aboutsummaryrefslogtreecommitdiff
path: root/src/PushButtonSynthesis
Commit message (Collapse)AuthorAge
* partition -> Partition.partition to prevent confusion with List.partitionGravatar jadep2019-04-03
|
* fix typoGravatar jadep2019-04-03
|
* fix up imports in SmallExamples.vGravatar jadep2019-04-03
|
* update import statementsGravatar jadep2019-04-03
|
* rename some thingsGravatar jadep2019-04-03
|
* fix imports and qualifiers so everything buildsGravatar jadep2019-04-03
|
* remove extraneous module identifiersGravatar jadep2019-04-03
|
* split up Arithmetic (imports etc. not yet fixed, does not build)Gravatar jadep2019-04-03
|
* Add constr_fail and constr_fail_withGravatar Jason Gross2019-03-31
| | | | | | | | | | | | | Rather than taking the convention that failures during constr construction emit a type error from `I : I` with the actual error message `idtac`d above them (because Coq has no way to emit multiple things on stderr), we instead factor everything through a new `constr_fail` or `constr_fail_with msg_tac`, which emit more helpful messages instructing the user to look in `*coq*` or to use `Fail`/`try` to see the more informative error message. When we can make our own version that does both `idtac` and `fail` (c.f. https://github.com/coq/coq/issues/3913), then we can do something a bit more sane, hopefully.
* fix montgomeryGravatar jadep2019-03-25
|
* finish proofsGravatar jadep2019-03-25
|
* Get new Barrett proofs to generate Fancy code as beforeGravatar jadep2019-03-25
|
* adapt barrett to new glue codeGravatar jadep2019-02-21
|
* Make Qed not take foreverGravatar jadep2019-02-21
|
* start adapting Montgomery to new glue codeGravatar jadep2019-02-21
|
* Use Preconditions: Postconditions:, rather than /\ and ->Gravatar Jason Gross2019-02-02
|
* More minor improvements in docstringsGravatar Jason Gross2019-02-02
|
* Rename docstring generator based on Andres' suggestionGravatar Jason Gross2019-02-02
|
* Update with davidben's and Andres' suggestionsGravatar Jason Gross2019-02-02
|
* Drop `map λ` bits in docstringsGravatar Jason Gross2019-02-02
| | | | | | | | | They are redundant with the bounds pre- and post-conditions in WBW montgomery. Also drop the fiat_p... prefix from the `from_montgomery` bits in most of the docstrings, under the assumption that shorter strings with less repetition are more readable.
* Address code review comments to improve docstringsGravatar Jason Gross2019-02-02
|
* Add autogenerated docstrings to synthesized codeGravatar Jason Gross2019-02-02
| | | | | | | | | We now stringify the correctness conditions to generate docstrings for the synthesized code. Closes #512 Time commitment: about 6 hours
* Also display the carry chain in a commentGravatar Jason Gross2019-01-26
|
* Add better computation of carry chainGravatar Jason Gross2019-01-26
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | We port the computation of the carry chain from generate_parameters.py to Coq, for unsaturated solinas. Note that while we now bounds-check p448, we do not yet support goldilocks nor karatsuba. However, there is still an issue with the synthesized p448 code, which is that on 64-bit, it tries to use 256-bit and 512-bit integers. I'm not sure what's up with that. Partial progress towards #507 After | File Name | Before || Change | % Change ------------------------------------------------------------------------------------------ 8m51.64s | Total | 8m34.60s || +0m17.04s | +3.31% ------------------------------------------------------------------------------------------ 0m15.16s | p448_solinas_64.c | N/A || +0m15.16s | ∞ 3m09.12s | p384_32.c | 3m09.36s || -0m00.24s | -0.12% 0m44.99s | ExtractionHaskell/word_by_word_montgomery | 0m44.91s || +0m00.08s | +0.17% 0m39.58s | p521_32.c | 0m39.22s || +0m00.35s | +0.91% 0m32.54s | p521_64.c | 0m32.49s || +0m00.04s | +0.15% 0m30.87s | ExtractionHaskell/unsaturated_solinas | 0m31.04s || -0m00.16s | -0.54% 0m24.31s | ExtractionHaskell/saturated_solinas | 0m24.32s || -0m00.01s | -0.04% 0m18.62s | PushButtonSynthesis/UnsaturatedSolinas.vo | 0m17.90s || +0m00.72s | +4.02% 0m17.53s | ExtractionOCaml/word_by_word_montgomery | 0m17.44s || +0m00.08s | +0.51% 0m13.36s | secp256k1_32.c | 0m13.58s || -0m00.22s | -1.62% 0m13.21s | p256_32.c | 0m13.15s || +0m00.06s | +0.45% 0m11.47s | p484_64.c | 0m11.39s || +0m00.08s | +0.70% 0m11.27s | ExtractionOCaml/unsaturated_solinas | 0m10.71s || +0m00.55s | +5.22% 0m10.48s | ExtractionOCaml/word_by_word_montgomery.ml | 0m10.34s || +0m00.14s | +1.35% 0m07.97s | ExtractionOCaml/saturated_solinas | 0m07.98s || -0m00.01s | -0.12% 0m07.05s | ExtractionOCaml/unsaturated_solinas.ml | 0m06.98s || +0m00.06s | +1.00% 0m06.58s | ExtractionHaskell/word_by_word_montgomery.hs | 0m06.48s || +0m00.09s | +1.54% 0m06.09s | p224_32.c | 0m06.04s || +0m00.04s | +0.82% 0m05.24s | p384_64.c | 0m05.34s || -0m00.09s | -1.87% 0m05.13s | ExtractionOCaml/saturated_solinas.ml | 0m05.19s || -0m00.06s | -1.15% 0m05.00s | ExtractionHaskell/unsaturated_solinas.hs | 0m04.98s || +0m00.01s | +0.40% 0m04.14s | ExtractionHaskell/saturated_solinas.hs | 0m04.04s || +0m00.09s | +2.47% 0m02.22s | curve25519_32.c | 0m02.22s || +0m00.00s | +0.00% 0m01.49s | curve25519_64.c | 0m01.53s || -0m00.04s | -2.61% 0m01.46s | CLI.vo | 0m01.44s || +0m00.02s | +1.38% 0m01.29s | SlowPrimeSynthesisExamples.vo | 0m01.24s || +0m00.05s | +4.03% 0m01.08s | p256_64.c | 0m01.00s || +0m00.08s | +8.00% 0m01.06s | StandaloneOCamlMain.vo | 0m00.96s || +0m00.10s | +10.41% 0m01.06s | secp256k1_64.c | 0m01.17s || -0m00.10s | -9.40% 0m01.02s | p224_64.c | 0m01.08s || -0m00.06s | -5.55% 0m00.99s | StandaloneHaskellMain.vo | 0m01.08s || -0m00.09s | -8.33% 0m00.27s | TAPSort.vo | N/A || +0m00.27s | ∞
* Split up PushButtonSynthesis.vGravatar Jason Gross2019-01-18
| | | | Closes #497
* move src/Experiments/NewPipeline/ to src/Gravatar Andres Erbsen2019-01-09
generated by cgit v1.2.3 (git 2.39.1) at 2024-05-17 23:32:58 +0000