The theory here is that we're already doing sufficient validation key loading, and this is purely duplicative.
Note that there's at least _some_ validationg that was previously occurring only ECDH, the LowOrderPublic check that can be seen in wycheproof.
* add support for centos9-fips
Requires a variety of new FIPS constraints on our tests, including the
addition of rsa_encryption_supported
* review comments
* add unsafe_skip_rsa_key_validation
This allows users to skip RSA key validation when calling
load_pem_private_key, load_der_private_key, and
RSAPrivateNumbers.private_key. This is a significant performance
improvement but is **only safe if you know the key is valid**. If you
use this when the key is invalid OpenSSL makes no guarantees about what
might happen. Infinite loops, crashes, and all manner of terrible things
become possible if that occurs. Beware, beware, beware.
* Apply suggestions from code review
Co-authored-by: Alex Gaynor <alex.gaynor@gmail.com>
* remove unneeded variable
Co-authored-by: Alex Gaynor <alex.gaynor@gmail.com>
* Drop supported for older LibreSSL
3.1 covers the oldest version used by versions of Alpine and OpenBSD that are supported by their upstreams.
* Remove tests that are now unused
* Update CHANGELOG.rst
Co-authored-by: Paul Kehrer <paul.l.kehrer@gmail.com>
Co-authored-by: Paul Kehrer <paul.l.kehrer@gmail.com>
* Speed up RSA tests in 3.0.0
RSA_check_key is slower in OpenSSL 3.0.0 due to improved
primality checking. In normal use this is unlikely to be a problem
since users don't load new keys constantly, but we do in our tests.
This adds some private flags to allow skipping those checks for
performance reasons.
On my laptop with this patch it takes 16s to run test_rsa.py. The
previous commit takes 72s.
* black
* different approach
* skip rsa key checks in wycheproof
wycheproof's tets don't rely on broken keys
* rsa type hinting
* remove unused import
* missed return type
* type fixes
* ignores no longer required
* black gets me every time
* Update src/cryptography/hazmat/backends/openssl/rsa.py
Co-authored-by: Alex Gaynor <alex.gaynor@gmail.com>
Co-authored-by: Alex Gaynor <alex.gaynor@gmail.com>
* GCM IV size limits
OpenSSL 3.0.0 is going to enforce these size limits so we might as well
put them in now.
* fix the tests
* black
* these cases can't happen if we're limiting IV size already
* Refs #5075 -- use x448_test.json from wycheproof
* Fixed test
* crypto libraries from people who can't math, it's fine
* Skip teh weirdo 57 byte public keys
* Remove irrelevant DHBackend test conditions
DHBackend provides functions for plain finite-field Diffie-Hellman.
X25519 and X448 are their own algorithms, and Ed25519 and Ed448 aren't
even Diffie-Hellman primitives.
* Add missing backend support checks.
Some new AES and EC tests did not check for whether the corresponding
mode or curve was supported by the backend.
* Add a DummyMode for coverage
