Merge tag 'locking-core-2025-03-22' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull locking updates from Ingo Molnar:
 "Locking primitives:
   - Micro-optimize percpu_{,try_}cmpxchg{64,128}_op() and
     {,try_}cmpxchg{64,128} on x86 (Uros Bizjak)
   - mutexes: extend debug checks in mutex_lock() (Yunhui Cui)
   - Misc cleanups (Uros Bizjak)

  Lockdep:
   - Fix might_fault() lockdep check of current->mm->mmap_lock (Peter
     Zijlstra)
   - Don't disable interrupts on RT in disable_irq_nosync_lockdep.*()
     (Sebastian Andrzej Siewior)
   - Disable KASAN instrumentation of lockdep.c (Waiman Long)
   - Add kasan_check_byte() check in lock_acquire() (Waiman Long)
   - Misc cleanups (Sebastian Andrzej Siewior)

  Rust runtime integration:
   - Use Pin for all LockClassKey usages (Mitchell Levy)
   - sync: Add accessor for the lock behind a given guard (Alice Ryhl)
   - sync: condvar: Add wait_interruptible_freezable() (Alice Ryhl)
   - sync: lock: Add an example for Guard:: Lock_ref() (Boqun Feng)

  Split-lock detection feature (x86):
   - Fix warning mode with disabled mitigation mode (Maksim Davydov)

  Locking events:
   - Add locking events for rtmutex slow paths (Waiman Long)
   - Add locking events for lockdep (Waiman Long)"

* tag 'locking-core-2025-03-22' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  lockdep: Remove disable_irq_lockdep()
  lockdep: Don't disable interrupts on RT in disable_irq_nosync_lockdep.*()
  rust: lockdep: Use Pin for all LockClassKey usages
  rust: sync: condvar: Add wait_interruptible_freezable()
  rust: sync: lock: Add an example for Guard:: Lock_ref()
  rust: sync: Add accessor for the lock behind a given guard
  locking/lockdep: Add kasan_check_byte() check in lock_acquire()
  locking/lockdep: Disable KASAN instrumentation of lockdep.c
  locking/lock_events: Add locking events for lockdep
  locking/lock_events: Add locking events for rtmutex slow paths
  x86/split_lock: Fix the delayed detection logic
  lockdep/mm: Fix might_fault() lockdep check of current->mm->mmap_lock
  x86/locking: Remove semicolon from "lock" prefix
  locking/mutex: Add MUTEX_WARN_ON() into fast path
  x86/locking: Use asm_inline for {,try_}cmpxchg{64,128} emulations
  x86/locking: Use ALT_OUTPUT_SP() for percpu_{,try_}cmpxchg{64,128}_op()

Author: torvalds

arch/x86/include/asm/alternative.h

@@ -48,7 +48,7 @@
 		".popsection\n"				\
 		"671:"
 
-#define LOCK_PREFIX LOCK_PREFIX_HERE "\n\tlock; "
+#define LOCK_PREFIX LOCK_PREFIX_HERE "\n\tlock "
 
 #else /* ! CONFIG_SMP */
 #define LOCK_PREFIX_HERE ""

arch/x86/include/asm/barrier.h

@@ -12,11 +12,11 @@
  */
 
 #ifdef CONFIG_X86_32
-#define mb() asm volatile(ALTERNATIVE("lock; addl $0,-4(%%esp)", "mfence", \
+#define mb() asm volatile(ALTERNATIVE("lock addl $0,-4(%%esp)", "mfence", \
 				      X86_FEATURE_XMM2) ::: "memory", "cc")
-#define rmb() asm volatile(ALTERNATIVE("lock; addl $0,-4(%%esp)", "lfence", \
+#define rmb() asm volatile(ALTERNATIVE("lock addl $0,-4(%%esp)", "lfence", \
 				       X86_FEATURE_XMM2) ::: "memory", "cc")
-#define wmb() asm volatile(ALTERNATIVE("lock; addl $0,-4(%%esp)", "sfence", \
+#define wmb() asm volatile(ALTERNATIVE("lock addl $0,-4(%%esp)", "sfence", \
 				       X86_FEATURE_XMM2) ::: "memory", "cc")
 #else
 #define __mb()	asm volatile("mfence":::"memory")
@@ -50,7 +50,7 @@
 #define __dma_rmb()	barrier()
 #define __dma_wmb()	barrier()
 
-#define __smp_mb()	asm volatile("lock; addl $0,-4(%%" _ASM_SP ")" ::: "memory", "cc")
+#define __smp_mb()	asm volatile("lock addl $0,-4(%%" _ASM_SP ")" ::: "memory", "cc")
 
 #define __smp_rmb()	dma_rmb()
 #define __smp_wmb()	barrier()

arch/x86/include/asm/cmpxchg.h

@@ -134,7 +134,7 @@ extern void __add_wrong_size(void)
 	__raw_cmpxchg((ptr), (old), (new), (size), LOCK_PREFIX)
 
 #define __sync_cmpxchg(ptr, old, new, size)				\
-	__raw_cmpxchg((ptr), (old), (new), (size), "lock; ")
+	__raw_cmpxchg((ptr), (old), (new), (size), "lock ")
 
 #define __cmpxchg_local(ptr, old, new, size)				\
 	__raw_cmpxchg((ptr), (old), (new), (size), "")
@@ -222,7 +222,7 @@ extern void __add_wrong_size(void)
 	__raw_try_cmpxchg((ptr), (pold), (new), (size), LOCK_PREFIX)
 
 #define __sync_try_cmpxchg(ptr, pold, new, size)			\
-	__raw_try_cmpxchg((ptr), (pold), (new), (size), "lock; ")
+	__raw_try_cmpxchg((ptr), (pold), (new), (size), "lock ")
 
 #define __try_cmpxchg_local(ptr, pold, new, size)			\
 	__raw_try_cmpxchg((ptr), (pold), (new), (size), "")

arch/x86/include/asm/cmpxchg_32.h

@@ -91,19 +91,21 @@ static __always_inline bool __try_cmpxchg64_local(volatile u64 *ptr, u64 *oldp,
 	union __u64_halves o = { .full = (_old), },			\
 			   n = { .full = (_new), };			\
 									\
-	asm volatile(ALTERNATIVE(_lock_loc				\
-				 "call cmpxchg8b_emu",			\
-				 _lock "cmpxchg8b %a[ptr]", X86_FEATURE_CX8) \
-		     : ALT_OUTPUT_SP("+a" (o.low), "+d" (o.high))	\
-		     : "b" (n.low), "c" (n.high), [ptr] "S" (_ptr)	\
-		     : "memory");					\
+	asm_inline volatile(						\
+		ALTERNATIVE(_lock_loc					\
+			    "call cmpxchg8b_emu",			\
+			    _lock "cmpxchg8b %a[ptr]", X86_FEATURE_CX8)	\
+		: ALT_OUTPUT_SP("+a" (o.low), "+d" (o.high))		\
+		: "b" (n.low), "c" (n.high),				\
+		  [ptr] "S" (_ptr)					\
+		: "memory");						\
 									\
 	o.full;								\
 })
 
 static __always_inline u64 arch_cmpxchg64(volatile u64 *ptr, u64 old, u64 new)
 {
-	return __arch_cmpxchg64_emu(ptr, old, new, LOCK_PREFIX_HERE, "lock; ");
+	return __arch_cmpxchg64_emu(ptr, old, new, LOCK_PREFIX_HERE, "lock ");
 }
 #define arch_cmpxchg64 arch_cmpxchg64
 
@@ -119,14 +121,16 @@ static __always_inline u64 arch_cmpxchg64_local(volatile u64 *ptr, u64 old, u64
 			   n = { .full = (_new), };			\
 	bool ret;							\
 									\
-	asm volatile(ALTERNATIVE(_lock_loc				\
-				 "call cmpxchg8b_emu",			\
-				 _lock "cmpxchg8b %a[ptr]", X86_FEATURE_CX8) \
-		     CC_SET(e)						\
-		     : ALT_OUTPUT_SP(CC_OUT(e) (ret),			\
-				     "+a" (o.low), "+d" (o.high))	\
-		     : "b" (n.low), "c" (n.high), [ptr] "S" (_ptr)	\
-		     : "memory");					\
+	asm_inline volatile(						\
+		ALTERNATIVE(_lock_loc					\
+			    "call cmpxchg8b_emu",			\
+			    _lock "cmpxchg8b %a[ptr]", X86_FEATURE_CX8) \
+		CC_SET(e)						\
+		: ALT_OUTPUT_SP(CC_OUT(e) (ret),			\
+				"+a" (o.low), "+d" (o.high))		\
+		: "b" (n.low), "c" (n.high),				\
+		  [ptr] "S" (_ptr)					\
+		: "memory");						\
 									\
 	if (unlikely(!ret))						\
 		*(_oldp) = o.full;					\
@@ -136,7 +140,7 @@ static __always_inline u64 arch_cmpxchg64_local(volatile u64 *ptr, u64 old, u64
 
 static __always_inline bool arch_try_cmpxchg64(volatile u64 *ptr, u64 *oldp, u64 new)
 {
-	return __arch_try_cmpxchg64_emu(ptr, oldp, new, LOCK_PREFIX_HERE, "lock; ");
+	return __arch_try_cmpxchg64_emu(ptr, oldp, new, LOCK_PREFIX_HERE, "lock ");
 }
 #define arch_try_cmpxchg64 arch_try_cmpxchg64
 

arch/x86/include/asm/edac.h

@@ -13,7 +13,7 @@ static inline void edac_atomic_scrub(void *va, u32 size)
 	 * are interrupt, DMA and SMP safe.
 	 */
 	for (i = 0; i < size / 4; i++, virt_addr++)
-		asm volatile("lock; addl $0, %0"::"m" (*virt_addr));
+		asm volatile("lock addl $0, %0"::"m" (*virt_addr));
 }
 
 #endif /* _ASM_X86_EDAC_H */

arch/x86/include/asm/percpu.h

@@ -348,15 +348,14 @@ do {									\
 	old__.var = _oval;						\
 	new__.var = _nval;						\
 									\
-	asm qual (ALTERNATIVE("call this_cpu_cmpxchg8b_emu",		\
-			      "cmpxchg8b " __percpu_arg([var]), X86_FEATURE_CX8) \
-		  : [var] "+m" (__my_cpu_var(_var)),			\
-		    "+a" (old__.low),					\
-		    "+d" (old__.high)					\
-		  : "b" (new__.low),					\
-		    "c" (new__.high),					\
-		    "S" (&(_var))					\
-		  : "memory");						\
+	asm_inline qual (						\
+		ALTERNATIVE("call this_cpu_cmpxchg8b_emu",		\
+			    "cmpxchg8b " __percpu_arg([var]), X86_FEATURE_CX8) \
+		: ALT_OUTPUT_SP([var] "+m" (__my_cpu_var(_var)),	\
+				"+a" (old__.low), "+d" (old__.high))	\
+		: "b" (new__.low), "c" (new__.high),			\
+		  "S" (&(_var))						\
+		: "memory");						\
 									\
 	old__.var;							\
 })
@@ -378,17 +377,16 @@ do {									\
 	old__.var = *_oval;						\
 	new__.var = _nval;						\
 									\
-	asm qual (ALTERNATIVE("call this_cpu_cmpxchg8b_emu",		\
-			      "cmpxchg8b " __percpu_arg([var]), X86_FEATURE_CX8) \
-		  CC_SET(z)						\
-		  : CC_OUT(z) (success),				\
-		    [var] "+m" (__my_cpu_var(_var)),			\
-		    "+a" (old__.low),					\
-		    "+d" (old__.high)					\
-		  : "b" (new__.low),					\
-		    "c" (new__.high),					\
-		    "S" (&(_var))					\
-		  : "memory");						\
+	asm_inline qual (						\
+		ALTERNATIVE("call this_cpu_cmpxchg8b_emu",		\
+			    "cmpxchg8b " __percpu_arg([var]), X86_FEATURE_CX8) \
+		CC_SET(z)						\
+		: ALT_OUTPUT_SP(CC_OUT(z) (success),			\
+				[var] "+m" (__my_cpu_var(_var)),	\
+				"+a" (old__.low), "+d" (old__.high))	\
+		: "b" (new__.low), "c" (new__.high),			\
+		  "S" (&(_var))						\
+		: "memory");						\
 	if (unlikely(!success))						\
 		*_oval = old__.var;					\
 									\
@@ -419,15 +417,14 @@ do {									\
 	old__.var = _oval;						\
 	new__.var = _nval;						\
 									\
-	asm qual (ALTERNATIVE("call this_cpu_cmpxchg16b_emu",		\
-			      "cmpxchg16b " __percpu_arg([var]), X86_FEATURE_CX16) \
-		  : [var] "+m" (__my_cpu_var(_var)),			\
-		    "+a" (old__.low),					\
-		    "+d" (old__.high)					\
-		  : "b" (new__.low),					\
-		    "c" (new__.high),					\
-		    "S" (&(_var))					\
-		  : "memory");						\
+	asm_inline qual (						\
+		ALTERNATIVE("call this_cpu_cmpxchg16b_emu",		\
+			    "cmpxchg16b " __percpu_arg([var]), X86_FEATURE_CX16) \
+		: ALT_OUTPUT_SP([var] "+m" (__my_cpu_var(_var)),	\
+				"+a" (old__.low), "+d" (old__.high))	\
+		: "b" (new__.low), "c" (new__.high),			\
+		  "S" (&(_var))						\
+		: "memory");						\
 									\
 	old__.var;							\
 })
@@ -449,19 +446,19 @@ do {									\
 	old__.var = *_oval;						\
 	new__.var = _nval;						\
 									\
-	asm qual (ALTERNATIVE("call this_cpu_cmpxchg16b_emu",		\
-			      "cmpxchg16b " __percpu_arg([var]), X86_FEATURE_CX16) \
-		  CC_SET(z)						\
-		  : CC_OUT(z) (success),				\
-		    [var] "+m" (__my_cpu_var(_var)),			\
-		    "+a" (old__.low),					\
-		    "+d" (old__.high)					\
-		  : "b" (new__.low),					\
-		    "c" (new__.high),					\
-		    "S" (&(_var))					\
-		  : "memory");						\
+	asm_inline qual (						\
+		ALTERNATIVE("call this_cpu_cmpxchg16b_emu",		\
+			    "cmpxchg16b " __percpu_arg([var]), X86_FEATURE_CX16) \
+		CC_SET(z)						\
+		: ALT_OUTPUT_SP(CC_OUT(z) (success),			\
+				[var] "+m" (__my_cpu_var(_var)),	\
+				"+a" (old__.low), "+d" (old__.high))	\
+		: "b" (new__.low), "c" (new__.high),			\
+		  "S" (&(_var))						\
+		: "memory");						\
 	if (unlikely(!success))						\
 		*_oval = old__.var;					\
+									\
 	likely(success);						\
 })
 

arch/x86/include/asm/sync_bitops.h

@@ -31,7 +31,7 @@
  */
 static inline void sync_set_bit(long nr, volatile unsigned long *addr)
 {
-	asm volatile("lock; " __ASM_SIZE(bts) " %1,%0"
+	asm volatile("lock " __ASM_SIZE(bts) " %1,%0"
 		     : "+m" (ADDR)
 		     : "Ir" (nr)
 		     : "memory");
@@ -49,7 +49,7 @@ static inline void sync_set_bit(long nr, volatile unsigned long *addr)
  */
 static inline void sync_clear_bit(long nr, volatile unsigned long *addr)
 {
-	asm volatile("lock; " __ASM_SIZE(btr) " %1,%0"
+	asm volatile("lock " __ASM_SIZE(btr) " %1,%0"
 		     : "+m" (ADDR)
 		     : "Ir" (nr)
 		     : "memory");
@@ -66,7 +66,7 @@ static inline void sync_clear_bit(long nr, volatile unsigned long *addr)
  */
 static inline void sync_change_bit(long nr, volatile unsigned long *addr)
 {
-	asm volatile("lock; " __ASM_SIZE(btc) " %1,%0"
+	asm volatile("lock " __ASM_SIZE(btc) " %1,%0"
 		     : "+m" (ADDR)
 		     : "Ir" (nr)
 		     : "memory");
@@ -82,7 +82,7 @@ static inline void sync_change_bit(long nr, volatile unsigned long *addr)
  */
 static inline bool sync_test_and_set_bit(long nr, volatile unsigned long *addr)
 {
-	return GEN_BINARY_RMWcc("lock; " __ASM_SIZE(bts), *addr, c, "Ir", nr);
+	return GEN_BINARY_RMWcc("lock " __ASM_SIZE(bts), *addr, c, "Ir", nr);
 }
 
 /**
@@ -95,7 +95,7 @@ static inline bool sync_test_and_set_bit(long nr, volatile unsigned long *addr)
  */
 static inline int sync_test_and_clear_bit(long nr, volatile unsigned long *addr)
 {
-	return GEN_BINARY_RMWcc("lock; " __ASM_SIZE(btr), *addr, c, "Ir", nr);
+	return GEN_BINARY_RMWcc("lock " __ASM_SIZE(btr), *addr, c, "Ir", nr);
 }
 
 /**
@@ -108,7 +108,7 @@ static inline int sync_test_and_clear_bit(long nr, volatile unsigned long *addr)
  */
 static inline int sync_test_and_change_bit(long nr, volatile unsigned long *addr)
 {
-	return GEN_BINARY_RMWcc("lock; " __ASM_SIZE(btc), *addr, c, "Ir", nr);
+	return GEN_BINARY_RMWcc("lock " __ASM_SIZE(btc), *addr, c, "Ir", nr);
 }
 
 #define sync_test_bit(nr, addr) test_bit(nr, addr)

arch/x86/kernel/cpu/bus_lock.c

@@ -192,7 +192,13 @@ static void __split_lock_reenable(struct work_struct *work)
 {
 	sld_update_msr(true);
 }
-static DECLARE_DELAYED_WORK(sl_reenable, __split_lock_reenable);
+/*
+ * In order for each CPU to schedule its delayed work independently of the
+ * others, delayed work struct must be per-CPU. This is not required when
+ * sysctl_sld_mitigate is enabled because of the semaphore that limits
+ * the number of simultaneously scheduled delayed works to 1.
+ */
+static DEFINE_PER_CPU(struct delayed_work, sl_reenable);
 
 /*
  * If a CPU goes offline with pending delayed work to re-enable split lock
@@ -213,7 +219,7 @@ static int splitlock_cpu_offline(unsigned int cpu)
 
 static void split_lock_warn(unsigned long ip)
 {
-	struct delayed_work *work;
+	struct delayed_work *work = NULL;
 	int cpu;
 
 	if (!current->reported_split_lock)
@@ -235,11 +241,17 @@ static void split_lock_warn(unsigned long ip)
 		if (down_interruptible(&buslock_sem) == -EINTR)
 			return;
 		work = &sl_reenable_unlock;
-	} else {
-		work = &sl_reenable;
 	}
 
 	cpu = get_cpu();
+
+	if (!work) {
+		work = this_cpu_ptr(&sl_reenable);
+		/* Deferred initialization of per-CPU struct */
+		if (!work->work.func)
+			INIT_DELAYED_WORK(work, __split_lock_reenable);
+	}
+
 	schedule_delayed_work_on(cpu, work, 2);
 
 	/* Disable split lock detection on this CPU to make progress */

include/linux/interrupt.h

@@ -448,38 +448,30 @@ irq_calc_affinity_vectors(unsigned int minvec, unsigned int maxvec,
 static inline void disable_irq_nosync_lockdep(unsigned int irq)
 {
 	disable_irq_nosync(irq);
-#ifdef CONFIG_LOCKDEP
+#if defined(CONFIG_LOCKDEP) && !defined(CONFIG_PREEMPT_RT)
 	local_irq_disable();
 #endif
 }
 
 static inline void disable_irq_nosync_lockdep_irqsave(unsigned int irq, unsigned long *flags)
 {
 	disable_irq_nosync(irq);
-#ifdef CONFIG_LOCKDEP
+#if defined(CONFIG_LOCKDEP) && !defined(CONFIG_PREEMPT_RT)
 	local_irq_save(*flags);
 #endif
 }
 
-static inline void disable_irq_lockdep(unsigned int irq)
-{
-	disable_irq(irq);
-#ifdef CONFIG_LOCKDEP
-	local_irq_disable();
-#endif
-}
-
 static inline void enable_irq_lockdep(unsigned int irq)
 {
-#ifdef CONFIG_LOCKDEP
+#if defined(CONFIG_LOCKDEP) && !defined(CONFIG_PREEMPT_RT)
 	local_irq_enable();
 #endif
 	enable_irq(irq);
 }
 
 static inline void enable_irq_lockdep_irqrestore(unsigned int irq, unsigned long *flags)
 {
-#ifdef CONFIG_LOCKDEP
+#if defined(CONFIG_LOCKDEP) && !defined(CONFIG_PREEMPT_RT)
 	local_irq_restore(*flags);
 #endif
 	enable_irq(irq);

kernel/locking/Makefile

@@ -5,7 +5,8 @@ KCOV_INSTRUMENT		:= n
 
 obj-y += mutex.o semaphore.o rwsem.o percpu-rwsem.o
 
-# Avoid recursion lockdep -> sanitizer -> ... -> lockdep.
+# Avoid recursion lockdep -> sanitizer -> ... -> lockdep & improve performance.
+KASAN_SANITIZE_lockdep.o := n
 KCSAN_SANITIZE_lockdep.o := n
 
 ifdef CONFIG_FUNCTION_TRACER