diff --git a/arch/arc/Kconfig b/arch/arc/Kconfig
index 376366a7db81..d750b302d5ab 100644
--- a/arch/arc/Kconfig
+++ b/arch/arc/Kconfig
@@ -191,7 +191,6 @@ config NR_CPUS
 
 config ARC_SMP_HALT_ON_RESET
 	bool "Enable Halt-on-reset boot mode"
-	default y if ARC_UBOOT_SUPPORT
 	help
 	  In SMP configuration cores can be configured as Halt-on-reset
 	  or they could all start at same time. For Halt-on-reset, non
@@ -407,6 +406,14 @@ config ARC_HAS_ACCL_REGS
 	  (also referred to as r58:r59). These can also be used by gcc as GPR so
 	  kernel needs to save/restore per process
 
+config ARC_IRQ_NO_AUTOSAVE
+	bool "Disable hardware autosave regfile on interrupts"
+	default n
+	help
+	  On HS cores, taken interrupt auto saves the regfile on stack.
+	  This is programmable and can be optionally disabled in which case
+	  software INTERRUPT_PROLOGUE/EPILGUE do the needed work
+
 endif	# ISA_ARCV2
 
 endmenu   # "ARC CPU Configuration"
@@ -515,17 +522,6 @@ config ARC_DBG_TLB_PARANOIA
 
 endif
 
-config ARC_UBOOT_SUPPORT
-	bool "Support uboot arg Handling"
-	help
-	  ARC Linux by default checks for uboot provided args as pointers to
-	  external cmdline or DTB. This however breaks in absence of uboot,
-	  when booting from Metaware debugger directly, as the registers are
-	  not zeroed out on reset by mdb and/or ARCv2 based cores. The bogus
-	  registers look like uboot args to kernel which then chokes.
-	  So only enable the uboot arg checking/processing if users are sure
-	  of uboot being in play.
-
 config ARC_BUILTIN_DTB_NAME
 	string "Built in DTB"
 	help
diff --git a/arch/arc/configs/nps_defconfig b/arch/arc/configs/nps_defconfig
index 6e84060e7c90..621f59407d76 100644
--- a/arch/arc/configs/nps_defconfig
+++ b/arch/arc/configs/nps_defconfig
@@ -31,7 +31,6 @@ CONFIG_ARC_CACHE_LINE_SHIFT=5
 # CONFIG_ARC_HAS_LLSC is not set
 CONFIG_ARC_KVADDR_SIZE=402
 CONFIG_ARC_EMUL_UNALIGNED=y
-CONFIG_ARC_UBOOT_SUPPORT=y
 CONFIG_PREEMPT=y
 CONFIG_NET=y
 CONFIG_UNIX=y
diff --git a/arch/arc/configs/vdk_hs38_defconfig b/arch/arc/configs/vdk_hs38_defconfig
index 1e59a2e9c602..e447ace6fa1c 100644
--- a/arch/arc/configs/vdk_hs38_defconfig
+++ b/arch/arc/configs/vdk_hs38_defconfig
@@ -13,7 +13,6 @@ CONFIG_PARTITION_ADVANCED=y
 CONFIG_ARC_PLAT_AXS10X=y
 CONFIG_AXS103=y
 CONFIG_ISA_ARCV2=y
-CONFIG_ARC_UBOOT_SUPPORT=y
 CONFIG_ARC_BUILTIN_DTB_NAME="vdk_hs38"
 CONFIG_PREEMPT=y
 CONFIG_NET=y
diff --git a/arch/arc/configs/vdk_hs38_smp_defconfig b/arch/arc/configs/vdk_hs38_smp_defconfig
index b5c3f6c54b03..c82cdb10aaf4 100644
--- a/arch/arc/configs/vdk_hs38_smp_defconfig
+++ b/arch/arc/configs/vdk_hs38_smp_defconfig
@@ -15,8 +15,6 @@ CONFIG_AXS103=y
 CONFIG_ISA_ARCV2=y
 CONFIG_SMP=y
 # CONFIG_ARC_TIMERS_64BIT is not set
-# CONFIG_ARC_SMP_HALT_ON_RESET is not set
-CONFIG_ARC_UBOOT_SUPPORT=y
 CONFIG_ARC_BUILTIN_DTB_NAME="vdk_hs38_smp"
 CONFIG_PREEMPT=y
 CONFIG_NET=y
diff --git a/arch/arc/include/asm/arcregs.h b/arch/arc/include/asm/arcregs.h
index f1b86cef0905..a27eafdc8260 100644
--- a/arch/arc/include/asm/arcregs.h
+++ b/arch/arc/include/asm/arcregs.h
@@ -151,6 +151,14 @@ struct bcr_isa_arcv2 {
 #endif
 };
 
+struct bcr_uarch_build_arcv2 {
+#ifdef CONFIG_CPU_BIG_ENDIAN
+	unsigned int pad:8, prod:8, maj:8, min:8;
+#else
+	unsigned int min:8, maj:8, prod:8, pad:8;
+#endif
+};
+
 struct bcr_mpy {
 #ifdef CONFIG_CPU_BIG_ENDIAN
 	unsigned int pad:8, x1616:8, dsp:4, cycles:2, type:2, ver:8;
diff --git a/arch/arc/include/asm/cache.h b/arch/arc/include/asm/cache.h
index f393b663413e..2ad77fb43639 100644
--- a/arch/arc/include/asm/cache.h
+++ b/arch/arc/include/asm/cache.h
@@ -52,6 +52,17 @@
 #define cache_line_size()	SMP_CACHE_BYTES
 #define ARCH_DMA_MINALIGN	SMP_CACHE_BYTES
 
+/*
+ * Make sure slab-allocated buffers are 64-bit aligned when atomic64_t uses
+ * ARCv2 64-bit atomics (LLOCKD/SCONDD). This guarantess runtime 64-bit
+ * alignment for any atomic64_t embedded in buffer.
+ * Default ARCH_SLAB_MINALIGN is __alignof__(long long) which has a relaxed
+ * value of 4 (and not 8) in ARC ABI.
+ */
+#if defined(CONFIG_ARC_HAS_LL64) && defined(CONFIG_ARC_HAS_LLSC)
+#define ARCH_SLAB_MINALIGN	8
+#endif
+
 extern void arc_cache_init(void);
 extern char *arc_cache_mumbojumbo(int cpu_id, char *buf, int len);
 extern void read_decode_cache_bcr(void);
diff --git a/arch/arc/include/asm/entry-arcv2.h b/arch/arc/include/asm/entry-arcv2.h
index 309f4e6721b3..225e7df2d8ed 100644
--- a/arch/arc/include/asm/entry-arcv2.h
+++ b/arch/arc/include/asm/entry-arcv2.h
@@ -17,6 +17,33 @@
 	;
 	; Now manually save: r12, sp, fp, gp, r25
 
+#ifdef CONFIG_ARC_IRQ_NO_AUTOSAVE
+.ifnc \called_from, exception
+	st.as	r9, [sp, -10]	; save r9 in it's final stack slot
+	sub	sp, sp, 12	; skip JLI, LDI, EI
+
+	PUSH	lp_count
+	PUSHAX	lp_start
+	PUSHAX	lp_end
+	PUSH	blink
+
+	PUSH	r11
+	PUSH	r10
+
+	sub	sp, sp, 4	; skip r9
+
+	PUSH	r8
+	PUSH	r7
+	PUSH	r6
+	PUSH	r5
+	PUSH	r4
+	PUSH	r3
+	PUSH	r2
+	PUSH	r1
+	PUSH	r0
+.endif
+#endif
+
 #ifdef CONFIG_ARC_HAS_ACCL_REGS
 	PUSH	r59
 	PUSH	r58
@@ -86,6 +113,33 @@
 	POP	r59
 #endif
 
+#ifdef CONFIG_ARC_IRQ_NO_AUTOSAVE
+.ifnc \called_from, exception
+	POP	r0
+	POP	r1
+	POP	r2
+	POP	r3
+	POP	r4
+	POP	r5
+	POP	r6
+	POP	r7
+	POP	r8
+	POP	r9
+	POP	r10
+	POP	r11
+
+	POP	blink
+	POPAX	lp_end
+	POPAX	lp_start
+
+	POP	r9
+	mov	lp_count, r9
+
+	add	sp, sp, 12	; skip JLI, LDI, EI
+	ld.as	r9, [sp, -10]	; reload r9 which got clobbered
+.endif
+#endif
+
 .endm
 
 /*------------------------------------------------------------------------*/
diff --git a/arch/arc/include/asm/uaccess.h b/arch/arc/include/asm/uaccess.h
index c9173c02081c..eabc3efa6c6d 100644
--- a/arch/arc/include/asm/uaccess.h
+++ b/arch/arc/include/asm/uaccess.h
@@ -207,7 +207,7 @@ raw_copy_from_user(void *to, const void __user *from, unsigned long n)
 		*/
 		  "=&r" (tmp), "+r" (to), "+r" (from)
 		:
-		: "lp_count", "lp_start", "lp_end", "memory");
+		: "lp_count", "memory");
 
 		return n;
 	}
@@ -433,7 +433,7 @@ raw_copy_to_user(void __user *to, const void *from, unsigned long n)
 		 */
 		  "=&r" (tmp), "+r" (to), "+r" (from)
 		:
-		: "lp_count", "lp_start", "lp_end", "memory");
+		: "lp_count", "memory");
 
 		return n;
 	}
@@ -653,7 +653,7 @@ static inline unsigned long __arc_clear_user(void __user *to, unsigned long n)
 	"	.previous			\n"
 	: "+r"(d_char), "+r"(res)
 	: "i"(0)
-	: "lp_count", "lp_start", "lp_end", "memory");
+	: "lp_count", "memory");
 
 	return res;
 }
@@ -686,7 +686,7 @@ __arc_strncpy_from_user(char *dst, const char __user *src, long count)
 	"	.previous			\n"
 	: "+r"(res), "+r"(dst), "+r"(src), "=r"(val)
 	: "g"(-EFAULT), "r"(count)
-	: "lp_count", "lp_start", "lp_end", "memory");
+	: "lp_count", "memory");
 
 	return res;
 }
diff --git a/arch/arc/kernel/entry-arcv2.S b/arch/arc/kernel/entry-arcv2.S
index cc558a25b8fa..562089d62d9d 100644
--- a/arch/arc/kernel/entry-arcv2.S
+++ b/arch/arc/kernel/entry-arcv2.S
@@ -209,7 +209,9 @@ restore_regs:
 ;####### Return from Intr #######
 
 debug_marker_l1:
-	bbit1.nt r0, STATUS_DE_BIT, .Lintr_ret_to_delay_slot
+	; bbit1.nt r0, STATUS_DE_BIT, .Lintr_ret_to_delay_slot
+	btst	r0, STATUS_DE_BIT		; Z flag set if bit clear
+	bnz	.Lintr_ret_to_delay_slot	; branch if STATUS_DE_BIT set
 
 .Lisr_ret_fast_path:
 	; Handle special case #1: (Entry via Exception, Return via IRQ)
diff --git a/arch/arc/kernel/head.S b/arch/arc/kernel/head.S
index 8b90d25a15cc..30e090625916 100644
--- a/arch/arc/kernel/head.S
+++ b/arch/arc/kernel/head.S
@@ -17,6 +17,7 @@
 #include <asm/entry.h>
 #include <asm/arcregs.h>
 #include <asm/cache.h>
+#include <asm/irqflags.h>
 
 .macro CPU_EARLY_SETUP
 
@@ -47,6 +48,15 @@
 	sr	r5, [ARC_REG_DC_CTRL]
 
 1:
+
+#ifdef CONFIG_ISA_ARCV2
+	; Unaligned access is disabled at reset, so re-enable early as
+	; gcc 7.3.1 (ARC GNU 2018.03) onwards generates unaligned access
+	; by default
+	lr	r5, [status32]
+	bset	r5, r5, STATUS_AD_BIT
+	kflag	r5
+#endif
 .endm
 
 	.section .init.text, "ax",@progbits
@@ -90,15 +100,13 @@ ENTRY(stext)
 	st.ab   0, [r5, 4]
 1:
 
-#ifdef CONFIG_ARC_UBOOT_SUPPORT
 	; Uboot - kernel ABI
 	;    r0 = [0] No uboot interaction, [1] cmdline in r2, [2] DTB in r2
-	;    r1 = magic number (board identity, unused as of now
+	;    r1 = magic number (always zero as of now)
 	;    r2 = pointer to uboot provided cmdline or external DTB in mem
-	; These are handled later in setup_arch()
+	; These are handled later in handle_uboot_args()
 	st	r0, [@uboot_tag]
 	st	r2, [@uboot_arg]
-#endif
 
 	; setup "current" tsk and optionally cache it in dedicated r25
 	mov	r9, @init_task
diff --git a/arch/arc/kernel/intc-arcv2.c b/arch/arc/kernel/intc-arcv2.c
index 067ea362fb3e..cf18b3e5a934 100644
--- a/arch/arc/kernel/intc-arcv2.c
+++ b/arch/arc/kernel/intc-arcv2.c
@@ -49,11 +49,13 @@ void arc_init_IRQ(void)
 
 	*(unsigned int *)&ictrl = 0;
 
+#ifndef CONFIG_ARC_IRQ_NO_AUTOSAVE
 	ictrl.save_nr_gpr_pairs = 6;	/* r0 to r11 (r12 saved manually) */
 	ictrl.save_blink = 1;
 	ictrl.save_lp_regs = 1;		/* LP_COUNT, LP_START, LP_END */
 	ictrl.save_u_to_u = 0;		/* user ctxt saved on kernel stack */
 	ictrl.save_idx_regs = 1;	/* JLI, LDI, EI */
+#endif
 
 	WRITE_AUX(AUX_IRQ_CTRL, ictrl);
 
diff --git a/arch/arc/kernel/setup.c b/arch/arc/kernel/setup.c
index feb90093e6b1..7b2340996cf8 100644
--- a/arch/arc/kernel/setup.c
+++ b/arch/arc/kernel/setup.c
@@ -199,20 +199,36 @@ static void read_arc_build_cfg_regs(void)
 		cpu->bpu.ret_stk = 4 << bpu.rse;
 
 		if (cpu->core.family >= 0x54) {
-			unsigned int exec_ctrl;
 
-			READ_BCR(AUX_EXEC_CTRL, exec_ctrl);
-			cpu->extn.dual_enb = !(exec_ctrl & 1);
+			struct bcr_uarch_build_arcv2 uarch;
 
-			/* dual issue always present for this core */
-			cpu->extn.dual = 1;
+			/*
+			 * The first 0x54 core (uarch maj:min 0:1 or 0:2) was
+			 * dual issue only (HS4x). But next uarch rev (1:0)
+			 * allows it be configured for single issue (HS3x)
+			 * Ensure we fiddle with dual issue only on HS4x
+			 */
+			READ_BCR(ARC_REG_MICRO_ARCH_BCR, uarch);
+
+			if (uarch.prod == 4) {
+				unsigned int exec_ctrl;
+
+				/* dual issue hardware always present */
+				cpu->extn.dual = 1;
+
+				READ_BCR(AUX_EXEC_CTRL, exec_ctrl);
+
+				/* dual issue hardware enabled ? */
+				cpu->extn.dual_enb = !(exec_ctrl & 1);
+
+			}
 		}
 	}
 
 	READ_BCR(ARC_REG_AP_BCR, ap);
 	if (ap.ver) {
 		cpu->extn.ap_num = 2 << ap.num;
-		cpu->extn.ap_full = !!ap.min;
+		cpu->extn.ap_full = !ap.min;
 	}
 
 	READ_BCR(ARC_REG_SMART_BCR, bcr);
@@ -462,43 +478,78 @@ void setup_processor(void)
 	arc_chk_core_config();
 }
 
-static inline int is_kernel(unsigned long addr)
+static inline bool uboot_arg_invalid(unsigned long addr)
 {
-	if (addr >= (unsigned long)_stext && addr <= (unsigned long)_end)
-		return 1;
-	return 0;
+	/*
+	 * Check that it is a untranslated address (although MMU is not enabled
+	 * yet, it being a high address ensures this is not by fluke)
+	 */
+	if (addr < PAGE_OFFSET)
+		return true;
+
+	/* Check that address doesn't clobber resident kernel image */
+	return addr >= (unsigned long)_stext && addr <= (unsigned long)_end;
+}
+
+#define IGNORE_ARGS		"Ignore U-boot args: "
+
+/* uboot_tag values for U-boot - kernel ABI revision 0; see head.S */
+#define UBOOT_TAG_NONE		0
+#define UBOOT_TAG_CMDLINE	1
+#define UBOOT_TAG_DTB		2
+
+void __init handle_uboot_args(void)
+{
+	bool use_embedded_dtb = true;
+	bool append_cmdline = false;
+
+	/* check that we know this tag */
+	if (uboot_tag != UBOOT_TAG_NONE &&
+	    uboot_tag != UBOOT_TAG_CMDLINE &&
+	    uboot_tag != UBOOT_TAG_DTB) {
+		pr_warn(IGNORE_ARGS "invalid uboot tag: '%08x'\n", uboot_tag);
+		goto ignore_uboot_args;
+	}
+
+	if (uboot_tag != UBOOT_TAG_NONE &&
+            uboot_arg_invalid((unsigned long)uboot_arg)) {
+		pr_warn(IGNORE_ARGS "invalid uboot arg: '%px'\n", uboot_arg);
+		goto ignore_uboot_args;
+	}
+
+	/* see if U-boot passed an external Device Tree blob */
+	if (uboot_tag == UBOOT_TAG_DTB) {
+		machine_desc = setup_machine_fdt((void *)uboot_arg);
+
+		/* external Device Tree blob is invalid - use embedded one */
+		use_embedded_dtb = !machine_desc;
+	}
+
+	if (uboot_tag == UBOOT_TAG_CMDLINE)
+		append_cmdline = true;
+
+ignore_uboot_args:
+
+	if (use_embedded_dtb) {
+		machine_desc = setup_machine_fdt(__dtb_start);
+		if (!machine_desc)
+			panic("Embedded DT invalid\n");
+	}
+
+	/*
+	 * NOTE: @boot_command_line is populated by setup_machine_fdt() so this
+	 * append processing can only happen after.
+	 */
+	if (append_cmdline) {
+		/* Ensure a whitespace between the 2 cmdlines */
+		strlcat(boot_command_line, " ", COMMAND_LINE_SIZE);
+		strlcat(boot_command_line, uboot_arg, COMMAND_LINE_SIZE);
+	}
 }
 
 void __init setup_arch(char **cmdline_p)
 {
-#ifdef CONFIG_ARC_UBOOT_SUPPORT
-	/* make sure that uboot passed pointer to cmdline/dtb is valid */
-	if (uboot_tag && is_kernel((unsigned long)uboot_arg))
-		panic("Invalid uboot arg\n");
-
-	/* See if u-boot passed an external Device Tree blob */
-	machine_desc = setup_machine_fdt(uboot_arg);	/* uboot_tag == 2 */
-	if (!machine_desc)
-#endif
-	{
-		/* No, so try the embedded one */
-		machine_desc = setup_machine_fdt(__dtb_start);
-		if (!machine_desc)
-			panic("Embedded DT invalid\n");
-
-		/*
-		 * If we are here, it is established that @uboot_arg didn't
-		 * point to DT blob. Instead if u-boot says it is cmdline,
-		 * append to embedded DT cmdline.
-		 * setup_machine_fdt() would have populated @boot_command_line
-		 */
-		if (uboot_tag == 1) {
-			/* Ensure a whitespace between the 2 cmdlines */
-			strlcat(boot_command_line, " ", COMMAND_LINE_SIZE);
-			strlcat(boot_command_line, uboot_arg,
-				COMMAND_LINE_SIZE);
-		}
-	}
+	handle_uboot_args();
 
 	/* Save unparsed command line copy for /proc/cmdline */
 	*cmdline_p = boot_command_line;
diff --git a/arch/arc/lib/memcpy-archs.S b/arch/arc/lib/memcpy-archs.S
index d61044dd8b58..ea14b0bf3116 100644
--- a/arch/arc/lib/memcpy-archs.S
+++ b/arch/arc/lib/memcpy-archs.S
@@ -25,15 +25,11 @@
 #endif
 
 #ifdef CONFIG_ARC_HAS_LL64
-# define PREFETCH_READ(RX)	prefetch    [RX, 56]
-# define PREFETCH_WRITE(RX)	prefetchw   [RX, 64]
 # define LOADX(DST,RX)		ldd.ab	DST, [RX, 8]
 # define STOREX(SRC,RX)		std.ab	SRC, [RX, 8]
 # define ZOLSHFT		5
 # define ZOLAND			0x1F
 #else
-# define PREFETCH_READ(RX)	prefetch    [RX, 28]
-# define PREFETCH_WRITE(RX)	prefetchw   [RX, 32]
 # define LOADX(DST,RX)		ld.ab	DST, [RX, 4]
 # define STOREX(SRC,RX)		st.ab	SRC, [RX, 4]
 # define ZOLSHFT		4
@@ -41,8 +37,6 @@
 #endif
 
 ENTRY_CFI(memcpy)
-	prefetch [r1]		; Prefetch the read location
-	prefetchw [r0]		; Prefetch the write location
 	mov.f	0, r2
 ;;; if size is zero
 	jz.d	[blink]
@@ -72,8 +66,6 @@ ENTRY_CFI(memcpy)
 	lpnz	@.Lcopy32_64bytes
 	;; LOOP START
 	LOADX (r6, r1)
-	PREFETCH_READ (r1)
-	PREFETCH_WRITE (r3)
 	LOADX (r8, r1)
 	LOADX (r10, r1)
 	LOADX (r4, r1)
@@ -117,9 +109,7 @@ ENTRY_CFI(memcpy)
 	lpnz	@.Lcopy8bytes_1
 	;; LOOP START
 	ld.ab	r6, [r1, 4]
-	prefetch [r1, 28]	;Prefetch the next read location
 	ld.ab	r8, [r1,4]
-	prefetchw [r3, 32]	;Prefetch the next write location
 
 	SHIFT_1	(r7, r6, 24)
 	or	r7, r7, r5
@@ -162,9 +152,7 @@ ENTRY_CFI(memcpy)
 	lpnz	@.Lcopy8bytes_2
 	;; LOOP START
 	ld.ab	r6, [r1, 4]
-	prefetch [r1, 28]	;Prefetch the next read location
 	ld.ab	r8, [r1,4]
-	prefetchw [r3, 32]	;Prefetch the next write location
 
 	SHIFT_1	(r7, r6, 16)
 	or	r7, r7, r5
@@ -204,9 +192,7 @@ ENTRY_CFI(memcpy)
 	lpnz	@.Lcopy8bytes_3
 	;; LOOP START
 	ld.ab	r6, [r1, 4]
-	prefetch [r1, 28]	;Prefetch the next read location
 	ld.ab	r8, [r1,4]
-	prefetchw [r3, 32]	;Prefetch the next write location
 
 	SHIFT_1	(r7, r6, 8)
 	or	r7, r7, r5
diff --git a/arch/arc/plat-hsdk/Kconfig b/arch/arc/plat-hsdk/Kconfig
index f25c085b9874..23e00216e5a5 100644
--- a/arch/arc/plat-hsdk/Kconfig
+++ b/arch/arc/plat-hsdk/Kconfig
@@ -9,6 +9,7 @@ menuconfig ARC_SOC_HSDK
 	bool "ARC HS Development Kit SOC"
 	depends on ISA_ARCV2
 	select ARC_HAS_ACCL_REGS
+	select ARC_IRQ_NO_AUTOSAVE
 	select CLK_HSDK
 	select RESET_HSDK
 	select HAVE_PCI