Commit message (Collapse) | Author | Age | |
---|---|---|---|
* | Add sand-leek-cl to gitignore | David Phillips | 2018-04-24 |
| | |||
* | Fix incorrect grouping in device-side ROL | David Phillips | 2017-11-19 |
| | |||
* | Resize device result buffer | David Phillips | 2017-10-03 |
| | |||
* | Dump WIP of opencl port | David Phillips | 2017-10-03 |
| | |||
* | Add more targets to -cl | David Phillips | 2017-10-02 |
| | |||
* | Add opencl CFLAGS and targets to makefile | David Phillips | 2017-10-02 |
| | |||
* | Remove partly broken sand-leek-cl target from all | David Phillips | 2017-10-01 |
| | |||
* | Factor out magic sha1 chunk size constant | David Phillips | 2017-10-01 |
| | |||
* | Add NIH SHA1 implementation mockup for porting | David Phillips | 2017-10-01 |
| | | | | | | | | | | | Mocked this up in host code to become familiar with the algorithm, and to iron out most bugs before writing/porting to OpenCL. For laughs, I switched sand-leek to use this implementation instead of OpenSSL's. Performance decrease is about 90% with no change in CFLAGS, and about 60% with -O3 instead of -O2. Of course this holds no weight, since the implementation is only meant to be ported to OpenCL and run on graphics cards where SIMD will skyrocket performance. | ||
* | Import my CL trampoline from fractal-gen project | David Phillips | 2017-09-27 |
| | |||
* | Tidy unnecessary magic constants away | David Phillips | 2017-09-20 |
| | |||
* | Fix accidental regression with never-zero loop count | David Phillips | 2017-09-14 |
| | |||
* | Clean up test runner script | David Phillips | 2017-09-12 |
| | |||
* | Free bignum_e at exit, don't leak that memory | David Phillips | 2017-09-12 |
| | | | | Chur alyptik for reminding me | ||
* | Don't use GNU extensions on rm | David Phillips | 2017-08-23 |
| | |||
* | Thread count can never be zero, don't worry about /0 from it | David Phillips | 2017-08-09 |
| | |||
* | Remove division by zero if no loops complete | David Phillips | 2017-08-09 |
| | |||
* | Change from sh to bash for pushd | David Phillips | 2017-07-27 |
| | |||
* | Add more tests, pull simple test runner from other project | David Phillips | 2017-07-27 |
| | |||
* | Add more function docs to onion_base32.c | David Phillips | 2017-07-03 |
| | |||
* | Correct equality check in test.sh | David Phillips | 2017-06-27 |
| | |||
* | Add test recipe to travis run | David Phillips | 2017-06-27 |
| | |||
* | Add test recipe to Makefile | David Phillips | 2017-06-27 |
| | |||
* | Expand \r to \n before parsing in test.sh | David Phillips | 2017-06-27 |
| | |||
* | Don't trample worker thread output with performance string | David Phillips | 2017-06-27 |
| | |||
* | Add basic sanity check script | David Phillips | 2017-06-27 |
| | |||
* | Correct old outdated error message | David Phillips | 2017-06-27 |
| | |||
* | Add travis build label to readme | David Phillips | 2017-06-27 |
| | |||
* | Fix LDFLAGS order | David Phillips | 2017-06-27 |
| | |||
* | Fix indentation in .travis.yml | David Phillips | 2017-06-27 |
| | |||
* | Add .travis.yml for CI | David Phillips | 2017-06-27 |
| | |||
* | Change char array to unsigned to avoid (pedantic) overflow | David Phillips | 2017-06-26 |
| | |||
* | Remove outdated key sanity comment | David Phillips | 2017-06-24 |
| | |||
* | Update benchmark | David Phillips | 2017-06-20 |
| | |||
* | Remove debug spin halt | David Phillips | 2017-05-29 |
| | |||
* | Free alloced objects, move volatile to start of decl | David Phillips | 2017-05-29 |
| | |||
* | Remove unnecessary include from key_update.h | David Phillips | 2017-05-29 |
| | |||
* | Remove bitmask and raw_len computation out of tight loop | David Phillips | 2017-05-29 |
| | |||
* | Roll back two accidental changes in last commit | David Phillips | 2017-05-28 |
| | |||
* | Openssl-1.0 compat | David Phillips | 2017-05-28 |
| | |||
* | More benchmarks | David Phillips | 2017-05-28 |
| | |||
* | Update benchmarks | David Phillips | 2017-05-28 |
| | |||
* | Tidy the raw_len switch slightly | David Phillips | 2017-05-28 |
| | |||
* | Revert "Change raw_len from hard-coded to equation" | David Phillips | 2017-05-28 |
| | | | | | | This reverts commit ff244210acaf78beb6530c9f9d64a5f6965fdc75. This actually made the algorithm slightly slower | ||
* | Change raw_len from hard-coded to equation | David Phillips | 2017-05-28 |
| | |||
* | Make bitmask array static | David Phillips | 2017-05-28 |
| | |||
* | Remove the need for base32 in the inner loop | David Phillips | 2017-05-28 |
| | |||
* | Remove ssse3 algorithm, tighten decoder | David Phillips | 2017-05-28 |
| | |||
* | Remove SSSE3 note | David Phillips | 2017-05-28 |
| | |||
* | Add naive base32 decoder in preparation for new algo | David Phillips | 2017-05-28 |
| |