Commit Graph

9366 Commits

Author SHA1 Message Date
danh-arm
c64d847127
Merge pull request #3358 from ronald-cron-arm/tool_versions-2.16
Backport 2.16: Add output of make, cmake and python3 versions
2020-05-27 17:38:32 +01:00
Janos Follath
2c6089f52a
Merge pull request #3362 from paul-elliott-arm/fix-contributing-2.16
Backport 2.16: Fix contributing link to changelog howto
2020-05-27 07:44:33 +01:00
Paul Elliott
98aa53d0e4 Change Changelog link to point at Changelog readme
Make the contributing document link to how to create a changelog rather
than just linking to the Changelog itself. Backport to 2.16

Signed-off-by: Paul Elliott <paul.elliott@arm.com>
2020-05-26 16:27:10 +01:00
Ronald Cron
0665677265 Add output of python3 version
Add output of python3 version to output_env.sh.
Added in addition to the version of `python` as some
project's scripts try both executable names.

Signed-off-by: Ronald Cron <ronald.cron@arm.com>
2020-05-26 14:20:12 +02:00
Ronald Cron
2eefeeebdb Add output of make and cmake versions
Add output of make and cmake versions to output_env.sh.
That way we can see their versions in the CI.

Signed-off-by: Ronald Cron <ronald.cron@arm.com>
2020-05-26 14:19:40 +02:00
Jonas
c98ca37cb9 Add test cases to check rng failure
Signed-off-by: Jonas <jonas.lejeune4420@gmail.com>
2020-05-25 13:55:09 +02:00
Jonas
6645fd31e7 Fix potential memory leak in EC multiplication
Signed-off-by: Jonas <jonas.lejeune4420@gmail.com>
2020-05-25 13:53:15 +02:00
Gilles Peskine
8b34fef0a8
Merge pull request #2710 from k-stachowiak/unified-exit-in-examples-2.16
Backport 2.16: Unify the example programs' termination
2020-05-12 10:46:36 +02:00
Gilles Peskine
543e6c5a7e Add changelog entry file
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-05-11 22:17:08 +02:00
k-stachowiak
ff8ae11159 Adjust the zeroization test to the new length of zeroize.c 2020-05-11 22:16:42 +02:00
Janos Follath
b58d07f892
Merge pull request #3307 from yanesca/update_contributing_discussion_link-2.16
Backport 2.16: Update link in contributing guide
2020-05-06 16:04:51 +01:00
Janos Follath
01836bffe0 Update link in contributing guide
The link pointed to the website, this information is out of date, the
correct place to start discussions is the mailing list.

Signed-off-by: Janos Follath <janos.follath@arm.com>
2020-05-06 12:45:42 +01:00
Gilles Peskine
0f593416e0
Merge pull request #3247 from gilles-peskine-arm/travis-rationalize-2.16
Backport 2.16: Rationalize Travis builds
2020-05-04 12:10:55 +02:00
danh-arm
21f2ba54f7
Merge pull request #3286 from gilles-peskine-arm/config-full-clarify-2.16
Backport 2.16: Clarify that the full config enables everything that can be tested together
2020-05-04 10:29:41 +01:00
Gilles Peskine
43e068c9eb Travis: do both a Cortex-A build and a Cortex-M0+ build
The Cortex-A build is similar to Debian armel. The Cortex-M0+ is a
handy point of comparison for code size. Put that one last so that
it's easy to find in the log.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-04-30 23:19:47 +02:00
Gilles Peskine
e6c0c7d369 all.sh: on arm builds (GCC or Arm Compiler), show the code size
Just show the code size in the logs, for human consumption.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-04-30 23:19:44 +02:00
Gilles Peskine
dac156b631 all.sh: add a Cortex-M0+ build
It's pretty fast and adds a little variety.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-04-30 23:19:44 +02:00
Gilles Peskine
fcccfbc64a all.sh: build_arm_none_eabi_gcc: do optimize
Otherwise the bignum assembly code is not used.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-04-30 23:18:24 +02:00
Gilles Peskine
a3c6c8a945 Rename --arm-gcc-prefix to --arm-none-eabi-gcc-prefix
This is supposed to be for GCC (or a compiler with a compatible
command line interface) targeting arm-none-eabi, so name it
accordingly.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-04-30 18:43:21 +02:00
Gilles Peskine
77b1f3051a all.sh: reorganize testing around deprecated features
build_deprecated combined the testing of deprecated features, and
testing of the build without deprecated features. Also, it violated the
component naming convention by being called build_xxx but running tests.

Replace it by:
* test_default_no_deprecated: check that you can remove deprecated
  features from the default build.
* test_full_deprecated_warning: check that enabling DEPRECATED_WARNING
  doesn't cause any warning from our own code.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-04-30 18:24:08 +02:00
Gilles Peskine
1e8cafcd3d Exclude DEPRECATED_WARNING in config full
In the full config, don't set MBEDTLS_DEPRECATED_WARNING. This is debatable:
the full config does not enable deprecated features in this branch, so
MBEDTLS_DEPRECATED_WARNING is compatible with the other features.
Exclude it to keep LTS branches closer to development.

In any case, baremetal and full should have the same settings regarding
deprecated features, so don't do anything about DEPRECATED_xxx in baremetal.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-04-30 09:04:40 +02:00
Gilles Peskine
313bb50db3 Strict C99: check it in the full config
Ensure that there is a build with -pedantic in the full config, not
just in "exotic" configurations.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-04-28 15:02:51 +02:00
Gilles Peskine
d135bbd7a1 Strict C99: don't use extremely large string literals
Don't use string literals that are longer than 4095 bytes, which is
the minimum that C99 compilers are required to support. Compilers are
extremely likely to support longer literals, but `gcc -std=c99 -pedantic`
complains.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-04-28 15:02:50 +02:00
Gilles Peskine
5ac979309b Strict C99: make sure that fileno() is declared
<stdio.h> only declares the non-ISO-C function fileno() if an
appropriate POSIX symbol is defined or if using a compiler such as GCC
in non-pedantic mode. Define the appropriate POSIX symbol.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-04-28 15:02:50 +02:00
Gilles Peskine
971e5e9f98 Strict C99: don't use a signed* when an unsigned* is expected
It works in practice on almost every platform, given that we're only
using the wrong type in cases where the value is guaranteed to stay
within the value bits of a signed int. But even in this case it may or
may not be strictly conforming. Anyway `gcc -std=c99 -pedantic`
rejects it.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-04-28 15:02:50 +02:00
Gilles Peskine
1e557b7f17 Fix build failure with MBEDTLS_PLATFORM_NV_SEED_ALT
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-04-28 15:02:50 +02:00
Gilles Peskine
df4f7c1274 Document and fix the MBEDTLS_xxx_ALT logic for the full config
The intended logic around MBEDTLS_xxx_ALT is to exclude them from full
because they require the alternative implementation of one or more
library functions, except that MBEDTLS_PLATFORM_xxx_ALT are different:
they're alternative implementations of a platform function and they
have a built-in default, so they should be included in full. Document
this.

Fix a bug whereby MBEDTLS_PLATFORM_xxx_ALT didn't catch symbols where
xxx contains an underscore. As a consequence,
MBEDTLS_PLATFORM_GMTIME_R_ALT and MBEDTLS_PLATFORM_NV_SEED_ALT are now
enabled in the full config. Explicitly exclude
MBEDTLS_PLATFORM_SETUP_TEARDOWN_ALT because it behaves like the
non-platform ones, requiring an extra build-time dependency.
Explicitly exclude MBEDTLS_PLATFORM_NV_SEED_ALT from baremetal
because it requires MBEDTLS_ENTROPY_NV_SEED, and likewise explicitly
unset it from builds that unset MBEDTLS_ENTROPY_NV_SEED.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-04-28 15:02:50 +02:00
Gilles Peskine
4796c19516
Merge pull request #3283 from gilles-peskine-arm/changelog-20200421-2.16
2.16 only: Add missing changelog entry for PRs since 2.16.6
2020-04-28 13:10:40 +02:00
Manuel Pégourié-Gonnard
899f63805a
Merge pull request #3205 from irwir/fix_vcxproj-2.16
[backport 2.16] Finalize fixing bugs in .vcxproj files
2020-04-28 13:03:30 +02:00
Gilles Peskine
08fc4aaec0 Add changelog entry for #3197: x509_crt max_pathlen int overflow fix
Backport of #3192

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-04-28 12:13:10 +02:00
Gilles Peskine
d6189e115d Sort lists for easier maintenance
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-04-28 11:50:06 +02:00
Gilles Peskine
108ebf2b26 Explain the rationale for inclusion in "full" and "baremetal"
Remove the duplicated, and often out-of-date, list in the comments.
Instead explain in a comment, and have a single copy of the list which
is in the code.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-04-28 11:50:06 +02:00
Gilles Peskine
df5614ef1d Enable X509_ALLOW_EXTENSIONS_NON_V3 in config full
Enable MBEDTLS_X509_ALLOW_EXTENSIONS_NON_V3 in the full config. There's
no reason to keep it out. We weren't testing it at all on the CI.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-04-28 10:23:59 +02:00
Gilles Peskine
989d05ffd8 Remove obsolete options from config.pl
These options haven't existed for a long time.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-04-28 10:23:59 +02:00
Gilles Peskine
eb59af14b0 Travis: rebalance the Linux builds
Make the running time more even.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-04-27 12:52:23 +02:00
Gilles Peskine
2ccdf897d8 Travis: Windows: target Visual Studio 2017
Only the Visual Studio 2017 toolset is currently preinstalled on Travis.
Use this, instead of our solution's default which is VS 2010.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-04-27 09:14:42 +00:00
Gilles Peskine
e335816bf6 Travis: add a build with Visual Studio on Windows
Travis now offers a Windows environment. Do a build with Visual
Studio. This brings diversity into the Travis CI which otherwise only
uses GCC and Clang.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-04-27 09:14:42 +00:00
Gilles Peskine
468adcc20b Add a simple build script for Windows with Visual Studio
Keep it simple and mostly non-parametrizable for now.
A path to Visual Studio 2017 is hard-coded.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-04-27 09:14:42 +00:00
Gilles Peskine
7c8c9c9531 check-files: support Windows .bat files
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-04-27 09:14:42 +00:00
Gilles Peskine
317e7a19eb Travis: add a build on macOS
Just do the default build with Clang and run the unit tests. The
objective is to have one build on a Unix-like platform other than
Linux.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-04-27 09:14:42 +00:00
Gilles Peskine
1c2c1bd7b0 Travis: add a baremetal build
Add a baremetal build to Travis, to catch inadvertent dependencies on
platform functions.

The exact choice of target platform doesn't matter for this purpose.
Pick one that's present in all.sh, that uses a compiler that's
available in the Travis build environment (Ubuntu 16.04), and that
happens to be close to the Debian "armel" distribution.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-04-27 09:14:42 +00:00
Gilles Peskine
025747bd81 Travis: call all depends_* tests
Call all.sh to run all the available test_depends_* components. This
adds a run of depends-hashes.pl and depends-pkgalgs.pl.

Keep invoking test-ref-configs.pl rather than via all.sh so that it
doesn't run with ASan. This saves some time and ASan there doesn't
turn up much more than in the full config.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-04-27 09:14:42 +00:00
Gilles Peskine
4b29e7d682 Travis: install gnutls-bin for all jobs
Some jobs don't actually test against GnuTLS, but all.sh checks its
presence in all test jobs, so it needs to be installed regardless.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-04-27 09:14:42 +00:00
Gilles Peskine
9a0e0f2655 Travis: replace "default configuration by "full configuration"
For the one long job with ASan, use the full configuration.

We get more coverage this way, at the cost of a slightly longer
runtime which we can afford since the "enumerated configurations" job
is slower.

Add a default-configuration build to the "basic checks" job. This job
is fairly quick (no ASan, no SSL testing).

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-04-27 09:14:42 +00:00
Gilles Peskine
5a711aba01 Travis: call all.sh for the default-configuration build
This way anything we change in all.sh, such as adding tests for
programs/*/*, will be reflected here.

The build now uses GCC instead of Clang, which doesn't make much
difference in practice. The build now enables ASan and UBSan.

The tests now run compat.sh and ssl-opt.sh fully.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-04-27 09:14:42 +00:00
Gilles Peskine
ec4c956372 Travis: call all.sh for sanity checks
Different releases have different sets of sanity checks. Keep the list
in one place, namely all.sh.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-04-27 09:14:42 +00:00
Gilles Peskine
9d0cee0d08 Travis: run Pylint
Declare an explicit Python version. Pick 3.5 which is the default
version on Ubuntu 16.04. This is necessary on Travis to have a working
pip for Python 3.

Install Pylint 2.4.4. There's nothing special about this version, it's
just the latest version.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-04-27 09:14:42 +00:00
Gilles Peskine
034ee2b848 Travis: move doxygen dependencies into the "sanity" job
Only this job uses doxygen and graphviz.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-04-27 09:14:42 +00:00
Gilles Peskine
7c4912d758 Travis: split the build into three parallel jobs
Split the build between:
* Basic checks
* A build in the default configuration with extensive tests
* Builds in other configurations with less testing

The intent is to have one shorter job with basic tests, and two longer
jobs that take roughly the same amount of time (split as evenly as
possible while keeping an easy-to-understand separation).

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-04-27 09:14:42 +00:00
Gilles Peskine
f2cf40ba1a Travis: don't test with both gcc and clang
In practice, we hardly ever get different outcomes, so there is no
gain in running tests with different compilers.

Experimentally, with the builds and tests we currently do and with the
compiler versions on a Travis Ubuntu 16.04, gcc jobs are significantly
faster than clang jobs (13 min vs 24 min). So use gcc.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2020-04-27 09:14:42 +00:00