diff --git a/block/Kconfig b/block/Kconfig
index 1f2469a0123ceb1f36103c0db9eb71d140556193..2b352d772793d4fda8ea16a8b2790fa9bfad101f 100644
--- a/block/Kconfig
+++ b/block/Kconfig
@@ -50,6 +50,9 @@ config LBDAF
 
 	  If unsure, say Y.
 
+config BLK_RQ_ALLOC_TIME
+	bool
+
 config BLK_SCSI_REQUEST
 	bool
 
diff --git a/block/blk-mq.c b/block/blk-mq.c
index 684acaa96db7e11893b8ddd56b05de75e05ffaef..39e2e46214728ec425c024fabac2e4a221474bf2 100644
--- a/block/blk-mq.c
+++ b/block/blk-mq.c
@@ -275,7 +275,7 @@ bool blk_mq_can_queue(struct blk_mq_hw_ctx *hctx)
 EXPORT_SYMBOL(blk_mq_can_queue);
 
 static struct request *blk_mq_rq_ctx_init(struct blk_mq_alloc_data *data,
-		unsigned int tag, unsigned int op)
+		unsigned int tag, unsigned int op, u64 alloc_time_ns)
 {
 	struct blk_mq_tags *tags = blk_mq_tags_from_data(data);
 	struct request *rq = tags->static_rqs[tag];
@@ -309,6 +309,9 @@ static struct request *blk_mq_rq_ctx_init(struct blk_mq_alloc_data *data,
 	RB_CLEAR_NODE(&rq->rb_node);
 	rq->rq_disk = NULL;
 	rq->part = NULL;
+#ifdef CONFIG_BLK_RQ_ALLOC_TIME
+	rq->alloc_time_ns = alloc_time_ns;
+#endif
 	rq->start_time_ns = ktime_get_ns();
 	rq->io_start_time_ns = 0;
 	rq->nr_phys_segments = 0;
@@ -344,8 +347,14 @@ static struct request *blk_mq_get_request(struct request_queue *q,
 	struct request *rq;
 	unsigned int tag;
 	bool put_ctx_on_error = false;
+	u64 alloc_time_ns = 0;
 
 	blk_queue_enter_live(q);
+
+	/* alloc_time includes depth and tag waits */
+	if (blk_queue_rq_alloc_time(q))
+		alloc_time_ns = ktime_get_ns();
+
 	data->q = q;
 	if (likely(!data->ctx)) {
 		data->ctx = blk_mq_get_ctx(q);
@@ -381,7 +390,7 @@ static struct request *blk_mq_get_request(struct request_queue *q,
 		return NULL;
 	}
 
-	rq = blk_mq_rq_ctx_init(data, tag, op);
+	rq = blk_mq_rq_ctx_init(data, tag, op, alloc_time_ns);
 	if (!op_is_flush(op)) {
 		rq->elv.icq = NULL;
 		if (e && e->type->ops.mq.prepare_request) {
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index d51e10f50e7551944adea98100499bdb5aac9218..ea647c91c4480411ec9db9f65f8063d5c68ffc8b 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -212,7 +212,11 @@ struct request {
 
 	struct gendisk *rq_disk;
 	struct hd_struct *part;
-	/* Time that I/O was submitted to the kernel. */
+#ifdef CONFIG_BLK_RQ_ALLOC_TIME
+	/* Time that the first bio started allocating this request. */
+	u64 alloc_time_ns;
+#endif
+	/* Time that this request was allocated for this IO. */
 	u64 start_time_ns;
 	/* Time that I/O was submitted to the device. */
 	u64 io_start_time_ns;
@@ -704,6 +708,7 @@ struct request_queue {
 #define QUEUE_FLAG_REGISTERED  26	/* queue has been registered to a disk */
 #define QUEUE_FLAG_SCSI_PASSTHROUGH 27	/* queue supports SCSI commands */
 #define QUEUE_FLAG_QUIESCED    28	/* queue has been quiesced */
+#define QUEUE_FLAG_RQ_ALLOC_TIME 29	/* record rq->alloc_time_ns */
 
 #define QUEUE_FLAG_DEFAULT	((1 << QUEUE_FLAG_IO_STAT) |		\
 				 (1 << QUEUE_FLAG_SAME_COMP)	|	\
@@ -736,6 +741,12 @@ bool blk_queue_flag_test_and_clear(unsigned int flag, struct request_queue *q);
 #define blk_queue_dax(q)	test_bit(QUEUE_FLAG_DAX, &(q)->queue_flags)
 #define blk_queue_scsi_passthrough(q)	\
 	test_bit(QUEUE_FLAG_SCSI_PASSTHROUGH, &(q)->queue_flags)
+#ifdef CONFIG_BLK_RQ_ALLOC_TIME
+#define blk_queue_rq_alloc_time(q)	\
+	test_bit(QUEUE_FLAG_RQ_ALLOC_TIME, &(q)->queue_flags)
+#else
+#define blk_queue_rq_alloc_time(q)	false
+#endif
 
 #define blk_noretry_request(rq) \
 	((rq)->cmd_flags & (REQ_FAILFAST_DEV|REQ_FAILFAST_TRANSPORT| \