[arm] Add support for 64-bit ARM (Aarch64)

Signed-off-by: Michael Brown <mcb30@ipxe.org>

2 files changed, 159 insertions, 0 deletions

diff --git a/src/arch/arm64/core/arm64_bigint.c b/src/arch/arm64/core/arm64_bigint.c
new file mode 100644
index 000000000..bc4ee9a00
--- /dev/null
+++ b/src/arch/arm64/core/arm64_bigint.c
@@ -0,0 +1,103 @@
+/*
+ * Copyright (C) 2016 Michael Brown <mbrown@fensystems.co.uk>.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; either version 2 of the
+ * License, or any later version.
+ *
+ * This program 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
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ *
+ * You can also choose to distribute this program under the terms of
+ * the Unmodified Binary Distribution Licence (as given in the file
+ * COPYING.UBDL), provided that you have satisfied its requirements.
+ */
+
+FILE_LICENCE ( GPL2_OR_LATER_OR_UBDL );
+
+#include <stdint.h>
+#include <string.h>
+#include <ipxe/bigint.h>
+
+/** @file
+ *
+ * Big integer support
+ */
+
+/**
+ * Multiply big integers
+ *
+ * @v multiplicand0	Element 0 of big integer to be multiplied
+ * @v multiplier0	Element 0 of big integer to be multiplied
+ * @v result0		Element 0 of big integer to hold result
+ * @v size		Number of elements
+ */
+void bigint_multiply_raw ( const uint64_t *multiplicand0,
+			   const uint64_t *multiplier0,
+			   uint64_t *result0, unsigned int size ) {
+	const bigint_t ( size ) __attribute__ (( may_alias )) *multiplicand =
+		( ( const void * ) multiplicand0 );
+	const bigint_t ( size ) __attribute__ (( may_alias )) *multiplier =
+		( ( const void * ) multiplier0 );
+	bigint_t ( size * 2 ) __attribute__ (( may_alias )) *result =
+		( ( void * ) result0 );
+	unsigned int i;
+	unsigned int j;
+	uint64_t multiplicand_element;
+	uint64_t multiplier_element;
+	uint64_t *result_elements;
+	uint64_t discard_low;
+	uint64_t discard_high;
+	uint64_t discard_temp_low;
+	uint64_t discard_temp_high;
+
+	/* Zero result */
+	memset ( result, 0, sizeof ( *result ) );
+
+	/* Multiply integers one element at a time */
+	for ( i = 0 ; i < size ; i++ ) {
+		multiplicand_element = multiplicand->element[i];
+		for ( j = 0 ; j < size ; j++ ) {
+			multiplier_element = multiplier->element[j];
+			result_elements = &result->element[ i + j ];
+			/* Perform a single multiply, and add the
+			 * resulting double-element into the result,
+			 * carrying as necessary.  The carry can
+			 * never overflow beyond the end of the
+			 * result, since:
+			 *
+			 *     a < 2^{n}, b < 2^{n} => ab < 2^{2n}
+			 */
+			__asm__ __volatile__ ( "mul %1, %6, %7\n\t"
+					       "umulh %2, %6, %7\n\t"
+					       "ldp %3, %4, [%0]\n\t"
+					       "adds %3, %3, %1\n\t"
+					       "adcs %4, %4, %2\n\t"
+					       "stp %3, %4, [%0], #16\n\t"
+					       "bcc 2f\n\t"
+					       "\n1:\n\t"
+					       "ldr %3, [%0]\n\t"
+					       "adcs %3, %3, xzr\n\t"
+					       "str %3, [%0], #8\n\t"
+					       "bcs 1b\n\t"
+					       "\n2:\n\t"
+					       : "+r" ( result_elements ),
+						 "=&r" ( discard_low ),
+						 "=&r" ( discard_high ),
+						 "=r" ( discard_temp_low ),
+						 "=r" ( discard_temp_high ),
+						 "+m" ( *result )
+					       : "r" ( multiplicand_element ),
+						 "r" ( multiplier_element )
+					       : "cc" );
+		}
+	}
+}
diff --git a/src/arch/arm64/core/setjmp.S b/src/arch/arm64/core/setjmp.S
new file mode 100644
index 000000000..fa47aa0af
--- /dev/null
+++ b/src/arch/arm64/core/setjmp.S
@@ -0,0 +1,56 @@
+FILE_LICENCE ( GPL2_OR_LATER_OR_UBDL )
+
+	.text
+
+	/* Must match jmp_buf structure layout */
+	.struct	0
+env_x19_x20:	.quad	0, 0
+env_x21_x22:	.quad	0, 0
+env_x23_x24:	.quad	0, 0
+env_x25_x26:	.quad	0, 0
+env_x27_x28:	.quad	0, 0
+env_x29_x30:	.quad	0, 0
+env_sp:		.quad	0
+	.previous
+
+/*
+ * Save stack context for non-local goto
+ */
+	.globl	setjmp
+	.type	setjmp, %function
+setjmp:
+	/* Store registers */
+	stp	x19, x20, [x0, #env_x19_x20]
+	stp	x21, x22, [x0, #env_x21_x22]
+	stp	x23, x24, [x0, #env_x23_x24]
+	stp	x25, x26, [x0, #env_x25_x26]
+	stp	x27, x28, [x0, #env_x27_x28]
+	stp	x29, x30, [x0, #env_x29_x30]
+	mov	x16, sp
+	str	x16, [x0, #env_sp]
+	/* Return 0 when returning as setjmp() */
+	mov	x0, #0
+	ret
+	.size	setjmp, . - setjmp
+
+/*
+ * Non-local jump to a saved stack context
+ */
+	.globl	longjmp
+	.type	longjmp, %function
+longjmp:
+	/* Restore registers */
+	ldp	x19, x20, [x0, #env_x19_x20]
+	ldp	x21, x22, [x0, #env_x21_x22]
+	ldp	x23, x24, [x0, #env_x23_x24]
+	ldp	x25, x26, [x0, #env_x25_x26]
+	ldp	x27, x28, [x0, #env_x27_x28]
+	ldp	x29, x30, [x0, #env_x29_x30]
+	ldr	x16, [x0, #env_sp]
+	mov	sp, x16
+	/* Force result to non-zero */
+	cmp	w1, #0
+	csinc	w0, w1, w1, ne
+	/* Return to setjmp() caller */
+	br	x30
+	.size	longjmp, . - longjmp