diff --git a/arch/arc/Kconfig b/arch/arc/Kconfig
index c75d29077e4a..283099c9560a 100644
--- a/arch/arc/Kconfig
+++ b/arch/arc/Kconfig
@@ -29,7 +29,7 @@ config ARC
 	select HAVE_KPROBES
 	select HAVE_KRETPROBES
 	select HAVE_MEMBLOCK
-	select HAVE_MOD_ARCH_SPECIFIC if ARC_DW2_UNWIND
+	select HAVE_MOD_ARCH_SPECIFIC
 	select HAVE_OPROFILE
 	select HAVE_PERF_EVENTS
 	select HANDLE_DOMAIN_IRQ
diff --git a/arch/arc/include/asm/cache.h b/arch/arc/include/asm/cache.h
index b3410ff6a62d..5008021fba98 100644
--- a/arch/arc/include/asm/cache.h
+++ b/arch/arc/include/asm/cache.h
@@ -67,7 +67,7 @@ extern unsigned long perip_base, perip_end;
 #define ARC_REG_IC_PTAG_HI	0x1F
 
 /* Bit val in IC_CTRL */
-#define IC_CTRL_CACHE_DISABLE   0x1
+#define IC_CTRL_DIS		0x1
 
 /* Data cache related Auxiliary registers */
 #define ARC_REG_DC_BCR		0x72	/* Build Config reg */
@@ -80,8 +80,9 @@ extern unsigned long perip_base, perip_end;
 #define ARC_REG_DC_PTAG_HI	0x5F
 
 /* Bit val in DC_CTRL */
-#define DC_CTRL_INV_MODE_FLUSH  0x40
-#define DC_CTRL_FLUSH_STATUS    0x100
+#define DC_CTRL_DIS		0x001
+#define DC_CTRL_INV_MODE_FLUSH	0x040
+#define DC_CTRL_FLUSH_STATUS	0x100
 
 /*System-level cache (L2 cache) related Auxiliary registers */
 #define ARC_REG_SLC_CFG		0x901
@@ -92,8 +93,8 @@ extern unsigned long perip_base, perip_end;
 #define ARC_REG_SLC_RGN_END	0x916
 
 /* Bit val in SLC_CONTROL */
+#define SLC_CTRL_DIS		0x001
 #define SLC_CTRL_IM		0x040
-#define SLC_CTRL_DISABLE	0x001
 #define SLC_CTRL_BUSY		0x100
 #define SLC_CTRL_RGN_OP_INV	0x200
 
diff --git a/arch/arc/include/asm/entry-arcv2.h b/arch/arc/include/asm/entry-arcv2.h
index b5ff87e6f4b7..aee1a77934cf 100644
--- a/arch/arc/include/asm/entry-arcv2.h
+++ b/arch/arc/include/asm/entry-arcv2.h
@@ -16,6 +16,7 @@
 	;
 	; Now manually save: r12, sp, fp, gp, r25
 
+	PUSH	r30
 	PUSH	r12
 
 	; Saving pt_regs->sp correctly requires some extra work due to the way
@@ -72,6 +73,7 @@
 	POPAX	AUX_USER_SP
 1:
 	POP	r12
+	POP	r30
 
 .endm
 
diff --git a/arch/arc/include/asm/module.h b/arch/arc/include/asm/module.h
index 6e91d8b339c3..567590ea8f6c 100644
--- a/arch/arc/include/asm/module.h
+++ b/arch/arc/include/asm/module.h
@@ -14,13 +14,13 @@
 
 #include <asm-generic/module.h>
 
-#ifdef CONFIG_ARC_DW2_UNWIND
 struct mod_arch_specific {
+#ifdef CONFIG_ARC_DW2_UNWIND
 	void *unw_info;
 	int unw_sec_idx;
+#endif
 	const char *secstr;
 };
-#endif
 
 #define MODULE_PROC_FAMILY "ARC700"
 
diff --git a/arch/arc/include/asm/ptrace.h b/arch/arc/include/asm/ptrace.h
index 69095da1fcfd..47111d565a95 100644
--- a/arch/arc/include/asm/ptrace.h
+++ b/arch/arc/include/asm/ptrace.h
@@ -84,7 +84,7 @@ struct pt_regs {
 	unsigned long fp;
 	unsigned long sp;	/* user/kernel sp depending on where we came from  */
 
-	unsigned long r12;
+	unsigned long r12, r30;
 
 	/*------- Below list auto saved by h/w -----------*/
 	unsigned long r0, r1, r2, r3, r4, r5, r6, r7, r8, r9, r10, r11;
diff --git a/arch/arc/include/asm/setup.h b/arch/arc/include/asm/setup.h
index cb954cdab070..c568a9df82b1 100644
--- a/arch/arc/include/asm/setup.h
+++ b/arch/arc/include/asm/setup.h
@@ -31,6 +31,7 @@ extern int root_mountflags, end_mem;
 
 void setup_processor(void);
 void __init setup_arch_memory(void);
+long __init arc_get_mem_sz(void);
 
 /* Helpers used in arc_*_mumbojumbo routines */
 #define IS_AVAIL1(v, s)		((v) ? s : "")
diff --git a/arch/arc/kernel/intc-arcv2.c b/arch/arc/kernel/intc-arcv2.c
index 994dca7014db..ecef0fb0b66c 100644
--- a/arch/arc/kernel/intc-arcv2.c
+++ b/arch/arc/kernel/intc-arcv2.c
@@ -77,20 +77,20 @@ void arc_init_IRQ(void)
 
 static void arcv2_irq_mask(struct irq_data *data)
 {
-	write_aux_reg(AUX_IRQ_SELECT, data->irq);
+	write_aux_reg(AUX_IRQ_SELECT, data->hwirq);
 	write_aux_reg(AUX_IRQ_ENABLE, 0);
 }
 
 static void arcv2_irq_unmask(struct irq_data *data)
 {
-	write_aux_reg(AUX_IRQ_SELECT, data->irq);
+	write_aux_reg(AUX_IRQ_SELECT, data->hwirq);
 	write_aux_reg(AUX_IRQ_ENABLE, 1);
 }
 
 void arcv2_irq_enable(struct irq_data *data)
 {
 	/* set default priority */
-	write_aux_reg(AUX_IRQ_SELECT, data->irq);
+	write_aux_reg(AUX_IRQ_SELECT, data->hwirq);
 	write_aux_reg(AUX_IRQ_PRIORITY, ARCV2_IRQ_DEF_PRIO);
 
 	/*
diff --git a/arch/arc/kernel/intc-compact.c b/arch/arc/kernel/intc-compact.c
index ce9deb953ca9..8c1fd5c00782 100644
--- a/arch/arc/kernel/intc-compact.c
+++ b/arch/arc/kernel/intc-compact.c
@@ -57,7 +57,7 @@ static void arc_irq_mask(struct irq_data *data)
 	unsigned int ienb;
 
 	ienb = read_aux_reg(AUX_IENABLE);
-	ienb &= ~(1 << data->irq);
+	ienb &= ~(1 << data->hwirq);
 	write_aux_reg(AUX_IENABLE, ienb);
 }
 
@@ -66,7 +66,7 @@ static void arc_irq_unmask(struct irq_data *data)
 	unsigned int ienb;
 
 	ienb = read_aux_reg(AUX_IENABLE);
-	ienb |= (1 << data->irq);
+	ienb |= (1 << data->hwirq);
 	write_aux_reg(AUX_IENABLE, ienb);
 }
 
diff --git a/arch/arc/kernel/mcip.c b/arch/arc/kernel/mcip.c
index 560c4afc2af4..9274f8ade8c7 100644
--- a/arch/arc/kernel/mcip.c
+++ b/arch/arc/kernel/mcip.c
@@ -10,6 +10,7 @@
 
 #include <linux/smp.h>
 #include <linux/irq.h>
+#include <linux/irqchip/chained_irq.h>
 #include <linux/spinlock.h>
 #include <soc/arc/mcip.h>
 #include <asm/irqflags-arcv2.h>
@@ -221,10 +222,13 @@ static irq_hw_number_t idu_first_hwirq;
 static void idu_cascade_isr(struct irq_desc *desc)
 {
 	struct irq_domain *idu_domain = irq_desc_get_handler_data(desc);
+	struct irq_chip *core_chip = irq_desc_get_chip(desc);
 	irq_hw_number_t core_hwirq = irqd_to_hwirq(irq_desc_get_irq_data(desc));
 	irq_hw_number_t idu_hwirq = core_hwirq - idu_first_hwirq;
 
+	chained_irq_enter(core_chip, desc);
 	generic_handle_irq(irq_find_mapping(idu_domain, idu_hwirq));
+	chained_irq_exit(core_chip, desc);
 }
 
 static int idu_irq_map(struct irq_domain *d, unsigned int virq, irq_hw_number_t hwirq)
diff --git a/arch/arc/kernel/module.c b/arch/arc/kernel/module.c
index 42e964db2967..3d99a6091332 100644
--- a/arch/arc/kernel/module.c
+++ b/arch/arc/kernel/module.c
@@ -32,8 +32,8 @@ int module_frob_arch_sections(Elf_Ehdr *hdr, Elf_Shdr *sechdrs,
 #ifdef CONFIG_ARC_DW2_UNWIND
 	mod->arch.unw_sec_idx = 0;
 	mod->arch.unw_info = NULL;
-	mod->arch.secstr = secstr;
 #endif
+	mod->arch.secstr = secstr;
 	return 0;
 }
 
@@ -113,8 +113,10 @@ int apply_relocate_add(Elf32_Shdr *sechdrs,
 
 	}
 
+#ifdef CONFIG_ARC_DW2_UNWIND
 	if (strcmp(module->arch.secstr+sechdrs[tgtsec].sh_name, ".eh_frame") == 0)
 		module->arch.unw_sec_idx = tgtsec;
+#endif
 
 	return 0;
 
diff --git a/arch/arc/mm/cache.c b/arch/arc/mm/cache.c
index ec86ac0e3321..d408fa21a07c 100644
--- a/arch/arc/mm/cache.c
+++ b/arch/arc/mm/cache.c
@@ -23,7 +23,7 @@
 
 static int l2_line_sz;
 static int ioc_exists;
-int slc_enable = 1, ioc_enable = 0;
+int slc_enable = 1, ioc_enable = 1;
 unsigned long perip_base = ARC_UNCACHED_ADDR_SPACE; /* legacy value for boot */
 unsigned long perip_end = 0xFFFFFFFF; /* legacy value */
 
@@ -271,7 +271,11 @@ void __cache_line_loop_v2(phys_addr_t paddr, unsigned long vaddr,
 
 /*
  * For ARC700 MMUv3 I-cache and D-cache flushes
- * Also reused for HS38 aliasing I-cache configuration
+ *  - ARC700 programming model requires paddr and vaddr be passed in seperate
+ *    AUX registers (*_IV*L and *_PTAG respectively) irrespective of whether the
+ *    caches actually alias or not.
+ * -  For HS38, only the aliasing I-cache configuration uses the PTAG reg
+ *    (non aliasing I-cache version doesn't; while D-cache can't possibly alias)
  */
 static inline
 void __cache_line_loop_v3(phys_addr_t paddr, unsigned long vaddr,
@@ -458,6 +462,21 @@ static inline void __dc_entire_op(const int op)
 	__after_dc_op(op);
 }
 
+static inline void __dc_disable(void)
+{
+	const int r = ARC_REG_DC_CTRL;
+
+	__dc_entire_op(OP_FLUSH_N_INV);
+	write_aux_reg(r, read_aux_reg(r) | DC_CTRL_DIS);
+}
+
+static void __dc_enable(void)
+{
+	const int r = ARC_REG_DC_CTRL;
+
+	write_aux_reg(r, read_aux_reg(r) & ~DC_CTRL_DIS);
+}
+
 /* For kernel mappings cache operation: index is same as paddr */
 #define __dc_line_op_k(p, sz, op)	__dc_line_op(p, p, sz, op)
 
@@ -483,6 +502,8 @@ static inline void __dc_line_op(phys_addr_t paddr, unsigned long vaddr,
 #else
 
 #define __dc_entire_op(op)
+#define __dc_disable()
+#define __dc_enable()
 #define __dc_line_op(paddr, vaddr, sz, op)
 #define __dc_line_op_k(paddr, sz, op)
 
@@ -597,6 +618,40 @@ noinline void slc_op(phys_addr_t paddr, unsigned long sz, const int op)
 #endif
 }
 
+noinline static void slc_entire_op(const int op)
+{
+	unsigned int ctrl, r = ARC_REG_SLC_CTRL;
+
+	ctrl = read_aux_reg(r);
+
+	if (!(op & OP_FLUSH))		/* i.e. OP_INV */
+		ctrl &= ~SLC_CTRL_IM;	/* clear IM: Disable flush before Inv */
+	else
+		ctrl |= SLC_CTRL_IM;
+
+	write_aux_reg(r, ctrl);
+
+	write_aux_reg(ARC_REG_SLC_INVALIDATE, 1);
+
+	/* Important to wait for flush to complete */
+	while (read_aux_reg(r) & SLC_CTRL_BUSY);
+}
+
+static inline void arc_slc_disable(void)
+{
+	const int r = ARC_REG_SLC_CTRL;
+
+	slc_entire_op(OP_FLUSH_N_INV);
+	write_aux_reg(r, read_aux_reg(r) | SLC_CTRL_DIS);
+}
+
+static inline void arc_slc_enable(void)
+{
+	const int r = ARC_REG_SLC_CTRL;
+
+	write_aux_reg(r, read_aux_reg(r) & ~SLC_CTRL_DIS);
+}
+
 /***********************************************************
  * Exported APIs
  */
@@ -923,21 +978,54 @@ SYSCALL_DEFINE3(cacheflush, uint32_t, start, uint32_t, sz, uint32_t, flags)
 	return 0;
 }
 
-void arc_cache_init(void)
+/*
+ * IO-Coherency (IOC) setup rules:
+ *
+ * 1. Needs to be at system level, so only once by Master core
+ *    Non-Masters need not be accessing caches at that time
+ *    - They are either HALT_ON_RESET and kick started much later or
+ *    - if run on reset, need to ensure that arc_platform_smp_wait_to_boot()
+ *      doesn't perturb caches or coherency unit
+ *
+ * 2. caches (L1 and SLC) need to be purged (flush+inv) before setting up IOC,
+ *    otherwise any straggler data might behave strangely post IOC enabling
+ *
+ * 3. All Caches need to be disabled when setting up IOC to elide any in-flight
+ *    Coherency transactions
+ */
+noinline void __init arc_ioc_setup(void)
 {
-	unsigned int __maybe_unused cpu = smp_processor_id();
-	char str[256];
+	unsigned int ap_sz;
 
-	printk(arc_cache_mumbojumbo(0, str, sizeof(str)));
+	/* Flush + invalidate + disable L1 dcache */
+	__dc_disable();
+
+	/* Flush + invalidate SLC */
+	if (read_aux_reg(ARC_REG_SLC_BCR))
+		slc_entire_op(OP_FLUSH_N_INV);
+
+	/* IOC Aperture start: TDB: handle non default CONFIG_LINUX_LINK_BASE */
+	write_aux_reg(ARC_REG_IO_COH_AP0_BASE, 0x80000);
 
 	/*
-	 * Only master CPU needs to execute rest of function:
-	 *  - Assume SMP so all cores will have same cache config so
-	 *    any geomtry checks will be same for all
-	 *  - IOC setup / dma callbacks only need to be setup once
+	 * IOC Aperture size:
+	 *   decoded as 2 ^ (SIZE + 2) KB: so setting 0x11 implies 512M
+	 * TBD: fix for PGU + 1GB of low mem
+	 * TBD: fix for PAE
 	 */
-	if (cpu)
-		return;
+	ap_sz = order_base_2(arc_get_mem_sz()/1024) - 2;
+	write_aux_reg(ARC_REG_IO_COH_AP0_SIZE, ap_sz);
+
+	write_aux_reg(ARC_REG_IO_COH_PARTIAL, 1);
+	write_aux_reg(ARC_REG_IO_COH_ENABLE, 1);
+
+	/* Re-enable L1 dcache */
+	__dc_enable();
+}
+
+void __init arc_cache_init_master(void)
+{
+	unsigned int __maybe_unused cpu = smp_processor_id();
 
 	if (IS_ENABLED(CONFIG_ARC_HAS_ICACHE)) {
 		struct cpuinfo_arc_cache *ic = &cpuinfo_arc700[cpu].icache;
@@ -985,30 +1073,14 @@ void arc_cache_init(void)
 		}
 	}
 
-	if (is_isa_arcv2() && l2_line_sz && !slc_enable) {
+	/* Note that SLC disable not formally supported till HS 3.0 */
+	if (is_isa_arcv2() && l2_line_sz && !slc_enable)
+		arc_slc_disable();
 
-		/* IM set : flush before invalidate */
-		write_aux_reg(ARC_REG_SLC_CTRL,
-			read_aux_reg(ARC_REG_SLC_CTRL) | SLC_CTRL_IM);
-
-		write_aux_reg(ARC_REG_SLC_INVALIDATE, 1);
-
-		/* Important to wait for flush to complete */
-		while (read_aux_reg(ARC_REG_SLC_CTRL) & SLC_CTRL_BUSY);
-		write_aux_reg(ARC_REG_SLC_CTRL,
-			read_aux_reg(ARC_REG_SLC_CTRL) | SLC_CTRL_DISABLE);
-	}
+	if (is_isa_arcv2() && ioc_enable)
+		arc_ioc_setup();
 
 	if (is_isa_arcv2() && ioc_enable) {
-		/* IO coherency base - 0x8z */
-		write_aux_reg(ARC_REG_IO_COH_AP0_BASE, 0x80000);
-		/* IO coherency aperture size - 512Mb: 0x8z-0xAz */
-		write_aux_reg(ARC_REG_IO_COH_AP0_SIZE, 0x11);
-		/* Enable partial writes */
-		write_aux_reg(ARC_REG_IO_COH_PARTIAL, 1);
-		/* Enable IO coherency */
-		write_aux_reg(ARC_REG_IO_COH_ENABLE, 1);
-
 		__dma_cache_wback_inv = __dma_cache_wback_inv_ioc;
 		__dma_cache_inv = __dma_cache_inv_ioc;
 		__dma_cache_wback = __dma_cache_wback_ioc;
@@ -1022,3 +1094,20 @@ void arc_cache_init(void)
 		__dma_cache_wback = __dma_cache_wback_l1;
 	}
 }
+
+void __ref arc_cache_init(void)
+{
+	unsigned int __maybe_unused cpu = smp_processor_id();
+	char str[256];
+
+	printk(arc_cache_mumbojumbo(0, str, sizeof(str)));
+
+	/*
+	 * Only master CPU needs to execute rest of function:
+	 *  - Assume SMP so all cores will have same cache config so
+	 *    any geomtry checks will be same for all
+	 *  - IOC setup / dma callbacks only need to be setup once
+	 */
+	if (!cpu)
+		arc_cache_init_master();
+}
diff --git a/arch/arc/mm/init.c b/arch/arc/mm/init.c
index 399e2f223d25..8c9415ed6280 100644
--- a/arch/arc/mm/init.c
+++ b/arch/arc/mm/init.c
@@ -40,6 +40,11 @@ struct pglist_data node_data[MAX_NUMNODES] __read_mostly;
 EXPORT_SYMBOL(node_data);
 #endif
 
+long __init arc_get_mem_sz(void)
+{
+	return low_mem_sz;
+}
+
 /* User can over-ride above with "mem=nnn[KkMm]" in cmdline */
 static int __init setup_mem_sz(char *str)
 {