[PATCH v2 0/4] Add MMC software queue support

[PATCH v2 0/4] Add MMC software queue support

[Baolin Wang]

Hi All,

Now the MMC read/write stack will always wait for previous request is
completed by mmc_blk_rw_wait(), before sending a new request to hardware,
or queue a work to complete request, that will bring context switching
overhead, especially for high I/O per second rates, to affect the IO
performance.

Thus this patch set will introduce the MMC software command queue support
based on command queue engine's interfaces, and set the queue depth as 2,
that means we do not need wait for previous request is completed and can
queue 2 requests in flight. It is enough to let the irq handler always
trigger the next request without a context switch and then ask the blk_mq
layer for the next one to get queued, as well as avoiding a long latency.

Moreover we can expand the MMC software queue interface to support
MMC packed request or packed command instead of adding new interfaces,
according to previosus discussion.

Below are some comparison data with fio tool. The fio command I used
is like below with changing the '--rw' parameter and enabling the direct
IO flag to measure the actual hardware transfer speed in 4K block size.

./fio --filename=/dev/mmcblk0p30 --direct=1 --iodepth=20 --rw=read --bs=4K --size=512M --group_reporting --numjobs=20 --name=test_read

My eMMC card working at HS400 Enhanced strobe mode:
[    2.229856] mmc0: new HS400 Enhanced strobe MMC card at address 0001
[    2.237566] mmcblk0: mmc0:0001 HBG4a2 29.1 GiB 
[    2.242621] mmcblk0boot0: mmc0:0001 HBG4a2 partition 1 4.00 MiB
[    2.249110] mmcblk0boot1: mmc0:0001 HBG4a2 partition 2 4.00 MiB
[    2.255307] mmcblk0rpmb: mmc0:0001 HBG4a2 partition 3 4.00 MiB, chardev (248:0)

1. Without MMC software queue
I tested 3 times for each case and output a average speed.

1) Sequential read:
Speed: 28.9MiB/s, 26.4MiB/s, 30.9MiB/s
Average speed: 28.7MiB/s

2) Random read:
Speed: 18.2MiB/s, 8.9MiB/s, 15.8MiB/s
Average speed: 14.3MiB/s

3) Sequential write:
Speed: 21.1MiB/s, 27.9MiB/s, 25MiB/s
Average speed: 24.7MiB/s

4) Random write:
Speed: 21.5MiB/s, 18.1MiB/s, 18.1MiB/s
Average speed: 19.2MiB/s

2. With MMC software queue
I tested 3 times for each case and output a average speed.

1) Sequential read:
Speed: 44.1MiB/s, 42.3MiB/s, 44.4MiB/s
Average speed: 43.6MiB/s

2) Random read:
Speed: 30.6MiB/s, 30.9MiB/s, 30.5MiB/s
Average speed: 30.6MiB/s

3) Sequential write:
Speed: 44.1MiB/s, 45.9MiB/s, 44.2MiB/s
Average speed: 44.7MiB/s

4) Random write:
Speed: 45.1MiB/s, 43.3MiB/s, 42.4MiB/s
Average speed: 43.6MiB/s

Form above data, we can see the MMC software queue can help to improve the
performance obviously.

Any comments are welcome. Thanks a lot.

Changes from v1:
 - Add request_done ops for sdhci_ops.
 - Replace virtual command queue with software queue for functions and
 variables.
 - Rename the software queue file and add sqhci.h header file.

Baolin Wang (4):
  mmc: host: cqhci: Move the struct cqhci_slot into header file
  mmc: Add MMC software queue support
  mmc: host: sdhci: Add request_done ops for struct sdhci_ops
  mmc: host: sdhci-sprd: Add software queue support

 drivers/mmc/core/block.c      |   61 ++++++++
 drivers/mmc/core/mmc.c        |   13 +-
 drivers/mmc/core/queue.c      |   25 ++-
 drivers/mmc/host/Kconfig      |    9 ++
 drivers/mmc/host/Makefile     |    1 +
 drivers/mmc/host/cqhci.c      |   10 --
 drivers/mmc/host/cqhci.h      |   17 +-
 drivers/mmc/host/sdhci-sprd.c |   26 ++++
 drivers/mmc/host/sdhci.c      |   12 +-
 drivers/mmc/host/sdhci.h      |    2 +
 drivers/mmc/host/sqhci.c      |  345 +++++++++++++++++++++++++++++++++++++++++
 drivers/mmc/host/sqhci.h      |   35 +++++
 include/linux/mmc/host.h      |    3 +
 13 files changed, 536 insertions(+), 23 deletions(-)
 create mode 100644 drivers/mmc/host/sqhci.c
 create mode 100644 drivers/mmc/host/sqhci.h

-- 
1.7.9.5

[PATCH v2 1/4] mmc: host: cqhci: Move the struct cqhci_slot into header file

[Baolin Wang]

The struct cqhci_slot will be used by MMC software queue introducing by
following patches, thus move it to the header file.

Signed-off-by: Baolin Wang <baolin.wang@linaro.org>
---
 drivers/mmc/host/cqhci.c |   10 ----------
 drivers/mmc/host/cqhci.h |   11 ++++++++++-
 2 files changed, 10 insertions(+), 11 deletions(-)

diff --git a/drivers/mmc/host/cqhci.c b/drivers/mmc/host/cqhci.c
index f7bdae5..57ff1cc 100644
--- a/drivers/mmc/host/cqhci.c
+++ b/drivers/mmc/host/cqhci.c
@@ -21,16 +21,6 @@
 #define DCMD_SLOT 31
 #define NUM_SLOTS 32
 
-struct cqhci_slot {
-	struct mmc_request *mrq;
-	unsigned int flags;
-#define CQHCI_EXTERNAL_TIMEOUT	BIT(0)
-#define CQHCI_COMPLETED		BIT(1)
-#define CQHCI_HOST_CRC		BIT(2)
-#define CQHCI_HOST_TIMEOUT	BIT(3)
-#define CQHCI_HOST_OTHER	BIT(4)
-};
-
 static inline u8 *get_desc(struct cqhci_host *cq_host, u8 tag)
 {
 	return cq_host->desc_base + (tag * cq_host->slot_sz);
diff --git a/drivers/mmc/host/cqhci.h b/drivers/mmc/host/cqhci.h
index def76e9..7b07bf24f 100644
--- a/drivers/mmc/host/cqhci.h
+++ b/drivers/mmc/host/cqhci.h
@@ -141,7 +141,16 @@
 struct cqhci_host_ops;
 struct mmc_host;
 struct mmc_request;
-struct cqhci_slot;
+
+struct cqhci_slot {
+	struct mmc_request *mrq;
+	unsigned int flags;
+#define CQHCI_EXTERNAL_TIMEOUT	BIT(0)
+#define CQHCI_COMPLETED		BIT(1)
+#define CQHCI_HOST_CRC		BIT(2)
+#define CQHCI_HOST_TIMEOUT	BIT(3)
+#define CQHCI_HOST_OTHER	BIT(4)
+};
 
 struct cqhci_host {
 	const struct cqhci_host_ops *ops;
-- 
1.7.9.5

[PATCH v2 2/4] mmc: Add MMC software queue support

[Baolin Wang]

Now the MMC read/write stack will always wait for previous request is
completed by mmc_blk_rw_wait(), before sending a new request to hardware,
or queue a work to complete request, that will bring context switching
overhead, especially for high I/O per second rates, to affect the IO
performance.

Thus this patch introduces MMC software queue interface based on the
hardware command queue engine's interfaces, which is similar with the
hardware command queue engine's idea, that can remove the context
switching. Moreover we set the queue depth as 2 for software queue,
that is enough to let the irq handler always trigger the next request
without a context switch and then ask the blk_mq layer for the next one
to get queued, as well as avoiding a long latency.

From the fio testing data in cover letter, we can see the software
queue can improve performance obviously with 4K block size, increasing
about 52% for sequential read, increasing about 114% for random read,
increasing about 81% for sequential write, and increasing about 127%
for random write.

Moreover we can expand the software queue interface to support MMC
packed request or packed command in future.

Signed-off-by: Baolin Wang <baolin.wang@linaro.org>
---
 drivers/mmc/core/block.c  |   61 ++++++++
 drivers/mmc/core/mmc.c    |   13 +-
 drivers/mmc/core/queue.c  |   25 +++-
 drivers/mmc/host/Kconfig  |    8 ++
 drivers/mmc/host/Makefile |    1 +
 drivers/mmc/host/cqhci.h  |    6 +
 drivers/mmc/host/sqhci.c  |  345 +++++++++++++++++++++++++++++++++++++++++++++
 drivers/mmc/host/sqhci.h  |   35 +++++
 include/linux/mmc/host.h  |    3 +
 9 files changed, 487 insertions(+), 10 deletions(-)
 create mode 100644 drivers/mmc/host/sqhci.c
 create mode 100644 drivers/mmc/host/sqhci.h

diff --git a/drivers/mmc/core/block.c b/drivers/mmc/core/block.c
index 2c71a43..870462c 100644
--- a/drivers/mmc/core/block.c
+++ b/drivers/mmc/core/block.c
@@ -168,6 +168,11 @@ struct mmc_rpmb_data {
 
 static inline int mmc_blk_part_switch(struct mmc_card *card,
 				      unsigned int part_type);
+static void mmc_blk_rw_rq_prep(struct mmc_queue_req *mqrq,
+			       struct mmc_card *card,
+			       int disable_multi,
+			       struct mmc_queue *mq);
+static void mmc_blk_swq_req_done(struct mmc_request *mrq);
 
 static struct mmc_blk_data *mmc_blk_get(struct gendisk *disk)
 {
@@ -1569,9 +1574,30 @@ static int mmc_blk_cqe_issue_flush(struct mmc_queue *mq, struct request *req)
 	return mmc_blk_cqe_start_req(mq->card->host, mrq);
 }
 
+static int mmc_blk_swq_issue_rw_rq(struct mmc_queue *mq, struct request *req)
+{
+	struct mmc_queue_req *mqrq = req_to_mmc_queue_req(req);
+	struct mmc_host *host = mq->card->host;
+	int err;
+
+	mmc_blk_rw_rq_prep(mqrq, mq->card, 0, mq);
+	mqrq->brq.mrq.done = mmc_blk_swq_req_done;
+	mmc_pre_req(host, &mqrq->brq.mrq);
+
+	err = mmc_cqe_start_req(host, &mqrq->brq.mrq);
+	if (err)
+		mmc_post_req(host, &mqrq->brq.mrq, err);
+
+	return err;
+}
+
 static int mmc_blk_cqe_issue_rw_rq(struct mmc_queue *mq, struct request *req)
 {
 	struct mmc_queue_req *mqrq = req_to_mmc_queue_req(req);
+	struct mmc_host *host = mq->card->host;
+
+	if (host->swq_enabled)
+		return mmc_blk_swq_issue_rw_rq(mq, req);
 
 	mmc_blk_data_prep(mq, mqrq, 0, NULL, NULL);
 
@@ -1957,6 +1983,41 @@ static void mmc_blk_urgent_bkops(struct mmc_queue *mq,
 		mmc_run_bkops(mq->card);
 }
 
+static void mmc_blk_swq_req_done(struct mmc_request *mrq)
+{
+	struct mmc_queue_req *mqrq =
+		container_of(mrq, struct mmc_queue_req, brq.mrq);
+	struct request *req = mmc_queue_req_to_req(mqrq);
+	struct request_queue *q = req->q;
+	struct mmc_queue *mq = q->queuedata;
+	struct mmc_host *host = mq->card->host;
+	unsigned long flags;
+
+	if (mmc_blk_rq_error(&mqrq->brq) ||
+	    mmc_blk_urgent_bkops_needed(mq, mqrq)) {
+		spin_lock_irqsave(&mq->lock, flags);
+		mq->recovery_needed = true;
+		mq->recovery_req = req;
+		spin_unlock_irqrestore(&mq->lock, flags);
+
+		host->cqe_ops->cqe_recovery_start(host);
+
+		schedule_work(&mq->recovery_work);
+		return;
+	}
+
+	mmc_blk_rw_reset_success(mq, req);
+
+	/*
+	 * Block layer timeouts race with completions which means the normal
+	 * completion path cannot be used during recovery.
+	 */
+	if (mq->in_recovery)
+		mmc_blk_cqe_complete_rq(mq, req);
+	else
+		blk_mq_complete_request(req);
+}
+
 void mmc_blk_mq_complete(struct request *req)
 {
 	struct mmc_queue *mq = req->q->queuedata;
diff --git a/drivers/mmc/core/mmc.c b/drivers/mmc/core/mmc.c
index c880489..8eac1a2 100644
--- a/drivers/mmc/core/mmc.c
+++ b/drivers/mmc/core/mmc.c
@@ -1852,15 +1852,22 @@ static int mmc_init_card(struct mmc_host *host, u32 ocr,
 	 */
 	card->reenable_cmdq = card->ext_csd.cmdq_en;
 
-	if (card->ext_csd.cmdq_en && !host->cqe_enabled) {
+	if (host->cqe_ops && !host->cqe_enabled) {
 		err = host->cqe_ops->cqe_enable(host, card);
 		if (err) {
 			pr_err("%s: Failed to enable CQE, error %d\n",
 				mmc_hostname(host), err);
 		} else {
 			host->cqe_enabled = true;
-			pr_info("%s: Command Queue Engine enabled\n",
-				mmc_hostname(host));
+
+			if (card->ext_csd.cmdq_en) {
+				pr_info("%s: Command Queue Engine enabled\n",
+					mmc_hostname(host));
+			} else {
+				host->swq_enabled = true;
+				pr_info("%s: Software Queue enabled\n",
+					mmc_hostname(host));
+			}
 		}
 	}
 
diff --git a/drivers/mmc/core/queue.c b/drivers/mmc/core/queue.c
index 7102e2e..2c93c29 100644
--- a/drivers/mmc/core/queue.c
+++ b/drivers/mmc/core/queue.c
@@ -60,7 +60,7 @@ enum mmc_issue_type mmc_issue_type(struct mmc_queue *mq, struct request *req)
 {
 	struct mmc_host *host = mq->card->host;
 
-	if (mq->use_cqe)
+	if (mq->use_cqe && !host->swq_enabled)
 		return mmc_cqe_issue_type(host, req);
 
 	if (req_op(req) == REQ_OP_READ || req_op(req) == REQ_OP_WRITE)
@@ -122,12 +122,14 @@ static enum blk_eh_timer_return mmc_mq_timed_out(struct request *req,
 {
 	struct request_queue *q = req->q;
 	struct mmc_queue *mq = q->queuedata;
+	struct mmc_card *card = mq->card;
+	struct mmc_host *host = card->host;
 	unsigned long flags;
 	int ret;
 
 	spin_lock_irqsave(&mq->lock, flags);
 
-	if (mq->recovery_needed || !mq->use_cqe)
+	if (mq->recovery_needed || !mq->use_cqe || host->swq_enabled)
 		ret = BLK_EH_RESET_TIMER;
 	else
 		ret = mmc_cqe_timed_out(req);
@@ -142,12 +144,13 @@ static void mmc_mq_recovery_handler(struct work_struct *work)
 	struct mmc_queue *mq = container_of(work, struct mmc_queue,
 					    recovery_work);
 	struct request_queue *q = mq->queue;
+	struct mmc_host *host = mq->card->host;
 
 	mmc_get_card(mq->card, &mq->ctx);
 
 	mq->in_recovery = true;
 
-	if (mq->use_cqe)
+	if (mq->use_cqe && !host->swq_enabled)
 		mmc_blk_cqe_recovery(mq);
 	else
 		mmc_blk_mq_recovery(mq);
@@ -158,6 +161,9 @@ static void mmc_mq_recovery_handler(struct work_struct *work)
 	mq->recovery_needed = false;
 	spin_unlock_irq(&mq->lock);
 
+	if (host->swq_enabled)
+		host->cqe_ops->cqe_recovery_finish(host);
+
 	mmc_put_card(mq->card, &mq->ctx);
 
 	blk_mq_run_hw_queues(q, true);
@@ -407,11 +413,16 @@ int mmc_init_queue(struct mmc_queue *mq, struct mmc_card *card)
 	 * The queue depth for CQE must match the hardware because the request
 	 * tag is used to index the hardware queue.
 	 */
-	if (mq->use_cqe)
-		mq->tag_set.queue_depth =
-			min_t(int, card->ext_csd.cmdq_depth, host->cqe_qdepth);
-	else
+	if (mq->use_cqe) {
+		if (host->swq_enabled)
+			mq->tag_set.queue_depth = host->cqe_qdepth;
+		else
+			mq->tag_set.queue_depth =
+				min_t(int, card->ext_csd.cmdq_depth, host->cqe_qdepth);
+	} else {
 		mq->tag_set.queue_depth = MMC_QUEUE_DEPTH;
+	}
+
 	mq->tag_set.numa_node = NUMA_NO_NODE;
 	mq->tag_set.flags = BLK_MQ_F_SHOULD_MERGE | BLK_MQ_F_BLOCKING;
 	mq->tag_set.nr_hw_queues = 1;
diff --git a/drivers/mmc/host/Kconfig b/drivers/mmc/host/Kconfig
index 14d89a1..d117f18 100644
--- a/drivers/mmc/host/Kconfig
+++ b/drivers/mmc/host/Kconfig
@@ -923,6 +923,14 @@ config MMC_CQHCI
 
 	  If unsure, say N.
 
+config MMC_SQHCI
+	bool "Software Queue Host Controller Interface support"
+	help
+	  This selects the Software Queue Host Controller Interface (SQHCI)
+	  support.
+
+	  If unsure, say N.
+
 config MMC_TOSHIBA_PCI
 	tristate "Toshiba Type A SD/MMC Card Interface Driver"
 	depends on PCI
diff --git a/drivers/mmc/host/Makefile b/drivers/mmc/host/Makefile
index 7357871..a3588d4 100644
--- a/drivers/mmc/host/Makefile
+++ b/drivers/mmc/host/Makefile
@@ -97,6 +97,7 @@ obj-$(CONFIG_MMC_SDHCI_BRCMSTB)		+= sdhci-brcmstb.o
 obj-$(CONFIG_MMC_SDHCI_OMAP)		+= sdhci-omap.o
 obj-$(CONFIG_MMC_SDHCI_SPRD)		+= sdhci-sprd.o
 obj-$(CONFIG_MMC_CQHCI)			+= cqhci.o
+obj-$(CONFIG_MMC_SQHCI)			+= sqhci.o
 
 ifeq ($(CONFIG_CB710_DEBUG),y)
 	CFLAGS-cb710-mmc	+= -DDEBUG
diff --git a/drivers/mmc/host/cqhci.h b/drivers/mmc/host/cqhci.h
index 7b07bf24f..1769715 100644
--- a/drivers/mmc/host/cqhci.h
+++ b/drivers/mmc/host/cqhci.h
@@ -205,6 +205,12 @@ struct cqhci_host {
 	struct completion halt_comp;
 	wait_queue_head_t wait_queue;
 	struct cqhci_slot *slot;
+
+#ifdef CONFIG_MMC_SQHCI
+	/* Software queue support */
+	int next_tag;
+	struct mmc_request *mrq;
+#endif
 };
 
 struct cqhci_host_ops {
diff --git a/drivers/mmc/host/sqhci.c b/drivers/mmc/host/sqhci.c
new file mode 100644
index 0000000..e9c73d1
--- /dev/null
+++ b/drivers/mmc/host/sqhci.c
@@ -0,0 +1,345 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * MMC software queue support based on command queue interfaces
+ *
+ * Copyright (C) 2019 Linaro, Inc.
+ * Author: Baolin Wang <baolin.wang@linaro.org>
+ */
+
+#include <linux/mmc/card.h>
+#include <linux/mmc/host.h>
+#include <linux/mmc/mmc.h>
+
+#include "sqhci.h"
+
+#define SQHCI_NUM_SLOTS		2
+#define SQHCI_INVALID_TAG	SQHCI_NUM_SLOTS
+
+static void sqhci_pump_requests(struct cqhci_host *sq_host)
+{
+	struct mmc_host *mmc = sq_host->mmc;
+	struct cqhci_slot *slot;
+	unsigned long flags;
+
+	spin_lock_irqsave(&sq_host->lock, flags);
+
+	/* Make sure we are not already running a request now */
+	if (sq_host->mrq) {
+		spin_unlock_irqrestore(&sq_host->lock, flags);
+		return;
+	}
+
+	/* Make sure there are remain requests need to pump */
+	if (!sq_host->qcnt || !sq_host->enabled) {
+		spin_unlock_irqrestore(&sq_host->lock, flags);
+		return;
+	}
+
+	slot = &sq_host->slot[sq_host->next_tag];
+	sq_host->mrq = slot->mrq;
+	sq_host->qcnt--;
+
+	spin_unlock_irqrestore(&sq_host->lock, flags);
+
+	mmc->ops->request(mmc, sq_host->mrq);
+}
+
+static void sqhci_update_next_tag(struct cqhci_host *sq_host, int remains)
+{
+	struct cqhci_slot *slot;
+	int tag;
+
+	/*
+	 * If there are no remain requests in software queue, then set a invalid
+	 * tag.
+	 */
+	if (!remains) {
+		sq_host->next_tag = SQHCI_INVALID_TAG;
+		return;
+	}
+
+	/*
+	 * Increasing the next tag and check if the corresponding request is
+	 * available, if yes, then we found a candidate request.
+	 */
+	if (++sq_host->next_tag != SQHCI_INVALID_TAG) {
+		slot = &sq_host->slot[sq_host->next_tag];
+		if (slot->mrq)
+			return;
+	}
+
+	/* Othersie we should iterate all slots to find a available tag. */
+	for (tag = 0; tag < SQHCI_NUM_SLOTS; tag++) {
+		slot = &sq_host->slot[tag];
+		if (slot->mrq)
+			break;
+	}
+
+	if (tag == SQHCI_NUM_SLOTS)
+		tag = SQHCI_INVALID_TAG;
+
+	sq_host->next_tag = tag;
+}
+
+static void sqhci_post_request(struct cqhci_host *sq_host)
+{
+	unsigned long flags;
+	int remains;
+
+	spin_lock_irqsave(&sq_host->lock, flags);
+
+	remains = sq_host->qcnt;
+	sq_host->mrq = NULL;
+
+	/* Update the next available tag to be queued. */
+	sqhci_update_next_tag(sq_host, remains);
+
+	if (sq_host->waiting_for_idle && !remains) {
+		sq_host->waiting_for_idle = false;
+		wake_up(&sq_host->wait_queue);
+	}
+
+	/* Do not pump new request in recovery mode. */
+	if (sq_host->recovery_halt) {
+		spin_unlock_irqrestore(&sq_host->lock, flags);
+		return;
+	}
+
+	spin_unlock_irqrestore(&sq_host->lock, flags);
+
+	 /*
+	  * Try to pump new request to host controller as fast as possible,
+	  * after completing previous request.
+	  */
+	if (remains > 0)
+		sqhci_pump_requests(sq_host);
+}
+
+/**
+ * sqhci_finalize_request - finalize one request if the request is done
+ * @mmc: the host controller
+ * @mrq: the request need to be finalized
+ *
+ * Return true if we finalized the corresponding request in software queue,
+ * otherwise return false.
+ */
+bool sqhci_finalize_request(struct mmc_host *mmc, struct mmc_request *mrq)
+{
+	struct cqhci_host *sq_host = mmc->cqe_private;
+	unsigned long flags;
+
+	spin_lock_irqsave(&sq_host->lock, flags);
+
+	if (!sq_host->enabled || !sq_host->mrq || sq_host->mrq != mrq) {
+		spin_unlock_irqrestore(&sq_host->lock, flags);
+		return false;
+	}
+
+	/*
+	 * Clear current completed slot request to make a room for new request.
+	 */
+	sq_host->slot[sq_host->next_tag].mrq = NULL;
+
+	spin_unlock_irqrestore(&sq_host->lock, flags);
+
+	mmc_cqe_request_done(mmc, sq_host->mrq);
+
+	sqhci_post_request(sq_host);
+
+	return true;
+}
+EXPORT_SYMBOL_GPL(sqhci_finalize_request);
+
+static void sqhci_recovery_start(struct mmc_host *mmc)
+{
+	struct cqhci_host *sq_host = mmc->cqe_private;
+	unsigned long flags;
+
+	spin_lock_irqsave(&sq_host->lock, flags);
+
+	sq_host->recovery_halt = true;
+
+	spin_unlock_irqrestore(&sq_host->lock, flags);
+}
+
+static void sqhci_recovery_finish(struct mmc_host *mmc)
+{
+	struct cqhci_host *sq_host = mmc->cqe_private;
+	int remains;
+
+	spin_lock_irq(&sq_host->lock);
+
+	sq_host->recovery_halt = false;
+	remains = sq_host->qcnt;
+
+	spin_unlock_irq(&sq_host->lock);
+
+	/*
+	 * Try to pump new request if there are request pending in software
+	 * queue after finishing recovery.
+	 */
+	if (remains > 0)
+		sqhci_pump_requests(sq_host);
+}
+
+static int sqhci_request(struct mmc_host *mmc, struct mmc_request *mrq)
+{
+	struct cqhci_host *sq_host = mmc->cqe_private;
+	int tag = mrq->tag;
+
+	spin_lock_irq(&sq_host->lock);
+
+	if (!sq_host->enabled) {
+		spin_unlock_irq(&sq_host->lock);
+		return -ESHUTDOWN;
+	}
+
+	/* Do not queue any new requests in recovery mode. */
+	if (sq_host->recovery_halt) {
+		spin_unlock_irq(&sq_host->lock);
+		return -EBUSY;
+	}
+
+	sq_host->slot[tag].mrq = mrq;
+
+	/*
+	 * Set the next tag as current request tag if no available
+	 * next tag.
+	 */
+	if (sq_host->next_tag == SQHCI_INVALID_TAG)
+		sq_host->next_tag = tag;
+
+	sq_host->qcnt++;
+
+	spin_unlock_irq(&sq_host->lock);
+
+	sqhci_pump_requests(sq_host);
+
+	return 0;
+}
+
+static void sqhci_post_req(struct mmc_host *mmc, struct mmc_request *mrq)
+{
+	if (mmc->ops->post_req)
+		mmc->ops->post_req(mmc, mrq, 0);
+}
+
+static bool sqhci_queue_is_idle(struct cqhci_host *sq_host, int *ret)
+{
+	bool is_idle;
+
+	spin_lock_irq(&sq_host->lock);
+
+	is_idle = (!sq_host->mrq && !sq_host->qcnt) ||
+		sq_host->recovery_halt;
+
+	*ret = sq_host->recovery_halt ? -EBUSY : 0;
+	sq_host->waiting_for_idle = !is_idle;
+
+	spin_unlock_irq(&sq_host->lock);
+
+	return is_idle;
+}
+
+static int sqhci_wait_for_idle(struct mmc_host *mmc)
+{
+	struct cqhci_host *sq_host = mmc->cqe_private;
+	int ret;
+
+	wait_event(sq_host->wait_queue,
+		   sqhci_queue_is_idle(sq_host, &ret));
+
+	return ret;
+}
+
+static void sqhci_disable(struct mmc_host *mmc)
+{
+	struct cqhci_host *sq_host = mmc->cqe_private;
+	u32 timeout = 500;
+	int ret;
+
+	spin_lock_irq(&sq_host->lock);
+
+	if (!sq_host->enabled) {
+		spin_unlock_irq(&sq_host->lock);
+		return;
+	}
+
+	spin_unlock_irq(&sq_host->lock);
+
+	ret = wait_event_timeout(sq_host->wait_queue,
+				 sqhci_queue_is_idle(sq_host, &ret),
+				 msecs_to_jiffies(timeout));
+	if (ret == 0) {
+		pr_warn("could not stop mmc software queue\n");
+		return;
+	}
+
+	spin_lock_irq(&sq_host->lock);
+
+	sq_host->enabled = false;
+
+	spin_unlock_irq(&sq_host->lock);
+}
+
+static int sqhci_enable(struct mmc_host *mmc, struct mmc_card *card)
+{
+	struct cqhci_host *sq_host = mmc->cqe_private;
+
+	spin_lock_irq(&sq_host->lock);
+
+	if (sq_host->enabled) {
+		spin_unlock_irq(&sq_host->lock);
+		return -EBUSY;
+	}
+
+	sq_host->enabled = true;
+
+	spin_unlock_irq(&sq_host->lock);
+
+	return 0;
+}
+
+static const struct mmc_cqe_ops sqhci_ops = {
+	.cqe_enable = sqhci_enable,
+	.cqe_disable = sqhci_disable,
+	.cqe_request = sqhci_request,
+	.cqe_post_req = sqhci_post_req,
+	.cqe_wait_for_idle = sqhci_wait_for_idle,
+	.cqe_recovery_start = sqhci_recovery_start,
+	.cqe_recovery_finish = sqhci_recovery_finish,
+};
+
+int sqhci_init(struct cqhci_host *sq_host, struct mmc_host *mmc)
+{
+	sq_host->num_slots = SQHCI_NUM_SLOTS;
+	sq_host->next_tag = SQHCI_INVALID_TAG;
+	mmc->cqe_qdepth = SQHCI_NUM_SLOTS;
+
+	sq_host->slot = devm_kcalloc(mmc_dev(mmc), sq_host->num_slots,
+				     sizeof(struct cqhci_slot), GFP_KERNEL);
+	if (!sq_host->slot)
+		return -ENOMEM;
+
+	sq_host->mmc = mmc;
+	sq_host->mmc->cqe_private = sq_host;
+	mmc->cqe_ops = &sqhci_ops;
+
+	spin_lock_init(&sq_host->lock);
+	init_waitqueue_head(&sq_host->wait_queue);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(sqhci_init);
+
+void sqhci_suspend(struct mmc_host *mmc)
+{
+	sqhci_disable(mmc);
+}
+EXPORT_SYMBOL_GPL(sqhci_suspend);
+
+int sqhci_resume(struct mmc_host *mmc)
+{
+	return sqhci_enable(mmc, NULL);
+}
+EXPORT_SYMBOL_GPL(sqhci_resume);
diff --git a/drivers/mmc/host/sqhci.h b/drivers/mmc/host/sqhci.h
new file mode 100644
index 0000000..fdaaed0
--- /dev/null
+++ b/drivers/mmc/host/sqhci.h
@@ -0,0 +1,35 @@
+// SPDX-License-Identifier: GPL-2.0
+#ifndef LINUX_MMC_SQHCI_H
+#define LINUX_MMC_SQHCI_H
+
+#include "cqhci.h"
+
+#ifdef CONFIG_MMC_SQHCI
+int sqhci_init(struct cqhci_host *cqv_host, struct mmc_host *mmc);
+void sqhci_suspend(struct mmc_host *mmc);
+int sqhci_resume(struct mmc_host *mmc);
+bool sqhci_finalize_request(struct mmc_host *mmc, struct mmc_request *mrq);
+#else
+static inline int sqhci_init(struct cqhci_host *cqv_host, struct mmc_host *mmc)
+{
+	return -EINVAL;
+}
+
+static inline void sqhci_suspend(struct mmc_host *mmc)
+{
+}
+
+static inline int sqhci_resume(struct mmc_host *mmc)
+{
+	return 0;
+}
+
+static inline bool sqhci_finalize_request(struct mmc_host *mmc,
+					  struct mmc_request *mrq)
+{
+	return false;
+}
+
+#endif
+
+#endif
diff --git a/include/linux/mmc/host.h b/include/linux/mmc/host.h
index 4a351cb..706e2bc 100644
--- a/include/linux/mmc/host.h
+++ b/include/linux/mmc/host.h
@@ -459,6 +459,9 @@ struct mmc_host {
 	bool			cqe_enabled;
 	bool			cqe_on;
 
+	/* Software Queue support */
+	bool			swq_enabled;
+
 	unsigned long		private[0] ____cacheline_aligned;
 };
 
-- 
1.7.9.5

[PATCH v2 3/4] mmc: host: sdhci: Add request_done ops for struct sdhci_ops

[Baolin Wang]

Add request_done ops for struct sdhci_ops as a preparation in case some
host controllers have different method to complete one request, such as
supporting request completion of MMC software queue.

Signed-off-by: Baolin Wang <baolin.wang@linaro.org>
---
 drivers/mmc/host/sdhci.c |   12 ++++++++++--
 drivers/mmc/host/sdhci.h |    2 ++
 2 files changed, 12 insertions(+), 2 deletions(-)

diff --git a/drivers/mmc/host/sdhci.c b/drivers/mmc/host/sdhci.c
index a5dc5aa..b2c8695 100644
--- a/drivers/mmc/host/sdhci.c
+++ b/drivers/mmc/host/sdhci.c
@@ -2710,7 +2710,10 @@ static bool sdhci_request_done(struct sdhci_host *host)
 
 	spin_unlock_irqrestore(&host->lock, flags);
 
-	mmc_request_done(host->mmc, mrq);
+	if (host->ops->request_done)
+		host->ops->request_done(host, mrq);
+	else
+		mmc_request_done(host->mmc, mrq);
 
 	return false;
 }
@@ -3133,7 +3136,12 @@ static irqreturn_t sdhci_irq(int irq, void *dev_id)
 
 	/* Process mrqs ready for immediate completion */
 	for (i = 0; i < SDHCI_MAX_MRQS; i++) {
-		if (mrqs_done[i])
+		if (!mrqs_done[i])
+			continue;
+
+		if (host->ops->request_done)
+			host->ops->request_done(host, mrqs_done[i]);
+		else
 			mmc_request_done(host->mmc, mrqs_done[i]);
 	}
 
diff --git a/drivers/mmc/host/sdhci.h b/drivers/mmc/host/sdhci.h
index 902f855..e9a476f 100644
--- a/drivers/mmc/host/sdhci.h
+++ b/drivers/mmc/host/sdhci.h
@@ -643,6 +643,8 @@ struct sdhci_ops {
 	void	(*voltage_switch)(struct sdhci_host *host);
 	void	(*adma_write_desc)(struct sdhci_host *host, void **desc,
 				   dma_addr_t addr, int len, unsigned int cmd);
+	void	(*request_done)(struct sdhci_host *host,
+				struct mmc_request *mrq);
 };
 
 #ifdef CONFIG_MMC_SDHCI_IO_ACCESSORS
-- 
1.7.9.5

[PATCH v2 4/4] mmc: host: sdhci-sprd: Add software queue support

[Baolin Wang]

Add software queue support to improve the performance.

Signed-off-by: Baolin Wang <baolin.wang@linaro.org>
---
 drivers/mmc/host/Kconfig      |    1 +
 drivers/mmc/host/sdhci-sprd.c |   26 ++++++++++++++++++++++++++
 2 files changed, 27 insertions(+)

diff --git a/drivers/mmc/host/Kconfig b/drivers/mmc/host/Kconfig
index d117f18..862e8e9 100644
--- a/drivers/mmc/host/Kconfig
+++ b/drivers/mmc/host/Kconfig
@@ -619,6 +619,7 @@ config MMC_SDHCI_SPRD
 	depends on ARCH_SPRD
 	depends on MMC_SDHCI_PLTFM
 	select MMC_SDHCI_IO_ACCESSORS
+	select MMC_SQHCI
 	help
 	  This selects the SDIO Host Controller in Spreadtrum
 	  SoCs, this driver supports R11(IP version: R11P0).
diff --git a/drivers/mmc/host/sdhci-sprd.c b/drivers/mmc/host/sdhci-sprd.c
index d07b979..f6f157f 100644
--- a/drivers/mmc/host/sdhci-sprd.c
+++ b/drivers/mmc/host/sdhci-sprd.c
@@ -19,6 +19,7 @@
 #include <linux/slab.h>
 
 #include "sdhci-pltfm.h"
+#include "sqhci.h"
 
 /* SDHCI_ARGUMENT2 register high 16bit */
 #define SDHCI_SPRD_ARG2_STUFF		GENMASK(31, 16)
@@ -379,6 +380,16 @@ static unsigned int sdhci_sprd_get_ro(struct sdhci_host *host)
 	return 0;
 }
 
+static void sdhci_sprd_request_done(struct sdhci_host *host,
+				    struct mmc_request *mrq)
+{
+	/* Validate if the request was from software queue firstly. */
+	if (sqhci_finalize_request(host->mmc, mrq))
+		return;
+
+	 mmc_request_done(host->mmc, mrq);
+}
+
 static struct sdhci_ops sdhci_sprd_ops = {
 	.read_l = sdhci_sprd_readl,
 	.write_l = sdhci_sprd_writel,
@@ -392,6 +403,7 @@ static unsigned int sdhci_sprd_get_ro(struct sdhci_host *host)
 	.hw_reset = sdhci_sprd_hw_reset,
 	.get_max_timeout_count = sdhci_sprd_get_max_timeout_count,
 	.get_ro = sdhci_sprd_get_ro,
+	.request_done = sdhci_sprd_request_done,
 };
 
 static void sdhci_sprd_request(struct mmc_host *mmc, struct mmc_request *mrq)
@@ -521,6 +533,7 @@ static int sdhci_sprd_probe(struct platform_device *pdev)
 {
 	struct sdhci_host *host;
 	struct sdhci_sprd_host *sprd_host;
+	struct cqhci_host *sq_host;
 	struct clk *clk;
 	int ret = 0;
 
@@ -631,6 +644,16 @@ static int sdhci_sprd_probe(struct platform_device *pdev)
 
 	sprd_host->flags = host->flags;
 
+	sq_host = devm_kzalloc(&pdev->dev, sizeof(*sq_host), GFP_KERNEL);
+	if (!sq_host) {
+		ret = -ENOMEM;
+		goto err_cleanup_host;
+	}
+
+	ret = sqhci_init(sq_host, host->mmc);
+	if (ret)
+		goto err_cleanup_host;
+
 	ret = __sdhci_add_host(host);
 	if (ret)
 		goto err_cleanup_host;
@@ -689,6 +712,7 @@ static int sdhci_sprd_runtime_suspend(struct device *dev)
 	struct sdhci_host *host = dev_get_drvdata(dev);
 	struct sdhci_sprd_host *sprd_host = TO_SPRD_HOST(host);
 
+	sqhci_suspend(host->mmc);
 	sdhci_runtime_suspend_host(host);
 
 	clk_disable_unprepare(sprd_host->clk_sdio);
@@ -717,6 +741,8 @@ static int sdhci_sprd_runtime_resume(struct device *dev)
 		goto clk_disable;
 
 	sdhci_runtime_resume_host(host, 1);
+	sqhci_resume(host->mmc);
+
 	return 0;
 
 clk_disable:
-- 
1.7.9.5