Dr. Stephen Henson [Wed, 21 May 2014 09:39:11 +0000 (10:39 +0100)]
Change default cipher in smime app to des3.
PR#3357
(cherry picked from commit
ca3ffd9670f2b589bf8cc04923f953e06d6fbc58)
Peter Mosmans [Tue, 27 May 2014 22:26:11 +0000 (23:26 +0100)]
Fix for test_bn regular expression to work on Windows using MSYS. PR#3346
Matt Caswell [Sun, 25 May 2014 22:37:53 +0000 (23:37 +0100)]
Fixed error in args for SSL_set_msg_callback and SSL_set_msg_callback_arg
Matt Caswell [Sat, 24 May 2014 22:55:27 +0000 (23:55 +0100)]
Fix for non compilation with TLS_DEBUG defined
Matt Caswell [Wed, 21 May 2014 23:07:35 +0000 (00:07 +0100)]
Fixed minor copy&paste error, and stray space causing rendering problem
Janpopan [Sun, 4 May 2014 16:13:43 +0000 (18:13 +0200)]
Fix a wrong parameter count ERR_add_error_data
Matt Caswell [Thu, 15 May 2014 20:19:46 +0000 (21:19 +0100)]
Removed note in BUGS section about AEAD ciphers - inadvertently added to wrong branch
Jeffrey Walton [Wed, 14 May 2014 21:49:30 +0000 (22:49 +0100)]
Fix grammar error in verify pod. PR#3355
Jeffrey Walton [Wed, 14 May 2014 21:48:26 +0000 (22:48 +0100)]
Add information to BUGS section of enc documentation. PR#3354
Michal Bozon [Wed, 14 May 2014 20:07:51 +0000 (21:07 +0100)]
Corrected POD syntax errors. PR#3353
Kurt Roeckx [Mon, 12 May 2014 16:19:14 +0000 (18:19 +0200)]
Check sk_SSL_CIPHER_num() after assigning sk.
Jean-Paul Calderone [Thu, 3 Apr 2014 00:47:38 +0000 (20:47 -0400)]
Correct the return type on the signature for X509_STORE_CTX_get_ex_data given in the pod file.
Serguei E. Leontiev [Sun, 11 May 2014 18:46:42 +0000 (19:46 +0100)]
Replace manual ASN1 decoder with ASN1_get_object
Replace manual ASN.1 decoder with ASN1_get object. This
will decode the tag and length properly and check against
it does not exceed the supplied buffer length.
PR#3335
(cherry picked from commit
b0308dddd1cc6a8e1de803ef29ba6da25ee072c2)
Matt Caswell [Sun, 11 May 2014 23:38:37 +0000 (00:38 +0100)]
Fixed NULL pointer dereference. See PR#3321
Günther Noack [Thu, 1 May 2014 11:33:11 +0000 (13:33 +0200)]
Avoid out-of-bounds write in SSL_get_shared_ciphers
PR: 3317
Viktor Dukhovni [Sun, 11 May 2014 19:28:56 +0000 (20:28 +0100)]
Fix infinite loop. PR#3347
Tim Hudson [Sun, 11 May 2014 12:29:59 +0000 (13:29 +0100)]
safety check to ensure we dont send out beyond the users buffer
Dr. Stephen Henson [Thu, 8 May 2014 12:17:11 +0000 (13:17 +0100)]
Return an error if no recipient type matches.
If the key type does not match any CMS recipient type return
an error instead of using a random key (MMA mitigation). This
does not leak any useful information to an attacker.
PR#3348
(cherry picked from commit
83a3182e0560f76548f4378325393461f6275493)
Tim Hudson [Mon, 5 May 2014 00:53:39 +0000 (10:53 +1000)]
coverity 966576 - close socket in error path
Tim Hudson [Mon, 5 May 2014 00:39:30 +0000 (10:39 +1000)]
PR#3342 fix resource leak coverity issue 966577
Tim Hudson [Sun, 4 May 2014 22:22:42 +0000 (08:22 +1000)]
fix coverity issue 966597 - error line is not always initialised
Matt Caswell [Wed, 7 May 2014 22:21:02 +0000 (23:21 +0100)]
Fixed NULL pointer dereference in PKCS7_dataDecode reported by David Ramos in PR#3339
Geoff Thorpe [Sun, 4 May 2014 22:44:14 +0000 (18:44 -0400)]
evp: prevent underflow in base64 decoding
This patch resolves RT ticket #2608.
Thanks to Robert Dugal for originally spotting this, and to David
Ramos for noticing that the ball had been dropped.
Signed-off-by: Geoff Thorpe <geoff@openssl.org>
Geoff Thorpe [Sun, 4 May 2014 20:19:22 +0000 (16:19 -0400)]
bignum: allow concurrent BN_MONT_CTX_set_locked()
The lazy-initialisation of BN_MONT_CTX was serialising all threads, as
noted by Daniel Sands and co at Sandia. This was to handle the case that
2 or more threads race to lazy-init the same context, but stunted all
scalability in the case where 2 or more threads are doing unrelated
things! We favour the latter case by punishing the former. The init work
gets done by each thread that finds the context to be uninitialised, and
we then lock the "set" logic after that work is done - the winning
thread's work gets used, the losing threads throw away what they've done.
Signed-off-by: Geoff Thorpe <geoff@openssl.org>
Geoff Thorpe [Sun, 27 Apr 2014 20:06:50 +0000 (16:06 -0400)]
dso: eliminate VMS code on non-VMS systems
Even though the meat of dso_vms.c is compiled out on non-VMS builds,
the (pre-)compiler still traverses some of the macro handling. This
trips up at least one non-VMS build configuration, so this commit
makes the skip-VMS case more robust.
Signed-off-by: Geoff Thorpe <geoff@openssl.org>
Dr. Stephen Henson [Tue, 6 May 2014 13:07:37 +0000 (14:07 +0100)]
Initialize num properly.
PR#3289
PR#3345
(cherry picked from commit
3ba1e406c2309adb427ced9815ebf05f5b58d155)
Dr. Stephen Henson [Tue, 6 May 2014 13:02:17 +0000 (14:02 +0100)]
Set Enveloped data version to 2 if ktri version not zero.
(cherry picked from commit
9c5d953a07f472452ae2cb578e39eddea2de2b9c)
David Ramos [Sat, 3 May 2014 10:00:27 +0000 (12:00 +0200)]
Double free in i2o_ECPublicKey
PR: 3338
Jeff Trawick [Sun, 13 Apr 2014 13:10:17 +0000 (15:10 +0200)]
typo in SSL_get_peer_cert_chain docs
RT: 3304
Matt Caswell [Wed, 30 Apr 2014 22:58:24 +0000 (23:58 +0100)]
Fixed missing =back commands
Lubomir Rintel [Mon, 21 Oct 2013 09:03:01 +0000 (11:03 +0200)]
POD: Fix item numbering
Newer pod2man considers =item [1-9] part of a numbered list, while =item
0 starts an unnumbered list. Add a zero effect formatting mark to override
this.
doc/apps/smime.pod around line 315: Expected text after =item, not a
number
...
PR#3146
mancha [Fri, 25 Apr 2014 14:58:49 +0000 (14:58 +0000)]
Fix version documentation.
Specify -f is for compilation flags. Add -d to synopsis section.
(cherry picked from commit
006397ea62bbcae22c8664d53c2222b808c4bdd1)
Closes #80.
mancha [Thu, 24 Apr 2014 19:06:20 +0000 (19:06 +0000)]
Fix eckey_priv_encode()
Fix eckey_priv_encode to return an error on failure of i2d_ECPrivateKey.
Steve Marquess [Thu, 24 Apr 2014 11:13:05 +0000 (07:13 -0400)]
Add new sponsors
(cherry picked from commit
351f0a124bffaa94d2a8abdec2e7dde5ae9c457d)
Ben Laurie [Wed, 23 Apr 2014 06:24:03 +0000 (07:24 +0100)]
Fix use after free.
mancha [Tue, 22 Apr 2014 12:11:56 +0000 (13:11 +0100)]
Fix double frees.
Conflicts:
CHANGES
Dr. Stephen Henson [Wed, 16 Apr 2014 11:15:43 +0000 (12:15 +0100)]
Document -debug_decrypt option.
(cherry picked from commit
0dd5b94aeb77c2982bdf6886962b7a8491c6c9ed)
Dr. Stephen Henson [Tue, 15 Apr 2014 17:48:54 +0000 (18:48 +0100)]
Extension checking fixes.
When looking for an extension we need to set the last found
position to -1 to properly search all extensions.
PR#3309.
(cherry picked from commit
300b9f0b704048f60776881f1d378c74d9c32fbd)
Dr. Stephen Henson [Tue, 15 Apr 2014 17:17:12 +0000 (18:17 +0100)]
Clarify CMS_decrypt behaviour.
(cherry picked from commit
5f8e9a477a18551052f2019c1f374061acbaa5e6)
Dr. Stephen Henson [Fri, 11 Apr 2014 01:50:51 +0000 (02:50 +0100)]
Add new key fingerprint.
(cherry picked from commit
3143a332e8f2f5ca1a6f0262a1a1a66103f2adf7)
Dr. Stephen Henson [Wed, 9 Apr 2014 14:42:40 +0000 (15:42 +0100)]
Fix free errors in ocsp utility.
Keep copy of any host, path and port values allocated by
OCSP_parse_url and free as necessary.
(cherry picked from commit
5219d3dd350cc74498dd49daef5e6ee8c34d9857)
Dr. Stephen Henson [Mon, 7 Apr 2014 12:02:10 +0000 (13:02 +0100)]
Document -verify_return_error option.
(cherry picked from commit
4e6c12f3088d3ee5747ec9e16d03fc671b8f40be)
Andy Polyakov [Sun, 6 Apr 2014 15:23:55 +0000 (17:23 +0200)]
crypto/modes: strict aliasing fixes from master.
Dr. Stephen Henson [Fri, 4 Apr 2014 11:44:43 +0000 (12:44 +0100)]
Use correct length when prompting for password.
Use bufsiz - 1 not BUFSIZ - 1 when prompting for a password in
the openssl utility.
Thanks to Rob Mackinnon, Leviathan Security for reporting this issue.
(cherry picked from commit
7ba08a4d73c1bdfd3aced09a628b1d7d7747cdca)
Dr. Stephen Henson [Fri, 4 Apr 2014 11:46:39 +0000 (12:46 +0100)]
Update FAQ.
(cherry picked from commit
6cc0068430d0a4abdef0b466d422e6a4d154a5fe)
Dr. Stephen Henson [Thu, 3 Apr 2014 12:27:08 +0000 (13:27 +0100)]
Document new crl option.
(cherry picked from commit
dbb7654dc189992966ecd95ca66f7a3bb011ab9b)
Tim Hudson [Thu, 3 Apr 2014 12:23:51 +0000 (13:23 +0100)]
Add option to generate old hash format.
New -hash_old to generate CRL hashes using old
(before OpenSSL 1.0.0) algorithm.
(cherry picked from commit
de2d97cd799f38024d70847bab37d91aa5a2536e)
Eric Young [Wed, 2 Apr 2014 18:50:33 +0000 (19:50 +0100)]
Fix base64 decoding bug.
A short PEM encoded sequence if passed to the BIO, and the file
had 2 \n following would fail.
PR#3289
(cherry picked from commit
10378fb5f4c67270b800e8f7c600cd0548874811)
Dr. Stephen Henson [Wed, 12 Mar 2014 14:43:03 +0000 (14:43 +0000)]
update NEWS
Dr. Stephen Henson [Wed, 12 Mar 2014 14:35:54 +0000 (14:35 +0000)]
Update ordinals.
Use a previously unused value as we will be updating multiple released
branches.
Dr. Stephen Henson [Wed, 12 Mar 2014 14:16:19 +0000 (14:16 +0000)]
Fix for CVE-2014-0076
Fix for the attack described in the paper "Recovering OpenSSL
ECDSA Nonces Using the FLUSH+RELOAD Cache Side-channel Attack"
by Yuval Yarom and Naomi Benger. Details can be obtained from:
http://eprint.iacr.org/2014/140
Thanks to Yuval Yarom and Naomi Benger for discovering this
flaw and to Yuval Yarom for supplying a fix.
Dr. Stephen Henson [Mon, 10 Mar 2014 15:45:30 +0000 (15:45 +0000)]
Andy Polyakov [Fri, 7 Mar 2014 10:02:25 +0000 (11:02 +0100)]
engines/ccgost/gosthash.c: simplify and avoid SEGV.
PR: 3275
(cherry picked from commit
ea38f020492042bc1d1adc26ef197b0b9cadf968)
Dr. Stephen Henson [Wed, 26 Feb 2014 15:28:20 +0000 (15:28 +0000)]
Fix for WIN32 builds with KRB5
(cherry picked from commit
3eddd1706a30cdf3dc9278692d8ee9038eac8a0d)
Andy Polyakov [Tue, 25 Feb 2014 21:21:54 +0000 (22:21 +0100)]
ssl/t1_enc.c: check EVP_MD_CTX_copy return value.
PR: 3201
(cherry picked from commit
03da57fe14f2de5bde9d4496a2ae9a4ae8879f88)
Dr. Stephen Henson [Tue, 25 Feb 2014 12:11:22 +0000 (12:11 +0000)]
Avoid Windows 8 Getversion deprecated errors.
Windows 8 SDKs complain that GetVersion() is deprecated.
We only use GetVersion like this:
(GetVersion() < 0x80000000)
which checks if the Windows version is NT based. Use a macro check_winnt()
which uses GetVersion() on older SDK versions and true otherwise.
(cherry picked from commit
a4cc3c8041104896d51ae12ef7b678c31808ce52)
Andy Polyakov [Mon, 24 Feb 2014 15:54:10 +0000 (16:54 +0100)]
BC-32.pl: pre-1.0.2-specific refresh for Borland C.
PR: 3251
Suggested by: Thorsten Schöning
(cherry picked from commit
3ae1b53496acd3f25e9b9089ed5fb6000e41a557)
Andy Polyakov [Mon, 24 Feb 2014 15:42:40 +0000 (16:42 +0100)]
BC-32.pl: refresh Borland C support.
PR: 3251
Suggested by: Thorsten Schöning
(cherry picked from commit
779c51c6446f384c2f2a7bd5cc4c3e0366baf628)
Andy Polyakov [Mon, 24 Feb 2014 14:16:56 +0000 (15:16 +0100)]
x509/by_dir.c: fix run-away pointer (and potential SEGV)
when adding duplicates in add_cert_dir.
PR: 3261
Reported by: Marian Done
(cherry picked from commit
758954e0d8232d370ed72b7f86640e40443e1778)
Dr. Stephen Henson [Sat, 15 Feb 2014 01:27:56 +0000 (01:27 +0000)]
Remove duplicate statement.
(cherry picked from commit
5a7652c3e585e970e5b778074c92e617e48fde38)
Kurt Roeckx [Mon, 23 Dec 2013 18:45:26 +0000 (19:45 +0100)]
Use defaults bits in req when not given
If you use "-newkey rsa" it's supposed to read the default number of bits from the
config file. However the value isn't used to generate the key, but it does
print it's generating such a key. The set_keygen_ctx() doesn't call
EVP_PKEY_CTX_set_rsa_keygen_bits() and you end up with the default set in
pkey_rsa_init() (1024). Afterwards the number of bits gets read from the config
file, but nothing is done with that anymore.
We now read the config first and use the value from the config file when no size
is given.
PR: 2592
(cherry picked from commit
3343220327664680420d4068e1fbe46d2236f1b0)
Kurt Roeckx [Mon, 23 Dec 2013 18:41:28 +0000 (19:41 +0100)]
Fix additional pod errors with numbered items.
(cherry picked from commit
e547c45f1c74e976656c042ec9d873f6eea0e756)
Scott Schaefer [Mon, 23 Dec 2013 18:28:30 +0000 (19:28 +0100)]
Fix various spelling errors
(cherry picked from commit
2b4ffc659eabec29f76821f0ac624a2b8c19e4c7)
Scott Schaefer [Mon, 23 Dec 2013 18:01:38 +0000 (19:01 +0100)]
Document pkcs12 -password behavior
apps/pkcs12.c accepts -password as an argument. The document author
almost certainly meant to write "-password, -passin".
However, that is not correct, either. Actually the code treats
-password as equivalent to -passin, EXCEPT when -export is also
specified, in which case -password as equivalent to -passout.
(cherry picked from commit
856c6dfb09d69fc82ada2611c6cd792dfc60e355)
Dr. Stephen Henson [Wed, 29 Jan 2014 00:59:35 +0000 (00:59 +0000)]
Clarify docs.
Remove reference to ERR_TXT_MALLOCED in the error library as that is
only used internally. Indicate that returned error data must not be
freed.
(cherry picked from commit
f2d678e6e89b6508147086610e985d4e8416e867)
Dr. Stephen Henson [Tue, 28 Jan 2014 15:35:16 +0000 (15:35 +0000)]
Dr. Stephen Henson [Tue, 28 Jan 2014 15:15:26 +0000 (15:15 +0000)]
Fix demo comment: 0.9.9 never released.
(cherry picked from commit
717cc8589540b95122a652dee68e6a75b6262d93)
Dr. Stephen Henson [Tue, 28 Jan 2014 13:51:58 +0000 (13:51 +0000)]
Add loaded dynamic ENGINEs to list.
Always add a dynamically loaded ENGINE to list. Otherwise it can cause
problems when multiply loaded, especially if it adds new public key methods.
For all current engines we only want a single implementation anyway.
(cherry picked from commit
e933f91f50108a43c0198cdc63ecdfdbc77b4d0d)
Dr. Stephen Henson [Thu, 23 Jan 2014 18:36:33 +0000 (18:36 +0000)]
Use default digest implementation in dgst.c
Use default instead of ENGINE version of digest. Without this
errors will occur if you use an ENGINE for a private key and
it doesn't implement the digest in question.
(cherry picked from commit
4eedf86a163284478140c46edb7b462d33898c93)
Dr. Stephen Henson [Thu, 23 Jan 2014 17:14:48 +0000 (17:14 +0000)]
make update
Kaspar Brand [Thu, 16 Jan 2014 13:49:38 +0000 (13:49 +0000)]
Omit initial status request callback check.
PR#3178
(cherry picked from commit
eb85ee9a8867b605cd7fb427869d0e50caa80a3f)
Zoltan Arpadffy [Sat, 11 Jan 2014 22:42:15 +0000 (22:42 +0000)]
VMS fixes
Dr. Stephen Henson [Fri, 10 Jan 2014 23:03:47 +0000 (23:03 +0000)]
fix shell syntax PR#3216
Dr. Stephen Henson [Thu, 9 Jan 2014 22:47:22 +0000 (22:47 +0000)]
Fix bug in X509_V_FLAG_IGNORE_CRITICAL CRL handling.
(cherry picked from commit
8f4077ca69076cebaca51b7b666db1ed49e46b9e)
Dr. Stephen Henson [Mon, 6 Jan 2014 15:02:46 +0000 (15:02 +0000)]
Prepare for 1.0.0m-dev
Dr. Stephen Henson [Mon, 6 Jan 2014 15:02:02 +0000 (15:02 +0000)]
Prepare for 1.0.0l release
Dr. Stephen Henson [Mon, 6 Jan 2014 13:36:25 +0000 (13:36 +0000)]
make update
Dr. Stephen Henson [Sat, 4 Jan 2014 13:50:52 +0000 (13:50 +0000)]
Restore SSL_OP_MSIE_SSLV2_RSA_PADDING
The flag SSL_OP_MSIE_SSLV2_RSA_PADDING hasn't done anything since OpenSSL
0.9.7h but deleting it will break source compatibility with any software
that references it. Restore it but #define to zero.
(cherry picked from commit
b17d6b8d1d49fa4732deff17cfd1833616af0d9c)
Dr. Stephen Henson [Thu, 2 Jan 2014 19:03:58 +0000 (19:03 +0000)]
update NEWS
Dr. Stephen Henson [Fri, 20 Dec 2013 15:26:50 +0000 (15:26 +0000)]
Fix DTLS retransmission from previous session.
For DTLS we might need to retransmit messages from the previous session
so keep a copy of write context in DTLS retransmission buffers instead
of replacing it after sending CCS. CVE-2013-6450.
(cherry picked from commit
34628967f1e65dc8f34e000f0f5518e21afbfc7b)
Conflicts:
ssl/ssl_locl.h
Dr. Stephen Henson [Fri, 20 Dec 2013 15:12:26 +0000 (15:12 +0000)]
Ignore NULL parameter in EVP_MD_CTX_destroy.
(cherry picked from commit
a6c62f0c25a756c263a80ce52afbae888028e986)
Dr. Stephen Henson [Tue, 10 Dec 2013 00:11:00 +0000 (00:11 +0000)]
remove obsolete STATUS file
Dr. Stephen Henson [Mon, 9 Dec 2013 23:54:37 +0000 (23:54 +0000)]
Add release dates to NEWS
Dr. Stephen Henson [Wed, 27 Nov 2013 15:37:00 +0000 (15:37 +0000)]
Simplify and update openssl.spec
Dr. Stephen Henson [Mon, 11 Nov 2013 22:39:40 +0000 (22:39 +0000)]
Fix memory leak.
(cherry picked from commit
16bc45ba956fdf07c7cda7feda88de597569df63)
Dr. Stephen Henson [Thu, 7 Nov 2013 15:15:20 +0000 (15:15 +0000)]
Check for missing components in RSA_check.
(cherry picked from commit
01be36ef70525e81fc358d2e559bdd0a0d9427a5)
Dr. Stephen Henson [Thu, 7 Nov 2013 17:27:07 +0000 (17:27 +0000)]
Document RSAPublicKey_{in,out} options.
(cherry picked from commit
7040d73d22987532faa503630d6616cf2788c975)
Andy Polyakov [Fri, 8 Nov 2013 22:00:35 +0000 (23:00 +0100)]
engines/ccgost/gost89.h: make word32 defintion unconditional.
Original definition depended on __LONG_MAX__ that is not guaranteed to
be present. As we don't support platforms with int narrower that 32 bits
it's appropriate to make defition inconditional.
PR: 3165
(cherry picked from commit
96180cac04591abfe50fc86096365553484bde65)
Dr. Stephen Henson [Wed, 6 Nov 2013 13:16:50 +0000 (13:16 +0000)]
Initialise context before using it.
(cherry picked from commit
a4947e4e064d2d5bb622ac64cf13edc4a46ed196)
Ben Laurie [Sun, 3 Nov 2013 17:23:50 +0000 (17:23 +0000)]
PBKDF2 should be efficient. Contributed by Christian Heimes
<christian@python.org>.
Conflicts:
crypto/evp/p5_crpt2.c
Ben Laurie [Fri, 4 Oct 2013 13:43:43 +0000 (14:43 +0100)]
Merge branch 'rob-100' into OpenSSL_1_0_0-stable
Rob Stradling [Thu, 12 Sep 2013 21:02:04 +0000 (22:02 +0100)]
Update CHANGES.
Andy Polyakov [Thu, 3 Oct 2013 08:55:49 +0000 (10:55 +0200)]
evp/e_des3.c: fix typo with potential integer overflow on 32-bit platforms.
Submitted by: Yuriy Kaminskiy
(cherry picked from commit
524b00c0da42b129ed8622dfb3f5eab9cc5d6617)
Resolved conflicts:
crypto/evp/e_des3.c
(cherry picked from commit
eb22b7ec7515b42a5865af7b2a1da232ce100e97)
Ben Laurie [Tue, 1 Oct 2013 13:51:04 +0000 (14:51 +0100)]
Constification.
Conflicts:
crypto/buffer/buffer.c
Dr. Stephen Henson [Wed, 17 Jul 2013 13:19:40 +0000 (14:19 +0100)]
Bodo Moeller [Mon, 16 Sep 2013 12:46:57 +0000 (14:46 +0200)]
Sync CHANGES and NEWS files.
Bodo Moeller [Mon, 16 Sep 2013 11:03:27 +0000 (13:03 +0200)]
Fix overly lenient comparisons:
- EC_GROUP_cmp shouldn't consider curves equal just because
the curve name is the same. (They really *should* be the same
in this case, but there's an EC_GROUP_set_curve_name API,
which could be misused.)
- EC_POINT_cmp shouldn't return 0 for ERR_R_SHOULD_NOT_HAVE_BEEN_CALLED
or EC_R_INCOMPATIBLE_OBJECTS errors because in a cmp API, 0 indicates
equality (not an error).
Reported by: king cope
(cherry picked from commit
312a46791ab465cfa3bf26764361faed0e5df014)
Rob Stradling [Tue, 10 Sep 2013 10:38:52 +0000 (11:38 +0100)]
Remove empty line.
Rob Stradling [Tue, 10 Sep 2013 09:13:00 +0000 (10:13 +0100)]
Tidy up comments.
Rob Stradling [Tue, 10 Sep 2013 09:05:17 +0000 (10:05 +0100)]
Use TLS version supplied by client when fingerprinting Safari.