RFR: 8350459: MontgomeryIntegerPolynomialP256 multiply intrinsic with AVX2 on x86_64

[Sandhya Viswanathan]

On Thu, 20 Feb 2025 21:49:42 GMT, Volodymyr Paprotski <vpaprotski at openjdk.org> wrote:

> Add AVX2 montgomery multiplication intrinsic. (About 60-80% gain)
> 
> Also add reduction to existing AVX512 multiplication (this was left-over from https://github.com/openjdk/jdk/pull/19893 where a quick fix was required). This is mostly for cleanup, but there is about 1-2% gain.
> 
> Before (no AVX512)
> 
> Benchmark                        (algorithm)  (dataSize)  (keyLength)  (provider)   Mode  Cnt      Score     Error  Units
> SignatureBench.ECDSA.sign    SHA256withECDSA        1024          256              thrpt   40   3720.589 ±  17.879  ops/s
> SignatureBench.ECDSA.sign    SHA256withECDSA       16384          256              thrpt   40   3605.940 ±  15.807  ops/s
> SignatureBench.ECDSA.verify  SHA256withECDSA        1024          256              thrpt   40   1076.502 ±   4.190  ops/s
> SignatureBench.ECDSA.verify  SHA256withECDSA       16384          256              thrpt   40   1069.624 ±   2.484  ops/s
> Benchmark                             (algorithm)  (keyLength)  (kpgAlgorithm)  (provider)   Mode  Cnt     Score   Error  Units
> KeyAgreementBench.EC.generateSecret          ECDH          256              EC              thrpt   40   830.448 ± 2.285  ops/s
> 
> After (with AVX2)
> 
> Benchmark                        (algorithm)  (dataSize)  (keyLength)  (provider)   Mode  Cnt      Score     Error  Units
> SignatureBench.ECDSA.sign    SHA256withECDSA        1024          256              thrpt   40   6000.496 ±  39.923  ops/s
> SignatureBench.ECDSA.sign    SHA256withECDSA       16384          256              thrpt   40   5739.878 ±  34.838  ops/s
> SignatureBench.ECDSA.verify  SHA256withECDSA        1024          256              thrpt   40   1942.437 ±  12.179  ops/s
> SignatureBench.ECDSA.verify  SHA256withECDSA       16384          256              thrpt   40   1921.770 ±   8.992  ops/s
> Benchmark                             (algorithm)  (keyLength)  (kpgAlgorithm)  (provider)   Mode  Cnt     Score   Error  Units
> KeyAgreementBench.EC.generateSecret          ECDH          256              EC              thrpt   40  1399.761 ± 6.238  ops/s
> 
> 
> Before (with AVX512):
> 
> Benchmark                        (algorithm)  (dataSize)  (keyLength)  (provider)   Mode  Cnt       Score     Error  Units
> SignatureBench.ECDSA.sign    SHA256withECDSA        1024          256              thrpt   40    9621.950 ±  27.260  ops/s
> SignatureBench.ECDSA.sign    SHA256withECDSA       16384          256              thrpt   40    8975.654 ±  26.707  ops/s
> SignatureBench.ECDSA.verify  SHA256withECDSA        102...

src/java.base/share/classes/sun/security/util/math/intpoly/MontgomeryIntegerPolynomialP256.java line 423:

> 421:         r[2] = ((c7 & mask) | (c2 & ~mask));
> 422:         r[3] = ((c8 & mask) | (c3 & ~mask));
> 423:         r[4] = ((c9 & mask) | (c4 & ~mask));

It would be good to add a comment here indicating that if the result (c9 - c5) had overflown by one modulus, result - modulus (c4-c0) would be positive else it would be negative.  i.e. Upper bits of c4 would be all zeroes on overflow otherwise upper bits of c4 would be all ones.  Thus on overflow, return value "r" should be set to result - modulus (c4 - c0) else it should be set to result (c9-c5).

test/jdk/com/sun/security/util/math/intpoly/MontgomeryPolynomialFuzzTest.java line 2:

> 1: /*
> 2:  * Copyright (c) 2025, Intel Corporation. All rights reserved.

This should be Copyright (c) 2024, 2025, Intel Corporation. All rights reserved.

-------------

PR Review Comment: https://git.openjdk.org/jdk/pull/23719#discussion_r1972301843
PR Review Comment: https://git.openjdk.org/jdk/pull/23719#discussion_r1972267785