Merge tag 'next.2025.05.17a' of git://git.kernel.org/pub/scm/linux/kernel/git/rcu/linux

Pull RCU updates from Joel Fernandes:
 - Removed swake_up_one_online() workaround
 - Reverted an incorrect rcuog wake-up fix from offline softirq
 - Rust RCU Guard methods marked as inline
 - Updated MAINTAINERS with Joel’s and Zqiang's new email address
 - Replaced magic constant in rcu_seq_done_exact() with named constant
 - Added warning mechanism to validate rcu_seq_done_exact()
 - Switched SRCU polling API to use rcu_seq_done_exact()
 - Commented on redundant delta check in rcu_seq_done_exact()
 - Made ->gpwrap tests in rcutorture more frequent
 - Fixed reuse of ARM64 images in rcutorture
 - rcutorture improved to check Kconfig and reader conflict handling
 - Extracted logic from rcu_torture_one_read() for clarity
 - Updated LWN RCU API documentation links
 - Enabled --do-rt in torture.sh for CONFIG_PREEMPT_RT
 - Added tests for SRCU up/down reader primitives
 - Added comments and delays checks in rcutorture
 - Deprecated srcu_read_lock_lite() and srcu_read_unlock_lite() via checkpatch
 - Added --do-normal and --do-no-normal to torture.sh
 - Added RCU Rust binding tests to torture.sh
 - Reduced CPU overcommit and removed MAXSMP/CPUMASK_OFFSTACK in TREE01
 - Replaced kmalloc() with kcalloc() in rcuscale
 - Refined listRCU example code for stale data elimination
 - Fixed hardirq count bug for x86 in cpu_stall_cputime
 - Added safety checks in rcu/nocb for offloaded rdp access
 - Other miscellaneous changes

* tag 'next.2025.05.17a' of git://git.kernel.org/pub/scm/linux/kernel/git/rcu/linux: (27 commits)
  rcutorture: Fix issue with re-using old images on ARM64
  rcutorture: Remove MAXSMP and CPUMASK_OFFSTACK from TREE01
  rcutorture: Reduce TREE01 CPU overcommit
  torture: Check for "Call trace:" as well as "Call Trace:"
  rcutorture: Perform more frequent testing of ->gpwrap
  torture: Add testing of RCU's Rust bindings to torture.sh
  torture: Add --do-{,no-}normal to torture.sh
  checkpatch: Deprecate srcu_read_lock_lite() and srcu_read_unlock_lite()
  rcutorture: Comment invocations of tick_dep_set_task()
  rcu/nocb: Add Safe checks for access offloaded rdp
  rcuscale: using kcalloc() to relpace kmalloc()
  doc/RCU/listRCU: refine example code for eliminating stale data
  doc: Update LWN RCU API links in whatisRCU.rst
  Revert "rcu/nocb: Fix rcuog wake-up from offline softirq"
  rust: sync: rcu: Mark Guard methods as inline
  rcu/cpu_stall_cputime: fix the hardirq count for x86 architecture
  rcu: Remove swake_up_one_online() bandaid
  MAINTAINERS: Update Zqiang's email address
  rcutorture: Make torture.sh --do-rt use CONFIG_PREEMPT_RT
  srcu: Use rcu_seq_done_exact() for polling API
  ...

Author: torvalds

Documentation/RCU/listRCU.rst

@@ -334,7 +334,7 @@ If the system-call audit module were to ever need to reject stale data, one way
 to accomplish this would be to add a ``deleted`` flag and a ``lock`` spinlock to the
 ``audit_entry`` structure, and modify audit_filter_task() as follows::
 
-	static enum audit_state audit_filter_task(struct task_struct *tsk)
+	static struct audit_entry *audit_filter_task(struct task_struct *tsk, char **key)
 	{
 		struct audit_entry *e;
 		enum audit_state   state;
@@ -346,16 +346,18 @@ to accomplish this would be to add a ``deleted`` flag and a ``lock`` spinlock to
 				if (e->deleted) {
 					spin_unlock(&e->lock);
 					rcu_read_unlock();
-					return AUDIT_BUILD_CONTEXT;
+					return NULL;
 				}
 				rcu_read_unlock();
 				if (state == AUDIT_STATE_RECORD)
 					*key = kstrdup(e->rule.filterkey, GFP_ATOMIC);
-				return state;
+				/* As long as e->lock is held, e is valid and
+				 * its value is not stale */
+				return e;
 			}
 		}
 		rcu_read_unlock();
-		return AUDIT_BUILD_CONTEXT;
+		return NULL;
 	}
 
 The ``audit_del_rule()`` function would need to set the ``deleted`` flag under the

Documentation/RCU/whatisRCU.rst

@@ -15,6 +15,9 @@ to start learning about RCU:
 |	2014 Big API Table           https://lwn.net/Articles/609973/
 | 6.	The RCU API, 2019 Edition    https://lwn.net/Articles/777036/
 |	2019 Big API Table           https://lwn.net/Articles/777165/
+| 7.	The RCU API, 2024 Edition    https://lwn.net/Articles/988638/
+|       2024 Background Information  https://lwn.net/Articles/988641/
+|	2024 Big API Table           https://lwn.net/Articles/988666/
 
 For those preferring video:
 

Documentation/admin-guide/kernel-parameters.txt

@@ -5672,6 +5672,31 @@
 			are zero, rcutorture acts as if is interpreted
 			they are all non-zero.
 
+	rcutorture.gpwrap_lag= [KNL]
+			Enable grace-period wrap lag testing. Setting
+			to false prevents the gpwrap lag test from
+			running. Default is true.
+
+	rcutorture.gpwrap_lag_gps= [KNL]
+			Set the value for grace-period wrap lag during
+			active lag testing periods. This controls how many
+			grace periods differences we tolerate between
+			rdp and rnp's gp_seq before setting overflow flag.
+			The default is always set to 8.
+
+	rcutorture.gpwrap_lag_cycle_mins= [KNL]
+			Set the total cycle duration for gpwrap lag
+			testing in minutes. This is the total time for
+			one complete cycle of active and inactive
+			testing periods. Default is 30 minutes.
+
+	rcutorture.gpwrap_lag_active_mins= [KNL]
+			Set the duration for which gpwrap lag is active
+			within each cycle, in minutes. During this time,
+			the grace-period wrap lag will be set to the
+			value specified by gpwrap_lag_gps. Default is
+			5 minutes.
+
 	rcutorture.irqreader= [KNL]
 			Run RCU readers from irq handlers, or, more
 			accurately, from a timer handler.  Not all RCU

MAINTAINERS

@@ -1682,7 +1682,7 @@ M:	Greg Kroah-Hartman <gregkh@linuxfoundation.org>
 M:	Arve Hjønnevåg <arve@android.com>
 M:	Todd Kjos <tkjos@android.com>
 M:	Martijn Coenen <maco@android.com>
-M:	Joel Fernandes <joel@joelfernandes.org>
+M:	Joel Fernandes <joelagnelf@nvidia.com>
 M:	Christian Brauner <christian@brauner.io>
 M:	Carlos Llamas <cmllamas@google.com>
 M:	Suren Baghdasaryan <surenb@google.com>
@@ -13746,7 +13746,7 @@ M:	Luc Maranget <luc.maranget@inria.fr>
 M:	"Paul E. McKenney" <paulmck@kernel.org>
 R:	Akira Yokosawa <akiyks@gmail.com>
 R:	Daniel Lustig <dlustig@nvidia.com>
-R:	Joel Fernandes <joel@joelfernandes.org>
+R:	Joel Fernandes <joelagnelf@nvidia.com>
 L:	linux-kernel@vger.kernel.org
 L:	linux-arch@vger.kernel.org
 L:	lkmm@lists.linux.dev
@@ -20514,14 +20514,14 @@ READ-COPY UPDATE (RCU)
 M:	"Paul E. McKenney" <paulmck@kernel.org>
 M:	Frederic Weisbecker <frederic@kernel.org> (kernel/rcu/tree_nocb.h)
 M:	Neeraj Upadhyay <neeraj.upadhyay@kernel.org> (kernel/rcu/tasks.h)
-M:	Joel Fernandes <joel@joelfernandes.org>
+M:	Joel Fernandes <joelagnelf@nvidia.com>
 M:	Josh Triplett <josh@joshtriplett.org>
 M:	Boqun Feng <boqun.feng@gmail.com>
 M:	Uladzislau Rezki <urezki@gmail.com>
 R:	Steven Rostedt <rostedt@goodmis.org>
 R:	Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
 R:	Lai Jiangshan <jiangshanlai@gmail.com>
-R:	Zqiang <qiang.zhang1211@gmail.com>
+R:	Zqiang <qiang.zhang@linux.dev>
 L:	rcu@vger.kernel.org
 S:	Supported
 W:	http://www.rdrop.com/users/paulmck/RCU/

kernel/rcu/rcu.h

@@ -57,6 +57,9 @@
 /* Low-order bit definition for polled grace-period APIs. */
 #define RCU_GET_STATE_COMPLETED	0x1
 
+/* A complete grace period count */
+#define RCU_SEQ_GP (RCU_SEQ_STATE_MASK + 1)
+
 extern int sysctl_sched_rt_runtime;
 
 /*
@@ -157,12 +160,21 @@ static inline bool rcu_seq_done(unsigned long *sp, unsigned long s)
  * Given a snapshot from rcu_seq_snap(), determine whether or not a
  * full update-side operation has occurred, but do not allow the
  * (ULONG_MAX / 2) safety-factor/guard-band.
+ *
+ * The token returned by get_state_synchronize_rcu_full() is based on
+ * rcu_state.gp_seq but it is tested in poll_state_synchronize_rcu_full()
+ * against the root rnp->gp_seq. Since rcu_seq_start() is first called
+ * on rcu_state.gp_seq and only later reflected on the root rnp->gp_seq,
+ * it is possible that rcu_seq_snap(rcu_state.gp_seq) returns 2 full grace
+ * periods ahead of the root rnp->gp_seq. To prevent false-positives with the
+ * full polling API that a wrap around instantly completed the GP, when nothing
+ * like that happened, adjust for the 2 GPs in the ULONG_CMP_LT().
  */
 static inline bool rcu_seq_done_exact(unsigned long *sp, unsigned long s)
 {
 	unsigned long cur_s = READ_ONCE(*sp);
 
-	return ULONG_CMP_GE(cur_s, s) || ULONG_CMP_LT(cur_s, s - (3 * RCU_SEQ_STATE_MASK + 1));
+	return ULONG_CMP_GE(cur_s, s) || ULONG_CMP_LT(cur_s, s - (2 * RCU_SEQ_GP));
 }
 
 /*
@@ -572,6 +584,8 @@ void do_trace_rcu_torture_read(const char *rcutorturename,
 			       unsigned long c_old,
 			       unsigned long c);
 void rcu_gp_set_torture_wait(int duration);
+void rcu_set_gpwrap_lag(unsigned long lag);
+int rcu_get_gpwrap_count(int cpu);
 #else
 static inline void rcutorture_get_gp_data(int *flags, unsigned long *gp_seq)
 {
@@ -589,6 +603,8 @@ void do_trace_rcu_torture_read(const char *rcutorturename,
 	do { } while (0)
 #endif
 static inline void rcu_gp_set_torture_wait(int duration) { }
+static inline void rcu_set_gpwrap_lag(unsigned long lag) { }
+static inline int rcu_get_gpwrap_count(int cpu) { return 0; }
 #endif
 unsigned long long rcutorture_gather_gp_seqs(void);
 void rcutorture_format_gp_seqs(unsigned long long seqs, char *cp, size_t len);

kernel/rcu/rcuscale.c

@@ -762,7 +762,7 @@ kfree_scale_thread(void *arg)
 		}
 
 		for (i = 0; i < kfree_alloc_num; i++) {
-			alloc_ptr = kmalloc(kfree_mult * sizeof(struct kfree_obj), GFP_KERNEL);
+			alloc_ptr = kcalloc(kfree_mult, sizeof(struct kfree_obj), GFP_KERNEL);
 			if (!alloc_ptr)
 				return -ENOMEM;