The paths to this header in clean: targets were missed previously when
changing paths to reflect the move of the header to the framework.
Signed-off-by: David Horstmann <david.horstmann@arm.com>
These are not moving to the framework, so their paths are fine to stay
as they are, undo the update done through global search-replace.
Signed-off-by: David Horstmann <david.horstmann@arm.com>
The find and replace of tests/src -> framework/tests/src and
tests/include -> framework/tests/include accidentally caused an
incorrect path substitution to be performed in tests/Makefile.
Undo this change and make the substitution correct again.
Signed-off-by: David Horstmann <david.horstmann@arm.com>
These need to include from the framework path as this is where most test
helpers have been moved to.
Signed-off-by: David Horstmann <david.horstmann@arm.com>
Everything not TLS-related and not ALT-interface-related has been moved
out of tests/src and tests/include and into the framework, except for
the PSA test wrappers.
Signed-off-by: David Horstmann <david.horstmann@arm.com>
Non generated code was unfortnuately added to psa_sim_crypto_server.c,
fix that by adding it to the generation scripts.
Signed-off-by: Paul Elliott <paul.elliott@arm.com>
Incorrect usage of psa_key_id_t breaks ability to regenerate psa_sim
headers properly, should have been mbedtls_svc_key_id_t in the first
place, so replace with that.
Signed-off-by: Paul Elliott <paul.elliott@arm.com>
Fix everest build following the merge of
9490. By the way removed an unnecessary
link of the everest lib to mbedtls test
executable. It comes through the mbedtls
library that gets it from the tfpsacrypto
library.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
Added `MBEDTLS_CIPHER_MODE_XTS` as a dependency for `PSA_WANT_ALG_XTS`.
Otherwise, `MBEDTLS_CIPHER_MODE_XTS` is always enabled which enables a
lot of code we would like to be disabled when testing CFB/CTR... only.
Signed-off-by: Gabor Mezei <gabor.mezei@arm.com>
All cipher padding methods depend on CBC. To aviod switching it off
add this dependency to all of the methods and handle it as a common dependency.
Signed-off-by: Gabor Mezei <gabor.mezei@arm.com>
When elements of an exclusive group have dependencies in common turning them off
breaks the elements build. Support added to handle and ignore these dependencies
when only one of the elements is enabled.
Signed-off-by: Gabor Mezei <gabor.mezei@arm.com>
It's a crypto test suite, but it was added in the main tree in a careless
forward port from 3.6.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Remove the configuration option MBEDTLS_KEY_EXCHANGE_RSA_PSK_ENABLED and all
code guarded by it. This remove support for the RSA-PSK key exchange in TLS
1.2.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
The test cases removed by this commit are specific to RSA-PSK, not
incidentally using RSA-PSK when testing other features, so there is no loss
of test coverage.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
generate_test_code.py doesn't support UTF-8 in .function files (but does
in .data files) when run in a non-UTF-8 locale with Python <=3.6.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Run the function on a few different curves with shorter, just-right and
larger buffer sizes. Perform some basic sanity checks on the output (which
is random).
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Skip 6144-bit and 8192-bit FFDH tests in MSan or Valgrind test
configurations. These test cases are slow and sometimes trigger GnuTLS's
built-in timeout (which cannot be configured until more recent versions).
Resolves#9742.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Now that it's always on, this is no longer necessary.
Also this is a step towards no longer being able to set it, because it soon
won't be a selectable configuration option anymore.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Now that MBEDTLS_USE_PSA_CRYPTO is always on, remove code that handles the
case when it's disabled.
Also this is a step towards no longer expecting it to be reported by
query_compile_time_config.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Now that MBEDTLS_USE_PSA_CRYPTO is always on, we don't need to check for it.
Also this is a step towards no longer expecting it to be reported by
query_compile_time_config.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
We no longer have two (only partially distinct) implementations of ECJ-PAKE
cipher suites in TLS, now that the non-MBEDTLS_USE_PSA_CRYPTO implementation
is being removed.
We may want to add this testing back in the future, but we'll have to use an
old Mbed TLS instead of a differently-built one.
https://github.com/Mbed-TLS/mbedtls/issues/9740
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
The Thumb-1 tests are by far the slowest, so split the -O0 and -Os tests
into their own components.
Signed-off-by: Bence Szépkúti <bence.szepkuti@arm.com>
Since PSA is always on in 4.x, *_PSA_INIT() and *_PSA_DONE() can be
simply aliased to PSA_INIT() and PSA_DONE() until such a time as we
remove them.
Simplify the login of these PSA_INIT/DONE variants by aliasing them in
4.x and keeping the more complex 3.6 logic entirely separate.
Signed-off-by: David Horstmann <david.horstmann@arm.com>
Remove all.sh components that explicitly disable MBEDTLS_USE_PSA_CRYPTO, and
for which there is another component with MBEDTLS_USE_PSA_CRYPTO enabled
that does the same or more testing.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Variables that are in any way destructed on exit should be initialised
prior to any tests that might jump to exit, to save potential
uninitialised memory accesses.
Signed-off-by: Paul Elliott <paul.elliott@arm.com>
In order to avoid changing the CI job's definition (which fewer team
members understand, compared to shell scripts), just have a wrapper so
that all.sh does both mbedtls and tf-psa-crypto under the hood for now.
When tf-psa-crypto has its own CI running its own all.sh (with enough
components migrated to get sufficient coverage) we can remove this hack.
Rejected strategy: modify all-core.sh so that when running in mbedtls,
it also sources components-*.sh from tf-psa-crypto, remembers which
components come from Mbed TLS and which come from crypto, and magically
adjust the environment for each component it runs. Rejected because it's
hard to be confident we're adjusting everything that needs adjusting in
the environment. Having separate processes seems much safer.
The downside is we get more complexity around error handling (check for
components requested but not available, --keep-going). When using
--keep-going with failing component(s) in mbedtls and some components in
tf-psa-crypto, the output is not satisfying for humans as we don't have
a nice summary of all errors at the end like we normally would.
IMO this is acceptable since:
- this wrapper is transitional and should be removed in a few months;
- it is mainly for the benefit of the CI; humans can always invoke the
underlying commands directly.
Signed-off-by: Manuel Pégourié-Gonnard <manuel.pegourie-gonnard@arm.com>
The existing check only took care of CONFIG_H. This was both not enough
and too much:
- not enough because config.py can also modify CRYPTO_CONFIG_H and we
want to know about it just as much as CONFIG_H;
- too much because CONFIG_H does not exist in tf-psa-crypto.
Check a list of files instead of a single one, and adjust that list.
Also update an outdated comment about Makefiles
Signed-off-by: Manuel Pégourié-Gonnard <manuel.pegourie-gonnard@arm.com>
In preparation for adding tf-psa-crypto/test/scripts/all.sh which will
run from tf-psa-crypto.
Use paths relative to the currently sourced file when including common
files (ie, those that will soon be moved to the framework). Otherwise,
use paths relative to the current directory, aka project's root.
Document that test/script/all.sh must be invoked from the project's root
(that was already the case, but implicit so far).
Signed-off-by: Manuel Pégourié-Gonnard <manuel.pegourie-gonnard@arm.com>
With PSA_WANT_KEY_TYPE_ECC_KEY_PAIR_DERIVE disabled, test TLS 1.3 and
USE_PSA TLS 1.2.
With PSA_WANT_KEY_TYPE_ECC_KEY_PAIR_GENERATE disabled, just test crypto,
because the TLS code needs that to generate ephemeral ECDH keys but this is
not tracked properly (the ephemeral ECDH code is only gated on having ECDH).
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Do not assume that tests with !PSA_WANT_ALG_TLS12_PRF will not
be executed in AnalyzeCoverage task.
Signed-off-by: Valerio Setti <valerio.setti@nordicsemi.no>
MBEDTLS_PSA_KEY_BUFFER_MAX_SIZE is only used in tests so it should
not be defined in a public header such as "crypto_extra.h".
"psa_crypto_helpers.h" is a better option.
Signed-off-by: Valerio Setti <valerio.setti@nordicsemi.no>
PSA_KEY_EXPORT_RSA_KEY_PAIR_MAX_SIZE() is not defined when there
is no MBEDTLS_PSA_CRYPTO_CLIENT so we need this guard to
define MBEDTLS_TEST_STATIC_KEY_SLOTS_SUPPORT_RSA_[2048/4096].
Signed-off-by: Valerio Setti <valerio.setti@nordicsemi.no>
- MBEDTLS_TEST_STATIC_KEY_SLOTS_SUPPORT_RSA_[2048/4096] are always
defined because they are only used in test_suite_psa_crypto
tests.
- MBEDTLS_TEST_ALLOW_RSA_4096 was renamed as
MBEDTLS_TEST_PK_ALLOW_RSA_KEY_PAIR_4096 because this is only used in
PK related test suites.
Signed-off-by: Valerio Setti <valerio.setti@nordicsemi.no>
This value should be:
- OK for all EC/FFDH key pairs/public keys;
- OK for all supported public RSA keys;
- OK for RSA key pairs up to 2048 bits;
- FAIL for RSA key pairs above 2048 bits.
Signed-off-by: Valerio Setti <valerio.setti@nordicsemi.no>
Intentionally set MBEDTLS_PSA_STATIC_KEY_SLOT_BUFFER_SIZE slightly
smaller than the maximum RSA key pair size for an RSA key of 4096
bits. Also add a test in test_suite_psa_crypto to verify this
condition.
Signed-off-by: Valerio Setti <valerio.setti@nordicsemi.no>
Disable all MBEDTLS_xxx symbols (keeping only the relevant ones enabled)
when building the main library.
Signed-off-by: Valerio Setti <valerio.setti@nordicsemi.no>
This commit also fixes related errors in test suites. In all
cases those failures are related to the use of raw keys whose
size cannot be determined a-priori.
Signed-off-by: Valerio Setti <valerio.setti@nordicsemi.no>
With the `depend.py` using the crypto config the `PBKDF2_HMAC` can be
enabled so thest cases can be run.
The equivalence (synonym) between `PSA_WANT_ALG_RSA_PSS_ANY_SALT` and
`PSA_WANT_ALG_RSA_PSS` is now done properly, the test can be run.
Dependencies for `MBEDTLS_CIPHER_PADDING_PKCS7' has been updated and
now it can be actually disabled.
Signed-off-by: Gabor Mezei <gabor.mezei@arm.com>
This commit fixes a linking error with psa_sim_crypto_server.c where
mbedtls_test_hook_error_add was undefined.
Signed-off-by: Harry Ramsey <harry.ramsey@arm.com>
In 4.x the semantics of MBEDTLS_PSA_CRYPTO_C and
MBEDTLS_PSA_CRYPTO_CLIENT are different compared with 3.6.
Where this is a problem, make guards more version-specific so that we
are guarding with MBEDTLS_PSA_CRYPTO_C for 3.6 and
MBEDTLS_PSA_CRYPTO_CLIENT for 4.x, keeping each branch the same as it
was formerly.
Signed-off-by: David Horstmann <david.horstmann@arm.com>
Allow tests/src and include/src to be Mbed TLS version-agnostic by:
* Sometimes accepting both an MBEDTLS_ and a PSA_ config option
* Sometimes using the version number to gate alternatives
Signed-off-by: David Horstmann <david.horstmann@arm.com>
The PSA and MbedTLS options can switch the same functionality separately
so add a check to ensure the cpnsistency.
Signed-off-by: Gabor Mezei <gabor.mezei@arm.com>
This commit fixes the error test suite to depend on MBEDTLS_ERROR_C
rather than MBEDTLS_ERRORCOMMON_C.
Signed-off-by: Harry Ramsey <harry.ramsey@arm.com>
We cannot add the equivalent of
cmake_package_install yet as the build in
tf-psa-crypto still references some headers
in ./include/mbedtls like mbedtls_config.h.
We cannot add the equivalent of
cmake_subproject yet as currently only
the case of the Mbed TLS parent project is
supported.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
Let the projects using some CMakeLists.txt files
without the top CMakeLists.txt define the variables
that the top CMakeLists.txt set for all.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
Without quotes, when the variable is empty, the shell will see three
tokens: [, -n, ]. After skipping ] as usual, it will see a single token,
so it will consider it not as command, but a string to be tested for "is
it empty", and since "-n" is not empty, the command will return true.
With quotes it see 4 tokens: [, -n, <empty string>, ] and interprets -n
as desired.
Signed-off-by: Manuel Pégourié-Gonnard <manuel.pegourie-gonnard@arm.com>
Replace obvious additions of an `MBEDTLS_ERR_xxx` constant by a call to
`MBEDTLS_ERROR_ADD`.
Skip `case` statements since `MBEDTLS_ERROR_ADD(pp_constant)` is not a
preprocessor constant.
This commit does not replace additions split over lines. Those will be
handled in a subsequent commit.
```
git ls-files '*.h' '*.c' '*.function' '*.data' |
xargs perl -i -pe '
next if /\bcase\b/;
s/\b(MBEDTLS_ERR_\w+)\s*\+\s*(\w+)\b/MBEDTLS_ERROR_ADD($1, $2)/g;
s/\b(\w+)\s*\+\s*(MBEDTLS_ERR_\w+)\b/MBEDTLS_ERROR_ADD($1, $2)/g'
```
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
This aligns development and 3.6 in preparation for moving to the
framework repo.
Signed-off-by: Manuel Pégourié-Gonnard <manuel.pegourie-gonnard@arm.com>
Use config_adjust_test_accelerators.h
to adjust the configuration needed for
test_psa_crypto_drivers all.sh component
with PSA_CRYPTO_CONFIG enabled.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
As we move to PSA_CRYPTO_CONFIG always on,
the way to configure the build with both
the builtin drivers and the transparent test
drivers (that are wrappers around the
builtin drivers) cannot be done through the
MBEDTLS_USER_CONFIG_FILE mechanism anymore.
With this mechanism and PSA_CRYPTO_CONFIG
enabled, the PSA_ACCEL_ macros are defined
before including config_adjust_legacy_from_psa.h
and the builtin implementations are removed.
Thus, we will rather define the PSA_ACCEL_
just after the inclusion of
config_adjust_legacy_from_psa.h through the
dedicated config_adjust_test_accelerators.h
header introduced by this commit.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
TLS only supports actual restartable ECDH with the legacy code that's going
away, not with the MBEDTLS_USE_PSA_CRYPTO code that's becoming the only
variant. This leaves a few test cases that validate restartable ECDH in TLS
as desirable, but not currently able to pass.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
For each ignore list entry, link to a GitHub issue for its resolution,
except for a few configurations which there is a good reason to leave
uncovered.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
For each ignore list entry, link to a GitHub issue for its resolution,
except for ssl-opt Valgrind tests which we never intend to run on the CI.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Ignore certain test cases which either should not be generated or should be
executed. For each ignore list entry, link to a GitHub issue whose
definition of done includes removing the entry.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Some negative tests involving unsupported mechanisms are executed, because
they're testing what happens if the mechanism is unsupported. Refine the
ignore list for `test_suite_psa_crypto_generate_key.generated` and
`test_suite_psa_crypto_op_fail.generated` accordingly.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
This clears more than half of the test cases that are not executed.
This also captures a few negative test cases that are executed.
Subsequent commits will refine the filtering.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Now the output of
grep '^[^ {}#]' tests/scripts/all-*.sh | grep -v '()'
is clean, with the only results being:
- shopt -s extglob which is needing for parsing the rest
- usage message
Signed-off-by: Manuel Pégourié-Gonnard <manuel.pegourie-gonnard@arm.com>
Currently, many test cases are not executed. A follow-up pull request will
take care of that. In the meantime, continue allowing partial test coverage.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Move `collect_test_cases.py` (split from `check_test_cases.py`),
`check_test_cases.py`, and `outcome_analysis.py` (split from
`analyze_outcomes.py`) to the framework repository.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
The only occurrences found by git grep were the definitions and the
comment saying these are kept for backwards compatibility.
Signed-off-by: Manuel Pégourié-Gonnard <manuel.pegourie-gonnard@arm.com>
Don't interleave defining functions with running some code.
The only exception is calling shopt, which needs to come first as it
affects how the following function definitions are parsed.
Signed-off-by: Manuel Pégourié-Gonnard <manuel.pegourie-gonnard@arm.com>
This commit fixes failing psasim tests by ensuring that
MBEDTLS_VERSION_C is defined before attempting to access version
information.
Signed-off-by: Harry Ramsey <harry.ramsey@arm.com>
This commit moves version.h back into /include/mbedtls from
tf-psa-crypto/drivers/builtin/include. This commit also changes the
necessary build files and scripts to generate version_features.c
Signed-off-by: Harry Ramsey <harry.ramsey@arm.com>
Place the code of outcome analysis (auxiliary functions, tasks, command line
entry point) into a separate module, which will be moved to the
version-independent framework repository so that it can be shared between
maintained branches. Keep the branch-specific list of driver components and
ignore lists in the per-repository script.
We keep the executable script at `tests/scripts/analyze_outcomes.py`. It's
simpler that way, because that path is hard-coded in CI scripts.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Always have tasks_list be a list, not potentially some fancier iterable.
Bypass mypy's somewhat legitimate complaint about REFERENCE and DRIVER in
task_class: they could potentially be instance attributes, but we rely on
them being class attributes. Python does normally guarantee their existence
as class attributes (unless a derived class explicitly deletes them), but
they could be overridden by an instance attribute; that's just something
we don't do, so the class attribute's value is legitimate. We can't
expect mypy to know that, so work around its complaint.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Use different names for task name, a task class and a task instance. The
interpreter doesn't care, but it's less confusing for both humans and type
checkers.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
The ignore list for coverage only has two test cases out of ~10000 that are
currently reported as not executed. This is a drop in the sea and not
useful. Remove them so that the class can be used generically. A follow-up
will construct a comprehensive ignore list.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Move the test case collection code out of check_test_cases.py and into its
own module. This allows outcome analysis to depend only on the new module
and not on check_test_cases.py.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Now that dtls_server doesn't print "error" when it receives stray messages
while it's waiting for a second handshake, have the tests fail if "error" is
printed for some other reason.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Stop testing configurations without PSA (MBEDTLS_PSA_CRYPTO_C or at least
MBEDTLS_PSA_CRYPTO_CLIENT). No future release from this branch will support
such configurations, and we can no longer build the SSL sample programs
without psa_crypto_init.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
When building with `configs/config-suite-b.h`, the SSL I/O buffer size is
1024 bytes. Experimentally, this isn't quite enough for the test certificate
that we use: the server aborts the handshake with
`MBEDTLS_ERR_SSL_BUFFER_TOO_SMALL` raised from
`mbedtls_ssl_write_certificate()`. State an ad hoc minimum output buffer
size to skip testing `ssl_server` in `config-suite-b`.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
dtls_client connects to "localhost", which is usually IPv6 on modern
systems. On our CI, $OPENSSL is OpenSSL 1.0.2g which doesn't support IPv6.
Pitching dtls_client against $OPENSSL works on the CI at the moment, but
only because the CI runs in Docker with default network settings which has
IPv6 disabled. This would stop working if we changed the CI's Docker setup,
and the test case is likely to fail on a developer machine. So switch the
test case to using $OPENSSL_NEXT (which is a version of OpenSSL that has
IPv6 support).
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
GnuTLS 3.4.x doesn't allow repeated `-p PORT` arguments.
OpenSSL 1.0.2 has different logs. For TLS 1.2 test cases, use a line that
is present in logs from OpenSSL 1.0.2g, 3.3.0 and presumably all versions
in between.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Test against both OpenSSL and GnuTLS.
Don't use a proxy. It's not particularly useful here, and would complicate
figuring out port numbers.
Clean up compile-time requirements dtls_server.c: any certificate-based key
exchange is ok, so don't insist on built-in RSA.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Test ssl_fork_server with both TLS 1.2 and TLS 1.3.
Test against both OpenSSL and GnuTLS.
In the server, flush more often. Otherwise, when stdout is redirected to a
file, the server gets killed before it writes important information, such as
the logs that we expect in the test cases.
In the server, only write output for 10 seconds, not 100. That's enough time
to start concurrent clients if desired. 100 seconds causes ssl-opt to take a
very long time when the client actually listens to the whole input (which
`gnutls-cli` does, but not `openssl s_client`).
Clean up compile-time requirements in ssl_fork_server.c: any certificate-based
key exchange is ok, so don't insist on built-in RSA.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Test ssl_pthread_server with both TLS 1.2 and TLS 1.3.
Test against both OpenSSL and GnuTLS.
In the server, flush more often. Otherwise, when stdout is redirected to a
file, the server gets killed before it writes important information, such as
the logs that we expect in the test cases.
Clean up compile-time requirements in ssl_pthread_server.c: any certificate-based
key exchange is ok, so don't insist on built-in RSA.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Test ssl_server with both TLS 1.2 and TLS 1.3.
Test against both OpenSSL and GnuTLS.
Clean up compile-time requirements in ssl_server.c: any certificate-based
key exchange is ok, so don't insist on built-in RSA.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Test against both OpenSSL and GnuTLS.
Don't use a proxy. It's not particularly useful here, and would complicate
figuring out port numbers.
Clean up compile-time requirements in dtls_client.c: any certificate-based
key exchange is ok, so don't insist on built-in RSA.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Test ssl_client1 with both TLS 1.2 and TLS 1.3.
Test against both OpenSSL and GnuTLS.
Clean up compile-time requirements in ssl_client1.c: any certificate-based
key exchange is ok, so don't insist on built-in RSA.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Work on the stringly typed KNOWN_TASKS by classes for each category
of tasks, with a structure that matches the behavior.
This commit migrates the data for driver-vs-reference analysis and gets rid
of the transitional code that was using the old form of the data.
No intended behavior change.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
