[PATCH v13 0/3] sched: Restructure task_mm_cid_work for predictability

[PATCH v13 0/3] sched: Restructure task_mm_cid_work for predictability

[Gabriele Monaco]

This patchset moves the task_mm_cid_work to a preemptible and migratable
context. This reduces the impact of this work to the scheduling latency
of real time tasks.
The change makes the recurrence of the task a bit more predictable.

The behaviour causing latency was introduced in commit 223baf9d17f2
("sched: Fix performance regression introduced by mm_cid") which
introduced a task work tied to the scheduler tick.
That approach presents two possible issues:
* the task work runs before returning to user and causes, in fact, a
  scheduling latency (with order of magnitude significant in PREEMPT_RT)
* periodic tasks with short runtime are less likely to run during the
  tick, hence they might not run the task work at all

Patch 1 add support for prev_sum_exec_runtime to the RT, deadline and
sched_ext classes as it is supported by fair, this is required to avoid
calling rseq_preempt on tick if the runtime is below a threshold.

Patch 2 contains the main changes, removing the task_work on the
scheduler tick and using a work_struct scheduled more reliably during
__rseq_handle_notify_resume.

Patch 3 adds a selftest to validate the functionality of the
task_mm_cid_work (i.e. to compact the mm_cids).

Changes since V12:
* Ensure the tick schedules the mm_cid compaction only once for tasks
  executing longer than 100ms (until the scan expires again)
* Execute an rseq_preempt from the tick only after compaction was done
  and the cid assignation changed

Changes since V11:
* Remove variable to make mm_cid_needs_scan more compact
* All patches reviewed

Changes since V10:
* Fix compilation errors with RSEQ and/or MM_CID disabled

Changes since V9:
* Simplify and move checks from task_queue_mm_cid to its call site

Changes since V8 [1]:
* Add support for prev_sum_exec_runtime to RT, deadline and sched_ext
* Avoid rseq_preempt on ticks unless executing for more than 100ms
* Queue the work on the unbound workqueue

Changes since V7:
* Schedule mm_cid compaction and update at every tick too
* mmgrab before scheduling the work

Changes since V6 [2]:
* Switch to a simple work_struct instead of a delayed work
* Schedule the work_struct in __rseq_handle_notify_resume
* Asynchronously disable the work but make sure mm is there while we run
* Remove first patch as merged independently
* Fix commit tag for test

Changes since V5:
* Punctuation

Changes since V4 [3]:
* Fixes on the selftest
    * Polished memory allocation and cleanup
    * Handle the test failure in main

Changes since V3 [4]:
* Fixes on the selftest
    * Minor style issues in comments and indentation
    * Use of perror where possible
    * Add a barrier to align threads execution
    * Improve test failure and error handling

Changes since V2 [5]:
* Change the order of the patches
* Merge patches changing the main delayed_work logic
* Improved self-test to spawn 1 less thread and use the main one instead

Changes since V1 [6]:
* Re-arm the delayed_work at each invocation
* Cancel the work synchronously at mmdrop
* Remove next scan fields and completely rely on the delayed_work
* Shrink mm_cid allocation with nr thread/affinity (Mathieu Desnoyers)
* Add self test

[1] - https://lore.kernel.org/lkml/20250220102639.141314-1-gmonaco@redhat.com
[2] - https://lore.kernel.org/lkml/20250210153253.460471-1-gmonaco@redhat.com
[3] - https://lore.kernel.org/lkml/20250113074231.61638-4-gmonaco@redhat.com
[4] - https://lore.kernel.org/lkml/20241216130909.240042-1-gmonaco@redhat.com
[5] - https://lore.kernel.org/lkml/20241213095407.271357-1-gmonaco@redhat.com
[6] - https://lore.kernel.org/lkml/20241205083110.180134-2-gmonaco@redhat.com

To: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
To: Peter Zijlstra <peterz@infradead.org>
To: Ingo Molnar <mingo@redhat.org>

Gabriele Monaco (3):
  sched: Add prev_sum_exec_runtime support for RT, DL and SCX classes
  sched: Move task_mm_cid_work to mm work_struct
  selftests/rseq: Add test for mm_cid compaction

 include/linux/mm_types.h                      |  26 +++
 include/linux/sched.h                         |   8 +-
 kernel/rseq.c                                 |   2 +
 kernel/sched/core.c                           |  75 ++++---
 kernel/sched/deadline.c                       |   1 +
 kernel/sched/ext.c                            |   1 +
 kernel/sched/rt.c                             |   1 +
 kernel/sched/sched.h                          |   6 +-
 tools/testing/selftests/rseq/.gitignore       |   1 +
 tools/testing/selftests/rseq/Makefile         |   2 +-
 .../selftests/rseq/mm_cid_compaction_test.c   | 200 ++++++++++++++++++
 11 files changed, 294 insertions(+), 29 deletions(-)
 create mode 100644 tools/testing/selftests/rseq/mm_cid_compaction_test.c


base-commit: 8ffd015db85fea3e15a77027fda6c02ced4d2444
-- 
2.49.0

[PATCH v13 1/3] sched: Add prev_sum_exec_runtime support for RT, DL and SCX classes

[Gabriele Monaco]

The fair scheduling class relies on prev_sum_exec_runtime to compute the
duration of the task's runtime since it was last scheduled. This value
is currently not required by other scheduling classes but can be useful
to understand long running tasks and take certain actions (e.g. during a
scheduler tick).

Add support for prev_sum_exec_runtime to the RT, deadline and sched_ext
classes by simply assigning the sum_exec_runtime at each set_next_task.

Reviewed-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Signed-off-by: Gabriele Monaco <gmonaco@redhat.com>
---
 kernel/sched/deadline.c | 1 +
 kernel/sched/ext.c      | 1 +
 kernel/sched/rt.c       | 1 +
 3 files changed, 3 insertions(+)

diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c
index ad45a8fea245e..8387006396c8a 100644
--- a/kernel/sched/deadline.c
+++ b/kernel/sched/deadline.c
@@ -2389,6 +2389,7 @@ static void set_next_task_dl(struct rq *rq, struct task_struct *p, bool first)
 	p->se.exec_start = rq_clock_task(rq);
 	if (on_dl_rq(&p->dl))
 		update_stats_wait_end_dl(dl_rq, dl_se);
+	p->se.prev_sum_exec_runtime = p->se.sum_exec_runtime;
 
 	/* You can't push away the running task */
 	dequeue_pushable_dl_task(rq, p);
diff --git a/kernel/sched/ext.c b/kernel/sched/ext.c
index 66bcd40a28ca1..4bb4f0ee23d75 100644
--- a/kernel/sched/ext.c
+++ b/kernel/sched/ext.c
@@ -3109,6 +3109,7 @@ static void set_next_task_scx(struct rq *rq, struct task_struct *p, bool first)
 	}
 
 	p->se.exec_start = rq_clock_task(rq);
+	p->se.prev_sum_exec_runtime = p->se.sum_exec_runtime;
 
 	/* see dequeue_task_scx() on why we skip when !QUEUED */
 	if (SCX_HAS_OP(running) && (p->scx.flags & SCX_TASK_QUEUED))
diff --git a/kernel/sched/rt.c b/kernel/sched/rt.c
index fa03ec3ed56a2..1ed4a133fdc79 100644
--- a/kernel/sched/rt.c
+++ b/kernel/sched/rt.c
@@ -1682,6 +1682,7 @@ static inline void set_next_task_rt(struct rq *rq, struct task_struct *p, bool f
 	p->se.exec_start = rq_clock_task(rq);
 	if (on_rt_rq(&p->rt))
 		update_stats_wait_end_rt(rt_rq, rt_se);
+	p->se.prev_sum_exec_runtime = p->se.sum_exec_runtime;
 
 	/* The running task is never eligible for pushing */
 	dequeue_pushable_task(rq, p);
-- 
2.49.0

[PATCH v13 2/3] sched: Move task_mm_cid_work to mm work_struct

[Gabriele Monaco]

Currently, the task_mm_cid_work function is called in a task work
triggered by a scheduler tick to frequently compact the mm_cids of each
process. This can delay the execution of the corresponding thread for
the entire duration of the function, negatively affecting the response
in case of real time tasks. In practice, we observe task_mm_cid_work
increasing the latency of 30-35us on a 128 cores system, this order of
magnitude is meaningful under PREEMPT_RT.

Run the task_mm_cid_work in a new work_struct connected to the
mm_struct rather than in the task context before returning to
userspace.

This work_struct is initialised with the mm and disabled before freeing
it. The queuing of the work happens while returning to userspace in
__rseq_handle_notify_resume, maintaining the checks to avoid running
more frequently than MM_CID_SCAN_DELAY.
To make sure this happens predictably also on long running tasks, we
trigger a call to __rseq_handle_notify_resume also from the scheduler
tick if the runtime exceeded a 100ms threshold.

The main advantage of this change is that the function can be offloaded
to a different CPU and even preempted by RT tasks.

Moreover, this new behaviour is more predictable with periodic tasks
with short runtime, which may rarely run during a scheduler tick.
Now, the work is always scheduled when the task returns to userspace.

The work is disabled during mmdrop, since the function cannot sleep in
all kernel configurations, we cannot wait for possibly running work
items to terminate. We make sure the mm is valid in case the task is
terminating by reserving it with mmgrab/mmdrop, returning prematurely if
we are really the last user while the work gets to run.
This situation is unlikely since we don't schedule the work for exiting
tasks, but we cannot rule it out.

Fixes: 223baf9d17f2 ("sched: Fix performance regression introduced by mm_cid")
Signed-off-by: Gabriele Monaco <gmonaco@redhat.com>
---
 include/linux/mm_types.h | 26 ++++++++++++++
 include/linux/sched.h    |  8 ++++-
 kernel/rseq.c            |  2 ++
 kernel/sched/core.c      | 75 ++++++++++++++++++++++++++--------------
 kernel/sched/sched.h     |  6 ++--
 5 files changed, 89 insertions(+), 28 deletions(-)

diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
index 56d07edd01f91..e4ae9295508cf 100644
--- a/include/linux/mm_types.h
+++ b/include/linux/mm_types.h
@@ -982,6 +982,10 @@ struct mm_struct {
 		 * mm nr_cpus_allowed updates.
 		 */
 		raw_spinlock_t cpus_allowed_lock;
+		/*
+		 * @cid_work: Work item to run the mm_cid scan.
+		 */
+		struct work_struct cid_work;
 #endif
 #ifdef CONFIG_MMU
 		atomic_long_t pgtables_bytes;	/* size of all page tables */
@@ -1282,6 +1286,8 @@ enum mm_cid_state {
 	MM_CID_LAZY_PUT = (1U << 31),
 };
 
+extern void task_mm_cid_work(struct work_struct *work);
+
 static inline bool mm_cid_is_unset(int cid)
 {
 	return cid == MM_CID_UNSET;
@@ -1354,12 +1360,14 @@ static inline int mm_alloc_cid_noprof(struct mm_struct *mm, struct task_struct *
 	if (!mm->pcpu_cid)
 		return -ENOMEM;
 	mm_init_cid(mm, p);
+	INIT_WORK(&mm->cid_work, task_mm_cid_work);
 	return 0;
 }
 #define mm_alloc_cid(...)	alloc_hooks(mm_alloc_cid_noprof(__VA_ARGS__))
 
 static inline void mm_destroy_cid(struct mm_struct *mm)
 {
+	disable_work(&mm->cid_work);
 	free_percpu(mm->pcpu_cid);
 	mm->pcpu_cid = NULL;
 }
@@ -1381,6 +1389,16 @@ static inline void mm_set_cpus_allowed(struct mm_struct *mm, const struct cpumas
 	WRITE_ONCE(mm->nr_cpus_allowed, cpumask_weight(mm_allowed));
 	raw_spin_unlock(&mm->cpus_allowed_lock);
 }
+
+static inline bool mm_cid_needs_scan(struct mm_struct *mm)
+{
+	return mm && !time_before(jiffies, READ_ONCE(mm->mm_cid_next_scan));
+}
+
+static inline bool mm_cid_scan_pending(struct mm_struct *mm)
+{
+	return mm && work_pending(&mm->cid_work);
+}
 #else /* CONFIG_SCHED_MM_CID */
 static inline void mm_init_cid(struct mm_struct *mm, struct task_struct *p) { }
 static inline int mm_alloc_cid(struct mm_struct *mm, struct task_struct *p) { return 0; }
@@ -1391,6 +1409,14 @@ static inline unsigned int mm_cid_size(void)
 	return 0;
 }
 static inline void mm_set_cpus_allowed(struct mm_struct *mm, const struct cpumask *cpumask) { }
+static inline bool mm_cid_needs_scan(struct mm_struct *mm)
+{
+	return false;
+}
+static inline bool mm_cid_scan_pending(struct mm_struct *mm)
+{
+	return false;
+}
 #endif /* CONFIG_SCHED_MM_CID */
 
 struct mmu_gather;
diff --git a/include/linux/sched.h b/include/linux/sched.h
index f96ac19828934..3ffdb96ef6b0a 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1424,7 +1424,7 @@ struct task_struct {
 	int				last_mm_cid;	/* Most recent cid in mm */
 	int				migrate_from_cpu;
 	int				mm_cid_active;	/* Whether cid bitmap is active */
-	struct callback_head		cid_work;
+	unsigned long			last_cid_reset;	/* Time of last reset in jiffies */
 #endif
 
 	struct tlbflush_unmap_batch	tlb_ubc;
@@ -2281,4 +2281,10 @@ static __always_inline void alloc_tag_restore(struct alloc_tag *tag, struct allo
 #define alloc_tag_restore(_tag, _old)		do {} while (0)
 #endif
 
+#ifdef CONFIG_SCHED_MM_CID
+extern void task_queue_mm_cid(struct task_struct *curr);
+#else
+static inline void task_queue_mm_cid(struct task_struct *curr) { }
+#endif
+
 #endif
diff --git a/kernel/rseq.c b/kernel/rseq.c
index b7a1ec327e811..383db2ccad4d0 100644
--- a/kernel/rseq.c
+++ b/kernel/rseq.c
@@ -441,6 +441,8 @@ void __rseq_handle_notify_resume(struct ksignal *ksig, struct pt_regs *regs)
 	}
 	if (unlikely(rseq_update_cpu_node_id(t)))
 		goto error;
+	if (mm_cid_needs_scan(t->mm))
+		task_queue_mm_cid(t);
 	return;
 
 error:
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index c81cf642dba05..02b18649e6a09 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -10566,22 +10566,16 @@ static void sched_mm_cid_remote_clear_weight(struct mm_struct *mm, int cpu,
 	sched_mm_cid_remote_clear(mm, pcpu_cid, cpu);
 }
 
-static void task_mm_cid_work(struct callback_head *work)
+void task_mm_cid_work(struct work_struct *work)
 {
 	unsigned long now = jiffies, old_scan, next_scan;
-	struct task_struct *t = current;
 	struct cpumask *cidmask;
-	struct mm_struct *mm;
+	struct mm_struct *mm = container_of(work, struct mm_struct, cid_work);
 	int weight, cpu;
 
-	WARN_ON_ONCE(t != container_of(work, struct task_struct, cid_work));
-
-	work->next = work;	/* Prevent double-add */
-	if (t->flags & PF_EXITING)
-		return;
-	mm = t->mm;
-	if (!mm)
-		return;
+	/* We are the last user, process already terminated. */
+	if (atomic_read(&mm->mm_count) == 1)
+		goto out_drop;
 	old_scan = READ_ONCE(mm->mm_cid_next_scan);
 	next_scan = now + msecs_to_jiffies(MM_CID_SCAN_DELAY);
 	if (!old_scan) {
@@ -10594,9 +10588,9 @@ static void task_mm_cid_work(struct callback_head *work)
 			old_scan = next_scan;
 	}
 	if (time_before(now, old_scan))
-		return;
+		goto out_drop;
 	if (!try_cmpxchg(&mm->mm_cid_next_scan, &old_scan, next_scan))
-		return;
+		goto out_drop;
 	cidmask = mm_cidmask(mm);
 	/* Clear cids that were not recently used. */
 	for_each_possible_cpu(cpu)
@@ -10608,6 +10602,8 @@ static void task_mm_cid_work(struct callback_head *work)
 	 */
 	for_each_possible_cpu(cpu)
 		sched_mm_cid_remote_clear_weight(mm, cpu, weight);
+out_drop:
+	mmdrop(mm);
 }
 
 void init_sched_mm_cid(struct task_struct *t)
@@ -10620,23 +10616,52 @@ void init_sched_mm_cid(struct task_struct *t)
 		if (mm_users == 1)
 			mm->mm_cid_next_scan = jiffies + msecs_to_jiffies(MM_CID_SCAN_DELAY);
 	}
-	t->cid_work.next = &t->cid_work;	/* Protect against double add */
-	init_task_work(&t->cid_work, task_mm_cid_work);
 }
 
-void task_tick_mm_cid(struct rq *rq, struct task_struct *curr)
+void task_tick_mm_cid(struct rq *rq, struct task_struct *t)
 {
-	struct callback_head *work = &curr->cid_work;
-	unsigned long now = jiffies;
+	u64 rtime = t->se.sum_exec_runtime - t->se.prev_sum_exec_runtime;
 
-	if (!curr->mm || (curr->flags & (PF_EXITING | PF_KTHREAD)) ||
-	    work->next != work)
-		return;
-	if (time_before(now, READ_ONCE(curr->mm->mm_cid_next_scan)))
-		return;
+	/*
+	 * If a task is running unpreempted for a long time, it won't get its
+	 * mm_cid compacted and won't update its mm_cid value after a
+	 * compaction occurs.
+	 * For such a task, this function does two things:
+	 * A) trigger the mm_cid recompaction,
+	 * B) trigger an update of the task's rseq->mm_cid field at some point
+	 * after recompaction, so it can get a mm_cid value closer to 0.
+	 * A change in the mm_cid triggers an rseq_preempt.
+	 *
+	 * A occurs only once after the scan time elapsed, until the next scan
+	 * expires as well.
+	 * B occurs once after the compaction work completes, that is when scan
+	 * is no longer needed (it occurred for this mm) but the last rseq
+	 * preempt was done before the last mm_cid scan.
+	 */
+	if (t->mm && rtime > RSEQ_UNPREEMPTED_THRESHOLD) {
+		if (mm_cid_needs_scan(t->mm) && !mm_cid_scan_pending(t->mm))
+			rseq_set_notify_resume(t);
+		else if (time_after(jiffies, t->last_cid_reset +
+				      msecs_to_jiffies(MM_CID_SCAN_DELAY))) {
+			int old_cid = t->mm_cid;
+
+			if (!t->mm_cid_active)
+				return;
+			mm_cid_snapshot_time(rq, t->mm);
+			mm_cid_put_lazy(t);
+			t->last_mm_cid = t->mm_cid = mm_cid_get(rq, t, t->mm);
+			if (old_cid != t->mm_cid)
+				rseq_preempt(t);
+		}
+	}
+}
 
-	/* No page allocation under rq lock */
-	task_work_add(curr, work, TWA_RESUME);
+/* Call only when curr is a user thread. */
+void task_queue_mm_cid(struct task_struct *curr)
+{
+	/* Ensure the mm exists when we run. */
+	mmgrab(curr->mm);
+	queue_work(system_unbound_wq, &curr->mm->cid_work);
 }
 
 void sched_mm_cid_exit_signals(struct task_struct *t)
diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
index 47972f34ea701..c0f8fd4c575c3 100644
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -3582,13 +3582,14 @@ extern const char *preempt_modes[];
 
 #define SCHED_MM_CID_PERIOD_NS	(100ULL * 1000000)	/* 100ms */
 #define MM_CID_SCAN_DELAY	100			/* 100ms */
+#define RSEQ_UNPREEMPTED_THRESHOLD	SCHED_MM_CID_PERIOD_NS
 
 extern raw_spinlock_t cid_lock;
 extern int use_cid_lock;
 
 extern void sched_mm_cid_migrate_from(struct task_struct *t);
 extern void sched_mm_cid_migrate_to(struct rq *dst_rq, struct task_struct *t);
-extern void task_tick_mm_cid(struct rq *rq, struct task_struct *curr);
+extern void task_tick_mm_cid(struct rq *rq, struct task_struct *t);
 extern void init_sched_mm_cid(struct task_struct *t);
 
 static inline void __mm_cid_put(struct mm_struct *mm, int cid)
@@ -3798,6 +3799,7 @@ static inline int mm_cid_get(struct rq *rq, struct task_struct *t,
 	cid = __mm_cid_get(rq, t, mm);
 	__this_cpu_write(pcpu_cid->cid, cid);
 	__this_cpu_write(pcpu_cid->recent_cid, cid);
+	t->last_cid_reset = jiffies;
 
 	return cid;
 }
@@ -3857,7 +3859,7 @@ static inline void switch_mm_cid(struct rq *rq,
 static inline void switch_mm_cid(struct rq *rq, struct task_struct *prev, struct task_struct *next) { }
 static inline void sched_mm_cid_migrate_from(struct task_struct *t) { }
 static inline void sched_mm_cid_migrate_to(struct rq *dst_rq, struct task_struct *t) { }
-static inline void task_tick_mm_cid(struct rq *rq, struct task_struct *curr) { }
+static inline void task_tick_mm_cid(struct rq *rq, struct task_struct *t) { }
 static inline void init_sched_mm_cid(struct task_struct *t) { }
 #endif /* !CONFIG_SCHED_MM_CID */
 
-- 
2.49.0

[PATCH v13 3/3] selftests/rseq: Add test for mm_cid compaction

[Gabriele Monaco]

A task in the kernel (task_mm_cid_work) runs somewhat periodically to
compact the mm_cid for each process. Add a test to validate that it runs
correctly and timely.

The test spawns 1 thread pinned to each CPU, then each thread, including
the main one, runs in short bursts for some time. During this period, the
mm_cids should be spanning all numbers between 0 and nproc.

At the end of this phase, a thread with high enough mm_cid (>= nproc/2)
is selected to be the new leader, all other threads terminate.

After some time, the only running thread should see 0 as mm_cid, if that
doesn't happen, the compaction mechanism didn't work and the test fails.

The test never fails if only 1 core is available, in which case, we
cannot test anything as the only available mm_cid is 0.

Reviewed-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Signed-off-by: Gabriele Monaco <gmonaco@redhat.com>
---
 tools/testing/selftests/rseq/.gitignore       |   1 +
 tools/testing/selftests/rseq/Makefile         |   2 +-
 .../selftests/rseq/mm_cid_compaction_test.c   | 200 ++++++++++++++++++
 3 files changed, 202 insertions(+), 1 deletion(-)
 create mode 100644 tools/testing/selftests/rseq/mm_cid_compaction_test.c

diff --git a/tools/testing/selftests/rseq/.gitignore b/tools/testing/selftests/rseq/.gitignore
index 0fda241fa62b0..b3920c59bf401 100644
--- a/tools/testing/selftests/rseq/.gitignore
+++ b/tools/testing/selftests/rseq/.gitignore
@@ -3,6 +3,7 @@ basic_percpu_ops_test
 basic_percpu_ops_mm_cid_test
 basic_test
 basic_rseq_op_test
+mm_cid_compaction_test
 param_test
 param_test_benchmark
 param_test_compare_twice
diff --git a/tools/testing/selftests/rseq/Makefile b/tools/testing/selftests/rseq/Makefile
index 0d0a5fae59547..bc4d940f66d40 100644
--- a/tools/testing/selftests/rseq/Makefile
+++ b/tools/testing/selftests/rseq/Makefile
@@ -17,7 +17,7 @@ OVERRIDE_TARGETS = 1
 TEST_GEN_PROGS = basic_test basic_percpu_ops_test basic_percpu_ops_mm_cid_test param_test \
 		param_test_benchmark param_test_compare_twice param_test_mm_cid \
 		param_test_mm_cid_benchmark param_test_mm_cid_compare_twice \
-		syscall_errors_test
+		syscall_errors_test mm_cid_compaction_test
 
 TEST_GEN_PROGS_EXTENDED = librseq.so
 
diff --git a/tools/testing/selftests/rseq/mm_cid_compaction_test.c b/tools/testing/selftests/rseq/mm_cid_compaction_test.c
new file mode 100644
index 0000000000000..7ddde3b657dd6
--- /dev/null
+++ b/tools/testing/selftests/rseq/mm_cid_compaction_test.c
@@ -0,0 +1,200 @@
+// SPDX-License-Identifier: LGPL-2.1
+#define _GNU_SOURCE
+#include <assert.h>
+#include <pthread.h>
+#include <sched.h>
+#include <stdint.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <stddef.h>
+
+#include "../kselftest.h"
+#include "rseq.h"
+
+#define VERBOSE 0
+#define printf_verbose(fmt, ...)                    \
+	do {                                        \
+		if (VERBOSE)                        \
+			printf(fmt, ##__VA_ARGS__); \
+	} while (0)
+
+/* 0.5 s */
+#define RUNNER_PERIOD 500000
+/* Number of runs before we terminate or get the token */
+#define THREAD_RUNS 5
+
+/*
+ * Number of times we check that the mm_cid were compacted.
+ * Checks are repeated every RUNNER_PERIOD.
+ */
+#define MM_CID_COMPACT_TIMEOUT 10
+
+struct thread_args {
+	int cpu;
+	int num_cpus;
+	pthread_mutex_t *token;
+	pthread_barrier_t *barrier;
+	pthread_t *tinfo;
+	struct thread_args *args_head;
+};
+
+static void __noreturn *thread_runner(void *arg)
+{
+	struct thread_args *args = arg;
+	int i, ret, curr_mm_cid;
+	cpu_set_t cpumask;
+
+	CPU_ZERO(&cpumask);
+	CPU_SET(args->cpu, &cpumask);
+	ret = pthread_setaffinity_np(pthread_self(), sizeof(cpumask), &cpumask);
+	if (ret) {
+		errno = ret;
+		perror("Error: failed to set affinity");
+		abort();
+	}
+	pthread_barrier_wait(args->barrier);
+
+	for (i = 0; i < THREAD_RUNS; i++)
+		usleep(RUNNER_PERIOD);
+	curr_mm_cid = rseq_current_mm_cid();
+	/*
+	 * We select one thread with high enough mm_cid to be the new leader.
+	 * All other threads (including the main thread) will terminate.
+	 * After some time, the mm_cid of the only remaining thread should
+	 * converge to 0, if not, the test fails.
+	 */
+	if (curr_mm_cid >= args->num_cpus / 2 &&
+	    !pthread_mutex_trylock(args->token)) {
+		printf_verbose(
+			"cpu%d has mm_cid=%d and will be the new leader.\n",
+			sched_getcpu(), curr_mm_cid);
+		for (i = 0; i < args->num_cpus; i++) {
+			if (args->tinfo[i] == pthread_self())
+				continue;
+			ret = pthread_join(args->tinfo[i], NULL);
+			if (ret) {
+				errno = ret;
+				perror("Error: failed to join thread");
+				abort();
+			}
+		}
+		pthread_barrier_destroy(args->barrier);
+		free(args->tinfo);
+		free(args->token);
+		free(args->barrier);
+		free(args->args_head);
+
+		for (i = 0; i < MM_CID_COMPACT_TIMEOUT; i++) {
+			curr_mm_cid = rseq_current_mm_cid();
+			printf_verbose("run %d: mm_cid=%d on cpu%d.\n", i,
+				       curr_mm_cid, sched_getcpu());
+			if (curr_mm_cid == 0)
+				exit(EXIT_SUCCESS);
+			usleep(RUNNER_PERIOD);
+		}
+		exit(EXIT_FAILURE);
+	}
+	printf_verbose("cpu%d has mm_cid=%d and is going to terminate.\n",
+		       sched_getcpu(), curr_mm_cid);
+	pthread_exit(NULL);
+}
+
+int test_mm_cid_compaction(void)
+{
+	cpu_set_t affinity;
+	int i, j, ret = 0, num_threads;
+	pthread_t *tinfo;
+	pthread_mutex_t *token;
+	pthread_barrier_t *barrier;
+	struct thread_args *args;
+
+	sched_getaffinity(0, sizeof(affinity), &affinity);
+	num_threads = CPU_COUNT(&affinity);
+	tinfo = calloc(num_threads, sizeof(*tinfo));
+	if (!tinfo) {
+		perror("Error: failed to allocate tinfo");
+		return -1;
+	}
+	args = calloc(num_threads, sizeof(*args));
+	if (!args) {
+		perror("Error: failed to allocate args");
+		ret = -1;
+		goto out_free_tinfo;
+	}
+	token = malloc(sizeof(*token));
+	if (!token) {
+		perror("Error: failed to allocate token");
+		ret = -1;
+		goto out_free_args;
+	}
+	barrier = malloc(sizeof(*barrier));
+	if (!barrier) {
+		perror("Error: failed to allocate barrier");
+		ret = -1;
+		goto out_free_token;
+	}
+	if (num_threads == 1) {
+		fprintf(stderr, "Cannot test on a single cpu. "
+				"Skipping mm_cid_compaction test.\n");
+		/* only skipping the test, this is not a failure */
+		goto out_free_barrier;
+	}
+	pthread_mutex_init(token, NULL);
+	ret = pthread_barrier_init(barrier, NULL, num_threads);
+	if (ret) {
+		errno = ret;
+		perror("Error: failed to initialise barrier");
+		goto out_free_barrier;
+	}
+	for (i = 0, j = 0; i < CPU_SETSIZE && j < num_threads; i++) {
+		if (!CPU_ISSET(i, &affinity))
+			continue;
+		args[j].num_cpus = num_threads;
+		args[j].tinfo = tinfo;
+		args[j].token = token;
+		args[j].barrier = barrier;
+		args[j].cpu = i;
+		args[j].args_head = args;
+		if (!j) {
+			/* The first thread is the main one */
+			tinfo[0] = pthread_self();
+			++j;
+			continue;
+		}
+		ret = pthread_create(&tinfo[j], NULL, thread_runner, &args[j]);
+		if (ret) {
+			errno = ret;
+			perror("Error: failed to create thread");
+			abort();
+		}
+		++j;
+	}
+	printf_verbose("Started %d threads.\n", num_threads);
+
+	/* Also main thread will terminate if it is not selected as leader */
+	thread_runner(&args[0]);
+
+	/* only reached in case of errors */
+out_free_barrier:
+	free(barrier);
+out_free_token:
+	free(token);
+out_free_args:
+	free(args);
+out_free_tinfo:
+	free(tinfo);
+
+	return ret;
+}
+
+int main(int argc, char **argv)
+{
+	if (!rseq_mm_cid_available()) {
+		fprintf(stderr, "Error: rseq_mm_cid unavailable\n");
+		return -1;
+	}
+	if (test_mm_cid_compaction())
+		return -1;
+	return 0;
+}
-- 
2.49.0

Re: [PATCH v13 2/3] sched: Move task_mm_cid_work to mm work_struct

[Mathieu Desnoyers]

On 2025-04-14 08:36, Gabriele Monaco wrote:
> Currently, the task_mm_cid_work function is called in a task work
> triggered by a scheduler tick to frequently compact the mm_cids of each
> process. This can delay the execution of the corresponding thread for
> the entire duration of the function, negatively affecting the response
> in case of real time tasks. In practice, we observe task_mm_cid_work
> increasing the latency of 30-35us on a 128 cores system, this order of
> magnitude is meaningful under PREEMPT_RT.
> 
> Run the task_mm_cid_work in a new work_struct connected to the
> mm_struct rather than in the task context before returning to
> userspace.
> 
> This work_struct is initialised with the mm and disabled before freeing
> it. The queuing of the work happens while returning to userspace in
> __rseq_handle_notify_resume, maintaining the checks to avoid running
> more frequently than MM_CID_SCAN_DELAY.
> To make sure this happens predictably also on long running tasks, we
> trigger a call to __rseq_handle_notify_resume also from the scheduler
> tick if the runtime exceeded a 100ms threshold.
> 
> The main advantage of this change is that the function can be offloaded
> to a different CPU and even preempted by RT tasks.
> 
> Moreover, this new behaviour is more predictable with periodic tasks
> with short runtime, which may rarely run during a scheduler tick.
> Now, the work is always scheduled when the task returns to userspace.
> 
> The work is disabled during mmdrop, since the function cannot sleep in
> all kernel configurations, we cannot wait for possibly running work
> items to terminate. We make sure the mm is valid in case the task is
> terminating by reserving it with mmgrab/mmdrop, returning prematurely if
> we are really the last user while the work gets to run.
> This situation is unlikely since we don't schedule the work for exiting
> tasks, but we cannot rule it out.

The implementation looks good to me. Peter, how does it look from your end ?

Thanks,

Mathieu

> 
> Fixes: 223baf9d17f2 ("sched: Fix performance regression introduced by mm_cid")
> Signed-off-by: Gabriele Monaco <gmonaco@redhat.com>
> ---
>   include/linux/mm_types.h | 26 ++++++++++++++
>   include/linux/sched.h    |  8 ++++-
>   kernel/rseq.c            |  2 ++
>   kernel/sched/core.c      | 75 ++++++++++++++++++++++++++--------------
>   kernel/sched/sched.h     |  6 ++--
>   5 files changed, 89 insertions(+), 28 deletions(-)
> 
> diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
> index 56d07edd01f91..e4ae9295508cf 100644
> --- a/include/linux/mm_types.h
> +++ b/include/linux/mm_types.h
> @@ -982,6 +982,10 @@ struct mm_struct {
>   		 * mm nr_cpus_allowed updates.
>   		 */
>   		raw_spinlock_t cpus_allowed_lock;
> +		/*
> +		 * @cid_work: Work item to run the mm_cid scan.
> +		 */
> +		struct work_struct cid_work;
>   #endif
>   #ifdef CONFIG_MMU
>   		atomic_long_t pgtables_bytes;	/* size of all page tables */
> @@ -1282,6 +1286,8 @@ enum mm_cid_state {
>   	MM_CID_LAZY_PUT = (1U << 31),
>   };
>   
> +extern void task_mm_cid_work(struct work_struct *work);
> +
>   static inline bool mm_cid_is_unset(int cid)
>   {
>   	return cid == MM_CID_UNSET;
> @@ -1354,12 +1360,14 @@ static inline int mm_alloc_cid_noprof(struct mm_struct *mm, struct task_struct *
>   	if (!mm->pcpu_cid)
>   		return -ENOMEM;
>   	mm_init_cid(mm, p);
> +	INIT_WORK(&mm->cid_work, task_mm_cid_work);
>   	return 0;
>   }
>   #define mm_alloc_cid(...)	alloc_hooks(mm_alloc_cid_noprof(__VA_ARGS__))
>   
>   static inline void mm_destroy_cid(struct mm_struct *mm)
>   {
> +	disable_work(&mm->cid_work);
>   	free_percpu(mm->pcpu_cid);
>   	mm->pcpu_cid = NULL;
>   }
> @@ -1381,6 +1389,16 @@ static inline void mm_set_cpus_allowed(struct mm_struct *mm, const struct cpumas
>   	WRITE_ONCE(mm->nr_cpus_allowed, cpumask_weight(mm_allowed));
>   	raw_spin_unlock(&mm->cpus_allowed_lock);
>   }
> +
> +static inline bool mm_cid_needs_scan(struct mm_struct *mm)
> +{
> +	return mm && !time_before(jiffies, READ_ONCE(mm->mm_cid_next_scan));
> +}
> +
> +static inline bool mm_cid_scan_pending(struct mm_struct *mm)
> +{
> +	return mm && work_pending(&mm->cid_work);
> +}
>   #else /* CONFIG_SCHED_MM_CID */
>   static inline void mm_init_cid(struct mm_struct *mm, struct task_struct *p) { }
>   static inline int mm_alloc_cid(struct mm_struct *mm, struct task_struct *p) { return 0; }
> @@ -1391,6 +1409,14 @@ static inline unsigned int mm_cid_size(void)
>   	return 0;
>   }
>   static inline void mm_set_cpus_allowed(struct mm_struct *mm, const struct cpumask *cpumask) { }
> +static inline bool mm_cid_needs_scan(struct mm_struct *mm)
> +{
> +	return false;
> +}
> +static inline bool mm_cid_scan_pending(struct mm_struct *mm)
> +{
> +	return false;
> +}
>   #endif /* CONFIG_SCHED_MM_CID */
>   
>   struct mmu_gather;
> diff --git a/include/linux/sched.h b/include/linux/sched.h
> index f96ac19828934..3ffdb96ef6b0a 100644
> --- a/include/linux/sched.h
> +++ b/include/linux/sched.h
> @@ -1424,7 +1424,7 @@ struct task_struct {
>   	int				last_mm_cid;	/* Most recent cid in mm */
>   	int				migrate_from_cpu;
>   	int				mm_cid_active;	/* Whether cid bitmap is active */
> -	struct callback_head		cid_work;
> +	unsigned long			last_cid_reset;	/* Time of last reset in jiffies */
>   #endif
>   
>   	struct tlbflush_unmap_batch	tlb_ubc;
> @@ -2281,4 +2281,10 @@ static __always_inline void alloc_tag_restore(struct alloc_tag *tag, struct allo
>   #define alloc_tag_restore(_tag, _old)		do {} while (0)
>   #endif
>   
> +#ifdef CONFIG_SCHED_MM_CID
> +extern void task_queue_mm_cid(struct task_struct *curr);
> +#else
> +static inline void task_queue_mm_cid(struct task_struct *curr) { }
> +#endif
> +
>   #endif
> diff --git a/kernel/rseq.c b/kernel/rseq.c
> index b7a1ec327e811..383db2ccad4d0 100644
> --- a/kernel/rseq.c
> +++ b/kernel/rseq.c
> @@ -441,6 +441,8 @@ void __rseq_handle_notify_resume(struct ksignal *ksig, struct pt_regs *regs)
>   	}
>   	if (unlikely(rseq_update_cpu_node_id(t)))
>   		goto error;
> +	if (mm_cid_needs_scan(t->mm))
> +		task_queue_mm_cid(t);
>   	return;
>   
>   error:
> diff --git a/kernel/sched/core.c b/kernel/sched/core.c
> index c81cf642dba05..02b18649e6a09 100644
> --- a/kernel/sched/core.c
> +++ b/kernel/sched/core.c
> @@ -10566,22 +10566,16 @@ static void sched_mm_cid_remote_clear_weight(struct mm_struct *mm, int cpu,
>   	sched_mm_cid_remote_clear(mm, pcpu_cid, cpu);
>   }
>   
> -static void task_mm_cid_work(struct callback_head *work)
> +void task_mm_cid_work(struct work_struct *work)
>   {
>   	unsigned long now = jiffies, old_scan, next_scan;
> -	struct task_struct *t = current;
>   	struct cpumask *cidmask;
> -	struct mm_struct *mm;
> +	struct mm_struct *mm = container_of(work, struct mm_struct, cid_work);
>   	int weight, cpu;
>   
> -	WARN_ON_ONCE(t != container_of(work, struct task_struct, cid_work));
> -
> -	work->next = work;	/* Prevent double-add */
> -	if (t->flags & PF_EXITING)
> -		return;
> -	mm = t->mm;
> -	if (!mm)
> -		return;
> +	/* We are the last user, process already terminated. */
> +	if (atomic_read(&mm->mm_count) == 1)
> +		goto out_drop;
>   	old_scan = READ_ONCE(mm->mm_cid_next_scan);
>   	next_scan = now + msecs_to_jiffies(MM_CID_SCAN_DELAY);
>   	if (!old_scan) {
> @@ -10594,9 +10588,9 @@ static void task_mm_cid_work(struct callback_head *work)
>   			old_scan = next_scan;
>   	}
>   	if (time_before(now, old_scan))
> -		return;
> +		goto out_drop;
>   	if (!try_cmpxchg(&mm->mm_cid_next_scan, &old_scan, next_scan))
> -		return;
> +		goto out_drop;
>   	cidmask = mm_cidmask(mm);
>   	/* Clear cids that were not recently used. */
>   	for_each_possible_cpu(cpu)
> @@ -10608,6 +10602,8 @@ static void task_mm_cid_work(struct callback_head *work)
>   	 */
>   	for_each_possible_cpu(cpu)
>   		sched_mm_cid_remote_clear_weight(mm, cpu, weight);
> +out_drop:
> +	mmdrop(mm);
>   }
>   
>   void init_sched_mm_cid(struct task_struct *t)
> @@ -10620,23 +10616,52 @@ void init_sched_mm_cid(struct task_struct *t)
>   		if (mm_users == 1)
>   			mm->mm_cid_next_scan = jiffies + msecs_to_jiffies(MM_CID_SCAN_DELAY);
>   	}
> -	t->cid_work.next = &t->cid_work;	/* Protect against double add */
> -	init_task_work(&t->cid_work, task_mm_cid_work);
>   }
>   
> -void task_tick_mm_cid(struct rq *rq, struct task_struct *curr)
> +void task_tick_mm_cid(struct rq *rq, struct task_struct *t)
>   {
> -	struct callback_head *work = &curr->cid_work;
> -	unsigned long now = jiffies;
> +	u64 rtime = t->se.sum_exec_runtime - t->se.prev_sum_exec_runtime;
>   
> -	if (!curr->mm || (curr->flags & (PF_EXITING | PF_KTHREAD)) ||
> -	    work->next != work)
> -		return;
> -	if (time_before(now, READ_ONCE(curr->mm->mm_cid_next_scan)))
> -		return;
> +	/*
> +	 * If a task is running unpreempted for a long time, it won't get its
> +	 * mm_cid compacted and won't update its mm_cid value after a
> +	 * compaction occurs.
> +	 * For such a task, this function does two things:
> +	 * A) trigger the mm_cid recompaction,
> +	 * B) trigger an update of the task's rseq->mm_cid field at some point
> +	 * after recompaction, so it can get a mm_cid value closer to 0.
> +	 * A change in the mm_cid triggers an rseq_preempt.
> +	 *
> +	 * A occurs only once after the scan time elapsed, until the next scan
> +	 * expires as well.
> +	 * B occurs once after the compaction work completes, that is when scan
> +	 * is no longer needed (it occurred for this mm) but the last rseq
> +	 * preempt was done before the last mm_cid scan.
> +	 */
> +	if (t->mm && rtime > RSEQ_UNPREEMPTED_THRESHOLD) {
> +		if (mm_cid_needs_scan(t->mm) && !mm_cid_scan_pending(t->mm))
> +			rseq_set_notify_resume(t);
> +		else if (time_after(jiffies, t->last_cid_reset +
> +				      msecs_to_jiffies(MM_CID_SCAN_DELAY))) {
> +			int old_cid = t->mm_cid;
> +
> +			if (!t->mm_cid_active)
> +				return;
> +			mm_cid_snapshot_time(rq, t->mm);
> +			mm_cid_put_lazy(t);
> +			t->last_mm_cid = t->mm_cid = mm_cid_get(rq, t, t->mm);
> +			if (old_cid != t->mm_cid)
> +				rseq_preempt(t);
> +		}
> +	}
> +}
>   
> -	/* No page allocation under rq lock */
> -	task_work_add(curr, work, TWA_RESUME);
> +/* Call only when curr is a user thread. */
> +void task_queue_mm_cid(struct task_struct *curr)
> +{
> +	/* Ensure the mm exists when we run. */
> +	mmgrab(curr->mm);
> +	queue_work(system_unbound_wq, &curr->mm->cid_work);
>   }
>   
>   void sched_mm_cid_exit_signals(struct task_struct *t)
> diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
> index 47972f34ea701..c0f8fd4c575c3 100644
> --- a/kernel/sched/sched.h
> +++ b/kernel/sched/sched.h
> @@ -3582,13 +3582,14 @@ extern const char *preempt_modes[];
>   
>   #define SCHED_MM_CID_PERIOD_NS	(100ULL * 1000000)	/* 100ms */
>   #define MM_CID_SCAN_DELAY	100			/* 100ms */
> +#define RSEQ_UNPREEMPTED_THRESHOLD	SCHED_MM_CID_PERIOD_NS
>   
>   extern raw_spinlock_t cid_lock;
>   extern int use_cid_lock;
>   
>   extern void sched_mm_cid_migrate_from(struct task_struct *t);
>   extern void sched_mm_cid_migrate_to(struct rq *dst_rq, struct task_struct *t);
> -extern void task_tick_mm_cid(struct rq *rq, struct task_struct *curr);
> +extern void task_tick_mm_cid(struct rq *rq, struct task_struct *t);
>   extern void init_sched_mm_cid(struct task_struct *t);
>   
>   static inline void __mm_cid_put(struct mm_struct *mm, int cid)
> @@ -3798,6 +3799,7 @@ static inline int mm_cid_get(struct rq *rq, struct task_struct *t,
>   	cid = __mm_cid_get(rq, t, mm);
>   	__this_cpu_write(pcpu_cid->cid, cid);
>   	__this_cpu_write(pcpu_cid->recent_cid, cid);
> +	t->last_cid_reset = jiffies;
>   
>   	return cid;
>   }
> @@ -3857,7 +3859,7 @@ static inline void switch_mm_cid(struct rq *rq,
>   static inline void switch_mm_cid(struct rq *rq, struct task_struct *prev, struct task_struct *next) { }
>   static inline void sched_mm_cid_migrate_from(struct task_struct *t) { }
>   static inline void sched_mm_cid_migrate_to(struct rq *dst_rq, struct task_struct *t) { }
> -static inline void task_tick_mm_cid(struct rq *rq, struct task_struct *curr) { }
> +static inline void task_tick_mm_cid(struct rq *rq, struct task_struct *t) { }
>   static inline void init_sched_mm_cid(struct task_struct *t) { }
>   #endif /* !CONFIG_SCHED_MM_CID */
>   


-- 
Mathieu Desnoyers
EfficiOS Inc.
https://www.efficios.com

Re: [PATCH v13 2/3] sched: Move task_mm_cid_work to mm work_struct

[Gabriele Monaco]

2025-04-14T15:28:34Z Mathieu Desnoyers <mathieu.desnoyers@efficios.com>:

> On 2025-04-14 08:36, Gabriele Monaco wrote:
>> Currently, the task_mm_cid_work function is called in a task work
>> triggered by a scheduler tick to frequently compact the mm_cids of each
>> process. This can delay the execution of the corresponding thread for
>> the entire duration of the function, negatively affecting the response
>> in case of real time tasks. In practice, we observe task_mm_cid_work
>> increasing the latency of 30-35us on a 128 cores system, this order of
>> magnitude is meaningful under PREEMPT_RT.
>> Run the task_mm_cid_work in a new work_struct connected to the
>> mm_struct rather than in the task context before returning to
>> userspace.
>> This work_struct is initialised with the mm and disabled before freeing
>> it. The queuing of the work happens while returning to userspace in
>> __rseq_handle_notify_resume, maintaining the checks to avoid running
>> more frequently than MM_CID_SCAN_DELAY.
>> To make sure this happens predictably also on long running tasks, we
>> trigger a call to __rseq_handle_notify_resume also from the scheduler
>> tick if the runtime exceeded a 100ms threshold.
>> The main advantage of this change is that the function can be offloaded
>> to a different CPU and even preempted by RT tasks.
>> Moreover, this new behaviour is more predictable with periodic tasks
>> with short runtime, which may rarely run during a scheduler tick.
>> Now, the work is always scheduled when the task returns to userspace.
>> The work is disabled during mmdrop, since the function cannot sleep in
>> all kernel configurations, we cannot wait for possibly running work
>> items to terminate. We make sure the mm is valid in case the task is
>> terminating by reserving it with mmgrab/mmdrop, returning prematurely if
>> we are really the last user while the work gets to run.
>> This situation is unlikely since we don't schedule the work for exiting
>> tasks, but we cannot rule it out.
>
> The implementation looks good to me. Peter, how does it look from your end ?
>

Peter, what do you think about this version? Can we bring it in?

Thanks,
Gabriele

> Thanks,
>
> Mathieu
>
>> Fixes: 223baf9d17f2 ("sched: Fix performance regression introduced by mm_cid")
>> Signed-off-by: Gabriele Monaco <gmonaco@redhat.com>
>> ---
>>   include/linux/mm_types.h | 26 ++++++++++++++
>>   include/linux/sched.h    |  8 ++++-
>>   kernel/rseq.c            |  2 ++
>>   kernel/sched/core.c      | 75 ++++++++++++++++++++++++++--------------
>>   kernel/sched/sched.h     |  6 ++--
>>   5 files changed, 89 insertions(+), 28 deletions(-)
>> diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
>> index 56d07edd01f91..e4ae9295508cf 100644
>> --- a/include/linux/mm_types.h
>> +++ b/include/linux/mm_types.h
>> @@ -982,6 +982,10 @@ struct mm_struct {
>>          * mm nr_cpus_allowed updates.
>>          */
>>         raw_spinlock_t cpus_allowed_lock;
>> +       /*
>> +        * @cid_work: Work item to run the mm_cid scan.
>> +        */
>> +       struct work_struct cid_work;
>>   #endif
>>   #ifdef CONFIG_MMU
>>         atomic_long_t pgtables_bytes;   /* size of all page tables */
>> @@ -1282,6 +1286,8 @@ enum mm_cid_state {
>>     MM_CID_LAZY_PUT = (1U << 31),
>>   };
>>   +extern void task_mm_cid_work(struct work_struct *work);
>> +
>>   static inline bool mm_cid_is_unset(int cid)
>>   {
>>     return cid == MM_CID_UNSET;
>> @@ -1354,12 +1360,14 @@ static inline int mm_alloc_cid_noprof(struct mm_struct *mm, struct task_struct *
>>     if (!mm->pcpu_cid)
>>         return -ENOMEM;
>>     mm_init_cid(mm, p);
>> +   INIT_WORK(&mm->cid_work, task_mm_cid_work);
>>     return 0;
>>   }
>>   #define mm_alloc_cid(...) alloc_hooks(mm_alloc_cid_noprof(__VA_ARGS__))
>>     static inline void mm_destroy_cid(struct mm_struct *mm)
>>   {
>> +   disable_work(&mm->cid_work);
>>     free_percpu(mm->pcpu_cid);
>>     mm->pcpu_cid = NULL;
>>   }
>> @@ -1381,6 +1389,16 @@ static inline void mm_set_cpus_allowed(struct mm_struct *mm, const struct cpumas
>>     WRITE_ONCE(mm->nr_cpus_allowed, cpumask_weight(mm_allowed));
>>     raw_spin_unlock(&mm->cpus_allowed_lock);
>>   }
>> +
>> +static inline bool mm_cid_needs_scan(struct mm_struct *mm)
>> +{
>> +   return mm && !time_before(jiffies, READ_ONCE(mm->mm_cid_next_scan));
>> +}
>> +
>> +static inline bool mm_cid_scan_pending(struct mm_struct *mm)
>> +{
>> +   return mm && work_pending(&mm->cid_work);
>> +}
>>   #else /* CONFIG_SCHED_MM_CID */
>>   static inline void mm_init_cid(struct mm_struct *mm, struct task_struct *p) { }
>>   static inline int mm_alloc_cid(struct mm_struct *mm, struct task_struct *p) { return 0; }
>> @@ -1391,6 +1409,14 @@ static inline unsigned int mm_cid_size(void)
>>     return 0;
>>   }
>>   static inline void mm_set_cpus_allowed(struct mm_struct *mm, const struct cpumask *cpumask) { }
>> +static inline bool mm_cid_needs_scan(struct mm_struct *mm)
>> +{
>> +   return false;
>> +}
>> +static inline bool mm_cid_scan_pending(struct mm_struct *mm)
>> +{
>> +   return false;
>> +}
>>   #endif /* CONFIG_SCHED_MM_CID */
>>     struct mmu_gather;
>> diff --git a/include/linux/sched.h b/include/linux/sched.h
>> index f96ac19828934..3ffdb96ef6b0a 100644
>> --- a/include/linux/sched.h
>> +++ b/include/linux/sched.h
>> @@ -1424,7 +1424,7 @@ struct task_struct {
>>     int             last_mm_cid;    /* Most recent cid in mm */
>>     int             migrate_from_cpu;
>>     int             mm_cid_active;  /* Whether cid bitmap is active */
>> -   struct callback_head        cid_work;
>> +   unsigned long           last_cid_reset; /* Time of last reset in jiffies */
>>   #endif
>>         struct tlbflush_unmap_batch tlb_ubc;
>> @@ -2281,4 +2281,10 @@ static __always_inline void alloc_tag_restore(struct alloc_tag *tag, struct allo
>>   #define alloc_tag_restore(_tag, _old)     do {} while (0)
>>   #endif
>>   +#ifdef CONFIG_SCHED_MM_CID
>> +extern void task_queue_mm_cid(struct task_struct *curr);
>> +#else
>> +static inline void task_queue_mm_cid(struct task_struct *curr) { }
>> +#endif
>> +
>>   #endif
>> diff --git a/kernel/rseq.c b/kernel/rseq.c
>> index b7a1ec327e811..383db2ccad4d0 100644
>> --- a/kernel/rseq.c
>> +++ b/kernel/rseq.c
>> @@ -441,6 +441,8 @@ void __rseq_handle_notify_resume(struct ksignal *ksig, struct pt_regs *regs)
>>     }
>>     if (unlikely(rseq_update_cpu_node_id(t)))
>>         goto error;
>> +   if (mm_cid_needs_scan(t->mm))
>> +       task_queue_mm_cid(t);
>>     return;
>>     error:
>> diff --git a/kernel/sched/core.c b/kernel/sched/core.c
>> index c81cf642dba05..02b18649e6a09 100644
>> --- a/kernel/sched/core.c
>> +++ b/kernel/sched/core.c
>> @@ -10566,22 +10566,16 @@ static void sched_mm_cid_remote_clear_weight(struct mm_struct *mm, int cpu,
>>     sched_mm_cid_remote_clear(mm, pcpu_cid, cpu);
>>   }
>>   -static void task_mm_cid_work(struct callback_head *work)
>> +void task_mm_cid_work(struct work_struct *work)
>>   {
>>     unsigned long now = jiffies, old_scan, next_scan;
>> -   struct task_struct *t = current;
>>     struct cpumask *cidmask;
>> -   struct mm_struct *mm;
>> +   struct mm_struct *mm = container_of(work, struct mm_struct, cid_work);
>>     int weight, cpu;
>>   - WARN_ON_ONCE(t != container_of(work, struct task_struct, cid_work));
>> -
>> -   work->next = work;  /* Prevent double-add */
>> -   if (t->flags & PF_EXITING)
>> -       return;
>> -   mm = t->mm;
>> -   if (!mm)
>> -       return;
>> +   /* We are the last user, process already terminated. */
>> +   if (atomic_read(&mm->mm_count) == 1)
>> +       goto out_drop;
>>     old_scan = READ_ONCE(mm->mm_cid_next_scan);
>>     next_scan = now + msecs_to_jiffies(MM_CID_SCAN_DELAY);
>>     if (!old_scan) {
>> @@ -10594,9 +10588,9 @@ static void task_mm_cid_work(struct callback_head *work)
>>             old_scan = next_scan;
>>     }
>>     if (time_before(now, old_scan))
>> -       return;
>> +       goto out_drop;
>>     if (!try_cmpxchg(&mm->mm_cid_next_scan, &old_scan, next_scan))
>> -       return;
>> +       goto out_drop;
>>     cidmask = mm_cidmask(mm);
>>     /* Clear cids that were not recently used. */
>>     for_each_possible_cpu(cpu)
>> @@ -10608,6 +10602,8 @@ static void task_mm_cid_work(struct callback_head *work)
>>      */
>>     for_each_possible_cpu(cpu)
>>         sched_mm_cid_remote_clear_weight(mm, cpu, weight);
>> +out_drop:
>> +   mmdrop(mm);
>>   }
>>     void init_sched_mm_cid(struct task_struct *t)
>> @@ -10620,23 +10616,52 @@ void init_sched_mm_cid(struct task_struct *t)
>>         if (mm_users == 1)
>>             mm->mm_cid_next_scan = jiffies + msecs_to_jiffies(MM_CID_SCAN_DELAY);
>>     }
>> -   t->cid_work.next = &t->cid_work;    /* Protect against double add */
>> -   init_task_work(&t->cid_work, task_mm_cid_work);
>>   }
>>   -void task_tick_mm_cid(struct rq *rq, struct task_struct *curr)
>> +void task_tick_mm_cid(struct rq *rq, struct task_struct *t)
>>   {
>> -   struct callback_head *work = &curr->cid_work;
>> -   unsigned long now = jiffies;
>> +   u64 rtime = t->se.sum_exec_runtime - t->se.prev_sum_exec_runtime;
>>   - if (!curr->mm || (curr->flags & (PF_EXITING | PF_KTHREAD)) ||
>> -       work->next != work)
>> -       return;
>> -   if (time_before(now, READ_ONCE(curr->mm->mm_cid_next_scan)))
>> -       return;
>> +   /*
>> +    * If a task is running unpreempted for a long time, it won't get its
>> +    * mm_cid compacted and won't update its mm_cid value after a
>> +    * compaction occurs.
>> +    * For such a task, this function does two things:
>> +    * A) trigger the mm_cid recompaction,
>> +    * B) trigger an update of the task's rseq->mm_cid field at some point
>> +    * after recompaction, so it can get a mm_cid value closer to 0.
>> +    * A change in the mm_cid triggers an rseq_preempt.
>> +    *
>> +    * A occurs only once after the scan time elapsed, until the next scan
>> +    * expires as well.
>> +    * B occurs once after the compaction work completes, that is when scan
>> +    * is no longer needed (it occurred for this mm) but the last rseq
>> +    * preempt was done before the last mm_cid scan.
>> +    */
>> +   if (t->mm && rtime > RSEQ_UNPREEMPTED_THRESHOLD) {
>> +       if (mm_cid_needs_scan(t->mm) && !mm_cid_scan_pending(t->mm))
>> +           rseq_set_notify_resume(t);
>> +       else if (time_after(jiffies, t->last_cid_reset +
>> +                     msecs_to_jiffies(MM_CID_SCAN_DELAY))) {
>> +           int old_cid = t->mm_cid;
>> +
>> +           if (!t->mm_cid_active)
>> +               return;
>> +           mm_cid_snapshot_time(rq, t->mm);
>> +           mm_cid_put_lazy(t);
>> +           t->last_mm_cid = t->mm_cid = mm_cid_get(rq, t, t->mm);
>> +           if (old_cid != t->mm_cid)
>> +               rseq_preempt(t);
>> +       }
>> +   }
>> +}
>>   - /* No page allocation under rq lock */
>> -   task_work_add(curr, work, TWA_RESUME);
>> +/* Call only when curr is a user thread. */
>> +void task_queue_mm_cid(struct task_struct *curr)
>> +{
>> +   /* Ensure the mm exists when we run. */
>> +   mmgrab(curr->mm);
>> +   queue_work(system_unbound_wq, &curr->mm->cid_work);
>>   }
>>     void sched_mm_cid_exit_signals(struct task_struct *t)
>> diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
>> index 47972f34ea701..c0f8fd4c575c3 100644
>> --- a/kernel/sched/sched.h
>> +++ b/kernel/sched/sched.h
>> @@ -3582,13 +3582,14 @@ extern const char *preempt_modes[];
>>     #define SCHED_MM_CID_PERIOD_NS  (100ULL * 1000000)  /* 100ms */
>>   #define MM_CID_SCAN_DELAY 100         /* 100ms */
>> +#define RSEQ_UNPREEMPTED_THRESHOLD SCHED_MM_CID_PERIOD_NS
>>     extern raw_spinlock_t cid_lock;
>>   extern int use_cid_lock;
>>     extern void sched_mm_cid_migrate_from(struct task_struct *t);
>>   extern void sched_mm_cid_migrate_to(struct rq *dst_rq, struct task_struct *t);
>> -extern void task_tick_mm_cid(struct rq *rq, struct task_struct *curr);
>> +extern void task_tick_mm_cid(struct rq *rq, struct task_struct *t);
>>   extern void init_sched_mm_cid(struct task_struct *t);
>>     static inline void __mm_cid_put(struct mm_struct *mm, int cid)
>> @@ -3798,6 +3799,7 @@ static inline int mm_cid_get(struct rq *rq, struct task_struct *t,
>>     cid = __mm_cid_get(rq, t, mm);
>>     __this_cpu_write(pcpu_cid->cid, cid);
>>     __this_cpu_write(pcpu_cid->recent_cid, cid);
>> +   t->last_cid_reset = jiffies;
>>         return cid;
>>   }
>> @@ -3857,7 +3859,7 @@ static inline void switch_mm_cid(struct rq *rq,
>>   static inline void switch_mm_cid(struct rq *rq, struct task_struct *prev, struct task_struct *next) { }
>>   static inline void sched_mm_cid_migrate_from(struct task_struct *t) { }
>>   static inline void sched_mm_cid_migrate_to(struct rq *dst_rq, struct task_struct *t) { }
>> -static inline void task_tick_mm_cid(struct rq *rq, struct task_struct *curr) { }
>> +static inline void task_tick_mm_cid(struct rq *rq, struct task_struct *t) { }
>>   static inline void init_sched_mm_cid(struct task_struct *t) { }
>>   #endif /* !CONFIG_SCHED_MM_CID */
>>  
>
>
> --
> Mathieu Desnoyers
> EfficiOS Inc.
> https://www.efficios.com

Re: [PATCH v13 2/3] sched: Move task_mm_cid_work to mm work_struct

[Gabriele Monaco]

On Mon, 2025-04-14 at 11:28 -0400, Mathieu Desnoyers wrote:
> On 2025-04-14 08:36, Gabriele Monaco wrote:
> > Currently, the task_mm_cid_work function is called in a task work
> > triggered by a scheduler tick to frequently compact the mm_cids of
> > each
> > process. This can delay the execution of the corresponding thread
> > for
> > the entire duration of the function, negatively affecting the
> > response
> > in case of real time tasks. In practice, we observe
> > task_mm_cid_work
> > increasing the latency of 30-35us on a 128 cores system, this order
> > of
> > magnitude is meaningful under PREEMPT_RT.
> > 
> > Run the task_mm_cid_work in a new work_struct connected to the
> > mm_struct rather than in the task context before returning to
> > userspace.
> > 
> > This work_struct is initialised with the mm and disabled before
> > freeing
> > it. The queuing of the work happens while returning to userspace in
> > __rseq_handle_notify_resume, maintaining the checks to avoid
> > running
> > more frequently than MM_CID_SCAN_DELAY.
> > To make sure this happens predictably also on long running tasks,
> > we
> > trigger a call to __rseq_handle_notify_resume also from the
> > scheduler
> > tick if the runtime exceeded a 100ms threshold.
> > 
> > The main advantage of this change is that the function can be
> > offloaded
> > to a different CPU and even preempted by RT tasks.
> > 
> > Moreover, this new behaviour is more predictable with periodic
> > tasks
> > with short runtime, which may rarely run during a scheduler tick.
> > Now, the work is always scheduled when the task returns to
> > userspace.
> > 
> > The work is disabled during mmdrop, since the function cannot sleep
> > in
> > all kernel configurations, we cannot wait for possibly running work
> > items to terminate. We make sure the mm is valid in case the task
> > is
> > terminating by reserving it with mmgrab/mmdrop, returning
> > prematurely if
> > we are really the last user while the work gets to run.
> > This situation is unlikely since we don't schedule the work for
> > exiting
> > tasks, but we cannot rule it out.
> 
> The implementation looks good to me. Peter, how does it look from
> your end ?
> 
> Thanks,
> 
> Mathieu
> 
> 

Gentle ping.

Peter, did you have some time to have a look at this patch?

Thanks,
Gabriele

Re: [PATCH v13 2/3] sched: Move task_mm_cid_work to mm work_struct

[Gabriele Monaco]

On Mon, 2025-04-14 at 11:28 -0400, Mathieu Desnoyers wrote:
> On 2025-04-14 08:36, Gabriele Monaco wrote:
> > Currently, the task_mm_cid_work function is called in a task work
> > triggered by a scheduler tick to frequently compact the mm_cids of
> > each
> > process. This can delay the execution of the corresponding thread
> > for
> > the entire duration of the function, negatively affecting the
> > response
> > in case of real time tasks. In practice, we observe
> > task_mm_cid_work
> > increasing the latency of 30-35us on a 128 cores system, this order
> > of
> > magnitude is meaningful under PREEMPT_RT.
> > 
> > Run the task_mm_cid_work in a new work_struct connected to the
> > mm_struct rather than in the task context before returning to
> > userspace.
> > 
> > This work_struct is initialised with the mm and disabled before
> > freeing
> > it. The queuing of the work happens while returning to userspace in
> > __rseq_handle_notify_resume, maintaining the checks to avoid
> > running
> > more frequently than MM_CID_SCAN_DELAY.
> > To make sure this happens predictably also on long running tasks,
> > we
> > trigger a call to __rseq_handle_notify_resume also from the
> > scheduler
> > tick if the runtime exceeded a 100ms threshold.
> > 
> > The main advantage of this change is that the function can be
> > offloaded
> > to a different CPU and even preempted by RT tasks.
> > 
> > Moreover, this new behaviour is more predictable with periodic
> > tasks
> > with short runtime, which may rarely run during a scheduler tick.
> > Now, the work is always scheduled when the task returns to
> > userspace.
> > 
> > The work is disabled during mmdrop, since the function cannot sleep
> > in
> > all kernel configurations, we cannot wait for possibly running work
> > items to terminate. We make sure the mm is valid in case the task
> > is
> > terminating by reserving it with mmgrab/mmdrop, returning
> > prematurely if
> > we are really the last user while the work gets to run.
> > This situation is unlikely since we don't schedule the work for
> > exiting
> > tasks, but we cannot rule it out.
> 
> The implementation looks good to me. Peter, how does it look from
> your end ?
> 

Peter, Ingo, this could we bring this series in?

Thanks,
Gabriele