Commit message (Collapse) | Author | Age | |
---|---|---|---|
* | Pass -mbmi2 to gcc | Jason Gross | 2018-01-19 |
| | | | | | This will hopefully fix the issue where travis gcc complains about not being able to inline mulx. I hope. | ||
* | Generate fecarry for solinas | Jason Gross | 2018-01-10 |
| | | | | | | | This is a one-line change in generate_parameters.py (plus some whitespace trimming), and running `make regenerate-curves` This handles part of #294 | ||
* | changes to parameter-generation script | jadep | 2017-11-12 |
| | |||
* | clang -fbracket-depth=999999 | Andres Erbsen | 2017-11-03 |
| | |||
* | use clang | Andres Erbsen | 2017-11-03 |
| | |||
* | Make compiler*.sh stable under rebuilding | Jason Gross | 2017-11-03 |
| | | | | | Oops, we were previously relying on python's iteration order of dicts for which order we passed the various -D flags to gcc. | ||
* | Actually remove trailing whitespace from the prime in primes.txt | Jason Gross | 2017-11-03 |
| | |||
* | fibe: Only define bitwidth on the command line | Jason Gross | 2017-11-03 |
| | | | | | | | | | | We generate the PRI macros and the limb_t type from the bitwidth in the C preprocessor My reasoning here is that generate_parameters.py and the generated *.json files are more user-facing than fibe.c, so we should optimize for cleanliness of that user-facing code. If this is the wrong direction to go in, then this commit should be reverted. | ||
* | Add PRI{u,x}limb for printf formats | Jason Gross | 2017-11-03 |
| | | | | | | | | | | | | | | This is to fix ``` src/Specific/Framework/bench/fibe.c: In function ‘fe_print’: src/Specific/Framework/bench/fibe.c:130:5: warning: format ‘%llx’ expects argument of type ‘long long unsigned int’, but argument 2 has type ‘uint32_t’ [-Wformat=] printf("0x%016llx)<< %lu) + ", x[i], limb_weight_gaps[i-1]); ^ src/Specific/Framework/bench/fibe.c:130:5: warning: format ‘%lu’ expects argument of type ‘long unsigned int’, but argument 3 has type ‘uint32_t’ [-Wformat=] src/Specific/Framework/bench/fibe.c:132:3: warning: format ‘%llx’ expects argument of type ‘long long unsigned int’, but argument 2 has type ‘uint32_t’ [-Wformat=] printf("0x%016llx", x[0]); ^ ``` | ||
* | Add a24_val, a24_hex compiler params | Jason Gross | 2017-11-03 |
| | | | | This is so that we pass a24 correctly to gmp{var,sec,xx}. | ||
* | Add montgomery fesquare files | Jason Gross | 2017-11-03 |
| | |||
* | generate_parameters.py: supply dummy a24 of correct length | Andres Erbsen | 2017-11-02 |
| | |||
* | Update generate parameters to generate feadd, fesub | Jason Gross | 2017-11-02 |
| | |||
* | Fix generate_parameters | Jason Gross | 2017-11-02 |
| | |||
* | Update fibe | Jason Gross | 2017-11-02 |
| | |||
* | Fix generation of q hex string | Jason Gross | 2017-11-02 |
| | |||
* | Also define modulus_array | Jason Gross | 2017-11-02 |
| | |||
* | Add gmpxx compilation | Jason Gross | 2017-11-02 |
| | |||
* | Add -D compiler arguments for gmpxx.cpp | Jason Gross | 2017-11-02 |
| | |||
* | Don't print ".0" for integer bases in the json files | Jason Gross | 2017-10-16 |
| | |||
* | bugfixes from messy rebase; remade json files | jadep | 2017-10-16 |
| | |||
* | change limit for max # limbs to allow, add commented-out pretty-printing | jadep | 2017-10-16 |
| | |||
* | express montgomery-friendly moduli in a more script-friendly way, since we ↵ | jadep | 2017-10-16 |
| | | | | don't have extra support for them anyway | ||
* | Fix base generation | Jason Gross | 2017-10-16 |
| | |||
* | Don't error if we can't open a file | Jason Gross | 2017-10-16 |
| | |||
* | Handle more primes | Jason Gross | 2017-10-16 |
| | |||
* | Fix various issues with generate_parameters.py | Jason Gross | 2017-10-16 |
| | | | | | | | Some of the sizes were too small because python3 generates floats on division, while python2 rounds down. Also add better error handling | ||
* | generate_parameters.py: Don't overwrite files with identical contents | Jason Gross | 2017-10-16 |
| | |||
* | Prettier json file generation | Jason Gross | 2017-10-14 |
| | |||
* | Generate sz as an int in python2 | Jason Gross | 2017-10-14 |
| | |||
* | Add generated json files from generate_parameters.py | Jason Gross | 2017-10-14 |
| | |||
* | Support python2 and python3 in json generation | Jason Gross | 2017-10-14 |
| | |||
* | add file input/output to json-generating script | jadep | 2017-10-13 |
| | |||
* | add support for unsaturated limbs in json-generation | jadep | 2017-10-13 |
| | |||
* | add examples to json-generation file for reference | jadep | 2017-10-13 |
| | |||
* | add script to generate json files given a prime | jadep | 2017-10-13 |