Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | improve async ops | Hannes Rantzsch | 2019-04-16 | 6 | -66/+149 |
| | |||||
* | do not use make_error_code and smaller PR review issues | Tim Oesterreich | 2019-04-16 | 5 | -62/+63 |
| | |||||
* | review: use specific buffer type where applicable | Tim Oesterreich | 2019-04-16 | 5 | -25/+26 |
| | |||||
* | catch std::exception rather than (...) and exceptions as const | Hannes Rantzsch | 2019-04-16 | 4 | -32/+32 |
| | |||||
* | review: c'tors and handler type deduction in async ops | Hannes Rantzsch | 2019-04-16 | 4 | -114/+73 |
| | |||||
* | fix #6: write_some to buffer_sequence | Hannes Rantzsch | 2019-04-16 | 1 | -15/+25 |
| | |||||
* | FIX: reference to temporary | Hannes Rantzsch | 2019-04-16 | 1 | -1/+1 |
| | |||||
* | add documentation | René Meusel | 2019-04-16 | 2 | -10/+22 |
| | |||||
* | cleanup asio::ssl::stream compatibility interfaces | Hannes Rantzsch | 2019-04-16 | 2 | -77/+81 |
| | |||||
* | cleanup async ops | Hannes Rantzsch | 2019-04-16 | 3 | -9/+7 |
| | |||||
* | don't call async_write handler directly | Hannes Rantzsch | 2019-04-16 | 2 | -7/+8 |
| | |||||
* | FIX: add legal information | René Meusel | 2019-04-16 | 9 | -0/+72 |
| | |||||
* | rename member variables like 'm_***' | René Meusel | 2019-04-16 | 6 | -107/+108 |
| | |||||
* | move stream implemention into TLS namespace | René Meusel | 2019-04-16 | 8 | -54/+96 |
| | |||||
* | move stream implementation to tls subdir | René Meusel | 2019-04-16 | 10 | -18/+21 |
| | |||||
* | add a dummy c'tor for asio::ssl::context | René Meusel | 2019-04-16 | 1 | -5/+30 |
| | |||||
* | make the stream noncopyable | René Meusel | 2019-04-16 | 1 | -0/+6 |
| | |||||
* | adapt Stream<> API to be compatibile to OpenSSL | René Meusel | 2019-04-16 | 3 | -29/+168 |
| | |||||
* | fix: const copy c'tors in async ops | Hannes Rantzsch | 2019-04-16 | 3 | -3/+3 |
| | |||||
* | add asio stream implementation | Hannes Rantzsch | 2019-04-16 | 10 | -0/+966 |
| | |||||
* | Add a wrapper class that wraps macOS or Linux certificate stores. | Jack Lloyd | 2019-04-13 | 7 | -9/+145 |
| | |||||
* | Fix a few warnings in Boost specific code | Jack Lloyd | 2019-04-13 | 1 | -1/+1 |
| | |||||
* | consolidate into Flatfile_Certificate_Store | Patrick Schmidt | 2019-04-12 | 5 | -76/+26 |
| | |||||
* | add Certstore for linux | Patrick Schmidt | 2019-04-12 | 3 | -0/+61 |
| | |||||
* | add Flatfile_Certificate_Store | Patrick Schmidt | 2019-04-12 | 3 | -0/+232 |
| | |||||
* | Support verifying from a v1 root | Jack Lloyd | 2019-04-11 | 1 | -0/+9 |
| | | | | | | Discovered in #1885, the "Verisign Class 3 Public Primary Certification Authority - G3" is a v1 certificate and is still included in distro trust stores. | ||||
* | Remove tab characters | Jack Lloyd | 2019-04-10 | 8 | -509/+509 |
| | |||||
* | FIX: disable conflicting apple assertion macros | René Meusel | 2019-04-08 | 1 | -0/+1 |
| | |||||
* | Using available cores rather than total for the pools. | David Carlier | 2019-04-07 | 3 | -1/+38 |
| | | | | | | Issue with the C++ api it might not reflect the reality of the H/W, and on Android, in battery saving mode for example, might be more reasonable to use what the system really offers. | ||||
* | Add some more todos for Memory_Pool | Jack Lloyd | 2019-03-28 | 1 | -0/+20 |
| | |||||
* | Merge GH #1864 Use thread pool for XMSS signatures | Jack Lloyd | 2019-03-28 | 5 | -172/+48 |
|\ | |||||
| * | Remove unnecessary cmath includes | Jack Lloyd | 2019-03-26 | 2 | -6/+9 |
| | | |||||
| * | Remove previous runtime testing for thread counters | Jack Lloyd | 2019-03-26 | 2 | -132/+0 |
| | | |||||
| * | Use the global thread pool for XMSS signatures | Jack Lloyd | 2019-03-26 | 2 | -34/+39 |
| | | | | | | | | | | | | | | * i7-6700K (4 core w/SMT): 10% improvements * Ryzen 7 2700 (8 core w/SMT): 25% improvement except SHA-256 which is over twice as fast. * POWER8 (160 cores w/SMT): between 25 and 60% faster | ||||
* | | Fix some extra semicolons | Jack Lloyd | 2019-03-27 | 3 | -3/+3 |
| | | |||||
* | | Rename 'darwin' target to 'macos' | Jack Lloyd | 2019-03-27 | 7 | -14/+14 |
|/ | | | | | Both because that's the more common term, and because iOS/watchOS also uses the Darwin kernel, but we have a distinct target for mobile. | ||||
* | Work around problem with GCC 4.8 | Jack Lloyd | 2019-03-26 | 2 | -2/+4 |
| | |||||
* | Fix some warnings from GCC 9 | Jack Lloyd | 2019-03-25 | 3 | -3/+3 |
| | | | | New redundant-move and pessimizing-move warnings found some | ||||
* | Avoid a warning when building in single file amalgamation mode | Jack Lloyd | 2019-03-13 | 1 | -1/+1 |
| | |||||
* | Replace upper_bound with lower_bound in pgp_s2k | Evgeny Pokhilko | 2019-03-11 | 1 | -1/+1 |
| | | | | | | | | Fix: RFC4880_encode_count doesn't return consistent results when processing exact iterations. It returns RFC4880 code + 1. Update PGP_S2K_Iter test to verify PGP formula Add test to verify that encoded values match the PGP formula | ||||
* | Fix Coverity warnings | Jack Lloyd | 2019-03-07 | 2 | -20/+38 |
| | | | | | | | | Checking a ptr against null after dereferencing it. Allowing exception throw to escape a noexcept function. Both harmless. | ||||
* | Add references for these addition chains [ci skip] | Jack Lloyd | 2019-03-07 | 1 | -0/+4 |
| | |||||
* | Merge GH #1849 Prefer to_string over as_string | Jack Lloyd | 2019-03-03 | 18 | -28/+41 |
|\ | |||||
| * | Fix MSVC and avoid deprecated function call | Jack Lloyd | 2019-03-01 | 1 | -1/+1 |
| | | |||||
| * | s/as_string/to_string/ | Jack Lloyd | 2019-03-01 | 18 | -28/+41 |
| | | | | | | | | | | A few older APIs use as_string where everywhere else uses to_string. Add to_string's where missing, and deprecate X::as_string. | ||||
* | | Handle invalid public key length in Ed25519 | Jack Lloyd | 2019-03-03 | 2 | -14/+20 |
|/ | | | | Closes #1850 | ||||
* | Constify a few things in X509_Certificate::to_string() | Jack Lloyd | 2019-03-01 | 1 | -5/+12 |
| | |||||
* | Split CLI utils.cpp into more parts | Jack Lloyd | 2019-02-28 | 1 | -4/+7 |
| | | | | | | Add base58 encoding/decoding CLI Use decrypt_or_random in pk_decrypt | ||||
* | Prevent a warning with old GCC | Jack Lloyd | 2019-02-27 | 1 | -4/+5 |
| | | | | GCC 4.8 seems to dislike returns_nonnull attribute, GCC 5 is ok. | ||||
* | s/Blake2b/BLAKE2b/ | Jack Lloyd | 2019-02-22 | 2 | -18/+20 |
| | | | | As that is the proper name of the hash. Add a typedef for compat. |