Commit Graph

1761 Commits

Author SHA1 Message Date
David Benjamin
e8261a0100 Convert base64_test to C++.
Not actually much of a change, but consistency.

Change-Id: If2ef7a8b698a229f5c494822d870767e1a61476e
Reviewed-on: https://boringssl-review.googlesource.com/4127
Reviewed-by: Adam Langley <agl@google.com>
2015-04-01 19:54:13 +00:00
David Benjamin
a5974bfae7 Convert bn_test to C++.
Along the way, fix a host of missing failure checks. This will save some
headache when it comes time to run these under the malloc failure tests.

Change-Id: I3fd589bd094178723398e793d6bc578884e99b67
Reviewed-on: https://boringssl-review.googlesource.com/4126
Reviewed-by: Adam Langley <agl@google.com>
2015-04-01 19:53:51 +00:00
David Benjamin
e12c4378e9 Fix leak in ssl_test.
SSL_CIPHER_get_rfc_name still returns an allocated string.

Change-Id: Ie2f14626c1ff22d0ea613b22439b7de5c04c9062
Reviewed-on: https://boringssl-review.googlesource.com/4190
Reviewed-by: Adam Langley <agl@google.com>
2015-04-01 18:22:23 +00:00
David Benjamin
5a3162a521 Convert hmac_test to C++.
Change-Id: I50db70385634c51ed692ac0ebf9732f46130ca41
Reviewed-on: https://boringssl-review.googlesource.com/4125
Reviewed-by: Adam Langley <agl@google.com>
2015-03-31 23:13:49 +00:00
David Benjamin
f93d737c86 Convert pbkdf_test to C++11.
We don't really gain much from this one, but consistency.

Change-Id: I3f830c6d1ad65263bd1cc09372a5b810a8f690c0
Reviewed-on: https://boringssl-review.googlesource.com/4124
Reviewed-by: Adam Langley <agl@google.com>
2015-03-31 23:12:11 +00:00
David Benjamin
94e597a51c Convert digest_test to C++.
Change-Id: Ibc8c3d8740cbd03dfb0d1ff9834fed66601dd479
Reviewed-on: https://boringssl-review.googlesource.com/4123
Reviewed-by: Adam Langley <agl@google.com>
2015-03-31 23:10:24 +00:00
David Benjamin
c2b45a164a Convert evp_test to C++.
Change-Id: I6e51815db2f600f4d9fb4d8b01cc92e128b31bbb
Reviewed-on: https://boringssl-review.googlesource.com/4122
Reviewed-by: Adam Langley <agl@google.com>
2015-03-31 23:08:53 +00:00
David Benjamin
1d77e56b29 Convert ssl_test to C++.
Change-Id: Ic8f3cd5c6a89e07bbae43b1599a01fedf119b081
Reviewed-on: https://boringssl-review.googlesource.com/4121
Reviewed-by: Adam Langley <agl@google.com>
2015-03-31 23:03:54 +00:00
David Benjamin
45fb1be33e Remove std::unique_ptr dependency on bssl_shim's scoped types.
This is in preparation for using RAII in the unit tests. Those tests are built
in Chromium as well, but Chromium does not have C++11 library support across
all its toolchains. Compiler support is available, so add a partial
reimplementation of std::unique_ptr and std::move under crypto/test/. The
scopers for the crypto/ library are also moved there while the ones for ssl/
stay in ssl/test/.

Change-Id: I38f769acbc16a870db34649928575c7314b6e9f6
Reviewed-on: https://boringssl-review.googlesource.com/4120
Reviewed-by: Adam Langley <agl@google.com>
2015-03-31 23:03:06 +00:00
Adam Langley
d7c5dfb233 Add native support for onces and thread-local storage.
Historically, OpenSSL has used callbacks for anything thread related,
but we don't actually have that many threading libraries to worry about:
just pthreads and Windows (I hope).

That suggests that it's quite reasonable to handle threading ourselves,
and eliminate the need for users to remember to install the thread
callbacks.

The first user of this would be ERR, which currently simulates
thread-local storage using a lock around a hash table keyed by the TID.
(Although I suspect that change will need some CMake work in order that
libpthread is automatically included with libcrypto when linking tests
etc, but not on Windows and without lots of ifs.)

Change-Id: I4dd088e3794506747f875c1f3e92b9bc6700fad2
Reviewed-on: https://boringssl-review.googlesource.com/4010
Reviewed-by: Adam Langley <agl@google.com>
2015-03-31 22:37:12 +00:00
Adam Langley
c11e13a78b Support MIPS64.
MIPS64 confusingly sets __mips__, but it's not a 32-bit platform. This
change updates the defines in base.h to recognise MIPS64 based on both
__mips__ and __LP64__ being defined.

Change-Id: I220f5d9c8f1cd7d3089cc013348e6f95cdee76d9
Reviewed-on: https://boringssl-review.googlesource.com/4093
Reviewed-by: Adam Langley <agl@google.com>
2015-03-31 22:27:10 +00:00
Adam Langley
20b64fd19d Export the PSS padding functions.
(system/keymaster is using them now.)

Change-Id: I8fba501005b9318b7d3a76bf1715fb772b23c49d
Reviewed-on: https://boringssl-review.googlesource.com/4092
Reviewed-by: Adam Langley <agl@google.com>
2015-03-31 22:26:57 +00:00
Adam Langley
32ab7b0e74 Use libmingwex for gmtime_s.
gmtime_s first appeared in MSVCR80, but libmingwex has a helper function
that tries to find the symbol or falls back to an internal
implementation.

(Patch by Kenny Root.)

Change-Id: I96ef9cd7459d7e8202831a4e687dfbc055c9f50b
Reviewed-on: https://boringssl-review.googlesource.com/4091
Reviewed-by: Adam Langley <agl@google.com>
2015-03-31 22:21:57 +00:00
Adam Langley
3e719319be Lowercase some Windows headers.
MinGW on Linux needs lowercase include files. On Windows this doesn't
matter since the filesystems are case-insensitive, but building
BoringSSL on Linux with MinGW has case-sensitive filesystems.

Change-Id: Id9c120d819071b041341fbb978352812d6d073bc
Reviewed-on: https://boringssl-review.googlesource.com/4090
Reviewed-by: Adam Langley <agl@google.com>
2015-03-31 22:21:42 +00:00
Adam Langley
e2e13265ff Don't set NEON_FUNCTIONAL in the getauxval path.
NEON_FUNCTIONAL is set by default in crypto.c. Chromium calls
|CRYPTO_set_NEON_functional| before |SSL_library_init| and thus the
getauxval path for CPU-feature detection was resetting the functional
flag, even on broken processors.

This change means that, apart from the default, only
|CRYPTO_set_NEON_functional| will change the NEON_FUNCTIONAL flag.

BUG=469511

Change-Id: I3d4dbbd9f4a5e33539f8559f90289e706ad17451
Reviewed-on: https://boringssl-review.googlesource.com/4170
Reviewed-by: David Benjamin <davidben@chromium.org>
Reviewed-by: Adam Langley <agl@google.com>
2015-03-31 00:26:25 +00:00
David Benjamin
4616bb1e6e Build 32-bit assembly with SSE2 enabled.
This affects bignum and sha. Also now that we're passing the SSE2 flag, revert
the change to ghash-x86.pl which unconditionally sets $sse2, just to minimize
upstream divergence. Chromium assumes SSE2 support, so relying on it is okay.
See https://crbug.com/349320.

Note: this change needs to be mirrored in Chromium to take.

bssl speed numbers:

SSE2:
Did 552 RSA 2048 signing operations in 3007814us (183.5 ops/sec)
Did 19003 RSA 2048 verify operations in 3070779us (6188.3 ops/sec)
Did 72 RSA 4096 signing operations in 3055885us (23.6 ops/sec)
Did 4650 RSA 4096 verify operations in 3024926us (1537.2 ops/sec)

Without SSE2:
Did 350 RSA 2048 signing operations in 3042021us (115.1 ops/sec)
Did 11760 RSA 2048 verify operations in 3003197us (3915.8 ops/sec)
Did 46 RSA 4096 signing operations in 3042692us (15.1 ops/sec)
Did 3400 RSA 4096 verify operations in 3083035us (1102.8 ops/sec)

SSE2:
Did 16407000 SHA-1 (16 bytes) operations in 3000141us (5468743.0 ops/sec): 87.5 MB/s
Did 4367000 SHA-1 (256 bytes) operations in 3000436us (1455455.1 ops/sec): 372.6 MB/s
Did 185000 SHA-1 (8192 bytes) operations in 3002666us (61611.9 ops/sec): 504.7 MB/s
Did 9444000 SHA-256 (16 bytes) operations in 3000052us (3147945.4 ops/sec): 50.4 MB/s
Did 2283000 SHA-256 (256 bytes) operations in 3000457us (760884.1 ops/sec): 194.8 MB/s
Did 89000 SHA-256 (8192 bytes) operations in 3016024us (29509.0 ops/sec): 241.7 MB/s
Did 5550000 SHA-512 (16 bytes) operations in 3000350us (1849784.2 ops/sec): 29.6 MB/s
Did 1820000 SHA-512 (256 bytes) operations in 3001039us (606456.6 ops/sec): 155.3 MB/s
Did 93000 SHA-512 (8192 bytes) operations in 3007874us (30918.8 ops/sec): 253.3 MB/s

Without SSE2:
Did 10573000 SHA-1 (16 bytes) operations in 3000261us (3524026.7 ops/sec): 56.4 MB/s
Did 2937000 SHA-1 (256 bytes) operations in 3000621us (978797.4 ops/sec): 250.6 MB/s
Did 123000 SHA-1 (8192 bytes) operations in 3033202us (40551.2 ops/sec): 332.2 MB/s
Did 5846000 SHA-256 (16 bytes) operations in 3000294us (1948475.7 ops/sec): 31.2 MB/s
Did 1377000 SHA-256 (256 bytes) operations in 3000335us (458948.8 ops/sec): 117.5 MB/s
Did 54000 SHA-256 (8192 bytes) operations in 3027962us (17833.8 ops/sec): 146.1 MB/s
Did 2075000 SHA-512 (16 bytes) operations in 3000967us (691443.8 ops/sec): 11.1 MB/s
Did 638000 SHA-512 (256 bytes) operations in 3000576us (212625.8 ops/sec): 54.4 MB/s
Did 30000 SHA-512 (8192 bytes) operations in 3042797us (9859.3 ops/sec): 80.8 MB/s

BUG=430237

Change-Id: I47d1c1ffcd71afe4f4a192272f8cb92af9505ee1
Reviewed-on: https://boringssl-review.googlesource.com/4130
Reviewed-by: Adam Langley <agl@google.com>
2015-03-26 23:35:56 +00:00
David Benjamin
87ac295d5d Remove unused fields from EVP_PKEY.
None of these are ever set.

Change-Id: If92554a4c560bf8fa2d630c419a946711db30839
Reviewed-on: https://boringssl-review.googlesource.com/4114
Reviewed-by: Adam Langley <agl@google.com>
2015-03-25 15:32:33 +00:00
David Benjamin
05a5beb35c Add support for finding depot_tools' MSVC toolchain on the bots.
- Pull in a trimmed down version of vs_toolchain.py from Chromium.

- Drop in toolchain_vs2013.hash from Chromium to use Chromium's
  current toolchain.

- Add a very hacky vs_env.py to pull in Visual Studio. This is
  loosely based off a handful of lines of Chromium's
  tools/clang/scripts/update.py. This (and vs_toolchain.py) depends
  on gyp which is now pulled in via DEPS.

BUG=430237

Change-Id: Ic29cbb15e19a99616cfe778d0778b9a71c45338a
Reviewed-on: https://boringssl-review.googlesource.com/3900
Reviewed-by: Adam Langley <agl@google.com>
2015-03-25 15:31:55 +00:00
David Benjamin
340d5ed295 Test that warning alerts are ignored.
Partly inspired by the new state exposed in
dc3da93899, stress this codepath by spamming our
poor shim with warning alerts.

Change-Id: I876c6e52911b6eb57493cf3e1782b37ea96d01f8
Reviewed-on: https://boringssl-review.googlesource.com/4112
Reviewed-by: Adam Langley <agl@google.com>
2015-03-25 15:25:28 +00:00
David Benjamin
0570923ed2 Add additional features to bssl client.
This exposes the features needed to mimic Chrome's ClientHello, which is useful
in testing. Also use bssl_shim's scopers for SSL objects.

Change-Id: Icb88bb00c0a05c27610134d618f466a24f7f757a
Reviewed-on: https://boringssl-review.googlesource.com/4113
Reviewed-by: Adam Langley <agl@google.com>
2015-03-23 23:16:30 +00:00
David Benjamin
0d4db50a54 Use C++11 inline initialization.
Google C++ style allows these. It's also considerably less tedious and
error-prone than defining an out-of-line constructor.

Change-Id: Ib76ccf6079be383722433046ac5c5d796dd1f525
Reviewed-on: https://boringssl-review.googlesource.com/4111
Reviewed-by: Adam Langley <agl@google.com>
2015-03-23 23:09:11 +00:00
David Benjamin
e5a3ac2cac Fix fail_second_ddos_callback flag.
It was failing only on 32-bit for some reason. Part of TestConfig wasn't
getting initialized.

Change-Id: I2a3747a347a47b47e2357f34d32f8db86d6cc629
Reviewed-on: https://boringssl-review.googlesource.com/4110
Reviewed-by: Adam Langley <agl@google.com>
2015-03-23 23:08:48 +00:00
David Benjamin
8056343a58 Fix CRYPTO_add_lock documentation.
It's not actually CRYPTO_add_locked, despite the name. I guess they just needed
a name that didn't clash with CRYPTO_add.

Change-Id: I3fdee08bf75e9a4e1b5e75630707c0be5792599b
Reviewed-on: https://boringssl-review.googlesource.com/4102
Reviewed-by: Adam Langley <agl@google.com>
2015-03-23 23:08:20 +00:00
David Benjamin
b6d0c6db5e Remove the stats block in SSL_CTX.
Within the library, only ssl_update_cache read them, so add a dedicated field
to replace that use.

The APIs have a handful of uninteresting callers so I've left them in for now,
but they now always return zero.

Change-Id: Ie4e36fd4ab18f9bff544541d042bf3c098a46933
Reviewed-on: https://boringssl-review.googlesource.com/4101
Reviewed-by: Adam Langley <agl@google.com>
2015-03-23 23:07:56 +00:00
David Benjamin
90fa69aaae Remove unnecessary -ldl and clean up includes for malloc tests.
I'm guessing a previous iteration used dlsym to look up the real malloc.

Change-Id: I18be9ef4db4ed059400074c8507d4e2fea882fbc
Reviewed-on: https://boringssl-review.googlesource.com/4100
Reviewed-by: Adam Langley <agl@google.com>
2015-03-21 00:07:42 +00:00
Håvard Molland
ab2479a08a Clean up error reporting.
Quite a few functions reported wrong function names when pushing
to the error stack.

Change-Id: I84d89dbefd2ecdc89ffb09799e673bae17be0e0f
Reviewed-on: https://boringssl-review.googlesource.com/4080
Reviewed-by: David Benjamin <davidben@chromium.org>
Reviewed-by: Adam Langley <agl@google.com>
2015-03-20 22:12:59 +00:00
Adam Langley
32156b974d Add some more Android compatibility declarations.
Change-Id: Ia1de2b746dec680eb59dc614295a1da289c5741a
2015-03-20 11:50:29 -07:00
David Benjamin
72dc7834af Test that signature_algorithm preferences are enforced.
Both on the client and the server.

Change-Id: I9892c6dbbb29938154aba4f53b10e8b5231f9c47
Reviewed-on: https://boringssl-review.googlesource.com/4071
Reviewed-by: Adam Langley <agl@google.com>
2015-03-20 18:23:54 +00:00
David Benjamin
67d1fb59ad Test that client cipher preferences are enforced.
Change-Id: I6e760cfd785c0c5688da6f7d3d3092a8add40409
Reviewed-on: https://boringssl-review.googlesource.com/4070
Reviewed-by: Adam Langley <agl@google.com>
2015-03-19 22:44:49 +00:00
David Benjamin
642f1498d0 evp: prevent underflow in base64 decoding
This patch resolves RT ticket #2608.

(Imported from upstream's d0666f289ac013094bbbf547bfbcd616199b7d2d.)

Change-Id: If0c8f583f3f643f3db6b6158cbc9ad7541bab8ec
Reviewed-on: https://boringssl-review.googlesource.com/4060
Reviewed-by: Adam Langley <agl@google.com>
2015-03-19 21:59:28 +00:00
David Benjamin
7061e28dc2 Rename EECDH and EDH to ECDHE and DHE.
Align with upstream's renames from a while ago. These names are considerably
more standard. This also aligns with upstream in that both "ECDHE" and "EECDH"
are now accepted in the various cipher string parsing bits.

Change-Id: I84c3daeacf806f79f12bc661c314941828656b04
Reviewed-on: https://boringssl-review.googlesource.com/4053
Reviewed-by: Adam Langley <agl@google.com>
2015-03-19 19:54:58 +00:00
David Benjamin
3c9746a6d7 Regression test for CVE-2015-0291.
This is really just scar tissue with https://crbug.com/468889 being the real
underlying problem. But the test is pretty easy.

Change-Id: I5eca18fdcbde8665c0e6c3ac419a28152647d66f
Reviewed-on: https://boringssl-review.googlesource.com/4052
Reviewed-by: Adam Langley <agl@google.com>
2015-03-19 19:52:59 +00:00
David Benjamin
b85a4c2923 Remove unnecessary NULL initializations in ssl_cert_dup.
A casual grep would suggest this function has the same problems as
CVE-2015-0291, but the structure is memset to 0, so the calls are unnecessary.
Also use BUF_memdup rather than an OPENSSL_malloc + mempcy pair.

Change-Id: Id605374d99cff32e2dccb7f9b8a9da226faf7715
Reviewed-on: https://boringssl-review.googlesource.com/4051
Reviewed-by: Adam Langley <agl@google.com>
2015-03-19 19:52:57 +00:00
David Benjamin
cdea40c3e2 Add tests for full handshakes under renegotiation.
In verifying the fix for CVE-2015-0291, I noticed we don't actually have any
test coverage for full handshakes on renegotiation. All our tests always do
resumptions.

Change-Id: Ia9b701e8a50ba9353fefb8cc4fb86e78065d0b40
Reviewed-on: https://boringssl-review.googlesource.com/4050
Reviewed-by: Adam Langley <agl@google.com>
2015-03-19 19:51:16 +00:00
David Benjamin
4b1510c71e Fix a failure to NULL a pointer freed on error.
Reported by the LibreSSL project as a follow on to CVE-2015-0209

(Imported from upstream's 5e5d53d341fd9a9b9cc0a58eb3690832ca7a511f.)

Change-Id: Ic2e5dc5c96e316c55f76bedc6ea55b416be3287a
Reviewed-on: https://boringssl-review.googlesource.com/4049
Reviewed-by: Adam Langley <agl@google.com>
2015-03-19 19:50:32 +00:00
David Benjamin
7a8e62dbd9 Fix ASN1_TYPE_cmp
Fix segmentation violation when ASN1_TYPE_cmp is passed a boolean type. This
can be triggered during certificate verification so could be a DoS attack
against a client or a server enabling client authentication.

CVE-2015-0286

(Imported from upstream's e677e8d13595f7b3287f8feef7676feb301b0e8a.)

Change-Id: I5faefc190568504bb5895ed9816a6d80432cfa45
Reviewed-on: https://boringssl-review.googlesource.com/4048
Reviewed-by: Adam Langley <agl@google.com>
2015-03-19 19:48:41 +00:00
David Benjamin
61c0d4e8b2 Always reset sigalgslen when NULLing sigalgs.
See also upstream's 34e3edbf3a10953cb407288101fd56a629af22f9. This fixes
CVE-2015-0291. Also bubble up malloc failures in tls1_set_shared_sigalgs. Tidy
up style a bit and remove unnecessary check (it actually is unnecessary; see
https://boringssl-review.googlesource.com/4042).

Change-Id: Idfb31a90fb3e56ef6fe7701464748a5c1603f064
Reviewed-on: https://boringssl-review.googlesource.com/4047
Reviewed-by: Adam Langley <agl@google.com>
2015-03-19 19:46:28 +00:00
David Benjamin
c12d377e24 Free up ADB and CHOICE if already initialised.
CVE-2015-0287

(Imported from upstream's b485d976340d3ca080060c3c7dee9102e2200762.)

Change-Id: Ie8dac46ccc95a4eee9ef535d8620217202f13ca1
Reviewed-on: https://boringssl-review.googlesource.com/4046
Reviewed-by: Adam Langley <agl@google.com>
2015-03-19 19:42:27 +00:00
David Benjamin
7e3b5840c2 Fix more missing error checks in v3_cpols.c
I only imported half of upstream's c5f2b5336ab72e40ab91e2ca85639f51fa3178c6 on
accident.

Change-Id: Ice8185ca6770f915eb859e918f5db7d5ccdc7cc7
Reviewed-on: https://boringssl-review.googlesource.com/4045
Reviewed-by: Adam Langley <agl@google.com>
2015-03-19 19:41:28 +00:00
Adam Langley
078abceb29 Don't install SIGILL handler until after sigsetjmp.
Previously, if SIGILL was received between the signal handler being
installed and before the sigsetjmp, the process would longjmp to a
random location.

Change-Id: I9e6143a17ff3db0e1b00ece68fce161801461010
Reviewed-on: https://boringssl-review.googlesource.com/3950
Reviewed-by: David Benjamin <davidben@chromium.org>
Reviewed-by: Adam Langley <agl@google.com>
2015-03-19 11:53:07 +00:00
David Benjamin
8b368412d3 Minor formatting fixes.
Noticed these as I was poking around.

Change-Id: I93833a152583feced374c9febf7485bec7abc1c7
Reviewed-on: https://boringssl-review.googlesource.com/3973
Reviewed-by: Adam Langley <agl@google.com>
2015-03-19 11:52:44 +00:00
David Benjamin
9e13e1a31d Move the is_dtls bit from SSL3_ENC_METHOD to SSL_PROTOCOL_METHOD.
This too isn't version-specific. This removes the final difference between TLS
and DTLS SSL3_ENC_METHODs and we can fold them together. (We should be able to
fold away the version-specific differences too, but all in due time.)

Change-Id: I6652d3942a0970273d46d28d7052629c81f848b5
Reviewed-on: https://boringssl-review.googlesource.com/3771
Reviewed-by: Adam Langley <agl@google.com>
2015-03-19 11:51:49 +00:00
David Benjamin
cfdd6b1aef Account for partial reads in PacketedBio.
This fixes test flakiness on Windows.

BUG=467767

Change-Id: Ie69b5b43ddd524aadb15c53705f6ec860e928786
Reviewed-on: https://boringssl-review.googlesource.com/4001
Reviewed-by: Adam Langley <agl@google.com>
2015-03-19 11:49:37 +00:00
David Benjamin
506fa4f770 Allocate string types directly.
Allocate and free ASN.1 string types directly instead of going through
the ASN.1 item code.

(Imported from upstream's 3d6aa6d441fe8124d247dffee5c68c2e5efd8258.)

Change-Id: I617283e67071a792f219ed08f19078afc223e2f5
Reviewed-on: https://boringssl-review.googlesource.com/4041
Reviewed-by: Adam Langley <agl@google.com>
2015-03-19 11:47:52 +00:00
David Benjamin
42574efaff Avoid undefined behavior in probable_prime.
(Imported from upstream's e4676e900f165f5272991443225813002300b09b.)

Change-Id: I678e158c223daf2f7f9114f4e743d531fe2e2a93
Reviewed-on: https://boringssl-review.googlesource.com/4044
Reviewed-by: Adam Langley <agl@google.com>
2015-03-19 11:36:49 +00:00
David Benjamin
cdcecf546b Fix regression in ASN1_UTCTIME_cmp_time_t
Previously, ASN1_UTCTIME_cmp_time_t would return 1 if s > t, -1 if
s < t, and 0 if s == t.

This behavior was broken in upstream's
904348a4922333106b613754136305db229475ea, resulting in the opposite time
comparison behavior.

PR#3706

(Imported from upstream's da27006df06853a33b132133699a7aa9d4277920.)

Change-Id: I9c1c28ba21d82ff4b587e33b262f46be4e846ff9
Reviewed-on: https://boringssl-review.googlesource.com/4043
Reviewed-by: Adam Langley <agl@google.com>
2015-03-19 11:36:10 +00:00
David Benjamin
a5a3eeb9cc Remove ssl_cert_inst()
It created the cert structure in SSL_CTX or SSL if it was NULL, but they can
never be NULL as the comments already said.

(Imported from upstream's 2c3823491d8812560922a58677e3ad2db4b2ec8d.)

Change-Id: I97c7bb306d6f3c18597850db9f08023b2ef74839
Reviewed-on: https://boringssl-review.googlesource.com/4042
Reviewed-by: Adam Langley <agl@google.com>
2015-03-19 11:35:46 +00:00
David Benjamin
ced555394f Fix undefined behavior in shifts.
Td4 is an array of u8. A u8 << int promotes the u8 to an int first then shifts.
If the mathematical result of a shift (as modelled by lhs * 2^{rhs}) is not
representable in an integer, behaviour is undefined. In other words, you can't
shift into the sign bit of a signed integer. Fix this by casting to u32
whenever we're shifting left by 24.

(For consistency, cast other shifts, too.)

Caught by -fsanitize=shift

Submitted by Nick Lewycky (Google)

(Imported from upstream's 8b37e5c14f0eddb10c7f91ef91004622d90ef361.)

Change-Id: Id0f98d1d65738533c6ddcc3c21bc38b569d74793
Reviewed-on: https://boringssl-review.googlesource.com/4040
Reviewed-by: Adam Langley <agl@google.com>
2015-03-19 11:33:46 +00:00
David Benjamin
34860739de ASN.1 print fix.
When printing out an ASN.1 structure, if the type is an item template don't
fall through and attempt to interpret as a primitive type.

(Imported from upstream's 5dc1247a7494f50c88ce7492518bbe0ce6f124fa.)

Change-Id: Ica39757792cbf3f83879953b67838927ddbdb809
Reviewed-on: https://boringssl-review.googlesource.com/4009
Reviewed-by: Adam Langley <agl@google.com>
2015-03-19 11:32:45 +00:00
David Benjamin
859a04b7ff Fix missing return checks in v3_cpols.c
Fixed assorted missing return value checks in v3_cpols.c

(Imported from upstream's c5f2b5336ab72e40ab91e2ca85639f51fa3178c6.)

Change-Id: I95e4157fc689201ca32e1c8e51d04801ba478685
Reviewed-on: https://boringssl-review.googlesource.com/4008
Reviewed-by: Adam Langley <agl@google.com>
2015-03-19 11:31:03 +00:00