mm: free swap cache aggressively if memcg swap is full

Swap cache pages are freed aggressively if swap is nearly full (>50%
currently), because otherwise we are likely to stop scanning anonymous
when we near the swap limit even if there is plenty of freeable swap cache
pages.  We should follow the same trend in case of memory cgroup, which
has its own swap limit.

Signed-off-by: Vladimir Davydov <vdavydov@virtuozzo.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: Michal Hocko <mhocko@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

5 files changed, 32 insertions, 3 deletions

diff --git a/include/linux/swap.h b/include/linux/swap.h
index a587050..d18b65c 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -552,6 +552,7 @@ extern void mem_cgroup_swapout(struct page *page, swp_entry_t entry);
 extern int mem_cgroup_try_charge_swap(struct page *page, swp_entry_t entry);
 extern void mem_cgroup_uncharge_swap(swp_entry_t entry);
 extern long mem_cgroup_get_nr_swap_pages(struct mem_cgroup *memcg);
+extern bool mem_cgroup_swap_full(struct page *page);
 #else
 static inline void mem_cgroup_swapout(struct page *page, swp_entry_t entry)
 {
@@ -571,6 +572,11 @@ static inline long mem_cgroup_get_nr_swap_pages(struct mem_cgroup *memcg)
 {
 	return get_nr_swap_pages();
 }
+
+static inline bool mem_cgroup_swap_full(struct page *page)
+{
+	return vm_swap_full();
+}
 #endif
 
 #endif /* __KERNEL__*/
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index bcb3871..6a00079 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -5761,6 +5761,28 @@ long mem_cgroup_get_nr_swap_pages(struct mem_cgroup *memcg)
 	return nr_swap_pages;
 }
 
+bool mem_cgroup_swap_full(struct page *page)
+{
+	struct mem_cgroup *memcg;
+
+	VM_BUG_ON_PAGE(!PageLocked(page), page);
+
+	if (vm_swap_full())
+		return true;
+	if (!do_swap_account || !cgroup_subsys_on_dfl(memory_cgrp_subsys))
+		return false;
+
+	memcg = page->mem_cgroup;
+	if (!memcg)
+		return false;
+
+	for (; memcg != root_mem_cgroup; memcg = parent_mem_cgroup(memcg))
+		if (page_counter_read(&memcg->swap) * 2 >= memcg->swap.limit)
+			return true;
+
+	return false;
+}
+
 /* for remember boot option*/
 #ifdef CONFIG_MEMCG_SWAP_ENABLED
 static int really_do_swap_account __initdata = 1;
diff --git a/mm/memory.c b/mm/memory.c
index ff17850..30991f8 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -2582,7 +2582,8 @@ static int do_swap_page(struct mm_struct *mm, struct vm_area_struct *vma,
 	}
 
 	swap_free(entry);
-	if (vm_swap_full() || (vma->vm_flags & VM_LOCKED) || PageMlocked(page))
+	if (mem_cgroup_swap_full(page) ||
+	    (vma->vm_flags & VM_LOCKED) || PageMlocked(page))
 		try_to_free_swap(page);
 	unlock_page(page);
 	if (page != swapcache) {
diff --git a/mm/swapfile.c b/mm/swapfile.c
index 22a7a1f..c43f654 100644
--- a/mm/swapfile.c
+++ b/mm/swapfile.c
@@ -1006,7 +1006,7 @@ int free_swap_and_cache(swp_entry_t entry)
 		 * Also recheck PageSwapCache now page is locked (above).
 		 */
 		if (PageSwapCache(page) && !PageWriteback(page) &&
-				(!page_mapped(page) || vm_swap_full())) {
+		    (!page_mapped(page) || mem_cgroup_swap_full(page))) {
 			delete_from_swap_cache(page);
 			SetPageDirty(page);
 		}
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 3be5f9d..bd620b6 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -1214,7 +1214,7 @@ cull_mlocked:
 
 activate_locked:
 		/* Not a candidate for swapping, so reclaim swap space. */
-		if (PageSwapCache(page) && vm_swap_full())
+		if (PageSwapCache(page) && mem_cgroup_swap_full(page))
 			try_to_free_swap(page);
 		VM_BUG_ON_PAGE(PageActive(page), page);
 		SetPageActive(page);