Henry de Valence
e693d7f020
fixup AVX2 ascii blocks so they don't run as doctests
2019-02-12 12:52:02 -08:00
Henry de Valence
8b0ad2b03d
Add vector implementation of precomputation.
2019-02-12 12:52:02 -08:00
Henry de Valence
c6acdfd5e2
Add serial implementation of precomputation.
2019-02-12 12:52:02 -08:00
Henry de Valence
5daff66079
Move cfgs into outer module.
2019-02-12 12:52:02 -08:00
Henry de Valence
00675b4c56
Add benchmarks for precomputed multiscalar multiplication.
2019-02-12 12:52:02 -08:00
Henry de Valence
98e713ef91
Add a trait for multiscalar multiplication with precomputation.
2019-02-12 12:52:02 -08:00
Sean Bowe
d71b6650d6
Check the equality of EdwardsPoints in the projective coordinates to avoid expensive compressions.
2019-02-06 17:45:25 -07:00
Henry de Valence
ff0dc4a3db
Merge branch 'master' into develop
2019-01-27 23:19:44 -08:00
Henry de Valence
4bdccd7b7c
Merge branch 'release/1.0.3'
2019-01-27 23:18:35 -08:00
Henry de Valence
9cd4551b40
Bump patch version
2019-01-27 23:17:27 -08:00
Henry de Valence
33e7d26513
Merge branch 'add-compressed-cteq-impls' into release/1.0.3
2019-01-27 23:16:48 -08:00
Henry de Valence
5ddc30061c
Implement ConstantTimeEq on compressed points.
2019-01-27 23:13:58 -08:00
Henry de Valence
6d7d94372e
Merge pull request #227 from isislovecruft/fix/weird-pub
...
I found a weird pub, not the kind you get cidre in.
2019-01-24 14:12:10 -08:00
Isis Lovecruft
0d23260a75
I found a weird pub, not the kind you get cidre in.
2019-01-24 21:36:00 +00:00
Henry de Valence
8e5062521f
Merge pull request #225 from isislovecruft/fix/invsqrt-doc-typos
...
Fix two typos in FieldElement::invsqrt documentation
2019-01-21 18:50:14 -08:00
Isis Lovecruft
4270d4ff5f
Fix two typos in FieldElement::invsqrt documentation.
2019-01-22 01:01:08 +00:00
Henry de Valence
353e3fe81a
Merge pull request #215 from hdevalence/ifma
...
Restructure vector backends and add a new IFMA backend
2019-01-19 21:51:13 -08:00
Henry de Valence
e3ca8961fc
Fix rustdoc feature flags to compile on stable.
2019-01-18 14:19:39 -08:00
Henry de Valence
9ddbf1ba83
Add notes on the IFMA implementation.
2019-01-18 14:19:39 -08:00
Henry de Valence
dc8431429f
Refactor documentation to use doc_cfg.
2019-01-18 14:19:39 -08:00
Henry de Valence
5700add507
Clean up warnings.
2019-01-18 14:19:30 -08:00
Henry de Valence
fa42708257
Change naming scheme for multiplication accumulators.
2019-01-18 14:19:30 -08:00
Henry de Valence
a84f9eafe4
Allow non_snake_case in the IFMA field and edwards modules.
2019-01-18 14:19:30 -08:00
Henry de Valence
06766bfdea
Delay shuffles that cross the "Great Wall of Intel".
2019-01-18 14:19:30 -08:00
Henry de Valence
33aa544cf4
Add IFMA constants file.
2019-01-18 14:19:30 -08:00
Henry de Valence
f825426ff4
Add impl Default, Identity for ExtendedPoint, CachedPoint.
2019-01-18 14:19:30 -08:00
Henry de Valence
da62569355
Add lookup table support.
2019-01-18 14:19:30 -08:00
Henry de Valence
6cec313f16
Add impl of Sub and Neg for Edwards points.
2019-01-18 14:19:30 -08:00
Henry de Valence
6faaef21df
Add mul_by_pow_2 for IFMA
2019-01-18 14:19:30 -08:00
Henry de Valence
8e38ff2859
Add Edwards doubling implementation.
2019-01-18 14:19:30 -08:00
Henry de Valence
c4617b28a9
Add squaring implementation.
2019-01-18 14:19:30 -08:00
Henry de Valence
a62571b003
Reduce field elements during conversion to vector form.
...
The IFMA types have tighter bounds than the u64 types.
2019-01-18 14:19:30 -08:00
Henry de Valence
3f21bb3b58
Add skeleton of parallel Edwards formulas.
2019-01-18 14:19:30 -08:00
Henry de Valence
5880fced2b
Add skeleton of shuffle and blend functions.
2019-01-18 14:19:30 -08:00
Henry de Valence
d90bb779ca
Delay all computations using x[0], y[0]
...
Since computation of the 0 term in reduction requires a multiplication with a
4-cycle latency, this ensures that the rest of the computation can start before
the 0 term is finished
2019-01-18 14:19:30 -08:00
Henry de Valence
c925c76342
Change the strategy for reducing the high half of the product.
2019-01-18 14:19:30 -08:00
Henry de Valence
607b839032
Change size of limbs in the reduction test.
2019-01-18 14:19:30 -08:00
Henry de Valence
14163131f9
Lift the reduction state of the field elements into the type system.
...
This splits the `FieldElement51x4` type into two types:
- `F51x4Reduced` (with reduced limbs)
- `F51x4Unreduced` (with unreduced limbs)
The reduction is implemented as a `From` impl to convert one type to the other.
The output of a multiplication is now a `F51x4Unreduced`. The reason is that
the inputs to IFMA operations must be at most 52 bits, so it's not possible to
perform an addition of (51+epsilon)-bit values and still be small enough to be
used as an input to multiplication. So, it doesn't make sense to perform a
reduction at the end of a multiplication, because the reduced values will be
fed into an addition or subtraction, which then needs to be re-reduced.
2019-01-18 14:19:30 -08:00
Henry de Valence
ff49673214
Add impl Mul<(u32, u32, u32, u32)> for FieldElement51x4.
2019-01-18 14:19:30 -08:00
Henry de Valence
70199d6094
Add impl Mul<FieldElement51x4> for FieldElement51x4.
2019-01-18 14:19:30 -08:00
Henry de Valence
aa73d7b1bc
Add a FieldElement51x4 type.
2019-01-18 14:19:30 -08:00
Henry de Valence
3233e2eac6
Fix: cfg_target_feature is stable since 1.27.
2019-01-18 14:19:20 -08:00
Henry de Valence
be77af1cd8
Remove unnecessary feature gate.
...
The vector backend doesn't generate constants in build.rs, so this isn't necessary.
2019-01-18 14:18:55 -08:00
Henry de Valence
e91cc685b1
Don't generate unused constants in build.rs.
2019-01-18 14:15:28 -08:00
Henry de Valence
47a164da4d
Replace avx2_backend with simd_backend (autoselects AVX2/IFMA)
2019-01-18 01:52:17 -08:00
Henry de Valence
9ed2128a10
Add stub code for IFMA intrinsics in Rust.
2019-01-18 01:52:17 -08:00
Henry de Valence
0a97f5fe8c
Don't compile the serial scalar_mul when using the vector one.
2019-01-18 01:52:07 -08:00
Henry de Valence
f1d2b5182b
Restructure source tree into serial and vector backends.
...
This begins to attempt to restructure the source tree so that the common parts
are common and the different parts are different.
The backend is now split into two parts:
- serial (containing the implementation using serial formulas and mixed-model arithmetic).
- vector (containing the implementation using parallel formulas and single-model arithmetic).
The serial scalar_mul tree is now under backend::serial::scalar_mul.
The avx2 scalar_mul tree is now under backend::avx2::scalar_mul.
2019-01-18 01:49:40 -08:00
Henry de Valence
94cb3e7842
Rename field types based on radix.
...
`FieldElement32` -> `FieldElement2625`
`FieldElement64` -> `FieldElement51`
`Scalar32` -> `Scalar29`
`Scalar64` -> `Scalar52`
This naming is more accurate and would let us add an ADX backend later.
2019-01-17 22:21:55 -08:00
Henry de Valence
4814242d00
Merge branch 'master' into develop
2019-01-17 19:26:50 -08:00
