Commit graph

604 commits

Author SHA1 Message Date
Henry de Valence
1e74cb3e56 Replace Scalar::from_u64 with From impls
Unfortunately, Rust selects `i32` as the type for an integer literal
when the literal has no other type constraints.  This means that someone
cannot write `Scalar::from(1)`, as Rust will choose `i32` as the type for
`1`, and we don't `impl From<i32> for Scalar`.

We could implement `From` conversions for signed integers, but since
`Scalar` operations should be constant-time by default, this would
require us to extract the sign bit of the integer and use it to
conditionally select between the positive and negative of Scalar
constructed from the value bits.  This is more expensive than the
unsigned operation, and I don't think it's what anyone really wants.

Making API consumers specify that their literals are unsigned is
slightly annoying, but better than the above alternative.

It would also be nice to change `Scalar::from_hash` to be
`impl<D: Digest<OutputSize = U64>> From<D> for Scalar`,
but this isn't currently allowed by Rust (since that `impl` "could"
conflict with the `impl From<u8>` if someone decided that `u8` should
`impl Digest`).
2018-07-19 08:39:09 -07:00
Isis Lovecruft
46c98224f5
Remove erroneous and extraneous alloc import from edwards module.
The "alloc" feature doesn't compile otherwise.

 * FIXES #160.
2018-07-17 00:28:04 +00:00
Henry de Valence
5b009a033e Remove extra line in doctest 2018-07-05 13:34:14 -07:00
Henry de Valence
0ab60b93ee Update wording on Scalar::invert to use self 2018-07-05 13:34:02 -07:00
Henry de Valence
b70b32a0c5 Change Scalar example to use the hasher functions 2018-07-05 13:27:09 -07:00
Isis Lovecruft
f43f4f9770
Update year in copyright notices to 2018. 2018-07-05 00:30:27 +00:00
Isis Lovecruft
03154d47ec
Add an example doctest for Scalar.invert(). 2018-07-05 00:30:27 +00:00
Isis Lovecruft
faf8609246
Copy the inversions of 0 warning to the invert() method. 2018-07-05 00:14:55 +00:00
Isis Lovecruft
626e070896
Document Scalar contructors with doctests. 2018-07-04 23:57:04 +00:00
Isis Lovecruft
5b263dabd0
Line wrap some docstrings in scalar.rs. 2018-07-04 21:43:13 +00:00
Isis Lovecruft
f4669c8b4d
Move the Scalar constructor documentation to the module level. 2018-07-04 21:29:36 +00:00
Isis Lovecruft
11aa71fb8d
Merge remote-tracking branch 'ebfull/sequential-montgomery-trick' into develop 2018-07-04 20:24:19 +00:00
Sean Bowe
61d6d89cd8
Fix comment describing Montgomery adjustment factor's value. 2018-07-02 10:41:45 -06:00
Sean Bowe
c4f86b231c
Only test debug assertion in batch_invert when debug assertions are enabled. 2018-07-01 15:07:04 -06:00
Sean Bowe
02af12b81a
Replace batch inversion for FieldElement with sequential variant of Montgomery's trick. 2018-07-01 15:07:04 -06:00
Sean Bowe
6294c02b52
Replace batch inversion implementation for Scalar with sequential variant of Montgomery's trick. 2018-07-01 15:07:04 -06:00
Sean Bowe
611fc40318
Add test that an empty vector field inversion returns one. 2018-06-30 16:29:02 -06:00
Sean Bowe
11b1dc142f
Add test for behavior of Scalar::batch_invert(). 2018-06-30 09:49:54 -06:00
Henry de Valence
16f39c82e5 Remove yolocrypto from avx2_backend 2018-06-18 13:32:04 -07:00
Henry de Valence
d791047aac Rewrite notes and documentation. 2018-06-18 13:22:03 -07:00
Henry de Valence
15f97221ba Suppress extraneous warnings 2018-06-15 13:44:31 -07:00
Henry de Valence
7198719419 Document bounds on FieldElement32x4 functions 2018-06-15 13:36:38 -07:00
Henry de Valence
9f5bd8c4c0 Rename reduce32 to reduce and have it return its result.
This means that all FieldElement32x4 operations return values, vs mutating interior state.
2018-06-14 15:54:36 -07:00
Henry de Valence
bab1ebbeb8 Replace scale_by_curve_constants by a Mul<(u32,u32,u32,u32)> impl 2018-06-14 15:42:20 -07:00
Henry de Valence
6ef9e9dcfb Make publicity a little more consistent 2018-06-14 15:22:16 -07:00
Henry de Valence
97292fef91 Move packing functions to top of the module 2018-06-14 15:21:51 -07:00
Henry de Valence
4dc219910a Move blend_lanes into the blend function 2018-06-14 15:13:53 -07:00
Henry de Valence
64b1b481ba Rewrite diff_sum in terms of shuffle, blend, negate 2018-06-14 14:59:11 -07:00
Henry de Valence
fe51adad31 Don't expose u32x8 unpacking functions 2018-06-14 14:32:13 -07:00
Henry de Valence
c46ec9638c Add documentation 2018-06-14 14:23:24 -07:00
Henry de Valence
eea3eadf5b Replace special-case swap_{AB,CD} methods with general shuffles 2018-06-14 14:23:24 -07:00
Henry de Valence
02296fafb5 Delete unused constant 2018-06-14 14:23:24 -07:00
Henry de Valence
25d9f3f6ca Eliminate vector constants from edwards module 2018-06-14 14:23:24 -07:00
Henry de Valence
cc8728b2a6 Add comment about rustc-constant-info to blend function 2018-06-14 14:23:24 -07:00
Henry de Valence
28f10bc183 Change Lanes::ALL to Lanes::ABCD for consistency 2018-06-14 14:23:24 -07:00
Henry de Valence
794ed5c8e3 Rewrite the doubling horrorshow 2018-06-14 14:23:24 -07:00
Henry de Valence
30a2b01c05 Add more selectors to the Lanes enum 2018-06-14 14:19:46 -07:00
Henry de Valence
64cb999866 Make platform-vector lanes constants private 2018-06-14 14:19:26 -07:00
Henry de Valence
00d8b6ea4f Change negate_D, negate_D_lazy to impl Neg, negate_lazy 2018-06-14 14:17:36 -07:00
Henry de Valence
14ce6d3da6 Add a shuffling abstraction for FieldElement32x4 2018-06-14 14:10:37 -07:00
Henry de Valence
c8dc2a6418 Implement addition for FieldElement32x4 2018-06-14 14:10:32 -07:00
Henry de Valence
bd1e3c5f3e some rustfmt changes 2018-06-14 14:05:52 -07:00
Isis Lovecruft
9a89a217f8
Merge remote-tracking branch 'dalek/multiscalar-trait-without-precomputation_r1' into develop 2018-05-15 20:23:05 +00:00
Henry de Valence
bbb64312f7 Use rand 0.5
Requires `0.5.0-pre.2`, which adds `impl CryptoRng for OsRng`.
2018-05-15 12:30:28 -07:00
Henry de Valence
bab642c5af Add doc comments 2018-05-15 11:33:38 -07:00
Henry de Valence
2eed24109e Move double-base scmul to the EdwardsPoint type 2018-05-15 11:33:38 -07:00
Henry de Valence
149c5004e8 Use multiscalar traits for the backend implementations. 2018-05-15 11:33:38 -07:00
Henry de Valence
e3bf9b0213 Add MultiscalarMul and VartimeMultiscalarMul traits.
These traits have the same interface, but with different names, so that it's
not possible to use them interchangeably.  (Constant-time and variable-time
routines should not be used interchangeably).

This commit changes the external API to use these traits, replacing
```
edwards::multiscalar_mul
edwards::vartime::multiscalar_mul
```
with
```
EdwardsPoint::multiscalar_mul (as an impl)
EdwardsPoint::vartime_multiscalar_mul (as an impl)
```
and similarly for Ristretto.

Refactoring the backend is for a later commit.

Multiscalar multiplication with precomputation is for a later commit.

The `edwards::vartime` module is retained since it's used for
`vartime_double_base_scalar_mul`.

It should be subsumed into the precomputation API in a later commit.
2018-05-15 11:33:38 -07:00
Henry de Valence
34c43c20a9 Rework backend selection code.
Each backend can now be selected by an individual feature:

- `u32_backend` for `backend::u32`;
- `u64_backend` for `backend::u64`;
- `avx2_backend` for `backend::avx2`;

The `u64_backend` is selected by default, since most people use X64 and we have
no way to select based on target (see discussion in #126).  However, these
changes mean that it is possible to select the backend explicitly, and if we
had the ability to select target-default features, we could do so easily.
2018-05-14 17:43:54 -07:00
Henry de Valence
9b6c932635 Rename 'precomputed_tables' to the more accurate 'stage2_build' 2018-05-14 15:41:45 -07:00