ARCv2: Support for ARCv2 ISA and HS38x cores

The notable features are:
    - SMP configurations of upto 4 cores with coherency
    - Optional L2 Cache and IO-Coherency
    - Revised Interrupt Architecture (multiple priorites, reg banks,
        auto stack switch, auto regfile save/restore)
    - MMUv4 (PIPT dcache, Huge Pages)
    - Instructions for
	* 64bit load/store: LDD, STD
	* Hardware assisted divide/remainder: DIV, REM
	* Function prologue/epilogue: ENTER_S, LEAVE_S
	* IRQ enable/disable: CLRI, SETI
	* pop count: FFS, FLS
	* SETcc, BMSKN, XBFU...

Signed-off-by: Vineet Gupta <vgupta@synopsys.com>

8 files changed, 278 insertions, 21 deletions

diff --git a/arch/arc/kernel/Makefile b/arch/arc/kernel/Makefile
index cc929c0..0be7ba0 100644
--- a/arch/arc/kernel/Makefile
+++ b/arch/arc/kernel/Makefile
@@ -10,7 +10,8 @@ CFLAGS_ptrace.o		+= -DUTS_MACHINE='"$(UTS_MACHINE)"'
 
 obj-y	:= arcksyms.o setup.o irq.o time.o reset.o ptrace.o process.o devtree.o
 obj-y	+= signal.o traps.o sys.o troubleshoot.o stacktrace.o disasm.o clk.o
-obj-y	+= entry-compact.o intc-compact.o
+obj-$(CONFIG_ISA_ARCOMPACT)		+= entry-compact.o intc-compact.o
+obj-$(CONFIG_ISA_ARCV2)			+= entry-arcv2.o intc-arcv2.o
 
 obj-$(CONFIG_MODULES)			+= arcksyms.o module.o
 obj-$(CONFIG_SMP) 			+= smp.o
diff --git a/arch/arc/kernel/entry-arcv2.S b/arch/arc/kernel/entry-arcv2.S
new file mode 100644
index 0000000..c59a396
--- /dev/null
+++ b/arch/arc/kernel/entry-arcv2.S
@@ -0,0 +1,189 @@
+/*
+ * ARCv2 ISA based core Low Level Intr/Traps/Exceptions(non-TLB) Handling
+ *
+ * Copyright (C) 2013 Synopsys, Inc. (www.synopsys.com)
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/linkage.h>   /* ARC_{EXTRY,EXIT} */
+#include <asm/entry.h>       /* SAVE_ALL_{INT1,INT2,TRAP...} */
+#include <asm/errno.h>
+#include <asm/arcregs.h>
+#include <asm/irqflags.h>
+
+	.cpu HS
+
+#define VECTOR	.word
+
+;############################ Vector Table #################################
+
+	.section .vector,"a",@progbits
+	.align 4
+
+# Initial 16 slots are Exception Vectors
+VECTOR	stext			; Restart Vector (jump to entry point)
+VECTOR	mem_service		; Mem exception
+VECTOR	instr_service		; Instrn Error
+VECTOR	EV_MachineCheck		; Fatal Machine check
+VECTOR	EV_TLBMissI		; Intruction TLB miss
+VECTOR	EV_TLBMissD		; Data TLB miss
+VECTOR	EV_TLBProtV		; Protection Violation
+VECTOR	EV_PrivilegeV		; Privilege Violation
+VECTOR	EV_SWI			; Software Breakpoint
+VECTOR	EV_Trap			; Trap exception
+VECTOR	EV_Extension		; Extn Instruction Exception
+VECTOR	EV_DivZero		; Divide by Zero
+VECTOR	EV_DCError		; Data Cache Error
+VECTOR	EV_Misaligned		; Misaligned Data Access
+VECTOR	reserved		; Reserved slots
+VECTOR	reserved		; Reserved slots
+
+# Begin Interrupt Vectors
+VECTOR	handle_interrupt	; (16) Timer0
+VECTOR	handle_interrupt	; unused (Timer1)
+VECTOR	handle_interrupt	; unused (WDT)
+VECTOR	handle_interrupt	; (19) ICI (inter core interrupt)
+VECTOR	handle_interrupt
+VECTOR	handle_interrupt
+VECTOR	handle_interrupt
+VECTOR	handle_interrupt	; (23) End of fixed IRQs
+
+.rept CONFIG_ARC_NUMBER_OF_INTERRUPTS - 8
+	VECTOR	handle_interrupt
+.endr
+
+	.section .text, "ax",@progbits
+
+res_service:		; processor restart
+	flag    0x1     ; not implemented
+	nop
+	nop
+
+reserved:		; processor restart
+	rtie            ; jump to processor initializations
+
+;##################### Interrupt Handling ##############################
+
+ENTRY(handle_interrupt)
+
+	INTERRUPT_PROLOGUE  irq
+
+	clri		; To make status32.IE agree with CPU internal state
+
+	lr  r0, [ICAUSE]
+
+	mov   blink, ret_from_exception
+
+	b.d  arch_do_IRQ
+	mov r1, sp
+
+END(handle_interrupt)
+
+;################### Non TLB Exception Handling #############################
+
+ENTRY(EV_SWI)
+	flag 1
+END(EV_SWI)
+
+ENTRY(EV_DivZero)
+	flag 1
+END(EV_DivZero)
+
+ENTRY(EV_DCError)
+	flag 1
+END(EV_DCError)
+
+ENTRY(EV_Misaligned)
+
+	EXCEPTION_PROLOGUE
+
+	lr  r0, [efa]	; Faulting Data address
+	mov r1, sp
+
+	FAKE_RET_FROM_EXCPN
+
+	SAVE_CALLEE_SAVED_USER
+	mov r2, sp              ; callee_regs
+
+	bl  do_misaligned_access
+
+	; TBD: optimize - do this only if a callee reg was involved
+	; either a dst of emulated LD/ST or src with address-writeback
+	RESTORE_CALLEE_SAVED_USER
+
+	b   ret_from_exception
+END(EV_Misaligned)
+
+; ---------------------------------------------
+; Protection Violation Exception Handler
+; ---------------------------------------------
+
+ENTRY(EV_TLBProtV)
+
+	EXCEPTION_PROLOGUE
+
+	lr  r0, [efa]	; Faulting Data address
+	mov r1, sp	; pt_regs
+
+	FAKE_RET_FROM_EXCPN
+
+	mov blink, ret_from_exception
+	b   do_page_fault
+
+END(EV_TLBProtV)
+
+; From Linux standpoint Slow Path I/D TLB Miss is same a ProtV as they
+; need to call do_page_fault().
+; ECR in pt_regs provides whether access was R/W/X
+
+.global        call_do_page_fault
+.set call_do_page_fault, EV_TLBProtV
+
+;############# Common Handlers for ARCompact and ARCv2 ##############
+
+#include "entry.S"
+
+;############# Return from Intr/Excp/Trap (ARCv2 ISA Specifics) ##############
+;
+; Restore the saved sys context (common exit-path for EXCPN/IRQ/Trap)
+; IRQ shd definitely not happen between now and rtie
+; All 2 entry points to here already disable interrupts
+
+.Lrestore_regs:
+
+	ld	r0, [sp, PT_status32]	; U/K mode at time of entry
+	lr	r10, [AUX_IRQ_ACT]
+
+	bmsk	r11, r10, 15	; AUX_IRQ_ACT.ACTIVE
+	breq	r11, 0, .Lexcept_ret	; No intr active, ret from Exception
+
+;####### Return from Intr #######
+
+debug_marker_l1:
+	; Handle special case #1: (Entry via Exception, Return via IRQ)
+	;
+	; Exception in U mode, preempted in kernel, Intr taken (K mode), orig
+	; task now returning to U mode (riding the Intr)
+	; AUX_IRQ_ACTIVE won't have U bit set (since intr in K mode), hence SP
+	; won't be switched to correct U mode value (from AUX_SP)
+	; So force AUX_IRQ_ACT.U for such a case
+
+	btst	r0, STATUS_U_BIT		; Z flag set if K (Z clear for U)
+	bset.nz	r11, r11, AUX_IRQ_ACT_BIT_U	; NZ means U
+	sr	r11, [AUX_IRQ_ACT]
+
+	INTERRUPT_EPILOGUE  irq
+	rtie
+
+;####### Return from Exception / pure kernel mode #######
+
+.Lexcept_ret:	; Expects r0 has PT_status32
+
+debug_marker_syscall:
+	EXCEPTION_EPILOGUE
+	rtie
+
+END(ret_from_exception)
diff --git a/arch/arc/kernel/head.S b/arch/arc/kernel/head.S
index 64a92e0..812f95e 100644
--- a/arch/arc/kernel/head.S
+++ b/arch/arc/kernel/head.S
@@ -49,8 +49,6 @@
 1:
 .endm
 
-	.cpu A7
-
 	.section .init.text, "ax",@progbits
 	.type stext, @function
 	.globl stext
diff --git a/arch/arc/kernel/process.c b/arch/arc/kernel/process.c
index b5426ba..5156043 100644
--- a/arch/arc/kernel/process.c
+++ b/arch/arc/kernel/process.c
@@ -44,7 +44,17 @@ SYSCALL_DEFINE0(arc_gettls)
 void arch_cpu_idle(void)
 {
 	/* sleep, but enable all interrupts before committing */
-	__asm__("sleep 0x3");
+	if (is_isa_arcompact()) {
+		__asm__("sleep 0x3");
+	} else {
+		/* default irq priority (<=) which can interrupt the doze */
+		const int arg = 0x10 | ARCV2_IRQ_DEF_PRIO;
+
+		__asm__ __volatile__(
+		"sleep %0	\n"
+		:
+		:"r"(arg));
+	}
 }
 
 asmlinkage void ret_from_fork(void);
@@ -166,7 +176,7 @@ void start_thread(struct pt_regs * regs, unsigned long pc, unsigned long usp)
 	 * [L] ZOL loop inhibited to begin with - cleared by a LP insn
 	 * Interrupts enabled
 	 */
-	regs->status32 = STATUS_U_MASK | STATUS_L_MASK | STATUS_IE_MASK;
+	regs->status32 = STATUS_U_MASK | STATUS_L_MASK | ISA_INIT_STATUS_BITS;
 
 	/* bogus seed values for debugging */
 	regs->lp_start = 0x10;
@@ -196,8 +206,11 @@ int elf_check_arch(const struct elf32_hdr *x)
 {
 	unsigned int eflags;
 
-	if (x->e_machine != EM_ARCOMPACT)
+	if (x->e_machine != EM_ARC_INUSE) {
+		pr_err("ELF not built for %s ISA\n",
+			is_isa_arcompact() ? "ARCompact":"ARCv2");
 		return 0;
+	}
 
 	eflags = x->e_flags;
 	if ((eflags & EF_ARC_OSABI_MSK) < EF_ARC_OSABI_CURRENT) {
diff --git a/arch/arc/kernel/ptrace.c b/arch/arc/kernel/ptrace.c
index 4dd9e3a..4442204 100644
--- a/arch/arc/kernel/ptrace.c
+++ b/arch/arc/kernel/ptrace.c
@@ -200,7 +200,7 @@ static const struct user_regset arc_regsets[] = {
 
 static const struct user_regset_view user_arc_view = {
 	.name		= UTS_MACHINE,
-	.e_machine	= EM_ARCOMPACT,
+	.e_machine	= EM_ARC_INUSE,
 	.regsets	= arc_regsets,
 	.n		= ARRAY_SIZE(arc_regsets)
 };
diff --git a/arch/arc/kernel/setup.c b/arch/arc/kernel/setup.c
index 96d4480..d6fe800 100644
--- a/arch/arc/kernel/setup.c
+++ b/arch/arc/kernel/setup.c
@@ -96,7 +96,7 @@ static void read_arc_build_cfg_regs(void)
 	read_decode_mmu_bcr();
 	read_decode_cache_bcr();
 
-	{
+	if (is_isa_arcompact()) {
 		struct bcr_fp_arcompact sp, dp;
 		struct bcr_bpu_arcompact bpu;
 
@@ -112,6 +112,19 @@ static void read_arc_build_cfg_regs(void)
 			cpu->bpu.num_cache = 256 << (bpu.ent - 1);
 			cpu->bpu.num_pred = 256 << (bpu.ent - 1);
 		}
+	} else {
+		struct bcr_fp_arcv2 spdp;
+		struct bcr_bpu_arcv2 bpu;
+
+		READ_BCR(ARC_REG_FP_V2_BCR, spdp);
+		cpu->extn.fpu_sp = spdp.sp ? 1 : 0;
+		cpu->extn.fpu_dp = spdp.dp ? 1 : 0;
+
+		READ_BCR(ARC_REG_BPU_BCR, bpu);
+		cpu->bpu.ver = bpu.ver;
+		cpu->bpu.full = bpu.ft;
+		cpu->bpu.num_cache = 256 << bpu.bce;
+		cpu->bpu.num_pred = 2048 << bpu.pte;
 	}
 
 	READ_BCR(ARC_REG_AP_BCR, bcr);
@@ -131,6 +144,7 @@ static const struct cpuinfo_data arc_cpu_tbl[] = {
 	{ {0x30, "ARC 700"      }, 0x33},
 	{ {0x34, "ARC 700 R4.10"}, 0x34},
 	{ {0x35, "ARC 700 R4.11"}, 0x35},
+	{ {0x50, "ARC HS38"	}, 0x51},
 	{ {0x00, NULL		} }
 };
 
@@ -149,13 +163,17 @@ static char *arc_cpu_mumbojumbo(int cpu_id, char *buf, int len)
 
 	FIX_PTR(cpu);
 
-	{
+	if (is_isa_arcompact()) {
 		isa_nm = "ARCompact";
 		be = IS_ENABLED(CONFIG_CPU_BIG_ENDIAN);
 
 		atomic = cpu->isa.atomic1;
 		if (!cpu->isa.ver)	/* ISA BCR absent, use Kconfig info */
 			atomic = IS_ENABLED(CONFIG_ARC_HAS_LLSC);
+	} else {
+		isa_nm = "ARCv2";
+		be = cpu->isa.be;
+		atomic = cpu->isa.atomic;
 	}
 
 	n += scnprintf(buf + n, len - n,
@@ -184,14 +202,31 @@ static char *arc_cpu_mumbojumbo(int cpu_id, char *buf, int len)
 		       IS_AVAIL1(cpu->timers.t0, "Timer0 "),
 		       IS_AVAIL1(cpu->timers.t1, "Timer1 "));
 
-	n += i = scnprintf(buf + n, len - n, "%s%s",
-			   IS_AVAIL2(atomic, "atomic ", CONFIG_ARC_HAS_LLSC));
+	n += i = scnprintf(buf + n, len - n, "%s%s%s%s%s",
+			   IS_AVAIL2(atomic, "atomic ", CONFIG_ARC_HAS_LLSC),
+			   IS_AVAIL2(cpu->isa.ldd, "ll64 ", CONFIG_ARC_HAS_LL64),
+			   IS_AVAIL1(cpu->isa.unalign, "unalign (not used)"));
 
 	if (i)
 		n += scnprintf(buf + n, len - n, "\n\t\t: ");
 
+	if (cpu->extn_mpy.ver) {
+		if (cpu->extn_mpy.ver <= 0x2) {	/* ARCompact */
+			n += scnprintf(buf + n, len - n, "mpy ");
+		} else {
+			int opt = 2;	/* stock MPY/MPYH */
+
+			if (cpu->extn_mpy.dsp)	/* OPT 7-9 */
+				opt = cpu->extn_mpy.dsp + 6;
+
+			n += scnprintf(buf + n, len - n, "mpy[opt %d] ", opt);
+		}
+		n += scnprintf(buf + n, len - n, "%s",
+			       IS_USED(CONFIG_ARC_HAS_HW_MPY));
+	}
+
 	n += scnprintf(buf + n, len - n, "%s%s%s%s%s%s%s%s\n",
-		       IS_AVAIL1(cpu->extn_mpy.ver, "mpy "),
+		       IS_AVAIL1(cpu->isa.div_rem, "div_rem "),
 		       IS_AVAIL1(cpu->extn.norm, "norm "),
 		       IS_AVAIL1(cpu->extn.barrel, "barrel-shift "),
 		       IS_AVAIL1(cpu->extn.swap, "swap "),
diff --git a/arch/arc/kernel/signal.c b/arch/arc/kernel/signal.c
index b15d2fe..004b7f0 100644
--- a/arch/arc/kernel/signal.c
+++ b/arch/arc/kernel/signal.c
@@ -336,7 +336,7 @@ static void arc_restart_syscall(struct k_sigaction *ka, struct pt_regs *regs)
 		 * their orig user space value when we ret from kernel
 		 */
 		regs->r0 = regs->orig_r0;
-		regs->ret -= 4;
+		regs->ret -= is_isa_arcv2() ? 2 : 4;
 		break;
 	}
 }
@@ -377,10 +377,10 @@ void do_signal(struct pt_regs *regs)
 		if (regs->r0 == -ERESTARTNOHAND ||
 		    regs->r0 == -ERESTARTSYS || regs->r0 == -ERESTARTNOINTR) {
 			regs->r0 = regs->orig_r0;
-			regs->ret -= 4;
+			regs->ret -= is_isa_arcv2() ? 2 : 4;
 		} else if (regs->r0 == -ERESTART_RESTARTBLOCK) {
 			regs->r8 = __NR_restart_syscall;
-			regs->ret -= 4;
+			regs->ret -= is_isa_arcv2() ? 2 : 4;
 		}
 		syscall_wont_restart(regs);	/* No more restarts */
 	}
diff --git a/arch/arc/kernel/troubleshoot.c b/arch/arc/kernel/troubleshoot.c
index e00a018..e0cf998 100644
--- a/arch/arc/kernel/troubleshoot.c
+++ b/arch/arc/kernel/troubleshoot.c
@@ -14,6 +14,7 @@
 #include <linux/proc_fs.h>
 #include <linux/file.h>
 #include <asm/arcregs.h>
+#include <asm/irqflags.h>
 
 /*
  * Common routine to print scratch regs (r0-r12) or callee regs (r13-r25)
@@ -34,7 +35,10 @@ static noinline void print_reg_file(long *reg_rev, int start_num)
 			n += scnprintf(buf + n, len - n, "\n");
 
 		/* because pt_regs has regs reversed: r12..r0, r25..r13 */
-		reg_rev--;
+		if (is_isa_arcv2() && start_num == 0)
+			reg_rev++;
+		else
+			reg_rev--;
 	}
 
 	if (start_num != 0)
@@ -152,6 +156,15 @@ static void show_ecr_verbose(struct pt_regs *regs)
 				((cause_code == 0x02) ? "Write" : "EX"));
 	} else if (vec == ECR_V_INSN_ERR) {
 		pr_cont("Illegal Insn\n");
+#ifdef CONFIG_ISA_ARCV2
+	} else if (vec == ECR_V_MEM_ERR) {
+		if (cause_code == 0x00)
+			pr_cont("Bus Error from Insn Mem\n");
+		else if (cause_code == 0x10)
+			pr_cont("Bus Error from Data Mem\n");
+		else
+			pr_cont("Bus Error, check PRM\n");
+#endif
 	} else {
 		pr_cont("Check Programmer's Manual\n");
 	}
@@ -185,12 +198,20 @@ void show_regs(struct pt_regs *regs)
 
 	pr_info("[STAT32]: 0x%08lx", regs->status32);
 
-#define STS_BIT(r, bit)	r->status32 & STATUS_##bit##_MASK ? #bit : ""
-	if (!user_mode(regs))
-		pr_cont(" : %2s %2s %2s %2s %2s\n",
-			STS_BIT(regs, AE), STS_BIT(regs, A2), STS_BIT(regs, A1),
-			STS_BIT(regs, E2), STS_BIT(regs, E1));
+#define STS_BIT(r, bit)	r->status32 & STATUS_##bit##_MASK ? #bit" " : ""
 
+#ifdef CONFIG_ISA_ARCOMPACT
+	pr_cont(" : %2s%2s%2s%2s%2s%2s%2s\n",
+			(regs->status32 & STATUS_U_MASK) ? "U " : "K ",
+			STS_BIT(regs, DE), STS_BIT(regs, AE),
+			STS_BIT(regs, A2), STS_BIT(regs, A1),
+			STS_BIT(regs, E2), STS_BIT(regs, E1));
+#else
+	pr_cont(" : %2s%2s%2s%2s\n",
+			STS_BIT(regs, IE),
+			(regs->status32 & STATUS_U_MASK) ? "U " : "K ",
+			STS_BIT(regs, DE), STS_BIT(regs, AE));
+#endif
 	pr_info("BTA: 0x%08lx\t SP: 0x%08lx\t FP: 0x%08lx\n",
 		regs->bta, regs->sp, regs->fp);
 	pr_info("LPS: 0x%08lx\tLPE: 0x%08lx\tLPC: 0x%08lx\n",