ksm: count ksm merging pages for each process

Some applications or containers want to use KSM by calling madvise() to
advise areas of address space to be MERGEABLE.  But they may not know
which applications are more likely to cause real merges in the
deployment.  If this patch is applied, it helps them know their
corresponding number of merged pages, and then optimize their app code.

As current KSM only counts the number of KSM merging pages(e.g. 
ksm_pages_sharing and ksm_pages_shared) of the whole system, we cannot see
the more fine-grained KSM merging, for the upper application optimization,
the merging area cannot be set easily according to the KSM page merging
probability of each process.  Therefore, it is necessary to add extra
statistical means so that the upper level users can know the detailed KSM
merging information of each process.

We add a new proc file named as ksm_merging_pages under /proc/<pid>/ to
indicate the involved ksm merging pages of this process.

[[email protected]: fix comment typo, remove BUG_ON()s]
Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: xu xin <[email protected]>
Reported-by: kernel test robot <[email protected]>
Reviewed-by: Yang Yang <[email protected]>
Reviewed-by: Ran Xiaokai <[email protected]>
Reported-by: Zeal Robot <[email protected]>
Cc: Kees Cook <[email protected]>
Cc: Alexey Dobriyan <[email protected]>
Cc: Stephen Rothwell <[email protected]>
Cc: Ohhoon Kwon <[email protected]>
Cc: Matthew Wilcox (Oracle) <[email protected]>
Cc: Stephen Brennan <[email protected]>
Cc: Vlastimil Babka <[email protected]>
Cc: Feng Tang <[email protected]>
Cc: Yang Yang <[email protected]>
Cc: Ran Xiaokai <[email protected]>
Cc: Zeal Robot <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>

fs/proc/base.c

@@ -3154,6 +3154,22 @@ static int proc_pid_patch_state(struct seq_file *m, struct pid_namespace *ns,
 }
 #endif /* CONFIG_LIVEPATCH */
 
+#ifdef CONFIG_KSM
+static int proc_pid_ksm_merging_pages(struct seq_file *m, struct pid_namespace *ns,
+				struct pid *pid, struct task_struct *task)
+{
+	struct mm_struct *mm;
+
+	mm = get_task_mm(task);
+	if (mm) {
+		seq_printf(m, "%lu\n", mm->ksm_merging_pages);
+		mmput(mm);
+	}
+
+	return 0;
+}
+#endif /* CONFIG_KSM */
+
 #ifdef CONFIG_STACKLEAK_METRICS
 static int proc_stack_depth(struct seq_file *m, struct pid_namespace *ns,
 				struct pid *pid, struct task_struct *task)
@@ -3285,6 +3301,9 @@ static const struct pid_entry tgid_base_stuff[] = {
 #ifdef CONFIG_SECCOMP_CACHE_DEBUG
 	ONE("seccomp_cache", S_IRUSR, proc_pid_seccomp_cache),
 #endif
+#ifdef CONFIG_KSM
+	ONE("ksm_merging_pages",  S_IRUSR, proc_pid_ksm_merging_pages),
+#endif
 };
 
 static int proc_tgid_base_readdir(struct file *file, struct dir_context *ctx)
@@ -3618,6 +3637,9 @@ static const struct pid_entry tid_base_stuff[] = {
 #ifdef CONFIG_SECCOMP_CACHE_DEBUG
 	ONE("seccomp_cache", S_IRUSR, proc_pid_seccomp_cache),
 #endif
+#ifdef CONFIG_KSM
+	ONE("ksm_merging_pages",  S_IRUSR, proc_pid_ksm_merging_pages),
+#endif
 };
 
 static int proc_tid_base_readdir(struct file *file, struct dir_context *ctx)

include/linux/mm_types.h

@@ -654,6 +654,13 @@ struct mm_struct {
 
 #ifdef CONFIG_IOMMU_SVA
 		u32 pasid;
+#endif
+#ifdef CONFIG_KSM
+		/*
+		 * Represent how many pages of this process are involved in KSM
+		 * merging.
+		 */
+		unsigned long ksm_merging_pages;
 #endif
 	} __randomize_layout;
 

mm/ksm.c

@@ -638,6 +638,9 @@ static void remove_node_from_stable_tree(struct stable_node *stable_node)
 			ksm_pages_sharing--;
 		else
 			ksm_pages_shared--;
+
+		rmap_item->mm->ksm_merging_pages--;
+
 		VM_BUG_ON(stable_node->rmap_hlist_len <= 0);
 		stable_node->rmap_hlist_len--;
 		put_anon_vma(rmap_item->anon_vma);
@@ -785,6 +788,9 @@ static void remove_rmap_item_from_tree(struct rmap_item *rmap_item)
 			ksm_pages_sharing--;
 		else
 			ksm_pages_shared--;
+
+		rmap_item->mm->ksm_merging_pages--;
+
 		VM_BUG_ON(stable_node->rmap_hlist_len <= 0);
 		stable_node->rmap_hlist_len--;
 
@@ -2007,6 +2013,8 @@ static void stable_tree_append(struct rmap_item *rmap_item,
 		ksm_pages_sharing++;
 	else
 		ksm_pages_shared++;
+
+	rmap_item->mm->ksm_merging_pages++;
 }
 
 /*