Add aarch64 (arm64) mpi assembly

* mpi/aarch64/mpi-asm-defs.h: New.
* mpi/aarch64/mpih-add1.S: New.
* mpi/aarch64/mpih-mul1.S: New.
* mpi/aarch64/mpih-mul2.S: New.
* mpi/aarch64/mpih-mul3.S: New.
* mpi/aarch64/mpih-sub1.S: New.
* mpi/config.links [host=aarch64-*-*]: Add configguration for aarch64
assembly.
* mpi/longlong.h [__aarch64__] (add_ssaaaa, sub_ddmmss, umul_ppmm)
(count_leading_zeros): New.
--

Add preliminary aarch64 assembly implementations for mpi.

Signed-off-by: Jussi Kivilinna <jussi.kivilinna@iki.fi>

1 files changed, 108 insertions, 0 deletions

diff --git a/mpi/aarch64/mpih-mul2.S b/mpi/aarch64/mpih-mul2.S
new file mode 100644
index 00000000..bfb35718
--- /dev/null
+++ b/mpi/aarch64/mpih-mul2.S
@@ -0,0 +1,108 @@
+/* ARM64 mul_2 -- Multiply a limb vector with a limb and add the result to
+ *                a second limb vector.
+ *
+ *      Copyright © 2013 Jussi Kivilinna <jussi.kivilinna@iki.fi>
+ *
+ * This file is part of Libgcrypt.
+ *
+ * Libgcrypt is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU Lesser General Public License as
+ * published by the Free Software Foundation; either version 2.1 of
+ * the License, or (at your option) any later version.
+ *
+ * Libgcrypt is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this program; if not, see <http://www.gnu.org/licenses/>.
+ *
+ */
+
+#include "sysdep.h"
+#include "asm-syntax.h"
+
+/*******************
+ * mpi_limb_t
+ * _gcry_mpih_addmul_1( mpi_ptr_t res_ptr,	x0
+ *		     mpi_ptr_t s1_ptr,		x1
+ *		     mpi_size_t s1_size,	x2
+ *		     mpi_limb_t s2_limb)	x3
+ */
+
+.text
+
+.globl _gcry_mpih_addmul_1
+.type  _gcry_mpih_addmul_1,%function
+_gcry_mpih_addmul_1:
+	and	x5, x2, #3;
+	mov	x6, xzr;
+	mov	x7, xzr;
+
+	cbz	x5, .Large_loop;
+
+.Loop:
+	ldr	x5, [x1], #8;
+
+	mul	x12, x5, x3;
+	ldr	x4, [x0];
+	umulh	x13, x5, x3;
+	sub	x2, x2, #1;
+
+	adds	x12, x12, x4;
+	and	x5, x2, #3;
+	adc	x13, x13, x7;
+	adds	x12, x12, x6;
+	str	x12, [x0], #8;
+	adc	x6, x7, x13;
+
+	cbz	x2, .Lend;
+	cbnz	x5, .Loop;
+
+.Large_loop:
+	ldp	x5, x9, [x1], #16;
+	sub	x2, x2, #4;
+	ldp	x4, x8, [x0];
+
+	mul	x12, x5, x3;
+	umulh	x13, x5, x3;
+
+	adds	x12, x12, x4;
+	mul	x14, x9, x3;
+	adc	x13, x13, x7;
+	adds	x12, x12, x6;
+	umulh	x15, x9, x3;
+	str	x12, [x0], #8;
+	adc	x6, x7, x13;
+
+	adds	x14, x14, x8;
+	ldp	x5, x9, [x1], #16;
+	adc	x15, x15, x7;
+	adds	x14, x14, x6;
+	mul	x12, x5, x3;
+	str	x14, [x0], #8;
+	ldp	x4, x8, [x0];
+	umulh	x13, x5, x3;
+	adc	x6, x7, x15;
+
+	adds	x12, x12, x4;
+	mul	x14, x9, x3;
+	adc	x13, x13, x7;
+	adds	x12, x12, x6;
+	umulh	x15, x9, x3;
+	str	x12, [x0], #8;
+	adc	x6, x7, x13;
+
+	adds	x14, x14, x8;
+	adc	x15, x15, x7;
+	adds	x14, x14, x6;
+	str	x14, [x0], #8;
+	adc	x6, x7, x15;
+
+	cbnz	x2, .Large_loop;
+
+.Lend:
+	mov	x0, x6;
+	ret;
+.size _gcry_mpih_addmul_1,.-_gcry_mpih_addmul_1;