aboutsummaryrefslogtreecommitdiffstats
path: root/src/pubkey/gost_3410
Commit message (Collapse)AuthorAgeFilesLines
* Replace PointGFp::check_invaraints, which would either return silentlylloyd2010-03-191-8/+4
| | | | | | | | | | | or throw an exception, with PointGFp::on_the_curve, which returns a bool. Update callers. This showed several cases where check_invaraints was being called multiple times, for instance when decoding a point with OS2ECP, check_invaraints was called; many callers of OS2ECP would then call check_invaraints again on the same object.
* Add a couple of verification tests for GOST 34.10lloyd2010-03-161-3/+16
| | | | | | | | | | | Generating the test vectors found yet another inane (and, of course, undocumented) behavior in the GOST implementation included in OpenSSL; it treats the hash inputs as little endian. Just out of curiousity, I checked RFC 5832, which supposedly specifies this algorithm; not a peep about endian conversions. The more I deal with standards coming out of the CryptoPro people, the less confidence I have in them.
* Remove iostream/stdio includeslloyd2010-03-131-3/+0
|
* Fix GOST 34.10 pub key loading (uses little endian format, what the fsck?)lloyd2010-03-131-6/+25
|
* Fix GOST, wasn't getting found in enginelloyd2010-03-132-4/+4
|
* Remove the now no-op classes PK_Encrypting_Key,lloyd2010-03-081-4/+2
| | | | | PK_Decrypting_Key, PK_Signing_Key, PK_Verifying_with_MR_Key, and PK_Verifying_wo_MR_Key.
* Constify sign and verify opslloyd2010-03-052-6/+5
|
* Remove sign and verify ops from key typeslloyd2010-03-052-43/+0
|
* Rename PK_Ops::Signature_Operation to PK_Ops::Signaturelloyd2010-03-051-1/+1
| | | | Rename PK_Ops::KA_Operation to PK_Ops::Key_Agreement
* Add verification ops for all signature key typeslloyd2010-03-052-3/+62
|
* Remove the sign() operation from the public key objects, totally replacedlloyd2010-03-052-49/+0
| | | | | | by using the ops. Add real ECDSA test vectors (two found in ANSI X9.62)
* Add signature generation operation classes. Remove sign() fromlloyd2010-03-052-1/+57
| | | | | | PK_Signing_Key, though for the moment the class remains because there are a few pieces of code that use it to detect if signatures are supported, or for passing to functions in look_pk
* Fix GOST pubkey encoding when x.bytes() != y.bytes()lloyd2010-03-041-1/+1
|
* Fix exception textlloyd2010-03-041-1/+1
|
* Quite the hack, here.lloyd2010-03-041-0/+3
| | | | | | | | | | | | | | | | | GOST 34.10 public keys use a funky encoding. There is no standard for PKCS #8 format private keys, so the obvious choice is to act exactly the same as ECDSA/ECDH (following the rule of thumb that if you're going to make up a random non-standard thing, at least try to copy something that's standard for something else). However the public key encoding uses a weird scheme for encoding the OID in the algorithm identifier, which we don't want to use for the PKCS #8 encoding. Add a new function to Private_Key, pkcs8_algorithm_identifier, which by default just calls algorithm_identifier(). However GOST_3410_PrivateKey overrides it, and calls EC_PublicKey::algorithm_identifier(), basically skipping over the virtual function hierarchy, so it doesn't pick up the funky format from the public key's version of algorithm_identifier().
* Fix GOST 34.10 pubkey encodinglloyd2010-03-042-3/+13
|
* Remove load hooks from ECC classes, unusedlloyd2010-03-041-1/+8
|
* Add similar decoding constructors to the private keyslloyd2010-03-041-5/+4
|
* Remove X509_Decoder. Fix GOST-34.10 DER constructor (was default to normal ECC)lloyd2010-03-042-41/+14
|
* Add a new constructor to each public key algorithm (only the publiclloyd2010-03-041-15/+19
| | | | | | | keys so far, private keys not changed) that takes an AlgorithmIdentifier and a MemoryRegion<byte>&. This performs the X.509 decoding. It is not possible anymore to create uninitialized PK objects.
* Add a new function to public key x509_subject_public_key which returnslloyd2010-03-042-33/+10
| | | | | what x509_encoder()->key_bits() used to return. This is much simpler than using the explicit encoder objects. Remove X509_Encoder entirely.
* Small cleanupslloyd2010-03-022-22/+13
|
* Add some simple constructors to the EC_ base key types to simplifylloyd2010-03-022-28/+5
| | | | the various implementations
* Remove auto_ptr from ECC key typeslloyd2010-03-022-167/+25
|
* Remove a fairly useless member of EC_PublicKey that was only used forlloyd2010-03-022-44/+1
| | | | | | | handling ImplicitCA ECDSA keys in the CVC code. Currently dealt with in CVC by simply commenting out the calls - CVC is already pretty broken and I'd much rather have ECC sane and under control and CVC totally broken than ECC remaining in its current state.
* Kill get_EC_Dom_Pars_by_oidlloyd2010-03-011-3/+1
|
* Clean up EC_Domain_Paramslloyd2010-03-011-5/+3
|
* Move contents of gfpmath to numbertheory. Adjust dependencies.lloyd2010-02-251-1/+0
|
* Convert PointGFp::get_affine_{x,y} to return just the BigInt valuelloyd2010-02-251-5/+5
|
* Convert 3-arg constructoor of PointGFp to take BigInts instead oflloyd2010-02-251-2/+1
| | | | | | GFpElements. Clean up OS2ECP
* Remove PointGFp::mult_this_securelloyd2010-02-241-5/+4
|
* Remove debug printlloyd2010-02-241-5/+0
|
* Of _course_, GOST 34.10 uses a non-standard X.509 encoding, and _of course_lloyd2009-12-122-5/+105
| | | | | | | | it's ridiculously poorly thought out. The PKCS #8 format isn't documented anywhere I can find, so I'm declaring the standard GOST-34.10-2001 to be identical to the ECDSA/ECDH format, thus defining the problem away. :)
* Add an implementation of GOST 34.10-2001 as described inlloyd2009-11-193-0/+434
draft-dolmatov-cryptocom-gost34102001-06 Known problem: GOST's X.509 (and PKCS #8?) formats are different from ECDSA. ECDSA uses compressed points, GOST uses a completely raw pair of points (with, OF COURSE, no leading uncompressed pair identifier, because using something that already exists would just be too much).
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-15 22:41:52 +0000