Merge tag 'locking-core-2024-01-08' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull locking updates from Ingo Molar:
 "Lock guards:

   - Use lock guards in the ptrace code

   - Introduce conditional guards to extend to conditional lock
     primitives like mutex_trylock()/mutex_lock_interruptible()/etc.

  lockdep:

   - Optimize 'struct lock_class' to be smaller

   - Update file patterns in MAINTAINERS

  mutexes:

   - Document mutex lifetime rules a bit more"

* tag 'locking-core-2024-01-08' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  locking/mutex: Clarify that mutex_unlock(), and most other sleeping locks, can still use the lock object after it's unlocked
  locking/mutex: Document that mutex_unlock() is non-atomic
  ptrace: Convert ptrace_attach() to use lock guards
  locking/lockdep: Slightly reorder 'struct lock_class' to save some memory
  MAINTAINERS: Add include/linux/lockdep*.h
  cleanup: Add conditional guard support

Documentation/locking/mutex-design.rst

@@ -101,6 +101,24 @@ features that make lock debugging easier and faster:
     - Detects multi-task circular deadlocks and prints out all affected
       locks and tasks (and only those tasks).
 
+Mutexes - and most other sleeping locks like rwsems - do not provide an
+implicit reference for the memory they occupy, which reference is released
+with mutex_unlock().
+
+[ This is in contrast with spin_unlock() [or completion_done()], which
+  APIs can be used to guarantee that the memory is not touched by the
+  lock implementation after spin_unlock()/completion_done() releases
+  the lock. ]
+
+mutex_unlock() may access the mutex structure even after it has internally
+released the lock already - so it's not safe for another context to
+acquire the mutex and assume that the mutex_unlock() context is not using
+the structure anymore.
+
+The mutex user must ensure that the mutex is not destroyed while a
+release operation is still in progress - in other words, callers of
+mutex_unlock() must ensure that the mutex stays alive until mutex_unlock()
+has returned.
 
 Interfaces
 ----------

MAINTAINERS

@@ -12424,7 +12424,7 @@ S:	Maintained
 T:	git git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git locking/core
 F:	Documentation/locking/
 F:	arch/*/include/asm/spinlock*.h
-F:	include/linux/lockdep.h
+F:	include/linux/lockdep*.h
 F:	include/linux/mutex*.h
 F:	include/linux/rwlock*.h
 F:	include/linux/rwsem*.h

include/linux/cleanup.h

@@ -125,25 +125,55 @@ static inline class_##_name##_t class_##_name##ext##_constructor(_init_args) \
  *	trivial wrapper around DEFINE_CLASS() above specifically
  *	for locks.
  *
+ * DEFINE_GUARD_COND(name, ext, condlock)
+ *	wrapper around EXTEND_CLASS above to add conditional lock
+ *	variants to a base class, eg. mutex_trylock() or
+ *	mutex_lock_interruptible().
+ *
  * guard(name):
- *	an anonymous instance of the (guard) class
+ *	an anonymous instance of the (guard) class, not recommended for
+ *	conditional locks.
  *
  * scoped_guard (name, args...) { }:
  *	similar to CLASS(name, scope)(args), except the variable (with the
  *	explicit name 'scope') is declard in a for-loop such that its scope is
  *	bound to the next (compound) statement.
  *
+ *	for conditional locks the loop body is skipped when the lock is not
+ *	acquired.
+ *
+ * scoped_cond_guard (name, fail, args...) { }:
+ *      similar to scoped_guard(), except it does fail when the lock
+ *      acquire fails.
+ *
  */
 
 #define DEFINE_GUARD(_name, _type, _lock, _unlock) \
-	DEFINE_CLASS(_name, _type, _unlock, ({ _lock; _T; }), _type _T)
+	DEFINE_CLASS(_name, _type, if (_T) { _unlock; }, ({ _lock; _T; }), _type _T); \
+	static inline void * class_##_name##_lock_ptr(class_##_name##_t *_T) \
+	{ return *_T; }
+
+#define DEFINE_GUARD_COND(_name, _ext, _condlock) \
+	EXTEND_CLASS(_name, _ext, \
+		     ({ void *_t = _T; if (_T && !(_condlock)) _t = NULL; _t; }), \
+		     class_##_name##_t _T) \
+	static inline void * class_##_name##_ext##_lock_ptr(class_##_name##_t *_T) \
+	{ return class_##_name##_lock_ptr(_T); }
 
 #define guard(_name) \
 	CLASS(_name, __UNIQUE_ID(guard))
 
+#define __guard_ptr(_name) class_##_name##_lock_ptr
+
 #define scoped_guard(_name, args...)					\
 	for (CLASS(_name, scope)(args),					\
-	     *done = NULL; !done; done = (void *)1)
+	     *done = NULL; __guard_ptr(_name)(&scope) && !done; done = (void *)1)
+
+#define scoped_cond_guard(_name, _fail, args...) \
+	for (CLASS(_name, scope)(args), \
+	     *done = NULL; !done; done = (void *)1) \
+		if (!__guard_ptr(_name)(&scope)) _fail; \
+		else
 
 /*
  * Additional helper macros for generating lock guards with types, either for
@@ -152,6 +182,7 @@ static inline class_##_name##_t class_##_name##ext##_constructor(_init_args) \
  *
  * DEFINE_LOCK_GUARD_0(name, lock, unlock, ...)
  * DEFINE_LOCK_GUARD_1(name, type, lock, unlock, ...)
+ * DEFINE_LOCK_GUARD_1_COND(name, ext, condlock)
  *
  * will result in the following type:
  *
@@ -173,6 +204,11 @@ typedef struct {							\
 static inline void class_##_name##_destructor(class_##_name##_t *_T)	\
 {									\
 	if (_T->lock) { _unlock; }					\
+}									\
+									\
+static inline void *class_##_name##_lock_ptr(class_##_name##_t *_T)	\
+{									\
+	return _T->lock;						\
 }
 
 
@@ -201,4 +237,14 @@ __DEFINE_LOCK_GUARD_1(_name, _type, _lock)
 __DEFINE_UNLOCK_GUARD(_name, void, _unlock, __VA_ARGS__)		\
 __DEFINE_LOCK_GUARD_0(_name, _lock)
 
+#define DEFINE_LOCK_GUARD_1_COND(_name, _ext, _condlock)		\
+	EXTEND_CLASS(_name, _ext,					\
+		     ({ class_##_name##_t _t = { .lock = l }, *_T = &_t;\
+		        if (_T->lock && !(_condlock)) _T->lock = NULL;	\
+			_t; }),						\
+		     typeof_member(class_##_name##_t, lock) l)		\
+	static inline void * class_##_name##_ext##_lock_ptr(class_##_name##_t *_T) \
+	{ return class_##_name##_lock_ptr(_T); }
+
+
 #endif /* __LINUX_GUARDS_H */

include/linux/lockdep_types.h

@@ -127,12 +127,12 @@ struct lock_class {
 	unsigned long			usage_mask;
 	const struct lock_trace		*usage_traces[LOCK_TRACE_STATES];
 
+	const char			*name;
 	/*
 	 * Generation counter, when doing certain classes of graph walking,
 	 * to ensure that we check one node only once:
 	 */
 	int				name_version;
-	const char			*name;
 
 	u8				wait_type_inner;
 	u8				wait_type_outer;

include/linux/mutex.h

@@ -221,6 +221,7 @@ extern void mutex_unlock(struct mutex *lock);
 extern int atomic_dec_and_mutex_lock(atomic_t *cnt, struct mutex *lock);
 
 DEFINE_GUARD(mutex, struct mutex *, mutex_lock(_T), mutex_unlock(_T))
-DEFINE_FREE(mutex, struct mutex *, if (_T) mutex_unlock(_T))
+DEFINE_GUARD_COND(mutex, _try, mutex_trylock(_T))
+DEFINE_GUARD_COND(mutex, _intr, mutex_lock_interruptible(_T) == 0)
 
 #endif /* __LINUX_MUTEX_H */

include/linux/rwsem.h

@@ -203,11 +203,11 @@ extern void up_read(struct rw_semaphore *sem);
 extern void up_write(struct rw_semaphore *sem);
 
 DEFINE_GUARD(rwsem_read, struct rw_semaphore *, down_read(_T), up_read(_T))
-DEFINE_GUARD(rwsem_write, struct rw_semaphore *, down_write(_T), up_write(_T))
-
-DEFINE_FREE(up_read, struct rw_semaphore *, if (_T) up_read(_T))
-DEFINE_FREE(up_write, struct rw_semaphore *, if (_T) up_write(_T))
+DEFINE_GUARD_COND(rwsem_read, _try, down_read_trylock(_T))
+DEFINE_GUARD_COND(rwsem_read, _intr, down_read_interruptible(_T) == 0)
 
+DEFINE_GUARD(rwsem_write, struct rw_semaphore *, down_write(_T), up_write(_T))
+DEFINE_GUARD_COND(rwsem_write, _try, down_write_trylock(_T))
 
 /*
  * downgrade write lock to read lock

include/linux/sched/task.h

@@ -226,4 +226,6 @@ static inline void task_unlock(struct task_struct *p)
 	spin_unlock(&p->alloc_lock);
 }
 
+DEFINE_GUARD(task_lock, struct task_struct *, task_lock(_T), task_unlock(_T))
+
 #endif /* _LINUX_SCHED_TASK_H */

include/linux/spinlock.h

@@ -507,6 +507,8 @@ DEFINE_LOCK_GUARD_1(raw_spinlock, raw_spinlock_t,
 		    raw_spin_lock(_T->lock),
 		    raw_spin_unlock(_T->lock))
 
+DEFINE_LOCK_GUARD_1_COND(raw_spinlock, _try, raw_spin_trylock(_T->lock))
+
 DEFINE_LOCK_GUARD_1(raw_spinlock_nested, raw_spinlock_t,
 		    raw_spin_lock_nested(_T->lock, SINGLE_DEPTH_NESTING),
 		    raw_spin_unlock(_T->lock))
@@ -515,23 +517,62 @@ DEFINE_LOCK_GUARD_1(raw_spinlock_irq, raw_spinlock_t,
 		    raw_spin_lock_irq(_T->lock),
 		    raw_spin_unlock_irq(_T->lock))
 
+DEFINE_LOCK_GUARD_1_COND(raw_spinlock_irq, _try, raw_spin_trylock_irq(_T->lock))
+
 DEFINE_LOCK_GUARD_1(raw_spinlock_irqsave, raw_spinlock_t,
 		    raw_spin_lock_irqsave(_T->lock, _T->flags),
 		    raw_spin_unlock_irqrestore(_T->lock, _T->flags),
 		    unsigned long flags)
 
+DEFINE_LOCK_GUARD_1_COND(raw_spinlock_irqsave, _try,
+			 raw_spin_trylock_irqsave(_T->lock, _T->flags))
+
 DEFINE_LOCK_GUARD_1(spinlock, spinlock_t,
 		    spin_lock(_T->lock),
 		    spin_unlock(_T->lock))
 
+DEFINE_LOCK_GUARD_1_COND(spinlock, _try, spin_trylock(_T->lock))
+
 DEFINE_LOCK_GUARD_1(spinlock_irq, spinlock_t,
 		    spin_lock_irq(_T->lock),
 		    spin_unlock_irq(_T->lock))
 
+DEFINE_LOCK_GUARD_1_COND(spinlock_irq, _try,
+			 spin_trylock_irq(_T->lock))
+
 DEFINE_LOCK_GUARD_1(spinlock_irqsave, spinlock_t,
 		    spin_lock_irqsave(_T->lock, _T->flags),
 		    spin_unlock_irqrestore(_T->lock, _T->flags),
 		    unsigned long flags)
 
+DEFINE_LOCK_GUARD_1_COND(spinlock_irqsave, _try,
+			 spin_trylock_irqsave(_T->lock, _T->flags))
+
+DEFINE_LOCK_GUARD_1(read_lock, rwlock_t,
+		    read_lock(_T->lock),
+		    read_unlock(_T->lock))
+
+DEFINE_LOCK_GUARD_1(read_lock_irq, rwlock_t,
+		    read_lock_irq(_T->lock),
+		    read_unlock_irq(_T->lock))
+
+DEFINE_LOCK_GUARD_1(read_lock_irqsave, rwlock_t,
+		    read_lock_irqsave(_T->lock, _T->flags),
+		    read_unlock_irqrestore(_T->lock, _T->flags),
+		    unsigned long flags)
+
+DEFINE_LOCK_GUARD_1(write_lock, rwlock_t,
+		    write_lock(_T->lock),
+		    write_unlock(_T->lock))
+
+DEFINE_LOCK_GUARD_1(write_lock_irq, rwlock_t,
+		    write_lock_irq(_T->lock),
+		    write_unlock_irq(_T->lock))
+
+DEFINE_LOCK_GUARD_1(write_lock_irqsave, rwlock_t,
+		    write_lock_irqsave(_T->lock, _T->flags),
+		    write_unlock_irqrestore(_T->lock, _T->flags),
+		    unsigned long flags)
+
 #undef __LINUX_INSIDE_SPINLOCK_H
 #endif /* __LINUX_SPINLOCK_H */

kernel/locking/mutex.c

@@ -532,6 +532,11 @@ static noinline void __sched __mutex_unlock_slowpath(struct mutex *lock, unsigne
  * This function must not be used in interrupt context. Unlocking
  * of a not locked mutex is not allowed.
  *
+ * The caller must ensure that the mutex stays alive until this function has
+ * returned - mutex_unlock() can NOT directly be used to release an object such
+ * that another concurrent task can free it.
+ * Mutexes are different from spinlocks & refcounts in this aspect.
+ *
  * This function is similar to (but not equivalent to) up().
  */
 void __sched mutex_unlock(struct mutex *lock)

kernel/ptrace.c

@@ -386,6 +386,34 @@ static int check_ptrace_options(unsigned long data)
 	return 0;
 }
 
+static inline void ptrace_set_stopped(struct task_struct *task)
+{
+	guard(spinlock)(&task->sighand->siglock);
+
+	/*
+	 * If the task is already STOPPED, set JOBCTL_TRAP_STOP and
+	 * TRAPPING, and kick it so that it transits to TRACED.  TRAPPING
+	 * will be cleared if the child completes the transition or any
+	 * event which clears the group stop states happens.  We'll wait
+	 * for the transition to complete before returning from this
+	 * function.
+	 *
+	 * This hides STOPPED -> RUNNING -> TRACED transition from the
+	 * attaching thread but a different thread in the same group can
+	 * still observe the transient RUNNING state.  IOW, if another
+	 * thread's WNOHANG wait(2) on the stopped tracee races against
+	 * ATTACH, the wait(2) may fail due to the transient RUNNING.
+	 *
+	 * The following task_is_stopped() test is safe as both transitions
+	 * in and out of STOPPED are protected by siglock.
+	 */
+	if (task_is_stopped(task) &&
+	    task_set_jobctl_pending(task, JOBCTL_TRAP_STOP | JOBCTL_TRAPPING)) {
+		task->jobctl &= ~JOBCTL_STOPPED;
+		signal_wake_up_state(task, __TASK_STOPPED);
+	}
+}
+
 static int ptrace_attach(struct task_struct *task, long request,
 			 unsigned long addr,
 			 unsigned long flags)
@@ -393,17 +421,17 @@ static int ptrace_attach(struct task_struct *task, long request,
 	bool seize = (request == PTRACE_SEIZE);
 	int retval;
 
-	retval = -EIO;
 	if (seize) {
 		if (addr != 0)
-			goto out;
+			return -EIO;
 		/*
 		 * This duplicates the check in check_ptrace_options() because
 		 * ptrace_attach() and ptrace_setoptions() have historically
 		 * used different error codes for unknown ptrace options.
 		 */
 		if (flags & ~(unsigned long)PTRACE_O_MASK)
-			goto out;
+			return -EIO;
+
 		retval = check_ptrace_options(flags);
 		if (retval)
 			return retval;
@@ -414,88 +442,54 @@ static int ptrace_attach(struct task_struct *task, long request,
 
 	audit_ptrace(task);
 
-	retval = -EPERM;
 	if (unlikely(task->flags & PF_KTHREAD))
-		goto out;
+		return -EPERM;
 	if (same_thread_group(task, current))
-		goto out;
+		return -EPERM;
 
 	/*
 	 * Protect exec's credential calculations against our interference;
 	 * SUID, SGID and LSM creds get determined differently
 	 * under ptrace.
 	 */
-	retval = -ERESTARTNOINTR;
-	if (mutex_lock_interruptible(&task->signal->cred_guard_mutex))
-		goto out;
+	scoped_cond_guard (mutex_intr, return -ERESTARTNOINTR,
+			   &task->signal->cred_guard_mutex) {
 
-	task_lock(task);
-	retval = __ptrace_may_access(task, PTRACE_MODE_ATTACH_REALCREDS);
-	task_unlock(task);
-	if (retval)
-		goto unlock_creds;
+		scoped_guard (task_lock, task) {
+			retval = __ptrace_may_access(task, PTRACE_MODE_ATTACH_REALCREDS);
+			if (retval)
+				return retval;
+		}
 
-	write_lock_irq(&tasklist_lock);
-	retval = -EPERM;
-	if (unlikely(task->exit_state))
-		goto unlock_tasklist;
-	if (task->ptrace)
-		goto unlock_tasklist;
+		scoped_guard (write_lock_irq, &tasklist_lock) {
+			if (unlikely(task->exit_state))
+				return -EPERM;
+			if (task->ptrace)
+				return -EPERM;
 
-	task->ptrace = flags;
+			task->ptrace = flags;
 
-	ptrace_link(task, current);
+			ptrace_link(task, current);
 
-	/* SEIZE doesn't trap tracee on attach */
-	if (!seize)
-		send_sig_info(SIGSTOP, SEND_SIG_PRIV, task);
+			/* SEIZE doesn't trap tracee on attach */
+			if (!seize)
+				send_sig_info(SIGSTOP, SEND_SIG_PRIV, task);
 
-	spin_lock(&task->sighand->siglock);
+			ptrace_set_stopped(task);
+		}
+	}
 
 	/*
-	 * If the task is already STOPPED, set JOBCTL_TRAP_STOP and
-	 * TRAPPING, and kick it so that it transits to TRACED.  TRAPPING
-	 * will be cleared if the child completes the transition or any
-	 * event which clears the group stop states happens.  We'll wait
-	 * for the transition to complete before returning from this
-	 * function.
-	 *
-	 * This hides STOPPED -> RUNNING -> TRACED transition from the
-	 * attaching thread but a different thread in the same group can
-	 * still observe the transient RUNNING state.  IOW, if another
-	 * thread's WNOHANG wait(2) on the stopped tracee races against
-	 * ATTACH, the wait(2) may fail due to the transient RUNNING.
-	 *
-	 * The following task_is_stopped() test is safe as both transitions
-	 * in and out of STOPPED are protected by siglock.
+	 * We do not bother to change retval or clear JOBCTL_TRAPPING
+	 * if wait_on_bit() was interrupted by SIGKILL. The tracer will
+	 * not return to user-mode, it will exit and clear this bit in
+	 * __ptrace_unlink() if it wasn't already cleared by the tracee;
+	 * and until then nobody can ptrace this task.
 	 */
-	if (task_is_stopped(task) &&
-	    task_set_jobctl_pending(task, JOBCTL_TRAP_STOP | JOBCTL_TRAPPING)) {
-		task->jobctl &= ~JOBCTL_STOPPED;
-		signal_wake_up_state(task, __TASK_STOPPED);
-	}
-
-	spin_unlock(&task->sighand->siglock);
-
-	retval = 0;
-unlock_tasklist:
-	write_unlock_irq(&tasklist_lock);
-unlock_creds:
-	mutex_unlock(&task->signal->cred_guard_mutex);
-out:
-	if (!retval) {
-		/*
-		 * We do not bother to change retval or clear JOBCTL_TRAPPING
-		 * if wait_on_bit() was interrupted by SIGKILL. The tracer will
-		 * not return to user-mode, it will exit and clear this bit in
-		 * __ptrace_unlink() if it wasn't already cleared by the tracee;
-		 * and until then nobody can ptrace this task.
-		 */
-		wait_on_bit(&task->jobctl, JOBCTL_TRAPPING_BIT, TASK_KILLABLE);
-		proc_ptrace_connector(task, PTRACE_ATTACH);
-	}
+	wait_on_bit(&task->jobctl, JOBCTL_TRAPPING_BIT, TASK_KILLABLE);
+	proc_ptrace_connector(task, PTRACE_ATTACH);
 
-	return retval;
+	return 0;
 }
 
 /**