESTS_MAC(emit_unit_test_comma, openssh_e256, OPENSSH_MAC_CMD)
|
|
|
|
|
|
|
|
|
|
PKDTESTS_DEFAULT(emit_unit_test_comma, openssh_ed, OPENSSH_CMD)
|
|
|
|
|
PKDTESTS_KEX(emit_unit_test_comma, openssh_ed, OPENSSH_KEX_CMD)
|
|
|
|
|
PKDTESTS_CIPHER(emit_unit_test_comma, openssh_ed, OPENSSH_CIPHER_CMD)
|
2017-08-17 10:18:20 +03:00
|
|
|
PKDTESTS_CIPHER_OPENSSHONLY(emit_unit_test_comma, openssh_ed, OPENSSH_CIPHER_CMD)
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
PKDTESTS_MAC(emit_unit_test_comma, openssh_ed, OPENSSH_MAC_CMD)
|
|
|
|
|
};
|
|
|
|
|
|
2015-09-07 13:44:23 +03:00
|
|
|
const struct CMUnitTest dropbear_tests[] = {
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
PKDTESTS_DEFAULT(emit_unit_test_comma, dropbear, DROPBEAR_CMD)
|
|
|
|
|
PKDTESTS_CIPHER(emit_unit_test_comma, dropbear, DROPBEAR_CIPHER_CMD)
|
|
|
|
|
PKDTESTS_MAC(emit_unit_test_comma, dropbear, DROPBEAR_MAC_CMD)
|
|
|
|
|
};
|
|
|
|
|
|
2015-09-07 13:44:23 +03:00
|
|
|
const struct CMUnitTest noop_tests[] = {
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
emit_unit_test(client, noop, "", setup_noop, teardown)
|
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
/* Test list is populated depending on which clients are enabled. */
|
2015-09-07 13:44:23 +03:00
|
|
|
struct CMUnitTest all_tests[(sizeof(openssh_tests) / sizeof(openssh_tests[0])) +
|
2017-08-17 10:18:20 +03:00
|
|
|
(sizeof(dropbear_tests) / sizeof(dropbear_tests[0])) +
|
|
|
|
|
(sizeof(noop_tests) / sizeof(noop_tests[0]))];
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
memset(&all_tests[0], 0x0, sizeof(all_tests));
|
|
|
|
|
|
|
|
|
|
/* Generate client keys and populate test list for each enabled client. */
|
|
|
|
|
if (is_openssh_client_enabled()) {
|
|
|
|
|
setup_openssh_client_keys();
|
|
|
|
|
memcpy(&all_tests[tindex], &openssh_tests[0], sizeof(openssh_tests));
|
|
|
|
|
tindex += (sizeof(openssh_tests) / sizeof(openssh_tests[0]));
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
if (is_dropbear_client_enabled()) {
|
|
|
|
|
setup_dropbear_client_rsa_key();
|
|
|
|
|
memcpy(&all_tests[tindex], &dropbear_tests[0], sizeof(dropbear_tests));
|
|
|
|
|
tindex += (sizeof(dropbear_tests) / sizeof(dropbear_tests[0]));
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
memcpy(&all_tests[tindex], &noop_tests[0], sizeof(noop_tests));
|
|
|
|
|
tindex += (sizeof(noop_tests) / sizeof(noop_tests[0]));
|
|
|
|
|
|
2017-08-17 10:18:20 +03:00
|
|
|
if ((pkd_dargs.opts.testname == NULL) &&
|
|
|
|
|
(pkd_dargs.opts.testmatch == NULL)) {
|
2015-09-07 13:44:23 +03:00
|
|
|
rc = _cmocka_run_group_tests("all tests", all_tests, tindex, NULL, NULL);
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
} else {
|
|
|
|
|
int i = 0;
|
2017-08-17 10:18:20 +03:00
|
|
|
int num_found = 0;
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
const char *testname = pkd_dargs.opts.testname;
|
2017-08-17 10:18:20 +03:00
|
|
|
const char *testmatch = pkd_dargs.opts.testmatch;
|
|
|
|
|
|
|
|
|
|
struct CMUnitTest matching_tests[sizeof(all_tests)];
|
|
|
|
|
memset(&matching_tests[0], 0x0, sizeof(matching_tests));
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
|
|
|
|
|
while (testmap[i].testname != NULL) {
|
2017-08-17 10:18:20 +03:00
|
|
|
if ((testname != NULL) &&
|
|
|
|
|
(strcmp(testmap[i].testname, testname) == 0)) {
|
|
|
|
|
memcpy(&matching_tests[0],
|
|
|
|
|
&testmap[i].test,
|
|
|
|
|
sizeof(struct CMUnitTest));
|
|
|
|
|
num_found += 1;
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
break;
|
|
|
|
|
}
|
2017-08-17 10:18:20 +03:00
|
|
|
|
|
|
|
|
if ((testmatch != NULL) &&
|
|
|
|
|
(strstr(testmap[i].testname, testmatch) != NULL)) {
|
|
|
|
|
memcpy(&matching_tests[num_found],
|
|
|
|
|
&testmap[i].test,
|
|
|
|
|
sizeof(struct CMUnitTest));
|
|
|
|
|
num_found += 1;
|
|
|
|
|
}
|
|
|
|
|
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
i += 1;
|
|
|
|
|
}
|
|
|
|
|
|
2017-08-17 10:18:20 +03:00
|
|
|
if (num_found > 0) {
|
|
|
|
|
rc = _cmocka_run_group_tests("found", matching_tests, num_found, NULL, NULL);
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
} else {
|
|
|
|
|
fprintf(stderr, "Did not find test '%s'\n", testname);
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/* Clean up client keys for each enabled client. */
|
|
|
|
|
if (is_dropbear_client_enabled()) {
|
|
|
|
|
cleanup_dropbear_client_rsa_key();
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
if (is_openssh_client_enabled()) {
|
|
|
|
|
cleanup_openssh_client_keys();
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/* Clean up any server keys that were generated. */
|
|
|
|
|
cleanup_rsa_key();
|
2017-12-28 13:10:43 +03:00
|
|
|
#ifdef HAVE_DSA
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
cleanup_dsa_key();
|
2017-12-28 13:10:43 +03:00
|
|
|
#endif
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
cleanup_ecdsa_keys();
|
|
|
|
|
|
|
|
|
|
return rc;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
int main(int argc, char **argv) {
|
|
|
|
|
int i = 0;
|
|
|
|
|
int rc = 0;
|
2017-09-10 06:12:33 +03:00
|
|
|
int exit_code = -1;
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
|
|
|
|
|
unsetenv("SSH_AUTH_SOCK");
|
|
|
|
|
|
|
|
|
|
rc = ssh_init();
|
|
|
|
|
if (rc != 0) {
|
|
|
|
|
rc = SSH_ERROR;
|
|
|
|
|
goto out;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
#ifdef HAVE_ARGP_H
|
|
|
|
|
argp_parse(&parser, argc, argv, 0, 0, NULL);
|
|
|
|
|
#else /* HAVE_ARGP_H */
|
|
|
|
|
(void) argc; (void) argv;
|
|
|
|
|
#endif /* HAVE_ARGP_H */
|
|
|
|
|
|
|
|
|
|
if (pkd_dargs.opts.list != 0) {
|
|
|
|
|
while (testmap[i].testname != NULL) {
|
|
|
|
|
printf("%s\n", testmap[i++].testname);
|
|
|
|
|
}
|
|
|
|
|
} else {
|
2017-09-10 06:12:33 +03:00
|
|
|
exit_code = pkd_run_tests();
|
|
|
|
|
if (exit_code != 0) {
|
|
|
|
|
fprintf(stderr, "pkd_run_tests failed: %d\n", exit_code);
|
2015-01-14 12:43:14 +03:00
|
|
|
}
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
rc = ssh_finalize();
|
|
|
|
|
if (rc != 0) {
|
|
|
|
|
fprintf(stderr, "ssh_finalize: %d\n", rc);
|
|
|
|
|
}
|
|
|
|
|
out:
|
2017-09-10 06:12:33 +03:00
|
|
|
return exit_code;
|
tests: introduce pkd_hello
Introduce a sample public-key testing daemon to the 'pkd' test directory,
and add support code for cycling through various combinations of different
key exchange, cipher, and MAC algorithms.
The goal of the 'pkd_hello' test is to make it easy to test interactions
between non-libssh clients and a libssh-server, and to provide a starting
point for testing new implementations for key types, ciphers, MACs, and
so on. The thinking is that testing a new algorithm should be as simple
as adding a new line for it in the PKDTESTS_* lists.
Macros are used to generate the tests and helper functions for a couple of
clients -- here, OpenSSH and dropbear are included for the first cut. If
binaries are found for these clients, their test lists will be enabled;
when binaries are not found for a given client, those tests are skipped.
Tests are run in one large batch by default, but can also be run individually
to help with tracking down things like signature bugs that may take many
iterations to reproduce.
Each test logs its stdout and stderr to its own file, which is cleaned up
when a test succeeds. For failures, those logs can be combined with verbose
libssh output from pkd itself to start debugging things.
Some example usages:
pkd_hello
Run all tests with default number of iterations.
pkd_hello --list
List available individual test names.
pkd_hello -i 1000 -t torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
Run only the torture_pkd_openssh_ecdsa_256_ecdh_sha2_nistp256
testcase 1000 times.
pkd_hello -v -v -v -v -e -o
Run all tests with maximum libssh and pkd logging.
Included in the tests are passes for all existing kex, cipher, and MAC
algorithms.
BUG: https://red.libssh.org/issues/144
Signed-off-by: Jon Simons <jon@jonsimons.org>
Reviewed-by: Andreas Schneider <asn@cryptomilk.org>
2014-10-10 23:38:45 +04:00
|
|
|
}
|