Merge tag 'sched-urgent-2025-02-22' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull rseq fixes from Ingo Molnar:

 - Fix overly spread-out RSEQ concurrency ID allocation pattern that
   regressed certain workloads

 - Fix RSEQ registration syscall behavior on -EFAULT errors when
   CONFIG_DEBUG_RSEQ=y (This debug option is disabled on most
   distributions)

* tag 'sched-urgent-2025-02-22' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  rseq: Fix rseq registration with CONFIG_DEBUG_RSEQ
  sched: Compact RSEQ concurrency IDs with reduced threads and affinity

Author: torvalds

include/linux/mm_types.h

@@ -875,10 +875,11 @@ struct mm_struct {
 		 */
 		unsigned int nr_cpus_allowed;
 		/**
-		 * @max_nr_cid: Maximum number of concurrency IDs allocated.
+		 * @max_nr_cid: Maximum number of allowed concurrency
+		 *              IDs allocated.
 		 *
-		 * Track the highest number of concurrency IDs allocated for the
-		 * mm.
+		 * Track the highest number of allowed concurrency IDs
+		 * allocated for the mm.
 		 */
 		atomic_t max_nr_cid;
 		/**

kernel/rseq.c

@@ -507,9 +507,6 @@ SYSCALL_DEFINE4(rseq, struct rseq __user *, rseq, u32, rseq_len,
 		return -EINVAL;
 	if (!access_ok(rseq, rseq_len))
 		return -EFAULT;
-	current->rseq = rseq;
-	current->rseq_len = rseq_len;
-	current->rseq_sig = sig;
 #ifdef CONFIG_DEBUG_RSEQ
 	/*
 	 * Initialize the in-kernel rseq fields copy for validation of
@@ -521,6 +518,14 @@ SYSCALL_DEFINE4(rseq, struct rseq __user *, rseq, u32, rseq_len,
 	    get_user(rseq_kernel_fields(current)->mm_cid, &rseq->mm_cid))
 		return -EFAULT;
 #endif
+	/*
+	 * Activate the registration by setting the rseq area address, length
+	 * and signature in the task struct.
+	 */
+	current->rseq = rseq;
+	current->rseq_len = rseq_len;
+	current->rseq_sig = sig;
+
 	/*
 	 * If rseq was previously inactive, and has just been
 	 * registered, ensure the cpu_id_start and cpu_id fields

kernel/sched/sched.h

@@ -3698,19 +3698,38 @@ static inline int __mm_cid_try_get(struct task_struct *t, struct mm_struct *mm)
 {
 	struct cpumask *cidmask = mm_cidmask(mm);
 	struct mm_cid __percpu *pcpu_cid = mm->pcpu_cid;
-	int cid = __this_cpu_read(pcpu_cid->recent_cid);
+	int cid, max_nr_cid, allowed_max_nr_cid;
 
+	/*
+	 * After shrinking the number of threads or reducing the number
+	 * of allowed cpus, reduce the value of max_nr_cid so expansion
+	 * of cid allocation will preserve cache locality if the number
+	 * of threads or allowed cpus increase again.
+	 */
+	max_nr_cid = atomic_read(&mm->max_nr_cid);
+	while ((allowed_max_nr_cid = min_t(int, READ_ONCE(mm->nr_cpus_allowed),
+					   atomic_read(&mm->mm_users))),
+	       max_nr_cid > allowed_max_nr_cid) {
+		/* atomic_try_cmpxchg loads previous mm->max_nr_cid into max_nr_cid. */
+		if (atomic_try_cmpxchg(&mm->max_nr_cid, &max_nr_cid, allowed_max_nr_cid)) {
+			max_nr_cid = allowed_max_nr_cid;
+			break;
+		}
+	}
 	/* Try to re-use recent cid. This improves cache locality. */
-	if (!mm_cid_is_unset(cid) && !cpumask_test_and_set_cpu(cid, cidmask))
+	cid = __this_cpu_read(pcpu_cid->recent_cid);
+	if (!mm_cid_is_unset(cid) && cid < max_nr_cid &&
+	    !cpumask_test_and_set_cpu(cid, cidmask))
 		return cid;
 	/*
 	 * Expand cid allocation if the maximum number of concurrency
 	 * IDs allocated (max_nr_cid) is below the number cpus allowed
 	 * and number of threads. Expanding cid allocation as much as
 	 * possible improves cache locality.
 	 */
-	cid = atomic_read(&mm->max_nr_cid);
+	cid = max_nr_cid;
 	while (cid < READ_ONCE(mm->nr_cpus_allowed) && cid < atomic_read(&mm->mm_users)) {
+		/* atomic_try_cmpxchg loads previous mm->max_nr_cid into cid. */
 		if (!atomic_try_cmpxchg(&mm->max_nr_cid, &cid, cid + 1))
 			continue;
 		if (!cpumask_test_and_set_cpu(cid, cidmask))