8 files changed, 121 insertions, 71 deletions

diff --git a/kernel/bpf/bpf_lsm.c b/kernel/bpf/bpf_lsm.c
index 9ea42a45da47..a4a41ee3e80b 100644
--- a/kernel/bpf/bpf_lsm.c
+++ b/kernel/bpf/bpf_lsm.c
@@ -351,8 +351,10 @@ BTF_ID(func, bpf_lsm_bpf_prog_alloc_security)
 BTF_ID(func, bpf_lsm_bpf_prog_free_security)
 BTF_ID(func, bpf_lsm_file_alloc_security)
 BTF_ID(func, bpf_lsm_file_free_security)
+#ifdef CONFIG_SECURITY_NETWORK
 BTF_ID(func, bpf_lsm_sk_alloc_security)
 BTF_ID(func, bpf_lsm_sk_free_security)
+#endif /* CONFIG_SECURITY_NETWORK */
 BTF_ID(func, bpf_lsm_task_free)
 BTF_SET_END(untrusted_lsm_hooks)
 
diff --git a/kernel/bpf/task_iter.c b/kernel/bpf/task_iter.c
index c2a2182ce570..c4ab9d6cdbe9 100644
--- a/kernel/bpf/task_iter.c
+++ b/kernel/bpf/task_iter.c
@@ -438,6 +438,7 @@ struct bpf_iter_seq_task_vma_info {
 	 */
 	struct bpf_iter_seq_task_common common;
 	struct task_struct *task;
+	struct mm_struct *mm;
 	struct vm_area_struct *vma;
 	u32 tid;
 	unsigned long prev_vm_start;
@@ -456,16 +457,19 @@ task_vma_seq_get_next(struct bpf_iter_seq_task_vma_info *info)
 	enum bpf_task_vma_iter_find_op op;
 	struct vm_area_struct *curr_vma;
 	struct task_struct *curr_task;
+	struct mm_struct *curr_mm;
 	u32 saved_tid = info->tid;
 
 	/* If this function returns a non-NULL vma, it holds a reference to
-	 * the task_struct, and holds read lock on vma->mm->mmap_lock.
+	 * the task_struct, holds a refcount on mm->mm_users, and holds
+	 * read lock on vma->mm->mmap_lock.
 	 * If this function returns NULL, it does not hold any reference or
 	 * lock.
 	 */
 	if (info->task) {
 		curr_task = info->task;
 		curr_vma = info->vma;
+		curr_mm = info->mm;
 		/* In case of lock contention, drop mmap_lock to unblock
 		 * the writer.
 		 *
@@ -504,13 +508,15 @@ task_vma_seq_get_next(struct bpf_iter_seq_task_vma_info *info)
 		 *    4.2) VMA2 and VMA2' covers different ranges, process
 		 *         VMA2'.
 		 */
-		if (mmap_lock_is_contended(curr_task->mm)) {
+		if (mmap_lock_is_contended(curr_mm)) {
 			info->prev_vm_start = curr_vma->vm_start;
 			info->prev_vm_end = curr_vma->vm_end;
 			op = task_vma_iter_find_vma;
-			mmap_read_unlock(curr_task->mm);
-			if (mmap_read_lock_killable(curr_task->mm))
+			mmap_read_unlock(curr_mm);
+			if (mmap_read_lock_killable(curr_mm)) {
+				mmput(curr_mm);
 				goto finish;
+			}
 		} else {
 			op = task_vma_iter_next_vma;
 		}
@@ -535,42 +541,47 @@ again:
 			op = task_vma_iter_find_vma;
 		}
 
-		if (!curr_task->mm)
+		curr_mm = get_task_mm(curr_task);
+		if (!curr_mm)
 			goto next_task;
 
-		if (mmap_read_lock_killable(curr_task->mm))
+		if (mmap_read_lock_killable(curr_mm)) {
+			mmput(curr_mm);
 			goto finish;
+		}
 	}
 
 	switch (op) {
 	case task_vma_iter_first_vma:
-		curr_vma = find_vma(curr_task->mm, 0);
+		curr_vma = find_vma(curr_mm, 0);
 		break;
 	case task_vma_iter_next_vma:
-		curr_vma = find_vma(curr_task->mm, curr_vma->vm_end);
+		curr_vma = find_vma(curr_mm, curr_vma->vm_end);
 		break;
 	case task_vma_iter_find_vma:
 		/* We dropped mmap_lock so it is necessary to use find_vma
 		 * to find the next vma. This is similar to the  mechanism
 		 * in show_smaps_rollup().
 		 */
-		curr_vma = find_vma(curr_task->mm, info->prev_vm_end - 1);
+		curr_vma = find_vma(curr_mm, info->prev_vm_end - 1);
 		/* case 1) and 4.2) above just use curr_vma */
 
 		/* check for case 2) or case 4.1) above */
 		if (curr_vma &&
 		    curr_vma->vm_start == info->prev_vm_start &&
 		    curr_vma->vm_end == info->prev_vm_end)
-			curr_vma = find_vma(curr_task->mm, curr_vma->vm_end);
+			curr_vma = find_vma(curr_mm, curr_vma->vm_end);
 		break;
 	}
 	if (!curr_vma) {
 		/* case 3) above, or case 2) 4.1) with vma->next == NULL */
-		mmap_read_unlock(curr_task->mm);
+		mmap_read_unlock(curr_mm);
+		mmput(curr_mm);
 		goto next_task;
 	}
 	info->task = curr_task;
 	info->vma = curr_vma;
+	info->mm = curr_mm;
 	return curr_vma;
 
 next_task:
@@ -579,6 +590,7 @@ next_task:
 
 	put_task_struct(curr_task);
 	info->task = NULL;
+	info->mm = NULL;
 	info->tid++;
 	goto again;
 
@@ -587,6 +599,7 @@ finish:
 		put_task_struct(curr_task);
 	info->task = NULL;
 	info->vma = NULL;
+	info->mm = NULL;
 	return NULL;
 }
 
@@ -658,7 +671,9 @@ static void task_vma_seq_stop(struct seq_file *seq, void *v)
 		 */
 		info->prev_vm_start = ~0UL;
 		info->prev_vm_end = info->vma->vm_end;
-		mmap_read_unlock(info->task->mm);
+		mmap_read_unlock(info->mm);
+		mmput(info->mm);
+		info->mm = NULL;
 		put_task_struct(info->task);
 		info->task = NULL;
 	}
diff --git a/kernel/bpf/trampoline.c b/kernel/bpf/trampoline.c
index 11f5ec0b8016..d0ed7d6f5eec 100644
--- a/kernel/bpf/trampoline.c
+++ b/kernel/bpf/trampoline.c
@@ -488,6 +488,10 @@ again:
 		/* reset fops->func and fops->trampoline for re-register */
 		tr->fops->func = NULL;
 		tr->fops->trampoline = 0;
+
+		/* reset im->image memory attr for arch_prepare_bpf_trampoline */
+		set_memory_nx((long)im->image, 1);
+		set_memory_rw((long)im->image, 1);
 		goto again;
 	}
 #endif
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index a5255a0dcbb6..85f96c1e9f62 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -1054,6 +1054,8 @@ static void print_insn_state(struct bpf_verifier_env *env,
  */
 static void *copy_array(void *dst, const void *src, size_t n, size_t size, gfp_t flags)
 {
+	size_t alloc_bytes;
+	void *orig = dst;
 	size_t bytes;
 
 	if (ZERO_OR_NULL_PTR(src))
@@ -1062,11 +1064,11 @@ static void *copy_array(void *dst, const void *src, size_t n, size_t size, gfp_t
 	if (unlikely(check_mul_overflow(n, size, &bytes)))
 		return NULL;
 
-	if (ksize(dst) < ksize(src)) {
-		kfree(dst);
-		dst = kmalloc_track_caller(kmalloc_size_roundup(bytes), flags);
-		if (!dst)
-			return NULL;
+	alloc_bytes = max(ksize(orig), kmalloc_size_roundup(bytes));
+	dst = krealloc(orig, alloc_bytes, flags);
+	if (!dst) {
+		kfree(orig);
+		return NULL;
 	}
 
 	memcpy(dst, src, bytes);
@@ -11822,10 +11824,17 @@ static int check_cond_jmp_op(struct bpf_verifier_env *env,
 	 *      register B - not null
 	 * for JNE A, B, ... - A is not null in the false branch;
 	 * for JEQ A, B, ... - A is not null in the true branch.
+	 *
+	 * Since PTR_TO_BTF_ID points to a kernel struct that does
+	 * not need to be null checked by the BPF program, i.e.,
+	 * could be null even without PTR_MAYBE_NULL marking, so
+	 * only propagate nullness when neither reg is that type.
 	 */
 	if (!is_jmp32 && BPF_SRC(insn->code) == BPF_X &&
 	    __is_pointer_value(false, src_reg) && __is_pointer_value(false, dst_reg) &&
-	    type_may_be_null(src_reg->type) != type_may_be_null(dst_reg->type)) {
+	    type_may_be_null(src_reg->type) != type_may_be_null(dst_reg->type) &&
+	    base_type(src_reg->type) != PTR_TO_BTF_ID &&
+	    base_type(dst_reg->type) != PTR_TO_BTF_ID) {
 		eq_branch_regs = NULL;
 		switch (opcode) {
 		case BPF_JEQ:
diff --git a/kernel/events/core.c b/kernel/events/core.c
index eacc3702654d..d56328e5080e 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -380,7 +380,6 @@ enum event_type_t {
 
 /*
  * perf_sched_events : >0 events exist
- * perf_cgroup_events: >0 per-cpu cgroup events exist on this cpu
  */
 
 static void perf_sched_delayed(struct work_struct *work);
@@ -389,7 +388,6 @@ static DECLARE_DELAYED_WORK(perf_sched_work, perf_sched_delayed);
 static DEFINE_MUTEX(perf_sched_mutex);
 static atomic_t perf_sched_count;
 
-static DEFINE_PER_CPU(atomic_t, perf_cgroup_events);
 static DEFINE_PER_CPU(struct pmu_event_list, pmu_sb_events);
 
 static atomic_t nr_mmap_events __read_mostly;
@@ -844,9 +842,16 @@ static void perf_cgroup_switch(struct task_struct *task)
 	struct perf_cpu_context *cpuctx = this_cpu_ptr(&perf_cpu_context);
 	struct perf_cgroup *cgrp;
 
-	cgrp = perf_cgroup_from_task(task, NULL);
+	/*
+	 * cpuctx->cgrp is set when the first cgroup event enabled,
+	 * and is cleared when the last cgroup event disabled.
+	 */
+	if (READ_ONCE(cpuctx->cgrp) == NULL)
+		return;
 
 	WARN_ON_ONCE(cpuctx->ctx.nr_cgroups == 0);
+
+	cgrp = perf_cgroup_from_task(task, NULL);
 	if (READ_ONCE(cpuctx->cgrp) == cgrp)
 		return;
 
@@ -3631,8 +3636,7 @@ void __perf_event_task_sched_out(struct task_struct *task,
 	 * to check if we have to switch out PMU state.
 	 * cgroup event are system-wide mode only
 	 */
-	if (atomic_read(this_cpu_ptr(&perf_cgroup_events)))
-		perf_cgroup_switch(next);
+	perf_cgroup_switch(next);
 }
 
 static bool perf_less_group_idx(const void *l, const void *r)
@@ -4974,15 +4978,6 @@ static void unaccount_pmu_sb_event(struct perf_event *event)
 		detach_sb_event(event);
 }
 
-static void unaccount_event_cpu(struct perf_event *event, int cpu)
-{
-	if (event->parent)
-		return;
-
-	if (is_cgroup_event(event))
-		atomic_dec(&per_cpu(perf_cgroup_events, cpu));
-}
-
 #ifdef CONFIG_NO_HZ_FULL
 static DEFINE_SPINLOCK(nr_freq_lock);
 #endif
@@ -5048,8 +5043,6 @@ static void unaccount_event(struct perf_event *event)
 			schedule_delayed_work(&perf_sched_work, HZ);
 	}
 
-	unaccount_event_cpu(event, event->cpu);
-
 	unaccount_pmu_sb_event(event);
 }
 
@@ -11679,15 +11672,6 @@ static void account_pmu_sb_event(struct perf_event *event)
 		attach_sb_event(event);
 }
 
-static void account_event_cpu(struct perf_event *event, int cpu)
-{
-	if (event->parent)
-		return;
-
-	if (is_cgroup_event(event))
-		atomic_inc(&per_cpu(perf_cgroup_events, cpu));
-}
-
 /* Freq events need the tick to stay alive (see perf_event_task_tick). */
 static void account_freq_event_nohz(void)
 {
@@ -11775,8 +11759,6 @@ static void account_event(struct perf_event *event)
 	}
 enabled:
 
-	account_event_cpu(event, event->cpu);
-
 	account_pmu_sb_event(event);
 }
 
@@ -12339,12 +12321,12 @@ SYSCALL_DEFINE5(perf_event_open,
 	if (flags & ~PERF_FLAG_ALL)
 		return -EINVAL;
 
-	/* Do we allow access to perf_event_open(2) ? */
-	err = security_perf_event_open(&attr, PERF_SECURITY_OPEN);
+	err = perf_copy_attr(attr_uptr, &attr);
 	if (err)
 		return err;
 
-	err = perf_copy_attr(attr_uptr, &attr);
+	/* Do we allow access to perf_event_open(2) ? */
+	err = security_perf_event_open(&attr, PERF_SECURITY_OPEN);
 	if (err)
 		return err;
 
@@ -12689,7 +12671,8 @@ SYSCALL_DEFINE5(perf_event_open,
 	return event_fd;
 
 err_context:
-	/* event->pmu_ctx freed by free_event() */
+	put_pmu_ctx(event->pmu_ctx);
+	event->pmu_ctx = NULL; /* _free_event() */
 err_locked:
 	mutex_unlock(&ctx->mutex);
 	perf_unpin_context(ctx);
@@ -12802,6 +12785,7 @@ perf_event_create_kernel_counter(struct perf_event_attr *attr, int cpu,
 
 err_pmu_ctx:
 	put_pmu_ctx(pmu_ctx);
+	event->pmu_ctx = NULL; /* _free_event() */
 err_unlock:
 	mutex_unlock(&ctx->mutex);
 	perf_unpin_context(ctx);
@@ -12822,13 +12806,11 @@ static void __perf_pmu_remove(struct perf_event_context *ctx,
 
 	perf_event_groups_for_cpu_pmu(event, groups, cpu, pmu) {
 		perf_remove_from_context(event, 0);
-		unaccount_event_cpu(event, cpu);
 		put_pmu_ctx(event->pmu_ctx);
 		list_add(&event->migrate_entry, events);
 
 		for_each_sibling_event(sibling, event) {
 			perf_remove_from_context(sibling, 0);
-			unaccount_event_cpu(sibling, cpu);
 			put_pmu_ctx(sibling->pmu_ctx);
 			list_add(&sibling->migrate_entry, events);
 		}
@@ -12847,7 +12829,6 @@ static void __perf_pmu_install_event(struct pmu *pmu,
 
 	if (event->state >= PERF_EVENT_STATE_OFF)
 		event->state = PERF_EVENT_STATE_INACTIVE;
-	account_event_cpu(event, cpu);
 	perf_install_in_context(ctx, event, cpu);
 }
 
@@ -13231,7 +13212,7 @@ inherit_event(struct perf_event *parent_event,
 	pmu_ctx = find_get_pmu_context(child_event->pmu, child_ctx, child_event);
 	if (IS_ERR(pmu_ctx)) {
 		free_event(child_event);
-		return NULL;
+		return ERR_CAST(pmu_ctx);
 	}
 	child_event->pmu_ctx = pmu_ctx;
 
@@ -13742,8 +13723,7 @@ static int __perf_cgroup_move(void *info)
 	struct task_struct *task = info;
 
 	preempt_disable();
-	if (atomic_read(this_cpu_ptr(&perf_cgroup_events)))
-		perf_cgroup_switch(task);
+	perf_cgroup_switch(task);
 	preempt_enable();
 
 	return 0;
diff --git a/kernel/futex/syscalls.c b/kernel/futex/syscalls.c
index 086a22d1adb7..a8074079b09e 100644
--- a/kernel/futex/syscalls.c
+++ b/kernel/futex/syscalls.c
@@ -286,19 +286,22 @@ SYSCALL_DEFINE5(futex_waitv, struct futex_waitv __user *, waiters,
 	}
 
 	futexv = kcalloc(nr_futexes, sizeof(*futexv), GFP_KERNEL);
-	if (!futexv)
-		return -ENOMEM;
+	if (!futexv) {
+		ret = -ENOMEM;
+		goto destroy_timer;
+	}
 
 	ret = futex_parse_waitv(futexv, waiters, nr_futexes);
 	if (!ret)
 		ret = futex_wait_multiple(futexv, nr_futexes, timeout ? &to : NULL);
 
+	kfree(futexv);
+
+destroy_timer:
 	if (timeout) {
 		hrtimer_cancel(&to.timer);
 		destroy_hrtimer_on_stack(&to.timer);
 	}
-
-	kfree(futexv);
 	return ret;
 }
 
diff --git a/kernel/locking/rtmutex.c b/kernel/locking/rtmutex.c
index 7779ee8abc2a..010cf4e6d0b8 100644
--- a/kernel/locking/rtmutex.c
+++ b/kernel/locking/rtmutex.c
@@ -89,15 +89,31 @@ static inline int __ww_mutex_check_kill(struct rt_mutex *lock,
  * set this bit before looking at the lock.
  */
 
-static __always_inline void
-rt_mutex_set_owner(struct rt_mutex_base *lock, struct task_struct *owner)
+static __always_inline struct task_struct *
+rt_mutex_owner_encode(struct rt_mutex_base *lock, struct task_struct *owner)
 {
 	unsigned long val = (unsigned long)owner;
 
 	if (rt_mutex_has_waiters(lock))
 		val |= RT_MUTEX_HAS_WAITERS;
 
-	WRITE_ONCE(lock->owner, (struct task_struct *)val);
+	return (struct task_struct *)val;
+}
+
+static __always_inline void
+rt_mutex_set_owner(struct rt_mutex_base *lock, struct task_struct *owner)
+{
+	/*
+	 * lock->wait_lock is held but explicit acquire semantics are needed
+	 * for a new lock owner so WRITE_ONCE is insufficient.
+	 */
+	xchg_acquire(&lock->owner, rt_mutex_owner_encode(lock, owner));
+}
+
+static __always_inline void rt_mutex_clear_owner(struct rt_mutex_base *lock)
+{
+	/* lock->wait_lock is held so the unlock provides release semantics. */
+	WRITE_ONCE(lock->owner, rt_mutex_owner_encode(lock, NULL));
 }
 
 static __always_inline void clear_rt_mutex_waiters(struct rt_mutex_base *lock)
@@ -106,7 +122,8 @@ static __always_inline void clear_rt_mutex_waiters(struct rt_mutex_base *lock)
 			((unsigned long)lock->owner & ~RT_MUTEX_HAS_WAITERS);
 }
 
-static __always_inline void fixup_rt_mutex_waiters(struct rt_mutex_base *lock)
+static __always_inline void
+fixup_rt_mutex_waiters(struct rt_mutex_base *lock, bool acquire_lock)
 {
 	unsigned long owner, *p = (unsigned long *) &lock->owner;
 
@@ -172,8 +189,21 @@ static __always_inline void fixup_rt_mutex_waiters(struct rt_mutex_base *lock)
 	 * still set.
 	 */
 	owner = READ_ONCE(*p);
-	if (owner & RT_MUTEX_HAS_WAITERS)
-		WRITE_ONCE(*p, owner & ~RT_MUTEX_HAS_WAITERS);
+	if (owner & RT_MUTEX_HAS_WAITERS) {
+		/*
+		 * See rt_mutex_set_owner() and rt_mutex_clear_owner() on
+		 * why xchg_acquire() is used for updating owner for
+		 * locking and WRITE_ONCE() for unlocking.
+		 *
+		 * WRITE_ONCE() would work for the acquire case too, but
+		 * in case that the lock acquisition failed it might
+		 * force other lockers into the slow path unnecessarily.
+		 */
+		if (acquire_lock)
+			xchg_acquire(p, owner & ~RT_MUTEX_HAS_WAITERS);
+		else
+			WRITE_ONCE(*p, owner & ~RT_MUTEX_HAS_WAITERS);
+	}
 }
 
 /*
@@ -208,6 +238,13 @@ static __always_inline void mark_rt_mutex_waiters(struct rt_mutex_base *lock)
 		owner = *p;
 	} while (cmpxchg_relaxed(p, owner,
 				 owner | RT_MUTEX_HAS_WAITERS) != owner);
+
+	/*
+	 * The cmpxchg loop above is relaxed to avoid back-to-back ACQUIRE
+	 * operations in the event of contention. Ensure the successful
+	 * cmpxchg is visible.
+	 */
+	smp_mb__after_atomic();
 }
 
 /*
@@ -1243,7 +1280,7 @@ static int __sched __rt_mutex_slowtrylock(struct rt_mutex_base *lock)
 	 * try_to_take_rt_mutex() sets the lock waiters bit
 	 * unconditionally. Clean this up.
 	 */
-	fixup_rt_mutex_waiters(lock);
+	fixup_rt_mutex_waiters(lock, true);
 
 	return ret;
 }
@@ -1604,7 +1641,7 @@ static int __sched __rt_mutex_slowlock(struct rt_mutex_base *lock,
 	 * try_to_take_rt_mutex() sets the waiter bit
 	 * unconditionally. We might have to fix that up.
 	 */
-	fixup_rt_mutex_waiters(lock);
+	fixup_rt_mutex_waiters(lock, true);
 
 	trace_contention_end(lock, ret);
 
@@ -1719,7 +1756,7 @@ static void __sched rtlock_slowlock_locked(struct rt_mutex_base *lock)
 	 * try_to_take_rt_mutex() sets the waiter bit unconditionally.
 	 * We might have to fix that up:
 	 */
-	fixup_rt_mutex_waiters(lock);
+	fixup_rt_mutex_waiters(lock, true);
 	debug_rt_mutex_free_waiter(&waiter);
 
 	trace_contention_end(lock, 0);
diff --git a/kernel/locking/rtmutex_api.c b/kernel/locking/rtmutex_api.c
index 900220941caa..cb9fdff76a8a 100644
--- a/kernel/locking/rtmutex_api.c
+++ b/kernel/locking/rtmutex_api.c
@@ -267,7 +267,7 @@ void __sched rt_mutex_init_proxy_locked(struct rt_mutex_base *lock,
 void __sched rt_mutex_proxy_unlock(struct rt_mutex_base *lock)
 {
 	debug_rt_mutex_proxy_unlock(lock);
-	rt_mutex_set_owner(lock, NULL);
+	rt_mutex_clear_owner(lock);
 }
 
 /**
@@ -382,7 +382,7 @@ int __sched rt_mutex_wait_proxy_lock(struct rt_mutex_base *lock,
 	 * try_to_take_rt_mutex() sets the waiter bit unconditionally. We might
 	 * have to fix that up.
 	 */
-	fixup_rt_mutex_waiters(lock);
+	fixup_rt_mutex_waiters(lock, true);
 	raw_spin_unlock_irq(&lock->wait_lock);
 
 	return ret;
@@ -438,7 +438,7 @@ bool __sched rt_mutex_cleanup_proxy_lock(struct rt_mutex_base *lock,
 	 * try_to_take_rt_mutex() sets the waiter bit unconditionally. We might
 	 * have to fix that up.
 	 */
-	fixup_rt_mutex_waiters(lock);
+	fixup_rt_mutex_waiters(lock, false);
 
 	raw_spin_unlock_irq(&lock->wait_lock);