[PATCH v4 01/12] blk-mq: Reduce blk_mq_hw_ctx size

[Bart Van Assche <bart.vanassche@xxxxxxx>]

From: Bart Van Assche <bart.vanassche@xxxxxxxxxxx>

Since the srcu structure is rather large (184 bytes on an x86-64
system with kernel debugging disabled), only allocate it if needed.

Reported-by: Ming Lei <ming.lei@xxxxxxxxxx>
Signed-off-by: Bart Van Assche <bart.vanassche@xxxxxxxxxxx>
Reviewed-by: Christoph Hellwig <hch@xxxxxx>
Reviewed-by: Ming Lei <ming.lei@xxxxxxxxxx>
Cc: Hannes Reinecke <hare@xxxxxxxx>
Cc: Omar Sandoval <osandov@xxxxxx>
---
 block/blk-mq.c         | 30 ++++++++++++++++++++++--------
 include/linux/blk-mq.h |  5 +++--
 2 files changed, 25 insertions(+), 10 deletions(-)

diff --git a/block/blk-mq.c b/block/blk-mq.c
index 80b14457cf6c..2dff85cdacba 100644
--- a/block/blk-mq.c
+++ b/block/blk-mq.c
@@ -186,7 +186,7 @@ void blk_mq_quiesce_queue(struct request_queue *q)
 
 	queue_for_each_hw_ctx(q, hctx, i) {
 		if (hctx->flags & BLK_MQ_F_BLOCKING)
-			synchronize_srcu(&hctx->queue_rq_srcu);
+			synchronize_srcu(hctx->queue_rq_srcu);
 		else
 			rcu = true;
 	}
@@ -1106,9 +1106,9 @@ static void __blk_mq_run_hw_queue(struct blk_mq_hw_ctx *hctx)
 	} else {
 		might_sleep();
 
-		srcu_idx = srcu_read_lock(&hctx->queue_rq_srcu);
+		srcu_idx = srcu_read_lock(hctx->queue_rq_srcu);
 		blk_mq_sched_dispatch_requests(hctx);
-		srcu_read_unlock(&hctx->queue_rq_srcu, srcu_idx);
+		srcu_read_unlock(hctx->queue_rq_srcu, srcu_idx);
 	}
 }
 
@@ -1517,9 +1517,9 @@ static void blk_mq_try_issue_directly(struct blk_mq_hw_ctx *hctx,
 
 		might_sleep();
 
-		srcu_idx = srcu_read_lock(&hctx->queue_rq_srcu);
+		srcu_idx = srcu_read_lock(hctx->queue_rq_srcu);
 		__blk_mq_try_issue_directly(hctx, rq, cookie, true);
-		srcu_read_unlock(&hctx->queue_rq_srcu, srcu_idx);
+		srcu_read_unlock(hctx->queue_rq_srcu, srcu_idx);
 	}
 }
 
@@ -1863,7 +1863,7 @@ static void blk_mq_exit_hctx(struct request_queue *q,
 		set->ops->exit_hctx(hctx, hctx_idx);
 
 	if (hctx->flags & BLK_MQ_F_BLOCKING)
-		cleanup_srcu_struct(&hctx->queue_rq_srcu);
+		cleanup_srcu_struct(hctx->queue_rq_srcu);
 
 	blk_mq_remove_cpuhp(hctx);
 	blk_free_flush_queue(hctx->fq);
@@ -1936,7 +1936,7 @@ static int blk_mq_init_hctx(struct request_queue *q,
 		goto free_fq;
 
 	if (hctx->flags & BLK_MQ_F_BLOCKING)
-		init_srcu_struct(&hctx->queue_rq_srcu);
+		init_srcu_struct(hctx->queue_rq_srcu);
 
 	blk_mq_debugfs_register_hctx(q, hctx);
 
@@ -2211,6 +2211,20 @@ struct request_queue *blk_mq_init_queue(struct blk_mq_tag_set *set)
 }
 EXPORT_SYMBOL(blk_mq_init_queue);
 
+static int blk_mq_hw_ctx_size(struct blk_mq_tag_set *tag_set)
+{
+	int hw_ctx_size = sizeof(struct blk_mq_hw_ctx);
+
+	BUILD_BUG_ON(ALIGN(offsetof(struct blk_mq_hw_ctx, queue_rq_srcu),
+			   __alignof__(struct blk_mq_hw_ctx)) !=
+		     sizeof(struct blk_mq_hw_ctx));
+
+	if (tag_set->flags & BLK_MQ_F_BLOCKING)
+		hw_ctx_size += sizeof(struct srcu_struct);
+
+	return hw_ctx_size;
+}
+
 static void blk_mq_realloc_hw_ctxs(struct blk_mq_tag_set *set,
 						struct request_queue *q)
 {
@@ -2225,7 +2239,7 @@ static void blk_mq_realloc_hw_ctxs(struct blk_mq_tag_set *set,
 			continue;
 
 		node = blk_mq_hw_queue_to_node(q->mq_map, i);
-		hctxs[i] = kzalloc_node(sizeof(struct blk_mq_hw_ctx),
+		hctxs[i] = kzalloc_node(blk_mq_hw_ctx_size(set),
 					GFP_KERNEL, node);
 		if (!hctxs[i])
 			break;
diff --git a/include/linux/blk-mq.h b/include/linux/blk-mq.h
index ed1b43045ed6..384a448a92a8 100644
--- a/include/linux/blk-mq.h
+++ b/include/linux/blk-mq.h
@@ -39,8 +39,6 @@ struct blk_mq_hw_ctx {
 	struct blk_mq_tags	*tags;
 	struct blk_mq_tags	*sched_tags;
 
-	struct srcu_struct	queue_rq_srcu;
-
 	unsigned long		queued;
 	unsigned long		run;
 #define BLK_MQ_MAX_DISPATCH_ORDER	7
@@ -62,6 +60,9 @@ struct blk_mq_hw_ctx {
 	struct dentry		*debugfs_dir;
 	struct dentry		*sched_debugfs_dir;
 #endif
+
+	/* Must be the last member - see also blk_mq_hw_ctx_size(). */
+	struct srcu_struct	queue_rq_srcu[0];
 };
 
 struct blk_mq_tag_set {
-- 
2.13.1