Re: [RFC PATCH V1 1/1] sched/numa: Enhance vma scanning logic

[Mel Gorman]

On Wed, Jan 25, 2023 at 12:48:16AM +0530, Raghavendra K T wrote:
> looks like we have to additionally handle numab initialization in
> vm_area_dup() code path. something like below fixed it (copied pasted
> from tty):
> 

Yep, it wasn't even boot tested. Better approach is something like this,
still not actually tested

 include/linux/mm.h       |  9 +++++++++
 include/linux/mm_types.h |  7 +++++++
 kernel/fork.c            |  2 ++
 kernel/sched/fair.c      | 17 +++++++++++++++++
 4 files changed, 35 insertions(+)

diff --git a/include/linux/mm.h b/include/linux/mm.h
index 8f857163ac89..481f90dc1983 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -612,6 +612,14 @@ struct vm_operations_struct {
 					  unsigned long addr);
 };
 
+#ifdef CONFIG_NUMA_BALANCING
+#define vma_numab_init(vma) do { (vma)->numab = NULL; } while (0)
+#define vma_numab_free(vma) do { kfree((vma)->numab); } while (0)
+#else
+static inline void vma_numab_init(struct vm_area_struct *vma) {}
+static inline void vma_numab_free(struct vm_area_struct *vma) {}
+#endif /* CONFIG_NUMA_BALANCING */
+
 static inline void vma_init(struct vm_area_struct *vma, struct mm_struct *mm)
 {
 	static const struct vm_operations_struct dummy_vm_ops = {};
@@ -620,6 +628,7 @@ static inline void vma_init(struct vm_area_struct *vma, struct mm_struct *mm)
 	vma->vm_mm = mm;
 	vma->vm_ops = &dummy_vm_ops;
 	INIT_LIST_HEAD(&vma->anon_vma_chain);
+	vma_numab_init(vma);
 }
 
 static inline void vma_set_anonymous(struct vm_area_struct *vma)
diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
index 9757067c3053..43ce363d5124 100644
--- a/include/linux/mm_types.h
+++ b/include/linux/mm_types.h
@@ -526,6 +526,10 @@ struct anon_vma_name {
 	char name[];
 };
 
+struct vma_numab {
+	unsigned long next_scan;
+};
+
 /*
  * This struct describes a virtual memory area. There is one of these
  * per VM-area/task. A VM area is any part of the process virtual memory
@@ -593,6 +597,9 @@ struct vm_area_struct {
 #endif
 #ifdef CONFIG_NUMA
 	struct mempolicy *vm_policy;	/* NUMA policy for the VMA */
+#endif
+#ifdef CONFIG_NUMA_BALANCING
+	struct vma_numab *numab;	/* NUMA Balancing state */
 #endif
 	struct vm_userfaultfd_ctx vm_userfaultfd_ctx;
 } __randomize_layout;
diff --git a/kernel/fork.c b/kernel/fork.c
index 9f7fe3541897..5a2e8c3cc410 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -474,6 +474,7 @@ struct vm_area_struct *vm_area_dup(struct vm_area_struct *orig)
 		 */
 		*new = data_race(*orig);
 		INIT_LIST_HEAD(&new->anon_vma_chain);
+		vma_numab_init(new);
 		dup_anon_vma_name(orig, new);
 	}
 	return new;
@@ -481,6 +482,7 @@ struct vm_area_struct *vm_area_dup(struct vm_area_struct *orig)
 
 void vm_area_free(struct vm_area_struct *vma)
 {
+	vma_numab_free(vma);
 	free_anon_vma_name(vma);
 	kmem_cache_free(vm_area_cachep, vma);
 }
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index c36aa54ae071..6a1cffdfc76b 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -3027,6 +3027,23 @@ static void task_numa_work(struct callback_head *work)
 		if (!vma_is_accessible(vma))
 			continue;
 
+		/* Initialise new per-VMA NUMAB state. */
+		if (!vma->numab) {
+			vma->numab = kzalloc(sizeof(struct vma_numab), GFP_KERNEL);
+			if (!vma->numab)
+				continue;
+
+			vma->numab->next_scan = now +
+				msecs_to_jiffies(sysctl_numa_balancing_scan_delay);
+		}
+
+		/*
+		 * After the first scan is complete, delay the balancing scan
+		 * for new VMAs.
+		 */
+		if (mm->numa_scan_seq && time_before(jiffies, vma->numab->next_scan))
+			continue;
+
 		do {
 			start = max(start, vma->vm_start);
 			end = ALIGN(start + (pages << PAGE_SHIFT), HPAGE_SIZE);