Fixes an issue on Windows where when source and build directory are on different drives hard-linking
to files or directory fails as it doesn't work across filesystem boundaries. Note that symlinking is also
not possible because it requires administrator privileges on Windows.
The solution copies the files using the built-in cmake `configure_file(src dest COPYONLY)` command.
As this command only operates on files, if a directory is specified the files will be globbed recursively
and through symlinks.
Signed-off-by: Dominik Gschwind <dominik.gschwind99@gmail.com>
The test cases using tcp_client.pl are failing only on Travis's bionic and
focal systems (I can't reproduce the failure locally). Exclude them.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Don't use an all.sh component because there isn't one that does what we
want (modern Clang with ASan, and test everything).
* We need to set CC explicitly or tweak PATH, because clang in $PATH on
Travis focal instances is Clang 7 which is too old (we want Clang 10).
* Travis lacks the array of versions of openssl and gnutls that we normally
use for testing, so we need to exclude some tests (or build our
own multiple versions of openssl and gnutls).
The SSL test exclusions are ad hoc and based on what currently works.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Restore compatibiltiy testing against OpenSSL for
(D)TLS versions smaller that 1.2.
. Fix the check for support in OpenSSL for these versions
. For test cases for (D)TLS version smaller than 1.2,
restore the configuration of OpenSSL client/server
with the given TLS version, as it was before #5660
that broke it.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
OpenSSL may be configured to support features such as cipher suites or
protocol versions that are disabled by default. Enable them all: we're
testing, we don't care about enabling insecure stuff. This is not needed
with the builds of OpenSSL that we're currently using on the Jenkins CI, but
it's needed with more recent versions such as typically found on developer
machines, and with future CI additions.
The syntax to do that was only introduced in OpenSSL 1.1.0; fortunately we
don't need to do anything special with earlier versions.
With OpenSSL 1.1.1f on Ubuntu 20.04, this is needed to enable TLS 1.0, TLS
1.1 and DTLS 1.0. This also allows SHA-1 in certificates, which is still
needed for a few test cases in ssl-opt.sh. Curiously, this is also needed
for the cipher suite TLS-DHE-PSK-WITH-ARIA-128-GCM-SHA256 (and no other,
including other DHE-PSK or ARIA cipher suites).
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Restore compatibiltiy testing against OpenSSL for
(D)TLS versions smaller that 1.2.
. Fix the check for support in OpenSSL for these versions
. For test cases for (D)TLS version smaller than 1.2,
restore the configuration of OpenSSL client/server
with the given TLS version, as it was before #5660
that broke it.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
psa_cipher_encrypt() and psa_cipher_decrypt() sometimes add a zero offset to
a null pointer when the cipher does not use an IV. This is undefined
behavior, although it works as naively expected on most platforms. This
can cause a crash with modern Clang+ASan (depending on compiler optimizations).
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
When x is the most negative value of a two's complement type,
`(unsigned_type)(-x)` has undefined behavior, whereas `-(unsigned_type)x`
has well-defined behavior and does what was intended.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Fix undefined behavior (typically harmless in practice) of
mbedtls_mpi_add_mpi(), mbedtls_mpi_add_abs() and mbedtls_mpi_add_int() when
both operands are 0 and the left operand is represented with 0 limbs.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Since they're part of the public API (even if only through a few functions),
they should be documented.
I deliberately skipped documenting how to configure the size of the type.
Right now, MBEDTLS_HAVE_INT32 and MBEDTLS_HAVE_INT64 have no Doxygen
documentation, so it's ambiguous whether they're part of the public API.
Resolving this ambiguity is out of scope of my current work.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Looking for the .data file doesn't work in out-of-tree builds. Use the
.datax file instead. `make clean` removes all .datax files, so this resolves
the issue of executables not present on the current branch being left behind
after a branch change followed by a `make clean`.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
The already existing "x509_csr_check()" function is extended in order
to support/test also CSR's extensions. The test is performed by
adding an extended key usage.
Signed-off-by: Valerio Setti <vsetti@baylibre.com>
Although negative zero is officially unsupported, we've had bugs related to
it in the past. So do test functions with a negative zero input.
There will likely be cases where we don't want to accept negative zero as if
it was valid, because it's too hard to handle. We'll add exceptions on a
case by case basis.
For the functions that are currently tested by the generated tests, the new
test cases pass.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
The bignum module does not officially support "negative zero" (an
mbedtls_mpi object with s=-1 and all limbs zero). However, we have a
history of bugs where a function that should produce an official
zero (with s=1), produces a negative zero in some circumstances. So it's
good to check that the bignum functions are robust when passed a negative
zero as input. And for that, we need a way to construct a negative zero
from test case arguments.
There are checks that functions don't produce negative zeros as output in
the test suite. Skip those checks if there's a negative zero input: we
don't want functions to _create_ negative zeros, but we don't mind if
they _propagate_ negative zeros.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
In mbedtls_mpi_add_mpi() and mbedtls_mpi_sub_mpi(), and by extention
mbedtls_mpi_add_int() and mbedtls_mpi_sub_int(), when the resulting value
was zero, the sign bit of the result was incorrectly set to -1 when the
left-hand operand was negative. This is not a valid mbedtls_mpi
representation. Fix this: always set the sign to +1 when the result is 0.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
This exposes a bug in mbedtls_mpi_add_mpi() and mbedtls_mpi_sub_mpi() which
will be fixed in a subsequent commit.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
mbedtls_mpi_add_mpi() and mbedtls_mpi_sub_mpi() have the same logic, just
with one bit to flip in the sign calculation. Move the shared logic to a new
auxiliary function. This slightly reduces the code size (if the compiler
doesn't inline) and reduces the maintenance burden.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
