GCC's __ARMEL__ and __ARMEB__ defines denote little- and big-endian arm,
respectively. They are not defined on aarch64, which instead use
__AARCH64EL__ and __AARCH64EB__.
However, OpenSSL's assembly originally used the 32-bit defines on both
platforms and even define __ARMEL__ and __ARMEB__ in arm_arch.h. This is
less portable and can even interfere with other headers, which use
__ARMEL__ to detect little-endian arm. (Our own base.h believes
__ARMEL__ implies 32-bit arm. We just happen to check __AARCH64EL__
first. base.h is probably also always included before arm_arch.h.)
Over time, the aarch64 assembly has switched to the correct defines,
such as in 32bbb62ea634239e7cb91d6450ba23517082bab6. This commit
finishes the job.
(There is an even more official endianness detector, __ARM_BIG_ENDIAN in
the Arm C Language Extensions. But I've stuck with the GCC ones here as
that would be a larger change.)
See also https://github.com/openssl/openssl/pull/17373
Change-Id: Ic04ff85782e6599cdeaeb33d12c2fa8edc882224
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/50848
Reviewed-by: Adam Langley <agl@google.com>
The latest version of ACLE splits __ARM_FEATURE_CRYPTO into two defines
to reflect that, starting ARMv8.2, the cryptography extension can
include {AES,PMULL} and {SHA1,SHA256} separately.
Also standardize on __ARM_NEON, which is the recommended symbol from
ACLE, and the only one defined on non-Apple aarch64 targets. Digging
through GCC history, __ARM_NEON__ is a bit older. __ARM_NEON was added
in GCC's 9e94a7fc5ab770928b9e6a2b74e292d35b4c94da from 2012, part of GCC
4.8.0.
I suspect we can stop paying attention to __ARM_NEON__ at this point,
but I've left both working for now. __ARM_FEATURE_{AES,SHA2} is definite
too new to fully replace __ARM_FEATURE_CRYPTO.
Tested on Linux that -march=armv8-a+aes now also drops the fallback AES
code. Previously, we would pick up -march=armv8-a+crypto, but not
-march=armv8-a+aes. Also tested that, on an OPENSSL_STATIC_ARMCAP build,
-march=armv8-a+sha2 sets the SHA-1 and SHA-256 features.
Change-Id: I749bdbc501ba2da23177ddb823547efcd77e5c98
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/50847
Reviewed-by: Adam Langley <agl@google.com>
These symbols were not marked OPENSSL_EXPORT, so they weren't really
usable externally anyway. They're also very sensitive to various build
configuration toggles, which don't always get reflected into projects
that include our headers. Move them to crypto/internal.h.
Change-Id: I79a1fcf0b24e398d75a9cc6473bae28ec85cb835
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/50846
Reviewed-by: Adam Langley <agl@google.com>
cpu.h contains almost entirely private symbols, which aren't reliably
usable outside the library because they lack OPENSSL_EXPORT. (And can't
have OPENSSL_EXPORT. The linker wants references to exported symbols to
go through the GOT, and our assembly doesn't do that.) In preparation
for unexporting them, move the few public APIs to crypto.h. They seem
similar in spirit to functions like CRYPTO_has_asm.
Update-Note: As part of this, I conditioned cpu-arm-linux.c on
OPENSSL_LINUX, so that the header files can have accurate conditions.
This means unrecognized ARM platforms that do not set
OPENSSL_STATIC_ARMCAP will fail to build, where previously we defaulted
to the Linux mechanisms. This matches cpu-aarch64-linux.c, which is
already gated on OPENSSL_LINUX. (And the file is quite Linux-specific.
Even if a non-Linux ELF target used getauxval for ARM capabilities, it's
unlikely that our hardcoded constants and /proc behavior applies
anyway.)
Change-Id: I1ee9eb72097be619d3f28a51b1ea058b3c37d05a
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/50845
Reviewed-by: Adam Langley <agl@google.com>
Commit-Queue: Adam Langley <agl@google.com>
We use underscores everywhere except these files, which use hyphens.
Switch them to be consistent.
Change-Id: I67eddbdae7caaf8405bdb4a0c1b65e6f3ca43916
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/50808
Reviewed-by: Adam Langley <agl@google.com>
Commit-Queue: David Benjamin <davidben@google.com>
This imports the changes to sha512-armv8.pl from
upstream's af0fcf7b4668218b24d9250b95e0b96939ccb4d1.
Tweaks needed:
- Add an explicit .text because we put .LK$BITS in .rodata for XOM
- .LK$bits and code are in separate sections, so use adrp/add instead of
plain adr
- Where glibc needs feature flags to *enable* pthread_rwlock, Apple
interprets _XOPEN_SOURCE as a request to *disable* Apple extensions.
Tighten the condition on the _XOPEN_SOURCE check.
Added support for macOS and Linux, tested manually on an ARM Mac and a
VM, respectively. Fuchsia and Windows do not currently have APIs to
expose this bit, so I've left in TODOs. Benchmarks from an Apple M1 Max:
Before:
Did 4647000 SHA-512 (16 bytes) operations in 1000103us (74.3 MB/sec)
Did 1614000 SHA-512 (256 bytes) operations in 1000379us (413.0 MB/sec)
Did 439000 SHA-512 (1350 bytes) operations in 1001694us (591.6 MB/sec)
Did 76000 SHA-512 (8192 bytes) operations in 1011821us (615.3 MB/sec)
Did 39000 SHA-512 (16384 bytes) operations in 1024311us (623.8 MB/sec)
After:
Did 10369000 SHA-512 (16 bytes) operations in 1000088us (165.9 MB/sec) [+123.1%]
Did 3650000 SHA-512 (256 bytes) operations in 1000079us (934.3 MB/sec) [+126.2%]
Did 1029000 SHA-512 (1350 bytes) operations in 1000521us (1388.4 MB/sec) [+134.7%]
Did 175000 SHA-512 (8192 bytes) operations in 1001874us (1430.9 MB/sec) [+132.5%]
Did 89000 SHA-512 (16384 bytes) operations in 1010314us (1443.3 MB/sec) [+131.4%]
(This doesn't seem to change the overall SHA-256 vs SHA-512 performance
question on ARM, when hashing perf matters. SHA-256 on the same chip
gets up to 2454.6 MB/s.)
In terms of build coverage, for now, we'll have build coverage
everywhere and test coverage on Chromium, which runs this code on macOS
CI. We should request a macOS ARM64 bot for our standalone CI. Longer
term, we need a QEMU-based builder to test various features. QEMU seems
to have pretty good coverage of all this, which will at least give us
Linux.
I haven't added an OPENSSL_STATIC_ARMCAP_SHA512 for now. Instead, we
just look at the standard __ARM_FEATURE_SHA512 define. Strangely, the
corresponding -march tag is not sha512. Neither GCC and nor Clang have
-march=armv8-a+sha512. Instead, -march=armv8-a+sha3 implies both
__ARM_FEATURE_SHA3 and __ARM_FEATURE_SHA512! Yet everything else seems
to describe the SHA512 extension as separate from SHA3.
https://developer.arm.com/architectures/system-architectures/software-standards/acle
Update-Note: Consumers with a different build setup may need to
limit -D_XOPEN_SOURCE=700 to Linux or non-Apple platforms. Otherwise,
<sys/types.h> won't define some typedef needed by <sys/sysctl.h>. If you
see a build error about u_char, etc., being undefined in some system
header, that is probably the cause.
Change-Id: Ia213d3796b84c71b7966bb68e0aec92e5d7d26f0
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/50807
Reviewed-by: Adam Langley <agl@google.com>
Commit-Queue: David Benjamin <davidben@google.com>
This imports 753316232243ccbf86b96c1c51ffcb41651d9ad5,
46f4e1bec51dc96fa275c168752aa34359d9ee51, and
32bbb62ea634239e7cb91d6450ba23517082bab6.
The last commit fixes a detection of big-endian aarch64 in the kernel,
which we do not support at all, but is imported to reduce the upstream
diff. Though it points out a messy part of arm_arch.h: __ARMEL__ and
__ARMEB__ are specific to 32-bit ARM. __AARCH64EB__ and __AARCH64EL__
are the 64-bit ones. But OpenSSL's arm_arch.h defines __ARME[LB]__ for
aarch64 and uses it in perlasm. We should fix the files upstream to
look at the aarch64 ones. (Indeed our own base.h assumes __ARMEL__
implies 32-bit ARM.)
Change-Id: I6c2241e103a97e8c3599cdfa43dcc6f30d4a2581
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/50806
Reviewed-by: Adam Langley <agl@google.com>
Commit-Queue: David Benjamin <davidben@google.com>
We currently have two aarch64 SHA-256 implementations: one using
general-purpose registers and one using the SHA-256 extensions.
Upstream's 866e505e0d663158b0fe63a7fb7455eebacc6470 added a NEON
version.
This CL syncs the transforms at the bottom of the file, to avoid
potential mistranslations in future imports. It doesn't change the
output for our current assembly.
Skips the NEON implementation itself for now. It only helps
processors without SHA-256 instructions. While Android does not
actually mandate the cryptography extensions on ARMv8, most devices
have it.
Additionally, this file does CPU dispatch in assembly, without taking
advantage of static information. We'd end up shipping both fallback
SHA-256 implementations. This is particularly silly because NEON is
mandatory in ARMv8-A anyway. (Does anyone build us on -R or -M? Probably
not?)
(If we later have a reason to import it, the binary size cost isn't that
significant. Moreover, the NEON fallback is actually slightly smaller
than the non-NEON fallback, so if we move CPU dispatch to C, importing
may even be worthwhile.)
Change-Id: I3c8ca6e77e4e6d1299f975c407cbcf4c9c240523
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/50805
Reviewed-by: Adam Langley <agl@google.com>
Commit-Queue: David Benjamin <davidben@google.com>
This fix isn't ideal, given the current space of possible version
values. But rather than make the printing code complicated, we should
make invalid versions impossible. I've left TODOs where that would be
needed.
Bug: 467, 450
Change-Id: I6c9ae97b8454182b0c1ab6ba2e070dc6d7d8b3f4
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/50767
Reviewed-by: Adam Langley <agl@google.com>
Commit-Queue: David Benjamin <davidben@google.com>
X509_print_ex tries to print negative serial numbers in decimal. In
doing so, it ends up passing a signed long to %lx and trips
-Wformat-signed.
A minimal fix would be to cast to unsigned long, but this unsigned long
is the absolute value of a signed long (l = -l). This is tricky because
-LONG_MIN does not fit in long. It all works because the length check
only allows one bit short of sizeof(long)*8 bits (ASN1_INTEGER is
sign-and-magnitude).
Still, this is a whole lot of subtlety to account for an invalid case.
Instead, send negative serial numbers down the generic path.
Bug: 450
Change-Id: Ib215fd23863de27e01f7ededf95578f9c800da37
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/50766
Reviewed-by: Adam Langley <agl@google.com>
Commit-Queue: David Benjamin <davidben@google.com>
GCC has a warning that complains about even more type mismatches in
printf. Some of these are a bit messy and will be fixed in separate CLs.
This covers the easy ones.
The .*s stuff is unfortunate, but printf has no size_t-clean string
printer. ALPN protocol lengths are bound by uint8_t, so it doesn't
really matter.
The IPv6 printing one is obnoxious and arguably a false positive. It's
really a C language flaw: all types smaller than int get converted to
int when you do arithmetic. So something like this first doesn't
overflow the shift because it computes over int, but then the result
overall is stored as an int.
uint8_t a, b;
(a << 8) | b
On the one hand, this fixes a "missing" cast to uint16_t before the
shift. At the same time, the incorrect final type means passing it to
%x, which expects unsigned int. The compiler has forgotten this value
actually fits in uint16_t and flags a warning. Mitigate this by storing
in a uint16_t first.
The story doesn't quite end here. Arguments passed to variadic functions
go through integer promotion[0], so the argument is still passed to
snprintf as an int! But then va_arg allows for a signedness mismatch[1],
provided the value is representable in both types. The combination means
that %x, though actually paired with unsigned, also accept uint8_t and
uint16_t, because those are guaranteed to promote to an int that meets
[1]. GCC recognizes [1] applies here.
(There's also PRI16x, but that's a bit tedious to use and, in glibc, is
defined as plain "x" anyway.)
[0] https://en.cppreference.com/w/c/language/conversion#Default_argument_promotions
[1] https://en.cppreference.com/w/c/variadic/va_arg
Bug: 450
Change-Id: Ic1d41356755a18ab922956dd2e07b560470341f4
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/50765
Reviewed-by: Adam Langley <agl@google.com>
Commit-Queue: Adam Langley <agl@google.com>
Update-Note: ".example.com" as an input DNS name will no longer match
"www.example.com" in a certificate. (Note this does not impact wildcard
certificates. Rather, it removes a non-standard "reverse wildcard" that
OpenSSL implemented.)
Fixed: 463
Change-Id: I627e1bd00b8e4b810e9bb756f424f6230a99496e
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/50726
Reviewed-by: Adam Langley <agl@google.com>
Commit-Queue: David Benjamin <davidben@google.com>
This replaces v3name_test.cc which is rather difficult to follow.
v3name_test.cc ran all pairs of names against each other, used a
default case-insensitivity rule, and then had a list of string
exceptions to that rule. This is hopefully easier for us to adjust
later. It also fixes a testing bug we wouldn't notice if an expected
"exception" didn't fire.
Sadly, we cannot use designated initializers in C++ yet. MSVC does not
support them until MSVC 2019.
Change-Id: Ia8e3bf5f57d33a9bf1fc929ba1e8cd2a270a8a24
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/50725
Reviewed-by: Adam Langley <agl@google.com>
Commit-Queue: Adam Langley <agl@google.com>
Always enable X509_CHECK_FLAG_NO_PARTIAL_WILDCARDS and never enable
X509_CHECK_FLAG_MULTI_LABEL_WILDCARDS.
Update-Note: BoringSSL will no longer accept wildcard patterns like
*www.example.com or www*.example.com. (It already did not accept
ww*w.example.com.) X509_CHECK_FLAG_MULTI_LABEL_WILDCARDS will also be
ignored and can no longer be used to allow foo.bar.example.com to match
*.example.com.
Fixes: 462
Change-Id: I004e087bf70f4c3f249235cd864d9e19cc9a5102
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/50705
Reviewed-by: Adam Langley <agl@google.com>
I believe, with this, we have aligned with OpenSSL 1.1.x on the
crypto/x509 and crypto/asn1 types that are now opaque. Strangely,
OpenSSL kept X509_ALGOR public. We may wish to hide that one too later,
but we can leave it for now.
Update-Note: Use X509_REVOKED accessors rather than reaching into the
struct.
Bug: 425
Change-Id: Ib47944648a8693ed7078ffe94f7b557022debe30
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/50685
Reviewed-by: Adam Langley <agl@google.com>
rust-openssl, rather than using Rust's existing error types, exposes the
OpenSSL error queue as the error type in its public callback types.
Supporting a simplified version of ERR_set_error_data is simple enough,
so it's easiest just to add this function.
Unlike OpenSSL's, we don't attempt to support non-string error data. We
also don't try to retain borrowed pointers. If the caller did not pass
ownership, make a copy internally.
Change-Id: I909eebc2867ab1f3b9975546a106ee1f762bf516
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/50625
Reviewed-by: Adam Langley <agl@google.com>
OpenSSL 1.1.0 made this structure opaque. I don't think we particularly
need to make it opaque, but external code uses it. Also add
RSA_test_flags.
Change-Id: I136d38e72ec4664c78f4d1720ec691f5760090c1
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/50605
Reviewed-by: Adam Langley <agl@google.com>
HPKE draft-12 has no changes from draft-08 except that the test vectors
were refreshed and some fields in the JSON file renamed. Also fix the
test vector reference to point to copy from the spec rather than the
(identical) copy from the reference implementation.
Change-Id: Icd4fd467672cc8701fcd2b262ac90c5adc05ac39
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/50465
Reviewed-by: Adam Langley <agl@google.com>
The non-_ex EVP_CIPHER_CTX Final functions are a bit interesting. Unlike
EVP_DigestFinal(_ex), where the non-_ex version calls EVP_MD_CTX_cleanup
for you, the EVP_CIPHER_CTX ones do not automatically cleanup.
EVP_CipherFinal and EVP_CipherFinal_ex are identical in all releases
where they exist.
This appears to date to OpenSSL 0.9.7:
Prior to OpenSSL 0.9.7, EVP_MD_CTX and EVP_CIPHER_CTX did not use void*
data fields. Instead, they just had a union of context structures for
every algorithm OpenSSL implemented.
EVP_MD_CTX was truly cleanup-less. There were no EVP_MD_CTX_init or
EVP_MD_CTX_cleanup functions at all. EVP_DigestInit filled things in
without reference to the previous state. EVP_DigestFinal didn't cleanup
because there was nothing to cleanup.
EVP_CIPHER_CTX was also a union, but for some reason did include
EVP_CIPHER_CTX_init and EVP_CIPHER_CTX_cleanup. EVP_CIPHER_CTX_init
seemed to be optional: EVP_CipherInit with non-NULL EVP_CIPHER similarly
didn't reference the previous state. EVP_CipherFinal did not call
EVP_CIPHER_CTX_cleanup, but EVP_CIPHER_CTX_cleanup didn't do anything.
It called an optional cleanup hook on the EVP_CIPHER, but as far as I
can tell, no EVP_CIPHER implemented it.
Then OpenSSL 0.9.7 introduced ENGINE. The union didn't work anymore, so
EVP_MD_CTX and EVP_CIPHER_CTX contained void* with allocated
type-specific data. The introduced EVP_MD_CTX_init and
EVP_MD_CTX_cleanup. For (imperfect!) backwards compatibility,
EVP_DigestInit and EVP_DigestFinal transparently called init/cleanup for
you. EVP_DigestInit_ex and EVP_DigestFinal_ex became the more flexible
versions that left init/cleanup to the caller.
EVP_CIPHER_CTX got the same treatment with
EVP_CipherInit/EVP_CipherInit_ex, but *not*
EVP_CipherFinal/EVP_CipherFinal_ex. The latter did the same thing. The
history seems to be that 581f1c84940d77451c2592e9fa470893f6c3c3eb
introduced the Final/Final_ex split, with the former doing an
auto-cleanup, then 544a2aea4ba1fad76f0802fb70d92a5a8e6ad85a undid it.
Looks like the motivation is that EVP_CIPHER_CTX objects are often
reused to do multiple operations with a single key. But they missed that
the split functions are now unnecessary.
Amusingly, OpenSSL's documentation incorrectly said that EVP_CipherFinal
cleaned up after the call until it was fixed in
538860a3ce0b9fd142a7f1a62e597cccb74475d3. The fix says that some
releases cleaned up, but there were, as far as I can tell, no actual
releases with that behavior.
I've put the new Final functions in the deprecated section, purely
because there is no sense in recommending two different versions of the
same function to users, and Final_ex seems to be more popular. But there
isn't actually anything wrong with plain Final.
Change-Id: Ic2bfda48fdcf30f292141add8c5f745348036852
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/50485
Reviewed-by: Adam Langley <agl@google.com>
d2i_ASN1_OBJECT had a similar set of bugs in as in
https://boringssl-review.googlesource.com/c/boringssl/+/49866.
This does not affect any other d2i functions. Those already go through
the ASN1_ITEM machinery.
Update-Note: d2i_ASN1_OBJECT will now notice more incorrect tags. It was
already checking for tag number 6, so it is unlikely anyone was relying
on this as a non-tag-checking parser.
Change-Id: I30f9ad28e3859aeb7a38c0ea299cd2e30002abce
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/50290
Reviewed-by: Adam Langley <agl@google.com>
Update-Note: PKCS#7 and PKCS#12 parsers will now reject BER constructed
BIT STRINGs. We were previously misparsing them, as was OpenSSL. Given
how long the incorrect parse has been out there, without anyone noticing
(other parsers handle it correctly), it is unlikely these exist.
Change-Id: I61d317461cc59480dc9f772f88edc7758206d20d
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/50289
Reviewed-by: Adam Langley <agl@google.com>
This simplifies the ASN1_get_object calling convention and removes
another significant source of tasn_dec.c complexity. This change does
not affect our PKCS#7 and PKCS#12 parsers.
Update-Note: Invalid certificates (and the few external structures using
asn1t.h) with BER indefinite lengths will now be rejected.
Bug: 354
Change-Id: I723036798fc3254d0a289c77b105fcbdcda309b2
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/50287
Reviewed-by: Adam Langley <agl@google.com>
Constructed strings are a BER mechanism where a string type can be
represented as a tree of constructed nodes and primitive leaves, that
then have to be concatenated by the parser. This is prohibited in DER
and a significant source of complexity in our parser.
Note this change does not affect our PKCS#7 and PKCS#12 parsers (where
BER is sadly necessary for interop) because those use CBS.
Update-Note: Invalid certificates (and the few external structures using
asn1t.h) with BER constructed strings will now be rejected.
Bug: 354
Change-Id: I5a8ee028ec89ed4f2d5c099a0588f2029b864580
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/50286
Reviewed-by: Adam Langley <agl@google.com>
X509V3_EXT_d2i should notice if an extension has extra data at the end.
Update-Note: Some previously accepted invalid certicates may be
rejected, either in certificate verification or in X509_get_ext_d2i.
Bug: 352
Change-Id: Iacbb74a52d15bf3318b4cb8271d44b0f0a2df137
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/50285
Reviewed-by: Adam Langley <agl@google.com>
pcy_int.h was especially weird because it is even missing include guards
and its dependencies.
Change-Id: Idccfe23b74b47641bcfc802f78f3ee6fe479b781
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/50245
Commit-Queue: Adam Langley <agl@google.com>
Reviewed-by: Adam Langley <agl@google.com>
Although the compiler will hopefully optimize it out, this is
technically a VLA. The new Android NDK now warns about this.
Change-Id: Ib9f38dc73c40e90ab61105f29a635c453f1477a1
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/50185
Commit-Queue: David Benjamin <davidben@google.com>
Commit-Queue: Adam Langley <agl@google.com>
Reviewed-by: Adam Langley <agl@google.com>
When making a CRYPTO_BUFFER from a static, const buffer, there is no
need to make a copy of the data. Instead, we can reference it directly.
The hope is this will save a bit of memory in Chromium, since root store
certs will already in static data.
Moreover, by letting static CRYPTO_BUFFERs participate in pooling, we
can extend the memory savings to yet other copies of these certs. For
instance, if we make the root store updatable via component updater,
most of the updated roots will likely already be in the binary's copy.
Pooling will transparently dedup those and avoid retaining an extra
copy.
(I haven't gone as far as to give static CRYPTO_BUFFERs strong
references from the pool, since that seems odd. But something like
Chromium probably wants to intentionally leak the initial static ones so
that, when all references go away, they're still available for pooling.)
Change-Id: I05c25c5ff618f9f7a6ed21e4575cf659e7c32811
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/50045
Commit-Queue: David Benjamin <davidben@google.com>
Reviewed-by: Adam Langley <agl@google.com>
ASN1_ENCODING can be unexported because all types using it are now
hidden. This does mean external uses of <openssl/asn1t.h> can no longer
use ASN1_SEQUENCE_enc, but there do not seem to be any such uses.
ASN1_TLC and ASN1_TEMPLATE typedefs are only necessary for users of
asn1t.h. I'm hopeful we can do away with ASN1_TLC once I get to
reworking tasn_dec.c. ASN1_TEMPLATE is somewhat stuck, though all
references should be hidden behind macros.
ASN1_generate_* appear to only referenced within the library. Remove the
unused one and move the other to x509/internal.h. (asn1_gen.c is
currently in crypto/x509 rather than crypto/asn1, so I put it in
x509/internal.h to match. I'll leave figuring out that file to later.)
Annoyingly, asn1/internal.h now pulls in asn1t.h, but so it goes.
Change-Id: I8b43de3fa9647883103006e27907730d5531fd7d
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/50106
Reviewed-by: Adam Langley <agl@google.com>
Commit-Queue: David Benjamin <davidben@google.com>
There are cases where people grep binaries for strings like OpenSSL
version strings in order to detect when out-dated versions of libraries
are being used. With BoringSSL you might find "OpenSSL 1.1.1
(compatible; BoringSSL)", if the linker didn't discard it, but that's
not very helpful for knowing how up-to-date BoringSSL is because we
hardly ever change it.
This change adds a distinct random value to search for that uniquely
identifies BoringSSL and includes a rough guide to how old the BoringSSL
copy is. The linker will hopefully not discard it because it's
refereneced from |OPENSSL_malloc|.
Change-Id: Ie2259fd17a55d249a538a8a161b0d755396dd7b8
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/49885
Reviewed-by: David Benjamin <davidben@google.com>
Commit-Queue: Adam Langley <agl@google.com>
While I'm here, add missing parentheses around the B_ASN1_* bitmasks.
I've tossed ASN1_PRINTABLE into the deprecated bucket, though X509_NAME
relies on it, because it is a mess.
Bug: 407, 426
Change-Id: I287f60e98d6c9f237908011e1a816f4b4fb4433e
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/50105
Reviewed-by: Adam Langley <agl@google.com>
Commit-Queue: David Benjamin <davidben@google.com>
After https://boringssl-review.googlesource.com/c/boringssl/+/45965,
X509_VAL became largely unusable. While it did still exist as an
ASN1_ITEM and we emitted d2i/i2d/new/free functions, there is no way to
access its contents. Thus, hide it entirely.
Interestingly, although we got that to stick a while ago, I missed that
OpenSSL actually keeps X509_VAL exported, so it's possible we'll find 3p
code that uses this later. Since a standalone X509_VAL isn't especially
useful to construct or encode, this is most likely to come up in code
defining new types with <openssl/asn1t.h>.
Still, if we need to rexport this later (revert this *and* bring back
the struct), it won't be a big deal. Nothing in the public API even
constrains X509 to use X509_VAL.
Update-Note: The last remnants of the now (barely usable) X509_VAL are
no longer exported. It is unlikely anyone was relying on this.
Bug: 425
Change-Id: I90975f2f7ec27753675d2b5fa18b5cc4716319f4
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/50085
Reviewed-by: Adam Langley <agl@google.com>
Outside the library, this function is practically useless. It creates an
empty ASN1_OBJECT, which can never be filled in because the struct is
private and there are no mutating setters.
(See https://boringssl-review.googlesource.com/c/boringssl/+/46164 and
https://boringssl-review.googlesource.com/c/boringssl/+/48326 for a
discussion on why it's important ASN1_OBJECTs are immutable.)
Update-Note: ASN1_OBJECT_new is no longer exported. While this function
does remain in OpenSSL, it is extremely unlikely anyone has found a use
for this function.
Bug: 452
Change-Id: I111a9a1ce3ca4d7aa717a3c3a03d34c05af8fdbd
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/50025
Reviewed-by: Adam Langley <agl@google.com>
Commit-Queue: David Benjamin <davidben@google.com>
If the header is valid, but the body is truncated, ASN1_get_object
intentionally preserves the indefinite-length and constructed output
bits. This means callers who check for error with == 0x80 may read off
the end of the buffer on accident.
This is unlikely to break callers: 0x80 was already a possible error
value, so callers already needed to handle it. The original function's
aim in returning more information is unlikely to matter because callers
cannot distinguish 0x80 (could not parse header) and 0x80 (header was
valid, definite-length, and primitive, but length was too long).
Update-Note: ASN1_get_object's calling convention is slightly
simplified.
Bug: 451
Change-Id: If2b45c47e6b8864aef9fd5e04f313219639991ed
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/50005
Reviewed-by: Adam Langley <agl@google.com>
Commit-Queue: David Benjamin <davidben@google.com>
We generate .S files for assembly, which means they run through the C
preprocessor first. In gas targets where # is the comment marker, there
is a conflict with cpp directives.
The comments actually rely on #This and #source not being directives. If
I begin a line with "if", the build fails. Since the C preprocessor is
responsible for removing C preprocessor comments, we should be able to
safely use // everywhere with less ambiguity.
(In fact, we were already relying on this for 32-bit ARM. The 32-bit ARM
gas line comment marker is @. 64-bit ARM uses //, and x86/x86_64/ppc64
use #.)
This reportedly causes issues for goma. See
https://bugs.chromium.org/p/boringssl/issues/detail?id=448#c3
Bug: 448
Change-Id: Ib58f3152691c1dbcccfc045f21f486b56824283d
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/49965
Reviewed-by: Adam Langley <agl@google.com>
Also fill in docs for some easy ASN1_STRING wrappers while I'm here.
(Not sure why they exist, but removing them is probably more trouble
than is worth it.)
Bug: 407, 426
Change-Id: Id12c5fbc84982728435d105d66a3b63e5f3a1d15
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/49945
Commit-Queue: David Benjamin <davidben@google.com>
Reviewed-by: Adam Langley <agl@google.com>
Some BIO_write failures weren't handled. Otherwise would successfully
write truncated results. The other i2a functions all report -1 on
truncation, so match. While I'm here, write a test to make sure I didn't
break this.
Change-Id: If17d0209e75c15b3f37bceb1cdfb480fd2c62c4d
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/49931
Commit-Queue: David Benjamin <davidben@google.com>
Reviewed-by: Adam Langley <agl@google.com>
They do the same thing, except i2a_ASN1_ENUMERATED has a bug and doesn't
handle negative values.
Change-Id: Ifb22aa4e4d6c441a39cf6b3702cce7f6d12a94ae
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/49929
Commit-Queue: David Benjamin <davidben@google.com>
Reviewed-by: Adam Langley <agl@google.com>
The comparison should notice differences in bit count.
Update-Note: ASN1_STRING_cmp no longer incorrectly treats BIT STRINGs
with different padding bits as equal.
Bug: 446
Change-Id: I22b3fcc5d369540d029ca234e9b3b02402cec4c3
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/49928
Reviewed-by: Adam Langley <agl@google.com>
Commit-Queue: David Benjamin <davidben@google.com>
ASN1_item_unpack was missing checks for trailing data. ASN1_item_pack's
error handling was all wrong. (Leaking the temporary on error, checking
the the wrong return value for i2d, would-be redundant check for NULL,
were the other check not wrong.)
Update-Note: ASN1_item_unpack now checks for trailing data.
Change-Id: Ibaa19ba2b264fca36dd21109e66f9558d373c58b
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/49927
Reviewed-by: Adam Langley <agl@google.com>
Commit-Queue: Adam Langley <agl@google.com>
In doing so, fix ASN1_item_pack to not use the ASN1_OCTET_STRING
typedef. The function makes an untyped ASN1_STRING.
With all these caveats, one might think that ASN1_BOOLEAN ASN1_ITEMs are
pretty useless. This is about right. They're really only usable embedded
as a field in another struct.
Bug: 426
Change-Id: Id7830b91b2d011038ce79ec848e17ad6241423e1
Reviewed-on: https://boringssl-review.googlesource.com/c/boringssl/+/49926
Reviewed-by: Adam Langley <agl@google.com>
David Benjamin
Adam Langley
Matt Mueller