k256/p256: use new SignPrimitive API; add BlindedScalar#99
Merged
Conversation
tarcieri
force-pushed
the
ecdsa-update-and-blinded-scalar
branch
from
e273c86 to
87efa5a
Compare
Updates both crates with the `SignPrimitive` changes from RustCrypto/signatures#107. These changes removed the `masking_scalar` from the trait, and replaced it with trait bounds that make it possible to substitute a blinded scalar. In the `p256` crate (which is the only one that presently implements a variable-time inversion) the `masking_scalar` is replaced with a `BlindedScalar` type that implements the previous blinded inversion. The implementation it uses (including `Scalar::invert_vartime`) could potentially be made generic and extracted into the `elliptic-curve` crate, allowing it to be used with any curve which implements the arithmetic primitives used in the blinded inversion implementation. However, for now, this PR leaves it in `p256`, and therefore at least has feature parity with the old implementation.
tarcieri
force-pushed
the
ecdsa-update-and-blinded-scalar
branch
from
87efa5a to
634f262
Compare
Codecov Report
@@ Coverage Diff @@
## master #99 +/- ##
==========================================
- Coverage 53.08% 52.83% -0.26%
==========================================
Files 18 19 +1
Lines 3415 3435 +20
==========================================
+ Hits 1813 1815 +2
- Misses 1602 1620 +18
Continue to review full report at Codecov.
|
tarcieri
commented
Jul 28, 2020
Comment on lines
+116
to
+127
| fn scalar_blinding() { | ||
| let vector = &ECDSA_TEST_VECTORS[0]; | ||
| let d = Scalar::from_bytes(vector.d.try_into().unwrap()).unwrap(); | ||
| let k = Scalar::from_bytes(vector.k.try_into().unwrap()).unwrap(); | ||
| let k_blinded = BlindedScalar::new(k, &mut OsRng); | ||
| let sig = d | ||
| .try_sign_prehashed(&k_blinded, GenericArray::from_slice(vector.m)) | ||
| .unwrap(); | ||
|
|
||
| assert_eq!(vector.r, sig.r().as_slice()); | ||
| assert_eq!(vector.s, sig.s().as_slice()); | ||
| } |
Member
Author
There was a problem hiding this comment.
FYI @nickray, here's the proposed alternative for masking_scalar.
I had it take an RNG argument directly, but we could also add an API to construct one from two Scalar values you generate however you want (perhaps that could be new, and the RNG-based one could be generate)
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Updates both crates with the
SignPrimitivechanges from RustCrypto/signatures#107.These changes removed the
masking_scalarargument from the trait, and replaced it with trait bounds that make it possible to substitute a blinded scalar.In the
p256crate (which is the only one that presently implements a variable-time inversion) themasking_scalaris replaced with aBlindedScalartype that implements the previous blinded inversion.The implementation it uses (including
Scalar::invert_vartime) could potentially be made generic and extracted into theelliptic-curvecrate, allowing it to be used with any curve which implements the arithmetic primitives used in the blinded inversion implementation.However, for now, this PR leaves it in
p256, and therefore at least has feature parity with the old implementation.