 Adam Ierymenko
|
166c7978ee
ALMOST compiles for ARM64 Windows, only remaining barrier is ring in zeroidc.
|
%!s(int64=3) %!d(string=hai) anos |
 staphen
|
87d17088cc
Disable Salsa20 SSE on MinGW 32-bit compiler
|
%!s(int64=3) %!d(string=hai) anos |
 Adam Ierymenko
|
41c187ba12
Another very small crypto optimization.
|
%!s(int64=8) %!d(string=hai) anos |
|
Adam Ierymenko
|
a8ced184dc
Some code cleanup and make sure any type punning is guarded with ZT_NO_TYPE_PUNNING.
|
%!s(int64=8) %!d(string=hai) anos |
|
Adam Ierymenko
|
4938e82795
Delete junk.
|
%!s(int64=8) %!d(string=hai) anos |
|
Adam Ierymenko
|
7a94f63058
Back out NaCl since the old one with xmm6 salsa2012 does not support multi-block use and the new one is slower.
|
%!s(int64=8) %!d(string=hai) anos |
|
Adam Ierymenko
|
df48738ac9
Enable use of NaCl for faster X64 Salsa20 implementations. Also include binary for OSX for easy build. Blazingly fast.
|
%!s(int64=8) %!d(string=hai) anos |
|
Adam Ierymenko
|
43182f8f57
Docs, code cleanup, and protect the extra new fields of HELLO with encryption as a precaution.
|
%!s(int64=8) %!d(string=hai) anos |
|
Adam Ierymenko
|
719233617c
Add uint16_t key to Hashtable, and make Salsa20 zero its keyspace on destruction.
|
%!s(int64=9) %!d(string=hai) anos |
|
Adam Ierymenko
|
0c498556d5
Unroll Salsa20 fully for a little more speed (non-SSE now almost as fast as SSE)
|
%!s(int64=9) %!d(string=hai) anos |
|
Adam Ierymenko
|
789046ca57
Speed up Salsa20 just a bit.
|
%!s(int64=9) %!d(string=hai) anos |
|
Adam Ierymenko
|
f19c3c51d3
Revert slow non-SSE Salsa20 modification since it did not fix Android/ARM issue. Also update Salsa20 comments and clean up a bit.
|
%!s(int64=10) %!d(string=hai) anos |
|
Adam Ierymenko
|
a9a390a930
Apply @marning suggested SSE auto-detect to make builds easier on non-x86 platforms: https://github.com/mwarning/ZeroTierOne/commit/aa2fd044cb69b97eef1ea38d5fefd18543aae22a
|
%!s(int64=10) %!d(string=hai) anos |
|
Adam Ierymenko
|
8d2e20ede6
Get rid of __align stuff in Salsa20 -- not portable, does not seem to help much on newer chips.
|
%!s(int64=11) %!d(string=hai) anos |
|
Adam Ierymenko
|
12692c551e
SSE optimized Salsa20 -- anywhere from 20% to 50% faster than plain C version
|
%!s(int64=11) %!d(string=hai) anos |
|
Adam Ierymenko
|
612c17240a
Dead code removal, fix for cleanup GitHub issue #28
|
%!s(int64=11) %!d(string=hai) anos |
|
Adam Ierymenko
|
8c9b73f67b
Make Salsa20 variable-round, allowing for Salsa20/12 to be used for Packet encrypt and decrypt. Profiling analysis found that Salsa20 encrypt was accounting for a nontrivial percentage of CPU time, so it makes sense to cut this load fundamentally. There are no published attacks against Salsa20/12, and DJB believes 20 rounds to be overkill. This should be more than enough for our needs. Obviously incorporating ASM Salsa20 is among the next steps for performance.
|
%!s(int64=11) %!d(string=hai) anos |
|
Adam Ierymenko
|
4c06fcfc9d
More include formatting cleanup.
|
%!s(int64=12) %!d(string=hai) anos |
|
Adam Ierymenko
|
150850b800
New git repository for release - version 0.2.0 tagged
|
%!s(int64=12) %!d(string=hai) anos |
