Merge tag 'locking_core_for_v5.17_rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull locking updates from Borislav Petkov:
 "Lots of cleanups and preparation. Highlights:

   - futex: Cleanup and remove runtime futex_cmpxchg detection

   - rtmutex: Some fixes for the PREEMPT_RT locking infrastructure

   - kcsan: Share owner_on_cpu() between mutex,rtmutex and rwsem and
     annotate the racy owner->on_cpu access *once*.

   - atomic64: Dead-Code-Elemination"

[ Description above by Peter Zijlstra ]

* tag 'locking_core_for_v5.17_rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  locking/atomic: atomic64: Remove unusable atomic ops
  futex: Fix additional regressions
  locking: Allow to include asm/spinlock_types.h from linux/spinlock_types_raw.h
  x86/mm: Include spinlock_t definition in pgtable.
  locking: Mark racy reads of owner->on_cpu
  locking: Make owner_on_cpu() into <linux/sched.h>
  lockdep/selftests: Adapt ww-tests for PREEMPT_RT
  lockdep/selftests: Skip the softirq related tests on PREEMPT_RT
  lockdep/selftests: Unbalanced migrate_disable() & rcu_read_lock().
  lockdep/selftests: Avoid using local_lock_{acquire|release}().
  lockdep: Remove softirq accounting on PREEMPT_RT.
  locking/rtmutex: Add rt_mutex_lock_nest_lock() and rt_mutex_lock_killable().
  locking/rtmutex: Squash self-deadlock check for ww_rt_mutex.
  locking: Remove rt_rwlock_is_contended().
  sched: Trigger warning if ->migration_disabled counter underflows.
  futex: Fix sparc32/m68k/nds32 build regression
  futex: Remove futex_cmpxchg detection
  futex: Ensure futex_atomic_cmpxchg_inatomic() is present
  kernel/locking: Use a pointer in ww_mutex_trylock().

arch/alpha/include/asm/spinlock_types.h

@@ -2,7 +2,7 @@
 #ifndef _ALPHA_SPINLOCK_TYPES_H
 #define _ALPHA_SPINLOCK_TYPES_H
 
-#ifndef __LINUX_SPINLOCK_TYPES_H
+#ifndef __LINUX_SPINLOCK_TYPES_RAW_H
 # error "please don't include this file directly"
 #endif
 

arch/arc/Kconfig

@@ -32,7 +32,6 @@ config ARC
 	select HAVE_ARCH_TRANSPARENT_HUGEPAGE if ARC_MMU_V4
 	select HAVE_DEBUG_STACKOVERFLOW
 	select HAVE_DEBUG_KMEMLEAK
-	select HAVE_FUTEX_CMPXCHG if FUTEX
 	select HAVE_IOREMAP_PROT
 	select HAVE_KERNEL_GZIP
 	select HAVE_KERNEL_LZMA

arch/arm/Kconfig

@@ -93,7 +93,6 @@ config ARM
 	select HAVE_FTRACE_MCOUNT_RECORD if !XIP_KERNEL
 	select HAVE_FUNCTION_GRAPH_TRACER if !THUMB2_KERNEL && !CC_IS_CLANG
 	select HAVE_FUNCTION_TRACER if !XIP_KERNEL && !(THUMB2_KERNEL && CC_IS_CLANG)
-	select HAVE_FUTEX_CMPXCHG if FUTEX
 	select HAVE_GCC_PLUGINS
 	select HAVE_HW_BREAKPOINT if PERF_EVENTS && (CPU_V6 || CPU_V6K || CPU_V7)
 	select HAVE_IRQ_TIME_ACCOUNTING

arch/arm/include/asm/spinlock_types.h

@@ -2,7 +2,7 @@
 #ifndef __ASM_SPINLOCK_TYPES_H
 #define __ASM_SPINLOCK_TYPES_H
 
-#ifndef __LINUX_SPINLOCK_TYPES_H
+#ifndef __LINUX_SPINLOCK_TYPES_RAW_H
 # error "please don't include this file directly"
 #endif
 

arch/arm64/Kconfig

@@ -196,7 +196,6 @@ config ARM64
 	select HAVE_REGS_AND_STACK_ACCESS_API
 	select HAVE_POSIX_CPU_TIMERS_TASK_WORK
 	select HAVE_FUNCTION_ARG_ACCESS_API
-	select HAVE_FUTEX_CMPXCHG if FUTEX
 	select MMU_GATHER_RCU_TABLE_FREE
 	select HAVE_RSEQ
 	select HAVE_STACKPROTECTOR

arch/arm64/include/asm/spinlock_types.h

@@ -5,7 +5,7 @@
 #ifndef __ASM_SPINLOCK_TYPES_H
 #define __ASM_SPINLOCK_TYPES_H
 
-#if !defined(__LINUX_SPINLOCK_TYPES_H) && !defined(__ASM_SPINLOCK_H)
+#if !defined(__LINUX_SPINLOCK_TYPES_RAW_H) && !defined(__ASM_SPINLOCK_H)
 # error "please don't include this file directly"
 #endif
 

arch/csky/Kconfig

@@ -52,7 +52,6 @@ config CSKY
 	select HAVE_FUNCTION_TRACER
 	select HAVE_FUNCTION_GRAPH_TRACER
 	select HAVE_FUNCTION_ERROR_INJECTION
-	select HAVE_FUTEX_CMPXCHG if FUTEX && SMP
 	select HAVE_FTRACE_MCOUNT_RECORD
 	select HAVE_KERNEL_GZIP
 	select HAVE_KERNEL_LZO

arch/csky/include/asm/spinlock_types.h

@@ -3,7 +3,7 @@
 #ifndef __ASM_CSKY_SPINLOCK_TYPES_H
 #define __ASM_CSKY_SPINLOCK_TYPES_H
 
-#ifndef __LINUX_SPINLOCK_TYPES_H
+#ifndef __LINUX_SPINLOCK_TYPES_RAW_H
 # error "please don't include this file directly"
 #endif
 

arch/hexagon/include/asm/spinlock_types.h

@@ -8,7 +8,7 @@
 #ifndef _ASM_SPINLOCK_TYPES_H
 #define _ASM_SPINLOCK_TYPES_H
 
-#ifndef __LINUX_SPINLOCK_TYPES_H
+#ifndef __LINUX_SPINLOCK_TYPES_RAW_H
 # error "please don't include this file directly"
 #endif
 

arch/ia64/include/asm/spinlock_types.h

@@ -2,7 +2,7 @@
 #ifndef _ASM_IA64_SPINLOCK_TYPES_H
 #define _ASM_IA64_SPINLOCK_TYPES_H
 
-#ifndef __LINUX_SPINLOCK_TYPES_H
+#ifndef __LINUX_SPINLOCK_TYPES_RAW_H
 # error "please don't include this file directly"
 #endif
 

arch/m68k/Kconfig

@@ -21,7 +21,6 @@ config M68K
 	select HAVE_ASM_MODVERSIONS
 	select HAVE_DEBUG_BUGVERBOSE
 	select HAVE_EFFICIENT_UNALIGNED_ACCESS if !CPU_HAS_NO_UNALIGNED
-	select HAVE_FUTEX_CMPXCHG if MMU && FUTEX
 	select HAVE_MOD_ARCH_SPECIFIC
 	select HAVE_UID16
 	select MMU_GATHER_NO_RANGE if MMU

arch/mips/include/asm/futex.h

@@ -19,7 +19,11 @@
 #include <asm/sync.h>
 #include <asm/war.h>
 
-#define __futex_atomic_op(insn, ret, oldval, uaddr, oparg)		\
+#define arch_futex_atomic_op_inuser arch_futex_atomic_op_inuser
+#define futex_atomic_cmpxchg_inatomic futex_atomic_cmpxchg_inatomic
+#include <asm-generic/futex.h>
+
+#define __futex_atomic_op(op, insn, ret, oldval, uaddr, oparg)		\
 {									\
 	if (cpu_has_llsc && IS_ENABLED(CONFIG_WAR_R10000_LLSC)) {	\
 		__asm__ __volatile__(					\
@@ -80,8 +84,10 @@
 		: "0" (0), GCC_OFF_SMALL_ASM() (*uaddr), "Jr" (oparg),	\
 		  "i" (-EFAULT)						\
 		: "memory");						\
-	} else								\
-		ret = -ENOSYS;						\
+	} else {							\
+		/* fallback for non-SMP */				\
+		ret = futex_atomic_op_inuser_local(op, oparg, oval, uaddr);	\
+	}								\
 }
 
 static inline int
@@ -94,23 +100,23 @@ arch_futex_atomic_op_inuser(int op, int oparg, int *oval, u32 __user *uaddr)
 
 	switch (op) {
 	case FUTEX_OP_SET:
-		__futex_atomic_op("move $1, %z5", ret, oldval, uaddr, oparg);
+		__futex_atomic_op(op, "move $1, %z5", ret, oldval, uaddr, oparg);
 		break;
 
 	case FUTEX_OP_ADD:
-		__futex_atomic_op("addu $1, %1, %z5",
+		__futex_atomic_op(op, "addu $1, %1, %z5",
 				  ret, oldval, uaddr, oparg);
 		break;
 	case FUTEX_OP_OR:
-		__futex_atomic_op("or	$1, %1, %z5",
+		__futex_atomic_op(op, "or	$1, %1, %z5",
 				  ret, oldval, uaddr, oparg);
 		break;
 	case FUTEX_OP_ANDN:
-		__futex_atomic_op("and	$1, %1, %z5",
+		__futex_atomic_op(op, "and	$1, %1, %z5",
 				  ret, oldval, uaddr, ~oparg);
 		break;
 	case FUTEX_OP_XOR:
-		__futex_atomic_op("xor	$1, %1, %z5",
+		__futex_atomic_op(op, "xor	$1, %1, %z5",
 				  ret, oldval, uaddr, oparg);
 		break;
 	default:
@@ -193,8 +199,9 @@ futex_atomic_cmpxchg_inatomic(u32 *uval, u32 __user *uaddr,
 		: GCC_OFF_SMALL_ASM() (*uaddr), "Jr" (oldval), "Jr" (newval),
 		  "i" (-EFAULT)
 		: "memory");
-	} else
-		return -ENOSYS;
+	} else {
+		return futex_atomic_cmpxchg_inatomic_local(uval, uaddr, oldval, newval);
+	}
 
 	*uval = val;
 	return ret;

arch/powerpc/include/asm/simple_spinlock_types.h

@@ -2,7 +2,7 @@
 #ifndef _ASM_POWERPC_SIMPLE_SPINLOCK_TYPES_H
 #define _ASM_POWERPC_SIMPLE_SPINLOCK_TYPES_H
 
-#ifndef __LINUX_SPINLOCK_TYPES_H
+#ifndef __LINUX_SPINLOCK_TYPES_RAW_H
 # error "please don't include this file directly"
 #endif
 

arch/powerpc/include/asm/spinlock_types.h

@@ -2,7 +2,7 @@
 #ifndef _ASM_POWERPC_SPINLOCK_TYPES_H
 #define _ASM_POWERPC_SPINLOCK_TYPES_H
 
-#ifndef __LINUX_SPINLOCK_TYPES_H
+#ifndef __LINUX_SPINLOCK_TYPES_RAW_H
 # error "please don't include this file directly"
 #endif
 

arch/riscv/Kconfig

@@ -83,7 +83,6 @@ config RISCV
 	select HAVE_DMA_CONTIGUOUS if MMU
 	select HAVE_EBPF_JIT if MMU
 	select HAVE_FUNCTION_ERROR_INJECTION
-	select HAVE_FUTEX_CMPXCHG if FUTEX
 	select HAVE_GCC_PLUGINS
 	select HAVE_GENERIC_VDSO if MMU && 64BIT
 	select HAVE_IRQ_TIME_ACCOUNTING

arch/riscv/include/asm/spinlock_types.h

@@ -6,7 +6,7 @@
 #ifndef _ASM_RISCV_SPINLOCK_TYPES_H
 #define _ASM_RISCV_SPINLOCK_TYPES_H
 
-#ifndef __LINUX_SPINLOCK_TYPES_H
+#ifndef __LINUX_SPINLOCK_TYPES_RAW_H
 # error "please don't include this file directly"
 #endif
 

arch/s390/Kconfig

@@ -165,7 +165,6 @@ config S390
 	select HAVE_FUNCTION_ERROR_INJECTION
 	select HAVE_FUNCTION_GRAPH_TRACER
 	select HAVE_FUNCTION_TRACER
-	select HAVE_FUTEX_CMPXCHG if FUTEX
 	select HAVE_GCC_PLUGINS
 	select HAVE_GENERIC_VDSO
 	select HAVE_IOREMAP_PROT if PCI

arch/s390/include/asm/spinlock_types.h

@@ -2,7 +2,7 @@
 #ifndef __ASM_SPINLOCK_TYPES_H
 #define __ASM_SPINLOCK_TYPES_H
 
-#ifndef __LINUX_SPINLOCK_TYPES_H
+#ifndef __LINUX_SPINLOCK_TYPES_RAW_H
 # error "please don't include this file directly"
 #endif
 

arch/sh/Kconfig

@@ -34,7 +34,6 @@ config SUPERH
 	select HAVE_FAST_GUP if MMU
 	select HAVE_FUNCTION_GRAPH_TRACER
 	select HAVE_FUNCTION_TRACER
-	select HAVE_FUTEX_CMPXCHG if FUTEX
 	select HAVE_FTRACE_MCOUNT_RECORD
 	select HAVE_HW_BREAKPOINT
 	select HAVE_IOREMAP_PROT if MMU && !X2TLB

arch/sh/include/asm/spinlock_types.h

@@ -2,7 +2,7 @@
 #ifndef __ASM_SH_SPINLOCK_TYPES_H
 #define __ASM_SH_SPINLOCK_TYPES_H
 
-#ifndef __LINUX_SPINLOCK_TYPES_H
+#ifndef __LINUX_SPINLOCK_TYPES_RAW_H
 # error "please don't include this file directly"
 #endif
 

arch/um/Kconfig

@@ -14,7 +14,6 @@ config UML
 	select HAVE_ARCH_SECCOMP_FILTER
 	select HAVE_ASM_MODVERSIONS
 	select HAVE_UID16
-	select HAVE_FUTEX_CMPXCHG if FUTEX
 	select HAVE_DEBUG_KMEMLEAK
 	select HAVE_DEBUG_BUGVERBOSE
 	select NO_DMA if !UML_DMA_EMULATION

arch/um/kernel/skas/uaccess.c

@@ -323,7 +323,6 @@ EXPORT_SYMBOL(arch_futex_atomic_op_inuser);
  * 0 - On success
  * -EFAULT - User access resulted in a page fault
  * -EAGAIN - Atomic operation was unable to complete due to contention
- * -ENOSYS - Function not implemented (only if !HAVE_FUTEX_CMPXCHG)
  */
 
 int futex_atomic_cmpxchg_inatomic(u32 *uval, u32 __user *uaddr,

arch/xtensa/Kconfig

@@ -31,7 +31,6 @@ config XTENSA
 	select HAVE_DMA_CONTIGUOUS
 	select HAVE_EXIT_THREAD
 	select HAVE_FUNCTION_TRACER
-	select HAVE_FUTEX_CMPXCHG if !MMU && FUTEX
 	select HAVE_HW_BREAKPOINT if PERF_EVENTS
 	select HAVE_IRQ_TIME_ACCOUNTING
 	select HAVE_PCI

arch/xtensa/include/asm/futex.h

@@ -16,6 +16,10 @@
 #include <linux/uaccess.h>
 #include <linux/errno.h>
 
+#define arch_futex_atomic_op_inuser arch_futex_atomic_op_inuser
+#define futex_atomic_cmpxchg_inatomic futex_atomic_cmpxchg_inatomic
+#include <asm-generic/futex.h>
+
 #if XCHAL_HAVE_EXCLUSIVE
 #define __futex_atomic_op(insn, ret, old, uaddr, arg)	\
 	__asm__ __volatile(				\
@@ -105,7 +109,7 @@ static inline int arch_futex_atomic_op_inuser(int op, int oparg, int *oval,
 
 	return ret;
 #else
-	return -ENOSYS;
+	return futex_atomic_op_inuser_local(op, oparg, oval, uaddr);
 #endif
 }
 
@@ -156,7 +160,7 @@ futex_atomic_cmpxchg_inatomic(u32 *uval, u32 __user *uaddr,
 
 	return ret;
 #else
-	return -ENOSYS;
+	return futex_atomic_cmpxchg_inatomic_local(uval, uaddr, oldval, newval);
 #endif
 }
 

arch/xtensa/include/asm/spinlock_types.h

@@ -2,7 +2,7 @@
 #ifndef __ASM_SPINLOCK_TYPES_H
 #define __ASM_SPINLOCK_TYPES_H
 
-#if !defined(__LINUX_SPINLOCK_TYPES_H) && !defined(__ASM_SPINLOCK_H)
+#if !defined(__LINUX_SPINLOCK_TYPES_RAW_H) && !defined(__ASM_SPINLOCK_H)
 # error "please don't include this file directly"
 #endif
 

include/asm-generic/futex.h

@@ -6,15 +6,22 @@
 #include <linux/uaccess.h>
 #include <asm/errno.h>
 
+#ifndef futex_atomic_cmpxchg_inatomic
 #ifndef CONFIG_SMP
 /*
  * The following implementation only for uniprocessor machines.
  * It relies on preempt_disable() ensuring mutual exclusion.
  *
  */
+#define futex_atomic_cmpxchg_inatomic(uval, uaddr, oldval, newval) \
+	futex_atomic_cmpxchg_inatomic_local(uval, uaddr, oldval, newval)
+#define arch_futex_atomic_op_inuser(op, oparg, oval, uaddr) \
+	futex_atomic_op_inuser_local(op, oparg, oval, uaddr)
+#endif /* CONFIG_SMP */
+#endif
 
 /**
- * arch_futex_atomic_op_inuser() - Atomic arithmetic operation with constant
+ * futex_atomic_op_inuser_local() - Atomic arithmetic operation with constant
  *			  argument and comparison of the previous
  *			  futex value with another constant.
  *
@@ -28,7 +35,7 @@
  * -ENOSYS - Operation not supported
  */
 static inline int
-arch_futex_atomic_op_inuser(int op, u32 oparg, int *oval, u32 __user *uaddr)
+futex_atomic_op_inuser_local(int op, u32 oparg, int *oval, u32 __user *uaddr)
 {
 	int oldval, ret;
 	u32 tmp;
@@ -75,7 +82,7 @@ arch_futex_atomic_op_inuser(int op, u32 oparg, int *oval, u32 __user *uaddr)
 }
 
 /**
- * futex_atomic_cmpxchg_inatomic() - Compare and exchange the content of the
+ * futex_atomic_cmpxchg_inatomic_local() - Compare and exchange the content of the
  *				uaddr with newval if the current value is
  *				oldval.
  * @uval:	pointer to store content of @uaddr
@@ -87,10 +94,9 @@ arch_futex_atomic_op_inuser(int op, u32 oparg, int *oval, u32 __user *uaddr)
  * 0 - On success
  * -EFAULT - User access resulted in a page fault
  * -EAGAIN - Atomic operation was unable to complete due to contention
- * -ENOSYS - Function not implemented (only if !HAVE_FUTEX_CMPXCHG)
  */
 static inline int
-futex_atomic_cmpxchg_inatomic(u32 *uval, u32 __user *uaddr,
+futex_atomic_cmpxchg_inatomic_local(u32 *uval, u32 __user *uaddr,
 			      u32 oldval, u32 newval)
 {
 	u32 val;
@@ -112,19 +118,4 @@ futex_atomic_cmpxchg_inatomic(u32 *uval, u32 __user *uaddr,
 	return 0;
 }
 
-#else
-static inline int
-arch_futex_atomic_op_inuser(int op, u32 oparg, int *oval, u32 __user *uaddr)
-{
-	return -ENOSYS;
-}
-
-static inline int
-futex_atomic_cmpxchg_inatomic(u32 *uval, u32 __user *uaddr,
-			      u32 oldval, u32 newval)
-{
-	return -ENOSYS;
-}
-
-#endif /* CONFIG_SMP */
 #endif

include/linux/irqflags.h

@@ -71,14 +71,6 @@ do {						\
 do {						\
 	__this_cpu_dec(hardirq_context);	\
 } while (0)
-# define lockdep_softirq_enter()		\
-do {						\
-	current->softirq_context++;		\
-} while (0)
-# define lockdep_softirq_exit()			\
-do {						\
-	current->softirq_context--;		\
-} while (0)
 
 # define lockdep_hrtimer_enter(__hrtimer)		\
 ({							\
@@ -140,6 +132,21 @@ do {						\
 # define lockdep_irq_work_exit(__work)		do { } while (0)
 #endif
 
+#if defined(CONFIG_TRACE_IRQFLAGS) && !defined(CONFIG_PREEMPT_RT)
+# define lockdep_softirq_enter()		\
+do {						\
+	current->softirq_context++;		\
+} while (0)
+# define lockdep_softirq_exit()			\
+do {						\
+	current->softirq_context--;		\
+} while (0)
+
+#else
+# define lockdep_softirq_enter()		do { } while (0)
+# define lockdep_softirq_exit()			do { } while (0)
+#endif
+
 #if defined(CONFIG_IRQSOFF_TRACER) || \
 	defined(CONFIG_PREEMPT_TRACER)
  extern void stop_critical_timings(void);

include/linux/ratelimit_types.h

@@ -4,7 +4,7 @@
 
 #include <linux/bits.h>
 #include <linux/param.h>
-#include <linux/spinlock_types.h>
+#include <linux/spinlock_types_raw.h>
 
 #define DEFAULT_RATELIMIT_INTERVAL	(5 * HZ)
 #define DEFAULT_RATELIMIT_BURST		10

include/linux/rtmutex.h

@@ -99,13 +99,22 @@ extern void __rt_mutex_init(struct rt_mutex *lock, const char *name, struct lock
 
 #ifdef CONFIG_DEBUG_LOCK_ALLOC
 extern void rt_mutex_lock_nested(struct rt_mutex *lock, unsigned int subclass);
+extern void _rt_mutex_lock_nest_lock(struct rt_mutex *lock, struct lockdep_map *nest_lock);
 #define rt_mutex_lock(lock) rt_mutex_lock_nested(lock, 0)
+#define rt_mutex_lock_nest_lock(lock, nest_lock)			\
+	do {								\
+		typecheck(struct lockdep_map *, &(nest_lock)->dep_map);	\
+		_rt_mutex_lock_nest_lock(lock, &(nest_lock)->dep_map);	\
+	} while (0)
+
 #else
 extern void rt_mutex_lock(struct rt_mutex *lock);
 #define rt_mutex_lock_nested(lock, subclass) rt_mutex_lock(lock)
+#define rt_mutex_lock_nest_lock(lock, nest_lock) rt_mutex_lock(lock)
 #endif
 
 extern int rt_mutex_lock_interruptible(struct rt_mutex *lock);
+extern int rt_mutex_lock_killable(struct rt_mutex *lock);
 extern int rt_mutex_trylock(struct rt_mutex *lock);
 
 extern void rt_mutex_unlock(struct rt_mutex *lock);

include/linux/sched.h

@@ -2178,6 +2178,15 @@ extern long sched_getaffinity(pid_t pid, struct cpumask *mask);
 #endif
 
 #ifdef CONFIG_SMP
+static inline bool owner_on_cpu(struct task_struct *owner)
+{
+	/*
+	 * As lock holder preemption issue, we both skip spinning if
+	 * task is not on cpu or its cpu is preempted
+	 */
+	return READ_ONCE(owner->on_cpu) && !vcpu_is_preempted(task_cpu(owner));
+}
+
 /* Returns effective CPU energy utilization, as seen by the scheduler */
 unsigned long sched_cpu_util(int cpu, unsigned long max);
 #endif /* CONFIG_SMP */

include/linux/spinlock_types_up.h

 #ifndef __LINUX_SPINLOCK_TYPES_UP_H
 #define __LINUX_SPINLOCK_TYPES_UP_H
 
-#ifndef __LINUX_SPINLOCK_TYPES_H
+#ifndef __LINUX_SPINLOCK_TYPES_RAW_H
 # error "please don't include this file directly"
 #endif
 

init/Kconfig

@@ -1579,6 +1579,7 @@ config BASE_FULL
 
 config FUTEX
 	bool "Enable futex support" if EXPERT
+	depends on !(SPARC32 && SMP)
 	default y
 	imply RT_MUTEXES
 	help
@@ -1591,14 +1592,6 @@ config FUTEX_PI
 	depends on FUTEX && RT_MUTEXES
 	default y
 
-config HAVE_FUTEX_CMPXCHG
-	bool
-	depends on FUTEX
-	help
-	  Architectures should select this if futex_atomic_cmpxchg_inatomic()
-	  is implemented and always working. This removes a couple of runtime
-	  checks.
-
 config EPOLL
 	bool "Enable eventpoll support" if EXPERT
 	default y

kernel/futex/core.c

@@ -41,11 +41,6 @@
 #include "futex.h"
 #include "../locking/rtmutex_common.h"
 
-#ifndef CONFIG_HAVE_FUTEX_CMPXCHG
-int  __read_mostly futex_cmpxchg_enabled;
-#endif
-
-
 /*
  * The base of the bucket array and its size are always used together
  * (after initialization only in futex_hash()), so ensure that they
@@ -776,9 +771,6 @@ static void exit_robust_list(struct task_struct *curr)
 	unsigned long futex_offset;
 	int rc;
 
-	if (!futex_cmpxchg_enabled)
-		return;
-
 	/*
 	 * Fetch the list head (which was registered earlier, via
 	 * sys_set_robust_list()):
@@ -874,9 +866,6 @@ static void compat_exit_robust_list(struct task_struct *curr)
 	compat_long_t futex_offset;
 	int rc;
 
-	if (!futex_cmpxchg_enabled)
-		return;
-
 	/*
 	 * Fetch the list head (which was registered earlier, via
 	 * sys_set_robust_list()):
@@ -950,8 +939,6 @@ static void exit_pi_state_list(struct task_struct *curr)
 	struct futex_hash_bucket *hb;
 	union futex_key key = FUTEX_KEY_INIT;
 
-	if (!futex_cmpxchg_enabled)
-		return;
 	/*
 	 * We are a ZOMBIE and nobody can enqueue itself on
 	 * pi_state_list anymore, but we have to be careful
@@ -1125,26 +1112,6 @@ void futex_exit_release(struct task_struct *tsk)
 	futex_cleanup_end(tsk, FUTEX_STATE_DEAD);
 }
 
-static void __init futex_detect_cmpxchg(void)
-{
-#ifndef CONFIG_HAVE_FUTEX_CMPXCHG
-	u32 curval;
-
-	/*
-	 * This will fail and we want it. Some arch implementations do
-	 * runtime detection of the futex_atomic_cmpxchg_inatomic()
-	 * functionality. We want to know that before we call in any
-	 * of the complex code paths. Also we want to prevent
-	 * registration of robust lists in that case. NULL is
-	 * guaranteed to fault and we get -EFAULT on functional
-	 * implementation, the non-functional ones will return
-	 * -ENOSYS.
-	 */
-	if (futex_cmpxchg_value_locked(&curval, NULL, 0, 0) == -EFAULT)
-		futex_cmpxchg_enabled = 1;
-#endif
-}
-
 static int __init futex_init(void)
 {
 	unsigned int futex_shift;
@@ -1163,8 +1130,6 @@ static int __init futex_init(void)
 					       futex_hashsize, futex_hashsize);
 	futex_hashsize = 1UL << futex_shift;
 
-	futex_detect_cmpxchg();
-
 	for (i = 0; i < futex_hashsize; i++) {
 		atomic_set(&futex_queues[i].waiters, 0);
 		plist_head_init(&futex_queues[i].chain);

kernel/futex/futex.h

@@ -27,12 +27,6 @@
 #define FLAGS_CLOCKRT		0x02
 #define FLAGS_HAS_TIMEOUT	0x04
 
-#ifdef CONFIG_HAVE_FUTEX_CMPXCHG
-#define futex_cmpxchg_enabled 1
-#else
-extern int  __read_mostly futex_cmpxchg_enabled;
-#endif
-
 #ifdef CONFIG_FAIL_FUTEX
 extern bool should_fail_futex(bool fshared);
 #else

kernel/futex/syscalls.c

@@ -29,8 +29,6 @@
 SYSCALL_DEFINE2(set_robust_list, struct robust_list_head __user *, head,
 		size_t, len)
 {
-	if (!futex_cmpxchg_enabled)
-		return -ENOSYS;
 	/*
 	 * The kernel knows only one size for now:
 	 */
@@ -56,9 +54,6 @@ SYSCALL_DEFINE3(get_robust_list, int, pid,
 	unsigned long ret;
 	struct task_struct *p;
 
-	if (!futex_cmpxchg_enabled)
-		return -ENOSYS;
-
 	rcu_read_lock();
 
 	ret = -ESRCH;
@@ -103,17 +98,6 @@ long do_futex(u32 __user *uaddr, int op, u32 val, ktime_t *timeout,
 			return -ENOSYS;
 	}
 
-	switch (cmd) {
-	case FUTEX_LOCK_PI:
-	case FUTEX_LOCK_PI2:
-	case FUTEX_UNLOCK_PI:
-	case FUTEX_TRYLOCK_PI:
-	case FUTEX_WAIT_REQUEUE_PI:
-	case FUTEX_CMP_REQUEUE_PI:
-		if (!futex_cmpxchg_enabled)
-			return -ENOSYS;
-	}
-
 	switch (cmd) {
 	case FUTEX_WAIT:
 		val3 = FUTEX_BITSET_MATCH_ANY;
@@ -323,9 +307,6 @@ COMPAT_SYSCALL_DEFINE2(set_robust_list,
 		struct compat_robust_list_head __user *, head,
 		compat_size_t, len)
 {
-	if (!futex_cmpxchg_enabled)
-		return -ENOSYS;
-
 	if (unlikely(len != sizeof(*head)))
 		return -EINVAL;
 
@@ -342,9 +323,6 @@ COMPAT_SYSCALL_DEFINE3(get_robust_list, int, pid,
 	unsigned long ret;
 	struct task_struct *p;
 
-	if (!futex_cmpxchg_enabled)
-		return -ENOSYS;
-
 	rcu_read_lock();
 
 	ret = -ESRCH;

kernel/locking/lockdep.c

@@ -5485,6 +5485,7 @@ static noinstr void check_flags(unsigned long flags)
 		}
 	}
 
+#ifndef CONFIG_PREEMPT_RT
 	/*
 	 * We dont accurately track softirq state in e.g.
 	 * hardirq contexts (such as on 4KSTACKS), so only
@@ -5499,6 +5500,7 @@ static noinstr void check_flags(unsigned long flags)
 			DEBUG_LOCKS_WARN_ON(!current->softirqs_enabled);
 		}
 	}
+#endif
 
 	if (!debug_locks)
 		print_irqtrace_events(current);

kernel/locking/mutex.c

@@ -367,8 +367,7 @@ bool mutex_spin_on_owner(struct mutex *lock, struct task_struct *owner,
 		/*
 		 * Use vcpu_is_preempted to detect lock holder preemption issue.
 		 */
-		if (!owner->on_cpu || need_resched() ||
-				vcpu_is_preempted(task_cpu(owner))) {
+		if (!owner_on_cpu(owner) || need_resched()) {
 			ret = false;
 			break;
 		}
@@ -403,14 +402,8 @@ static inline int mutex_can_spin_on_owner(struct mutex *lock)
 	 * structure won't go away during the spinning period.
 	 */
 	owner = __mutex_owner(lock);
-
-	/*
-	 * As lock holder preemption issue, we both skip spinning if task is not
-	 * on cpu or its cpu is preempted
-	 */
-
 	if (owner)
-		retval = owner->on_cpu && !vcpu_is_preempted(task_cpu(owner));
+		retval = owner_on_cpu(owner);
 
 	/*
 	 * If lock->owner is not set, the mutex has been released. Return true

kernel/locking/rtmutex.c

@@ -1103,8 +1103,11 @@ static int __sched task_blocks_on_rt_mutex(struct rt_mutex_base *lock,
 	 * the other will detect the deadlock and return -EDEADLOCK,
 	 * which is wrong, as the other waiter is not in a deadlock
 	 * situation.
+	 *
+	 * Except for ww_mutex, in that case the chain walk must already deal
+	 * with spurious cycles, see the comments at [3] and [6].
 	 */
-	if (owner == task)
+	if (owner == task && !(build_ww_mutex() && ww_ctx))
 		return -EDEADLK;
 
 	raw_spin_lock(&task->pi_lock);
@@ -1379,9 +1382,8 @@ static bool rtmutex_spin_on_owner(struct rt_mutex_base *lock,
 		 *    for CONFIG_PREEMPT_RCU=y)
 		 *  - the VCPU on which owner runs is preempted
 		 */
-		if (!owner->on_cpu || need_resched() ||
-		    !rt_mutex_waiter_is_top_waiter(lock, waiter) ||
-		    vcpu_is_preempted(task_cpu(owner))) {
+		if (!owner_on_cpu(owner) || need_resched() ||
+		    !rt_mutex_waiter_is_top_waiter(lock, waiter)) {
 			res = false;
 			break;
 		}

kernel/locking/rtmutex_api.c

@@ -21,12 +21,13 @@ int max_lock_depth = 1024;
  */
 static __always_inline int __rt_mutex_lock_common(struct rt_mutex *lock,
 						  unsigned int state,
+						  struct lockdep_map *nest_lock,
 						  unsigned int subclass)
 {
 	int ret;
 
 	might_sleep();
-	mutex_acquire(&lock->dep_map, subclass, 0, _RET_IP_);
+	mutex_acquire_nest(&lock->dep_map, subclass, 0, nest_lock, _RET_IP_);
 	ret = __rt_mutex_lock(&lock->rtmutex, state);
 	if (ret)
 		mutex_release(&lock->dep_map, _RET_IP_);
@@ -48,10 +49,16 @@ EXPORT_SYMBOL(rt_mutex_base_init);
  */
 void __sched rt_mutex_lock_nested(struct rt_mutex *lock, unsigned int subclass)
 {
-	__rt_mutex_lock_common(lock, TASK_UNINTERRUPTIBLE, subclass);
+	__rt_mutex_lock_common(lock, TASK_UNINTERRUPTIBLE, NULL, subclass);
 }
 EXPORT_SYMBOL_GPL(rt_mutex_lock_nested);
 
+void __sched _rt_mutex_lock_nest_lock(struct rt_mutex *lock, struct lockdep_map *nest_lock)
+{
+	__rt_mutex_lock_common(lock, TASK_UNINTERRUPTIBLE, nest_lock, 0);
+}
+EXPORT_SYMBOL_GPL(_rt_mutex_lock_nest_lock);
+
 #else /* !CONFIG_DEBUG_LOCK_ALLOC */
 
 /**
@@ -61,7 +68,7 @@ EXPORT_SYMBOL_GPL(rt_mutex_lock_nested);
  */
 void __sched rt_mutex_lock(struct rt_mutex *lock)
 {
-	__rt_mutex_lock_common(lock, TASK_UNINTERRUPTIBLE, 0);
+	__rt_mutex_lock_common(lock, TASK_UNINTERRUPTIBLE, NULL, 0);
 }
 EXPORT_SYMBOL_GPL(rt_mutex_lock);
 #endif
@@ -77,10 +84,25 @@ EXPORT_SYMBOL_GPL(rt_mutex_lock);
  */
 int __sched rt_mutex_lock_interruptible(struct rt_mutex *lock)
 {
-	return __rt_mutex_lock_common(lock, TASK_INTERRUPTIBLE, 0);
+	return __rt_mutex_lock_common(lock, TASK_INTERRUPTIBLE, NULL, 0);
 }
 EXPORT_SYMBOL_GPL(rt_mutex_lock_interruptible);
 
+/**
+ * rt_mutex_lock_killable - lock a rt_mutex killable
+ *
+ * @lock:		the rt_mutex to be locked
+ *
+ * Returns:
+ *  0		on success
+ * -EINTR	when interrupted by a signal
+ */
+int __sched rt_mutex_lock_killable(struct rt_mutex *lock)
+{
+	return __rt_mutex_lock_common(lock, TASK_KILLABLE, NULL, 0);
+}
+EXPORT_SYMBOL_GPL(rt_mutex_lock_killable);
+
 /**
  * rt_mutex_trylock - try to lock a rt_mutex
  *

kernel/locking/rwsem.c

@@ -658,15 +658,6 @@ static inline bool rwsem_try_write_lock_unqueued(struct rw_semaphore *sem)
 	return false;
 }
 
-static inline bool owner_on_cpu(struct task_struct *owner)
-{
-	/*
-	 * As lock holder preemption issue, we both skip spinning if
-	 * task is not on cpu or its cpu is preempted
-	 */
-	return owner->on_cpu && !vcpu_is_preempted(task_cpu(owner));
-}
-
 static inline bool rwsem_can_spin_on_owner(struct rw_semaphore *sem)
 {
 	struct task_struct *owner;

kernel/locking/spinlock_rt.c

@@ -257,12 +257,6 @@ void __sched rt_write_unlock(rwlock_t *rwlock)
 }
 EXPORT_SYMBOL(rt_write_unlock);
 
-int __sched rt_rwlock_is_contended(rwlock_t *rwlock)
-{
-	return rw_base_is_contended(&rwlock->rwbase);
-}
-EXPORT_SYMBOL(rt_rwlock_is_contended);
-
 #ifdef CONFIG_DEBUG_LOCK_ALLOC
 void __rt_rwlock_init(rwlock_t *rwlock, const char *name,
 		      struct lock_class_key *key)

kernel/locking/ww_rt_mutex.c

@@ -26,7 +26,7 @@ int ww_mutex_trylock(struct ww_mutex *lock, struct ww_acquire_ctx *ww_ctx)
 
 	if (__rt_mutex_trylock(&rtm->rtmutex)) {
 		ww_mutex_set_context_fastpath(lock, ww_ctx);
-		mutex_acquire_nest(&rtm->dep_map, 0, 1, ww_ctx->dep_map, _RET_IP_);
+		mutex_acquire_nest(&rtm->dep_map, 0, 1, &ww_ctx->dep_map, _RET_IP_);
 		return 1;
 	}
 

kernel/sched/core.c

@@ -2184,6 +2184,9 @@ void migrate_enable(void)
 		return;
 	}
 
+	if (WARN_ON_ONCE(!p->migration_disabled))
+		return;
+
 	/*
 	 * Ensure stop_task runs either before or after this, and that
 	 * __set_cpus_allowed_ptr(SCA_MIGRATE_ENABLE) doesn't schedule().

lib/atomic64.c

@@ -118,7 +118,6 @@ ATOMIC64_OPS(sub, -=)
 #undef ATOMIC64_OPS
 #define ATOMIC64_OPS(op, c_op)						\
 	ATOMIC64_OP(op, c_op)						\
-	ATOMIC64_OP_RETURN(op, c_op)					\
 	ATOMIC64_FETCH_OP(op, c_op)
 
 ATOMIC64_OPS(and, &=)
@@ -127,7 +126,6 @@ ATOMIC64_OPS(xor, ^=)
 
 #undef ATOMIC64_OPS
 #undef ATOMIC64_FETCH_OP
-#undef ATOMIC64_OP_RETURN
 #undef ATOMIC64_OP
 
 s64 generic_atomic64_dec_if_positive(atomic64_t *v)

lib/locking-selftest.c

@@ -26,6 +26,12 @@
 #include <linux/rtmutex.h>
 #include <linux/local_lock.h>
 
+#ifdef CONFIG_PREEMPT_RT
+# define NON_RT(...)
+#else
+# define NON_RT(...)	__VA_ARGS__
+#endif
+
 /*
  * Change this to 1 if you want to see the failure printouts:
  */
@@ -139,7 +145,7 @@ static DEFINE_RT_MUTEX(rtmutex_Z2);
 
 #endif
 
-static local_lock_t local_A = INIT_LOCAL_LOCK(local_A);
+static DEFINE_PER_CPU(local_lock_t, local_A);
 
 /*
  * non-inlined runtime initializers, to let separate locks share
@@ -712,12 +718,18 @@ GENERATE_TESTCASE(ABCDBCDA_rtmutex);
 
 #undef E
 
+#ifdef CONFIG_PREEMPT_RT
+# define RT_PREPARE_DBL_UNLOCK()	{ migrate_disable(); rcu_read_lock(); }
+#else
+# define RT_PREPARE_DBL_UNLOCK()
+#endif
 /*
  * Double unlock:
  */
 #define E()					\
 						\
 	LOCK(A);				\
+	RT_PREPARE_DBL_UNLOCK();		\
 	UNLOCK(A);				\
 	UNLOCK(A); /* fail */
 
@@ -802,6 +814,7 @@ GENERATE_PERMUTATIONS_2_EVENTS(irqsafe1_hard_rlock)
 #include "locking-selftest-wlock-hardirq.h"
 GENERATE_PERMUTATIONS_2_EVENTS(irqsafe1_hard_wlock)
 
+#ifndef CONFIG_PREEMPT_RT
 #include "locking-selftest-spin-softirq.h"
 GENERATE_PERMUTATIONS_2_EVENTS(irqsafe1_soft_spin)
 
@@ -810,10 +823,12 @@ GENERATE_PERMUTATIONS_2_EVENTS(irqsafe1_soft_rlock)
 
 #include "locking-selftest-wlock-softirq.h"
 GENERATE_PERMUTATIONS_2_EVENTS(irqsafe1_soft_wlock)
+#endif
 
 #undef E1
 #undef E2
 
+#ifndef CONFIG_PREEMPT_RT
 /*
  * Enabling hardirqs with a softirq-safe lock held:
  */
@@ -846,6 +861,8 @@ GENERATE_PERMUTATIONS_2_EVENTS(irqsafe2A_rlock)
 #undef E1
 #undef E2
 
+#endif
+
 /*
  * Enabling irqs with an irq-safe lock held:
  */
@@ -875,6 +892,7 @@ GENERATE_PERMUTATIONS_2_EVENTS(irqsafe2B_hard_rlock)
 #include "locking-selftest-wlock-hardirq.h"
 GENERATE_PERMUTATIONS_2_EVENTS(irqsafe2B_hard_wlock)
 
+#ifndef CONFIG_PREEMPT_RT
 #include "locking-selftest-spin-softirq.h"
 GENERATE_PERMUTATIONS_2_EVENTS(irqsafe2B_soft_spin)
 
@@ -883,6 +901,7 @@ GENERATE_PERMUTATIONS_2_EVENTS(irqsafe2B_soft_rlock)
 
 #include "locking-selftest-wlock-softirq.h"
 GENERATE_PERMUTATIONS_2_EVENTS(irqsafe2B_soft_wlock)
+#endif
 
 #undef E1
 #undef E2
@@ -921,6 +940,7 @@ GENERATE_PERMUTATIONS_3_EVENTS(irqsafe3_hard_rlock)
 #include "locking-selftest-wlock-hardirq.h"
 GENERATE_PERMUTATIONS_3_EVENTS(irqsafe3_hard_wlock)
 
+#ifndef CONFIG_PREEMPT_RT
 #include "locking-selftest-spin-softirq.h"
 GENERATE_PERMUTATIONS_3_EVENTS(irqsafe3_soft_spin)
 
@@ -929,6 +949,7 @@ GENERATE_PERMUTATIONS_3_EVENTS(irqsafe3_soft_rlock)
 
 #include "locking-selftest-wlock-softirq.h"
 GENERATE_PERMUTATIONS_3_EVENTS(irqsafe3_soft_wlock)
+#endif
 
 #undef E1
 #undef E2
@@ -969,6 +990,7 @@ GENERATE_PERMUTATIONS_3_EVENTS(irqsafe4_hard_rlock)
 #include "locking-selftest-wlock-hardirq.h"
 GENERATE_PERMUTATIONS_3_EVENTS(irqsafe4_hard_wlock)
 
+#ifndef CONFIG_PREEMPT_RT
 #include "locking-selftest-spin-softirq.h"
 GENERATE_PERMUTATIONS_3_EVENTS(irqsafe4_soft_spin)
 
@@ -977,6 +999,7 @@ GENERATE_PERMUTATIONS_3_EVENTS(irqsafe4_soft_rlock)
 
 #include "locking-selftest-wlock-softirq.h"
 GENERATE_PERMUTATIONS_3_EVENTS(irqsafe4_soft_wlock)
+#endif
 
 #undef E1
 #undef E2
@@ -1031,6 +1054,7 @@ GENERATE_PERMUTATIONS_3_EVENTS(irq_inversion_hard_rlock)
 #include "locking-selftest-wlock-hardirq.h"
 GENERATE_PERMUTATIONS_3_EVENTS(irq_inversion_hard_wlock)
 
+#ifndef CONFIG_PREEMPT_RT
 #include "locking-selftest-spin-softirq.h"
 GENERATE_PERMUTATIONS_3_EVENTS(irq_inversion_soft_spin)
 
@@ -1039,6 +1063,7 @@ GENERATE_PERMUTATIONS_3_EVENTS(irq_inversion_soft_rlock)
 
 #include "locking-selftest-wlock-softirq.h"
 GENERATE_PERMUTATIONS_3_EVENTS(irq_inversion_soft_wlock)
+#endif
 
 #undef E1
 #undef E2
@@ -1206,12 +1231,14 @@ GENERATE_PERMUTATIONS_3_EVENTS(irq_read_recursion_hard_rlock)
 #include "locking-selftest-wlock.h"
 GENERATE_PERMUTATIONS_3_EVENTS(irq_read_recursion_hard_wlock)
 
+#ifndef CONFIG_PREEMPT_RT
 #include "locking-selftest-softirq.h"
 #include "locking-selftest-rlock.h"
 GENERATE_PERMUTATIONS_3_EVENTS(irq_read_recursion_soft_rlock)
 
 #include "locking-selftest-wlock.h"
 GENERATE_PERMUTATIONS_3_EVENTS(irq_read_recursion_soft_wlock)
+#endif
 
 #undef E1
 #undef E2
@@ -1252,12 +1279,14 @@ GENERATE_PERMUTATIONS_3_EVENTS(irq_read_recursion2_hard_rlock)
 #include "locking-selftest-wlock.h"
 GENERATE_PERMUTATIONS_3_EVENTS(irq_read_recursion2_hard_wlock)
 
+#ifndef CONFIG_PREEMPT_RT
 #include "locking-selftest-softirq.h"
 #include "locking-selftest-rlock.h"
 GENERATE_PERMUTATIONS_3_EVENTS(irq_read_recursion2_soft_rlock)
 
 #include "locking-selftest-wlock.h"
 GENERATE_PERMUTATIONS_3_EVENTS(irq_read_recursion2_soft_wlock)
+#endif
 
 #undef E1
 #undef E2
@@ -1306,12 +1335,14 @@ GENERATE_PERMUTATIONS_3_EVENTS(irq_read_recursion3_hard_rlock)
 #include "locking-selftest-wlock.h"
 GENERATE_PERMUTATIONS_3_EVENTS(irq_read_recursion3_hard_wlock)
 
+#ifndef CONFIG_PREEMPT_RT
 #include "locking-selftest-softirq.h"
 #include "locking-selftest-rlock.h"
 GENERATE_PERMUTATIONS_3_EVENTS(irq_read_recursion3_soft_rlock)
 
 #include "locking-selftest-wlock.h"
 GENERATE_PERMUTATIONS_3_EVENTS(irq_read_recursion3_soft_wlock)
+#endif
 
 #ifdef CONFIG_DEBUG_LOCK_ALLOC
 # define I_SPINLOCK(x)	lockdep_reset_lock(&lock_##x.dep_map)
@@ -1320,7 +1351,7 @@ GENERATE_PERMUTATIONS_3_EVENTS(irq_read_recursion3_soft_wlock)
 # define I_MUTEX(x)	lockdep_reset_lock(&mutex_##x.dep_map)
 # define I_RWSEM(x)	lockdep_reset_lock(&rwsem_##x.dep_map)
 # define I_WW(x)	lockdep_reset_lock(&x.dep_map)
-# define I_LOCAL_LOCK(x) lockdep_reset_lock(&local_##x.dep_map)
+# define I_LOCAL_LOCK(x) lockdep_reset_lock(this_cpu_ptr(&local_##x.dep_map))
 #ifdef CONFIG_RT_MUTEXES
 # define I_RTMUTEX(x)	lockdep_reset_lock(&rtmutex_##x.dep_map)
 #endif
@@ -1380,7 +1411,7 @@ static void reset_locks(void)
 	init_shared_classes();
 	raw_spin_lock_init(&raw_lock_A);
 	raw_spin_lock_init(&raw_lock_B);
-	local_lock_init(&local_A);
+	local_lock_init(this_cpu_ptr(&local_A));
 
 	ww_mutex_init(&o, &ww_lockdep); ww_mutex_init(&o2, &ww_lockdep); ww_mutex_init(&o3, &ww_lockdep);
 	memset(&t, 0, sizeof(t)); memset(&t2, 0, sizeof(t2));
@@ -1398,7 +1429,13 @@ static int unexpected_testcase_failures;
 
 static void dotest(void (*testcase_fn)(void), int expected, int lockclass_mask)
 {
-	unsigned long saved_preempt_count = preempt_count();
+	int saved_preempt_count = preempt_count();
+#ifdef CONFIG_PREEMPT_RT
+#ifdef CONFIG_SMP
+	int saved_mgd_count = current->migration_disabled;
+#endif
+	int saved_rcu_count = current->rcu_read_lock_nesting;
+#endif
 
 	WARN_ON(irqs_disabled());
 
@@ -1432,6 +1469,18 @@ static void dotest(void (*testcase_fn)(void), int expected, int lockclass_mask)
 	 * count, so restore it:
 	 */
 	preempt_count_set(saved_preempt_count);
+
+#ifdef CONFIG_PREEMPT_RT
+#ifdef CONFIG_SMP
+	while (current->migration_disabled > saved_mgd_count)
+		migrate_enable();
+#endif
+
+	while (current->rcu_read_lock_nesting > saved_rcu_count)
+		rcu_read_unlock();
+	WARN_ON_ONCE(current->rcu_read_lock_nesting < saved_rcu_count);
+#endif
+
 #ifdef CONFIG_TRACE_IRQFLAGS
 	if (softirq_count())
 		current->softirqs_enabled = 0;
@@ -1499,7 +1548,7 @@ static inline void print_testname(const char *testname)
 
 #define DO_TESTCASE_2x2RW(desc, name, nr)			\
 	DO_TESTCASE_2RW("hard-"desc, name##_hard, nr)		\
-	DO_TESTCASE_2RW("soft-"desc, name##_soft, nr)		\
+	NON_RT(DO_TESTCASE_2RW("soft-"desc, name##_soft, nr))	\
 
 #define DO_TESTCASE_6x2x2RW(desc, name)				\
 	DO_TESTCASE_2x2RW(desc, name, 123);			\
@@ -1547,19 +1596,19 @@ static inline void print_testname(const char *testname)
 
 #define DO_TESTCASE_2I(desc, name, nr)				\
 	DO_TESTCASE_1("hard-"desc, name##_hard, nr);		\
-	DO_TESTCASE_1("soft-"desc, name##_soft, nr);
+	NON_RT(DO_TESTCASE_1("soft-"desc, name##_soft, nr));
 
 #define DO_TESTCASE_2IB(desc, name, nr)				\
 	DO_TESTCASE_1B("hard-"desc, name##_hard, nr);		\
-	DO_TESTCASE_1B("soft-"desc, name##_soft, nr);
+	NON_RT(DO_TESTCASE_1B("soft-"desc, name##_soft, nr));
 
 #define DO_TESTCASE_6I(desc, name, nr)				\
 	DO_TESTCASE_3("hard-"desc, name##_hard, nr);		\
-	DO_TESTCASE_3("soft-"desc, name##_soft, nr);
+	NON_RT(DO_TESTCASE_3("soft-"desc, name##_soft, nr));
 
 #define DO_TESTCASE_6IRW(desc, name, nr)			\
 	DO_TESTCASE_3RW("hard-"desc, name##_hard, nr);		\
-	DO_TESTCASE_3RW("soft-"desc, name##_soft, nr);
+	NON_RT(DO_TESTCASE_3RW("soft-"desc, name##_soft, nr));
 
 #define DO_TESTCASE_2x3(desc, name)				\
 	DO_TESTCASE_3(desc, name, 12);				\
@@ -1651,6 +1700,22 @@ static void ww_test_fail_acquire(void)
 #endif
 }
 
+#ifdef CONFIG_PREEMPT_RT
+#define ww_mutex_base_lock(b)			rt_mutex_lock(b)
+#define ww_mutex_base_trylock(b)		rt_mutex_trylock(b)
+#define ww_mutex_base_lock_nest_lock(b, b2)	rt_mutex_lock_nest_lock(b, b2)
+#define ww_mutex_base_lock_interruptible(b)	rt_mutex_lock_interruptible(b)
+#define ww_mutex_base_lock_killable(b)		rt_mutex_lock_killable(b)
+#define ww_mutex_base_unlock(b)			rt_mutex_unlock(b)
+#else
+#define ww_mutex_base_lock(b)			mutex_lock(b)
+#define ww_mutex_base_trylock(b)		mutex_trylock(b)
+#define ww_mutex_base_lock_nest_lock(b, b2)	mutex_lock_nest_lock(b, b2)
+#define ww_mutex_base_lock_interruptible(b)	mutex_lock_interruptible(b)
+#define ww_mutex_base_lock_killable(b)		mutex_lock_killable(b)
+#define ww_mutex_base_unlock(b)			mutex_unlock(b)
+#endif
+
 static void ww_test_normal(void)
 {
 	int ret;
@@ -1665,50 +1730,50 @@ static void ww_test_normal(void)
 
 	/* mutex_lock (and indirectly, mutex_lock_nested) */
 	o.ctx = (void *)~0UL;
-	mutex_lock(&o.base);
-	mutex_unlock(&o.base);
+	ww_mutex_base_lock(&o.base);
+	ww_mutex_base_unlock(&o.base);
 	WARN_ON(o.ctx != (void *)~0UL);
 
 	/* mutex_lock_interruptible (and *_nested) */
 	o.ctx = (void *)~0UL;
-	ret = mutex_lock_interruptible(&o.base);
+	ret = ww_mutex_base_lock_interruptible(&o.base);
 	if (!ret)
-		mutex_unlock(&o.base);
+		ww_mutex_base_unlock(&o.base);
 	else
 		WARN_ON(1);
 	WARN_ON(o.ctx != (void *)~0UL);
 
 	/* mutex_lock_killable (and *_nested) */
 	o.ctx = (void *)~0UL;
-	ret = mutex_lock_killable(&o.base);
+	ret = ww_mutex_base_lock_killable(&o.base);
 	if (!ret)
-		mutex_unlock(&o.base);
+		ww_mutex_base_unlock(&o.base);
 	else
 		WARN_ON(1);
 	WARN_ON(o.ctx != (void *)~0UL);
 
 	/* trylock, succeeding */
 	o.ctx = (void *)~0UL;
-	ret = mutex_trylock(&o.base);
+	ret = ww_mutex_base_trylock(&o.base);
 	WARN_ON(!ret);
 	if (ret)
-		mutex_unlock(&o.base);
+		ww_mutex_base_unlock(&o.base);
 	else
 		WARN_ON(1);
 	WARN_ON(o.ctx != (void *)~0UL);
 
 	/* trylock, failing */
 	o.ctx = (void *)~0UL;
-	mutex_lock(&o.base);
-	ret = mutex_trylock(&o.base);
+	ww_mutex_base_lock(&o.base);
+	ret = ww_mutex_base_trylock(&o.base);
 	WARN_ON(ret);
-	mutex_unlock(&o.base);
+	ww_mutex_base_unlock(&o.base);
 	WARN_ON(o.ctx != (void *)~0UL);
 
 	/* nest_lock */
 	o.ctx = (void *)~0UL;
-	mutex_lock_nest_lock(&o.base, &t);
-	mutex_unlock(&o.base);
+	ww_mutex_base_lock_nest_lock(&o.base, &t);
+	ww_mutex_base_unlock(&o.base);
 	WARN_ON(o.ctx != (void *)~0UL);
 }
 
@@ -1721,7 +1786,7 @@ static void ww_test_two_contexts(void)
 static void ww_test_diff_class(void)
 {
 	WWAI(&t);
-#ifdef CONFIG_DEBUG_MUTEXES
+#ifdef DEBUG_WW_MUTEXES
 	t.ww_class = NULL;
 #endif
 	WWL(&o, &t);
@@ -1785,7 +1850,7 @@ static void ww_test_edeadlk_normal(void)
 {
 	int ret;
 
-	mutex_lock(&o2.base);
+	ww_mutex_base_lock(&o2.base);
 	o2.ctx = &t2;
 	mutex_release(&o2.base.dep_map, _THIS_IP_);
 
@@ -1801,7 +1866,7 @@ static void ww_test_edeadlk_normal(void)
 
 	o2.ctx = NULL;
 	mutex_acquire(&o2.base.dep_map, 0, 1, _THIS_IP_);
-	mutex_unlock(&o2.base);
+	ww_mutex_base_unlock(&o2.base);
 	WWU(&o);
 
 	WWL(&o2, &t);
@@ -1811,7 +1876,7 @@ static void ww_test_edeadlk_normal_slow(void)
 {
 	int ret;
 
-	mutex_lock(&o2.base);
+	ww_mutex_base_lock(&o2.base);
 	mutex_release(&o2.base.dep_map, _THIS_IP_);
 	o2.ctx = &t2;
 
@@ -1827,7 +1892,7 @@ static void ww_test_edeadlk_normal_slow(void)
 
 	o2.ctx = NULL;
 	mutex_acquire(&o2.base.dep_map, 0, 1, _THIS_IP_);
-	mutex_unlock(&o2.base);
+	ww_mutex_base_unlock(&o2.base);
 	WWU(&o);
 
 	ww_mutex_lock_slow(&o2, &t);
@@ -1837,7 +1902,7 @@ static void ww_test_edeadlk_no_unlock(void)
 {
 	int ret;
 
-	mutex_lock(&o2.base);
+	ww_mutex_base_lock(&o2.base);
 	o2.ctx = &t2;
 	mutex_release(&o2.base.dep_map, _THIS_IP_);
 
@@ -1853,7 +1918,7 @@ static void ww_test_edeadlk_no_unlock(void)
 
 	o2.ctx = NULL;
 	mutex_acquire(&o2.base.dep_map, 0, 1, _THIS_IP_);
-	mutex_unlock(&o2.base);
+	ww_mutex_base_unlock(&o2.base);
 
 	WWL(&o2, &t);
 }
@@ -1862,7 +1927,7 @@ static void ww_test_edeadlk_no_unlock_slow(void)
 {
 	int ret;
 
-	mutex_lock(&o2.base);
+	ww_mutex_base_lock(&o2.base);
 	mutex_release(&o2.base.dep_map, _THIS_IP_);
 	o2.ctx = &t2;
 
@@ -1878,7 +1943,7 @@ static void ww_test_edeadlk_no_unlock_slow(void)
 
 	o2.ctx = NULL;
 	mutex_acquire(&o2.base.dep_map, 0, 1, _THIS_IP_);
-	mutex_unlock(&o2.base);
+	ww_mutex_base_unlock(&o2.base);
 
 	ww_mutex_lock_slow(&o2, &t);
 }
@@ -1887,7 +1952,7 @@ static void ww_test_edeadlk_acquire_more(void)
 {
 	int ret;
 
-	mutex_lock(&o2.base);
+	ww_mutex_base_lock(&o2.base);
 	mutex_release(&o2.base.dep_map, _THIS_IP_);
 	o2.ctx = &t2;
 
@@ -1908,7 +1973,7 @@ static void ww_test_edeadlk_acquire_more_slow(void)
 {
 	int ret;
 
-	mutex_lock(&o2.base);
+	ww_mutex_base_lock(&o2.base);
 	mutex_release(&o2.base.dep_map, _THIS_IP_);
 	o2.ctx = &t2;
 
@@ -1929,11 +1994,11 @@ static void ww_test_edeadlk_acquire_more_edeadlk(void)
 {
 	int ret;
 
-	mutex_lock(&o2.base);
+	ww_mutex_base_lock(&o2.base);
 	mutex_release(&o2.base.dep_map, _THIS_IP_);
 	o2.ctx = &t2;
 
-	mutex_lock(&o3.base);
+	ww_mutex_base_lock(&o3.base);
 	mutex_release(&o3.base.dep_map, _THIS_IP_);
 	o3.ctx = &t2;
 
@@ -1955,11 +2020,11 @@ static void ww_test_edeadlk_acquire_more_edeadlk_slow(void)
 {
 	int ret;
 
-	mutex_lock(&o2.base);
+	ww_mutex_base_lock(&o2.base);
 	mutex_release(&o2.base.dep_map, _THIS_IP_);
 	o2.ctx = &t2;
 
-	mutex_lock(&o3.base);
+	ww_mutex_base_lock(&o3.base);
 	mutex_release(&o3.base.dep_map, _THIS_IP_);
 	o3.ctx = &t2;
 
@@ -1980,7 +2045,7 @@ static void ww_test_edeadlk_acquire_wrong(void)
 {
 	int ret;
 
-	mutex_lock(&o2.base);
+	ww_mutex_base_lock(&o2.base);
 	mutex_release(&o2.base.dep_map, _THIS_IP_);
 	o2.ctx = &t2;
 
@@ -2005,7 +2070,7 @@ static void ww_test_edeadlk_acquire_wrong_slow(void)
 {
 	int ret;
 
-	mutex_lock(&o2.base);
+	ww_mutex_base_lock(&o2.base);
 	mutex_release(&o2.base.dep_map, _THIS_IP_);
 	o2.ctx = &t2;
 
@@ -2646,8 +2711,8 @@ static void wait_context_tests(void)
 
 static void local_lock_2(void)
 {
-	local_lock_acquire(&local_A);	/* IRQ-ON */
-	local_lock_release(&local_A);
+	local_lock(&local_A);	/* IRQ-ON */
+	local_unlock(&local_A);
 
 	HARDIRQ_ENTER();
 	spin_lock(&lock_A);		/* IN-IRQ */
@@ -2656,18 +2721,18 @@ static void local_lock_2(void)
 
 	HARDIRQ_DISABLE();
 	spin_lock(&lock_A);
-	local_lock_acquire(&local_A);	/* IN-IRQ <-> IRQ-ON cycle, false */
-	local_lock_release(&local_A);
+	local_lock(&local_A);	/* IN-IRQ <-> IRQ-ON cycle, false */
+	local_unlock(&local_A);
 	spin_unlock(&lock_A);
 	HARDIRQ_ENABLE();
 }
 
 static void local_lock_3A(void)
 {
-	local_lock_acquire(&local_A);	/* IRQ-ON */
+	local_lock(&local_A);	/* IRQ-ON */
 	spin_lock(&lock_B);		/* IRQ-ON */
 	spin_unlock(&lock_B);
-	local_lock_release(&local_A);
+	local_unlock(&local_A);
 
 	HARDIRQ_ENTER();
 	spin_lock(&lock_A);		/* IN-IRQ */
@@ -2676,18 +2741,18 @@ static void local_lock_3A(void)
 
 	HARDIRQ_DISABLE();
 	spin_lock(&lock_A);
-	local_lock_acquire(&local_A);	/* IN-IRQ <-> IRQ-ON cycle only if we count local_lock(), false */
-	local_lock_release(&local_A);
+	local_lock(&local_A);	/* IN-IRQ <-> IRQ-ON cycle only if we count local_lock(), false */
+	local_unlock(&local_A);
 	spin_unlock(&lock_A);
 	HARDIRQ_ENABLE();
 }
 
 static void local_lock_3B(void)
 {
-	local_lock_acquire(&local_A);	/* IRQ-ON */
+	local_lock(&local_A);	/* IRQ-ON */
 	spin_lock(&lock_B);		/* IRQ-ON */
 	spin_unlock(&lock_B);
-	local_lock_release(&local_A);
+	local_unlock(&local_A);
 
 	HARDIRQ_ENTER();
 	spin_lock(&lock_A);		/* IN-IRQ */
@@ -2696,8 +2761,8 @@ static void local_lock_3B(void)
 
 	HARDIRQ_DISABLE();
 	spin_lock(&lock_A);
-	local_lock_acquire(&local_A);	/* IN-IRQ <-> IRQ-ON cycle only if we count local_lock(), false */
-	local_lock_release(&local_A);
+	local_lock(&local_A);	/* IN-IRQ <-> IRQ-ON cycle only if we count local_lock(), false */
+	local_unlock(&local_A);
 	spin_unlock(&lock_A);
 	HARDIRQ_ENABLE();
 
@@ -2812,7 +2877,7 @@ void locking_selftest(void)
 	printk("------------------------\n");
 	printk("| Locking API testsuite:\n");
 	printk("----------------------------------------------------------------------------\n");
-	printk("                                 | spin |wlock |rlock |mutex | wsem | rsem |\n");
+	printk("                                 | spin |wlock |rlock |mutex | wsem | rsem |rtmutex\n");
 	printk("  --------------------------------------------------------------------------\n");
 
 	init_shared_classes();
@@ -2885,12 +2950,11 @@ void locking_selftest(void)
 	DO_TESTCASE_6x1RR("rlock W1R2/R2R3/W3W1", W1R2_R2R3_W3W1);
 
 	printk("  --------------------------------------------------------------------------\n");
-
 	/*
 	 * irq-context testcases:
 	 */
 	DO_TESTCASE_2x6("irqs-on + irq-safe-A", irqsafe1);
-	DO_TESTCASE_2x3("sirq-safe-A => hirqs-on", irqsafe2A);
+	NON_RT(DO_TESTCASE_2x3("sirq-safe-A => hirqs-on", irqsafe2A));
 	DO_TESTCASE_2x6("safe-A + irqs-on", irqsafe2B);
 	DO_TESTCASE_6x6("safe-A + unsafe-B #1", irqsafe3);
 	DO_TESTCASE_6x6("safe-A + unsafe-B #2", irqsafe4);