mm: consolidate page table accounting

Currently, we account page tables separately for each page table level,
but that's redundant -- we only make use of total memory allocated to
page tables for oom_badness calculation.  We also provide the
information to userspace, but it has dubious value there too.

This patch switches page table accounting to single counter.

mm->pgtables_bytes is now used to account all page table levels.  We use
bytes, because page table size for different levels of page table tree
may be different.

The change has user-visible effect: we don't have VmPMD and VmPUD
reported in /proc/[pid]/status.  Not sure if anybody uses them.  (As
alternative, we can always report 0 kB for them.)

OOM-killer report is also slightly changed: we now report pgtables_bytes
instead of nr_ptes, nr_pmd, nr_puds.

Apart from reducing number of counters per-mm, the benefit is that we
now calculate oom_badness() more correctly for machines which have
different size of page tables depending on level or where page tables
are less than a page in size.

The only downside can be debuggability because we do not know which page
table level could leak.  But I do not remember many bugs that would be
caught by separate counters so I wouldn't lose sleep over this.

[akpm@linux-foundation.org: fix mm/huge_memory.c]
Link: http://lkml.kernel.org/r/20171006100651.44742-2-kirill.shutemov@linux.intel.com
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Acked-by: Michal Hocko <mhocko@suse.com>
[kirill.shutemov@linux.intel.com: fix build]
  Link: http://lkml.kernel.org/r/20171016150113.ikfxy3e7zzfvsr4w@black.fi.intel.com
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

9 files changed, 32 insertions, 93 deletions

diff --git a/Documentation/filesystems/proc.txt b/Documentation/filesystems/proc.txt
index adba21b5ada7..ec571b9bb18a 100644
--- a/Documentation/filesystems/proc.txt
+++ b/Documentation/filesystems/proc.txt
@@ -250,7 +250,6 @@ Table 1-2: Contents of the status files (as of 4.8)
  VmExe                       size of text segment
  VmLib                       size of shared library code
  VmPTE                       size of page table entries
- VmPMD                       size of second level page tables
  VmSwap                      amount of swap used by anonymous private data
                              (shmem swap usage is not included)
  HugetlbPages                size of hugetlb memory portions
diff --git a/Documentation/sysctl/vm.txt b/Documentation/sysctl/vm.txt
index 2729e8db9492..3e579740b49f 100644
--- a/Documentation/sysctl/vm.txt
+++ b/Documentation/sysctl/vm.txt
@@ -629,10 +629,10 @@ oom_dump_tasks
 
 Enables a system-wide task dump (excluding kernel threads) to be produced
 when the kernel performs an OOM-killing and includes such information as
-pid, uid, tgid, vm size, rss, nr_ptes, nr_pmds, nr_puds, swapents,
-oom_score_adj score, and name.  This is helpful to determine why the OOM
-killer was invoked, to identify the rogue task that caused it, and to
-determine why the OOM killer chose the task it did to kill.
+pid, uid, tgid, vm size, rss, pgtables_bytes, swapents, oom_score_adj
+score, and name.  This is helpful to determine why the OOM killer was
+invoked, to identify the rogue task that caused it, and to determine why
+the OOM killer chose the task it did to kill.
 
 If this is set to zero, this information is suppressed.  On very
 large systems with thousands of tasks it may not be feasible to dump
diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c
index 9bd2a0294ac1..875231c36cb3 100644
--- a/fs/proc/task_mmu.c
+++ b/fs/proc/task_mmu.c
@@ -26,7 +26,7 @@
 
 void task_mem(struct seq_file *m, struct mm_struct *mm)
 {
-	unsigned long text, lib, swap, ptes, pmds, puds, anon, file, shmem;
+	unsigned long text, lib, swap, anon, file, shmem;
 	unsigned long hiwater_vm, total_vm, hiwater_rss, total_rss;
 
 	anon = get_mm_counter(mm, MM_ANONPAGES);
@@ -50,9 +50,6 @@ void task_mem(struct seq_file *m, struct mm_struct *mm)
 	text = (PAGE_ALIGN(mm->end_code) - (mm->start_code & PAGE_MASK)) >> 10;
 	lib = (mm->exec_vm << (PAGE_SHIFT-10)) - text;
 	swap = get_mm_counter(mm, MM_SWAPENTS);
-	ptes = PTRS_PER_PTE * sizeof(pte_t) * mm_nr_ptes(mm);
-	pmds = PTRS_PER_PMD * sizeof(pmd_t) * mm_nr_pmds(mm);
-	puds = PTRS_PER_PUD * sizeof(pud_t) * mm_nr_puds(mm);
 	seq_printf(m,
 		"VmPeak:\t%8lu kB\n"
 		"VmSize:\t%8lu kB\n"
@@ -68,8 +65,6 @@ void task_mem(struct seq_file *m, struct mm_struct *mm)
 		"VmExe:\t%8lu kB\n"
 		"VmLib:\t%8lu kB\n"
 		"VmPTE:\t%8lu kB\n"
-		"VmPMD:\t%8lu kB\n"
-		"VmPUD:\t%8lu kB\n"
 		"VmSwap:\t%8lu kB\n",
 		hiwater_vm << (PAGE_SHIFT-10),
 		total_vm << (PAGE_SHIFT-10),
@@ -82,9 +77,7 @@ void task_mem(struct seq_file *m, struct mm_struct *mm)
 		shmem << (PAGE_SHIFT-10),
 		mm->data_vm << (PAGE_SHIFT-10),
 		mm->stack_vm << (PAGE_SHIFT-10), text, lib,
-		ptes >> 10,
-		pmds >> 10,
-		puds >> 10,
+		mm_pgtables_bytes(mm) >> 10,
 		swap << (PAGE_SHIFT-10));
 	hugetlb_report_usage(m, mm);
 }
diff --git a/include/linux/mm.h b/include/linux/mm.h
index 2ca799f0d762..7c1e82a1aa77 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -1605,37 +1605,20 @@ static inline int __pud_alloc(struct mm_struct *mm, p4d_t *p4d,
 {
 	return 0;
 }
-
-static inline unsigned long mm_nr_puds(const struct mm_struct *mm)
-{
-	return 0;
-}
-
-static inline void mm_nr_puds_init(struct mm_struct *mm) {}
 static inline void mm_inc_nr_puds(struct mm_struct *mm) {}
 static inline void mm_dec_nr_puds(struct mm_struct *mm) {}
 
 #else
 int __pud_alloc(struct mm_struct *mm, p4d_t *p4d, unsigned long address);
 
-static inline void mm_nr_puds_init(struct mm_struct *mm)
-{
-	atomic_long_set(&mm->nr_puds, 0);
-}
-
-static inline unsigned long mm_nr_puds(const struct mm_struct *mm)
-{
-	return atomic_long_read(&mm->nr_puds);
-}
-
 static inline void mm_inc_nr_puds(struct mm_struct *mm)
 {
-	atomic_long_inc(&mm->nr_puds);
+	atomic_long_add(PTRS_PER_PUD * sizeof(pud_t), &mm->pgtables_bytes);
 }
 
 static inline void mm_dec_nr_puds(struct mm_struct *mm)
 {
-	atomic_long_dec(&mm->nr_puds);
+	atomic_long_sub(PTRS_PER_PUD * sizeof(pud_t), &mm->pgtables_bytes);
 }
 #endif
 
@@ -1646,64 +1629,47 @@ static inline int __pmd_alloc(struct mm_struct *mm, pud_t *pud,
 	return 0;
 }
 
-static inline void mm_nr_pmds_init(struct mm_struct *mm) {}
-
-static inline unsigned long mm_nr_pmds(const struct mm_struct *mm)
-{
-	return 0;
-}
-
 static inline void mm_inc_nr_pmds(struct mm_struct *mm) {}
 static inline void mm_dec_nr_pmds(struct mm_struct *mm) {}
 
 #else
 int __pmd_alloc(struct mm_struct *mm, pud_t *pud, unsigned long address);
 
-static inline void mm_nr_pmds_init(struct mm_struct *mm)
-{
-	atomic_long_set(&mm->nr_pmds, 0);
-}
-
-static inline unsigned long mm_nr_pmds(const struct mm_struct *mm)
-{
-	return atomic_long_read(&mm->nr_pmds);
-}
-
 static inline void mm_inc_nr_pmds(struct mm_struct *mm)
 {
-	atomic_long_inc(&mm->nr_pmds);
+	atomic_long_add(PTRS_PER_PMD * sizeof(pmd_t), &mm->pgtables_bytes);
 }
 
 static inline void mm_dec_nr_pmds(struct mm_struct *mm)
 {
-	atomic_long_dec(&mm->nr_pmds);
+	atomic_long_sub(PTRS_PER_PMD * sizeof(pmd_t), &mm->pgtables_bytes);
 }
 #endif
 
 #ifdef CONFIG_MMU
-static inline void mm_nr_ptes_init(struct mm_struct *mm)
+static inline void mm_pgtables_bytes_init(struct mm_struct *mm)
 {
-	atomic_long_set(&mm->nr_ptes, 0);
+	atomic_long_set(&mm->pgtables_bytes, 0);
 }
 
-static inline unsigned long mm_nr_ptes(const struct mm_struct *mm)
+static inline unsigned long mm_pgtables_bytes(const struct mm_struct *mm)
 {
-	return atomic_long_read(&mm->nr_ptes);
+	return atomic_long_read(&mm->pgtables_bytes);
 }
 
 static inline void mm_inc_nr_ptes(struct mm_struct *mm)
 {
-	atomic_long_inc(&mm->nr_ptes);
+	atomic_long_add(PTRS_PER_PTE * sizeof(pte_t), &mm->pgtables_bytes);
 }
 
 static inline void mm_dec_nr_ptes(struct mm_struct *mm)
 {
-	atomic_long_dec(&mm->nr_ptes);
+	atomic_long_sub(PTRS_PER_PTE * sizeof(pte_t), &mm->pgtables_bytes);
 }
 #else
-static inline void mm_nr_ptes_init(struct mm_struct *mm) {}
 
-static inline unsigned long mm_nr_ptes(const struct mm_struct *mm)
+static inline void mm_pgtables_bytes_init(struct mm_struct *mm) {}
+static inline unsigned long mm_pgtables_bytes(const struct mm_struct *mm)
 {
 	return 0;
 }
diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
index e42048020664..09643e0472fc 100644
--- a/include/linux/mm_types.h
+++ b/include/linux/mm_types.h
@@ -402,13 +402,7 @@ struct mm_struct {
 	atomic_t mm_count;
 
 #ifdef CONFIG_MMU
-	atomic_long_t nr_ptes;			/* PTE page table pages */
-#endif
-#if CONFIG_PGTABLE_LEVELS > 2
-	atomic_long_t nr_pmds;			/* PMD page table pages */
-#endif
-#if CONFIG_PGTABLE_LEVELS > 3
-	atomic_long_t nr_puds;			/* PUD page table pages */
+	atomic_long_t pgtables_bytes;		/* PTE page table pages */
 #endif
 	int map_count;				/* number of VMAs */
 
diff --git a/kernel/fork.c b/kernel/fork.c
index 946922a30ede..006dc5899a1a 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -817,9 +817,7 @@ static struct mm_struct *mm_init(struct mm_struct *mm, struct task_struct *p,
 	init_rwsem(&mm->mmap_sem);
 	INIT_LIST_HEAD(&mm->mmlist);
 	mm->core_state = NULL;
-	mm_nr_ptes_init(mm);
-	mm_nr_pmds_init(mm);
-	mm_nr_puds_init(mm);
+	mm_pgtables_bytes_init(mm);
 	mm->map_count = 0;
 	mm->locked_vm = 0;
 	mm->pinned_vm = 0;
@@ -873,15 +871,9 @@ static void check_mm(struct mm_struct *mm)
 					  "mm:%p idx:%d val:%ld\n", mm, i, x);
 	}
 
-	if (mm_nr_ptes(mm))
-		pr_alert("BUG: non-zero nr_ptes on freeing mm: %ld\n",
-				mm_nr_ptes(mm));
-	if (mm_nr_pmds(mm))
-		pr_alert("BUG: non-zero nr_pmds on freeing mm: %ld\n",
-				mm_nr_pmds(mm));
-	if (mm_nr_puds(mm))
-		pr_alert("BUG: non-zero nr_puds on freeing mm: %ld\n",
-				mm_nr_puds(mm));
+	if (mm_pgtables_bytes(mm))
+		pr_alert("BUG: non-zero pgtables_bytes on freeing mm: %ld\n",
+				mm_pgtables_bytes(mm));
 
 #if defined(CONFIG_TRANSPARENT_HUGEPAGE) && !USE_SPLIT_PMD_PTLOCKS
 	VM_BUG_ON_MM(mm->pmd_huge_pte, mm);
diff --git a/mm/debug.c b/mm/debug.c
index c9888a6d7875..d947f3e03b0d 100644
--- a/mm/debug.c
+++ b/mm/debug.c
@@ -105,8 +105,7 @@ void dump_mm(const struct mm_struct *mm)
 		"get_unmapped_area %p\n"
 #endif
 		"mmap_base %lu mmap_legacy_base %lu highest_vm_end %lu\n"
-		"pgd %p mm_users %d mm_count %d\n"
-		"nr_ptes %lu nr_pmds %lu nr_puds %lu map_count %d\n"
+		"pgd %p mm_users %d mm_count %d pgtables_bytes %lu map_count %d\n"
 		"hiwater_rss %lx hiwater_vm %lx total_vm %lx locked_vm %lx\n"
 		"pinned_vm %lx data_vm %lx exec_vm %lx stack_vm %lx\n"
 		"start_code %lx end_code %lx start_data %lx end_data %lx\n"
@@ -136,9 +135,7 @@ void dump_mm(const struct mm_struct *mm)
 		mm->mmap_base, mm->mmap_legacy_base, mm->highest_vm_end,
 		mm->pgd, atomic_read(&mm->mm_users),
 		atomic_read(&mm->mm_count),
-		mm_nr_ptes(mm),
-		mm_nr_pmds(mm),
-		mm_nr_puds(mm),
+		mm_pgtables_bytes(mm),
 		mm->map_count,
 		mm->hiwater_rss, mm->hiwater_vm, mm->total_vm, mm->locked_vm,
 		mm->pinned_vm, mm->data_vm, mm->exec_vm, mm->stack_vm,
diff --git a/mm/huge_memory.c b/mm/huge_memory.c
index 3610d81c062a..86fe697e8bfb 100644
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -942,7 +942,7 @@ int copy_huge_pmd(struct mm_struct *dst_mm, struct mm_struct *src_mm,
 			set_pmd_at(src_mm, addr, src_pmd, pmd);
 		}
 		add_mm_counter(dst_mm, MM_ANONPAGES, HPAGE_PMD_NR);
-		atomic_long_inc(&dst_mm->nr_ptes);
+		mm_inc_nr_ptes(dst_mm);
 		pgtable_trans_huge_deposit(dst_mm, dst_pmd, pgtable);
 		set_pmd_at(dst_mm, addr, dst_pmd, pmd);
 		ret = 0;
diff --git a/mm/oom_kill.c b/mm/oom_kill.c
index f9300141480e..26add8a0d1f7 100644
--- a/mm/oom_kill.c
+++ b/mm/oom_kill.c
@@ -221,7 +221,7 @@ unsigned long oom_badness(struct task_struct *p, struct mem_cgroup *memcg,
 	 * task's rss, pagetable and swap space use.
 	 */
 	points = get_mm_rss(p->mm) + get_mm_counter(p->mm, MM_SWAPENTS) +
-		mm_nr_ptes(p->mm) + mm_nr_pmds(p->mm) + mm_nr_puds(p->mm);
+		mm_pgtables_bytes(p->mm) / PAGE_SIZE;
 	task_unlock(p);
 
 	/*
@@ -389,15 +389,15 @@ static void select_bad_process(struct oom_control *oc)
  * Dumps the current memory state of all eligible tasks.  Tasks not in the same
  * memcg, not in the same cpuset, or bound to a disjoint set of mempolicy nodes
  * are not shown.
- * State information includes task's pid, uid, tgid, vm size, rss, nr_ptes,
- * swapents, oom_score_adj value, and name.
+ * State information includes task's pid, uid, tgid, vm size, rss,
+ * pgtables_bytes, swapents, oom_score_adj value, and name.
  */
 static void dump_tasks(struct mem_cgroup *memcg, const nodemask_t *nodemask)
 {
 	struct task_struct *p;
 	struct task_struct *task;
 
-	pr_info("[ pid ]   uid  tgid total_vm      rss nr_ptes nr_pmds nr_puds swapents oom_score_adj name\n");
+	pr_info("[ pid ]   uid  tgid total_vm      rss pgtables_bytes swapents oom_score_adj name\n");
 	rcu_read_lock();
 	for_each_process(p) {
 		if (oom_unkillable_task(p, memcg, nodemask))
@@ -413,12 +413,10 @@ static void dump_tasks(struct mem_cgroup *memcg, const nodemask_t *nodemask)
 			continue;
 		}
 
-		pr_info("[%5d] %5d %5d %8lu %8lu %7ld %7ld %7ld %8lu         %5hd %s\n",
+		pr_info("[%5d] %5d %5d %8lu %8lu %8ld %8lu         %5hd %s\n",
 			task->pid, from_kuid(&init_user_ns, task_uid(task)),
 			task->tgid, task->mm->total_vm, get_mm_rss(task->mm),
-			mm_nr_ptes(task->mm),
-			mm_nr_pmds(task->mm),
-			mm_nr_puds(task->mm),
+			mm_pgtables_bytes(task->mm),
 			get_mm_counter(task->mm, MM_SWAPENTS),
 			task->signal->oom_score_adj, task->comm);
 		task_unlock(task);