Andy Polyakov [Mon, 10 Jun 2013 20:34:06 +0000 (22:34 +0200)]
aesni-sha256-x86_64.pl: harmonize with latest sha512-x86_64.pl.
Andy Polyakov [Mon, 10 Jun 2013 20:30:34 +0000 (22:30 +0200)]
sha1-x86_64.pl: add AVX2+BMI code path.
Andy Polyakov [Mon, 10 Jun 2013 20:29:01 +0000 (22:29 +0200)]
sha512-x86_64.pl: upcoming-Atom-specific optimization.
Andy Polyakov [Mon, 10 Jun 2013 20:26:53 +0000 (22:26 +0200)]
sha[256|512]-586.pl: add more SIMD code paths.
Andy Polyakov [Mon, 10 Jun 2013 20:25:12 +0000 (22:25 +0200)]
ghash-x86_64.pl: add Haswell performance data.
Andy Polyakov [Mon, 10 Jun 2013 20:20:46 +0000 (22:20 +0200)]
x86cpuid.pl: fix extended feature flags detection.
Dr. Stephen Henson [Sun, 5 May 2013 12:34:03 +0000 (13:34 +0100)]
Fix PSS signature printing.
Fix PSS signature printing: consistently use 0x prefix for hex values for
padding length and trailer fields.
Dr. Stephen Henson [Fri, 3 May 2013 11:31:47 +0000 (12:31 +0100)]
Reencode with X509_CRL_ctx_sign too.
Adam Langley [Tue, 23 Apr 2013 18:36:06 +0000 (14:36 -0400)]
Ensure that, when generating small primes, the result is actually of the
requested size. Fixes OpenSSL #2701.
This change does not address the cases of generating safe primes, or
where the |add| parameter is non-NULL.
Conflicts:
crypto/bn/bn.h
crypto/bn/bn_err.c
Adam Langley [Tue, 23 Apr 2013 16:13:51 +0000 (12:13 -0400)]
Ensure that x**0 mod 1 = 0.
Adam Langley [Mon, 3 Jun 2013 19:45:11 +0000 (15:45 -0400)]
Add volatile qualifications to two blocks of inline asm to stop GCC from
eliminating them as dead code.
Both volatile and "memory" are used because of some concern that the compiler
may still cache values across the asm block without it, and because this was
such a painful debugging session that I wanted to ensure that it's never
repeated.
Ben Laurie [Tue, 4 Jun 2013 14:05:18 +0000 (15:05 +0100)]
Remove added ;.
Ben Laurie [Tue, 4 Jun 2013 14:14:18 +0000 (15:14 +0100)]
Missing prototypes.
Andy Polyakov [Sat, 1 Jun 2013 07:58:07 +0000 (09:58 +0200)]
crypto/bn/bn_exp.c: SPARC portability fix.
Andy Polyakov [Sat, 25 May 2013 17:23:09 +0000 (19:23 +0200)]
aesni-x86_64.pl: optimize XTS.
PR: 3042
Andy Polyakov [Sat, 25 May 2013 17:08:39 +0000 (19:08 +0200)]
aesni-sha1-x86_64.pl: Atom-specific optimization.
Andy Polyakov [Sat, 25 May 2013 17:02:57 +0000 (19:02 +0200)]
sha512-x86_64.pl: +16% optimization for Atom.
(and pending AVX2 changes).
Andy Polyakov [Sat, 25 May 2013 16:57:03 +0000 (18:57 +0200)]
vpaes-x86[_64].pl: minor Atom-specific optimization.
Andy Polyakov [Mon, 13 May 2013 20:49:58 +0000 (22:49 +0200)]
Add AES-SHA256 stitch.
Andy Polyakov [Mon, 13 May 2013 14:06:25 +0000 (16:06 +0200)]
x86_64-xlate.pl: minor size/performance improvement.
Andy Polyakov [Mon, 13 May 2013 13:49:03 +0000 (15:49 +0200)]
aesni-x86_64.pl: minor CTR performance improvement.
Ben Laurie [Sun, 5 May 2013 15:15:34 +0000 (16:15 +0100)]
Tests pass!
Ben Laurie [Sun, 5 May 2013 14:06:33 +0000 (15:06 +0100)]
Ugly hack to avoid recompiling the same thing multiple times in parallel.
Dr. Stephen Henson [Thu, 2 May 2013 11:18:46 +0000 (12:18 +0100)]
Reencode certificates in X509_sign_ctx.
Reencode certificates in X509_sign_ctx as well as X509_sign.
This was causing a problem in the x509 application when it modified an
existing certificate.
Andy Polyakov [Tue, 23 Apr 2013 15:52:14 +0000 (17:52 +0200)]
bsaes-armv7.pl: add bsaes_cbc_encrypt and bsaes_ctr32_encrypt_blocks.
Submitted by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Contributor claims ~50% improvement in CTR and ~9% in CBC decrypt
on Cortex-A15.
Andy Polyakov [Tue, 23 Apr 2013 15:49:54 +0000 (17:49 +0200)]
bsaes-armv7.pl: take it into build loop.
Dr. Stephen Henson [Sun, 14 Apr 2013 00:19:07 +0000 (01:19 +0100)]
Typo.
Andy Polyakov [Sat, 13 Apr 2013 21:03:31 +0000 (23:03 +0200)]
Fix Windows linking error in GOST test case.
Andy Polyakov [Sat, 13 Apr 2013 18:57:37 +0000 (20:57 +0200)]
crypto/modes/modes_lcl.h: let STRICT_ALIGNMENT be on ARMv7.
While ARMv7 in general is capable of unaligned access, not all instructions
actually are. And trouble is that compiler doesn't seem to differentiate
those capable and incapable of unaligned access. Side effect is that kernel
goes into endless loop retrying same instruction triggering unaligned trap.
Problem was observed in xts128.c and ccm128.c modules. It's possible to
resolve it by using (volatile u32*) casts, but letting STRICT_ALIGNMENT
be feels more appropriate.
Dr. Stephen Henson [Tue, 9 Apr 2013 14:57:39 +0000 (15:57 +0100)]
Suite B support for DTLS 1.2
Check for Suite B support using method flags instead of version numbers:
anything supporting TLS 1.2 cipher suites will also support Suite B.
Return an error if an attempt to use DTLS 1.0 is made in Suite B mode.
Dr. Stephen Henson [Tue, 9 Apr 2013 14:53:38 +0000 (15:53 +0100)]
Always return errors in ssl3_get_client_hello
If we successfully match a cookie don't set return value to 2 as this
results in other error conditions returning 2 as well.
Instead set return value to -2 which can be checked later if everything
else is OK.
Dr. Stephen Henson [Sat, 6 Apr 2013 14:50:12 +0000 (15:50 +0100)]
Dual DTLS version methods.
Add new methods DTLS_*_method() which support both DTLS 1.0 and DTLS 1.2 and
pick the highest version the peer supports during negotiation.
As with SSL/TLS options can change this behaviour specifically
SSL_OP_NO_DTLSv1 and SSL_OP_NO_DTLSv1_2.
Dr. Stephen Henson [Mon, 8 Apr 2013 17:03:12 +0000 (18:03 +0100)]
Set s->d1 to NULL after freeing it.
Dr. Stephen Henson [Mon, 8 Apr 2013 13:53:54 +0000 (14:53 +0100)]
Asm build portability.
Don't use Win32 specific options in mk1mf.pl to build assembly language
files.
Ben Laurie [Sun, 7 Apr 2013 15:52:54 +0000 (16:52 +0100)]
Make sure all tests are actually run, plus some fixups for things that
turn out to be made somewhere by existing Makefiles.
Ben Laurie [Sat, 6 Apr 2013 19:42:38 +0000 (20:42 +0100)]
Use original alltests target for definitive test list.
Ben Laurie [Sat, 6 Apr 2013 15:17:14 +0000 (16:17 +0100)]
Missing semicolon.
Dr. Stephen Henson [Sat, 6 Apr 2013 15:05:23 +0000 (16:05 +0100)]
Fix non-copy builds.
Only use -MMD and .sinclude in copy builds: other platforms don't
support them.
Ben Laurie [Sat, 6 Apr 2013 14:25:43 +0000 (15:25 +0100)]
Missing file.
Ben Laurie [Sat, 6 Apr 2013 14:22:28 +0000 (15:22 +0100)]
Make executable.
Ben Laurie [Sat, 6 Apr 2013 14:17:30 +0000 (15:17 +0100)]
Helper scripts for one makefile build.
Ben Laurie [Sat, 6 Apr 2013 14:15:11 +0000 (15:15 +0100)]
Merge, go back to copy-if-different.
Ben Laurie [Wed, 3 Apr 2013 05:50:47 +0000 (06:50 +0100)]
Fix test_ss.
Ben Laurie [Wed, 3 Apr 2013 05:38:33 +0000 (06:38 +0100)]
Add new asm target.
Ben Laurie [Wed, 3 Apr 2013 05:30:16 +0000 (06:30 +0100)]
Show start/end of tests.
Ben Laurie [Wed, 3 Apr 2013 05:05:33 +0000 (06:05 +0100)]
Avoid collisions.
Ben Laurie [Tue, 2 Apr 2013 05:04:58 +0000 (06:04 +0100)]
Make session ID test work.
Ben Laurie [Tue, 2 Apr 2013 05:02:13 +0000 (06:02 +0100)]
Make S/MIME test work.
Ben Laurie [Tue, 2 Apr 2013 04:38:57 +0000 (05:38 +0100)]
Make RSA test work.
Ben Laurie [Tue, 2 Apr 2013 04:36:21 +0000 (05:36 +0100)]
Make PKCS#7 test work.
Ben Laurie [Tue, 2 Apr 2013 04:33:15 +0000 (05:33 +0100)]
Make OCSP test work.
Ben Laurie [Mon, 1 Apr 2013 17:22:58 +0000 (18:22 +0100)]
More progress towards working tests.
Ben Laurie [Sat, 23 Mar 2013 19:05:19 +0000 (19:05 +0000)]
Get closer to a working single Makefile with test support.
Ben Laurie [Sat, 23 Mar 2013 19:04:46 +0000 (19:04 +0000)]
Fix warnings.
Dr. Stephen Henson [Thu, 4 Apr 2013 17:19:18 +0000 (18:19 +0100)]
Make TLS 1.2 ciphers work again.
Since s->method does not reflect the final client version when a client
hello is sent for SSLv23_client_method it can't be relied on to indicate
if TLS 1.2 ciphers should be used. So use the client version instead.
Andy Polyakov [Thu, 4 Apr 2013 13:57:43 +0000 (15:57 +0200)]
cryptlib.c: fix typo in OPENSSL_showfatal.
Andy Polyakov [Thu, 4 Apr 2013 13:56:23 +0000 (15:56 +0200)]
aesni-x86_64.pl: optimize CBC decrypt.
Give CBC decrypt approximately same treatment as to CTR and collect 25%.
Andy Polyakov [Thu, 4 Apr 2013 13:55:49 +0000 (15:55 +0200)]
e_aes.c: reserve for future extensions.
Andy Polyakov [Thu, 4 Apr 2013 13:54:58 +0000 (15:54 +0200)]
gcm128.c: fix linking problems in 32-bit Windows build.
Andy Polyakov [Thu, 4 Apr 2013 13:54:08 +0000 (15:54 +0200)]
dest4-sparcv9.pl: add clarification comment.
Andy Polyakov [Thu, 4 Apr 2013 13:53:01 +0000 (15:53 +0200)]
evptests.txt: add XTS test vectors
Dr. Stephen Henson [Wed, 3 Apr 2013 21:38:18 +0000 (22:38 +0100)]
Use $(PERL) when calling scripts in mk1mf.pl
Dr. Stephen Henson [Sun, 31 Mar 2013 16:42:46 +0000 (17:42 +0100)]
Typo.
Andy Polyakov [Sun, 31 Mar 2013 12:32:05 +0000 (14:32 +0200)]
Add support for SPARC T4 DES opcode.
Andy Polyakov [Sun, 31 Mar 2013 12:07:48 +0000 (14:07 +0200)]
des_enc.m4: add missing #include.
Submitted by: David Miller
Andy Polyakov [Fri, 29 Mar 2013 19:45:33 +0000 (20:45 +0100)]
Add AES-NI GCM stitch.
Andy Polyakov [Fri, 29 Mar 2013 17:54:24 +0000 (18:54 +0100)]
aesni-x86_64.pl: fix typo and optimize small block performance.
Dr. Stephen Henson [Thu, 28 Mar 2013 14:28:06 +0000 (14:28 +0000)]
Call RAND_cleanup in openssl application.
(cherry picked from commit
944bc29f9004cf8851427ebfa83ee70b8399da57)
Dr. Stephen Henson [Wed, 27 Mar 2013 19:54:48 +0000 (19:54 +0000)]
Enable TLS 1.2 ciphers in DTLS 1.2.
Port TLS 1.2 GCM code to DTLS. Enable use of TLS 1.2 only ciphers when in
DTLS 1.2 mode too.
Dr. Stephen Henson [Wed, 27 Mar 2013 16:05:10 +0000 (16:05 +0000)]
Update fixed DH requirements.
The relaxed signing requirements for fixed DH certificates apply to DTLS 1.2
too.
Dr. Stephen Henson [Wed, 27 Mar 2013 15:50:42 +0000 (15:50 +0000)]
DTLS 1.2 cached record support.
Add DTLS1.2 support for cached records when computing handshake macs
instead of the MD5+SHA1 case for DTLS < 1.2 (this is a port of the
equivalent TLS 1.2 code to DTLS).
Matt Caswell [Tue, 26 Mar 2013 15:39:50 +0000 (15:39 +0000)]
Make binary curve ASN.1 work in FIPS mode.
Don't check for binary curves by checking methods: the values will
be different in FIPS mode as they are redirected to the validated module
version.
Dr. Stephen Henson [Wed, 20 Mar 2013 15:49:14 +0000 (15:49 +0000)]
Provisional DTLS 1.2 support.
Add correct flags for DTLS 1.2, update s_server and s_client to handle
DTLS 1.2 methods.
Currently no support for version negotiation: i.e. if client/server selects
DTLS 1.2 it is that or nothing.
Dr. Stephen Henson [Tue, 26 Mar 2013 14:46:18 +0000 (14:46 +0000)]
Remove versions test from dtls1_buffer_message
Since this is always called from DTLS code it is safe to assume the header
length should be the DTLS value. This avoids the need to check the version
number and should work with any version of DTLS (not just 1.0).
Dr. Stephen Henson [Tue, 19 Mar 2013 15:51:26 +0000 (15:51 +0000)]
Extend DTLS method macros.
Extend DTLS method creation macros to support version numbers and encryption
methods. Update existing code.
Dr. Stephen Henson [Tue, 19 Mar 2013 15:49:35 +0000 (15:49 +0000)]
Enable various DTLS extensions.
Some TLS extensions were disabled for DTLS. Possibly because they caused
problems with the old duplicated code. Enable them again.
Andy Polyakov [Tue, 26 Mar 2013 13:29:18 +0000 (14:29 +0100)]
aesni-x86_64.pl: optimize CTR even further.
Based on suggestions from Shay Gueron and Vlad Krasnov.
PR: 3021
Andy Polyakov [Sun, 24 Mar 2013 22:44:35 +0000 (23:44 +0100)]
ghash-x86_64.pl: add AVX code path.
Andy Polyakov [Tue, 19 Mar 2013 19:03:02 +0000 (20:03 +0100)]
aesni-x86_64.pl: optimize CTR even further.
Andy Polyakov [Tue, 19 Mar 2013 19:02:11 +0000 (20:02 +0100)]
ghash-x86_64.pl: minor optimization.
Dr. Stephen Henson [Tue, 19 Mar 2013 13:43:06 +0000 (13:43 +0000)]
Disable compression for DTLS.
The only standard compression method is stateful and is incompatible with
DTLS.
Dr. Stephen Henson [Tue, 19 Mar 2013 12:41:13 +0000 (12:41 +0000)]
Typo.
Andy Polyakov [Mon, 18 Mar 2013 18:29:41 +0000 (19:29 +0100)]
e_aes_cbc_hmac_sha1.c: fix rare bad record mac on AES-NI plaforms.
PR: 3002
Dr. Stephen Henson [Wed, 13 Mar 2013 15:33:24 +0000 (15:33 +0000)]
Use enc_flags when deciding protocol variations.
Use the enc_flags field to determine whether we should use explicit IV,
signature algorithms or SHA256 default PRF instead of hard coding which
versions support each requirement.
Dr. Stephen Henson [Wed, 13 Mar 2013 13:30:44 +0000 (13:30 +0000)]
Use appropriate versions of SSL3_ENC_METHOD
Dr. Stephen Henson [Mon, 11 Mar 2013 15:34:28 +0000 (15:34 +0000)]
DTLS revision.
Revise DTLS code. There was a *lot* of code duplication in the
DTLS code that generates records. This makes it harder to maintain and
sometimes a TLS update is omitted by accident from the DTLS code.
Specifically almost all of the record generation functions have code like
this:
some_pointer = buffer + HANDSHAKE_HEADER_LENGTH;
... Record creation stuff ...
set_handshake_header(ssl, SSL_MT_SOMETHING, message_len);
...
write_handshake_message(ssl);
Where the "Record creation stuff" is identical between SSL/TLS and DTLS or
in some cases has very minor differences.
By adding a few fields to SSL3_ENC to include the header length, some flags
and function pointers for handshake header setting and handshake writing the
code can cope with both cases.
Note: although this passes "make test" and some simple DTLS tests there may
be some minor differences in the DTLS code that have to be accounted for.
Michael Tuexen [Mon, 18 Mar 2013 14:30:38 +0000 (14:30 +0000)]
Avoid unnecessary fragmentation.
Dr. Stephen Henson [Mon, 18 Mar 2013 14:19:40 +0000 (14:19 +0000)]
Encode INTEGER correctly.
If an ASN1_INTEGER structure is allocated but not explicitly set encode
it as zero: don't generate an invalid zero length INTEGER.
(cherry picked from commit
1643edc63c3e15b6db5a15a728bc288f2cc2bbc7)
Dr. Stephen Henson [Mon, 18 Mar 2013 13:58:32 +0000 (13:58 +0000)]
Dr. Stephen Henson [Fri, 8 Mar 2013 16:45:37 +0000 (16:45 +0000)]
DTLS trace support.
Add DTLS record header parsing, different client hello format and add
HelloVerifyRequest message type.
Add code to d1_pkt.c to send message headers to the message callback.
Andy Polyakov [Wed, 6 Mar 2013 18:24:05 +0000 (19:24 +0100)]
evptests.txt: additional GCM test vectors.
Dr. Stephen Henson [Wed, 6 Mar 2013 15:45:15 +0000 (15:45 +0000)]
GCM and CCM test support
Add code to support GCM an CCM modes in evp_test. On encrypt this
will compare the expected ciphertext and tag. On decrypt it will
compare the expected plaintext: tag comparison is done internally.
Add a simple CCM test case and convert all tests from crypto/modes/gcm128.c
Dr. Stephen Henson [Wed, 6 Mar 2013 15:41:29 +0000 (15:41 +0000)]
Add CCM ciphers to tables.
Andy Polyakov [Wed, 6 Mar 2013 09:42:21 +0000 (10:42 +0100)]
ghash-x86_64.pl: fix length handling bug.
Thanks to Shay Gueron & Vlad Krasnov for report.
Dr. Stephen Henson [Tue, 5 Mar 2013 21:19:39 +0000 (21:19 +0000)]
typo
Dr. Stephen Henson [Tue, 5 Mar 2013 18:29:14 +0000 (18:29 +0000)]
Initial CCM code.
Simple example of CCM code use: translated from the FIPS self tests.
Ben Laurie [Tue, 5 Mar 2013 05:55:19 +0000 (05:55 +0000)]
Use CFLAG for LFLAGS instead of the nonexistent CFLAGS.
Ben Laurie [Tue, 5 Mar 2013 04:29:41 +0000 (04:29 +0000)]
Ignore mk1mf.pl output directories.
Ben Laurie [Mon, 4 Mar 2013 22:48:38 +0000 (22:48 +0000)]
Remove unused variable.
Ben Laurie [Mon, 4 Mar 2013 20:31:59 +0000 (20:31 +0000)]
Actually comment out the cpuid asm!