memcg: limit change shrink usage

Shrinking memory usage at limit change. [akpm@linux-foundation.org: coding-style fixes] Acked-by: Balbir Singh <balbir@linux.vnet.ibm.com> Acked-by: Pavel Emelyanov <xemul@openvz.org> Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Paul Menage <menage@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

memcg: limit change shrink usage
Shrinking memory usage at limit change. [akpm@linux-foundation.org: coding-style fixes] Acked-by: Balbir Singh <balbir@linux.vnet.ibm.com> Acked-by: Pavel Emelyanov <xemul@openvz.org> Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Paul Menage <menage@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 12b9804419
Showing 2 changed files with 45 additions and 6 deletions Side-by-side Diff
Documentation/controllers/memory.txt
mm/memcontrol.c
@@ -242,8 +242,7 @@
 1. Add support for accounting huge pages (as a separate controller)
 2. Make per-cgroup scanner reclaim not-shared pages first
 3. Teach controller to account for shared-pages
-4. Start reclamation when the limit is lowered
-5. Start reclamation in the background when the limit is
+4. Start reclamation in the background when the limit is
    not yet hit but the usage is getting closer
  
 Summary
@@ -812,6 +812,30 @@
 	return 0;
 }
  
+int mem_cgroup_resize_limit(struct mem_cgroup *memcg, unsigned long long val)
+{
+
+	int retry_count = MEM_CGROUP_RECLAIM_RETRIES;
+	int progress;
+	int ret = 0;
+
+	while (res_counter_set_limit(&memcg->res, val)) {
+		if (signal_pending(current)) {
+			ret = -EINTR;
+			break;
+		}
+		if (!retry_count) {
+			ret = -EBUSY;
+			break;
+		}
+		progress = try_to_free_mem_cgroup_pages(memcg, GFP_KERNEL);
+		if (!progress)
+			retry_count--;
+	}
+	return ret;
+}
+
+
 /*
  * This routine traverse page_cgroup in given list and drop them all.
  * *And* this routine doesn't reclaim page itself, just removes page_cgroup.
  
@@ -896,13 +920,29 @@
 	return res_counter_read_u64(&mem_cgroup_from_cont(cont)->res,
 				    cft->private);
 }
-
+/*
+ * The user of this function is...
+ * RES_LIMIT.
+ */
 static int mem_cgroup_write(struct cgroup *cont, struct cftype *cft,
 			    const char *buffer)
 {
-	return res_counter_write(&mem_cgroup_from_cont(cont)->res,
-				 cft->private, buffer,
-				 res_counter_memparse_write_strategy);
+	struct mem_cgroup *memcg = mem_cgroup_from_cont(cont);
+	unsigned long long val;
+	int ret;
+
+	switch (cft->private) {
+	case RES_LIMIT:
+		/* This function does all necessary parse...reuse it */
+		ret = res_counter_memparse_write_strategy(buffer, &val);
+		if (!ret)
+			ret = mem_cgroup_resize_limit(memcg, val);
+		break;
+	default:
+		ret = -EINVAL; /* should be BUG() ? */
+		break;
+	}
+	return ret;
 }
  
 static int mem_cgroup_reset(struct cgroup *cont, unsigned int event)
...	...	@@ -242,8 +242,7 @@
242	242	1. Add support for accounting huge pages (as a separate controller)
243	243	2. Make per-cgroup scanner reclaim not-shared pages first
244	244	3. Teach controller to account for shared-pages
245		-4. Start reclamation when the limit is lowered
246		-5. Start reclamation in the background when the limit is
	245	+4. Start reclamation in the background when the limit is
247	246	not yet hit but the usage is getting closer
248	247
249	248	Summary
...	...	@@ -812,6 +812,30 @@
812	812	return 0;
813	813	}
814	814
	815	+int mem_cgroup_resize_limit(struct mem_cgroup *memcg, unsigned long long val)
	816	+{
	817	+
	818	+ int retry_count = MEM_CGROUP_RECLAIM_RETRIES;
	819	+ int progress;
	820	+ int ret = 0;
	821	+
	822	+ while (res_counter_set_limit(&memcg->res, val)) {
	823	+ if (signal_pending(current)) {
	824	+ ret = -EINTR;
	825	+ break;
	826	+ }
	827	+ if (!retry_count) {
	828	+ ret = -EBUSY;
	829	+ break;
	830	+ }
	831	+ progress = try_to_free_mem_cgroup_pages(memcg, GFP_KERNEL);
	832	+ if (!progress)
	833	+ retry_count--;
	834	+ }
	835	+ return ret;
	836	+}
	837	+
	838	+
815	839	/*
816	840	* This routine traverse page_cgroup in given list and drop them all.
817	841	* And this routine doesn't reclaim page itself, just removes page_cgroup.
818	842
...	...	@@ -896,13 +920,29 @@
896	920	return res_counter_read_u64(&mem_cgroup_from_cont(cont)->res,
897	921	cft->private);
898	922	}
899		-
	923	+/*
	924	+ * The user of this function is...
	925	+ * RES_LIMIT.
	926	+ */
900	927	static int mem_cgroup_write(struct cgroup cont, struct cftype cft,
901	928	const char *buffer)
902	929	{
903		- return res_counter_write(&mem_cgroup_from_cont(cont)->res,
904		- cft->private, buffer,
905		- res_counter_memparse_write_strategy);
	930	+ struct mem_cgroup *memcg = mem_cgroup_from_cont(cont);
	931	+ unsigned long long val;
	932	+ int ret;
	933	+
	934	+ switch (cft->private) {
	935	+ case RES_LIMIT:
	936	+ /* This function does all necessary parse...reuse it */
	937	+ ret = res_counter_memparse_write_strategy(buffer, &val);
	938	+ if (!ret)
	939	+ ret = mem_cgroup_resize_limit(memcg, val);
	940	+ break;
	941	+ default:
	942	+ ret = -EINVAL; /* should be BUG() ? */
	943	+ break;
	944	+ }
	945	+ return ret;
906	946	}
907	947
908	948	static int mem_cgroup_reset(struct cgroup *cont, unsigned int event)