memcg: remove a redundant check

Because of remove refcnt patch, it's very rare case to that mem_cgroup_charge_common() is called against a page which is accounted. mem_cgroup_charge_common() is called when. 1. a page is added into file cache. 2. an anon page is _newly_ mapped. A racy case is that a newly-swapped-in anonymous page is referred from prural threads in do_swap_page() at the same time. (a page is not Locked when mem_cgroup_charge() is called from do_swap_page.) Another case is shmem. It charges its page before calling add_to_page_cache(). Then, mem_cgroup_charge_cache() is called twice. This case is handled in mem_cgroup_cache_charge(). But this check may be too hacky... Signed-off-by : KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Balbir Singh <balbir@in.ibm.com> Cc: "Eric W. Biederman" <ebiederm@xmission.com> Cc: Pavel Emelyanov <xemul@openvz.org> Cc: Li Zefan <lizf@cn.fujitsu.com> Cc: Hugh Dickins <hugh@veritas.com> Cc: YAMAMOTO Takashi <yamamoto@valinux.co.jp> Cc: Paul Menage <menage@google.com> Cc: David Rientjes <rientjes@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

memcg: remove a redundant check
Because of remove refcnt patch, it's very rare case to that mem_cgroup_charge_common() is called against a page which is accounted. mem_cgroup_charge_common() is called when. 1. a page is added into file cache. 2. an anon page is _newly_ mapped. A racy case is that a newly-swapped-in anonymous page is referred from prural threads in do_swap_page() at the same time. (a page is not Locked when mem_cgroup_charge() is called from do_swap_page.) Another case is shmem. It charges its page before calling add_to_page_cache(). Then, mem_cgroup_charge_cache() is called twice. This case is handled in mem_cgroup_cache_charge(). But this check may be too hacky... Signed-off-by : KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Balbir Singh <balbir@in.ibm.com> Cc: "Eric W. Biederman" <ebiederm@xmission.com> Cc: Pavel Emelyanov <xemul@openvz.org> Cc: Li Zefan <lizf@cn.fujitsu.com> Cc: Hugh Dickins <hugh@veritas.com> Cc: YAMAMOTO Takashi <yamamoto@valinux.co.jp> Cc: Paul Menage <menage@google.com> Cc: David Rientjes <rientjes@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
KAMEZAWA Hiroyuki · Linus Torvalds
1 parent b76734e5e3
Showing 1 changed file with 25 additions and 28 deletions Side-by-side Diff
mm/memcontrol.c
@@ -536,28 +536,6 @@
 	if (mem_cgroup_subsys.disabled)
 		return 0;
  
-	/*
-	 * Should page_cgroup's go to their own slab?
-	 * One could optimize the performance of the charging routine
-	 * by saving a bit in the page_flags and using it as a lock
-	 * to see if the cgroup page already has a page_cgroup associated
-	 * with it
-	 */
-retry:
-	lock_page_cgroup(page);
-	pc = page_get_page_cgroup(page);
-	/*
-	 * The page_cgroup exists and
-	 * the page has already been accounted.
-	 */
-	if (unlikely(pc)) {
-		VM_BUG_ON(pc->page != page);
-		VM_BUG_ON(!pc->mem_cgroup);
-		unlock_page_cgroup(page);
-		goto done;
-	}
-	unlock_page_cgroup(page);
-
 	pc = kmem_cache_alloc(page_cgroup_cache, gfp_mask);
 	if (unlikely(pc == NULL))
 		goto err;
  
@@ -618,15 +596,10 @@
 	lock_page_cgroup(page);
 	if (unlikely(page_get_page_cgroup(page))) {
 		unlock_page_cgroup(page);
-		/*
-		 * Another charge has been added to this page already.
-		 * We take lock_page_cgroup(page) again and read
-		 * page->cgroup, increment refcnt.... just retry is OK.
-		 */
 		res_counter_uncharge(&mem->res, PAGE_SIZE);
 		css_put(&mem->css);
 		kmem_cache_free(page_cgroup_cache, pc);
-		goto retry;
+		goto done;
 	}
 	page_assign_page_cgroup(page, pc);
  
  
@@ -665,8 +638,32 @@
 int mem_cgroup_cache_charge(struct page *page, struct mm_struct *mm,
 				gfp_t gfp_mask)
 {
+	/*
+	 * Corner case handling. This is called from add_to_page_cache()
+	 * in usual. But some FS (shmem) precharges this page before calling it
+	 * and call add_to_page_cache() with GFP_NOWAIT.
+	 *
+	 * For GFP_NOWAIT case, the page may be pre-charged before calling
+	 * add_to_page_cache(). (See shmem.c) check it here and avoid to call
+	 * charge twice. (It works but has to pay a bit larger cost.)
+	 */
+	if (!(gfp_mask & __GFP_WAIT)) {
+		struct page_cgroup *pc;
+
+		lock_page_cgroup(page);
+		pc = page_get_page_cgroup(page);
+		if (pc) {
+			VM_BUG_ON(pc->page != page);
+			VM_BUG_ON(!pc->mem_cgroup);
+			unlock_page_cgroup(page);
+			return 0;
+		}
+		unlock_page_cgroup(page);
+	}
+
 	if (unlikely(!mm))
 		mm = &init_mm;
+
 	return mem_cgroup_charge_common(page, mm, gfp_mask,
 				MEM_CGROUP_CHARGE_TYPE_CACHE, NULL);
 }
...	...	@@ -536,28 +536,6 @@
536	536	if (mem_cgroup_subsys.disabled)
537	537	return 0;
538	538
539		- /*
540		- * Should page_cgroup's go to their own slab?
541		- * One could optimize the performance of the charging routine
542		- * by saving a bit in the page_flags and using it as a lock
543		- * to see if the cgroup page already has a page_cgroup associated
544		- * with it
545		- */
546		-retry:
547		- lock_page_cgroup(page);
548		- pc = page_get_page_cgroup(page);
549		- /*
550		- * The page_cgroup exists and
551		- * the page has already been accounted.
552		- */
553		- if (unlikely(pc)) {
554		- VM_BUG_ON(pc->page != page);
555		- VM_BUG_ON(!pc->mem_cgroup);
556		- unlock_page_cgroup(page);
557		- goto done;
558		- }
559		- unlock_page_cgroup(page);
560		-
561	539	pc = kmem_cache_alloc(page_cgroup_cache, gfp_mask);
562	540	if (unlikely(pc == NULL))
563	541	goto err;
564	542
...	...	@@ -618,15 +596,10 @@
618	596	lock_page_cgroup(page);
619	597	if (unlikely(page_get_page_cgroup(page))) {
620	598	unlock_page_cgroup(page);
621		- /*
622		- * Another charge has been added to this page already.
623		- * We take lock_page_cgroup(page) again and read
624		- * page->cgroup, increment refcnt.... just retry is OK.
625		- */
626	599	res_counter_uncharge(&mem->res, PAGE_SIZE);
627	600	css_put(&mem->css);
628	601	kmem_cache_free(page_cgroup_cache, pc);
629		- goto retry;
	602	+ goto done;
630	603	}
631	604	page_assign_page_cgroup(page, pc);
632	605
633	606
...	...	@@ -665,8 +638,32 @@
665	638	int mem_cgroup_cache_charge(struct page page, struct mm_struct mm,
666	639	gfp_t gfp_mask)
667	640	{
	641	+ /*
	642	+ * Corner case handling. This is called from add_to_page_cache()
	643	+ * in usual. But some FS (shmem) precharges this page before calling it
	644	+ * and call add_to_page_cache() with GFP_NOWAIT.
	645	+ *
	646	+ * For GFP_NOWAIT case, the page may be pre-charged before calling
	647	+ * add_to_page_cache(). (See shmem.c) check it here and avoid to call
	648	+ * charge twice. (It works but has to pay a bit larger cost.)
	649	+ */
	650	+ if (!(gfp_mask & __GFP_WAIT)) {
	651	+ struct page_cgroup *pc;
	652	+
	653	+ lock_page_cgroup(page);
	654	+ pc = page_get_page_cgroup(page);
	655	+ if (pc) {
	656	+ VM_BUG_ON(pc->page != page);
	657	+ VM_BUG_ON(!pc->mem_cgroup);
	658	+ unlock_page_cgroup(page);
	659	+ return 0;
	660	+ }
	661	+ unlock_page_cgroup(page);
	662	+ }
	663	+
668	664	if (unlikely(!mm))
669	665	mm = &init_mm;
	666	+
670	667	return mem_cgroup_charge_common(page, mm, gfp_mask,
671	668	MEM_CGROUP_CHARGE_TYPE_CACHE, NULL);
672	669	}