Merge tag 'sh-for-4.8' of git://git.libc.org/linux-sh

Pull arch/sh updates from Rich Felker:
 "These changes improve device tree support (including builtin DTB), add
  support for the J-Core J2 processor, an open source synthesizable
  reimplementation of the SH-2 ISA, resolve a longstanding sigcontext
  ABI mismatch issue, and fix various bugs including nommu-specific
  issues and minor regressions introduced in 4.6.

  The J-Core arch support is included here but to be usable it needs
  drivers that are waiting on approval/inclusion from their subsystem
  maintainers"

* tag 'sh-for-4.8' of git://git.libc.org/linux-sh: (23 commits)
  sh: add device tree source for J2 FPGA on Mimas v2 board
  sh: add defconfig for J-Core J2
  sh: use common clock framework with device tree boards
  sh: system call wire up
  sh: Delete unnecessary checks before the function call "mempool_destroy"
  sh: do not perform IPI-based cache flush except on boards that need it
  sh: add SMP support for J2
  sh: SMP support for SH2 entry.S
  sh: add working futex atomic ops on userspace addresses for smp
  sh: add J2 atomics using the cas.l instruction
  sh: add AT_HWCAP flag for J-Core cas.l instruction
  sh: add support for J-Core J2 processor
  sh: fix build regression with CONFIG_OF && !CONFIG_OF_FLATTREE
  sh: allow clocksource drivers to register sched_clock backends
  sh: make heartbeat driver explicitly non-modular
  sh: make board-secureedge5410 explicitly non-modular
  sh: make mm/asids-debugfs explicitly non-modular
  sh: make time.c explicitly non-modular
  sh: fix futex/robust_list on nommu models
  sh: disable aliased page logic on NOMMU models
  ...

19 files changed, 648 insertions, 347 deletions

diff --git a/arch/sh/include/asm/atomic.h b/arch/sh/include/asm/atomic.h
index c399e1c55685..8a7bd80c8b33 100644
--- a/arch/sh/include/asm/atomic.h
+++ b/arch/sh/include/asm/atomic.h
@@ -1,6 +1,12 @@
 #ifndef __ASM_SH_ATOMIC_H
 #define __ASM_SH_ATOMIC_H
 
+#if defined(CONFIG_CPU_J2)
+
+#include <asm-generic/atomic.h>
+
+#else
+
 /*
  * Atomic operations that C can't guarantee us.  Useful for
  * resource counting etc..
@@ -63,4 +69,6 @@ static inline int __atomic_add_unless(atomic_t *v, int a, int u)
 	return c;
 }
 
+#endif /* CONFIG_CPU_J2 */
+
 #endif /* __ASM_SH_ATOMIC_H */
diff --git a/arch/sh/include/asm/barrier.h b/arch/sh/include/asm/barrier.h
index 8a84e05adb2e..3c30b6e166b6 100644
--- a/arch/sh/include/asm/barrier.h
+++ b/arch/sh/include/asm/barrier.h
@@ -29,6 +29,11 @@
 #define wmb()		mb()
 #define ctrl_barrier()	__icbi(PAGE_OFFSET)
 #else
+#if defined(CONFIG_CPU_J2) && defined(CONFIG_SMP)
+#define __smp_mb()	do { int tmp = 0; __asm__ __volatile__ ("cas.l %0,%0,@%1" : "+r"(tmp) : "z"(&tmp) : "memory", "t"); } while(0)
+#define __smp_rmb()	__smp_mb()
+#define __smp_wmb()	__smp_mb()
+#endif
 #define ctrl_barrier()	__asm__ __volatile__ ("nop;nop;nop;nop;nop;nop;nop;nop")
 #endif
 
diff --git a/arch/sh/include/asm/bitops-cas.h b/arch/sh/include/asm/bitops-cas.h
new file mode 100644
index 000000000000..88f793c04d3c
--- /dev/null
+++ b/arch/sh/include/asm/bitops-cas.h
@@ -0,0 +1,93 @@
+#ifndef __ASM_SH_BITOPS_CAS_H
+#define __ASM_SH_BITOPS_CAS_H
+
+static inline unsigned __bo_cas(volatile unsigned *p, unsigned old, unsigned new)
+{
+	__asm__ __volatile__("cas.l %1,%0,@r0"
+		: "+r"(new)
+		: "r"(old), "z"(p)
+		: "t", "memory" );
+	return new;
+}
+
+static inline void set_bit(int nr, volatile void *addr)
+{
+	unsigned mask, old;
+	volatile unsigned *a = addr;
+
+	a += nr >> 5;
+	mask = 1U << (nr & 0x1f);
+
+	do old = *a;
+	while (__bo_cas(a, old, old|mask) != old);
+}
+
+static inline void clear_bit(int nr, volatile void *addr)
+{
+	unsigned mask, old;
+	volatile unsigned *a = addr;
+
+	a += nr >> 5;
+	mask = 1U << (nr & 0x1f);
+
+	do old = *a;
+	while (__bo_cas(a, old, old&~mask) != old);
+}
+
+static inline void change_bit(int nr, volatile void *addr)
+{
+	unsigned mask, old;
+	volatile unsigned *a = addr;
+
+	a += nr >> 5;
+	mask = 1U << (nr & 0x1f);
+
+	do old = *a;
+	while (__bo_cas(a, old, old^mask) != old);
+}
+
+static inline int test_and_set_bit(int nr, volatile void *addr)
+{
+	unsigned mask, old;
+	volatile unsigned *a = addr;
+
+	a += nr >> 5;
+	mask = 1U << (nr & 0x1f);
+
+	do old = *a;
+	while (__bo_cas(a, old, old|mask) != old);
+
+	return !!(old & mask);
+}
+
+static inline int test_and_clear_bit(int nr, volatile void *addr)
+{
+	unsigned mask, old;
+	volatile unsigned *a = addr;
+
+	a += nr >> 5;
+	mask = 1U << (nr & 0x1f);
+
+	do old = *a;
+	while (__bo_cas(a, old, old&~mask) != old);
+
+	return !!(old & mask);
+}
+
+static inline int test_and_change_bit(int nr, volatile void *addr)
+{
+	unsigned mask, old;
+	volatile unsigned *a = addr;
+
+	a += nr >> 5;
+	mask = 1U << (nr & 0x1f);
+
+	do old = *a;
+	while (__bo_cas(a, old, old^mask) != old);
+
+	return !!(old & mask);
+}
+
+#include <asm-generic/bitops/non-atomic.h>
+
+#endif /* __ASM_SH_BITOPS_CAS_H */
diff --git a/arch/sh/include/asm/bitops.h b/arch/sh/include/asm/bitops.h
index fc8e652cf173..a8699d60a8c4 100644
--- a/arch/sh/include/asm/bitops.h
+++ b/arch/sh/include/asm/bitops.h
@@ -18,6 +18,8 @@
 #include <asm/bitops-op32.h>
 #elif defined(CONFIG_CPU_SH4A)
 #include <asm/bitops-llsc.h>
+#elif defined(CONFIG_CPU_J2) && defined(CONFIG_SMP)
+#include <asm/bitops-cas.h>
 #else
 #include <asm-generic/bitops/atomic.h>
 #include <asm-generic/bitops/non-atomic.h>
diff --git a/arch/sh/include/asm/cmpxchg-cas.h b/arch/sh/include/asm/cmpxchg-cas.h
new file mode 100644
index 000000000000..d0d86649e8c1
--- /dev/null
+++ b/arch/sh/include/asm/cmpxchg-cas.h
@@ -0,0 +1,24 @@
+#ifndef __ASM_SH_CMPXCHG_CAS_H
+#define __ASM_SH_CMPXCHG_CAS_H
+
+static inline unsigned long
+__cmpxchg_u32(volatile u32 *m, unsigned long old, unsigned long new)
+{
+	__asm__ __volatile__("cas.l %1,%0,@r0"
+		: "+r"(new)
+		: "r"(old), "z"(m)
+		: "t", "memory" );
+	return new;
+}
+
+static inline unsigned long xchg_u32(volatile u32 *m, unsigned long val)
+{
+	unsigned long old;
+	do old = *m;
+	while (__cmpxchg_u32(m, old, val) != old);
+	return old;
+}
+
+#include <asm/cmpxchg-xchg.h>
+
+#endif /* __ASM_SH_CMPXCHG_CAS_H */
diff --git a/arch/sh/include/asm/cmpxchg-xchg.h b/arch/sh/include/asm/cmpxchg-xchg.h
index 7219719c23a3..1e881f5db659 100644
--- a/arch/sh/include/asm/cmpxchg-xchg.h
+++ b/arch/sh/include/asm/cmpxchg-xchg.h
@@ -21,7 +21,7 @@ static inline u32 __xchg_cmpxchg(volatile void *ptr, u32 x, int size)
 	int off = (unsigned long)ptr % sizeof(u32);
 	volatile u32 *p = ptr - off;
 #ifdef __BIG_ENDIAN
-	int bitoff = (sizeof(u32) - 1 - off) * BITS_PER_BYTE;
+	int bitoff = (sizeof(u32) - size - off) * BITS_PER_BYTE;
 #else
 	int bitoff = off * BITS_PER_BYTE;
 #endif
diff --git a/arch/sh/include/asm/cmpxchg.h b/arch/sh/include/asm/cmpxchg.h
index 5225916c1057..3dfe0467a773 100644
--- a/arch/sh/include/asm/cmpxchg.h
+++ b/arch/sh/include/asm/cmpxchg.h
@@ -13,6 +13,8 @@
 #include <asm/cmpxchg-grb.h>
 #elif defined(CONFIG_CPU_SH4A)
 #include <asm/cmpxchg-llsc.h>
+#elif defined(CONFIG_CPU_J2) && defined(CONFIG_SMP)
+#include <asm/cmpxchg-cas.h>
 #else
 #include <asm/cmpxchg-irq.h>
 #endif
diff --git a/arch/sh/include/asm/futex-cas.h b/arch/sh/include/asm/futex-cas.h
new file mode 100644
index 000000000000..267cb7a5f101
--- /dev/null
+++ b/arch/sh/include/asm/futex-cas.h
@@ -0,0 +1,34 @@
+#ifndef __ASM_SH_FUTEX_CAS_H
+#define __ASM_SH_FUTEX_CAS_H
+
+static inline int atomic_futex_op_cmpxchg_inatomic(u32 *uval,
+						   u32 __user *uaddr,
+						   u32 oldval, u32 newval)
+{
+	int err = 0;
+	__asm__ __volatile__(
+		"1:\n\t"
+		"cas.l	%2, %1, @r0\n"
+		"2:\n\t"
+#ifdef CONFIG_MMU
+		".section	.fixup,\"ax\"\n"
+		"3:\n\t"
+		"mov.l	4f, %0\n\t"
+		"jmp	@%0\n\t"
+		" mov	%3, %0\n\t"
+		".balign	4\n"
+		"4:	.long	2b\n\t"
+		".previous\n"
+		".section	__ex_table,\"a\"\n\t"
+		".long	1b, 3b\n\t"
+		".previous"
+#endif
+		:"+r" (err), "+r" (newval)
+		:"r" (oldval), "i" (-EFAULT), "z" (uaddr)
+		:"t", "memory");
+	if (err) return err;
+	*uval = newval;
+	return 0;
+}
+
+#endif /* __ASM_SH_FUTEX_CAS_H */
diff --git a/arch/sh/include/asm/futex-irq.h b/arch/sh/include/asm/futex-irq.h
index 63d33129ea23..ab01dbee0a82 100644
--- a/arch/sh/include/asm/futex-irq.h
+++ b/arch/sh/include/asm/futex-irq.h
@@ -1,92 +1,6 @@
 #ifndef __ASM_SH_FUTEX_IRQ_H
 #define __ASM_SH_FUTEX_IRQ_H
 
-
-static inline int atomic_futex_op_xchg_set(int oparg, u32 __user *uaddr,
-					   int *oldval)
-{
-	unsigned long flags;
-	int ret;
-
-	local_irq_save(flags);
-
-	ret = get_user(*oldval, uaddr);
-	if (!ret)
-		ret = put_user(oparg, uaddr);
-
-	local_irq_restore(flags);
-
-	return ret;
-}
-
-static inline int atomic_futex_op_xchg_add(int oparg, u32 __user *uaddr,
-					   int *oldval)
-{
-	unsigned long flags;
-	int ret;
-
-	local_irq_save(flags);
-
-	ret = get_user(*oldval, uaddr);
-	if (!ret)
-		ret = put_user(*oldval + oparg, uaddr);
-
-	local_irq_restore(flags);
-
-	return ret;
-}
-
-static inline int atomic_futex_op_xchg_or(int oparg, u32 __user *uaddr,
-					  int *oldval)
-{
-	unsigned long flags;
-	int ret;
-
-	local_irq_save(flags);
-
-	ret = get_user(*oldval, uaddr);
-	if (!ret)
-		ret = put_user(*oldval | oparg, uaddr);
-
-	local_irq_restore(flags);
-
-	return ret;
-}
-
-static inline int atomic_futex_op_xchg_and(int oparg, u32 __user *uaddr,
-					   int *oldval)
-{
-	unsigned long flags;
-	int ret;
-
-	local_irq_save(flags);
-
-	ret = get_user(*oldval, uaddr);
-	if (!ret)
-		ret = put_user(*oldval & oparg, uaddr);
-
-	local_irq_restore(flags);
-
-	return ret;
-}
-
-static inline int atomic_futex_op_xchg_xor(int oparg, u32 __user *uaddr,
-					   int *oldval)
-{
-	unsigned long flags;
-	int ret;
-
-	local_irq_save(flags);
-
-	ret = get_user(*oldval, uaddr);
-	if (!ret)
-		ret = put_user(*oldval ^ oparg, uaddr);
-
-	local_irq_restore(flags);
-
-	return ret;
-}
-
 static inline int atomic_futex_op_cmpxchg_inatomic(u32 *uval,
 						   u32 __user *uaddr,
 						   u32 oldval, u32 newval)
diff --git a/arch/sh/include/asm/futex-llsc.h b/arch/sh/include/asm/futex-llsc.h
new file mode 100644
index 000000000000..23591703bec0
--- /dev/null
+++ b/arch/sh/include/asm/futex-llsc.h
@@ -0,0 +1,41 @@
+#ifndef __ASM_SH_FUTEX_LLSC_H
+#define __ASM_SH_FUTEX_LLSC_H
+
+static inline int atomic_futex_op_cmpxchg_inatomic(u32 *uval,
+						   u32 __user *uaddr,
+						   u32 oldval, u32 newval)
+{
+	int err = 0;
+	__asm__ __volatile__(
+		"synco\n"
+		"1:\n\t"
+		"movli.l	@%2, r0\n\t"
+		"mov	r0, %1\n\t"
+		"cmp/eq	%1, %4\n\t"
+		"bf	2f\n\t"
+		"mov	%5, r0\n\t"
+		"movco.l	r0, @%2\n\t"
+		"bf	1b\n"
+		"2:\n\t"
+		"synco\n\t"
+#ifdef CONFIG_MMU
+		".section	.fixup,\"ax\"\n"
+		"3:\n\t"
+		"mov.l	4f, %0\n\t"
+		"jmp	@%0\n\t"
+		" mov	%3, %0\n\t"
+		".balign	4\n"
+		"4:	.long	2b\n\t"
+		".previous\n"
+		".section	__ex_table,\"a\"\n\t"
+		".long	1b, 3b\n\t"
+		".previous"
+#endif
+		:"+r" (err), "=&r" (*uval)
+		:"r" (uaddr), "i" (-EFAULT), "r" (oldval), "r" (newval)
+		:"t", "memory", "r0");
+	if (err) return err;
+	return 0;
+}
+
+#endif /* __ASM_SH_FUTEX_LLSC_H */
diff --git a/arch/sh/include/asm/futex.h b/arch/sh/include/asm/futex.h
index 7be39a646fbd..d0078747d308 100644
--- a/arch/sh/include/asm/futex.h
+++ b/arch/sh/include/asm/futex.h
@@ -7,16 +7,34 @@
 #include <linux/uaccess.h>
 #include <asm/errno.h>
 
-/* XXX: UP variants, fix for SH-4A and SMP.. */
+#if !defined(CONFIG_SMP)
 #include <asm/futex-irq.h>
+#elif defined(CONFIG_CPU_J2)
+#include <asm/futex-cas.h>
+#elif defined(CONFIG_CPU_SH4A)
+#include <asm/futex-llsc.h>
+#else
+#error SMP not supported on this configuration.
+#endif
+
+static inline int
+futex_atomic_cmpxchg_inatomic(u32 *uval, u32 __user *uaddr,
+			      u32 oldval, u32 newval)
+{
+	if (!access_ok(VERIFY_WRITE, uaddr, sizeof(u32)))
+		return -EFAULT;
+
+	return atomic_futex_op_cmpxchg_inatomic(uval, uaddr, oldval, newval);
+}
 
 static inline int futex_atomic_op_inuser(int encoded_op, u32 __user *uaddr)
 {
 	int op = (encoded_op >> 28) & 7;
 	int cmp = (encoded_op >> 24) & 15;
-	int oparg = (encoded_op << 8) >> 20;
-	int cmparg = (encoded_op << 20) >> 20;
-	int oldval = 0, ret;
+	u32 oparg = (encoded_op << 8) >> 20;
+	u32 cmparg = (encoded_op << 20) >> 20;
+	u32 oldval, newval, prev;
+	int ret;
 
 	if (encoded_op & (FUTEX_OP_OPARG_SHIFT << 28))
 		oparg = 1 << oparg;
@@ -26,26 +44,39 @@ static inline int futex_atomic_op_inuser(int encoded_op, u32 __user *uaddr)
 
 	pagefault_disable();
 
-	switch (op) {
-	case FUTEX_OP_SET:
-		ret = atomic_futex_op_xchg_set(oparg, uaddr, &oldval);
-		break;
-	case FUTEX_OP_ADD:
-		ret = atomic_futex_op_xchg_add(oparg, uaddr, &oldval);
-		break;
-	case FUTEX_OP_OR:
-		ret = atomic_futex_op_xchg_or(oparg, uaddr, &oldval);
-		break;
-	case FUTEX_OP_ANDN:
-		ret = atomic_futex_op_xchg_and(~oparg, uaddr, &oldval);
-		break;
-	case FUTEX_OP_XOR:
-		ret = atomic_futex_op_xchg_xor(oparg, uaddr, &oldval);
-		break;
-	default:
-		ret = -ENOSYS;
-		break;
-	}
+	do {
+		if (op == FUTEX_OP_SET)
+			ret = oldval = 0;
+		else
+			ret = get_user(oldval, uaddr);
+
+		if (ret) break;
+
+		switch (op) {
+		case FUTEX_OP_SET:
+			newval = oparg;
+			break;
+		case FUTEX_OP_ADD:
+			newval = oldval + oparg;
+			break;
+		case FUTEX_OP_OR:
+			newval = oldval | oparg;
+			break;
+		case FUTEX_OP_ANDN:
+			newval = oldval & ~oparg;
+			break;
+		case FUTEX_OP_XOR:
+			newval = oldval ^ oparg;
+			break;
+		default:
+			ret = -ENOSYS;
+			break;
+		}
+
+		if (ret) break;
+
+		ret = futex_atomic_cmpxchg_inatomic(&prev, uaddr, oldval, newval);
+	} while (!ret && prev != oldval);
 
 	pagefault_enable();
 
@@ -53,10 +84,10 @@ static inline int futex_atomic_op_inuser(int encoded_op, u32 __user *uaddr)
 		switch (cmp) {
 		case FUTEX_OP_CMP_EQ: ret = (oldval == cmparg); break;
 		case FUTEX_OP_CMP_NE: ret = (oldval != cmparg); break;
-		case FUTEX_OP_CMP_LT: ret = (oldval < cmparg); break;
-		case FUTEX_OP_CMP_GE: ret = (oldval >= cmparg); break;
-		case FUTEX_OP_CMP_LE: ret = (oldval <= cmparg); break;
-		case FUTEX_OP_CMP_GT: ret = (oldval > cmparg); break;
+		case FUTEX_OP_CMP_LT: ret = ((int)oldval < (int)cmparg); break;
+		case FUTEX_OP_CMP_GE: ret = ((int)oldval >= (int)cmparg); break;
+		case FUTEX_OP_CMP_LE: ret = ((int)oldval <= (int)cmparg); break;
+		case FUTEX_OP_CMP_GT: ret = ((int)oldval > (int)cmparg); break;
 		default: ret = -ENOSYS;
 		}
 	}
@@ -64,15 +95,5 @@ static inline int futex_atomic_op_inuser(int encoded_op, u32 __user *uaddr)
 	return ret;
 }
 
-static inline int
-futex_atomic_cmpxchg_inatomic(u32 *uval, u32 __user *uaddr,
-			      u32 oldval, u32 newval)
-{
-	if (!access_ok(VERIFY_WRITE, uaddr, sizeof(u32)))
-		return -EFAULT;
-
-	return atomic_futex_op_cmpxchg_inatomic(uval, uaddr, oldval, newval);
-}
-
 #endif /* __KERNEL__ */
 #endif /* __ASM_SH_FUTEX_H */
diff --git a/arch/sh/include/asm/processor.h b/arch/sh/include/asm/processor.h
index 1506897648aa..f9a09942a32d 100644
--- a/arch/sh/include/asm/processor.h
+++ b/arch/sh/include/asm/processor.h
@@ -15,7 +15,7 @@
  */
 enum cpu_type {
 	/* SH-2 types */
-	CPU_SH7619,
+	CPU_SH7619, CPU_J2,
 
 	/* SH-2A types */
 	CPU_SH7201, CPU_SH7203, CPU_SH7206, CPU_SH7263, CPU_SH7264, CPU_SH7269,
diff --git a/arch/sh/include/asm/spinlock-cas.h b/arch/sh/include/asm/spinlock-cas.h
new file mode 100644
index 000000000000..c46e8cc7b515
--- /dev/null
+++ b/arch/sh/include/asm/spinlock-cas.h
@@ -0,0 +1,117 @@
+/*
+ * include/asm-sh/spinlock-cas.h
+ *
+ * Copyright (C) 2015 SEI
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#ifndef __ASM_SH_SPINLOCK_CAS_H
+#define __ASM_SH_SPINLOCK_CAS_H
+
+#include <asm/barrier.h>
+#include <asm/processor.h>
+
+static inline unsigned __sl_cas(volatile unsigned *p, unsigned old, unsigned new)
+{
+	__asm__ __volatile__("cas.l %1,%0,@r0"
+		: "+r"(new)
+		: "r"(old), "z"(p)
+		: "t", "memory" );
+	return new;
+}
+
+/*
+ * Your basic SMP spinlocks, allowing only a single CPU anywhere
+ */
+
+#define arch_spin_is_locked(x)		((x)->lock <= 0)
+#define arch_spin_lock_flags(lock, flags) arch_spin_lock(lock)
+
+static inline void arch_spin_unlock_wait(arch_spinlock_t *lock)
+{
+	smp_cond_load_acquire(&lock->lock, VAL > 0);
+}
+
+static inline void arch_spin_lock(arch_spinlock_t *lock)
+{
+	while (!__sl_cas(&lock->lock, 1, 0));
+}
+
+static inline void arch_spin_unlock(arch_spinlock_t *lock)
+{
+	__sl_cas(&lock->lock, 0, 1);
+}
+
+static inline int arch_spin_trylock(arch_spinlock_t *lock)
+{
+	return __sl_cas(&lock->lock, 1, 0);
+}
+
+/*
+ * Read-write spinlocks, allowing multiple readers but only one writer.
+ *
+ * NOTE! it is quite common to have readers in interrupts but no interrupt
+ * writers. For those circumstances we can "mix" irq-safe locks - any writer
+ * needs to get a irq-safe write-lock, but readers can get non-irqsafe
+ * read-locks.
+ */
+
+/**
+ * read_can_lock - would read_trylock() succeed?
+ * @lock: the rwlock in question.
+ */
+#define arch_read_can_lock(x)	((x)->lock > 0)
+
+/**
+ * write_can_lock - would write_trylock() succeed?
+ * @lock: the rwlock in question.
+ */
+#define arch_write_can_lock(x)	((x)->lock == RW_LOCK_BIAS)
+
+static inline void arch_read_lock(arch_rwlock_t *rw)
+{
+	unsigned old;
+	do old = rw->lock;
+	while (!old || __sl_cas(&rw->lock, old, old-1) != old);
+}
+
+static inline void arch_read_unlock(arch_rwlock_t *rw)
+{
+	unsigned old;
+	do old = rw->lock;
+	while (__sl_cas(&rw->lock, old, old+1) != old);
+}
+
+static inline void arch_write_lock(arch_rwlock_t *rw)
+{
+	while (__sl_cas(&rw->lock, RW_LOCK_BIAS, 0) != RW_LOCK_BIAS);
+}
+
+static inline void arch_write_unlock(arch_rwlock_t *rw)
+{
+	__sl_cas(&rw->lock, 0, RW_LOCK_BIAS);
+}
+
+static inline int arch_read_trylock(arch_rwlock_t *rw)
+{
+	unsigned old;
+	do old = rw->lock;
+	while (old && __sl_cas(&rw->lock, old, old-1) != old);
+	return !!old;
+}
+
+static inline int arch_write_trylock(arch_rwlock_t *rw)
+{
+	return __sl_cas(&rw->lock, RW_LOCK_BIAS, 0) == RW_LOCK_BIAS;
+}
+
+#define arch_read_lock_flags(lock, flags) arch_read_lock(lock)
+#define arch_write_lock_flags(lock, flags) arch_write_lock(lock)
+
+#define arch_spin_relax(lock)	cpu_relax()
+#define arch_read_relax(lock)	cpu_relax()
+#define arch_write_relax(lock)	cpu_relax()
+
+#endif /* __ASM_SH_SPINLOCK_CAS_H */
diff --git a/arch/sh/include/asm/spinlock-llsc.h b/arch/sh/include/asm/spinlock-llsc.h
new file mode 100644
index 000000000000..cec78143fa83
--- /dev/null
+++ b/arch/sh/include/asm/spinlock-llsc.h
@@ -0,0 +1,224 @@
+/*
+ * include/asm-sh/spinlock-llsc.h
+ *
+ * Copyright (C) 2002, 2003 Paul Mundt
+ * Copyright (C) 2006, 2007 Akio Idehara
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#ifndef __ASM_SH_SPINLOCK_LLSC_H
+#define __ASM_SH_SPINLOCK_LLSC_H
+
+#include <asm/barrier.h>
+#include <asm/processor.h>
+
+/*
+ * Your basic SMP spinlocks, allowing only a single CPU anywhere
+ */
+
+#define arch_spin_is_locked(x)		((x)->lock <= 0)
+#define arch_spin_lock_flags(lock, flags) arch_spin_lock(lock)
+
+static inline void arch_spin_unlock_wait(arch_spinlock_t *lock)
+{
+	smp_cond_load_acquire(&lock->lock, VAL > 0);
+}
+
+/*
+ * Simple spin lock operations.  There are two variants, one clears IRQ's
+ * on the local processor, one does not.
+ *
+ * We make no fairness assumptions.  They have a cost.
+ */
+static inline void arch_spin_lock(arch_spinlock_t *lock)
+{
+	unsigned long tmp;
+	unsigned long oldval;
+
+	__asm__ __volatile__ (
+		"1:						\n\t"
+		"movli.l	@%2, %0	! arch_spin_lock	\n\t"
+		"mov		%0, %1				\n\t"
+		"mov		#0, %0				\n\t"
+		"movco.l	%0, @%2				\n\t"
+		"bf		1b				\n\t"
+		"cmp/pl		%1				\n\t"
+		"bf		1b				\n\t"
+		: "=&z" (tmp), "=&r" (oldval)
+		: "r" (&lock->lock)
+		: "t", "memory"
+	);
+}
+
+static inline void arch_spin_unlock(arch_spinlock_t *lock)
+{
+	unsigned long tmp;
+
+	__asm__ __volatile__ (
+		"mov		#1, %0 ! arch_spin_unlock	\n\t"
+		"mov.l		%0, @%1				\n\t"
+		: "=&z" (tmp)
+		: "r" (&lock->lock)
+		: "t", "memory"
+	);
+}
+
+static inline int arch_spin_trylock(arch_spinlock_t *lock)
+{
+	unsigned long tmp, oldval;
+
+	__asm__ __volatile__ (
+		"1:						\n\t"
+		"movli.l	@%2, %0	! arch_spin_trylock	\n\t"
+		"mov		%0, %1				\n\t"
+		"mov		#0, %0				\n\t"
+		"movco.l	%0, @%2				\n\t"
+		"bf		1b				\n\t"
+		"synco						\n\t"
+		: "=&z" (tmp), "=&r" (oldval)
+		: "r" (&lock->lock)
+		: "t", "memory"
+	);
+
+	return oldval;
+}
+
+/*
+ * Read-write spinlocks, allowing multiple readers but only one writer.
+ *
+ * NOTE! it is quite common to have readers in interrupts but no interrupt
+ * writers. For those circumstances we can "mix" irq-safe locks - any writer
+ * needs to get a irq-safe write-lock, but readers can get non-irqsafe
+ * read-locks.
+ */
+
+/**
+ * read_can_lock - would read_trylock() succeed?
+ * @lock: the rwlock in question.
+ */
+#define arch_read_can_lock(x)	((x)->lock > 0)
+
+/**
+ * write_can_lock - would write_trylock() succeed?
+ * @lock: the rwlock in question.
+ */
+#define arch_write_can_lock(x)	((x)->lock == RW_LOCK_BIAS)
+
+static inline void arch_read_lock(arch_rwlock_t *rw)
+{
+	unsigned long tmp;
+
+	__asm__ __volatile__ (
+		"1:						\n\t"
+		"movli.l	@%1, %0	! arch_read_lock	\n\t"
+		"cmp/pl		%0				\n\t"
+		"bf		1b				\n\t"
+		"add		#-1, %0				\n\t"
+		"movco.l	%0, @%1				\n\t"
+		"bf		1b				\n\t"
+		: "=&z" (tmp)
+		: "r" (&rw->lock)
+		: "t", "memory"
+	);
+}
+
+static inline void arch_read_unlock(arch_rwlock_t *rw)
+{
+	unsigned long tmp;
+
+	__asm__ __volatile__ (
+		"1:						\n\t"
+		"movli.l	@%1, %0	! arch_read_unlock	\n\t"
+		"add		#1, %0				\n\t"
+		"movco.l	%0, @%1				\n\t"
+		"bf		1b				\n\t"
+		: "=&z" (tmp)
+		: "r" (&rw->lock)
+		: "t", "memory"
+	);
+}
+
+static inline void arch_write_lock(arch_rwlock_t *rw)
+{
+	unsigned long tmp;
+
+	__asm__ __volatile__ (
+		"1:						\n\t"
+		"movli.l	@%1, %0	! arch_write_lock	\n\t"
+		"cmp/hs		%2, %0				\n\t"
+		"bf		1b				\n\t"
+		"sub		%2, %0				\n\t"
+		"movco.l	%0, @%1				\n\t"
+		"bf		1b				\n\t"
+		: "=&z" (tmp)
+		: "r" (&rw->lock), "r" (RW_LOCK_BIAS)
+		: "t", "memory"
+	);
+}
+
+static inline void arch_write_unlock(arch_rwlock_t *rw)
+{
+	__asm__ __volatile__ (
+		"mov.l		%1, @%0 ! arch_write_unlock	\n\t"
+		:
+		: "r" (&rw->lock), "r" (RW_LOCK_BIAS)
+		: "t", "memory"
+	);
+}
+
+static inline int arch_read_trylock(arch_rwlock_t *rw)
+{
+	unsigned long tmp, oldval;
+
+	__asm__ __volatile__ (
+		"1:						\n\t"
+		"movli.l	@%2, %0	! arch_read_trylock	\n\t"
+		"mov		%0, %1				\n\t"
+		"cmp/pl		%0				\n\t"
+		"bf		2f				\n\t"
+		"add		#-1, %0				\n\t"
+		"movco.l	%0, @%2				\n\t"
+		"bf		1b				\n\t"
+		"2:						\n\t"
+		"synco						\n\t"
+		: "=&z" (tmp), "=&r" (oldval)
+		: "r" (&rw->lock)
+		: "t", "memory"
+	);
+
+	return (oldval > 0);
+}
+
+static inline int arch_write_trylock(arch_rwlock_t *rw)
+{
+	unsigned long tmp, oldval;
+
+	__asm__ __volatile__ (
+		"1:						\n\t"
+		"movli.l	@%2, %0	! arch_write_trylock	\n\t"
+		"mov		%0, %1				\n\t"
+		"cmp/hs		%3, %0				\n\t"
+		"bf		2f				\n\t"
+		"sub		%3, %0				\n\t"
+		"2:						\n\t"
+		"movco.l	%0, @%2				\n\t"
+		"bf		1b				\n\t"
+		"synco						\n\t"
+		: "=&z" (tmp), "=&r" (oldval)
+		: "r" (&rw->lock), "r" (RW_LOCK_BIAS)
+		: "t", "memory"
+	);
+
+	return (oldval > (RW_LOCK_BIAS - 1));
+}
+
+#define arch_read_lock_flags(lock, flags) arch_read_lock(lock)
+#define arch_write_lock_flags(lock, flags) arch_write_lock(lock)
+
+#define arch_spin_relax(lock)	cpu_relax()
+#define arch_read_relax(lock)	cpu_relax()
+#define arch_write_relax(lock)	cpu_relax()
+
+#endif /* __ASM_SH_SPINLOCK_LLSC_H */
diff --git a/arch/sh/include/asm/spinlock.h b/arch/sh/include/asm/spinlock.h
index 416834b60ad0..c2c61ea6a8e2 100644
--- a/arch/sh/include/asm/spinlock.h
+++ b/arch/sh/include/asm/spinlock.h
@@ -11,222 +11,12 @@
 #ifndef __ASM_SH_SPINLOCK_H
 #define __ASM_SH_SPINLOCK_H
 
-/*
- * The only locking implemented here uses SH-4A opcodes. For others,
- * split this out as per atomic-*.h.
- */
-#ifndef CONFIG_CPU_SH4A
-#error "Need movli.l/movco.l for spinlocks"
+#if defined(CONFIG_CPU_SH4A)
+#include <asm/spinlock-llsc.h>
+#elif defined(CONFIG_CPU_J2)
+#include <asm/spinlock-cas.h>
+#else
+#error "The configured cpu type does not support spinlocks"
 #endif
 
-#include <asm/barrier.h>
-#include <asm/processor.h>
-
-/*
- * Your basic SMP spinlocks, allowing only a single CPU anywhere
- */
-
-#define arch_spin_is_locked(x)		((x)->lock <= 0)
-#define arch_spin_lock_flags(lock, flags) arch_spin_lock(lock)
-
-static inline void arch_spin_unlock_wait(arch_spinlock_t *lock)
-{
-	smp_cond_load_acquire(&lock->lock, VAL > 0);
-}
-
-/*
- * Simple spin lock operations.  There are two variants, one clears IRQ's
- * on the local processor, one does not.
- *
- * We make no fairness assumptions.  They have a cost.
- */
-static inline void arch_spin_lock(arch_spinlock_t *lock)
-{
-	unsigned long tmp;
-	unsigned long oldval;
-
-	__asm__ __volatile__ (
-		"1:						\n\t"
-		"movli.l	@%2, %0	! arch_spin_lock	\n\t"
-		"mov		%0, %1				\n\t"
-		"mov		#0, %0				\n\t"
-		"movco.l	%0, @%2				\n\t"
-		"bf		1b				\n\t"
-		"cmp/pl		%1				\n\t"
-		"bf		1b				\n\t"
-		: "=&z" (tmp), "=&r" (oldval)
-		: "r" (&lock->lock)
-		: "t", "memory"
-	);
-}
-
-static inline void arch_spin_unlock(arch_spinlock_t *lock)
-{
-	unsigned long tmp;
-
-	__asm__ __volatile__ (
-		"mov		#1, %0 ! arch_spin_unlock	\n\t"
-		"mov.l		%0, @%1				\n\t"
-		: "=&z" (tmp)
-		: "r" (&lock->lock)
-		: "t", "memory"
-	);
-}
-
-static inline int arch_spin_trylock(arch_spinlock_t *lock)
-{
-	unsigned long tmp, oldval;
-
-	__asm__ __volatile__ (
-		"1:						\n\t"
-		"movli.l	@%2, %0	! arch_spin_trylock	\n\t"
-		"mov		%0, %1				\n\t"
-		"mov		#0, %0				\n\t"
-		"movco.l	%0, @%2				\n\t"
-		"bf		1b				\n\t"
-		"synco						\n\t"
-		: "=&z" (tmp), "=&r" (oldval)
-		: "r" (&lock->lock)
-		: "t", "memory"
-	);
-
-	return oldval;
-}
-
-/*
- * Read-write spinlocks, allowing multiple readers but only one writer.
- *
- * NOTE! it is quite common to have readers in interrupts but no interrupt
- * writers. For those circumstances we can "mix" irq-safe locks - any writer
- * needs to get a irq-safe write-lock, but readers can get non-irqsafe
- * read-locks.
- */
-
-/**
- * read_can_lock - would read_trylock() succeed?
- * @lock: the rwlock in question.
- */
-#define arch_read_can_lock(x)	((x)->lock > 0)
-
-/**
- * write_can_lock - would write_trylock() succeed?
- * @lock: the rwlock in question.
- */
-#define arch_write_can_lock(x)	((x)->lock == RW_LOCK_BIAS)
-
-static inline void arch_read_lock(arch_rwlock_t *rw)
-{
-	unsigned long tmp;
-
-	__asm__ __volatile__ (
-		"1:						\n\t"
-		"movli.l	@%1, %0	! arch_read_lock	\n\t"
-		"cmp/pl		%0				\n\t"
-		"bf		1b				\n\t"
-		"add		#-1, %0				\n\t"
-		"movco.l	%0, @%1				\n\t"
-		"bf		1b				\n\t"
-		: "=&z" (tmp)
-		: "r" (&rw->lock)
-		: "t", "memory"
-	);
-}
-
-static inline void arch_read_unlock(arch_rwlock_t *rw)
-{
-	unsigned long tmp;
-
-	__asm__ __volatile__ (
-		"1:						\n\t"
-		"movli.l	@%1, %0	! arch_read_unlock	\n\t"
-		"add		#1, %0				\n\t"
-		"movco.l	%0, @%1				\n\t"
-		"bf		1b				\n\t"
-		: "=&z" (tmp)
-		: "r" (&rw->lock)
-		: "t", "memory"
-	);
-}
-
-static inline void arch_write_lock(arch_rwlock_t *rw)
-{
-	unsigned long tmp;
-
-	__asm__ __volatile__ (
-		"1:						\n\t"
-		"movli.l	@%1, %0	! arch_write_lock	\n\t"
-		"cmp/hs		%2, %0				\n\t"
-		"bf		1b				\n\t"
-		"sub		%2, %0				\n\t"
-		"movco.l	%0, @%1				\n\t"
-		"bf		1b				\n\t"
-		: "=&z" (tmp)
-		: "r" (&rw->lock), "r" (RW_LOCK_BIAS)
-		: "t", "memory"
-	);
-}
-
-static inline void arch_write_unlock(arch_rwlock_t *rw)
-{
-	__asm__ __volatile__ (
-		"mov.l		%1, @%0 ! arch_write_unlock	\n\t"
-		:
-		: "r" (&rw->lock), "r" (RW_LOCK_BIAS)
-		: "t", "memory"
-	);
-}
-
-static inline int arch_read_trylock(arch_rwlock_t *rw)
-{
-	unsigned long tmp, oldval;
-
-	__asm__ __volatile__ (
-		"1:						\n\t"
-		"movli.l	@%2, %0	! arch_read_trylock	\n\t"
-		"mov		%0, %1				\n\t"
-		"cmp/pl		%0				\n\t"
-		"bf		2f				\n\t"
-		"add		#-1, %0				\n\t"
-		"movco.l	%0, @%2				\n\t"
-		"bf		1b				\n\t"
-		"2:						\n\t"
-		"synco						\n\t"
-		: "=&z" (tmp), "=&r" (oldval)
-		: "r" (&rw->lock)
-		: "t", "memory"
-	);
-
-	return (oldval > 0);
-}
-
-static inline int arch_write_trylock(arch_rwlock_t *rw)
-{
-	unsigned long tmp, oldval;
-
-	__asm__ __volatile__ (
-		"1:						\n\t"
-		"movli.l	@%2, %0	! arch_write_trylock	\n\t"
-		"mov		%0, %1				\n\t"
-		"cmp/hs		%3, %0				\n\t"
-		"bf		2f				\n\t"
-		"sub		%3, %0				\n\t"
-		"2:						\n\t"
-		"movco.l	%0, @%2				\n\t"
-		"bf		1b				\n\t"
-		"synco						\n\t"
-		: "=&z" (tmp), "=&r" (oldval)
-		: "r" (&rw->lock), "r" (RW_LOCK_BIAS)
-		: "t", "memory"
-	);
-
-	return (oldval > (RW_LOCK_BIAS - 1));
-}
-
-#define arch_read_lock_flags(lock, flags) arch_read_lock(lock)
-#define arch_write_lock_flags(lock, flags) arch_write_lock(lock)
-
-#define arch_spin_relax(lock)	cpu_relax()
-#define arch_read_relax(lock)	cpu_relax()
-#define arch_write_relax(lock)	cpu_relax()
-
 #endif /* __ASM_SH_SPINLOCK_H */
diff --git a/arch/sh/include/uapi/asm/cpu-features.h b/arch/sh/include/uapi/asm/cpu-features.h
index 694abe490edb..2f1bc851042a 100644
--- a/arch/sh/include/uapi/asm/cpu-features.h
+++ b/arch/sh/include/uapi/asm/cpu-features.h
@@ -22,5 +22,6 @@
 #define CPU_HAS_L2_CACHE	0x0080	/* Secondary cache / URAM */
 #define CPU_HAS_OP32		0x0100	/* 32-bit instruction support */
 #define CPU_HAS_PTEAEX		0x0200	/* PTE ASID Extension support */
+#define CPU_HAS_CAS_L		0x0400	/* cas.l atomic compare-and-swap */
 
 #endif /* __ASM_SH_CPU_FEATURES_H */
diff --git a/arch/sh/include/uapi/asm/sigcontext.h b/arch/sh/include/uapi/asm/sigcontext.h
index 8ce1435bc0bf..faa5d0833412 100644
--- a/arch/sh/include/uapi/asm/sigcontext.h
+++ b/arch/sh/include/uapi/asm/sigcontext.h
@@ -25,8 +25,6 @@ struct sigcontext {
 	unsigned long sc_mach;
 	unsigned long sc_macl;
 
-#if defined(__SH4__) || defined(CONFIG_CPU_SH4) || \
-    defined(__SH2A__) || defined(CONFIG_CPU_SH2A)
 	/* FPU registers */
 	unsigned long sc_fpregs[16];
 	unsigned long sc_xfpregs[16];
@@ -34,7 +32,6 @@ struct sigcontext {
 	unsigned int sc_fpul;
 	unsigned int sc_ownedfp;
 #endif
-#endif
 };
 
 #endif /* __ASM_SH_SIGCONTEXT_H */
diff --git a/arch/sh/include/uapi/asm/unistd_32.h b/arch/sh/include/uapi/asm/unistd_32.h
index d13a1d623736..c801bde9e6f5 100644
--- a/arch/sh/include/uapi/asm/unistd_32.h
+++ b/arch/sh/include/uapi/asm/unistd_32.h
@@ -380,7 +380,21 @@
 #define __NR_process_vm_writev	366
 #define __NR_kcmp		367
 #define __NR_finit_module	368
+#define __NR_sched_getattr	369
+#define __NR_sched_setattr	370
+#define __NR_renameat2		371
+#define __NR_seccomp		372
+#define __NR_getrandom		373
+#define __NR_memfd_create	374
+#define __NR_bpf		375
+#define __NR_execveat		376
+#define __NR_userfaultfd	377
+#define __NR_membarrier		378
+#define __NR_mlock2		379
+#define __NR_copy_file_range	380
+#define __NR_preadv2		381
+#define __NR_pwritev2		382
 
-#define NR_syscalls 369
+#define NR_syscalls 383
 
 #endif /* __ASM_SH_UNISTD_32_H */
diff --git a/arch/sh/include/uapi/asm/unistd_64.h b/arch/sh/include/uapi/asm/unistd_64.h
index 47ebd5b5ed55..ce0cb3598b62 100644
--- a/arch/sh/include/uapi/asm/unistd_64.h
+++ b/arch/sh/include/uapi/asm/unistd_64.h
@@ -400,7 +400,21 @@
 #define __NR_process_vm_writev	377
 #define __NR_kcmp		378
 #define __NR_finit_module	379
+#define __NR_sched_getattr	380
+#define __NR_sched_setattr	381
+#define __NR_renameat2		382
+#define __NR_seccomp		383
+#define __NR_getrandom		384
+#define __NR_memfd_create	385
+#define __NR_bpf		386
+#define __NR_execveat		387
+#define __NR_userfaultfd	388
+#define __NR_membarrier		389
+#define __NR_mlock2		390
+#define __NR_copy_file_range	391
+#define __NR_preadv2		392
+#define __NR_pwritev2		393
 
-#define NR_syscalls 380
+#define NR_syscalls 394
 
 #endif /* __ASM_SH_UNISTD_64_H */