Merge pull request #6947 from SparkiDev/sp_arm64_mont_red_4_fix

SP ARM64 asm: fix Montgomery reduction by 4

Author: JacobBarthelmeh

wolfcrypt/src/sp_arm64.c

@@ -40512,6 +40512,8 @@ static void sp_256_mont_mul_order_4(sp_digit* r, const sp_digit* a, const sp_dig
         "adcs	x11, x11, x3\n\t"
         "umulh	x4, x16, x17\n\t"
         "adcs	x12, x12, x4\n\t"
+        "# x15 == -1\n\t"
+        "adcs	x19, x19, x15\n\t"
         "csel	x13, x13, xzr, cs\n\t"
         "csel	x14, x14, xzr, cs\n\t"
         "csel	x15, x15, xzr, cs\n\t"
@@ -40703,6 +40705,8 @@ static void sp_256_mont_sqr_order_4(sp_digit* r, const sp_digit* a)
         "adcs	x10, x10, x2\n\t"
         "umulh	x3, x15, x16\n\t"
         "adcs	x11, x11, x3\n\t"
+        "# x14 == -1\n\t"
+        "adcs	x17, x17, x14\n\t"
         "csel	x12, x12, xzr, cs\n\t"
         "csel	x13, x13, xzr, cs\n\t"
         "csel	x14, x14, xzr, cs\n\t"
@@ -40889,6 +40893,8 @@ static void sp_256_mont_sqr_n_order_4(sp_digit* r, const sp_digit* a, int n)
         "adcs	x10, x10, x2\n\t"
         "umulh	x3, x15, x16\n\t"
         "adcs	x11, x11, x3\n\t"
+        "# x14 == -1\n\t"
+        "adcs	x17, x17, x14\n\t"
         "csel	x12, x12, xzr, cs\n\t"
         "csel	x13, x13, xzr, cs\n\t"
         "csel	x14, x14, xzr, cs\n\t"