NFC: Move low-level Montgomery arithmetic out of bigint.
#1652
Conversation
Codecov Report
@@ Coverage Diff @@
## main #1652 +/- ##
==========================================
- Coverage 92.10% 92.09% -0.02%
==========================================
Files 132 132
Lines 18869 18869
Branches 196 196
==========================================
- Hits 17379 17377 -2
- Misses 1453 1455 +2
Partials 37 37
... and 1 file with indirect coverage changes 📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more |
briansmith
force-pushed
the
b/move-bn_mul_mont
branch
from
a0afa48
to
a17e62a
Compare
briansmith
force-pushed
the
b/move-bn_mul_mont
branch
from
a17e62a
to
2e4f0f9
Compare
|
I just pushed a new version to fix the |
briansmith
force-pushed
the
b/move-bn_mul_mont
branch
2 times, most recently
from
fc59e1e
to
1f4998d
Compare
src/arithmetic/montgomery.rs
Outdated
| // Nothing aliases `n` | ||
| let n = unsafe { core::slice::from_raw_parts(n, num_limbs) }; | ||
|
|
||
| let mut tmp = [0; 2 * MODULUS_MAX_LIMBS]; |
There was a problem hiding this comment.
I think this should be BIGINT_MODULUS_MAX_LIMBS, with this tweak it successfully builds and passes tests with --no-default-features for powerpc64le-unknown-linux-gnu.
When the `alloc` feature is disabled, on lesser-used targets we don't
build `bigint` but we still need some of the Montgomery arithmetic.
```
git diff \
HEAD^1:src/arithmetic/bigint/bn_mul_mont_fallback.rs \
src/arithmetic/montgomery.rs
```
```
git diff \
HEAD^1:src/arithmetic/bigint.rs \
src/arithmetic/montgomery.rs
```
briansmith
force-pushed
the
b/move-bn_mul_mont
branch
from
1f4998d
to
ee24519
Compare
|
@briansmith - I don't remember having an issue with In the meantime, I'll ask around with the other fuchsia people in case they remember something. |
|
@erickt Sorry, it was a typo. I meant to ping erichte-ibm. |
When the
allocfeature is disabled, on lesser-used targets we don't buildbigintbut we still need some of the Montgomery arithmetic.