blkcg: replace blkcg_policy->cpd_size with ->cpd_alloc/free_fn() methods

Each active policy has a cpd (blkcg_policy_data) on each blkcg.  The
cpd's were allocated by blkcg core and each policy could request to
allocate extra space at the end by setting blkcg_policy->cpd_size
larger than the size of cpd.

This is a bit unusual but blkg (blkcg_gq) policy data used to be
handled this way too so it made sense to be consistent; however, blkg
policy data switched to alloc/free callbacks.

This patch makes similar changes to cpd handling.
blkcg_policy->cpd_alloc/free_fn() are added to replace ->cpd_size.  As
cpd allocation is now done from policy side, it can simply allocate a
larger area which embeds cpd at the beginning.

As ->cpd_alloc_fn() may be able to perform all necessary
initializations, this patch makes ->cpd_init_fn() optional.

Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Vivek Goyal <vgoyal@redhat.com>
Cc: Arianna Avanzini <avanzini.arianna@gmail.com>
Signed-off-by: Jens Axboe <axboe@fb.com>

block/blk-cgroup.c

@@ -817,11 +817,15 @@ static void blkcg_css_free(struct cgroup_subsys_state *css)
 	int i;
 
 	mutex_lock(&blkcg_pol_mutex);
+
 	list_del(&blkcg->all_blkcgs_node);
-	mutex_unlock(&blkcg_pol_mutex);
 
 	for (i = 0; i < BLKCG_MAX_POLS; i++)
-		kfree(blkcg->cpd[i]);
+		if (blkcg->cpd[i])
+			blkcg_policy[i]->cpd_free_fn(blkcg->cpd[i]);
+
+	mutex_unlock(&blkcg_pol_mutex);
+
 	kfree(blkcg);
 }
 
@@ -854,11 +858,10 @@ blkcg_css_alloc(struct cgroup_subsys_state *parent_css)
 		 * check if the policy requires any specific per-cgroup
 		 * data: if it does, allocate and initialize it.
 		 */
-		if (!pol || !pol->cpd_size)
+		if (!pol || !pol->cpd_alloc_fn)
 			continue;
 
-		BUG_ON(blkcg->cpd[i]);
+		cpd = pol->cpd_alloc_fn(GFP_KERNEL);
-		cpd = kzalloc(pol->cpd_size, GFP_KERNEL);
 		if (!cpd) {
 			ret = ERR_PTR(-ENOMEM);
 			goto free_pd_blkcg;
@@ -866,7 +869,8 @@ blkcg_css_alloc(struct cgroup_subsys_state *parent_css)
 		blkcg->cpd[i] = cpd;
 		cpd->blkcg = blkcg;
 		cpd->plid = i;
-		pol->cpd_init_fn(cpd);
+		if (pol->cpd_init_fn)
+			pol->cpd_init_fn(cpd);
 	}
 
 	spin_lock_init(&blkcg->lock);
@@ -882,7 +886,8 @@ blkcg_css_alloc(struct cgroup_subsys_state *parent_css)
 
 free_pd_blkcg:
 	for (i--; i >= 0; i--)
-		kfree(blkcg->cpd[i]);
+		if (blkcg->cpd[i])
+			blkcg_policy[i]->cpd_free_fn(blkcg->cpd[i]);
 free_blkcg:
 	kfree(blkcg);
 	mutex_unlock(&blkcg_pol_mutex);
@@ -1159,11 +1164,11 @@ int blkcg_policy_register(struct blkcg_policy *pol)
 	blkcg_policy[pol->plid] = pol;
 
 	/* allocate and install cpd's */
-	if (pol->cpd_size) {
+	if (pol->cpd_alloc_fn) {
 		list_for_each_entry(blkcg, &all_blkcgs, all_blkcgs_node) {
 			struct blkcg_policy_data *cpd;
 
-			cpd = kzalloc(pol->cpd_size, GFP_KERNEL);
+			cpd = pol->cpd_alloc_fn(GFP_KERNEL);
 			if (!cpd) {
 				mutex_unlock(&blkcg_pol_mutex);
 				goto err_free_cpds;
@@ -1186,10 +1191,12 @@ int blkcg_policy_register(struct blkcg_policy *pol)
 	return 0;
 
 err_free_cpds:
-	if (pol->cpd_size) {
+	if (pol->cpd_alloc_fn) {
 		list_for_each_entry(blkcg, &all_blkcgs, all_blkcgs_node) {
-			kfree(blkcg->cpd[pol->plid]);
+			if (blkcg->cpd[pol->plid]) {
-			blkcg->cpd[pol->plid] = NULL;
+				pol->cpd_free_fn(blkcg->cpd[pol->plid]);
+				blkcg->cpd[pol->plid] = NULL;
+			}
 		}
 	}
 	blkcg_policy[pol->plid] = NULL;
@@ -1222,10 +1229,12 @@ void blkcg_policy_unregister(struct blkcg_policy *pol)
 	/* remove cpds and unregister */
 	mutex_lock(&blkcg_pol_mutex);
 
-	if (pol->cpd_size) {
+	if (pol->cpd_alloc_fn) {
 		list_for_each_entry(blkcg, &all_blkcgs, all_blkcgs_node) {
-			kfree(blkcg->cpd[pol->plid]);
+			if (blkcg->cpd[pol->plid]) {
-			blkcg->cpd[pol->plid] = NULL;
+				pol->cpd_free_fn(blkcg->cpd[pol->plid]);
+				blkcg->cpd[pol->plid] = NULL;
+			}
 		}
 	}
 	blkcg_policy[pol->plid] = NULL;

block/cfq-iosched.c

@@ -1568,6 +1568,16 @@ static void cfqg_stats_init(struct cfqg_stats *stats)
 #endif
 }
 
+static struct blkcg_policy_data *cfq_cpd_alloc(gfp_t gfp)
+{
+	struct cfq_group_data *cgd;
+
+	cgd = kzalloc(sizeof(*cgd), GFP_KERNEL);
+	if (!cgd)
+		return NULL;
+	return &cgd->cpd;
+}
+
 static void cfq_cpd_init(struct blkcg_policy_data *cpd)
 {
 	struct cfq_group_data *cgd = cpd_to_cfqgd(cpd);
@@ -1581,6 +1591,11 @@ static void cfq_cpd_init(struct blkcg_policy_data *cpd)
 	}
 }
 
+static void cfq_cpd_free(struct blkcg_policy_data *cpd)
+{
+	kfree(cpd_to_cfqgd(cpd));
+}
+
 static struct blkg_policy_data *cfq_pd_alloc(gfp_t gfp, int node)
 {
 	struct cfq_group *cfqg;
@@ -4649,10 +4664,12 @@ static struct elevator_type iosched_cfq = {
 
 #ifdef CONFIG_CFQ_GROUP_IOSCHED
 static struct blkcg_policy blkcg_policy_cfq = {
-	.cpd_size		= sizeof(struct cfq_group_data),
 	.cftypes		= cfq_blkcg_files,
 
+	.cpd_alloc_fn		= cfq_cpd_alloc,
 	.cpd_init_fn		= cfq_cpd_init,
+	.cpd_free_fn		= cfq_cpd_free,
+
 	.pd_alloc_fn		= cfq_pd_alloc,
 	.pd_init_fn		= cfq_pd_init,
 	.pd_offline_fn		= cfq_pd_offline,

include/linux/blk-cgroup.h

@@ -81,11 +81,11 @@ struct blkg_policy_data {
 };
 
 /*
- * Policies that need to keep per-blkcg data which is independent
+ * Policies that need to keep per-blkcg data which is independent from any
- * from any request_queue associated to it must specify its size
+ * request_queue associated to it should implement cpd_alloc/free_fn()
- * with the cpd_size field of the blkcg_policy structure and
+ * methods.  A policy can allocate private data area by allocating larger
- * embed a blkcg_policy_data in it.  cpd_init() is invoked to let
+ * data structure which embeds blkcg_policy_data at the beginning.
- * each policy handle per-blkcg data.
+ * cpd_init() is invoked to let each policy handle per-blkcg data.
  */
 struct blkcg_policy_data {
 	/* the blkcg and policy id this per-policy data belongs to */
@@ -124,7 +124,9 @@ struct blkcg_gq {
 	struct rcu_head			rcu_head;
 };
 
+typedef struct blkcg_policy_data *(blkcg_pol_alloc_cpd_fn)(gfp_t gfp);
 typedef void (blkcg_pol_init_cpd_fn)(struct blkcg_policy_data *cpd);
+typedef void (blkcg_pol_free_cpd_fn)(struct blkcg_policy_data *cpd);
 typedef struct blkg_policy_data *(blkcg_pol_alloc_pd_fn)(gfp_t gfp, int node);
 typedef void (blkcg_pol_init_pd_fn)(struct blkg_policy_data *pd);
 typedef void (blkcg_pol_online_pd_fn)(struct blkg_policy_data *pd);
@@ -134,13 +136,14 @@ typedef void (blkcg_pol_reset_pd_stats_fn)(struct blkg_policy_data *pd);
 
 struct blkcg_policy {
 	int				plid;
-	/* policy specific per-blkcg data size */
-	size_t				cpd_size;
 	/* cgroup files for the policy */
 	struct cftype			*cftypes;
 
 	/* operations */
+	blkcg_pol_alloc_cpd_fn		*cpd_alloc_fn;
 	blkcg_pol_init_cpd_fn		*cpd_init_fn;
+	blkcg_pol_free_cpd_fn		*cpd_free_fn;
+
 	blkcg_pol_alloc_pd_fn		*pd_alloc_fn;
 	blkcg_pol_init_pd_fn		*pd_init_fn;
 	blkcg_pol_online_pd_fn		*pd_online_fn;