crypto/internal/fips140/nistec/fiat: optimize P256mul on riscv64

src/crypto/internal/fips140/nistec/fiat/benchmark_test.go

@@ -34,6 +34,14 @@ func BenchmarkMul(b *testing.B) {
 			v.Mul(v, v)
 		}
 	})
+	b.Run("P256", func(b *testing.B) {
+		v := new(fiat.P256Element).One()
+		b.ReportAllocs()
+		b.ResetTimer()
+		for i := 0; i < b.N; i++ {
+			v.Mul(v, v)
+		}
+	})
 }
 
 func BenchmarkSquare(b *testing.B) {

src/crypto/internal/fips140/nistec/fiat/p256_impl.go

@@ -0,0 +1,11 @@
+// Copyright 2024 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+//go:build !riscv64
+
+package fiat
+
+func p256Mul(out, a, b *p256MontgomeryDomainFieldElement) {
+	p256MulGeneric(out, a, b)
+}

src/crypto/internal/fips140/nistec/fiat/p256_impl_riscv64.go

@@ -0,0 +1,10 @@
+// Copyright 2024 The Go Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+//go:build !purego
+
+package fiat
+
+//go:noescape
+func p256Mul(out, a, b *p256MontgomeryDomainFieldElement)