Age | Commit message (Collapse) | Author |
|
|
|
libnativeheader exports headers under nativeheader. These were
available before incorrectly as global headers in order to give
access to jni.h.
Test: modules using external/scrypt find headers
Bug: 63762847
Change-Id: I9baf896015a1920f8a26daa2a0c549169628890f
|
|
|
|
* Owners are selected from top CL approvals or owners.
They will be suggested to review/approve future CLs.
* OWNERS files are recognized by the new find-owners plugin, see .md files in
https://gerrit.googlesource.com/plugins/find-owners/+/master/src/main/resources/Documentation/
Test: build/make/tools/checkowners.py -c -v OWNERS
Change-Id: I41d1076af8926ae570cffd7f3341a4c76f2dc196
|
|
am: b442b13773
am: 32be729d37
Change-Id: I13fec30c977a102b14e94b326ee292a64f79a17f
|
|
am: b442b13773
Change-Id: I1b431113bfa436546c18f4bd361766ec237274d4
|
|
am: 4b2bd9665c
Change-Id: Icd4ce38a012e91a709de21b1c377c580cdc12599
|
|
|
|
There's no need to use a different name for target and host moudles. In
Soong, it's better to use the same for both, as target and host modules
can be defined at the same time.
Change-Id: I47f356b4fdde5b3a0b75862b1e9d91723ca06f27
|
|
3608597959
am: 540c0274be
* commit '540c0274bedb28050223c305f9a5499b06df328e':
scrypt doesn't need <machine/cpu-features.h>.
Change-Id: I6242ca18c85c3cd98b9f4a558d54f934c6c5409b
|
|
am: 3608597959
* commit '36085979592293853cb18fadb61358f5b7f4f717':
scrypt doesn't need <machine/cpu-features.h>.
Change-Id: I724c161acbb6b2caa6ba9d438c1325e95ad4b830
|
|
am: e6e09c358c
* commit 'e6e09c358c249191cf3afa0e44db95b45ea332c4':
scrypt doesn't need <machine/cpu-features.h>.
Change-Id: Ia87611af260d74930786c044475da06928ab5d6b
|
|
|
|
Bug: http://b/18556103
Change-Id: Ibccef334b4d1a2e68e327fc8b73e62e62d13df32
|
|
|
|
Bug: http://b/16574165
Change-Id: Ieb63fb6a73f0fb2ab3260bc5eabca7d86d892fbd
|
|
|
|
The libcrypto and libssl modules (and their respective static and host
versions) use LOCAL_EXPORT_C_INCLUDE_DIRS thus just including the module
is sufficient.
Change-Id: I1bc36094ecbc4ed2d9a4c62498f05580770f4a8a
|
|
|
|
The build system does this for us. Since libgtest is the stlport
version of the library, this causes build breakages when libc++ is the
default STL.
Bug: 15193147
Change-Id: Ief6f539de36f38d38bd1460b17efcdc09114d64d
|
|
|
|
Since some of the test's dependencies use clang, ASAN_ALL will cause
those to be built with ASAN instrumentation, meaning the final
executable needs to be linked with ASAN's runtime libraries.
The simple fix is to just move the test to clang to let it link with
ASAN when ASAN_ALL is provided.
Change-Id: I8d134667347ec45fe59d557e3c792520eaf62f1b
|
|
|
|
Change-Id: I075938a7bcbe66aac7c6d0ab874ed8816a89a2c9
|
|
|
|
Change-Id: I5697885dec49f14d717ab0e5f2cbe618e724c2d1
|
|
|
|
full_eng is a target that doesn't emulate NEON. Presumably because qemu
makes NEON slower than non-NEON.
Change-Id: I2153a62da6aef7efa75eae9892822af861ef4476
|
|
|
|
Reference implementation:
ScryptTest.TestVectors (2588 ms)
NEON acceleration:
ScryptTest.TestVectors (1169 ms)
Change-Id: I23974b2f0f64435696e75acfbdee1115a699abae
|
|
|
|
Make sure our scrypt library is passing the test vectors provided at:
http://tools.ietf.org/html/draft-josefsson-scrypt-kdf-00#page-11
This will allow us to substitute in new Salsa20/8, PBKDF2, etc
implementations that are faster without breaking compatibility.
Change-Id: I89024b637e0075e5018526e4147148e6fc8e4d96
|
|
|
|
bionic did not have SIZE_MAX in stdint.h but this was fixed by bionic
commit 7c89506e3af3ee97691a98329f0d7b5edef49218 so we no longer need
this patch.
Change-Id: I49006dafac6c09742004d2cf10bcf76b8ac2bcb5
|
|
* changes:
Use OpenSSL library for PBKDF2 functions
Import initial version of scrypt library
|
|
Change-Id: Ib3e9fcbda26ee95b26c4e4d1c9a4aba78b3ac713
|
|
Change-Id: Idcc14fed4878763ad26855cd99c5e6093ab5b3dd
|
|
|