arc: introduce U-Boot port for ARCv2 ISA

ARC HS and ARC EM are new cores based on ARCv2 ISA which is binary
incompatible with ISAv1 (AKA ARCompact).

Significant difference between ISAv2 and v1 is implementation of
interrupt vector table.

In v1 it is implemented in the same way as on many other architectures -
as a special location where user may put whether code executed in place
(if machine word of space is enough) or jump to a full-scale interrupt
handler.

In v2 interrupt table is just an array of adresses of real interrupt
handlers. That requires a separate section for IVT that is not encoded
as code by assembler.

This change adds support for following cores:
 * ARC EM6 (simple 32-bit microcontroller without MMU)
 * ARC HS36 (advanced 32-bit microcontroller without MMU)
 * ARC HS38 (advanced 32-bit microcontroller with MMU)

As a part of ARC HS38 new version of MMU (v4) was introduced.

Also this change adds AXS131 board which is the same DW ARC SDP base board but
with ARC HS38 CPU tile.

Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>

2 files changed, 261 insertions, 0 deletions

diff --git a/arch/arc/cpu/arcv2/Makefile b/arch/arc/cpu/arcv2/Makefile
new file mode 100644
index 0000000000..cc69e5a17e
--- /dev/null
+++ b/arch/arc/cpu/arcv2/Makefile
@@ -0,0 +1,7 @@
+#
+# Copyright (C) 2013-2015 Synopsys, Inc. All rights reserved.
+#
+# SPDX-License-Identifier:	GPL-2.0+
+#
+
+obj-y += start.o
diff --git a/arch/arc/cpu/arcv2/start.S b/arch/arc/cpu/arcv2/start.S
new file mode 100644
index 0000000000..3ce689675f
--- /dev/null
+++ b/arch/arc/cpu/arcv2/start.S
@@ -0,0 +1,254 @@
+/*
+ * Copyright (C) 2013-2015 Synopsys, Inc. All rights reserved.
+ *
+ * SPDX-License-Identifier:	GPL-2.0+
+ */
+
+#include <asm-offsets.h>
+#include <config.h>
+#include <asm/arcregs.h>
+
+/*
+ * Note on the LD/ST addressing modes with address register write-back
+ *
+ * LD.a same as LD.aw
+ *
+ * LD.a    reg1, [reg2, x]  => Pre Incr
+ *      Eff Addr for load = [reg2 + x]
+ *
+ * LD.ab   reg1, [reg2, x]  => Post Incr
+ *      Eff Addr for load = [reg2]
+ */
+
+.macro PUSH reg
+	st.a	\reg, [%sp, -4]
+.endm
+
+.macro PUSHAX aux
+	lr	%r9, [\aux]
+	PUSH	%r9
+.endm
+
+.macro  SAVE_R1_TO_R24
+	PUSH	%r1
+	PUSH	%r2
+	PUSH	%r3
+	PUSH	%r4
+	PUSH	%r5
+	PUSH	%r6
+	PUSH	%r7
+	PUSH	%r8
+	PUSH	%r9
+	PUSH	%r10
+	PUSH	%r11
+	PUSH	%r12
+	PUSH	%r13
+	PUSH	%r14
+	PUSH	%r15
+	PUSH	%r16
+	PUSH	%r17
+	PUSH	%r18
+	PUSH	%r19
+	PUSH	%r20
+	PUSH	%r21
+	PUSH	%r22
+	PUSH	%r23
+	PUSH	%r24
+.endm
+
+.macro SAVE_ALL_SYS
+	/* saving %r0 to reg->r0 in advance since weread %ecr into it */
+	st	%r0, [%sp, -8]
+	lr	%r0, [%ecr]	/* all stack addressing is manual so far */
+	st	%r0, [%sp]
+	st	%sp, [%sp, -4]
+	/* now move %sp to reg->r0 position so we can do "push" automatically */
+	sub	%sp, %sp, 8
+
+	SAVE_R1_TO_R24
+	PUSH	%r25
+	PUSH	%gp
+	PUSH	%fp
+	PUSH	%blink
+	PUSHAX	%eret
+	PUSHAX	%erstatus
+	PUSH	%lp_count
+	PUSHAX	%lp_end
+	PUSHAX	%lp_start
+	PUSHAX	%erbta
+.endm
+
+.macro SAVE_EXCEPTION_SOURCE
+#ifdef CONFIG_MMU
+	/* If MMU exists exception faulting address is loaded in EFA reg */
+	lr	%r0, [%efa]
+#else
+	/* Otherwise in ERET (exception return) reg */
+	lr	%r0, [%eret]
+#endif
+.endm
+
+.section .ivt, "a",@progbits
+.align 4
+	/* Critical system events */
+.word	_start			/* 0 - 0x000 */
+.word	memory_error		/* 1 - 0x008 */
+.word	instruction_error	/* 2 - 0x010 */
+
+	/* Exceptions */
+.word	EV_MachineCheck		/* 0x100, Fatal Machine check  (0x20) */
+.word	EV_TLBMissI		/* 0x108, Intruction TLB miss  (0x21) */
+.word	EV_TLBMissD		/* 0x110, Data TLB miss        (0x22) */
+.word	EV_TLBProtV		/* 0x118, Protection Violation (0x23)
+							or Misaligned Access  */
+.word	EV_PrivilegeV		/* 0x120, Privilege Violation  (0x24) */
+.word	EV_Trap			/* 0x128, Trap exception       (0x25) */
+.word	EV_Extension		/* 0x130, Extn Intruction Excp (0x26) */
+
+	/* Device interrupts */
+.rept	29
+	j	interrupt_handler	/* 3:31 - 0x018:0xF8 */
+.endr
+
+.text
+.globl _start
+_start:
+	/* Setup interrupt vector base that matches "__text_start" */
+	sr	__ivt_start, [ARC_AUX_INTR_VEC_BASE]
+
+	/* Setup stack pointer */
+	mov	%sp, CONFIG_SYS_INIT_SP_ADDR
+	mov	%fp, %sp
+
+	/* Clear bss */
+	mov	%r0, __bss_start
+	mov	%r1, __bss_end
+
+clear_bss:
+	st.ab	0, [%r0, 4]
+	brlt	%r0, %r1, clear_bss
+
+	/* Zero the one and only argument of "board_init_f" */
+	mov_s	%r0, 0
+	j	board_init_f
+
+memory_error:
+	SAVE_ALL_SYS
+	SAVE_EXCEPTION_SOURCE
+	mov	%r1, %sp
+	j	do_memory_error
+
+instruction_error:
+	SAVE_ALL_SYS
+	SAVE_EXCEPTION_SOURCE
+	mov	%r1, %sp
+	j	do_instruction_error
+
+interrupt_handler:
+	/* Todo - save and restore CPU context when interrupts will be in use */
+	bl	do_interrupt_handler
+	rtie
+
+EV_MachineCheck:
+	SAVE_ALL_SYS
+	SAVE_EXCEPTION_SOURCE
+	mov	%r1, %sp
+	j	do_machine_check_fault
+
+EV_TLBMissI:
+	SAVE_ALL_SYS
+	mov	%r0, %sp
+	j	do_itlb_miss
+
+EV_TLBMissD:
+	SAVE_ALL_SYS
+	mov	%r0, %sp
+	j	do_dtlb_miss
+
+EV_TLBProtV:
+	SAVE_ALL_SYS
+	SAVE_EXCEPTION_SOURCE
+	mov	%r1, %sp
+	j	do_tlb_prot_violation
+
+EV_PrivilegeV:
+	SAVE_ALL_SYS
+	mov	%r0, %sp
+	j	do_privilege_violation
+
+EV_Trap:
+	SAVE_ALL_SYS
+	mov	%r0, %sp
+	j	do_trap
+
+EV_Extension:
+	SAVE_ALL_SYS
+	mov	%r0, %sp
+	j	do_extension
+
+/*
+ * void relocate_code (addr_sp, gd, addr_moni)
+ *
+ * This "function" does not return, instead it continues in RAM
+ * after relocating the monitor code.
+ *
+ * r0 = start_addr_sp
+ * r1 = new__gd
+ * r2 = relocaddr
+ */
+.align 4
+.globl	relocate_code
+relocate_code:
+	/*
+	 * r0-r12 might be clobbered by C functions
+	 * so we use r13-r16 for storage here
+	 */
+	mov	%r13, %r0		/* save addr_sp */
+	mov	%r14, %r1		/* save addr of gd */
+	mov	%r15, %r2		/* save addr of destination */
+
+	mov	%r16, %r2		/* %r9 - relocation offset */
+	sub	%r16, %r16, __image_copy_start
+
+/* Set up the stack */
+stack_setup:
+	mov	%sp, %r13
+	mov	%fp, %sp
+
+/* Check if monitor is loaded right in place for relocation */
+	mov	%r0, __image_copy_start
+	cmp	%r0, %r15		/* skip relocation if code loaded */
+	bz	do_board_init_r		/* in target location already */
+
+/* Copy data (__image_copy_start - __image_copy_end) to new location */
+	mov	%r1, %r15
+	mov	%r2, __image_copy_end
+	sub	%r2, %r2, %r0		/* r3 <- amount of bytes to copy */
+	asr	%r2, %r2, 2		/* r3 <- amount of words to copy */
+	mov	%lp_count, %r2
+	lp	copy_end
+	ld.ab	%r2,[%r0,4]
+	st.ab	%r2,[%r1,4]
+copy_end:
+
+/* Fix relocations related issues */
+	bl	do_elf_reloc_fixups
+#ifndef CONFIG_SYS_ICACHE_OFF
+	bl	invalidate_icache_all
+#endif
+#ifndef CONFIG_SYS_DCACHE_OFF
+	bl	flush_dcache_all
+#endif
+
+/* Update position of intterupt vector table */
+	lr	%r0, [ARC_AUX_INTR_VEC_BASE]	/* Read current position */
+	add	%r0, %r0, %r16			/* Update address */
+	sr	%r0, [ARC_AUX_INTR_VEC_BASE]	/* Write new position */
+
+do_board_init_r:
+/* Prepare for exection of "board_init_r" in relocated monitor */
+	mov	%r2, board_init_r	/* old address of "board_init_r()" */
+	add	%r2, %r2, %r16		/* new address of "board_init_r()" */
+	mov	%r0, %r14		/* 1-st parameter: gd_t */
+	mov	%r1, %r15		/* 2-nd parameter: dest_addr */
+	j	[%r2]