[PATCH V8 00/14] mmc: Add Command Queue support

[PATCH V8 00/14] mmc: Add Command Queue support

[Adrian Hunter]

Hi

Here is V8 of the hardware command queue patches without the software
command queue patches, now using blk-mq and now with blk-mq support for
non-CQE I/O.

After the unacceptable debacle of the last release cycle, I expect an
immediate response to these patches.

HW CMDQ offers 25% - 50% better random multi-threaded I/O.  I see a slight
2% drop in sequential read speed but no change to sequential write.

Non-CQE blk-mq showed a 3% decrease in sequential read performance.  This
seemed to be coming from the inferior latency of running work items compared
with a dedicated thread.  Hacking blk-mq workqueue to be unbound reduced the
performance degradation from 3% to 1%.

While we should look at changing blk-mq to give better workqueue performance,
a bigger gain is likely to be made by adding a new host API to enable the
next already-prepared request to be issued directly from within ->done()
callback of the current request.


Changes since V7:
	Re-based
      mmc: core: Introduce host claiming by context
	Slightly simplified
      mmc: core: Add parameter use_blk_mq
	New patch.
      mmc: core: Remove unnecessary host claim
	New patch.
      mmc: core: Export mmc_start_bkops()
	New patch.
      mmc: core: Export mmc_start_request()
	New patch.
      mmc: block: Add CQE and blk-mq support
	Add blk-mq support for non_CQE requests

Changes since V6:
      mmc: core: Introduce host claiming by context
	New patch.
      mmc: core: Move mmc_start_areq() declaration
	Dropped because it has been applied
      mmc: block: Fix block status codes
	Dropped because it has been applied
      mmc: host: Add CQE interface
	Dropped because it has been applied
      mmc: core: Turn off CQE before sending commands
	Dropped because it has been applied
      mmc: block: Factor out mmc_setup_queue()
	New patch.
      mmc: block: Add CQE support
	Drop legacy support and add blk-mq support

Changes since V5:
	Re-based
      mmc: core: Add mmc_retune_hold_now()
	Dropped because it has been applied
      mmc: core: Add members to mmc_request and mmc_data for CQE's
	Dropped because it has been applied
      mmc: core: Move mmc_start_areq() declaration
	New patch at Ulf's request
      mmc: block: Fix block status codes
	Another un-related patch
      mmc: host: Add CQE interface
	Move recovery_notifier() callback to struct mmc_request
      mmc: core: Add support for handling CQE requests
	Roll __mmc_cqe_request_done() into mmc_cqe_request_done()
	Move function declarations requested by Ulf
      mmc: core: Remove unused MMC_CAP2_PACKED_CMD
	Dropped because it has been applied
      mmc: block: Add CQE support
	Add explanation to commit message
	Adjustment for changed recovery_notifier() callback
      mmc: cqhci: support for command queue enabled host
	Adjustment for changed recovery_notifier() callback
      mmc: sdhci-pci: Add CQHCI support for Intel GLK
	Add DCMD capability for Intel controllers except GLK

Changes since V4:
      mmc: core: Add mmc_retune_hold_now()
	Add explanation to commit message.
      mmc: host: Add CQE interface
	Add comments to callback declarations.
      mmc: core: Turn off CQE before sending commands
	Add explanation to commit message.
      mmc: core: Add support for handling CQE requests
	Add comments as requested by Ulf.
      mmc: core: Remove unused MMC_CAP2_PACKED_CMD
	New patch.
      mmc: mmc: Enable Command Queuing
	Adjust for removal of MMC_CAP2_PACKED_CMD.
	Add a comment about Packed Commands.
      mmc: mmc: Enable CQE's
	Remove un-necessary check for MMC_CAP2_CQE
      mmc: block: Use local variables in mmc_blk_data_prep()
	New patch.
      mmc: block: Prepare CQE data
	Adjust due to "mmc: block: Use local variables in mmc_blk_data_prep()"
	Remove priority setting.
	Add explanation to commit message.
      mmc: cqhci: support for command queue enabled host
	Fix transfer descriptor setting in cqhci_set_tran_desc() for 32-bit DMA

Changes since V3:
	Adjusted ...blk_end_request...() for new block status codes
	Fixed CQHCI transaction descriptor for "no DCMD" case

Changes since V2:
	Dropped patches that have been applied.
	Re-based
	Added "mmc: sdhci-pci: Add CQHCI support for Intel GLK"

Changes since V1:

	"Share mmc request array between partitions" is dependent
	on changes in "Introduce queue semantics", so added that
	and block fixes:

	Added "Fix is_waiting_last_req set incorrectly"
	Added "Fix cmd error reset failure path"
	Added "Use local var for mqrq_cur"
	Added "Introduce queue semantics"

Changes since RFC:

	Re-based on next.
	Added comment about command queue priority.
	Added some acks and reviews.


Adrian Hunter (13):
      mmc: core: Introduce host claiming by context
      mmc: core: Add support for handling CQE requests
      mmc: mmc: Enable Command Queuing
      mmc: mmc: Enable CQE's
      mmc: block: Use local variables in mmc_blk_data_prep()
      mmc: block: Prepare CQE data
      mmc: block: Factor out mmc_setup_queue()
      mmc: core: Add parameter use_blk_mq
      mmc: core: Remove unnecessary host claim
      mmc: core: Export mmc_start_bkops()
      mmc: core: Export mmc_start_request()
      mmc: block: Add CQE and blk-mq support
      mmc: sdhci-pci: Add CQHCI support for Intel GLK

Venkat Gopalakrishnan (1):
      mmc: cqhci: support for command queue enabled host

 drivers/mmc/Kconfig               |   11 +
 drivers/mmc/core/block.c          |  777 ++++++++++++++++++++++++-
 drivers/mmc/core/block.h          |    8 +
 drivers/mmc/core/bus.c            |    7 +
 drivers/mmc/core/core.c           |  265 ++++++++-
 drivers/mmc/core/core.h           |   14 +
 drivers/mmc/core/host.c           |    2 +
 drivers/mmc/core/host.h           |    4 +
 drivers/mmc/core/mmc.c            |   29 +
 drivers/mmc/core/mmc_ops.c        |    6 +-
 drivers/mmc/core/queue.c          |  479 +++++++++++++--
 drivers/mmc/core/queue.h          |   54 +-
 drivers/mmc/host/Kconfig          |   14 +
 drivers/mmc/host/Makefile         |    1 +
 drivers/mmc/host/cqhci.c          | 1154 +++++++++++++++++++++++++++++++++++++
 drivers/mmc/host/cqhci.h          |  240 ++++++++
 drivers/mmc/host/sdhci-pci-core.c |  154 ++++-
 include/linux/mmc/host.h          |   10 +-
 18 files changed, 3146 insertions(+), 83 deletions(-)
 create mode 100644 drivers/mmc/host/cqhci.c
 create mode 100644 drivers/mmc/host/cqhci.h


Regards
Adrian

[PATCH V8 01/14] mmc: core: Introduce host claiming by context

[Adrian Hunter]

Currently the host can be claimed by a task.  Change this so that the host
can be claimed by a context that may or may not be a task.  This provides
for the host to be claimed by a block driver queue to support blk-mq, while
maintaining compatibility with the existing use of mmc_claim_host().

Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
---
 drivers/mmc/core/core.c  | 92 ++++++++++++++++++++++++++++++++++++++++++++----
 drivers/mmc/core/core.h  |  6 ++++
 include/linux/mmc/host.h |  7 +++-
 3 files changed, 97 insertions(+), 8 deletions(-)

diff --git a/drivers/mmc/core/core.c b/drivers/mmc/core/core.c
index 66c9cf49ad2f..09fdc467b804 100644
--- a/drivers/mmc/core/core.c
+++ b/drivers/mmc/core/core.c
@@ -832,9 +832,38 @@ unsigned int mmc_align_data_size(struct mmc_card *card, unsigned int sz)
 }
 EXPORT_SYMBOL(mmc_align_data_size);
 
+/*
+ * Allow claiming an already claimed host if the context is the same or there is
+ * no context but the task is the same.
+ */
+static inline bool mmc_ctx_matches(struct mmc_host *host, struct mmc_ctx *ctx,
+				   struct task_struct *task)
+{
+	return host->claimer == ctx ||
+	       (!ctx && task && host->claimer->task == task);
+}
+
+static inline void mmc_ctx_set_claimer(struct mmc_host *host,
+				       struct mmc_ctx *ctx,
+				       struct task_struct *task)
+{
+	if (!host->claimer) {
+		if (ctx)
+			host->claimer = ctx;
+		else
+			host->claimer = &host->default_ctx;
+	}
+	if (task)
+		host->claimer->task = task;
+}
+
 /**
- *	__mmc_claim_host - exclusively claim a host
+ *	__mmc_ctx_task_claim_host - exclusively claim a host
  *	@host: mmc host to claim
+ *	@ctx: context that claims the host or NULL in which case the default
+ *	context will be used
+ *	@task: task that claims the host or NULL in which case @ctx must be
+ *	provided
  *	@abort: whether or not the operation should be aborted
  *
  *	Claim a host for a set of operations.  If @abort is non null and
@@ -842,7 +871,8 @@ unsigned int mmc_align_data_size(struct mmc_card *card, unsigned int sz)
  *	that non-zero value without acquiring the lock.  Returns zero
  *	with the lock held otherwise.
  */
-int __mmc_claim_host(struct mmc_host *host, atomic_t *abort)
+static int __mmc_ctx_task_claim_host(struct mmc_host *host, struct mmc_ctx *ctx,
+				     struct task_struct *task, atomic_t *abort)
 {
 	DECLARE_WAITQUEUE(wait, current);
 	unsigned long flags;
@@ -856,7 +886,7 @@ int __mmc_claim_host(struct mmc_host *host, atomic_t *abort)
 	while (1) {
 		set_current_state(TASK_UNINTERRUPTIBLE);
 		stop = abort ? atomic_read(abort) : 0;
-		if (stop || !host->claimed || host->claimer == current)
+		if (stop || !host->claimed || mmc_ctx_matches(host, ctx, task))
 			break;
 		spin_unlock_irqrestore(&host->lock, flags);
 		schedule();
@@ -865,7 +895,7 @@ int __mmc_claim_host(struct mmc_host *host, atomic_t *abort)
 	set_current_state(TASK_RUNNING);
 	if (!stop) {
 		host->claimed = 1;
-		host->claimer = current;
+		mmc_ctx_set_claimer(host, ctx, task);
 		host->claim_cnt += 1;
 		if (host->claim_cnt == 1)
 			pm = true;
@@ -879,8 +909,19 @@ int __mmc_claim_host(struct mmc_host *host, atomic_t *abort)
 
 	return stop;
 }
+
+int __mmc_claim_host(struct mmc_host *host, atomic_t *abort)
+{
+	return __mmc_ctx_task_claim_host(host, NULL, current, abort);
+}
 EXPORT_SYMBOL(__mmc_claim_host);
 
+void mmc_ctx_claim_host(struct mmc_host *host, struct mmc_ctx *ctx)
+{
+	__mmc_ctx_task_claim_host(host, ctx, NULL, NULL);
+}
+EXPORT_SYMBOL(mmc_ctx_claim_host);
+
 /**
  *	mmc_release_host - release a host
  *	@host: mmc host to release
@@ -888,18 +929,17 @@ int __mmc_claim_host(struct mmc_host *host, atomic_t *abort)
  *	Release a MMC host, allowing others to claim the host
  *	for their operations.
  */
-void mmc_release_host(struct mmc_host *host)
+static void __mmc_release_host(struct mmc_host *host)
 {
 	unsigned long flags;
 
-	WARN_ON(!host->claimed);
-
 	spin_lock_irqsave(&host->lock, flags);
 	if (--host->claim_cnt) {
 		/* Release for nested claim */
 		spin_unlock_irqrestore(&host->lock, flags);
 	} else {
 		host->claimed = 0;
+		host->claimer->task = NULL;
 		host->claimer = NULL;
 		spin_unlock_irqrestore(&host->lock, flags);
 		wake_up(&host->wq);
@@ -907,8 +947,23 @@ void mmc_release_host(struct mmc_host *host)
 		pm_runtime_put_autosuspend(mmc_dev(host));
 	}
 }
+
+void mmc_release_host(struct mmc_host *host)
+{
+	WARN_ON(!host->claimed);
+
+	__mmc_release_host(host);
+}
 EXPORT_SYMBOL(mmc_release_host);
 
+void mmc_ctx_release_host(struct mmc_host *host, struct mmc_ctx *ctx)
+{
+	WARN_ON(!host->claimed || host->claimer != ctx);
+
+	__mmc_release_host(host);
+}
+EXPORT_SYMBOL(mmc_ctx_release_host);
+
 /*
  * This is a helper function, which fetches a runtime pm reference for the
  * card device and also claims the host.
@@ -921,6 +976,17 @@ void mmc_get_card(struct mmc_card *card)
 EXPORT_SYMBOL(mmc_get_card);
 
 /*
+ * This is a helper function, which fetches a runtime pm reference for the
+ * card device and also claims the host for the mmc context.
+ */
+void mmc_ctx_get_card(struct mmc_card *card, struct mmc_ctx *ctx)
+{
+	pm_runtime_get_sync(&card->dev);
+	mmc_ctx_claim_host(card->host, ctx);
+}
+EXPORT_SYMBOL(mmc_ctx_get_card);
+
+/*
  * This is a helper function, which releases the host and drops the runtime
  * pm reference for the card device.
  */
@@ -933,6 +999,18 @@ void mmc_put_card(struct mmc_card *card)
 EXPORT_SYMBOL(mmc_put_card);
 
 /*
+ * This is a helper function, which releases the host and drops the runtime
+ * pm reference for the card device.
+ */
+void mmc_ctx_put_card(struct mmc_card *card, struct mmc_ctx *ctx)
+{
+	mmc_ctx_release_host(card->host, ctx);
+	pm_runtime_mark_last_busy(&card->dev);
+	pm_runtime_put_autosuspend(&card->dev);
+}
+EXPORT_SYMBOL(mmc_ctx_put_card);
+
+/*
  * Internal function that does the actual ios call to the host driver,
  * optionally printing some debug output.
  */
diff --git a/drivers/mmc/core/core.h b/drivers/mmc/core/core.h
index ca861091a776..a294c836a143 100644
--- a/drivers/mmc/core/core.h
+++ b/drivers/mmc/core/core.h
@@ -144,4 +144,10 @@ static inline void mmc_claim_host(struct mmc_host *host)
 	__mmc_claim_host(host, NULL);
 }
 
+void mmc_ctx_claim_host(struct mmc_host *host, struct mmc_ctx *ctx);
+void mmc_ctx_release_host(struct mmc_host *host, struct mmc_ctx *ctx);
+
+void mmc_ctx_get_card(struct mmc_card *card, struct mmc_ctx *ctx);
+void mmc_ctx_put_card(struct mmc_card *card, struct mmc_ctx *ctx);
+
 #endif
diff --git a/include/linux/mmc/host.h b/include/linux/mmc/host.h
index f3f2d07feb2a..228a493c3b25 100644
--- a/include/linux/mmc/host.h
+++ b/include/linux/mmc/host.h
@@ -255,6 +255,10 @@ struct mmc_supply {
 	struct regulator *vqmmc;	/* Optional Vccq supply */
 };
 
+struct mmc_ctx {
+	struct task_struct *task;
+};
+
 struct mmc_host {
 	struct device		*parent;
 	struct device		class_dev;
@@ -388,8 +392,9 @@ struct mmc_host {
 	struct mmc_card		*card;		/* device attached to this host */
 
 	wait_queue_head_t	wq;
-	struct task_struct	*claimer;	/* task that has host claimed */
+	struct mmc_ctx		*claimer;	/* context that has host claimed */
 	int			claim_cnt;	/* "claim" nesting count */
+	struct mmc_ctx		default_ctx;	/* default context */
 
 	struct delayed_work	detect;
 	int			detect_change;	/* card detect flag */
-- 
1.9.1

[PATCH V8 02/14] mmc: core: Add support for handling CQE requests

[Adrian Hunter]

Add core support for handling CQE requests, including starting, completing
and recovering.

Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
---
 drivers/mmc/core/core.c  | 163 +++++++++++++++++++++++++++++++++++++++++++++--
 drivers/mmc/core/core.h  |   4 ++
 include/linux/mmc/host.h |   2 +
 3 files changed, 164 insertions(+), 5 deletions(-)

diff --git a/drivers/mmc/core/core.c b/drivers/mmc/core/core.c
index 09fdc467b804..ef2d8aa1e7d2 100644
--- a/drivers/mmc/core/core.c
+++ b/drivers/mmc/core/core.c
@@ -266,7 +266,8 @@ static void __mmc_start_request(struct mmc_host *host, struct mmc_request *mrq)
 	host->ops->request(host, mrq);
 }
 
-static void mmc_mrq_pr_debug(struct mmc_host *host, struct mmc_request *mrq)
+static void mmc_mrq_pr_debug(struct mmc_host *host, struct mmc_request *mrq,
+			     bool cqe)
 {
 	if (mrq->sbc) {
 		pr_debug("<%s: starting CMD%u arg %08x flags %08x>\n",
@@ -275,9 +276,12 @@ static void mmc_mrq_pr_debug(struct mmc_host *host, struct mmc_request *mrq)
 	}
 
 	if (mrq->cmd) {
-		pr_debug("%s: starting CMD%u arg %08x flags %08x\n",
-			 mmc_hostname(host), mrq->cmd->opcode, mrq->cmd->arg,
-			 mrq->cmd->flags);
+		pr_debug("%s: starting %sCMD%u arg %08x flags %08x\n",
+			 mmc_hostname(host), cqe ? "CQE direct " : "",
+			 mrq->cmd->opcode, mrq->cmd->arg, mrq->cmd->flags);
+	} else if (cqe) {
+		pr_debug("%s: starting CQE transfer for tag %d blkaddr %u\n",
+			 mmc_hostname(host), mrq->tag, mrq->data->blk_addr);
 	}
 
 	if (mrq->data) {
@@ -342,7 +346,7 @@ static int mmc_start_request(struct mmc_host *host, struct mmc_request *mrq)
 	if (mmc_card_removed(host->card))
 		return -ENOMEDIUM;
 
-	mmc_mrq_pr_debug(host, mrq);
+	mmc_mrq_pr_debug(host, mrq, false);
 
 	WARN_ON(!host->claimed);
 
@@ -482,6 +486,155 @@ void mmc_wait_for_req_done(struct mmc_host *host, struct mmc_request *mrq)
 }
 EXPORT_SYMBOL(mmc_wait_for_req_done);
 
+/*
+ * mmc_cqe_start_req - Start a CQE request.
+ * @host: MMC host to start the request
+ * @mrq: request to start
+ *
+ * Start the request, re-tuning if needed and it is possible. Returns an error
+ * code if the request fails to start or -EBUSY if CQE is busy.
+ */
+int mmc_cqe_start_req(struct mmc_host *host, struct mmc_request *mrq)
+{
+	int err;
+
+	/*
+	 * CQE cannot process re-tuning commands. Caller must hold retuning
+	 * while CQE is in use.  Re-tuning can happen here only when CQE has no
+	 * active requests i.e. this is the first.  Note, re-tuning will call
+	 * ->cqe_off().
+	 */
+	err = mmc_retune(host);
+	if (err)
+		goto out_err;
+
+	mrq->host = host;
+
+	mmc_mrq_pr_debug(host, mrq, true);
+
+	err = mmc_mrq_prep(host, mrq);
+	if (err)
+		goto out_err;
+
+	err = host->cqe_ops->cqe_request(host, mrq);
+	if (err)
+		goto out_err;
+
+	trace_mmc_request_start(host, mrq);
+
+	return 0;
+
+out_err:
+	if (mrq->cmd) {
+		pr_debug("%s: failed to start CQE direct CMD%u, error %d\n",
+			 mmc_hostname(host), mrq->cmd->opcode, err);
+	} else {
+		pr_debug("%s: failed to start CQE transfer for tag %d, error %d\n",
+			 mmc_hostname(host), mrq->tag, err);
+	}
+	return err;
+}
+EXPORT_SYMBOL(mmc_cqe_start_req);
+
+/**
+ *	mmc_cqe_request_done - CQE has finished processing an MMC request
+ *	@host: MMC host which completed request
+ *	@mrq: MMC request which completed
+ *
+ *	CQE drivers should call this function when they have completed
+ *	their processing of a request.
+ */
+void mmc_cqe_request_done(struct mmc_host *host, struct mmc_request *mrq)
+{
+	mmc_should_fail_request(host, mrq);
+
+	/* Flag re-tuning needed on CRC errors */
+	if ((mrq->cmd && mrq->cmd->error == -EILSEQ) ||
+	    (mrq->data && mrq->data->error == -EILSEQ))
+		mmc_retune_needed(host);
+
+	trace_mmc_request_done(host, mrq);
+
+	if (mrq->cmd) {
+		pr_debug("%s: CQE req done (direct CMD%u): %d\n",
+			 mmc_hostname(host), mrq->cmd->opcode, mrq->cmd->error);
+	} else {
+		pr_debug("%s: CQE transfer done tag %d\n",
+			 mmc_hostname(host), mrq->tag);
+	}
+
+	if (mrq->data) {
+		pr_debug("%s:     %d bytes transferred: %d\n",
+			 mmc_hostname(host),
+			 mrq->data->bytes_xfered, mrq->data->error);
+	}
+
+	mrq->done(mrq);
+}
+EXPORT_SYMBOL(mmc_cqe_request_done);
+
+/**
+ *	mmc_cqe_post_req - CQE post process of a completed MMC request
+ *	@host: MMC host
+ *	@mrq: MMC request to be processed
+ */
+void mmc_cqe_post_req(struct mmc_host *host, struct mmc_request *mrq)
+{
+	if (host->cqe_ops->cqe_post_req)
+		host->cqe_ops->cqe_post_req(host, mrq);
+}
+EXPORT_SYMBOL(mmc_cqe_post_req);
+
+/* Arbitrary 1 second timeout */
+#define MMC_CQE_RECOVERY_TIMEOUT	1000
+
+/*
+ * mmc_cqe_recovery - Recover from CQE errors.
+ * @host: MMC host to recover
+ *
+ * Recovery consists of stopping CQE, stopping eMMC, discarding the queue in
+ * in eMMC, and discarding the queue in CQE. CQE must call
+ * mmc_cqe_request_done() on all requests. An error is returned if the eMMC
+ * fails to discard its queue.
+ */
+int mmc_cqe_recovery(struct mmc_host *host)
+{
+	struct mmc_command cmd;
+	int err;
+
+	mmc_retune_hold_now(host);
+
+	/*
+	 * Recovery is expected seldom, if at all, but it reduces performance,
+	 * so make sure it is not completely silent.
+	 */
+	pr_warn("%s: running CQE recovery\n", mmc_hostname(host));
+
+	host->cqe_ops->cqe_recovery_start(host);
+
+	memset(&cmd, 0, sizeof(cmd));
+	cmd.opcode       = MMC_STOP_TRANSMISSION,
+	cmd.flags        = MMC_RSP_R1B | MMC_CMD_AC,
+	cmd.flags       &= ~MMC_RSP_CRC; /* Ignore CRC */
+	cmd.busy_timeout = MMC_CQE_RECOVERY_TIMEOUT,
+	mmc_wait_for_cmd(host, &cmd, 0);
+
+	memset(&cmd, 0, sizeof(cmd));
+	cmd.opcode       = MMC_CMDQ_TASK_MGMT;
+	cmd.arg          = 1; /* Discard entire queue */
+	cmd.flags        = MMC_RSP_R1B | MMC_CMD_AC;
+	cmd.flags       &= ~MMC_RSP_CRC; /* Ignore CRC */
+	cmd.busy_timeout = MMC_CQE_RECOVERY_TIMEOUT,
+	err = mmc_wait_for_cmd(host, &cmd, 0);
+
+	host->cqe_ops->cqe_recovery_finish(host);
+
+	mmc_retune_release(host);
+
+	return err;
+}
+EXPORT_SYMBOL(mmc_cqe_recovery);
+
 /**
  *	mmc_is_req_done - Determine if a 'cap_cmd_during_tfr' request is done
  *	@host: MMC host
diff --git a/drivers/mmc/core/core.h b/drivers/mmc/core/core.h
index a294c836a143..e941342ed450 100644
--- a/drivers/mmc/core/core.h
+++ b/drivers/mmc/core/core.h
@@ -150,4 +150,8 @@ static inline void mmc_claim_host(struct mmc_host *host)
 void mmc_ctx_get_card(struct mmc_card *card, struct mmc_ctx *ctx);
 void mmc_ctx_put_card(struct mmc_card *card, struct mmc_ctx *ctx);
 
+int mmc_cqe_start_req(struct mmc_host *host, struct mmc_request *mrq);
+void mmc_cqe_post_req(struct mmc_host *host, struct mmc_request *mrq);
+int mmc_cqe_recovery(struct mmc_host *host);
+
 #endif
diff --git a/include/linux/mmc/host.h b/include/linux/mmc/host.h
index 228a493c3b25..54b0463443bd 100644
--- a/include/linux/mmc/host.h
+++ b/include/linux/mmc/host.h
@@ -474,6 +474,8 @@ static inline void *mmc_priv(struct mmc_host *host)
 void mmc_request_done(struct mmc_host *, struct mmc_request *);
 void mmc_command_done(struct mmc_host *host, struct mmc_request *mrq);
 
+void mmc_cqe_request_done(struct mmc_host *host, struct mmc_request *mrq);
+
 static inline void mmc_signal_sdio_irq(struct mmc_host *host)
 {
 	host->ops->enable_sdio_irq(host, 0);
-- 
1.9.1

[PATCH V8 03/14] mmc: mmc: Enable Command Queuing

[Adrian Hunter]

Enable the Command Queue if the host controller supports a command queue
engine. It is not compatible with Packed Commands, so make a note of that in the
comment.

Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
---
 drivers/mmc/core/mmc.c | 17 +++++++++++++++++
 1 file changed, 17 insertions(+)

diff --git a/drivers/mmc/core/mmc.c b/drivers/mmc/core/mmc.c
index a7eb623f8daa..194cf2082e73 100644
--- a/drivers/mmc/core/mmc.c
+++ b/drivers/mmc/core/mmc.c
@@ -1784,6 +1784,23 @@ static int mmc_init_card(struct mmc_host *host, u32 ocr,
 	}
 
 	/*
+	 * Enable Command Queue if supported. Note that Packed Commands cannot
+	 * be used with Command Queue.
+	 */
+	card->ext_csd.cmdq_en = false;
+	if (card->ext_csd.cmdq_support && host->caps2 & MMC_CAP2_CQE) {
+		err = mmc_cmdq_enable(card);
+		if (err && err != -EBADMSG)
+			goto free_card;
+		if (err) {
+			pr_warn("%s: Enabling CMDQ failed\n",
+				mmc_hostname(card->host));
+			card->ext_csd.cmdq_support = false;
+			card->ext_csd.cmdq_depth = 0;
+			err = 0;
+		}
+	}
+	/*
 	 * In some cases (e.g. RPMB or mmc_test), the Command Queue must be
 	 * disabled for a time, so a flag is needed to indicate to re-enable the
 	 * Command Queue.
-- 
1.9.1

[PATCH V8 04/14] mmc: mmc: Enable CQE's

[Adrian Hunter]

Enable or disable CQE when a card is added or removed respectively.

Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
---
 drivers/mmc/core/bus.c |  7 +++++++
 drivers/mmc/core/mmc.c | 12 ++++++++++++
 2 files changed, 19 insertions(+)

diff --git a/drivers/mmc/core/bus.c b/drivers/mmc/core/bus.c
index 301246513a37..a4b49e25fe96 100644
--- a/drivers/mmc/core/bus.c
+++ b/drivers/mmc/core/bus.c
@@ -369,10 +369,17 @@ int mmc_add_card(struct mmc_card *card)
  */
 void mmc_remove_card(struct mmc_card *card)
 {
+	struct mmc_host *host = card->host;
+
 #ifdef CONFIG_DEBUG_FS
 	mmc_remove_card_debugfs(card);
 #endif
 
+	if (host->cqe_enabled) {
+		host->cqe_ops->cqe_disable(host);
+		host->cqe_enabled = false;
+	}
+
 	if (mmc_card_present(card)) {
 		if (mmc_host_is_spi(card->host)) {
 			pr_info("%s: SPI card removed\n",
diff --git a/drivers/mmc/core/mmc.c b/drivers/mmc/core/mmc.c
index 194cf2082e73..0dcbd25126a1 100644
--- a/drivers/mmc/core/mmc.c
+++ b/drivers/mmc/core/mmc.c
@@ -1807,6 +1807,18 @@ 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) {
+		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 (!oldcard)
 		host->card = card;
 
-- 
1.9.1

[PATCH V8 05/14] mmc: block: Use local variables in mmc_blk_data_prep()

[Adrian Hunter]

Use local variables in mmc_blk_data_prep() in preparation for adding CQE
support which doesn't use the output variables.

Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
---
 drivers/mmc/core/block.c | 29 ++++++++++++++++++-----------
 1 file changed, 18 insertions(+), 11 deletions(-)

diff --git a/drivers/mmc/core/block.c b/drivers/mmc/core/block.c
index 29fc1e662891..1ec5f0bc98d2 100644
--- a/drivers/mmc/core/block.c
+++ b/drivers/mmc/core/block.c
@@ -1534,21 +1534,22 @@ static enum mmc_blk_status mmc_blk_err_check(struct mmc_card *card,
 }
 
 static void mmc_blk_data_prep(struct mmc_queue *mq, struct mmc_queue_req *mqrq,
-			      int disable_multi, bool *do_rel_wr,
-			      bool *do_data_tag)
+			      int disable_multi, bool *do_rel_wr_p,
+			      bool *do_data_tag_p)
 {
 	struct mmc_blk_data *md = mq->blkdata;
 	struct mmc_card *card = md->queue.card;
 	struct mmc_blk_request *brq = &mqrq->brq;
 	struct request *req = mmc_queue_req_to_req(mqrq);
+	bool do_rel_wr, do_data_tag;
 
 	/*
 	 * Reliable writes are used to implement Forced Unit Access and
 	 * are supported only on MMCs.
 	 */
-	*do_rel_wr = (req->cmd_flags & REQ_FUA) &&
-		     rq_data_dir(req) == WRITE &&
-		     (md->flags & MMC_BLK_REL_WR);
+	do_rel_wr = (req->cmd_flags & REQ_FUA) &&
+		    rq_data_dir(req) == WRITE &&
+		    (md->flags & MMC_BLK_REL_WR);
 
 	memset(brq, 0, sizeof(struct mmc_blk_request));
 
@@ -1596,18 +1597,18 @@ static void mmc_blk_data_prep(struct mmc_queue *mq, struct mmc_queue_req *mqrq,
 						brq->data.blocks);
 	}
 
-	if (*do_rel_wr)
+	if (do_rel_wr)
 		mmc_apply_rel_rw(brq, card, req);
 
 	/*
 	 * Data tag is used only during writing meta data to speed
 	 * up write and any subsequent read of this meta data
 	 */
-	*do_data_tag = card->ext_csd.data_tag_unit_size &&
-		       (req->cmd_flags & REQ_META) &&
-		       (rq_data_dir(req) == WRITE) &&
-		       ((brq->data.blocks * brq->data.blksz) >=
-			card->ext_csd.data_tag_unit_size);
+	do_data_tag = card->ext_csd.data_tag_unit_size &&
+		      (req->cmd_flags & REQ_META) &&
+		      (rq_data_dir(req) == WRITE) &&
+		      ((brq->data.blocks * brq->data.blksz) >=
+		       card->ext_csd.data_tag_unit_size);
 
 	mmc_set_data_timeout(&brq->data, card);
 
@@ -1636,6 +1637,12 @@ static void mmc_blk_data_prep(struct mmc_queue *mq, struct mmc_queue_req *mqrq,
 	mqrq->areq.mrq = &brq->mrq;
 
 	mmc_queue_bounce_pre(mqrq);
+
+	if (do_rel_wr_p)
+		*do_rel_wr_p = do_rel_wr;
+
+	if (do_data_tag_p)
+		*do_data_tag_p = do_data_tag;
 }
 
 static void mmc_blk_rw_rq_prep(struct mmc_queue_req *mqrq,
-- 
1.9.1

[PATCH V8 06/14] mmc: block: Prepare CQE data

[Adrian Hunter]

Enhance mmc_blk_data_prep() to support CQE requests. That means adding
some things that for non-CQE requests would be encoded into the command
arguments - such as the block address, reliable-write flag, and data tag
flag. Also the request tag is needed to provide the command queue task id,
and a comment is added to explain the future possibility of defining a
priority.

Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
---
 drivers/mmc/core/block.c | 16 +++++++++++++++-
 1 file changed, 15 insertions(+), 1 deletion(-)

diff --git a/drivers/mmc/core/block.c b/drivers/mmc/core/block.c
index 1ec5f0bc98d2..2424c7360687 100644
--- a/drivers/mmc/core/block.c
+++ b/drivers/mmc/core/block.c
@@ -1554,6 +1554,7 @@ static void mmc_blk_data_prep(struct mmc_queue *mq, struct mmc_queue_req *mqrq,
 	memset(brq, 0, sizeof(struct mmc_blk_request));
 
 	brq->mrq.data = &brq->data;
+	brq->mrq.tag = req->tag;
 
 	brq->stop.opcode = MMC_STOP_TRANSMISSION;
 	brq->stop.arg = 0;
@@ -1568,6 +1569,14 @@ static void mmc_blk_data_prep(struct mmc_queue *mq, struct mmc_queue_req *mqrq,
 
 	brq->data.blksz = 512;
 	brq->data.blocks = blk_rq_sectors(req);
+	brq->data.blk_addr = blk_rq_pos(req);
+
+	/*
+	 * The command queue supports 2 priorities: "high" (1) and "simple" (0).
+	 * The eMMC will give "high" priority tasks priority over "simple"
+	 * priority tasks. Here we always set "simple" priority by not setting
+	 * MMC_DATA_PRIO.
+	 */
 
 	/*
 	 * The block layer doesn't support all sector count
@@ -1597,8 +1606,10 @@ static void mmc_blk_data_prep(struct mmc_queue *mq, struct mmc_queue_req *mqrq,
 						brq->data.blocks);
 	}
 
-	if (do_rel_wr)
+	if (do_rel_wr) {
 		mmc_apply_rel_rw(brq, card, req);
+		brq->data.flags |= MMC_DATA_REL_WR;
+	}
 
 	/*
 	 * Data tag is used only during writing meta data to speed
@@ -1610,6 +1621,9 @@ static void mmc_blk_data_prep(struct mmc_queue *mq, struct mmc_queue_req *mqrq,
 		      ((brq->data.blocks * brq->data.blksz) >=
 		       card->ext_csd.data_tag_unit_size);
 
+	if (do_data_tag)
+		brq->data.flags |= MMC_DATA_DAT_TAG;
+
 	mmc_set_data_timeout(&brq->data, card);
 
 	brq->data.sg = mqrq->sg;
-- 
1.9.1

[PATCH V8 07/14] mmc: block: Factor out mmc_setup_queue()

[Adrian Hunter]

Factor out some common code that will also be used with blk-mq.

Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
---
 drivers/mmc/core/queue.c | 51 ++++++++++++++++++++++++++++--------------------
 1 file changed, 30 insertions(+), 21 deletions(-)

diff --git a/drivers/mmc/core/queue.c b/drivers/mmc/core/queue.c
index 74c663b1c0a7..da74fe4e638a 100644
--- a/drivers/mmc/core/queue.c
+++ b/drivers/mmc/core/queue.c
@@ -223,6 +223,35 @@ static void mmc_exit_request(struct request_queue *q, struct request *req)
 	mq_rq->sg = NULL;
 }
 
+static void mmc_setup_queue(struct mmc_queue *mq, struct mmc_card *card)
+{
+	struct mmc_host *host = card->host;
+	u64 limit = BLK_BOUNCE_HIGH;
+
+	if (mmc_dev(host)->dma_mask && *mmc_dev(host)->dma_mask)
+		limit = (u64)dma_max_pfn(mmc_dev(host)) << PAGE_SHIFT;
+
+	queue_flag_set_unlocked(QUEUE_FLAG_NONROT, mq->queue);
+	queue_flag_clear_unlocked(QUEUE_FLAG_ADD_RANDOM, mq->queue);
+	if (mmc_can_erase(card))
+		mmc_queue_setup_discard(mq->queue, card);
+
+	if (card->bouncesz) {
+		blk_queue_max_hw_sectors(mq->queue, card->bouncesz / 512);
+		blk_queue_max_segments(mq->queue, card->bouncesz / 512);
+		blk_queue_max_segment_size(mq->queue, card->bouncesz);
+	} else {
+		blk_queue_bounce_limit(mq->queue, limit);
+		blk_queue_max_hw_sectors(mq->queue,
+			min(host->max_blk_count, host->max_req_size / 512));
+		blk_queue_max_segments(mq->queue, host->max_segs);
+		blk_queue_max_segment_size(mq->queue, host->max_seg_size);
+	}
+
+	/* Initialize thread_sem even if it is not used */
+	sema_init(&mq->thread_sem, 1);
+}
+
 /**
  * mmc_init_queue - initialise a queue structure.
  * @mq: mmc queue
@@ -236,12 +265,8 @@ int mmc_init_queue(struct mmc_queue *mq, struct mmc_card *card,
 		   spinlock_t *lock, const char *subname)
 {
 	struct mmc_host *host = card->host;
-	u64 limit = BLK_BOUNCE_HIGH;
 	int ret = -ENOMEM;
 
-	if (mmc_dev(host)->dma_mask && *mmc_dev(host)->dma_mask)
-		limit = (u64)dma_max_pfn(mmc_dev(host)) << PAGE_SHIFT;
-
 	/*
 	 * mmc_init_request() depends on card->bouncesz so it must be calculated
 	 * before blk_init_allocated_queue() starts allocating requests.
@@ -266,24 +291,8 @@ int mmc_init_queue(struct mmc_queue *mq, struct mmc_card *card,
 	}
 
 	blk_queue_prep_rq(mq->queue, mmc_prep_request);
-	queue_flag_set_unlocked(QUEUE_FLAG_NONROT, mq->queue);
-	queue_flag_clear_unlocked(QUEUE_FLAG_ADD_RANDOM, mq->queue);
-	if (mmc_can_erase(card))
-		mmc_queue_setup_discard(mq->queue, card);
-
-	if (card->bouncesz) {
-		blk_queue_max_hw_sectors(mq->queue, card->bouncesz / 512);
-		blk_queue_max_segments(mq->queue, card->bouncesz / 512);
-		blk_queue_max_segment_size(mq->queue, card->bouncesz);
-	} else {
-		blk_queue_bounce_limit(mq->queue, limit);
-		blk_queue_max_hw_sectors(mq->queue,
-			min(host->max_blk_count, host->max_req_size / 512));
-		blk_queue_max_segments(mq->queue, host->max_segs);
-		blk_queue_max_segment_size(mq->queue, host->max_seg_size);
-	}
 
-	sema_init(&mq->thread_sem, 1);
+	mmc_setup_queue(mq, card);
 
 	mq->thread = kthread_run(mmc_queue_thread, mq, "mmcqd/%d%s",
 		host->index, subname ? subname : "");
-- 
1.9.1

[PATCH V8 08/14] mmc: core: Add parameter use_blk_mq

[Adrian Hunter]

Until mmc has blk-mq support fully implemented and tested, add a
parameter use_blk_mq, default to false unless config option MMC_MQ_DEFAULT
is selected.

Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
---
 drivers/mmc/Kconfig      | 11 +++++++++++
 drivers/mmc/core/core.c  |  7 +++++++
 drivers/mmc/core/core.h  |  2 ++
 drivers/mmc/core/host.c  |  2 ++
 drivers/mmc/core/host.h  |  4 ++++
 include/linux/mmc/host.h |  1 +
 6 files changed, 27 insertions(+)

diff --git a/drivers/mmc/Kconfig b/drivers/mmc/Kconfig
index ec21388311db..98202934bd29 100644
--- a/drivers/mmc/Kconfig
+++ b/drivers/mmc/Kconfig
@@ -12,6 +12,17 @@ menuconfig MMC
 	  If you want MMC/SD/SDIO support, you should say Y here and
 	  also to your specific host controller driver.
 
+config MMC_MQ_DEFAULT
+	bool "MMC: use blk-mq I/O path by default"
+	depends on MMC && BLOCK
+	---help---
+	  This option enables the new blk-mq based I/O path for MMC block
+	  devices by default.  With the option the mmc_core.use_blk_mq
+	  module/boot option defaults to Y, without it to N, but it can
+	  still be overridden either way.
+
+	  If unsure say N.
+
 if MMC
 
 source "drivers/mmc/core/Kconfig"
diff --git a/drivers/mmc/core/core.c b/drivers/mmc/core/core.c
index ef2d8aa1e7d2..3638ed4f0f9e 100644
--- a/drivers/mmc/core/core.c
+++ b/drivers/mmc/core/core.c
@@ -66,6 +66,13 @@
 bool use_spi_crc = 1;
 module_param(use_spi_crc, bool, 0);
 
+#ifdef CONFIG_MMC_MQ_DEFAULT
+bool mmc_use_blk_mq = true;
+#else
+bool mmc_use_blk_mq = false;
+#endif
+module_param_named(use_blk_mq, mmc_use_blk_mq, bool, S_IWUSR | S_IRUGO);
+
 static int mmc_schedule_delayed_work(struct delayed_work *work,
 				     unsigned long delay)
 {
diff --git a/drivers/mmc/core/core.h b/drivers/mmc/core/core.h
index e941342ed450..535539a9e7eb 100644
--- a/drivers/mmc/core/core.h
+++ b/drivers/mmc/core/core.h
@@ -35,6 +35,8 @@ struct mmc_bus_ops {
 	int (*reset)(struct mmc_host *);
 };
 
+extern bool mmc_use_blk_mq;
+
 void mmc_attach_bus(struct mmc_host *host, const struct mmc_bus_ops *ops);
 void mmc_detach_bus(struct mmc_host *host);
 
diff --git a/drivers/mmc/core/host.c b/drivers/mmc/core/host.c
index ad88deb2e8f3..b624dbb6cd15 100644
--- a/drivers/mmc/core/host.c
+++ b/drivers/mmc/core/host.c
@@ -398,6 +398,8 @@ struct mmc_host *mmc_alloc_host(int extra, struct device *dev)
 	host->max_blk_size = 512;
 	host->max_blk_count = PAGE_SIZE / 512;
 
+	host->use_blk_mq = mmc_use_blk_mq;
+
 	return host;
 }
 
diff --git a/drivers/mmc/core/host.h b/drivers/mmc/core/host.h
index 77d6f60d1bf9..170fe5947087 100644
--- a/drivers/mmc/core/host.h
+++ b/drivers/mmc/core/host.h
@@ -69,6 +69,10 @@ static inline bool mmc_card_hs400es(struct mmc_card *card)
 	return card->host->ios.enhanced_strobe;
 }
 
+static inline bool mmc_host_use_blk_mq(struct mmc_host *host)
+{
+	return host->use_blk_mq;
+}
 
 #endif
 
diff --git a/include/linux/mmc/host.h b/include/linux/mmc/host.h
index 54b0463443bd..5d1bd10991f7 100644
--- a/include/linux/mmc/host.h
+++ b/include/linux/mmc/host.h
@@ -378,6 +378,7 @@ struct mmc_host {
 	unsigned int		doing_retune:1;	/* re-tuning in progress */
 	unsigned int		retune_now:1;	/* do re-tuning at next req */
 	unsigned int		retune_paused:1; /* re-tuning is temporarily disabled */
+	unsigned int		use_blk_mq:1;	/* use blk-mq */
 
 	int			rescan_disable;	/* disable card detection */
 	int			rescan_entered;	/* used with nonremovable devices */
-- 
1.9.1

[PATCH V8 09/14] mmc: core: Remove unnecessary host claim

[Adrian Hunter]

Callers already have the host claimed, so remove the unnecessary
calls to mmc_claim_host() and mmc_release_host().

Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
---
 drivers/mmc/core/mmc_ops.c | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/drivers/mmc/core/mmc_ops.c b/drivers/mmc/core/mmc_ops.c
index 54686ca4bfb7..a6b0a232f24a 100644
--- a/drivers/mmc/core/mmc_ops.c
+++ b/drivers/mmc/core/mmc_ops.c
@@ -977,7 +977,6 @@ void mmc_start_bkops(struct mmc_card *card, bool from_exception)
 	    from_exception)
 		return;
 
-	mmc_claim_host(card->host);
 	if (card->ext_csd.raw_bkops_status >= EXT_CSD_BKOPS_LEVEL_2) {
 		timeout = MMC_OPS_TIMEOUT_MS;
 		use_busy_signal = true;
@@ -995,7 +994,7 @@ void mmc_start_bkops(struct mmc_card *card, bool from_exception)
 		pr_warn("%s: Error %d starting bkops\n",
 			mmc_hostname(card->host), err);
 		mmc_retune_release(card->host);
-		goto out;
+		return;
 	}
 
 	/*
@@ -1007,8 +1006,6 @@ void mmc_start_bkops(struct mmc_card *card, bool from_exception)
 		mmc_card_set_doing_bkops(card);
 	else
 		mmc_retune_release(card->host);
-out:
-	mmc_release_host(card->host);
 }
 
 /*
-- 
1.9.1

[PATCH V8 10/14] mmc: core: Export mmc_start_bkops()

[Adrian Hunter]

Export mmc_start_bkops() so that the block driver can use it.

Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
---
 drivers/mmc/core/mmc_ops.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/mmc/core/mmc_ops.c b/drivers/mmc/core/mmc_ops.c
index a6b0a232f24a..908e4db03535 100644
--- a/drivers/mmc/core/mmc_ops.c
+++ b/drivers/mmc/core/mmc_ops.c
@@ -1007,6 +1007,7 @@ void mmc_start_bkops(struct mmc_card *card, bool from_exception)
 	else
 		mmc_retune_release(card->host);
 }
+EXPORT_SYMBOL(mmc_start_bkops);
 
 /*
  * Flush the cache to the non-volatile storage.
-- 
1.9.1

[PATCH V8 11/14] mmc: core: Export mmc_start_request()

[Adrian Hunter]

Export mmc_start_request() so that the block driver can use it.

Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
---
 drivers/mmc/core/core.c | 3 ++-
 drivers/mmc/core/core.h | 2 ++
 2 files changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/mmc/core/core.c b/drivers/mmc/core/core.c
index 3638ed4f0f9e..24a73f387482 100644
--- a/drivers/mmc/core/core.c
+++ b/drivers/mmc/core/core.c
@@ -344,7 +344,7 @@ static int mmc_mrq_prep(struct mmc_host *host, struct mmc_request *mrq)
 	return 0;
 }
 
-static int mmc_start_request(struct mmc_host *host, struct mmc_request *mrq)
+int mmc_start_request(struct mmc_host *host, struct mmc_request *mrq)
 {
 	int err;
 
@@ -366,6 +366,7 @@ static int mmc_start_request(struct mmc_host *host, struct mmc_request *mrq)
 
 	return 0;
 }
+EXPORT_SYMBOL(mmc_start_request);
 
 /*
  * mmc_wait_data_done() - done callback for data request
diff --git a/drivers/mmc/core/core.h b/drivers/mmc/core/core.h
index 535539a9e7eb..03e0f8384b1c 100644
--- a/drivers/mmc/core/core.h
+++ b/drivers/mmc/core/core.h
@@ -109,6 +109,8 @@ static inline void mmc_unregister_pm_notifier(struct mmc_host *host) { }
 void mmc_wait_for_req_done(struct mmc_host *host, struct mmc_request *mrq);
 bool mmc_is_req_done(struct mmc_host *host, struct mmc_request *mrq);
 
+int mmc_start_request(struct mmc_host *host, struct mmc_request *mrq);
+
 struct mmc_async_req;
 
 struct mmc_async_req *mmc_start_areq(struct mmc_host *host,
-- 
1.9.1

[PATCH V8 12/14] mmc: block: Add CQE and blk-mq support

[Adrian Hunter]

Add CQE support to the block driver, including:
    - optionally using DCMD for flush requests
    - "manually" issuing discard requests
    - issuing read / write requests to the CQE
    - supporting block-layer timeouts
    - handling recovery
    - supporting re-tuning

CQE offers 25% - 50% better random multi-threaded I/O.  There is a slight
(e.g. 2%) drop in sequential read speed but no observable change to sequential
write.

CQE automatically sends the commands to complete requests.  However it only
supports reads / writes and so-called "direct commands" (DCMD).  Furthermore
DCMD is limited to one command at a time, but discards require 3 commands.
That makes issuing discards through CQE very awkward, but some CQE's don't
support DCMD anyway.  So for discards, the existing non-CQE approach is
taken, where the mmc core code issues the 3 commands one at a time i.e.
mmc_erase(). Where DCMD is used, is for issuing flushes.

For host controllers without CQE support, blk-mq support is extended to
synchronous reads/writes or, if the host supports CAP_WAIT_WHILE_BUSY,
asynchonous reads/writes.  The advantage of asynchronous reads/writes is
that it allows the preparation of the next request while the current
request is in progress.

Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
---
 drivers/mmc/core/block.c | 734 ++++++++++++++++++++++++++++++++++++++++++++++-
 drivers/mmc/core/block.h |   8 +
 drivers/mmc/core/queue.c | 428 +++++++++++++++++++++++++--
 drivers/mmc/core/queue.h |  54 +++-
 4 files changed, 1192 insertions(+), 32 deletions(-)

diff --git a/drivers/mmc/core/block.c b/drivers/mmc/core/block.c
index 2424c7360687..eca81502d09f 100644
--- a/drivers/mmc/core/block.c
+++ b/drivers/mmc/core/block.c
@@ -109,6 +109,7 @@ struct mmc_blk_data {
 #define MMC_BLK_WRITE		BIT(1)
 #define MMC_BLK_DISCARD		BIT(2)
 #define MMC_BLK_SECDISCARD	BIT(3)
+#define MMC_BLK_CQE_RECOVERY	BIT(4)
 
 	/*
 	 * Only set in main mmc_blk_data associated
@@ -1266,7 +1267,10 @@ static void mmc_blk_issue_discard_rq(struct mmc_queue *mq, struct request *req)
 	else
 		mmc_blk_reset_success(md, type);
 fail:
-	blk_end_request(req, status, blk_rq_bytes(req));
+	if (req->mq_ctx)
+		blk_mq_end_request(req, status);
+	else
+		blk_end_request(req, status, blk_rq_bytes(req));
 }
 
 static void mmc_blk_issue_secdiscard_rq(struct mmc_queue *mq,
@@ -1336,7 +1340,10 @@ static void mmc_blk_issue_secdiscard_rq(struct mmc_queue *mq,
 	if (!err)
 		mmc_blk_reset_success(md, type);
 out:
-	blk_end_request(req, status, blk_rq_bytes(req));
+	if (req->mq_ctx)
+		blk_mq_end_request(req, status);
+	else
+		blk_end_request(req, status, blk_rq_bytes(req));
 }
 
 static void mmc_blk_issue_flush(struct mmc_queue *mq, struct request *req)
@@ -1346,7 +1353,10 @@ static void mmc_blk_issue_flush(struct mmc_queue *mq, struct request *req)
 	int ret = 0;
 
 	ret = mmc_flush_cache(card);
-	blk_end_request_all(req, ret ? BLK_STS_IOERR : BLK_STS_OK);
+	if (req->mq_ctx)
+		blk_mq_end_request(req, ret ? BLK_STS_IOERR : BLK_STS_OK);
+	else
+		blk_end_request_all(req, ret ? BLK_STS_IOERR : BLK_STS_OK);
 }
 
 /*
@@ -1372,15 +1382,18 @@ static inline void mmc_apply_rel_rw(struct mmc_blk_request *brq,
 	}
 }
 
-#define CMD_ERRORS							\
-	(R1_OUT_OF_RANGE |	/* Command argument out of range */	\
-	 R1_ADDRESS_ERROR |	/* Misaligned address */		\
+#define CMD_ERRORS_EXCL_OOR						\
+	(R1_ADDRESS_ERROR |	/* Misaligned address */		\
 	 R1_BLOCK_LEN_ERROR |	/* Transferred block length incorrect */\
 	 R1_WP_VIOLATION |	/* Tried to write to protected block */	\
 	 R1_CARD_ECC_FAILED |	/* Card ECC failed */			\
 	 R1_CC_ERROR |		/* Card controller error */		\
 	 R1_ERROR)		/* General/unknown error */
 
+#define CMD_ERRORS							\
+	(CMD_ERRORS_EXCL_OOR |						\
+	 R1_OUT_OF_RANGE)	/* Command argument out of range */	\
+
 static void mmc_blk_eval_resp_error(struct mmc_blk_request *brq)
 {
 	u32 val;
@@ -1659,6 +1672,138 @@ static void mmc_blk_data_prep(struct mmc_queue *mq, struct mmc_queue_req *mqrq,
 		*do_data_tag_p = do_data_tag;
 }
 
+#define MMC_CQE_RETRIES 2
+
+static void mmc_blk_cqe_complete_rq(struct mmc_queue *mq, struct request *req)
+{
+	struct mmc_queue_req *mqrq = req_to_mmc_queue_req(req);
+	struct mmc_request *mrq = &mqrq->brq.mrq;
+	struct request_queue *q = req->q;
+	struct mmc_host *host = mq->card->host;
+	unsigned long flags;
+	bool put_card;
+	int err;
+
+	mmc_cqe_post_req(host, mrq);
+
+	if (mrq->cmd && mrq->cmd->error)
+		err = mrq->cmd->error;
+	else if (mrq->data && mrq->data->error)
+		err = mrq->data->error;
+	else
+		err = 0;
+
+	if (err) {
+		if (mqrq->retries++ < MMC_CQE_RETRIES)
+			blk_mq_requeue_request(req, true);
+		else
+			blk_mq_end_request(req, BLK_STS_IOERR);
+	} else if (mrq->data) {
+		if (blk_update_request(req, BLK_STS_OK, mrq->data->bytes_xfered))
+			blk_mq_requeue_request(req, true);
+		else
+			__blk_mq_end_request(req, BLK_STS_OK);
+	} else {
+		blk_mq_end_request(req, BLK_STS_OK);
+	}
+
+	spin_lock_irqsave(q->queue_lock, flags);
+
+	mq->in_flight[mmc_issue_type(mq, req)] -= 1;
+
+	put_card = mmc_tot_in_flight(mq) == 0;
+
+	mmc_cqe_check_busy(mq);
+
+	spin_unlock_irqrestore(q->queue_lock, flags);
+
+	if (!mq->cqe_busy)
+		blk_mq_run_hw_queues(q, true);
+
+	if (put_card)
+		mmc_ctx_put_card(mq->card, &mq->ctx);
+}
+
+void mmc_blk_cqe_recovery(struct mmc_queue *mq)
+{
+	struct mmc_card *card = mq->card;
+	struct mmc_host *host = card->host;
+	int err;
+
+	pr_debug("%s: CQE recovery start\n", mmc_hostname(host));
+
+	err = mmc_cqe_recovery(host);
+	if (err)
+		mmc_blk_reset(mq->blkdata, host, MMC_BLK_CQE_RECOVERY);
+	else
+		mmc_blk_reset_success(mq->blkdata, MMC_BLK_CQE_RECOVERY);
+
+	pr_debug("%s: CQE recovery done\n", mmc_hostname(host));
+}
+
+static void mmc_blk_cqe_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;
+
+	/*
+	 * 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);
+}
+
+static int mmc_blk_cqe_start_req(struct mmc_host *host, struct mmc_request *mrq)
+{
+	mrq->done		= mmc_blk_cqe_req_done;
+	mrq->recovery_notifier	= mmc_cqe_recovery_notifier;
+
+	return mmc_cqe_start_req(host, mrq);
+}
+
+static struct mmc_request *mmc_blk_cqe_prep_dcmd(struct mmc_queue_req *mqrq,
+						 struct request *req)
+{
+	struct mmc_blk_request *brq = &mqrq->brq;
+
+	memset(brq, 0, sizeof(*brq));
+
+	brq->mrq.cmd = &brq->cmd;
+	brq->mrq.tag = req->tag;
+
+	return &brq->mrq;
+}
+
+static int mmc_blk_cqe_issue_flush(struct mmc_queue *mq, struct request *req)
+{
+	struct mmc_queue_req *mqrq = req_to_mmc_queue_req(req);
+	struct mmc_request *mrq = mmc_blk_cqe_prep_dcmd(mqrq, req);
+
+	mrq->cmd->opcode = MMC_SWITCH;
+	mrq->cmd->arg = (MMC_SWITCH_MODE_WRITE_BYTE << 24) |
+			(EXT_CSD_FLUSH_CACHE << 16) |
+			(1 << 8) |
+			EXT_CSD_CMD_SET_NORMAL;
+	mrq->cmd->flags = MMC_CMD_AC | MMC_RSP_R1B;
+
+	return mmc_blk_cqe_start_req(mq->card->host, mrq);
+}
+
+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);
+
+	mmc_blk_data_prep(mq, mqrq, 0, NULL, NULL);
+
+	return mmc_blk_cqe_start_req(mq->card->host, &mqrq->brq.mrq);
+}
+
 static void mmc_blk_rw_rq_prep(struct mmc_queue_req *mqrq,
 			       struct mmc_card *card,
 			       int disable_multi,
@@ -1727,6 +1872,581 @@ static void mmc_blk_rw_rq_prep(struct mmc_queue_req *mqrq,
 	mqrq->areq.err_check = mmc_blk_err_check;
 }
 
+#define MMC_MAX_RETRIES		5
+#define MMC_DATA_RETRIES	2
+#define MMC_NO_RETRIES		(MMC_MAX_RETRIES + 1)
+
+/* Single sector read during recovery */
+static void mmc_blk_ss_read(struct mmc_queue *mq, struct request *req)
+{
+	struct mmc_queue_req *mqrq = req_to_mmc_queue_req(req);
+	blk_status_t status;
+
+	while (1) {
+		mmc_blk_rw_rq_prep(mqrq, mq->card, 1, mq);
+
+		mmc_wait_for_req(mq->card->host, &mqrq->brq.mrq);
+
+		/*
+		 * Not expecting command errors, so just give up in that case.
+		 * If there are retries remaining, the request will get
+		 * requeued.
+		 */
+		if (mqrq->brq.cmd.error)
+			return;
+
+		if (blk_rq_bytes(req) <= 512)
+			break;
+
+		status = mqrq->brq.data.error ? BLK_STS_IOERR : BLK_STS_OK;
+
+		blk_update_request(req, status, 512);
+	}
+
+	mqrq->retries = MMC_NO_RETRIES;
+}
+
+static inline bool mmc_blk_oor_valid(struct mmc_blk_request *brq)
+{
+	return !!brq->mrq.sbc;
+}
+
+static inline u32 mmc_blk_stop_err_bits(struct mmc_blk_request *brq)
+{
+	return mmc_blk_oor_valid(brq) ? CMD_ERRORS : CMD_ERRORS_EXCL_OOR;
+}
+
+static inline bool mmc_blk_in_tran_state(u32 status)
+{
+	/*
+	 * Some cards mishandle the status bits, so make sure to check both the
+	 * busy indication and the card state.
+	 */
+	return status & R1_READY_FOR_DATA &&
+	       (R1_CURRENT_STATE(status) == R1_STATE_TRAN);
+}
+
+/*
+ * Check for errors the host controller driver might not have seen such as
+ * response mode errors or invalid card state.
+ */
+static bool mmc_blk_status_error(struct request *req, u32 status)
+{
+	struct mmc_queue_req *mqrq = req_to_mmc_queue_req(req);
+	struct mmc_blk_request *brq = &mqrq->brq;
+	u32 stop_err_bits = mmc_blk_stop_err_bits(brq);
+
+	return brq->cmd.resp[0]  & CMD_ERRORS    ||
+	       brq->stop.resp[0] & stop_err_bits ||
+	       status            & stop_err_bits ||
+	       (rq_data_dir(req) == WRITE && !mmc_blk_in_tran_state(status));
+}
+
+static inline bool mmc_blk_cmd_started(struct mmc_blk_request *brq)
+{
+	return !brq->sbc.error && !brq->cmd.error &&
+	       !(brq->cmd.resp[0] & CMD_ERRORS);
+}
+
+static unsigned int mmc_blk_clock_khz(struct mmc_host *host)
+{
+	if (host->actual_clock)
+		return host->actual_clock / 1000;
+
+	/* Clock may be subject to a divisor, fudge it by a factor of 2. */
+	if (host->ios.clock)
+		return host->ios.clock / 2000;
+
+	/* How can there be no clock */
+	WARN_ON_ONCE(1);
+	return 100; /* 100 kHz is minimum possible value */
+}
+
+static unsigned long mmc_blk_data_timeout_jiffies(struct mmc_host *host,
+						  struct mmc_data *data)
+{
+	unsigned int ms = DIV_ROUND_UP(data->timeout_ns, 1000000);
+	unsigned int khz;
+
+	if (data->timeout_clks) {
+		khz = mmc_blk_clock_khz(host);
+		ms += DIV_ROUND_UP(data->timeout_clks, khz);
+	}
+
+	return msecs_to_jiffies(ms);
+}
+
+static int mmc_blk_card_stuck(struct mmc_card *card, struct request *req,
+			      u32 *resp_errs)
+{
+	struct mmc_queue_req *mqrq = req_to_mmc_queue_req(req);
+	struct mmc_data *data = &mqrq->brq.data;
+	unsigned long timeout;
+	u32 status;
+	int err;
+
+	timeout = jiffies + mmc_blk_data_timeout_jiffies(card->host, data);
+
+	while (1) {
+		bool done = time_after(jiffies, timeout);
+
+		err = __mmc_send_status(card, &status, 5);
+		if (err) {
+			pr_err("%s: error %d requesting status\n",
+			       req->rq_disk->disk_name, err);
+			break;
+		}
+
+		/* Accumulate any response error bits seen */
+		if (resp_errs)
+			*resp_errs |= status;
+
+		if (mmc_blk_in_tran_state(status))
+			break;
+
+		/* Timeout if the device never becomes ready */
+		if (done) {
+			pr_err("%s: Card stuck in wrong state! %s %s\n",
+				mmc_hostname(card->host),
+				req->rq_disk->disk_name, __func__);
+			err = -ETIMEDOUT;
+			break;
+		}
+	}
+
+	return err;
+}
+
+static int mmc_blk_send_stop(struct mmc_card *card)
+{
+	struct mmc_command cmd = {
+		.opcode = MMC_STOP_TRANSMISSION,
+		.flags = MMC_RSP_SPI_R1 | MMC_RSP_R1 | MMC_CMD_AC,
+	};
+
+	return mmc_wait_for_cmd(card->host, &cmd, 5);
+}
+
+static int mmc_blk_fix_state(struct mmc_card *card, struct request *req)
+{
+	int err;
+
+	mmc_retune_hold_now(card->host);
+
+	mmc_blk_send_stop(card);
+
+	err = mmc_blk_card_stuck(card, req, NULL);
+
+	mmc_retune_release(card->host);
+
+	return err;
+}
+
+static void mmc_blk_rw_recovery(struct mmc_queue *mq, struct request *req)
+{
+	int type = rq_data_dir(req) == READ ? MMC_BLK_READ : MMC_BLK_WRITE;
+	struct mmc_queue_req *mqrq = req_to_mmc_queue_req(req);
+	struct mmc_blk_request *brq = &mqrq->brq;
+	struct mmc_blk_data *md = mq->blkdata;
+	struct mmc_card *card = mq->card;
+	u32 status;
+	u32 blocks;
+	int err;
+
+	/*
+	 * Status error bits might get lost during re-tuning so don't allow
+	 * re-tuning yet.
+	 */
+	mmc_retune_hold_now(card->host);
+
+	/*
+	 * Some errors the host driver might not have seen. Set the number of
+	 * bytes transferred to zero in that case.
+	 */
+	err = __mmc_send_status(card, &status, 0);
+	if (err || mmc_blk_status_error(req, status))
+		brq->data.bytes_xfered = 0;
+
+	mmc_retune_release(card->host);
+
+	/*
+	 * Try again to get the status. This also provides an opportunity for
+	 * re-tuning.
+	 */
+	if (err)
+		err = __mmc_send_status(card, &status, 0);
+
+	/*
+	 * Nothing more to do after the number of bytes transferred has been
+	 * updated and there is no card.
+	 */
+	if (err && mmc_detect_card_removed(card->host))
+		return;
+
+	/* Try to get back to "tran" state */
+	if (err || !mmc_blk_in_tran_state(status))
+		err = mmc_blk_fix_state(mq->card, req);
+
+	/*
+	 * Special case for SD cards where the card might record the number of
+	 * blocks written.
+	 */
+	if (!err && mmc_blk_cmd_started(brq) && mmc_card_sd(card) &&
+	    rq_data_dir(req) == WRITE && !mmc_sd_num_wr_blocks(card, &blocks))
+		brq->data.bytes_xfered = blocks << 9;
+
+	/* Reset if the card is in a bad state */
+	if (err && mmc_blk_reset(md, card->host, type)) {
+		pr_err("%s: recovery failed!\n", req->rq_disk->disk_name);
+		mqrq->retries = MMC_NO_RETRIES;
+		return;
+	}
+
+	/*
+	 * If anything was done, just return and if there is anything remaining
+	 * on the request it will get requeued.
+	 */
+	if (brq->data.bytes_xfered)
+		return;
+
+	/* Reset before last retry */
+	if (mqrq->retries + 1 == MMC_MAX_RETRIES)
+		mmc_blk_reset(md, card->host, type);
+
+	/* Command errors fail fast, so use all MMC_MAX_RETRIES */
+	if (brq->sbc.error || brq->cmd.error)
+		return;
+
+	/* Reduce the remaining retries for data errors */
+	if (mqrq->retries < MMC_MAX_RETRIES - MMC_DATA_RETRIES) {
+		mqrq->retries = MMC_MAX_RETRIES - MMC_DATA_RETRIES;
+		return;
+	}
+
+	/* FIXME: Missing single sector read for large sector size */
+	if (rq_data_dir(req) == READ && !mmc_large_sector(card)) {
+		/* Read one sector at a time */
+		mmc_blk_ss_read(mq, req);
+		return;
+	}
+}
+
+static inline bool mmc_blk_rq_error(struct mmc_blk_request *brq)
+{
+	mmc_blk_eval_resp_error(brq);
+
+	return brq->sbc.error || brq->cmd.error || brq->stop.error ||
+	       brq->data.error || brq->cmd.resp[0] & CMD_ERRORS;
+}
+
+static int mmc_blk_card_busy(struct mmc_card *card, struct request *req)
+{
+	struct mmc_queue_req *mqrq = req_to_mmc_queue_req(req);
+	u32 status = 0;
+	int err;
+
+	if (mmc_host_is_spi(card->host) || rq_data_dir(req) == READ)
+		return 0;
+
+	mmc_retune_hold_now(card->host);
+
+	err = mmc_blk_card_stuck(card, req, &status);
+
+	mmc_retune_release(card->host);
+
+	/*
+	 * Do not assume data transferred correctly if there are any error bits
+	 * set.
+	 */
+	if (!err && status & mmc_blk_stop_err_bits(&mqrq->brq)) {
+		mqrq->brq.data.bytes_xfered = 0;
+		err = -EIO;
+	}
+
+	/* Copy the exception bit so it will be seen later on */
+	if (mmc_card_mmc(card) && status & R1_EXCEPTION_EVENT)
+		mqrq->brq.cmd.resp[0] |= R1_EXCEPTION_EVENT;
+
+	return err;
+}
+
+static inline void mmc_blk_rw_reset_success(struct mmc_queue *mq,
+					    struct request *req)
+{
+	int type = rq_data_dir(req) == READ ? MMC_BLK_READ : MMC_BLK_WRITE;
+
+	mmc_blk_reset_success(mq->blkdata, type);
+}
+
+static void mmc_blk_mq_complete_rq(struct mmc_queue *mq, struct request *req)
+{
+	struct mmc_queue_req *mqrq = req_to_mmc_queue_req(req);
+	unsigned int nr_bytes = mqrq->brq.data.bytes_xfered;
+
+	if (nr_bytes) {
+		if (blk_update_request(req, BLK_STS_OK, nr_bytes))
+			blk_mq_requeue_request(req, true);
+		else
+			__blk_mq_end_request(req, BLK_STS_OK);
+	} else if (mqrq->retries++ < MMC_MAX_RETRIES) {
+		blk_mq_requeue_request(req, true);
+	} else {
+		if (mmc_card_removed(mq->card))
+			req->rq_flags |= RQF_QUIET;
+		blk_mq_end_request(req, BLK_STS_IOERR);
+	}
+}
+
+static bool mmc_blk_urgent_bkops_needed(struct mmc_queue *mq,
+					struct mmc_queue_req *mqrq)
+{
+	return mmc_card_mmc(mq->card) &&
+	       (mqrq->brq.cmd.resp[0] & R1_EXCEPTION_EVENT ||
+		mqrq->brq.stop.resp[0] & R1_EXCEPTION_EVENT);
+}
+
+static void mmc_blk_urgent_bkops(struct mmc_queue *mq,
+				 struct mmc_queue_req *mqrq)
+{
+	if (mmc_blk_urgent_bkops_needed(mq, mqrq))
+		mmc_start_bkops(mq->card, true);
+}
+
+static int mmc_blk_card_busy_check(struct mmc_queue *mq, struct request *req)
+{
+	if (mmc_queue_rw_async(mq->card->host))
+		return 0;
+
+	return mmc_blk_card_busy(mq->card, req);
+}
+
+void mmc_blk_mq_complete(struct request *req)
+{
+	struct mmc_queue *mq = req->q->queuedata;
+
+	if (mq->use_cqe)
+		mmc_blk_cqe_complete_rq(mq, req);
+	else
+		mmc_blk_mq_complete_rq(mq, req);
+}
+
+static void mmc_blk_finish_rw_rq_blocking(struct mmc_queue *mq,
+					  struct request *req)
+{
+	struct mmc_queue_req *mqrq = req_to_mmc_queue_req(req);
+
+	if (mmc_blk_rq_error(&mqrq->brq) || mmc_blk_card_busy_check(mq, req))
+		mmc_blk_rw_recovery(mq, req);
+	else
+		mmc_blk_rw_reset_success(mq, req);
+
+	mmc_blk_urgent_bkops(mq, mqrq);
+
+	mq->rw_wait = false;
+
+	/*
+	 * Block layer timeouts race with completions which means the normal
+	 * completion path cannot be used during recovery.
+	 */
+	if (mq->in_recovery)
+		mmc_blk_mq_complete_rq(mq, req);
+	else
+		blk_mq_complete_request(req);
+}
+
+void mmc_blk_mq_recovery(struct mmc_queue *mq)
+{
+	struct request *req = mq->recovery_req;
+
+	mq->recovery_req = NULL;
+
+	mmc_blk_finish_rw_rq_blocking(mq, req);
+}
+
+static void mmc_blk_issue_rw_rq_blocking(struct mmc_queue *mq,
+					 struct request *req)
+{
+	struct mmc_queue_req *mqrq = req_to_mmc_queue_req(req);
+
+	mmc_blk_rw_rq_prep(mqrq, mq->card, 0, mq);
+
+	mmc_wait_for_req(mq->card->host, &mqrq->brq.mrq);
+
+	mmc_blk_finish_rw_rq_blocking(mq, req);
+}
+
+static void mmc_blk_mq_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;
+	bool put_card;
+
+	if (mmc_blk_rq_error(&mqrq->brq) ||
+	    mmc_blk_urgent_bkops_needed(mq, mqrq)) {
+		mq->recovery_needed = true;
+		mq->recovery_req = req;
+		wake_up(&mq->wait);
+		schedule_work(&mq->recovery_work);
+		return;
+	}
+
+	mmc_blk_rw_reset_success(mq, req);
+
+	if (host->ops->post_req)
+		host->ops->post_req(host, mrq, 0);
+
+	mq->rw_wait = false;
+	wake_up(&mq->wait);
+
+	blk_mq_complete_request(req);
+
+	spin_lock_irqsave(q->queue_lock, flags);
+
+	mq->in_flight[mmc_issue_type(mq, req)] -= 1;
+
+	put_card = mmc_tot_in_flight(mq) == 0;
+
+	spin_unlock_irqrestore(q->queue_lock, flags);
+
+	if (put_card)
+		mmc_ctx_put_card(mq->card, &mq->ctx);
+}
+
+static bool mmc_blk_rw_wait_cond(struct mmc_queue *mq, int *err)
+{
+	if (mq->recovery_needed) {
+		*err = -EBUSY;
+		return true;
+	}
+
+	return !mq->rw_wait;
+}
+
+static int mmc_blk_rw_wait(struct mmc_queue *mq)
+{
+	int err = 0;
+
+	wait_event(mq->wait, mmc_blk_rw_wait_cond(mq, &err));
+
+	return err;
+}
+
+static int mmc_blk_mq_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 = 0;
+
+	mmc_blk_rw_rq_prep(mqrq, mq->card, 0, mq);
+
+	mqrq->brq.mrq.done = mmc_blk_mq_req_done;
+
+	if (host->ops->pre_req)
+		host->ops->pre_req(host, &mqrq->brq.mrq);
+
+	err = mmc_blk_rw_wait(mq);
+	if (err)
+		goto out_post_req;
+
+	mq->rw_wait = true;
+
+	err = mmc_start_request(host, &mqrq->brq.mrq);
+
+	if (err)
+		mq->rw_wait = false;
+
+	/* Release re-tuning here where there is no synchronization required */
+	mmc_retune_release(host);
+
+out_post_req:
+	if (err && host->ops->post_req)
+		host->ops->post_req(host, &mqrq->brq.mrq, err);
+
+	return err;
+}
+
+static int mmc_blk_wait_for_idle(struct mmc_queue *mq, struct mmc_host *host)
+{
+	if (mq->use_cqe)
+		return host->cqe_ops->cqe_wait_for_idle(host);
+
+	if (mmc_queue_rw_async(host))
+		return mmc_blk_rw_wait(mq);
+
+	return 0;
+}
+
+enum mmc_issued mmc_blk_mq_issue_rq(struct mmc_queue *mq, struct request *req)
+{
+	struct mmc_blk_data *md = mq->blkdata;
+	struct mmc_card *card = md->queue.card;
+	struct mmc_host *host = card->host;
+	int ret;
+
+	ret = mmc_blk_part_switch(card, md->part_type);
+	if (ret)
+		return MMC_REQ_FAILED_TO_START;
+
+	switch (mmc_issue_type(mq, req)) {
+	case MMC_ISSUE_SYNC:
+		ret = mmc_blk_wait_for_idle(mq, host);
+		if (ret)
+			return MMC_REQ_BUSY;
+		switch (req_op(req)) {
+		case REQ_OP_DRV_IN:
+		case REQ_OP_DRV_OUT:
+			mmc_blk_issue_drv_op(mq, req);
+			break;
+		case REQ_OP_DISCARD:
+			mmc_blk_issue_discard_rq(mq, req);
+			break;
+		case REQ_OP_SECURE_ERASE:
+			mmc_blk_issue_secdiscard_rq(mq, req);
+			break;
+		case REQ_OP_FLUSH:
+			mmc_blk_issue_flush(mq, req);
+			break;
+		case REQ_OP_READ:
+		case REQ_OP_WRITE:
+			mmc_blk_issue_rw_rq_blocking(mq, req);
+			break;
+		default:
+			WARN_ON_ONCE(1);
+			return MMC_REQ_FAILED_TO_START;
+		}
+		return MMC_REQ_FINISHED;
+	case MMC_ISSUE_DCMD:
+	case MMC_ISSUE_ASYNC:
+		switch (req_op(req)) {
+		case REQ_OP_FLUSH:
+			ret = mmc_blk_cqe_issue_flush(mq, req);
+			break;
+		case REQ_OP_READ:
+		case REQ_OP_WRITE:
+			if (mq->use_cqe)
+				ret = mmc_blk_cqe_issue_rw_rq(mq, req);
+			else
+				ret = mmc_blk_mq_issue_rw_rq(mq, req);
+			break;
+		default:
+			WARN_ON_ONCE(1);
+			ret = -EINVAL;
+		}
+		if (!ret)
+			return MMC_REQ_STARTED;
+		return ret == -EBUSY ? MMC_REQ_BUSY : MMC_REQ_FAILED_TO_START;
+	default:
+		WARN_ON_ONCE(1);
+		return MMC_REQ_FAILED_TO_START;
+	}
+}
+
 static bool mmc_blk_rw_cmd_err(struct mmc_blk_data *md, struct mmc_card *card,
 			       struct mmc_blk_request *brq, struct request *req,
 			       bool old_req_pending)
@@ -2094,7 +2814,7 @@ static struct mmc_blk_data *mmc_blk_alloc_req(struct mmc_card *card,
 	INIT_LIST_HEAD(&md->part);
 	md->usage = 1;
 
-	ret = mmc_init_queue(&md->queue, card, &md->lock, subname);
+	ret = mmc_init_queue(&md->queue, card, &md->lock, subname, area_type);
 	if (ret)
 		goto err_putdisk;
 
diff --git a/drivers/mmc/core/block.h b/drivers/mmc/core/block.h
index 860ca7c8df86..742aa4d27cbf 100644
--- a/drivers/mmc/core/block.h
+++ b/drivers/mmc/core/block.h
@@ -6,4 +6,12 @@
 
 void mmc_blk_issue_rq(struct mmc_queue *mq, struct request *req);
 
+enum mmc_issued;
+
+void mmc_blk_cqe_recovery(struct mmc_queue *mq);
+
+enum mmc_issued mmc_blk_mq_issue_rq(struct mmc_queue *mq, struct request *req);
+void mmc_blk_mq_complete(struct request *req);
+void mmc_blk_mq_recovery(struct mmc_queue *mq);
+
 #endif
diff --git a/drivers/mmc/core/queue.c b/drivers/mmc/core/queue.c
index da74fe4e638a..c3144f0c2285 100644
--- a/drivers/mmc/core/queue.c
+++ b/drivers/mmc/core/queue.c
@@ -22,6 +22,7 @@
 #include "block.h"
 #include "core.h"
 #include "card.h"
+#include "host.h"
 
 #define MMC_QUEUE_BOUNCESZ	65536
 
@@ -36,10 +37,153 @@ static int mmc_prep_request(struct request_queue *q, struct request *req)
 		return BLKPREP_KILL;
 
 	req->rq_flags |= RQF_DONTPREP;
+	req_to_mmc_queue_req(req)->retries = 0;
 
 	return BLKPREP_OK;
 }
 
+static inline bool mmc_cqe_dcmd_busy(struct mmc_queue *mq)
+{
+	/* Allow only 1 DCMD at a time */
+	return mq->in_flight[MMC_ISSUE_DCMD];
+}
+
+void mmc_cqe_check_busy(struct mmc_queue *mq)
+{
+	if ((mq->cqe_busy & MMC_CQE_DCMD_BUSY) && !mmc_cqe_dcmd_busy(mq))
+		mq->cqe_busy &= ~MMC_CQE_DCMD_BUSY;
+
+	mq->cqe_busy &= ~MMC_CQE_QUEUE_FULL;
+}
+
+static inline bool mmc_cqe_can_dcmd(struct mmc_host *host)
+{
+	return host->caps2 & MMC_CAP2_CQE_DCMD;
+}
+
+enum mmc_issue_type mmc_cqe_issue_type(struct mmc_host *host,
+				       struct request *req)
+{
+	switch (req_op(req)) {
+	case REQ_OP_DRV_IN:
+	case REQ_OP_DRV_OUT:
+	case REQ_OP_DISCARD:
+	case REQ_OP_SECURE_ERASE:
+		return MMC_ISSUE_SYNC;
+	case REQ_OP_FLUSH:
+		return mmc_cqe_can_dcmd(host) ? MMC_ISSUE_DCMD : MMC_ISSUE_SYNC;
+	default:
+		return MMC_ISSUE_ASYNC;
+	}
+}
+
+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)
+		return mmc_cqe_issue_type(host, req);
+
+	if (mmc_queue_rw_async(host) &&
+	    (req_op(req) == REQ_OP_READ || req_op(req) == REQ_OP_WRITE))
+		return MMC_ISSUE_ASYNC;
+
+	return MMC_ISSUE_SYNC;
+}
+
+static void __mmc_cqe_recovery_notifier(struct mmc_queue *mq)
+{
+	if (!mq->recovery_needed) {
+		mq->recovery_needed = true;
+		schedule_work(&mq->recovery_work);
+	}
+}
+
+void mmc_cqe_recovery_notifier(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;
+	unsigned long flags;
+
+	spin_lock_irqsave(q->queue_lock, flags);
+	__mmc_cqe_recovery_notifier(mq);
+	spin_unlock_irqrestore(q->queue_lock, flags);
+}
+
+static enum blk_eh_timer_return mmc_cqe_timed_out(struct request *req)
+{
+	struct mmc_queue_req *mqrq = req_to_mmc_queue_req(req);
+	struct mmc_request *mrq = &mqrq->brq.mrq;
+	struct mmc_queue *mq = req->q->queuedata;
+	struct mmc_host *host = mq->card->host;
+	enum mmc_issue_type issue_type = mmc_issue_type(mq, req);
+	bool recovery_needed = false;
+
+	switch (issue_type) {
+	case MMC_ISSUE_ASYNC:
+	case MMC_ISSUE_DCMD:
+		if (host->cqe_ops->cqe_timeout(host, mrq, &recovery_needed)) {
+			if (recovery_needed)
+				__mmc_cqe_recovery_notifier(mq);
+			return BLK_EH_RESET_TIMER;
+		}
+		/* No timeout */
+		return BLK_EH_HANDLED;
+	default:
+		/* Timeout is handled by mmc core */
+		return BLK_EH_RESET_TIMER;
+	}
+}
+
+static enum blk_eh_timer_return mmc_mq_timed_out(struct request *req,
+						 bool reserved)
+{
+	struct request_queue *q = req->q;
+	struct mmc_queue *mq = q->queuedata;
+	unsigned long flags;
+	int ret;
+
+	spin_lock_irqsave(q->queue_lock, flags);
+
+	if (mq->recovery_needed || !mq->use_cqe)
+		ret = BLK_EH_RESET_TIMER;
+	else
+		ret = mmc_cqe_timed_out(req);
+
+	spin_unlock_irqrestore(q->queue_lock, flags);
+
+	return ret;
+}
+
+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;
+
+	mmc_ctx_get_card(mq->card, &mq->ctx);
+
+	mq->in_recovery = true;
+
+	if (mq->use_cqe)
+		mmc_blk_cqe_recovery(mq);
+	else
+		mmc_blk_mq_recovery(mq);
+
+	mq->in_recovery = false;
+
+	spin_lock_irq(q->queue_lock);
+	mq->recovery_needed = false;
+	spin_unlock_irq(q->queue_lock);
+
+	mmc_ctx_put_card(mq->card, &mq->ctx);
+
+	blk_mq_run_hw_queues(q, true);
+}
+
 static int mmc_queue_thread(void *d)
 {
 	struct mmc_queue *mq = d;
@@ -176,11 +320,10 @@ static unsigned int mmc_queue_calc_bouncesz(struct mmc_host *host)
  * @req: the request
  * @gfp: memory allocation policy
  */
-static int mmc_init_request(struct request_queue *q, struct request *req,
-			    gfp_t gfp)
+static int __mmc_init_request(struct mmc_queue *mq, struct request *req,
+			      gfp_t gfp)
 {
 	struct mmc_queue_req *mq_rq = req_to_mmc_queue_req(req);
-	struct mmc_queue *mq = q->queuedata;
 	struct mmc_card *card = mq->card;
 	struct mmc_host *host = card->host;
 
@@ -208,6 +351,12 @@ static int mmc_init_request(struct request_queue *q, struct request *req,
 	return 0;
 }
 
+static int mmc_init_request(struct request_queue *q, struct request *req,
+			    gfp_t gfp)
+{
+	return __mmc_init_request(q->queuedata, req, gfp);
+}
+
 static void mmc_exit_request(struct request_queue *q, struct request *req)
 {
 	struct mmc_queue_req *mq_rq = req_to_mmc_queue_req(req);
@@ -223,6 +372,125 @@ static void mmc_exit_request(struct request_queue *q, struct request *req)
 	mq_rq->sg = NULL;
 }
 
+static int mmc_mq_init_request(struct blk_mq_tag_set *set, struct request *req,
+			       unsigned int hctx_idx, unsigned int numa_node)
+{
+	return __mmc_init_request(set->driver_data, req, GFP_KERNEL);
+}
+
+static void mmc_mq_exit_request(struct blk_mq_tag_set *set, struct request *req,
+				unsigned int hctx_idx)
+{
+	struct mmc_queue *mq = set->driver_data;
+
+	mmc_exit_request(mq->queue, req);
+}
+
+static blk_status_t mmc_mq_queue_rq(struct blk_mq_hw_ctx *hctx,
+				    const struct blk_mq_queue_data *bd)
+{
+	struct request *req = bd->rq;
+	struct request_queue *q = req->q;
+	struct mmc_queue *mq = q->queuedata;
+	struct mmc_card *card = mq->card;
+	struct mmc_host *host = card->host;
+	enum mmc_issue_type issue_type;
+	enum mmc_issued issued;
+	bool get_card, cqe_retune_ok;
+	int ret;
+
+	if (mmc_card_removed(mq->card) ||
+	    (mmc_access_rpmb(mq) && !blk_rq_is_private(req))) {
+		req->rq_flags |= RQF_QUIET;
+		return BLK_STS_IOERR;
+	}
+
+	issue_type = mmc_issue_type(mq, req);
+
+	spin_lock_irq(q->queue_lock);
+
+	if (mq->recovery_needed) {
+		spin_unlock_irq(q->queue_lock);
+		return BLK_STS_RESOURCE;
+	}
+
+	switch (issue_type) {
+	case MMC_ISSUE_DCMD:
+		if (mmc_cqe_dcmd_busy(mq)) {
+			mq->cqe_busy |= MMC_CQE_DCMD_BUSY;
+			spin_unlock_irq(q->queue_lock);
+			return BLK_STS_RESOURCE;
+		}
+		break;
+	case MMC_ISSUE_ASYNC:
+		break;
+	default:
+		/*
+		 * Timeouts are handled by mmc core, so set a large value to
+		 * avoid races.
+		 */
+		req->timeout = 600 * HZ;
+		break;
+	}
+
+	mq->in_flight[issue_type] += 1;
+	get_card = mmc_tot_in_flight(mq) == 1;
+	cqe_retune_ok = mmc_cqe_qcnt(mq) == 1;
+
+	spin_unlock_irq(q->queue_lock);
+
+	if (!(req->rq_flags & RQF_DONTPREP)) {
+		req_to_mmc_queue_req(req)->retries = 0;
+		req->rq_flags |= RQF_DONTPREP;
+	}
+
+	if (get_card)
+		mmc_ctx_get_card(card, &mq->ctx);
+
+	if (mq->use_cqe) {
+		host->retune_now = host->need_retune && cqe_retune_ok &&
+				   !host->hold_retune;
+	}
+
+	blk_mq_start_request(req);
+
+	issued = mmc_blk_mq_issue_rq(mq, req);
+
+	switch (issued) {
+	case MMC_REQ_BUSY:
+		ret = BLK_STS_RESOURCE;
+		break;
+	case MMC_REQ_FAILED_TO_START:
+		ret = BLK_STS_IOERR;
+		break;
+	default:
+		ret = BLK_STS_OK;
+		break;
+	}
+
+	if (issued != MMC_REQ_STARTED) {
+		bool put_card = false;
+
+		spin_lock_irq(q->queue_lock);
+		mq->in_flight[issue_type] -= 1;
+		if (mmc_tot_in_flight(mq) == 0)
+			put_card = true;
+		spin_unlock_irq(q->queue_lock);
+		if (put_card)
+			mmc_ctx_put_card(card, &mq->ctx);
+	}
+
+	return ret;
+}
+
+static const struct blk_mq_ops mmc_mq_cqe_ops = {
+	.queue_rq	= mmc_mq_queue_rq,
+	.init_request	= mmc_mq_init_request,
+	.exit_request	= mmc_mq_exit_request,
+	.complete	= mmc_blk_mq_complete,
+	.timeout	= mmc_mq_timed_out,
+};
+
 static void mmc_setup_queue(struct mmc_queue *mq, struct mmc_card *card)
 {
 	struct mmc_host *host = card->host;
@@ -250,6 +518,72 @@ static void mmc_setup_queue(struct mmc_queue *mq, struct mmc_card *card)
 
 	/* Initialize thread_sem even if it is not used */
 	sema_init(&mq->thread_sem, 1);
+
+	/* Initialize recovery_work even if it is not used */
+	INIT_WORK(&mq->recovery_work, mmc_mq_recovery_handler);
+
+	init_waitqueue_head(&mq->wait);
+}
+
+static int mmc_mq_init_queue(struct mmc_queue *mq, int q_depth,
+			     const struct blk_mq_ops *mq_ops, spinlock_t *lock)
+{
+	int ret;
+
+	memset(&mq->tag_set, 0, sizeof(mq->tag_set));
+	mq->tag_set.ops = mq_ops;
+	mq->tag_set.queue_depth = q_depth;
+	mq->tag_set.numa_node = NUMA_NO_NODE;
+	mq->tag_set.flags = BLK_MQ_F_SHOULD_MERGE | BLK_MQ_F_SG_MERGE |
+			    BLK_MQ_F_BLOCKING;
+	mq->tag_set.nr_hw_queues = 1;
+	mq->tag_set.cmd_size = sizeof(struct mmc_queue_req);
+	mq->tag_set.driver_data = mq;
+
+	ret = blk_mq_alloc_tag_set(&mq->tag_set);
+	if (ret)
+		return ret;
+
+	mq->queue = blk_mq_init_queue(&mq->tag_set);
+	if (IS_ERR(mq->queue)) {
+		ret = PTR_ERR(mq->queue);
+		goto free_tag_set;
+	}
+
+	mq->queue->queue_lock = lock;
+	mq->queue->queuedata = mq;
+
+	return 0;
+
+free_tag_set:
+	blk_mq_free_tag_set(&mq->tag_set);
+
+	return ret;
+}
+
+#define MMC_QUEUE_DEPTH 64
+
+static int mmc_mq_init(struct mmc_queue *mq, struct mmc_card *card,
+			 spinlock_t *lock)
+{
+	struct mmc_host *host = card->host;
+	int q_depth;
+	int ret;
+
+	if (mq->use_cqe)
+		q_depth = min_t(int, card->ext_csd.cmdq_depth, host->cqe_qdepth);
+	else
+		q_depth = MMC_QUEUE_DEPTH;
+
+	ret = mmc_mq_init_queue(mq, q_depth, &mmc_mq_cqe_ops, lock);
+	if (ret)
+		return ret;
+
+	blk_queue_rq_timeout(mq->queue, 60 * HZ);
+
+	mmc_setup_queue(mq, card);
+
+	return 0;
 }
 
 /**
@@ -262,7 +596,7 @@ static void mmc_setup_queue(struct mmc_queue *mq, struct mmc_card *card)
  * Initialise a MMC card request queue.
  */
 int mmc_init_queue(struct mmc_queue *mq, struct mmc_card *card,
-		   spinlock_t *lock, const char *subname)
+		   spinlock_t *lock, const char *subname, int area_type)
 {
 	struct mmc_host *host = card->host;
 	int ret = -ENOMEM;
@@ -274,6 +608,12 @@ int mmc_init_queue(struct mmc_queue *mq, struct mmc_card *card,
 	card->bouncesz = mmc_queue_calc_bouncesz(host);
 
 	mq->card = card;
+
+	mq->use_cqe = host->cqe_enabled && area_type != MMC_BLK_DATA_AREA_RPMB;
+
+	if (mq->use_cqe || mmc_host_use_blk_mq(host))
+		return mmc_mq_init(mq, card, lock);
+
 	mq->queue = blk_alloc_queue(GFP_KERNEL);
 	if (!mq->queue)
 		return -ENOMEM;
@@ -309,11 +649,63 @@ int mmc_init_queue(struct mmc_queue *mq, struct mmc_card *card,
 	return ret;
 }
 
+static void mmc_mq_queue_suspend(struct mmc_queue *mq)
+{
+	blk_mq_quiesce_queue(mq->queue);
+
+	/*
+	 * The host remains claimed while there are outstanding requests, so
+	 * simply claiming and releasing here ensures there are none.
+	 */
+	mmc_claim_host(mq->card->host);
+	mmc_release_host(mq->card->host);
+}
+
+static void mmc_mq_queue_resume(struct mmc_queue *mq)
+{
+	blk_mq_unquiesce_queue(mq->queue);
+}
+
+static void __mmc_queue_suspend(struct mmc_queue *mq)
+{
+	struct request_queue *q = mq->queue;
+	unsigned long flags;
+
+	if (!mq->suspended) {
+		mq->suspended |= true;
+
+		spin_lock_irqsave(q->queue_lock, flags);
+		blk_stop_queue(q);
+		spin_unlock_irqrestore(q->queue_lock, flags);
+
+		down(&mq->thread_sem);
+	}
+}
+
+static void __mmc_queue_resume(struct mmc_queue *mq)
+{
+	struct request_queue *q = mq->queue;
+	unsigned long flags;
+
+	if (mq->suspended) {
+		mq->suspended = false;
+
+		up(&mq->thread_sem);
+
+		spin_lock_irqsave(q->queue_lock, flags);
+		blk_start_queue(q);
+		spin_unlock_irqrestore(q->queue_lock, flags);
+	}
+}
+
 void mmc_cleanup_queue(struct mmc_queue *mq)
 {
 	struct request_queue *q = mq->queue;
 	unsigned long flags;
 
+	if (q->mq_ops)
+		return;
+
 	/* Make sure the queue isn't suspended, as that will deadlock */
 	mmc_queue_resume(mq);
 
@@ -341,17 +733,11 @@ void mmc_cleanup_queue(struct mmc_queue *mq)
 void mmc_queue_suspend(struct mmc_queue *mq)
 {
 	struct request_queue *q = mq->queue;
-	unsigned long flags;
-
-	if (!mq->suspended) {
-		mq->suspended |= true;
-
-		spin_lock_irqsave(q->queue_lock, flags);
-		blk_stop_queue(q);
-		spin_unlock_irqrestore(q->queue_lock, flags);
 
-		down(&mq->thread_sem);
-	}
+	if (q->mq_ops)
+		mmc_mq_queue_suspend(mq);
+	else
+		__mmc_queue_suspend(mq);
 }
 
 /**
@@ -361,17 +747,11 @@ void mmc_queue_suspend(struct mmc_queue *mq)
 void mmc_queue_resume(struct mmc_queue *mq)
 {
 	struct request_queue *q = mq->queue;
-	unsigned long flags;
-
-	if (mq->suspended) {
-		mq->suspended = false;
 
-		up(&mq->thread_sem);
-
-		spin_lock_irqsave(q->queue_lock, flags);
-		blk_start_queue(q);
-		spin_unlock_irqrestore(q->queue_lock, flags);
-	}
+	if (q->mq_ops)
+		mmc_mq_queue_resume(mq);
+	else
+		__mmc_queue_resume(mq);
 }
 
 /*
diff --git a/drivers/mmc/core/queue.h b/drivers/mmc/core/queue.h
index 04fc89360a7a..7a9cdfcfb7b3 100644
--- a/drivers/mmc/core/queue.h
+++ b/drivers/mmc/core/queue.h
@@ -7,6 +7,20 @@
 #include <linux/mmc/core.h>
 #include <linux/mmc/host.h>
 
+enum mmc_issued {
+	MMC_REQ_STARTED,
+	MMC_REQ_BUSY,
+	MMC_REQ_FAILED_TO_START,
+	MMC_REQ_FINISHED,
+};
+
+enum mmc_issue_type {
+	MMC_ISSUE_SYNC,
+	MMC_ISSUE_DCMD,
+	MMC_ISSUE_ASYNC,
+	MMC_ISSUE_MAX,
+};
+
 static inline struct mmc_queue_req *req_to_mmc_queue_req(struct request *rq)
 {
 	return blk_mq_rq_to_pdu(rq);
@@ -57,12 +71,15 @@ struct mmc_queue_req {
 	int			drv_op_result;
 	void			*drv_op_data;
 	unsigned int		ioc_count;
+	int			retries;
 };
 
 struct mmc_queue {
 	struct mmc_card		*card;
 	struct task_struct	*thread;
 	struct semaphore	thread_sem;
+	struct mmc_ctx		ctx;
+	struct blk_mq_tag_set	tag_set;
 	bool			suspended;
 	bool			asleep;
 	struct mmc_blk_data	*blkdata;
@@ -74,10 +91,22 @@ struct mmc_queue {
 	 * associated mmc_queue_req data.
 	 */
 	int			qcnt;
+
+	int			in_flight[MMC_ISSUE_MAX];
+	unsigned int		cqe_busy;
+#define MMC_CQE_DCMD_BUSY	BIT(0)
+#define MMC_CQE_QUEUE_FULL	BIT(1)
+	bool			use_cqe;
+	bool			recovery_needed;
+	bool			in_recovery;
+	bool			rw_wait;
+	struct work_struct	recovery_work;
+	wait_queue_head_t	wait;
+	struct request		*recovery_req;
 };
 
 extern int mmc_init_queue(struct mmc_queue *, struct mmc_card *, spinlock_t *,
-			  const char *);
+			  const char *, int);
 extern void mmc_cleanup_queue(struct mmc_queue *);
 extern void mmc_queue_suspend(struct mmc_queue *);
 extern void mmc_queue_resume(struct mmc_queue *);
@@ -89,4 +118,27 @@ extern unsigned int mmc_queue_map_sg(struct mmc_queue *,
 
 extern int mmc_access_rpmb(struct mmc_queue *);
 
+void mmc_cqe_check_busy(struct mmc_queue *mq);
+void mmc_cqe_recovery_notifier(struct mmc_request *mrq);
+
+enum mmc_issue_type mmc_issue_type(struct mmc_queue *mq, struct request *req);
+
+static inline int mmc_tot_in_flight(struct mmc_queue *mq)
+{
+	return mq->in_flight[MMC_ISSUE_SYNC] +
+	       mq->in_flight[MMC_ISSUE_DCMD] +
+	       mq->in_flight[MMC_ISSUE_ASYNC];
+}
+
+static inline int mmc_cqe_qcnt(struct mmc_queue *mq)
+{
+	return mq->in_flight[MMC_ISSUE_DCMD] +
+	       mq->in_flight[MMC_ISSUE_ASYNC];
+}
+
+static inline bool mmc_queue_rw_async(struct mmc_host *host)
+{
+	return host->caps & MMC_CAP_WAIT_WHILE_BUSY;
+}
+
 #endif
-- 
1.9.1

[PATCH V8 13/14] mmc: cqhci: support for command queue enabled host

[Adrian Hunter]

From: Venkat Gopalakrishnan <venkatg@codeaurora.org>

This patch adds CMDQ support for command-queue compatible
hosts.

Command queue is added in eMMC-5.1 specification. This
enables the controller to process upto 32 requests at
a time.

Adrian Hunter contributed renaming to cqhci, recovery, suspend
and resume, cqhci_off, cqhci_wait_for_idle, and external timeout
handling.

Signed-off-by: Asutosh Das <asutoshd@codeaurora.org>
Signed-off-by: Sujit Reddy Thumma <sthumma@codeaurora.org>
Signed-off-by: Konstantin Dorfman <kdorfman@codeaurora.org>
Signed-off-by: Venkat Gopalakrishnan <venkatg@codeaurora.org>
Signed-off-by: Subhash Jadavani <subhashj@codeaurora.org>
Signed-off-by: Ritesh Harjani <riteshh@codeaurora.org>
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
---
 drivers/mmc/host/Kconfig  |   13 +
 drivers/mmc/host/Makefile |    1 +
 drivers/mmc/host/cqhci.c  | 1154 +++++++++++++++++++++++++++++++++++++++++++++
 drivers/mmc/host/cqhci.h  |  240 ++++++++++
 4 files changed, 1408 insertions(+)
 create mode 100644 drivers/mmc/host/cqhci.c
 create mode 100644 drivers/mmc/host/cqhci.h

diff --git a/drivers/mmc/host/Kconfig b/drivers/mmc/host/Kconfig
index 8c15637178ff..6db8e46b4e12 100644
--- a/drivers/mmc/host/Kconfig
+++ b/drivers/mmc/host/Kconfig
@@ -843,6 +843,19 @@ config MMC_SUNXI
 	  This selects support for the SD/MMC Host Controller on
 	  Allwinner sunxi SoCs.
 
+config MMC_CQHCI
+	tristate "Command Queue Host Controller Interface support"
+	depends on HAS_DMA
+	help
+	  This selects the Command Queue Host Controller Interface (CQHCI)
+	  support present in host controllers of Qualcomm Technologies, Inc
+	  amongst others.
+	  This controller supports eMMC devices with command queue support.
+
+	  If you have a controller with this interface, say Y or M here.
+
+	  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 303f5cd46cd9..4cec2a985a04 100644
--- a/drivers/mmc/host/Makefile
+++ b/drivers/mmc/host/Makefile
@@ -89,6 +89,7 @@ obj-$(CONFIG_MMC_SDHCI_MSM)		+= sdhci-msm.o
 obj-$(CONFIG_MMC_SDHCI_ST)		+= sdhci-st.o
 obj-$(CONFIG_MMC_SDHCI_MICROCHIP_PIC32)	+= sdhci-pic32.o
 obj-$(CONFIG_MMC_SDHCI_BRCMSTB)		+= sdhci-brcmstb.o
+obj-$(CONFIG_MMC_CQHCI)			+= cqhci.o
 
 ifeq ($(CONFIG_CB710_DEBUG),y)
 	CFLAGS-cb710-mmc	+= -DDEBUG
diff --git a/drivers/mmc/host/cqhci.c b/drivers/mmc/host/cqhci.c
new file mode 100644
index 000000000000..eb3c1695b0c7
--- /dev/null
+++ b/drivers/mmc/host/cqhci.c
@@ -0,0 +1,1154 @@
+/* Copyright (c) 2015, The Linux Foundation. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#include <linux/delay.h>
+#include <linux/highmem.h>
+#include <linux/io.h>
+#include <linux/module.h>
+#include <linux/dma-mapping.h>
+#include <linux/slab.h>
+#include <linux/scatterlist.h>
+#include <linux/platform_device.h>
+#include <linux/ktime.h>
+
+#include <linux/mmc/mmc.h>
+#include <linux/mmc/host.h>
+#include <linux/mmc/card.h>
+
+#include "cqhci.h"
+
+#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);
+}
+
+static inline u8 *get_link_desc(struct cqhci_host *cq_host, u8 tag)
+{
+	u8 *desc = get_desc(cq_host, tag);
+
+	return desc + cq_host->task_desc_len;
+}
+
+static inline dma_addr_t get_trans_desc_dma(struct cqhci_host *cq_host, u8 tag)
+{
+	return cq_host->trans_desc_dma_base +
+		(cq_host->mmc->max_segs * tag *
+		 cq_host->trans_desc_len);
+}
+
+static inline u8 *get_trans_desc(struct cqhci_host *cq_host, u8 tag)
+{
+	return cq_host->trans_desc_base +
+		(cq_host->trans_desc_len * cq_host->mmc->max_segs * tag);
+}
+
+static void setup_trans_desc(struct cqhci_host *cq_host, u8 tag)
+{
+	u8 *link_temp;
+	dma_addr_t trans_temp;
+
+	link_temp = get_link_desc(cq_host, tag);
+	trans_temp = get_trans_desc_dma(cq_host, tag);
+
+	memset(link_temp, 0, cq_host->link_desc_len);
+	if (cq_host->link_desc_len > 8)
+		*(link_temp + 8) = 0;
+
+	if (tag == DCMD_SLOT && (cq_host->mmc->caps2 & MMC_CAP2_CQE_DCMD)) {
+		*link_temp = CQHCI_VALID(0) | CQHCI_ACT(0) | CQHCI_END(1);
+		return;
+	}
+
+	*link_temp = CQHCI_VALID(1) | CQHCI_ACT(0x6) | CQHCI_END(0);
+
+	if (cq_host->dma64) {
+		__le64 *data_addr = (__le64 __force *)(link_temp + 4);
+
+		data_addr[0] = cpu_to_le64(trans_temp);
+	} else {
+		__le32 *data_addr = (__le32 __force *)(link_temp + 4);
+
+		data_addr[0] = cpu_to_le32(trans_temp);
+	}
+}
+
+static void cqhci_set_irqs(struct cqhci_host *cq_host, u32 set)
+{
+	u32 ier;
+
+	ier = cqhci_readl(cq_host, CQHCI_ISTE);
+	ier |= set;
+	cqhci_writel(cq_host, ier, CQHCI_ISTE);
+	cqhci_writel(cq_host, ier, CQHCI_ISGE);
+}
+
+#define DRV_NAME "cqhci"
+
+#define CQHCI_DUMP(f, x...) \
+	pr_err("%s: " DRV_NAME ": " f, mmc_hostname(mmc), ## x)
+
+static void cqhci_dumpregs(struct cqhci_host *cq_host)
+{
+	struct mmc_host *mmc = cq_host->mmc;
+
+	CQHCI_DUMP("============ CQHCI REGISTER DUMP ===========\n");
+
+	CQHCI_DUMP("Caps:      0x%08x | Version:  0x%08x\n",
+		   cqhci_readl(cq_host, CQHCI_CAP),
+		   cqhci_readl(cq_host, CQHCI_VER));
+	CQHCI_DUMP("Config:    0x%08x | Control:  0x%08x\n",
+		   cqhci_readl(cq_host, CQHCI_CFG),
+		   cqhci_readl(cq_host, CQHCI_CTL));
+	CQHCI_DUMP("Int stat:  0x%08x | Int enab: 0x%08x\n",
+		   cqhci_readl(cq_host, CQHCI_IS),
+		   cqhci_readl(cq_host, CQHCI_ISTE));
+	CQHCI_DUMP("Int sig:   0x%08x | Int Coal: 0x%08x\n",
+		   cqhci_readl(cq_host, CQHCI_ISGE),
+		   cqhci_readl(cq_host, CQHCI_IC));
+	CQHCI_DUMP("TDL base:  0x%08x | TDL up32: 0x%08x\n",
+		   cqhci_readl(cq_host, CQHCI_TDLBA),
+		   cqhci_readl(cq_host, CQHCI_TDLBAU));
+	CQHCI_DUMP("Doorbell:  0x%08x | TCN:      0x%08x\n",
+		   cqhci_readl(cq_host, CQHCI_TDBR),
+		   cqhci_readl(cq_host, CQHCI_TCN));
+	CQHCI_DUMP("Dev queue: 0x%08x | Dev Pend: 0x%08x\n",
+		   cqhci_readl(cq_host, CQHCI_DQS),
+		   cqhci_readl(cq_host, CQHCI_DPT));
+	CQHCI_DUMP("Task clr:  0x%08x | SSC1:     0x%08x\n",
+		   cqhci_readl(cq_host, CQHCI_TCLR),
+		   cqhci_readl(cq_host, CQHCI_SSC1));
+	CQHCI_DUMP("SSC2:      0x%08x | DCMD rsp: 0x%08x\n",
+		   cqhci_readl(cq_host, CQHCI_SSC2),
+		   cqhci_readl(cq_host, CQHCI_CRDCT));
+	CQHCI_DUMP("RED mask:  0x%08x | TERRI:    0x%08x\n",
+		   cqhci_readl(cq_host, CQHCI_RMEM),
+		   cqhci_readl(cq_host, CQHCI_TERRI));
+	CQHCI_DUMP("Resp idx:  0x%08x | Resp arg: 0x%08x\n",
+		   cqhci_readl(cq_host, CQHCI_CRI),
+		   cqhci_readl(cq_host, CQHCI_CRA));
+
+	if (cq_host->ops->dumpregs)
+		cq_host->ops->dumpregs(mmc);
+	else
+		CQHCI_DUMP(": ===========================================\n");
+}
+
+/**
+ * The allocated descriptor table for task, link & transfer descritors
+ * looks like:
+ * |----------|
+ * |task desc |  |->|----------|
+ * |----------|  |  |trans desc|
+ * |link desc-|->|  |----------|
+ * |----------|          .
+ *      .                .
+ *  no. of slots      max-segs
+ *      .           |----------|
+ * |----------|
+ * The idea here is to create the [task+trans] table and mark & point the
+ * link desc to the transfer desc table on a per slot basis.
+ */
+static int cqhci_host_alloc_tdl(struct cqhci_host *cq_host)
+{
+	int i = 0;
+
+	/* task descriptor can be 64/128 bit irrespective of arch */
+	if (cq_host->caps & CQHCI_TASK_DESC_SZ_128) {
+		cqhci_writel(cq_host, cqhci_readl(cq_host, CQHCI_CFG) |
+			       CQHCI_TASK_DESC_SZ, CQHCI_CFG);
+		cq_host->task_desc_len = 16;
+	} else {
+		cq_host->task_desc_len = 8;
+	}
+
+	/*
+	 * 96 bits length of transfer desc instead of 128 bits which means
+	 * ADMA would expect next valid descriptor at the 96th bit
+	 * or 128th bit
+	 */
+	if (cq_host->dma64) {
+		if (cq_host->quirks & CQHCI_QUIRK_SHORT_TXFR_DESC_SZ)
+			cq_host->trans_desc_len = 12;
+		else
+			cq_host->trans_desc_len = 16;
+		cq_host->link_desc_len = 16;
+	} else {
+		cq_host->trans_desc_len = 8;
+		cq_host->link_desc_len = 8;
+	}
+
+	/* total size of a slot: 1 task & 1 transfer (link) */
+	cq_host->slot_sz = cq_host->task_desc_len + cq_host->link_desc_len;
+
+	cq_host->desc_size = cq_host->slot_sz * cq_host->num_slots;
+
+	cq_host->data_size = cq_host->trans_desc_len * cq_host->mmc->max_segs *
+		(cq_host->num_slots - 1);
+
+	pr_debug("%s: cqhci: desc_size: %zu data_sz: %zu slot-sz: %d\n",
+		 mmc_hostname(cq_host->mmc), cq_host->desc_size, cq_host->data_size,
+		 cq_host->slot_sz);
+
+	/*
+	 * allocate a dma-mapped chunk of memory for the descriptors
+	 * allocate a dma-mapped chunk of memory for link descriptors
+	 * setup each link-desc memory offset per slot-number to
+	 * the descriptor table.
+	 */
+	cq_host->desc_base = dmam_alloc_coherent(mmc_dev(cq_host->mmc),
+						 cq_host->desc_size,
+						 &cq_host->desc_dma_base,
+						 GFP_KERNEL);
+	cq_host->trans_desc_base = dmam_alloc_coherent(mmc_dev(cq_host->mmc),
+					      cq_host->data_size,
+					      &cq_host->trans_desc_dma_base,
+					      GFP_KERNEL);
+	if (!cq_host->desc_base || !cq_host->trans_desc_base)
+		return -ENOMEM;
+
+	pr_debug("%s: cqhci: desc-base: 0x%p trans-base: 0x%p\n desc_dma 0x%llx trans_dma: 0x%llx\n",
+		 mmc_hostname(cq_host->mmc), cq_host->desc_base, cq_host->trans_desc_base,
+		(unsigned long long)cq_host->desc_dma_base,
+		(unsigned long long)cq_host->trans_desc_dma_base);
+
+	for (; i < (cq_host->num_slots); i++)
+		setup_trans_desc(cq_host, i);
+
+	return 0;
+}
+
+static void __cqhci_enable(struct cqhci_host *cq_host)
+{
+	struct mmc_host *mmc = cq_host->mmc;
+	u32 cqcfg;
+
+	cqcfg = cqhci_readl(cq_host, CQHCI_CFG);
+
+	/* Configuration must not be changed while enabled */
+	if (cqcfg & CQHCI_ENABLE) {
+		cqcfg &= ~CQHCI_ENABLE;
+		cqhci_writel(cq_host, cqcfg, CQHCI_CFG);
+	}
+
+	cqcfg &= ~(CQHCI_DCMD | CQHCI_TASK_DESC_SZ);
+
+	if (mmc->caps2 & MMC_CAP2_CQE_DCMD)
+		cqcfg |= CQHCI_DCMD;
+
+	if (cq_host->caps & CQHCI_TASK_DESC_SZ_128)
+		cqcfg |= CQHCI_TASK_DESC_SZ;
+
+	cqhci_writel(cq_host, cqcfg, CQHCI_CFG);
+
+	cqhci_writel(cq_host, lower_32_bits(cq_host->desc_dma_base),
+		     CQHCI_TDLBA);
+	cqhci_writel(cq_host, upper_32_bits(cq_host->desc_dma_base),
+		     CQHCI_TDLBAU);
+
+	cqhci_writel(cq_host, cq_host->rca, CQHCI_SSC2);
+
+	cqhci_set_irqs(cq_host, 0);
+
+	cqcfg |= CQHCI_ENABLE;
+
+	cqhci_writel(cq_host, cqcfg, CQHCI_CFG);
+
+	mmc->cqe_on = true;
+
+	if (cq_host->ops->enable)
+		cq_host->ops->enable(mmc);
+
+	/* Ensure all writes are done before interrupts are enabled */
+	wmb();
+
+	cqhci_set_irqs(cq_host, CQHCI_IS_MASK);
+
+	cq_host->activated = true;
+}
+
+static void __cqhci_disable(struct cqhci_host *cq_host)
+{
+	u32 cqcfg;
+
+	cqcfg = cqhci_readl(cq_host, CQHCI_CFG);
+	cqcfg &= ~CQHCI_ENABLE;
+	cqhci_writel(cq_host, cqcfg, CQHCI_CFG);
+
+	cq_host->mmc->cqe_on = false;
+
+	cq_host->activated = false;
+}
+
+int cqhci_suspend(struct mmc_host *mmc)
+{
+	struct cqhci_host *cq_host = mmc->cqe_private;
+
+	if (cq_host->enabled)
+		__cqhci_disable(cq_host);
+
+	return 0;
+}
+EXPORT_SYMBOL(cqhci_suspend);
+
+int cqhci_resume(struct mmc_host *mmc)
+{
+	/* Re-enable is done upon first request */
+	return 0;
+}
+EXPORT_SYMBOL(cqhci_resume);
+
+static int cqhci_enable(struct mmc_host *mmc, struct mmc_card *card)
+{
+	struct cqhci_host *cq_host = mmc->cqe_private;
+	int err;
+
+	if (cq_host->enabled)
+		return 0;
+
+	cq_host->rca = card->rca;
+
+	err = cqhci_host_alloc_tdl(cq_host);
+	if (err)
+		return err;
+
+	__cqhci_enable(cq_host);
+
+	cq_host->enabled = true;
+
+#ifdef DEBUG
+	cqhci_dumpregs(cq_host);
+#endif
+	return 0;
+}
+
+/* CQHCI is idle and should halt immediately, so set a small timeout */
+#define CQHCI_OFF_TIMEOUT 100
+
+static void cqhci_off(struct mmc_host *mmc)
+{
+	struct cqhci_host *cq_host = mmc->cqe_private;
+	ktime_t timeout;
+	bool timed_out;
+	u32 reg;
+
+	if (!cq_host->enabled || !mmc->cqe_on || cq_host->recovery_halt)
+		return;
+
+	if (cq_host->ops->disable)
+		cq_host->ops->disable(mmc, false);
+
+	cqhci_writel(cq_host, CQHCI_HALT, CQHCI_CTL);
+
+	timeout = ktime_add_us(ktime_get(), CQHCI_OFF_TIMEOUT);
+	while (1) {
+		timed_out = ktime_compare(ktime_get(), timeout) > 0;
+		reg = cqhci_readl(cq_host, CQHCI_CTL);
+		if ((reg & CQHCI_HALT) || timed_out)
+			break;
+	}
+
+	if (timed_out)
+		pr_err("%s: cqhci: CQE stuck on\n", mmc_hostname(mmc));
+	else
+		pr_debug("%s: cqhci: CQE off\n", mmc_hostname(mmc));
+
+	mmc->cqe_on = false;
+}
+
+static void cqhci_disable(struct mmc_host *mmc)
+{
+	struct cqhci_host *cq_host = mmc->cqe_private;
+
+	if (!cq_host->enabled)
+		return;
+
+	cqhci_off(mmc);
+
+	__cqhci_disable(cq_host);
+
+	dmam_free_coherent(mmc_dev(mmc), cq_host->data_size,
+			   cq_host->trans_desc_base,
+			   cq_host->trans_desc_dma_base);
+
+	dmam_free_coherent(mmc_dev(mmc), cq_host->desc_size,
+			   cq_host->desc_base,
+			   cq_host->desc_dma_base);
+
+	cq_host->trans_desc_base = NULL;
+	cq_host->desc_base = NULL;
+
+	cq_host->enabled = false;
+}
+
+static void cqhci_prep_task_desc(struct mmc_request *mrq,
+					u64 *data, bool intr)
+{
+	u32 req_flags = mrq->data->flags;
+
+	*data = CQHCI_VALID(1) |
+		CQHCI_END(1) |
+		CQHCI_INT(intr) |
+		CQHCI_ACT(0x5) |
+		CQHCI_FORCED_PROG(!!(req_flags & MMC_DATA_FORCED_PRG)) |
+		CQHCI_DATA_TAG(!!(req_flags & MMC_DATA_DAT_TAG)) |
+		CQHCI_DATA_DIR(!!(req_flags & MMC_DATA_READ)) |
+		CQHCI_PRIORITY(!!(req_flags & MMC_DATA_PRIO)) |
+		CQHCI_QBAR(!!(req_flags & MMC_DATA_QBR)) |
+		CQHCI_REL_WRITE(!!(req_flags & MMC_DATA_REL_WR)) |
+		CQHCI_BLK_COUNT(mrq->data->blocks) |
+		CQHCI_BLK_ADDR((u64)mrq->data->blk_addr);
+
+	pr_debug("%s: cqhci: tag %d task descriptor 0x016%llx\n",
+		 mmc_hostname(mrq->host), mrq->tag, (unsigned long long)*data);
+}
+
+static int cqhci_dma_map(struct mmc_host *host, struct mmc_request *mrq)
+{
+	int sg_count;
+	struct mmc_data *data = mrq->data;
+
+	if (!data)
+		return -EINVAL;
+
+	sg_count = dma_map_sg(mmc_dev(host), data->sg,
+			      data->sg_len,
+			      (data->flags & MMC_DATA_WRITE) ?
+			      DMA_TO_DEVICE : DMA_FROM_DEVICE);
+	if (!sg_count) {
+		pr_err("%s: sg-len: %d\n", __func__, data->sg_len);
+		return -ENOMEM;
+	}
+
+	return sg_count;
+}
+
+static void cqhci_set_tran_desc(u8 *desc, dma_addr_t addr, int len, bool end,
+				bool dma64)
+{
+	__le32 *attr = (__le32 __force *)desc;
+
+	*attr = (CQHCI_VALID(1) |
+		 CQHCI_END(end ? 1 : 0) |
+		 CQHCI_INT(0) |
+		 CQHCI_ACT(0x4) |
+		 CQHCI_DAT_LENGTH(len));
+
+	if (dma64) {
+		__le64 *dataddr = (__le64 __force *)(desc + 4);
+
+		dataddr[0] = cpu_to_le64(addr);
+	} else {
+		__le32 *dataddr = (__le32 __force *)(desc + 4);
+
+		dataddr[0] = cpu_to_le32(addr);
+	}
+}
+
+static int cqhci_prep_tran_desc(struct mmc_request *mrq,
+			       struct cqhci_host *cq_host, int tag)
+{
+	struct mmc_data *data = mrq->data;
+	int i, sg_count, len;
+	bool end = false;
+	bool dma64 = cq_host->dma64;
+	dma_addr_t addr;
+	u8 *desc;
+	struct scatterlist *sg;
+
+	sg_count = cqhci_dma_map(mrq->host, mrq);
+	if (sg_count < 0) {
+		pr_err("%s: %s: unable to map sg lists, %d\n",
+				mmc_hostname(mrq->host), __func__, sg_count);
+		return sg_count;
+	}
+
+	desc = get_trans_desc(cq_host, tag);
+
+	for_each_sg(data->sg, sg, sg_count, i) {
+		addr = sg_dma_address(sg);
+		len = sg_dma_len(sg);
+
+		if ((i+1) == sg_count)
+			end = true;
+		cqhci_set_tran_desc(desc, addr, len, end, dma64);
+		desc += cq_host->trans_desc_len;
+	}
+
+	return 0;
+}
+
+static void cqhci_prep_dcmd_desc(struct mmc_host *mmc,
+				   struct mmc_request *mrq)
+{
+	u64 *task_desc = NULL;
+	u64 data = 0;
+	u8 resp_type;
+	u8 *desc;
+	__le64 *dataddr;
+	struct cqhci_host *cq_host = mmc->cqe_private;
+	u8 timing;
+
+	if (!(mrq->cmd->flags & MMC_RSP_PRESENT)) {
+		resp_type = 0x0;
+		timing = 0x1;
+	} else {
+		if (mrq->cmd->flags & MMC_RSP_R1B) {
+			resp_type = 0x3;
+			timing = 0x0;
+		} else {
+			resp_type = 0x2;
+			timing = 0x1;
+		}
+	}
+
+	task_desc = (__le64 __force *)get_desc(cq_host, cq_host->dcmd_slot);
+	memset(task_desc, 0, cq_host->task_desc_len);
+	data |= (CQHCI_VALID(1) |
+		 CQHCI_END(1) |
+		 CQHCI_INT(1) |
+		 CQHCI_QBAR(1) |
+		 CQHCI_ACT(0x5) |
+		 CQHCI_CMD_INDEX(mrq->cmd->opcode) |
+		 CQHCI_CMD_TIMING(timing) | CQHCI_RESP_TYPE(resp_type));
+	*task_desc |= data;
+	desc = (u8 *)task_desc;
+	pr_debug("%s: cqhci: dcmd: cmd: %d timing: %d resp: %d\n",
+		 mmc_hostname(mmc), mrq->cmd->opcode, timing, resp_type);
+	dataddr = (__le64 __force *)(desc + 4);
+	dataddr[0] = cpu_to_le64((u64)mrq->cmd->arg);
+
+}
+
+static void cqhci_post_req(struct mmc_host *host, struct mmc_request *mrq)
+{
+	struct mmc_data *data = mrq->data;
+
+	if (data) {
+		dma_unmap_sg(mmc_dev(host), data->sg, data->sg_len,
+			     (data->flags & MMC_DATA_READ) ?
+			     DMA_FROM_DEVICE : DMA_TO_DEVICE);
+	}
+}
+
+static inline int cqhci_tag(struct mmc_request *mrq)
+{
+	return mrq->cmd ? DCMD_SLOT : mrq->tag;
+}
+
+static int cqhci_request(struct mmc_host *mmc, struct mmc_request *mrq)
+{
+	int err = 0;
+	u64 data = 0;
+	u64 *task_desc = NULL;
+	int tag = cqhci_tag(mrq);
+	struct cqhci_host *cq_host = mmc->cqe_private;
+	unsigned long flags;
+
+	if (!cq_host->enabled) {
+		pr_err("%s: cqhci: not enabled\n", mmc_hostname(mmc));
+		return -EINVAL;
+	}
+
+	/* First request after resume has to re-enable */
+	if (!cq_host->activated)
+		__cqhci_enable(cq_host);
+
+	if (!mmc->cqe_on) {
+		cqhci_writel(cq_host, 0, CQHCI_CTL);
+		mmc->cqe_on = true;
+		pr_debug("%s: cqhci: CQE on\n", mmc_hostname(mmc));
+		if (cqhci_readl(cq_host, CQHCI_CTL) && CQHCI_HALT) {
+			pr_err("%s: cqhci: CQE failed to exit halt state\n",
+			       mmc_hostname(mmc));
+		}
+		if (cq_host->ops->enable)
+			cq_host->ops->enable(mmc);
+	}
+
+	if (mrq->data) {
+		task_desc = (__le64 __force *)get_desc(cq_host, tag);
+		cqhci_prep_task_desc(mrq, &data, 1);
+		*task_desc = cpu_to_le64(data);
+		err = cqhci_prep_tran_desc(mrq, cq_host, tag);
+		if (err) {
+			pr_err("%s: cqhci: failed to setup tx desc: %d\n",
+			       mmc_hostname(mmc), err);
+			return err;
+		}
+	} else {
+		cqhci_prep_dcmd_desc(mmc, mrq);
+	}
+
+	spin_lock_irqsave(&cq_host->lock, flags);
+
+	if (cq_host->recovery_halt) {
+		err = -EBUSY;
+		goto out_unlock;
+	}
+
+	cq_host->slot[tag].mrq = mrq;
+	cq_host->slot[tag].flags = 0;
+
+	cq_host->qcnt += 1;
+
+	cqhci_writel(cq_host, 1 << tag, CQHCI_TDBR);
+	if (!(cqhci_readl(cq_host, CQHCI_TDBR) & (1 << tag)))
+		pr_debug("%s: cqhci: doorbell not set for tag %d\n",
+			 mmc_hostname(mmc), tag);
+out_unlock:
+	spin_unlock_irqrestore(&cq_host->lock, flags);
+
+	if (err)
+		cqhci_post_req(mmc, mrq);
+
+	return err;
+}
+
+static void cqhci_recovery_needed(struct mmc_host *mmc, struct mmc_request *mrq,
+				  bool notify)
+{
+	struct cqhci_host *cq_host = mmc->cqe_private;
+
+	if (!cq_host->recovery_halt) {
+		cq_host->recovery_halt = true;
+		pr_debug("%s: cqhci: recovery needed\n", mmc_hostname(mmc));
+		wake_up(&cq_host->wait_queue);
+		if (notify && mrq->recovery_notifier)
+			mrq->recovery_notifier(mrq);
+	}
+}
+
+static unsigned int cqhci_error_flags(int error1, int error2)
+{
+	int error = error1 ? error1 : error2;
+
+	switch (error) {
+	case -EILSEQ:
+		return CQHCI_HOST_CRC;
+	case -ETIMEDOUT:
+		return CQHCI_HOST_TIMEOUT;
+	default:
+		return CQHCI_HOST_OTHER;
+	}
+}
+
+static void cqhci_error_irq(struct mmc_host *mmc, u32 status, int cmd_error,
+			    int data_error)
+{
+	struct cqhci_host *cq_host = mmc->cqe_private;
+	struct cqhci_slot *slot;
+	u32 terri;
+	int tag;
+
+	spin_lock(&cq_host->lock);
+
+	terri = cqhci_readl(cq_host, CQHCI_TERRI);
+
+	pr_debug("%s: cqhci: error IRQ status: 0x%08x cmd error %d data error %d TERRI: 0x%08x\n",
+		 mmc_hostname(mmc), status, cmd_error, data_error, terri);
+
+	/* Forget about errors when recovery has already been triggered */
+	if (cq_host->recovery_halt)
+		goto out_unlock;
+
+	if (!cq_host->qcnt) {
+		WARN_ONCE(1, "%s: cqhci: error when idle. IRQ status: 0x%08x cmd error %d data error %d TERRI: 0x%08x\n",
+			  mmc_hostname(mmc), status, cmd_error, data_error,
+			  terri);
+		goto out_unlock;
+	}
+
+	if (CQHCI_TERRI_C_VALID(terri)) {
+		tag = CQHCI_TERRI_C_TASK(terri);
+		slot = &cq_host->slot[tag];
+		if (slot->mrq) {
+			slot->flags = cqhci_error_flags(cmd_error, data_error);
+			cqhci_recovery_needed(mmc, slot->mrq, true);
+		}
+	}
+
+	if (CQHCI_TERRI_D_VALID(terri)) {
+		tag = CQHCI_TERRI_D_TASK(terri);
+		slot = &cq_host->slot[tag];
+		if (slot->mrq) {
+			slot->flags = cqhci_error_flags(data_error, cmd_error);
+			cqhci_recovery_needed(mmc, slot->mrq, true);
+		}
+	}
+
+	if (!cq_host->recovery_halt) {
+		/*
+		 * The only way to guarantee forward progress is to mark at
+		 * least one task in error, so if none is indicated, pick one.
+		 */
+		for (tag = 0; tag < NUM_SLOTS; tag++) {
+			slot = &cq_host->slot[tag];
+			if (!slot->mrq)
+				continue;
+			slot->flags = cqhci_error_flags(data_error, cmd_error);
+			cqhci_recovery_needed(mmc, slot->mrq, true);
+			break;
+		}
+	}
+
+out_unlock:
+	spin_unlock(&cq_host->lock);
+}
+
+static void cqhci_finish_mrq(struct mmc_host *mmc, unsigned int tag)
+{
+	struct cqhci_host *cq_host = mmc->cqe_private;
+	struct cqhci_slot *slot = &cq_host->slot[tag];
+	struct mmc_request *mrq = slot->mrq;
+	struct mmc_data *data;
+
+	if (!mrq) {
+		WARN_ONCE(1, "%s: cqhci: spurious TCN for tag %d\n",
+			  mmc_hostname(mmc), tag);
+		return;
+	}
+
+	/* No completions allowed during recovery */
+	if (cq_host->recovery_halt) {
+		slot->flags |= CQHCI_COMPLETED;
+		return;
+	}
+
+	slot->mrq = NULL;
+
+	cq_host->qcnt -= 1;
+
+	data = mrq->data;
+	if (data) {
+		if (data->error)
+			data->bytes_xfered = 0;
+		else
+			data->bytes_xfered = data->blksz * data->blocks;
+	}
+
+	mmc_cqe_request_done(mmc, mrq);
+}
+
+irqreturn_t cqhci_irq(struct mmc_host *mmc, u32 intmask, int cmd_error,
+		      int data_error)
+{
+	u32 status;
+	unsigned long tag = 0, comp_status;
+	struct cqhci_host *cq_host = mmc->cqe_private;
+
+	status = cqhci_readl(cq_host, CQHCI_IS);
+	cqhci_writel(cq_host, status, CQHCI_IS);
+
+	pr_debug("%s: cqhci: IRQ status: 0x%08x\n", mmc_hostname(mmc), status);
+
+	if ((status & CQHCI_IS_RED) || cmd_error || data_error)
+		cqhci_error_irq(mmc, status, cmd_error, data_error);
+
+	if (status & CQHCI_IS_TCC) {
+		/* read TCN and complete the request */
+		comp_status = cqhci_readl(cq_host, CQHCI_TCN);
+		cqhci_writel(cq_host, comp_status, CQHCI_TCN);
+		pr_debug("%s: cqhci: TCN: 0x%08lx\n",
+			 mmc_hostname(mmc), comp_status);
+
+		spin_lock(&cq_host->lock);
+
+		for_each_set_bit(tag, &comp_status, cq_host->num_slots) {
+			/* complete the corresponding mrq */
+			pr_debug("%s: cqhci: completing tag %lu\n",
+				 mmc_hostname(mmc), tag);
+			cqhci_finish_mrq(mmc, tag);
+		}
+
+		if (cq_host->waiting_for_idle && !cq_host->qcnt) {
+			cq_host->waiting_for_idle = false;
+			wake_up(&cq_host->wait_queue);
+		}
+
+		spin_unlock(&cq_host->lock);
+	}
+
+	if (status & CQHCI_IS_TCL)
+		wake_up(&cq_host->wait_queue);
+
+	if (status & CQHCI_IS_HAC)
+		wake_up(&cq_host->wait_queue);
+
+	return IRQ_HANDLED;
+}
+EXPORT_SYMBOL(cqhci_irq);
+
+static bool cqhci_is_idle(struct cqhci_host *cq_host, int *ret)
+{
+	unsigned long flags;
+	bool is_idle;
+
+	spin_lock_irqsave(&cq_host->lock, flags);
+	is_idle = !cq_host->qcnt || cq_host->recovery_halt;
+	*ret = cq_host->recovery_halt ? -EBUSY : 0;
+	cq_host->waiting_for_idle = !is_idle;
+	spin_unlock_irqrestore(&cq_host->lock, flags);
+
+	return is_idle;
+}
+
+static int cqhci_wait_for_idle(struct mmc_host *mmc)
+{
+	struct cqhci_host *cq_host = mmc->cqe_private;
+	int ret;
+
+	wait_event(cq_host->wait_queue, cqhci_is_idle(cq_host, &ret));
+
+	return ret;
+}
+
+static bool cqhci_timeout(struct mmc_host *mmc, struct mmc_request *mrq,
+			  bool *recovery_needed)
+{
+	struct cqhci_host *cq_host = mmc->cqe_private;
+	int tag = cqhci_tag(mrq);
+	struct cqhci_slot *slot = &cq_host->slot[tag];
+	unsigned long flags;
+	bool timed_out;
+
+	spin_lock_irqsave(&cq_host->lock, flags);
+	timed_out = slot->mrq == mrq;
+	if (timed_out) {
+		slot->flags |= CQHCI_EXTERNAL_TIMEOUT;
+		cqhci_recovery_needed(mmc, mrq, false);
+		*recovery_needed = cq_host->recovery_halt;
+	}
+	spin_unlock_irqrestore(&cq_host->lock, flags);
+
+	if (timed_out) {
+		pr_err("%s: cqhci: timeout for tag %d\n",
+		       mmc_hostname(mmc), tag);
+		cqhci_dumpregs(cq_host);
+	}
+
+	return timed_out;
+}
+
+static bool cqhci_tasks_cleared(struct cqhci_host *cq_host)
+{
+	return !(cqhci_readl(cq_host, CQHCI_CTL) & CQHCI_CLEAR_ALL_TASKS);
+}
+
+static bool cqhci_clear_all_tasks(struct mmc_host *mmc, unsigned int timeout)
+{
+	struct cqhci_host *cq_host = mmc->cqe_private;
+	bool ret;
+	u32 ctl;
+
+	cqhci_set_irqs(cq_host, CQHCI_IS_TCL);
+
+	ctl = cqhci_readl(cq_host, CQHCI_CTL);
+	ctl |= CQHCI_CLEAR_ALL_TASKS;
+	cqhci_writel(cq_host, ctl, CQHCI_CTL);
+
+	wait_event_timeout(cq_host->wait_queue, cqhci_tasks_cleared(cq_host),
+			   msecs_to_jiffies(timeout) + 1);
+
+	cqhci_set_irqs(cq_host, 0);
+
+	ret = cqhci_tasks_cleared(cq_host);
+
+	if (!ret)
+		pr_debug("%s: cqhci: Failed to clear tasks\n",
+			 mmc_hostname(mmc));
+
+	return ret;
+}
+
+static bool cqhci_halted(struct cqhci_host *cq_host)
+{
+	return cqhci_readl(cq_host, CQHCI_CTL) & CQHCI_HALT;
+}
+
+static bool cqhci_halt(struct mmc_host *mmc, unsigned int timeout)
+{
+	struct cqhci_host *cq_host = mmc->cqe_private;
+	bool ret;
+	u32 ctl;
+
+	if (cqhci_halted(cq_host))
+		return true;
+
+	cqhci_set_irqs(cq_host, CQHCI_IS_HAC);
+
+	ctl = cqhci_readl(cq_host, CQHCI_CTL);
+	ctl |= CQHCI_HALT;
+	cqhci_writel(cq_host, ctl, CQHCI_CTL);
+
+	wait_event_timeout(cq_host->wait_queue, cqhci_halted(cq_host),
+			   msecs_to_jiffies(timeout) + 1);
+
+	cqhci_set_irqs(cq_host, 0);
+
+	ret = cqhci_halted(cq_host);
+
+	if (!ret)
+		pr_debug("%s: cqhci: Failed to halt\n", mmc_hostname(mmc));
+
+	return ret;
+}
+
+/*
+ * After halting we expect to be able to use the command line. We interpret the
+ * failure to halt to mean the data lines might still be in use (and the upper
+ * layers will need to send a STOP command), so we set the timeout based on a
+ * generous command timeout.
+ */
+#define CQHCI_START_HALT_TIMEOUT	5
+
+static void cqhci_recovery_start(struct mmc_host *mmc)
+{
+	struct cqhci_host *cq_host = mmc->cqe_private;
+
+	pr_debug("%s: cqhci: %s\n", mmc_hostname(mmc), __func__);
+
+	WARN_ON(!cq_host->recovery_halt);
+
+	cqhci_halt(mmc, CQHCI_START_HALT_TIMEOUT);
+
+	if (cq_host->ops->disable)
+		cq_host->ops->disable(mmc, true);
+
+	mmc->cqe_on = false;
+}
+
+static int cqhci_error_from_flags(unsigned int flags)
+{
+	if (!flags)
+		return 0;
+
+	/* CRC errors might indicate re-tuning so prefer to report that */
+	if (flags & CQHCI_HOST_CRC)
+		return -EILSEQ;
+
+	if (flags & (CQHCI_EXTERNAL_TIMEOUT | CQHCI_HOST_TIMEOUT))
+		return -ETIMEDOUT;
+
+	return -EIO;
+}
+
+static void cqhci_recover_mrq(struct cqhci_host *cq_host, unsigned int tag)
+{
+	struct cqhci_slot *slot = &cq_host->slot[tag];
+	struct mmc_request *mrq = slot->mrq;
+	struct mmc_data *data;
+
+	if (!mrq)
+		return;
+
+	slot->mrq = NULL;
+
+	cq_host->qcnt -= 1;
+
+	data = mrq->data;
+	if (data) {
+		data->bytes_xfered = 0;
+		data->error = cqhci_error_from_flags(slot->flags);
+	} else {
+		mrq->cmd->error = cqhci_error_from_flags(slot->flags);
+	}
+
+	mmc_cqe_request_done(cq_host->mmc, mrq);
+}
+
+static void cqhci_recover_mrqs(struct cqhci_host *cq_host)
+{
+	int i;
+
+	for (i = 0; i < cq_host->num_slots; i++)
+		cqhci_recover_mrq(cq_host, i);
+}
+
+/*
+ * By now the command and data lines should be unused so there is no reason for
+ * CQHCI to take a long time to halt, but if it doesn't halt there could be
+ * problems clearing tasks, so be generous.
+ */
+#define CQHCI_FINISH_HALT_TIMEOUT	20
+
+/* CQHCI could be expected to clear it's internal state pretty quickly */
+#define CQHCI_CLEAR_TIMEOUT		20
+
+static void cqhci_recovery_finish(struct mmc_host *mmc)
+{
+	struct cqhci_host *cq_host = mmc->cqe_private;
+	unsigned long flags;
+	u32 cqcfg;
+	bool ok;
+
+	pr_debug("%s: cqhci: %s\n", mmc_hostname(mmc), __func__);
+
+	WARN_ON(!cq_host->recovery_halt);
+
+	ok = cqhci_halt(mmc, CQHCI_FINISH_HALT_TIMEOUT);
+
+	if (!cqhci_clear_all_tasks(mmc, CQHCI_CLEAR_TIMEOUT))
+		ok = false;
+
+	/*
+	 * The specification contradicts itself, by saying that tasks cannot be
+	 * cleared if CQHCI does not halt, but if CQHCI does not halt, it should
+	 * be disabled/re-enabled, but not to disable before clearing tasks.
+	 * Have a go anyway.
+	 */
+	if (!ok) {
+		pr_debug("%s: cqhci: disable / re-enable\n", mmc_hostname(mmc));
+		cqcfg = cqhci_readl(cq_host, CQHCI_CFG);
+		cqcfg &= ~CQHCI_ENABLE;
+		cqhci_writel(cq_host, cqcfg, CQHCI_CFG);
+		cqcfg |= CQHCI_ENABLE;
+		cqhci_writel(cq_host, cqcfg, CQHCI_CFG);
+		/* Be sure that there are no tasks */
+		ok = cqhci_halt(mmc, CQHCI_FINISH_HALT_TIMEOUT);
+		if (!cqhci_clear_all_tasks(mmc, CQHCI_CLEAR_TIMEOUT))
+			ok = false;
+		WARN_ON(!ok);
+	}
+
+	cqhci_recover_mrqs(cq_host);
+
+	WARN_ON(cq_host->qcnt);
+
+	spin_lock_irqsave(&cq_host->lock, flags);
+	cq_host->qcnt = 0;
+	cq_host->recovery_halt = false;
+	mmc->cqe_on = false;
+	spin_unlock_irqrestore(&cq_host->lock, flags);
+
+	/* Ensure all writes are done before interrupts are re-enabled */
+	wmb();
+
+	cqhci_writel(cq_host, CQHCI_IS_HAC | CQHCI_IS_TCL, CQHCI_IS);
+
+	cqhci_set_irqs(cq_host, CQHCI_IS_MASK);
+
+	pr_debug("%s: cqhci: recovery done\n", mmc_hostname(mmc));
+}
+
+static const struct mmc_cqe_ops cqhci_cqe_ops = {
+	.cqe_enable = cqhci_enable,
+	.cqe_disable = cqhci_disable,
+	.cqe_request = cqhci_request,
+	.cqe_post_req = cqhci_post_req,
+	.cqe_off = cqhci_off,
+	.cqe_wait_for_idle = cqhci_wait_for_idle,
+	.cqe_timeout = cqhci_timeout,
+	.cqe_recovery_start = cqhci_recovery_start,
+	.cqe_recovery_finish = cqhci_recovery_finish,
+};
+
+struct cqhci_host *cqhci_pltfm_init(struct platform_device *pdev)
+{
+	struct cqhci_host *cq_host;
+	struct resource *cqhci_memres = NULL;
+
+	/* check and setup CMDQ interface */
+	cqhci_memres = platform_get_resource_byname(pdev, IORESOURCE_MEM,
+						   "cqhci_mem");
+	if (!cqhci_memres) {
+		dev_dbg(&pdev->dev, "CMDQ not supported\n");
+		return ERR_PTR(-EINVAL);
+	}
+
+	cq_host = devm_kzalloc(&pdev->dev, sizeof(*cq_host), GFP_KERNEL);
+	if (!cq_host)
+		return ERR_PTR(-ENOMEM);
+	cq_host->mmio = devm_ioremap(&pdev->dev,
+				     cqhci_memres->start,
+				     resource_size(cqhci_memres));
+	if (!cq_host->mmio) {
+		dev_err(&pdev->dev, "failed to remap cqhci regs\n");
+		return ERR_PTR(-EBUSY);
+	}
+	dev_dbg(&pdev->dev, "CMDQ ioremap: done\n");
+
+	return cq_host;
+}
+EXPORT_SYMBOL(cqhci_pltfm_init);
+
+static unsigned int cqhci_ver_major(struct cqhci_host *cq_host)
+{
+	return CQHCI_VER_MAJOR(cqhci_readl(cq_host, CQHCI_VER));
+}
+
+static unsigned int cqhci_ver_minor(struct cqhci_host *cq_host)
+{
+	u32 ver = cqhci_readl(cq_host, CQHCI_VER);
+
+	return CQHCI_VER_MINOR1(ver) * 10 + CQHCI_VER_MINOR2(ver);
+}
+
+int cqhci_init(struct cqhci_host *cq_host, struct mmc_host *mmc,
+	      bool dma64)
+{
+	int err;
+
+	cq_host->dma64 = dma64;
+	cq_host->mmc = mmc;
+	cq_host->mmc->cqe_private = cq_host;
+
+	cq_host->num_slots = NUM_SLOTS;
+	cq_host->dcmd_slot = DCMD_SLOT;
+
+	mmc->cqe_ops = &cqhci_cqe_ops;
+
+	mmc->cqe_qdepth = NUM_SLOTS;
+	if (mmc->caps2 & MMC_CAP2_CQE_DCMD)
+		mmc->cqe_qdepth -= 1;
+
+	cq_host->slot = devm_kcalloc(mmc_dev(mmc), cq_host->num_slots,
+				     sizeof(*cq_host->slot), GFP_KERNEL);
+	if (!cq_host->slot) {
+		err = -ENOMEM;
+		goto out_err;
+	}
+
+	spin_lock_init(&cq_host->lock);
+
+	init_completion(&cq_host->halt_comp);
+	init_waitqueue_head(&cq_host->wait_queue);
+
+	pr_info("%s: CQHCI version %u.%02u\n",
+		mmc_hostname(mmc), cqhci_ver_major(cq_host),
+		cqhci_ver_minor(cq_host));
+
+	return 0;
+
+out_err:
+	pr_err("%s: CQHCI version %u.%02u failed to initialize, error %d\n",
+	       mmc_hostname(mmc), cqhci_ver_major(cq_host),
+	       cqhci_ver_minor(cq_host), err);
+	return err;
+}
+EXPORT_SYMBOL(cqhci_init);
+
+MODULE_AUTHOR("Venkat Gopalakrishnan <venkatg@codeaurora.org>");
+MODULE_DESCRIPTION("Command Queue Host Controller Interface driver");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/mmc/host/cqhci.h b/drivers/mmc/host/cqhci.h
new file mode 100644
index 000000000000..2d39d361b322
--- /dev/null
+++ b/drivers/mmc/host/cqhci.h
@@ -0,0 +1,240 @@
+/* Copyright (c) 2015, The Linux Foundation. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+#ifndef LINUX_MMC_CQHCI_H
+#define LINUX_MMC_CQHCI_H
+
+#include <linux/compiler.h>
+#include <linux/bitops.h>
+#include <linux/spinlock_types.h>
+#include <linux/types.h>
+#include <linux/completion.h>
+#include <linux/wait.h>
+#include <linux/irqreturn.h>
+#include <asm/io.h>
+
+/* registers */
+/* version */
+#define CQHCI_VER			0x00
+#define CQHCI_VER_MAJOR(x)		(((x) & GENMASK(11, 8)) >> 8)
+#define CQHCI_VER_MINOR1(x)		(((x) & GENMASK(7, 4)) >> 4)
+#define CQHCI_VER_MINOR2(x)		((x) & GENMASK(3, 0))
+
+/* capabilities */
+#define CQHCI_CAP			0x04
+/* configuration */
+#define CQHCI_CFG			0x08
+#define CQHCI_DCMD			0x00001000
+#define CQHCI_TASK_DESC_SZ		0x00000100
+#define CQHCI_ENABLE			0x00000001
+
+/* control */
+#define CQHCI_CTL			0x0C
+#define CQHCI_CLEAR_ALL_TASKS		0x00000100
+#define CQHCI_HALT			0x00000001
+
+/* interrupt status */
+#define CQHCI_IS			0x10
+#define CQHCI_IS_HAC			BIT(0)
+#define CQHCI_IS_TCC			BIT(1)
+#define CQHCI_IS_RED			BIT(2)
+#define CQHCI_IS_TCL			BIT(3)
+
+#define CQHCI_IS_MASK (CQHCI_IS_TCC | CQHCI_IS_RED)
+
+/* interrupt status enable */
+#define CQHCI_ISTE			0x14
+
+/* interrupt signal enable */
+#define CQHCI_ISGE			0x18
+
+/* interrupt coalescing */
+#define CQHCI_IC			0x1C
+#define CQHCI_IC_ENABLE			BIT(31)
+#define CQHCI_IC_RESET			BIT(16)
+#define CQHCI_IC_ICCTHWEN		BIT(15)
+#define CQHCI_IC_ICCTH(x)		((x & 0x1F) << 8)
+#define CQHCI_IC_ICTOVALWEN		BIT(7)
+#define CQHCI_IC_ICTOVAL(x)		(x & 0x7F)
+
+/* task list base address */
+#define CQHCI_TDLBA			0x20
+
+/* task list base address upper */
+#define CQHCI_TDLBAU			0x24
+
+/* door-bell */
+#define CQHCI_TDBR			0x28
+
+/* task completion notification */
+#define CQHCI_TCN			0x2C
+
+/* device queue status */
+#define CQHCI_DQS			0x30
+
+/* device pending tasks */
+#define CQHCI_DPT			0x34
+
+/* task clear */
+#define CQHCI_TCLR			0x38
+
+/* send status config 1 */
+#define CQHCI_SSC1			0x40
+
+/* send status config 2 */
+#define CQHCI_SSC2			0x44
+
+/* response for dcmd */
+#define CQHCI_CRDCT			0x48
+
+/* response mode error mask */
+#define CQHCI_RMEM			0x50
+
+/* task error info */
+#define CQHCI_TERRI			0x54
+
+#define CQHCI_TERRI_C_INDEX(x)		((x) & GENMASK(5, 0))
+#define CQHCI_TERRI_C_TASK(x)		(((x) & GENMASK(12, 8)) >> 8)
+#define CQHCI_TERRI_C_VALID(x)		((x) & BIT(15))
+#define CQHCI_TERRI_D_INDEX(x)		(((x) & GENMASK(21, 16)) >> 16)
+#define CQHCI_TERRI_D_TASK(x)		(((x) & GENMASK(28, 24)) >> 24)
+#define CQHCI_TERRI_D_VALID(x)		((x) & BIT(31))
+
+/* command response index */
+#define CQHCI_CRI			0x58
+
+/* command response argument */
+#define CQHCI_CRA			0x5C
+
+#define CQHCI_INT_ALL			0xF
+#define CQHCI_IC_DEFAULT_ICCTH		31
+#define CQHCI_IC_DEFAULT_ICTOVAL	1
+
+/* attribute fields */
+#define CQHCI_VALID(x)			((x & 1) << 0)
+#define CQHCI_END(x)			((x & 1) << 1)
+#define CQHCI_INT(x)			((x & 1) << 2)
+#define CQHCI_ACT(x)			((x & 0x7) << 3)
+
+/* data command task descriptor fields */
+#define CQHCI_FORCED_PROG(x)		((x & 1) << 6)
+#define CQHCI_CONTEXT(x)		((x & 0xF) << 7)
+#define CQHCI_DATA_TAG(x)		((x & 1) << 11)
+#define CQHCI_DATA_DIR(x)		((x & 1) << 12)
+#define CQHCI_PRIORITY(x)		((x & 1) << 13)
+#define CQHCI_QBAR(x)			((x & 1) << 14)
+#define CQHCI_REL_WRITE(x)		((x & 1) << 15)
+#define CQHCI_BLK_COUNT(x)		((x & 0xFFFF) << 16)
+#define CQHCI_BLK_ADDR(x)		((x & 0xFFFFFFFF) << 32)
+
+/* direct command task descriptor fields */
+#define CQHCI_CMD_INDEX(x)		((x & 0x3F) << 16)
+#define CQHCI_CMD_TIMING(x)		((x & 1) << 22)
+#define CQHCI_RESP_TYPE(x)		((x & 0x3) << 23)
+
+/* transfer descriptor fields */
+#define CQHCI_DAT_LENGTH(x)		((x & 0xFFFF) << 16)
+#define CQHCI_DAT_ADDR_LO(x)		((x & 0xFFFFFFFF) << 32)
+#define CQHCI_DAT_ADDR_HI(x)		((x & 0xFFFFFFFF) << 0)
+
+struct cqhci_host_ops;
+struct mmc_host;
+struct cqhci_slot;
+
+struct cqhci_host {
+	const struct cqhci_host_ops *ops;
+	void __iomem *mmio;
+	struct mmc_host *mmc;
+
+	spinlock_t lock;
+
+	/* relative card address of device */
+	unsigned int rca;
+
+	/* 64 bit DMA */
+	bool dma64;
+	int num_slots;
+	int qcnt;
+
+	u32 dcmd_slot;
+	u32 caps;
+#define CQHCI_TASK_DESC_SZ_128		0x1
+
+	u32 quirks;
+#define CQHCI_QUIRK_SHORT_TXFR_DESC_SZ	0x1
+
+	bool enabled;
+	bool halted;
+	bool init_done;
+	bool activated;
+	bool waiting_for_idle;
+	bool recovery_halt;
+
+	size_t desc_size;
+	size_t data_size;
+
+	u8 *desc_base;
+
+	/* total descriptor size */
+	u8 slot_sz;
+
+	/* 64/128 bit depends on CQHCI_CFG */
+	u8 task_desc_len;
+
+	/* 64 bit on 32-bit arch, 128 bit on 64-bit */
+	u8 link_desc_len;
+
+	u8 *trans_desc_base;
+	/* same length as transfer descriptor */
+	u8 trans_desc_len;
+
+	dma_addr_t desc_dma_base;
+	dma_addr_t trans_desc_dma_base;
+
+	struct completion halt_comp;
+	wait_queue_head_t wait_queue;
+	struct cqhci_slot *slot;
+};
+
+struct cqhci_host_ops {
+	void (*dumpregs)(struct mmc_host *mmc);
+	void (*write_l)(struct cqhci_host *host, u32 val, int reg);
+	u32 (*read_l)(struct cqhci_host *host, int reg);
+	void (*enable)(struct mmc_host *mmc);
+	void (*disable)(struct mmc_host *mmc, bool recovery);
+};
+
+static inline void cqhci_writel(struct cqhci_host *host, u32 val, int reg)
+{
+	if (unlikely(host->ops->write_l))
+		host->ops->write_l(host, val, reg);
+	else
+		writel_relaxed(val, host->mmio + reg);
+}
+
+static inline u32 cqhci_readl(struct cqhci_host *host, int reg)
+{
+	if (unlikely(host->ops->read_l))
+		return host->ops->read_l(host, reg);
+	else
+		return readl_relaxed(host->mmio + reg);
+}
+
+struct platform_device;
+
+irqreturn_t cqhci_irq(struct mmc_host *mmc, u32 intmask, int cmd_error,
+		      int data_error);
+int cqhci_init(struct cqhci_host *cq_host, struct mmc_host *mmc, bool dma64);
+struct cqhci_host *cqhci_pltfm_init(struct platform_device *pdev);
+int cqhci_suspend(struct mmc_host *mmc);
+int cqhci_resume(struct mmc_host *mmc);
+
+#endif
-- 
1.9.1

[PATCH V8 14/14] mmc: sdhci-pci: Add CQHCI support for Intel GLK

[Adrian Hunter]

Add CQHCI initialization and implement CQHCI operations for Intel GLK.

Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
---
 drivers/mmc/host/Kconfig          |   1 +
 drivers/mmc/host/sdhci-pci-core.c | 154 +++++++++++++++++++++++++++++++++++++-
 2 files changed, 154 insertions(+), 1 deletion(-)

diff --git a/drivers/mmc/host/Kconfig b/drivers/mmc/host/Kconfig
index 6db8e46b4e12..23621c52d2f8 100644
--- a/drivers/mmc/host/Kconfig
+++ b/drivers/mmc/host/Kconfig
@@ -81,6 +81,7 @@ config MMC_SDHCI_BIG_ENDIAN_32BIT_BYTE_SWAPPER
 config MMC_SDHCI_PCI
 	tristate "SDHCI support on PCI bus"
 	depends on MMC_SDHCI && PCI
+	select MMC_CQHCI
 	help
 	  This selects the PCI Secure Digital Host Controller Interface.
 	  Most controllers found today are PCI devices.
diff --git a/drivers/mmc/host/sdhci-pci-core.c b/drivers/mmc/host/sdhci-pci-core.c
index bbaddf18a1b3..e22075f99707 100644
--- a/drivers/mmc/host/sdhci-pci-core.c
+++ b/drivers/mmc/host/sdhci-pci-core.c
@@ -30,6 +30,8 @@
 #include <linux/mmc/sdhci-pci-data.h>
 #include <linux/acpi.h>
 
+#include "cqhci.h"
+
 #include "sdhci.h"
 #include "sdhci-pci.h"
 #include "sdhci-pci-o2micro.h"
@@ -117,6 +119,28 @@ int sdhci_pci_resume_host(struct sdhci_pci_chip *chip)
 
 	return 0;
 }
+
+static int sdhci_cqhci_suspend(struct sdhci_pci_chip *chip)
+{
+	int ret;
+
+	ret = cqhci_suspend(chip->slots[0]->host->mmc);
+	if (ret)
+		return ret;
+
+	return sdhci_pci_suspend_host(chip);
+}
+
+static int sdhci_cqhci_resume(struct sdhci_pci_chip *chip)
+{
+	int ret;
+
+	ret = sdhci_pci_resume_host(chip);
+	if (ret)
+		return ret;
+
+	return cqhci_resume(chip->slots[0]->host->mmc);
+}
 #endif
 
 #ifdef CONFIG_PM
@@ -167,8 +191,48 @@ static int sdhci_pci_runtime_resume_host(struct sdhci_pci_chip *chip)
 
 	return 0;
 }
+
+static int sdhci_cqhci_runtime_suspend(struct sdhci_pci_chip *chip)
+{
+	int ret;
+
+	ret = cqhci_suspend(chip->slots[0]->host->mmc);
+	if (ret)
+		return ret;
+
+	return sdhci_pci_runtime_suspend_host(chip);
+}
+
+static int sdhci_cqhci_runtime_resume(struct sdhci_pci_chip *chip)
+{
+	int ret;
+
+	ret = sdhci_pci_runtime_resume_host(chip);
+	if (ret)
+		return ret;
+
+	return cqhci_resume(chip->slots[0]->host->mmc);
+}
 #endif
 
+static u32 sdhci_cqhci_irq(struct sdhci_host *host, u32 intmask)
+{
+	int cmd_error = 0;
+	int data_error = 0;
+
+	if (!sdhci_cqe_irq(host, intmask, &cmd_error, &data_error))
+		return intmask;
+
+	cqhci_irq(host->mmc, intmask, cmd_error, data_error);
+
+	return 0;
+}
+
+static void sdhci_pci_dumpregs(struct mmc_host *mmc)
+{
+	sdhci_dumpregs(mmc_priv(mmc));
+}
+
 /*****************************************************************************\
  *                                                                           *
  * Hardware specific quirk handling                                          *
@@ -567,6 +631,17 @@ static void intel_hs400_enhanced_strobe(struct mmc_host *mmc,
 	.hw_reset		= sdhci_pci_hw_reset,
 };
 
+static const struct sdhci_ops sdhci_intel_glk_ops = {
+	.set_clock		= sdhci_set_clock,
+	.set_power		= sdhci_intel_set_power,
+	.enable_dma		= sdhci_pci_enable_dma,
+	.set_bus_width		= sdhci_set_bus_width,
+	.reset			= sdhci_reset,
+	.set_uhs_signaling	= sdhci_set_uhs_signaling,
+	.hw_reset		= sdhci_pci_hw_reset,
+	.irq			= sdhci_cqhci_irq,
+};
+
 static void byt_read_dsm(struct sdhci_pci_slot *slot)
 {
 	struct intel_host *intel_host = sdhci_pci_priv(slot);
@@ -596,15 +671,83 @@ static int glk_emmc_probe_slot(struct sdhci_pci_slot *slot)
 {
 	int ret = byt_emmc_probe_slot(slot);
 
+	slot->host->mmc->caps2 |= MMC_CAP2_CQE;
+
 	if (slot->chip->pdev->device != PCI_DEVICE_ID_INTEL_GLK_EMMC) {
 		slot->host->mmc->caps2 |= MMC_CAP2_HS400_ES,
 		slot->host->mmc_host_ops.hs400_enhanced_strobe =
 						intel_hs400_enhanced_strobe;
+		slot->host->mmc->caps2 |= MMC_CAP2_CQE_DCMD;
 	}
 
 	return ret;
 }
 
+static void glk_cqe_enable(struct mmc_host *mmc)
+{
+	struct sdhci_host *host = mmc_priv(mmc);
+	u32 reg;
+
+	/*
+	 * CQE gets stuck if it sees Buffer Read Enable bit set, which can be
+	 * the case after tuning, so ensure the buffer is drained.
+	 */
+	reg = sdhci_readl(host, SDHCI_PRESENT_STATE);
+	while (reg & SDHCI_DATA_AVAILABLE) {
+		sdhci_readl(host, SDHCI_BUFFER);
+		reg = sdhci_readl(host, SDHCI_PRESENT_STATE);
+	}
+
+	sdhci_cqe_enable(mmc);
+}
+
+static const struct cqhci_host_ops glk_cqhci_ops = {
+	.enable		= glk_cqe_enable,
+	.disable	= sdhci_cqe_disable,
+	.dumpregs	= sdhci_pci_dumpregs,
+};
+
+static int glk_emmc_add_host(struct sdhci_pci_slot *slot)
+{
+	struct device *dev = &slot->chip->pdev->dev;
+	struct sdhci_host *host = slot->host;
+	struct cqhci_host *cq_host;
+	bool dma64;
+	int ret;
+
+	ret = sdhci_setup_host(host);
+	if (ret)
+		return ret;
+
+	cq_host = devm_kzalloc(dev, sizeof(*cq_host), GFP_KERNEL);
+	if (!cq_host) {
+		ret = -ENOMEM;
+		goto cleanup;
+	}
+
+	cq_host->mmio = host->ioaddr + 0x200;
+	cq_host->quirks |= CQHCI_QUIRK_SHORT_TXFR_DESC_SZ;
+	cq_host->ops = &glk_cqhci_ops;
+
+	dma64 = host->flags & SDHCI_USE_64_BIT_DMA;
+	if (dma64)
+		cq_host->caps |= CQHCI_TASK_DESC_SZ_128;
+
+	ret = cqhci_init(cq_host, host->mmc, dma64);
+	if (ret)
+		goto cleanup;
+
+	ret = __sdhci_add_host(host);
+	if (ret)
+		goto cleanup;
+
+	return 0;
+
+cleanup:
+	sdhci_cleanup_host(host);
+	return ret;
+}
+
 #ifdef CONFIG_ACPI
 static int ni_set_max_freq(struct sdhci_pci_slot *slot)
 {
@@ -683,11 +826,20 @@ static int byt_sd_probe_slot(struct sdhci_pci_slot *slot)
 static const struct sdhci_pci_fixes sdhci_intel_glk_emmc = {
 	.allow_runtime_pm	= true,
 	.probe_slot		= glk_emmc_probe_slot,
+	.add_host		= glk_emmc_add_host,
+#ifdef CONFIG_PM_SLEEP
+	.suspend		= sdhci_cqhci_suspend,
+	.resume			= sdhci_cqhci_resume,
+#endif
+#ifdef CONFIG_PM
+	.runtime_suspend	= sdhci_cqhci_runtime_suspend,
+	.runtime_resume		= sdhci_cqhci_runtime_resume,
+#endif
 	.quirks			= SDHCI_QUIRK_NO_ENDATTR_IN_NOPDESC,
 	.quirks2		= SDHCI_QUIRK2_PRESET_VALUE_BROKEN |
 				  SDHCI_QUIRK2_CAPS_BIT63_FOR_HS400 |
 				  SDHCI_QUIRK2_STOP_WITH_TC,
-	.ops			= &sdhci_intel_byt_ops,
+	.ops			= &sdhci_intel_glk_ops,
 	.priv_size		= sizeof(struct intel_host),
 };
 
-- 
1.9.1

[PATCH V9 14/14] mmc: sdhci-pci: Add CQHCI support for Intel GLK

[Adrian Hunter]

Add CQHCI initialization and implement CQHCI operations for Intel GLK.

Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
---


Changes since V8:

Re-base on top of patch "mmc: sdhci-pci: Fix voltage switch for some Intel host controllers"


 drivers/mmc/host/Kconfig          |   1 +
 drivers/mmc/host/sdhci-pci-core.c | 155 +++++++++++++++++++++++++++++++++++++-
 2 files changed, 155 insertions(+), 1 deletion(-)

diff --git a/drivers/mmc/host/Kconfig b/drivers/mmc/host/Kconfig
index 6db8e46b4e12..23621c52d2f8 100644
--- a/drivers/mmc/host/Kconfig
+++ b/drivers/mmc/host/Kconfig
@@ -81,6 +81,7 @@ config MMC_SDHCI_BIG_ENDIAN_32BIT_BYTE_SWAPPER
 config MMC_SDHCI_PCI
 	tristate "SDHCI support on PCI bus"
 	depends on MMC_SDHCI && PCI
+	select MMC_CQHCI
 	help
 	  This selects the PCI Secure Digital Host Controller Interface.
 	  Most controllers found today are PCI devices.
diff --git a/drivers/mmc/host/sdhci-pci-core.c b/drivers/mmc/host/sdhci-pci-core.c
index 5f3f7b51299f..36fc5c2226a1 100644
--- a/drivers/mmc/host/sdhci-pci-core.c
+++ b/drivers/mmc/host/sdhci-pci-core.c
@@ -30,6 +30,8 @@
 #include <linux/mmc/sdhci-pci-data.h>
 #include <linux/acpi.h>
 
+#include "cqhci.h"
+
 #include "sdhci.h"
 #include "sdhci-pci.h"
 #include "sdhci-pci-o2micro.h"
@@ -117,6 +119,28 @@ int sdhci_pci_resume_host(struct sdhci_pci_chip *chip)
 
 	return 0;
 }
+
+static int sdhci_cqhci_suspend(struct sdhci_pci_chip *chip)
+{
+	int ret;
+
+	ret = cqhci_suspend(chip->slots[0]->host->mmc);
+	if (ret)
+		return ret;
+
+	return sdhci_pci_suspend_host(chip);
+}
+
+static int sdhci_cqhci_resume(struct sdhci_pci_chip *chip)
+{
+	int ret;
+
+	ret = sdhci_pci_resume_host(chip);
+	if (ret)
+		return ret;
+
+	return cqhci_resume(chip->slots[0]->host->mmc);
+}
 #endif
 
 #ifdef CONFIG_PM
@@ -167,8 +191,48 @@ static int sdhci_pci_runtime_resume_host(struct sdhci_pci_chip *chip)
 
 	return 0;
 }
+
+static int sdhci_cqhci_runtime_suspend(struct sdhci_pci_chip *chip)
+{
+	int ret;
+
+	ret = cqhci_suspend(chip->slots[0]->host->mmc);
+	if (ret)
+		return ret;
+
+	return sdhci_pci_runtime_suspend_host(chip);
+}
+
+static int sdhci_cqhci_runtime_resume(struct sdhci_pci_chip *chip)
+{
+	int ret;
+
+	ret = sdhci_pci_runtime_resume_host(chip);
+	if (ret)
+		return ret;
+
+	return cqhci_resume(chip->slots[0]->host->mmc);
+}
 #endif
 
+static u32 sdhci_cqhci_irq(struct sdhci_host *host, u32 intmask)
+{
+	int cmd_error = 0;
+	int data_error = 0;
+
+	if (!sdhci_cqe_irq(host, intmask, &cmd_error, &data_error))
+		return intmask;
+
+	cqhci_irq(host->mmc, intmask, cmd_error, data_error);
+
+	return 0;
+}
+
+static void sdhci_pci_dumpregs(struct mmc_host *mmc)
+{
+	sdhci_dumpregs(mmc_priv(mmc));
+}
+
 /*****************************************************************************\
  *                                                                           *
  * Hardware specific quirk handling                                          *
@@ -582,6 +646,18 @@ static void sdhci_intel_voltage_switch(struct sdhci_host *host)
 	.voltage_switch		= sdhci_intel_voltage_switch,
 };
 
+static const struct sdhci_ops sdhci_intel_glk_ops = {
+	.set_clock		= sdhci_set_clock,
+	.set_power		= sdhci_intel_set_power,
+	.enable_dma		= sdhci_pci_enable_dma,
+	.set_bus_width		= sdhci_set_bus_width,
+	.reset			= sdhci_reset,
+	.set_uhs_signaling	= sdhci_set_uhs_signaling,
+	.hw_reset		= sdhci_pci_hw_reset,
+	.voltage_switch		= sdhci_intel_voltage_switch,
+	.irq			= sdhci_cqhci_irq,
+};
+
 static void byt_read_dsm(struct sdhci_pci_slot *slot)
 {
 	struct intel_host *intel_host = sdhci_pci_priv(slot);
@@ -611,12 +687,80 @@ static int glk_emmc_probe_slot(struct sdhci_pci_slot *slot)
 {
 	int ret = byt_emmc_probe_slot(slot);
 
+	slot->host->mmc->caps2 |= MMC_CAP2_CQE;
+
 	if (slot->chip->pdev->device != PCI_DEVICE_ID_INTEL_GLK_EMMC) {
 		slot->host->mmc->caps2 |= MMC_CAP2_HS400_ES,
 		slot->host->mmc_host_ops.hs400_enhanced_strobe =
 						intel_hs400_enhanced_strobe;
+		slot->host->mmc->caps2 |= MMC_CAP2_CQE_DCMD;
+	}
+
+	return ret;
+}
+
+static void glk_cqe_enable(struct mmc_host *mmc)
+{
+	struct sdhci_host *host = mmc_priv(mmc);
+	u32 reg;
+
+	/*
+	 * CQE gets stuck if it sees Buffer Read Enable bit set, which can be
+	 * the case after tuning, so ensure the buffer is drained.
+	 */
+	reg = sdhci_readl(host, SDHCI_PRESENT_STATE);
+	while (reg & SDHCI_DATA_AVAILABLE) {
+		sdhci_readl(host, SDHCI_BUFFER);
+		reg = sdhci_readl(host, SDHCI_PRESENT_STATE);
+	}
+
+	sdhci_cqe_enable(mmc);
+}
+
+static const struct cqhci_host_ops glk_cqhci_ops = {
+	.enable		= glk_cqe_enable,
+	.disable	= sdhci_cqe_disable,
+	.dumpregs	= sdhci_pci_dumpregs,
+};
+
+static int glk_emmc_add_host(struct sdhci_pci_slot *slot)
+{
+	struct device *dev = &slot->chip->pdev->dev;
+	struct sdhci_host *host = slot->host;
+	struct cqhci_host *cq_host;
+	bool dma64;
+	int ret;
+
+	ret = sdhci_setup_host(host);
+	if (ret)
+		return ret;
+
+	cq_host = devm_kzalloc(dev, sizeof(*cq_host), GFP_KERNEL);
+	if (!cq_host) {
+		ret = -ENOMEM;
+		goto cleanup;
 	}
 
+	cq_host->mmio = host->ioaddr + 0x200;
+	cq_host->quirks |= CQHCI_QUIRK_SHORT_TXFR_DESC_SZ;
+	cq_host->ops = &glk_cqhci_ops;
+
+	dma64 = host->flags & SDHCI_USE_64_BIT_DMA;
+	if (dma64)
+		cq_host->caps |= CQHCI_TASK_DESC_SZ_128;
+
+	ret = cqhci_init(cq_host, host->mmc, dma64);
+	if (ret)
+		goto cleanup;
+
+	ret = __sdhci_add_host(host);
+	if (ret)
+		goto cleanup;
+
+	return 0;
+
+cleanup:
+	sdhci_cleanup_host(host);
 	return ret;
 }
 
@@ -698,11 +842,20 @@ static int byt_sd_probe_slot(struct sdhci_pci_slot *slot)
 static const struct sdhci_pci_fixes sdhci_intel_glk_emmc = {
 	.allow_runtime_pm	= true,
 	.probe_slot		= glk_emmc_probe_slot,
+	.add_host		= glk_emmc_add_host,
+#ifdef CONFIG_PM_SLEEP
+	.suspend		= sdhci_cqhci_suspend,
+	.resume			= sdhci_cqhci_resume,
+#endif
+#ifdef CONFIG_PM
+	.runtime_suspend	= sdhci_cqhci_runtime_suspend,
+	.runtime_resume		= sdhci_cqhci_runtime_resume,
+#endif
 	.quirks			= SDHCI_QUIRK_NO_ENDATTR_IN_NOPDESC,
 	.quirks2		= SDHCI_QUIRK2_PRESET_VALUE_BROKEN |
 				  SDHCI_QUIRK2_CAPS_BIT63_FOR_HS400 |
 				  SDHCI_QUIRK2_STOP_WITH_TC,
-	.ops			= &sdhci_intel_byt_ops,
+	.ops			= &sdhci_intel_glk_ops,
 	.priv_size		= sizeof(struct intel_host),
 };
 
-- 
1.9.1

Re: [PATCH V8 01/14] mmc: core: Introduce host claiming by context

[Ulf Hansson]

On 13 September 2017 at 13:40, Adrian Hunter <adrian.hunter@intel.com> wrote:
> Currently the host can be claimed by a task.  Change this so that the host
> can be claimed by a context that may or may not be a task.  This provides
> for the host to be claimed by a block driver queue to support blk-mq, while
> maintaining compatibility with the existing use of mmc_claim_host().

As stated earlier, I think this is a good approach, which allows us to
move forward.

Some comments below, most related how to avoid us from adding more
wrapper functions.

>
> Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
> ---
>  drivers/mmc/core/core.c  | 92 ++++++++++++++++++++++++++++++++++++++++++++----
>  drivers/mmc/core/core.h  |  6 ++++
>  include/linux/mmc/host.h |  7 +++-
>  3 files changed, 97 insertions(+), 8 deletions(-)
>
> diff --git a/drivers/mmc/core/core.c b/drivers/mmc/core/core.c
> index 66c9cf49ad2f..09fdc467b804 100644
> --- a/drivers/mmc/core/core.c
> +++ b/drivers/mmc/core/core.c
> @@ -832,9 +832,38 @@ unsigned int mmc_align_data_size(struct mmc_card *card, unsigned int sz)
>  }
>  EXPORT_SYMBOL(mmc_align_data_size);
>
> +/*
> + * Allow claiming an already claimed host if the context is the same or there is
> + * no context but the task is the same.
> + */
> +static inline bool mmc_ctx_matches(struct mmc_host *host, struct mmc_ctx *ctx,
> +                                  struct task_struct *task)
> +{
> +       return host->claimer == ctx ||
> +              (!ctx && task && host->claimer->task == task);
> +}
> +
> +static inline void mmc_ctx_set_claimer(struct mmc_host *host,
> +                                      struct mmc_ctx *ctx,
> +                                      struct task_struct *task)
> +{
> +       if (!host->claimer) {
> +               if (ctx)
> +                       host->claimer = ctx;
> +               else
> +                       host->claimer = &host->default_ctx;
> +       }
> +       if (task)
> +               host->claimer->task = task;
> +}
> +
>  /**
> - *     __mmc_claim_host - exclusively claim a host
> + *     __mmc_ctx_task_claim_host - exclusively claim a host
>   *     @host: mmc host to claim
> + *     @ctx: context that claims the host or NULL in which case the default
> + *     context will be used
> + *     @task: task that claims the host or NULL in which case @ctx must be
> + *     provided
>   *     @abort: whether or not the operation should be aborted
>   *
>   *     Claim a host for a set of operations.  If @abort is non null and
> @@ -842,7 +871,8 @@ unsigned int mmc_align_data_size(struct mmc_card *card, unsigned int sz)
>   *     that non-zero value without acquiring the lock.  Returns zero
>   *     with the lock held otherwise.
>   */
> -int __mmc_claim_host(struct mmc_host *host, atomic_t *abort)
> +static int __mmc_ctx_task_claim_host(struct mmc_host *host, struct mmc_ctx *ctx,
> +                                    struct task_struct *task, atomic_t *abort)
>  {
>         DECLARE_WAITQUEUE(wait, current);
>         unsigned long flags;
> @@ -856,7 +886,7 @@ int __mmc_claim_host(struct mmc_host *host, atomic_t *abort)
>         while (1) {
>                 set_current_state(TASK_UNINTERRUPTIBLE);
>                 stop = abort ? atomic_read(abort) : 0;
> -               if (stop || !host->claimed || host->claimer == current)
> +               if (stop || !host->claimed || mmc_ctx_matches(host, ctx, task))
>                         break;
>                 spin_unlock_irqrestore(&host->lock, flags);
>                 schedule();
> @@ -865,7 +895,7 @@ int __mmc_claim_host(struct mmc_host *host, atomic_t *abort)
>         set_current_state(TASK_RUNNING);
>         if (!stop) {
>                 host->claimed = 1;
> -               host->claimer = current;
> +               mmc_ctx_set_claimer(host, ctx, task);
>                 host->claim_cnt += 1;
>                 if (host->claim_cnt == 1)
>                         pm = true;
> @@ -879,8 +909,19 @@ int __mmc_claim_host(struct mmc_host *host, atomic_t *abort)
>
>         return stop;
>  }
> +
> +int __mmc_claim_host(struct mmc_host *host, atomic_t *abort)
> +{
> +       return __mmc_ctx_task_claim_host(host, NULL, current, abort);
> +}
>  EXPORT_SYMBOL(__mmc_claim_host);

There is currently only two users of __mmc_claim_host().

Instead of adding yet another layer of wrapper functions, let's keep
the existing __mmc_claim_host(), but change its definition to take the
new parameters. Then change the existing two users.

>
> +void mmc_ctx_claim_host(struct mmc_host *host, struct mmc_ctx *ctx)
> +{
> +       __mmc_ctx_task_claim_host(host, ctx, NULL, NULL);
> +}
> +EXPORT_SYMBOL(mmc_ctx_claim_host);

There is no need to make this an exported function, better to keep it
static and only to core.c.

As a matter of fact, I suggest you remove this function altogether,
and just call __mmc_claim_host(), with the new parameters, from
mmc_ctx_get_card().

> +
>  /**
>   *     mmc_release_host - release a host
>   *     @host: mmc host to release
> @@ -888,18 +929,17 @@ int __mmc_claim_host(struct mmc_host *host, atomic_t *abort)
>   *     Release a MMC host, allowing others to claim the host
>   *     for their operations.
>   */
> -void mmc_release_host(struct mmc_host *host)
> +static void __mmc_release_host(struct mmc_host *host)
>  {
>         unsigned long flags;
>
> -       WARN_ON(!host->claimed);
> -
>         spin_lock_irqsave(&host->lock, flags);
>         if (--host->claim_cnt) {
>                 /* Release for nested claim */
>                 spin_unlock_irqrestore(&host->lock, flags);
>         } else {
>                 host->claimed = 0;
> +               host->claimer->task = NULL;
>                 host->claimer = NULL;
>                 spin_unlock_irqrestore(&host->lock, flags);
>                 wake_up(&host->wq);
> @@ -907,8 +947,23 @@ void mmc_release_host(struct mmc_host *host)
>                 pm_runtime_put_autosuspend(mmc_dev(host));
>         }
>  }
> +
> +void mmc_release_host(struct mmc_host *host)
> +{
> +       WARN_ON(!host->claimed);
> +
> +       __mmc_release_host(host);
> +}
>  EXPORT_SYMBOL(mmc_release_host);
>
> +void mmc_ctx_release_host(struct mmc_host *host, struct mmc_ctx *ctx)
> +{
> +       WARN_ON(!host->claimed || host->claimer != ctx);
> +
> +       __mmc_release_host(host);
> +}
> +EXPORT_SYMBOL(mmc_ctx_release_host);

>From the same reasoning as above, I think this should be converted to
a static function.

Or more preferably, remove it altogether, and just call
__mmc_release_host() from mmc_ctx_put_card().

> +
>  /*
>   * This is a helper function, which fetches a runtime pm reference for the
>   * card device and also claims the host.
> @@ -921,6 +976,17 @@ void mmc_get_card(struct mmc_card *card)
>  EXPORT_SYMBOL(mmc_get_card);
>
>  /*
> + * This is a helper function, which fetches a runtime pm reference for the
> + * card device and also claims the host for the mmc context.
> + */
> +void mmc_ctx_get_card(struct mmc_card *card, struct mmc_ctx *ctx)
> +{
> +       pm_runtime_get_sync(&card->dev);
> +       mmc_ctx_claim_host(card->host, ctx);
> +}
> +EXPORT_SYMBOL(mmc_ctx_get_card);

Again, please avoid adding yet another layer of wrapper functions.

Instead change the existing mmc_get_card() to take the parameter and
then change the existing three users of it.

> +
> +/*
>   * This is a helper function, which releases the host and drops the runtime
>   * pm reference for the card device.
>   */
> @@ -933,6 +999,18 @@ void mmc_put_card(struct mmc_card *card)
>  EXPORT_SYMBOL(mmc_put_card);
>
>  /*
> + * This is a helper function, which releases the host and drops the runtime
> + * pm reference for the card device.
> + */
> +void mmc_ctx_put_card(struct mmc_card *card, struct mmc_ctx *ctx)
> +{
> +       mmc_ctx_release_host(card->host, ctx);
> +       pm_runtime_mark_last_busy(&card->dev);
> +       pm_runtime_put_autosuspend(&card->dev);
> +}
> +EXPORT_SYMBOL(mmc_ctx_put_card);

Ditto.

> +
> +/*
>   * Internal function that does the actual ios call to the host driver,
>   * optionally printing some debug output.
>   */
> diff --git a/drivers/mmc/core/core.h b/drivers/mmc/core/core.h
> index ca861091a776..a294c836a143 100644
> --- a/drivers/mmc/core/core.h
> +++ b/drivers/mmc/core/core.h
> @@ -144,4 +144,10 @@ static inline void mmc_claim_host(struct mmc_host *host)
>         __mmc_claim_host(host, NULL);
>  }
>
> +void mmc_ctx_claim_host(struct mmc_host *host, struct mmc_ctx *ctx);
> +void mmc_ctx_release_host(struct mmc_host *host, struct mmc_ctx *ctx);
> +
> +void mmc_ctx_get_card(struct mmc_card *card, struct mmc_ctx *ctx);
> +void mmc_ctx_put_card(struct mmc_card *card, struct mmc_ctx *ctx);
> +
>  #endif
> diff --git a/include/linux/mmc/host.h b/include/linux/mmc/host.h
> index f3f2d07feb2a..228a493c3b25 100644
> --- a/include/linux/mmc/host.h
> +++ b/include/linux/mmc/host.h
> @@ -255,6 +255,10 @@ struct mmc_supply {
>         struct regulator *vqmmc;        /* Optional Vccq supply */
>  };
>
> +struct mmc_ctx {
> +       struct task_struct *task;
> +};
> +
>  struct mmc_host {
>         struct device           *parent;
>         struct device           class_dev;
> @@ -388,8 +392,9 @@ struct mmc_host {
>         struct mmc_card         *card;          /* device attached to this host */
>
>         wait_queue_head_t       wq;
> -       struct task_struct      *claimer;       /* task that has host claimed */
> +       struct mmc_ctx          *claimer;       /* context that has host claimed */
>         int                     claim_cnt;      /* "claim" nesting count */
> +       struct mmc_ctx          default_ctx;    /* default context */
>
>         struct delayed_work     detect;
>         int                     detect_change;  /* card detect flag */
> --
> 1.9.1
>

Kind regards
Uffe

Re: [PATCH V8 00/14] mmc: Add Command Queue support

[Ulf Hansson]

On 13 September 2017 at 13:40, Adrian Hunter <adrian.hunter@intel.com> wrote:
> Hi
>
> Here is V8 of the hardware command queue patches without the software
> command queue patches, now using blk-mq and now with blk-mq support for
> non-CQE I/O.
>
> After the unacceptable debacle of the last release cycle, I expect an
> immediate response to these patches.
>
> HW CMDQ offers 25% - 50% better random multi-threaded I/O.  I see a slight
> 2% drop in sequential read speed but no change to sequential write.
>
> Non-CQE blk-mq showed a 3% decrease in sequential read performance.  This
> seemed to be coming from the inferior latency of running work items compared
> with a dedicated thread.  Hacking blk-mq workqueue to be unbound reduced the
> performance degradation from 3% to 1%.
>
> While we should look at changing blk-mq to give better workqueue performance,
> a bigger gain is likely to be made by adding a new host API to enable the
> next already-prepared request to be issued directly from within ->done()
> callback of the current request.

Adrian, I am reviewing this series, however let me comment on each
change individually.

I have also run some test on my ux500 board and enabling the blkmq
path via the new MMC Kconfig option. My idea was to run some iozone
comparisons between the legacy path and the new blkmq path, but I just
couldn't get to that point because of the following errors.

I am using a Kingston 4GB SDHC card, which is detected and mounted
nicely. However, when I decide to do some writes to the card I get the
following errors.

root@ME:/mnt/sdcard dd if=/dev/zero of=testfile bs=8192 count=5000 conv=fsync
[  463.714294] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
[  464.722656] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
[  466.081481] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
[  467.111236] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
[  468.669647] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
[  469.685699] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
[  471.043334] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
[  472.052337] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
[  473.342651] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
[  474.323760] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
[  475.544769] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
[  476.539031] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
[  477.748474] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
[  478.724182] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!

I haven't yet got the point of investigating this any further, and
unfortunate I have a busy schedule with traveling next week. I will do
my best to look into this as soon as I can.

Perhaps you have some ideas?

Kind regards
Uffe

>
>
> Changes since V7:
>         Re-based
>       mmc: core: Introduce host claiming by context
>         Slightly simplified
>       mmc: core: Add parameter use_blk_mq
>         New patch.
>       mmc: core: Remove unnecessary host claim
>         New patch.
>       mmc: core: Export mmc_start_bkops()
>         New patch.
>       mmc: core: Export mmc_start_request()
>         New patch.
>       mmc: block: Add CQE and blk-mq support
>         Add blk-mq support for non_CQE requests
>
> Changes since V6:
>       mmc: core: Introduce host claiming by context
>         New patch.
>       mmc: core: Move mmc_start_areq() declaration
>         Dropped because it has been applied
>       mmc: block: Fix block status codes
>         Dropped because it has been applied
>       mmc: host: Add CQE interface
>         Dropped because it has been applied
>       mmc: core: Turn off CQE before sending commands
>         Dropped because it has been applied
>       mmc: block: Factor out mmc_setup_queue()
>         New patch.
>       mmc: block: Add CQE support
>         Drop legacy support and add blk-mq support
>
> Changes since V5:
>         Re-based
>       mmc: core: Add mmc_retune_hold_now()
>         Dropped because it has been applied
>       mmc: core: Add members to mmc_request and mmc_data for CQE's
>         Dropped because it has been applied
>       mmc: core: Move mmc_start_areq() declaration
>         New patch at Ulf's request
>       mmc: block: Fix block status codes
>         Another un-related patch
>       mmc: host: Add CQE interface
>         Move recovery_notifier() callback to struct mmc_request
>       mmc: core: Add support for handling CQE requests
>         Roll __mmc_cqe_request_done() into mmc_cqe_request_done()
>         Move function declarations requested by Ulf
>       mmc: core: Remove unused MMC_CAP2_PACKED_CMD
>         Dropped because it has been applied
>       mmc: block: Add CQE support
>         Add explanation to commit message
>         Adjustment for changed recovery_notifier() callback
>       mmc: cqhci: support for command queue enabled host
>         Adjustment for changed recovery_notifier() callback
>       mmc: sdhci-pci: Add CQHCI support for Intel GLK
>         Add DCMD capability for Intel controllers except GLK
>
> Changes since V4:
>       mmc: core: Add mmc_retune_hold_now()
>         Add explanation to commit message.
>       mmc: host: Add CQE interface
>         Add comments to callback declarations.
>       mmc: core: Turn off CQE before sending commands
>         Add explanation to commit message.
>       mmc: core: Add support for handling CQE requests
>         Add comments as requested by Ulf.
>       mmc: core: Remove unused MMC_CAP2_PACKED_CMD
>         New patch.
>       mmc: mmc: Enable Command Queuing
>         Adjust for removal of MMC_CAP2_PACKED_CMD.
>         Add a comment about Packed Commands.
>       mmc: mmc: Enable CQE's
>         Remove un-necessary check for MMC_CAP2_CQE
>       mmc: block: Use local variables in mmc_blk_data_prep()
>         New patch.
>       mmc: block: Prepare CQE data
>         Adjust due to "mmc: block: Use local variables in mmc_blk_data_prep()"
>         Remove priority setting.
>         Add explanation to commit message.
>       mmc: cqhci: support for command queue enabled host
>         Fix transfer descriptor setting in cqhci_set_tran_desc() for 32-bit DMA
>
> Changes since V3:
>         Adjusted ...blk_end_request...() for new block status codes
>         Fixed CQHCI transaction descriptor for "no DCMD" case
>
> Changes since V2:
>         Dropped patches that have been applied.
>         Re-based
>         Added "mmc: sdhci-pci: Add CQHCI support for Intel GLK"
>
> Changes since V1:
>
>         "Share mmc request array between partitions" is dependent
>         on changes in "Introduce queue semantics", so added that
>         and block fixes:
>
>         Added "Fix is_waiting_last_req set incorrectly"
>         Added "Fix cmd error reset failure path"
>         Added "Use local var for mqrq_cur"
>         Added "Introduce queue semantics"
>
> Changes since RFC:
>
>         Re-based on next.
>         Added comment about command queue priority.
>         Added some acks and reviews.
>
>
> Adrian Hunter (13):
>       mmc: core: Introduce host claiming by context
>       mmc: core: Add support for handling CQE requests
>       mmc: mmc: Enable Command Queuing
>       mmc: mmc: Enable CQE's
>       mmc: block: Use local variables in mmc_blk_data_prep()
>       mmc: block: Prepare CQE data
>       mmc: block: Factor out mmc_setup_queue()
>       mmc: core: Add parameter use_blk_mq
>       mmc: core: Remove unnecessary host claim
>       mmc: core: Export mmc_start_bkops()
>       mmc: core: Export mmc_start_request()
>       mmc: block: Add CQE and blk-mq support
>       mmc: sdhci-pci: Add CQHCI support for Intel GLK
>
> Venkat Gopalakrishnan (1):
>       mmc: cqhci: support for command queue enabled host
>
>  drivers/mmc/Kconfig               |   11 +
>  drivers/mmc/core/block.c          |  777 ++++++++++++++++++++++++-
>  drivers/mmc/core/block.h          |    8 +
>  drivers/mmc/core/bus.c            |    7 +
>  drivers/mmc/core/core.c           |  265 ++++++++-
>  drivers/mmc/core/core.h           |   14 +
>  drivers/mmc/core/host.c           |    2 +
>  drivers/mmc/core/host.h           |    4 +
>  drivers/mmc/core/mmc.c            |   29 +
>  drivers/mmc/core/mmc_ops.c        |    6 +-
>  drivers/mmc/core/queue.c          |  479 +++++++++++++--
>  drivers/mmc/core/queue.h          |   54 +-
>  drivers/mmc/host/Kconfig          |   14 +
>  drivers/mmc/host/Makefile         |    1 +
>  drivers/mmc/host/cqhci.c          | 1154 +++++++++++++++++++++++++++++++++++++
>  drivers/mmc/host/cqhci.h          |  240 ++++++++
>  drivers/mmc/host/sdhci-pci-core.c |  154 ++++-
>  include/linux/mmc/host.h          |   10 +-
>  18 files changed, 3146 insertions(+), 83 deletions(-)
>  create mode 100644 drivers/mmc/host/cqhci.c
>  create mode 100644 drivers/mmc/host/cqhci.h
>
>
> Regards
> Adrian

Re: [PATCH V8 00/14] mmc: Add Command Queue support

[Adrian Hunter]

On 21/09/17 12:01, Ulf Hansson wrote:
> On 13 September 2017 at 13:40, Adrian Hunter <adrian.hunter@intel.com> wrote:
>> Hi
>>
>> Here is V8 of the hardware command queue patches without the software
>> command queue patches, now using blk-mq and now with blk-mq support for
>> non-CQE I/O.
>>
>> After the unacceptable debacle of the last release cycle, I expect an
>> immediate response to these patches.
>>
>> HW CMDQ offers 25% - 50% better random multi-threaded I/O.  I see a slight
>> 2% drop in sequential read speed but no change to sequential write.
>>
>> Non-CQE blk-mq showed a 3% decrease in sequential read performance.  This
>> seemed to be coming from the inferior latency of running work items compared
>> with a dedicated thread.  Hacking blk-mq workqueue to be unbound reduced the
>> performance degradation from 3% to 1%.
>>
>> While we should look at changing blk-mq to give better workqueue performance,
>> a bigger gain is likely to be made by adding a new host API to enable the
>> next already-prepared request to be issued directly from within ->done()
>> callback of the current request.
> 
> Adrian, I am reviewing this series, however let me comment on each
> change individually.
> 
> I have also run some test on my ux500 board and enabling the blkmq
> path via the new MMC Kconfig option. My idea was to run some iozone
> comparisons between the legacy path and the new blkmq path, but I just
> couldn't get to that point because of the following errors.
> 
> I am using a Kingston 4GB SDHC card, which is detected and mounted
> nicely. However, when I decide to do some writes to the card I get the
> following errors.
> 
> root@ME:/mnt/sdcard dd if=/dev/zero of=testfile bs=8192 count=5000 conv=fsync
> [  463.714294] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
> [  464.722656] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
> [  466.081481] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
> [  467.111236] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
> [  468.669647] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
> [  469.685699] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
> [  471.043334] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
> [  472.052337] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
> [  473.342651] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
> [  474.323760] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
> [  475.544769] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
> [  476.539031] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
> [  477.748474] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
> [  478.724182] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
> 
> I haven't yet got the point of investigating this any further, and
> unfortunate I have a busy schedule with traveling next week. I will do
> my best to look into this as soon as I can.
> 
> Perhaps you have some ideas?

The behaviour depends on whether you have MMC_CAP_WAIT_WHILE_BUSY. Try
changing that and see if it makes a difference.

Re: [PATCH V8 08/14] mmc: core: Add parameter use_blk_mq

[Ulf Hansson]

On 13 September 2017 at 13:40, Adrian Hunter <adrian.hunter@intel.com> wrote:
> Until mmc has blk-mq support fully implemented and tested, add a
> parameter use_blk_mq, default to false unless config option MMC_MQ_DEFAULT
> is selected.
>
> Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
> ---
>  drivers/mmc/Kconfig      | 11 +++++++++++
>  drivers/mmc/core/core.c  |  7 +++++++
>  drivers/mmc/core/core.h  |  2 ++
>  drivers/mmc/core/host.c  |  2 ++
>  drivers/mmc/core/host.h  |  4 ++++
>  include/linux/mmc/host.h |  1 +
>  6 files changed, 27 insertions(+)
>
> diff --git a/drivers/mmc/Kconfig b/drivers/mmc/Kconfig
> index ec21388311db..98202934bd29 100644
> --- a/drivers/mmc/Kconfig
> +++ b/drivers/mmc/Kconfig
> @@ -12,6 +12,17 @@ menuconfig MMC
>           If you want MMC/SD/SDIO support, you should say Y here and
>           also to your specific host controller driver.
>
> +config MMC_MQ_DEFAULT
> +       bool "MMC: use blk-mq I/O path by default"
> +       depends on MMC && BLOCK
> +       ---help---
> +         This option enables the new blk-mq based I/O path for MMC block
> +         devices by default.  With the option the mmc_core.use_blk_mq
> +         module/boot option defaults to Y, without it to N, but it can
> +         still be overridden either way.
> +
> +         If unsure say N.
> +
>  if MMC

I asume the goal of adding this option is to enable us to move slowly
forward. In general that might be a good idea, however for this
particular case I am not sure.

The main reason is simply that I find it unlikely that people and
distributions will actually go in and change the default value, so in
the end we will just be adding new code, which isn't really going to
be much tested. That's what happened in scsi case.

As I also stated earlier, I do worry about the maintenance of the mmc
block device code, and this approach make it worse, at least short
term.

To me, the scsi case is also different, because the mq support was
added long time ago and at that point one could worry a bit of
maturity of the blkmq in general, that I assume have been sorted out
by know.

>
>  source "drivers/mmc/core/Kconfig"
> diff --git a/drivers/mmc/core/core.c b/drivers/mmc/core/core.c
> index ef2d8aa1e7d2..3638ed4f0f9e 100644
> --- a/drivers/mmc/core/core.c
> +++ b/drivers/mmc/core/core.c
> @@ -66,6 +66,13 @@
>  bool use_spi_crc = 1;
>  module_param(use_spi_crc, bool, 0);
>
> +#ifdef CONFIG_MMC_MQ_DEFAULT
> +bool mmc_use_blk_mq = true;
> +#else
> +bool mmc_use_blk_mq = false;
> +#endif
> +module_param_named(use_blk_mq, mmc_use_blk_mq, bool, S_IWUSR | S_IRUGO);
> +
>  static int mmc_schedule_delayed_work(struct delayed_work *work,
>                                      unsigned long delay)
>  {
> diff --git a/drivers/mmc/core/core.h b/drivers/mmc/core/core.h
> index e941342ed450..535539a9e7eb 100644
> --- a/drivers/mmc/core/core.h
> +++ b/drivers/mmc/core/core.h
> @@ -35,6 +35,8 @@ struct mmc_bus_ops {
>         int (*reset)(struct mmc_host *);
>  };
>
> +extern bool mmc_use_blk_mq;
> +
>  void mmc_attach_bus(struct mmc_host *host, const struct mmc_bus_ops *ops);
>  void mmc_detach_bus(struct mmc_host *host);
>
> diff --git a/drivers/mmc/core/host.c b/drivers/mmc/core/host.c
> index ad88deb2e8f3..b624dbb6cd15 100644
> --- a/drivers/mmc/core/host.c
> +++ b/drivers/mmc/core/host.c
> @@ -398,6 +398,8 @@ struct mmc_host *mmc_alloc_host(int extra, struct device *dev)
>         host->max_blk_size = 512;
>         host->max_blk_count = PAGE_SIZE / 512;
>
> +       host->use_blk_mq = mmc_use_blk_mq;
> +
>         return host;
>  }
>
> diff --git a/drivers/mmc/core/host.h b/drivers/mmc/core/host.h
> index 77d6f60d1bf9..170fe5947087 100644
> --- a/drivers/mmc/core/host.h
> +++ b/drivers/mmc/core/host.h
> @@ -69,6 +69,10 @@ static inline bool mmc_card_hs400es(struct mmc_card *card)
>         return card->host->ios.enhanced_strobe;
>  }
>
> +static inline bool mmc_host_use_blk_mq(struct mmc_host *host)
> +{
> +       return host->use_blk_mq;
> +}
>
>  #endif
>
> diff --git a/include/linux/mmc/host.h b/include/linux/mmc/host.h
> index 54b0463443bd..5d1bd10991f7 100644
> --- a/include/linux/mmc/host.h
> +++ b/include/linux/mmc/host.h
> @@ -378,6 +378,7 @@ struct mmc_host {
>         unsigned int            doing_retune:1; /* re-tuning in progress */
>         unsigned int            retune_now:1;   /* do re-tuning at next req */
>         unsigned int            retune_paused:1; /* re-tuning is temporarily disabled */
> +       unsigned int            use_blk_mq:1;   /* use blk-mq */
>
>         int                     rescan_disable; /* disable card detection */
>         int                     rescan_entered; /* used with nonremovable devices */
> --
> 1.9.1
>

Kind regards
Uffe

Re: [PATCH V8 12/14] mmc: block: Add CQE and blk-mq support

[Ulf Hansson]

On 13 September 2017 at 13:40, Adrian Hunter <adrian.hunter@intel.com> wrote:
> Add CQE support to the block driver, including:
>     - optionally using DCMD for flush requests
>     - "manually" issuing discard requests
>     - issuing read / write requests to the CQE
>     - supporting block-layer timeouts
>     - handling recovery
>     - supporting re-tuning
>
> CQE offers 25% - 50% better random multi-threaded I/O.  There is a slight
> (e.g. 2%) drop in sequential read speed but no observable change to sequential
> write.
>
> CQE automatically sends the commands to complete requests.  However it only
> supports reads / writes and so-called "direct commands" (DCMD).  Furthermore
> DCMD is limited to one command at a time, but discards require 3 commands.
> That makes issuing discards through CQE very awkward, but some CQE's don't
> support DCMD anyway.  So for discards, the existing non-CQE approach is
> taken, where the mmc core code issues the 3 commands one at a time i.e.
> mmc_erase(). Where DCMD is used, is for issuing flushes.
>
> For host controllers without CQE support, blk-mq support is extended to
> synchronous reads/writes or, if the host supports CAP_WAIT_WHILE_BUSY,
> asynchonous reads/writes.  The advantage of asynchronous reads/writes is
> that it allows the preparation of the next request while the current
> request is in progress.
>
> Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
> ---
>  drivers/mmc/core/block.c | 734 ++++++++++++++++++++++++++++++++++++++++++++++-
>  drivers/mmc/core/block.h |   8 +
>  drivers/mmc/core/queue.c | 428 +++++++++++++++++++++++++--
>  drivers/mmc/core/queue.h |  54 +++-
>  4 files changed, 1192 insertions(+), 32 deletions(-)

Adrian, this is just too hard for me to review. Please, can you split this up?

At least enabling mq support should be done in a separate and first
step, then you could add the CQE bits on top.

I think that is important, because we want to make sure the mq
deployment is done correctly first. Otherwise it becomes impossible to
narrow down problems, because those could then be either CQE related
or mq related.

[....]

Kind regards
Uffe

Re: [PATCH V8 12/14] mmc: block: Add CQE and blk-mq support

[Adrian Hunter]

On 21/09/17 12:59, Ulf Hansson wrote:
> On 13 September 2017 at 13:40, Adrian Hunter <adrian.hunter@intel.com> wrote:
>> Add CQE support to the block driver, including:
>>     - optionally using DCMD for flush requests
>>     - "manually" issuing discard requests
>>     - issuing read / write requests to the CQE
>>     - supporting block-layer timeouts
>>     - handling recovery
>>     - supporting re-tuning
>>
>> CQE offers 25% - 50% better random multi-threaded I/O.  There is a slight
>> (e.g. 2%) drop in sequential read speed but no observable change to sequential
>> write.
>>
>> CQE automatically sends the commands to complete requests.  However it only
>> supports reads / writes and so-called "direct commands" (DCMD).  Furthermore
>> DCMD is limited to one command at a time, but discards require 3 commands.
>> That makes issuing discards through CQE very awkward, but some CQE's don't
>> support DCMD anyway.  So for discards, the existing non-CQE approach is
>> taken, where the mmc core code issues the 3 commands one at a time i.e.
>> mmc_erase(). Where DCMD is used, is for issuing flushes.
>>
>> For host controllers without CQE support, blk-mq support is extended to
>> synchronous reads/writes or, if the host supports CAP_WAIT_WHILE_BUSY,
>> asynchonous reads/writes.  The advantage of asynchronous reads/writes is
>> that it allows the preparation of the next request while the current
>> request is in progress.
>>
>> Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
>> ---
>>  drivers/mmc/core/block.c | 734 ++++++++++++++++++++++++++++++++++++++++++++++-
>>  drivers/mmc/core/block.h |   8 +
>>  drivers/mmc/core/queue.c | 428 +++++++++++++++++++++++++--
>>  drivers/mmc/core/queue.h |  54 +++-
>>  4 files changed, 1192 insertions(+), 32 deletions(-)
> 
> Adrian, this is just too hard for me to review. Please, can you split this up?
> 
> At least enabling mq support should be done in a separate and first
> step, then you could add the CQE bits on top.
> 
> I think that is important, because we want to make sure the mq
> deployment is done correctly first. Otherwise it becomes impossible to
> narrow down problems, because those could then be either CQE related
> or mq related.

I looked at splitting it up, but it makes more sense together.  Things are
the way they are to support all the different issue types.  It reads much
better together.  The CQE functions are named appropriately and practically
all the code is new, so it is not as though there are any intermediate steps.

Re: [PATCH V8 00/14] mmc: Add Command Queue support

[Christoph Hellwig]

On Wed, Sep 13, 2017 at 02:40:00PM +0300, Adrian Hunter wrote:
> Non-CQE blk-mq showed a 3% decrease in sequential read performance.  This
> seemed to be coming from the inferior latency of running work items compared
> with a dedicated thread.  Hacking blk-mq workqueue to be unbound reduced the
> performance degradation from 3% to 1%.

Can you start a discussion on just this on the linux-block lists?

Re: [PATCH V8 01/14] mmc: core: Introduce host claiming by context

[Linus Walleij]

On Wed, Sep 13, 2017 at 1:40 PM, Adrian Hunter <adrian.hunter@intel.com> wrote:

> Currently the host can be claimed by a task.  Change this so that the host
> can be claimed by a context that may or may not be a task.  This provides
> for the host to be claimed by a block driver queue to support blk-mq, while
> maintaining compatibility with the existing use of mmc_claim_host().
>
> Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>

I think this is a reasonable intermediate step as well.
I am working to kill off the "big MMC host lock" by the following
plan:

DONE:
- Convert all ioctl() ops to custom block layer requests (done)
- Convert all debugfs ops to custom block layer requests (done)
- Convert RPMB access to a character device to avoid one
  partition switch (just merged)

REMAINS:
- Convert boot and "general" partitions to be part of the main
  block device. Because they are. (This is really tricky and
  may require changes in the core kernel partition handling.)
- Delete the dual mode card support. (No-one seems to be
  using it.)
- Split the world in block access and SDIO with something
  like a simple mutex: either the host is a block device OR
  it is SDIO, and that state has the same lifetime as a card,
  since they are now either block devices or SDIO not both
  at the same time.
- Implement a lighter lock for SDIO host claiming.
- Now the lock should be gone...

As you can see it is not entirely trivial :/

Linus

Re: [PATCH V8 08/14] mmc: core: Add parameter use_blk_mq

[Adrian Hunter]

On 21/09/17 12:47, Ulf Hansson wrote:
> On 13 September 2017 at 13:40, Adrian Hunter <adrian.hunter@intel.com> wrote:
>> Until mmc has blk-mq support fully implemented and tested, add a
>> parameter use_blk_mq, default to false unless config option MMC_MQ_DEFAULT
>> is selected.
>>
>> Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
>> ---
>>  drivers/mmc/Kconfig      | 11 +++++++++++
>>  drivers/mmc/core/core.c  |  7 +++++++
>>  drivers/mmc/core/core.h  |  2 ++
>>  drivers/mmc/core/host.c  |  2 ++
>>  drivers/mmc/core/host.h  |  4 ++++
>>  include/linux/mmc/host.h |  1 +
>>  6 files changed, 27 insertions(+)
>>
>> diff --git a/drivers/mmc/Kconfig b/drivers/mmc/Kconfig
>> index ec21388311db..98202934bd29 100644
>> --- a/drivers/mmc/Kconfig
>> +++ b/drivers/mmc/Kconfig
>> @@ -12,6 +12,17 @@ menuconfig MMC
>>           If you want MMC/SD/SDIO support, you should say Y here and
>>           also to your specific host controller driver.
>>
>> +config MMC_MQ_DEFAULT
>> +       bool "MMC: use blk-mq I/O path by default"
>> +       depends on MMC && BLOCK
>> +       ---help---
>> +         This option enables the new blk-mq based I/O path for MMC block
>> +         devices by default.  With the option the mmc_core.use_blk_mq
>> +         module/boot option defaults to Y, without it to N, but it can
>> +         still be overridden either way.
>> +
>> +         If unsure say N.
>> +
>>  if MMC
> 
> I asume the goal of adding this option is to enable us to move slowly
> forward. In general that might be a good idea, however for this
> particular case I am not sure.
> 
> The main reason is simply that I find it unlikely that people and
> distributions will actually go in and change the default value, so in
> the end we will just be adding new code, which isn't really going to
> be much tested. That's what happened in scsi case.

The argument that no one is going to test anyway so we shouldn't give them
the opportunity, is not a sustainable vision for the future.  Instead we
should reach out to relevant stakeholders and get them to do their testing
with blk-mq first.

> 
> As I also stated earlier, I do worry about the maintenance of the mmc
> block device code, and this approach make it worse, at least short
> term.

What maintenance?

> 
> To me, the scsi case is also different, because the mq support was
> added long time ago and at that point one could worry a bit of
> maturity of the blkmq in general, that I assume have been sorted out
> by know.

But blk-mq still isn't fully ready for all of scsi, so you seem to be
contradicting yourself.

> 
>>
>>  source "drivers/mmc/core/Kconfig"
>> diff --git a/drivers/mmc/core/core.c b/drivers/mmc/core/core.c
>> index ef2d8aa1e7d2..3638ed4f0f9e 100644
>> --- a/drivers/mmc/core/core.c
>> +++ b/drivers/mmc/core/core.c
>> @@ -66,6 +66,13 @@
>>  bool use_spi_crc = 1;
>>  module_param(use_spi_crc, bool, 0);
>>
>> +#ifdef CONFIG_MMC_MQ_DEFAULT
>> +bool mmc_use_blk_mq = true;
>> +#else
>> +bool mmc_use_blk_mq = false;
>> +#endif
>> +module_param_named(use_blk_mq, mmc_use_blk_mq, bool, S_IWUSR | S_IRUGO);
>> +
>>  static int mmc_schedule_delayed_work(struct delayed_work *work,
>>                                      unsigned long delay)
>>  {
>> diff --git a/drivers/mmc/core/core.h b/drivers/mmc/core/core.h
>> index e941342ed450..535539a9e7eb 100644
>> --- a/drivers/mmc/core/core.h
>> +++ b/drivers/mmc/core/core.h
>> @@ -35,6 +35,8 @@ struct mmc_bus_ops {
>>         int (*reset)(struct mmc_host *);
>>  };
>>
>> +extern bool mmc_use_blk_mq;
>> +
>>  void mmc_attach_bus(struct mmc_host *host, const struct mmc_bus_ops *ops);
>>  void mmc_detach_bus(struct mmc_host *host);
>>
>> diff --git a/drivers/mmc/core/host.c b/drivers/mmc/core/host.c
>> index ad88deb2e8f3..b624dbb6cd15 100644
>> --- a/drivers/mmc/core/host.c
>> +++ b/drivers/mmc/core/host.c
>> @@ -398,6 +398,8 @@ struct mmc_host *mmc_alloc_host(int extra, struct device *dev)
>>         host->max_blk_size = 512;
>>         host->max_blk_count = PAGE_SIZE / 512;
>>
>> +       host->use_blk_mq = mmc_use_blk_mq;
>> +
>>         return host;
>>  }
>>
>> diff --git a/drivers/mmc/core/host.h b/drivers/mmc/core/host.h
>> index 77d6f60d1bf9..170fe5947087 100644
>> --- a/drivers/mmc/core/host.h
>> +++ b/drivers/mmc/core/host.h
>> @@ -69,6 +69,10 @@ static inline bool mmc_card_hs400es(struct mmc_card *card)
>>         return card->host->ios.enhanced_strobe;
>>  }
>>
>> +static inline bool mmc_host_use_blk_mq(struct mmc_host *host)
>> +{
>> +       return host->use_blk_mq;
>> +}
>>
>>  #endif
>>
>> diff --git a/include/linux/mmc/host.h b/include/linux/mmc/host.h
>> index 54b0463443bd..5d1bd10991f7 100644
>> --- a/include/linux/mmc/host.h
>> +++ b/include/linux/mmc/host.h
>> @@ -378,6 +378,7 @@ struct mmc_host {
>>         unsigned int            doing_retune:1; /* re-tuning in progress */
>>         unsigned int            retune_now:1;   /* do re-tuning at next req */
>>         unsigned int            retune_paused:1; /* re-tuning is temporarily disabled */
>> +       unsigned int            use_blk_mq:1;   /* use blk-mq */
>>
>>         int                     rescan_disable; /* disable card detection */
>>         int                     rescan_entered; /* used with nonremovable devices */
>> --
>> 1.9.1
>>
> 
> Kind regards
> Uffe
> 

Re: [PATCH V8 08/14] mmc: core: Add parameter use_blk_mq

[Linus Walleij]

On Fri, Sep 22, 2017 at 3:30 PM, Adrian Hunter <adrian.hunter@intel.com> wrote:
> On 21/09/17 12:47, Ulf Hansson wrote:

>> I asume the goal of adding this option is to enable us to move slowly
>> forward. In general that might be a good idea, however for this
>> particular case I am not sure.
>>
>> The main reason is simply that I find it unlikely that people and
>> distributions will actually go in and change the default value, so in
>> the end we will just be adding new code, which isn't really going to
>> be much tested. That's what happened in scsi case.
>
> The argument that no one is going to test anyway so we shouldn't give them
> the opportunity, is not a sustainable vision for the future.  Instead we
> should reach out to relevant stakeholders and get them to do their testing
> with blk-mq first.

We *could* simply invert the option then. Default it to "y" and
only leave it as a debugging aid so that people can set it to
"n" if they want to test with MQ disabled.

This is also simple to revert by a oneliner just removing "default y"
if there are problems with it.

Yours,
Linus Walleij

Re: [PATCH V8 00/14] mmc: Add Command Queue support

[Ulf Hansson]

On 21 September 2017 at 11:44, Adrian Hunter <adrian.hunter@intel.com> wrote:
> On 21/09/17 12:01, Ulf Hansson wrote:
>> On 13 September 2017 at 13:40, Adrian Hunter <adrian.hunter@intel.com> wrote:
>>> Hi
>>>
>>> Here is V8 of the hardware command queue patches without the software
>>> command queue patches, now using blk-mq and now with blk-mq support for
>>> non-CQE I/O.
>>>
>>> After the unacceptable debacle of the last release cycle, I expect an
>>> immediate response to these patches.
>>>
>>> HW CMDQ offers 25% - 50% better random multi-threaded I/O.  I see a slight
>>> 2% drop in sequential read speed but no change to sequential write.
>>>
>>> Non-CQE blk-mq showed a 3% decrease in sequential read performance.  This
>>> seemed to be coming from the inferior latency of running work items compared
>>> with a dedicated thread.  Hacking blk-mq workqueue to be unbound reduced the
>>> performance degradation from 3% to 1%.
>>>
>>> While we should look at changing blk-mq to give better workqueue performance,
>>> a bigger gain is likely to be made by adding a new host API to enable the
>>> next already-prepared request to be issued directly from within ->done()
>>> callback of the current request.
>>
>> Adrian, I am reviewing this series, however let me comment on each
>> change individually.
>>
>> I have also run some test on my ux500 board and enabling the blkmq
>> path via the new MMC Kconfig option. My idea was to run some iozone
>> comparisons between the legacy path and the new blkmq path, but I just
>> couldn't get to that point because of the following errors.
>>
>> I am using a Kingston 4GB SDHC card, which is detected and mounted
>> nicely. However, when I decide to do some writes to the card I get the
>> following errors.
>>
>> root@ME:/mnt/sdcard dd if=/dev/zero of=testfile bs=8192 count=5000 conv=fsync
>> [  463.714294] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>> [  464.722656] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>> [  466.081481] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>> [  467.111236] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>> [  468.669647] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>> [  469.685699] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>> [  471.043334] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>> [  472.052337] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>> [  473.342651] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>> [  474.323760] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>> [  475.544769] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>> [  476.539031] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>> [  477.748474] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>> [  478.724182] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>
>> I haven't yet got the point of investigating this any further, and
>> unfortunate I have a busy schedule with traveling next week. I will do
>> my best to look into this as soon as I can.
>>
>> Perhaps you have some ideas?
>
> The behaviour depends on whether you have MMC_CAP_WAIT_WHILE_BUSY. Try
> changing that and see if it makes a difference.

Yes, it does! I disabled MMC_CAP_WAIT_WHILE_BUSY (and its
corresponding code in mmci.c) and the errors goes away.

When I use MMC_CAP_WAIT_WHILE_BUSY I get these problems:

[  223.820983] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
[  224.815795] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
[  226.034881] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
[  227.112884] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
[  227.220275] mmc0: Card stuck in wrong state! mmcblk0 mmc_blk_card_stuck
[  228.686798] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
[  229.892150] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
[  231.031890] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
[  232.239013] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
5000+0 records in
5000+0 records out
root@ME:/mnt/sdcard

I looked at the new blkmq code from patch v10 13/15. It seems like the
MMC_CAP_WAIT_WHILE_BUSY is used to determine whether the async request
mechanism should be used or not. Perhaps I didn't looked close enough,
but maybe you could elaborate on why this seems to be the case!?

Kind regards
Uffe

Re: [PATCH V8 00/14] mmc: Add Command Queue support

[Adrian Hunter]

On 10/10/17 15:12, Ulf Hansson wrote:
> On 21 September 2017 at 11:44, Adrian Hunter <adrian.hunter@intel.com> wrote:
>> On 21/09/17 12:01, Ulf Hansson wrote:
>>> On 13 September 2017 at 13:40, Adrian Hunter <adrian.hunter@intel.com> wrote:
>>>> Hi
>>>>
>>>> Here is V8 of the hardware command queue patches without the software
>>>> command queue patches, now using blk-mq and now with blk-mq support for
>>>> non-CQE I/O.
>>>>
>>>> After the unacceptable debacle of the last release cycle, I expect an
>>>> immediate response to these patches.
>>>>
>>>> HW CMDQ offers 25% - 50% better random multi-threaded I/O.  I see a slight
>>>> 2% drop in sequential read speed but no change to sequential write.
>>>>
>>>> Non-CQE blk-mq showed a 3% decrease in sequential read performance.  This
>>>> seemed to be coming from the inferior latency of running work items compared
>>>> with a dedicated thread.  Hacking blk-mq workqueue to be unbound reduced the
>>>> performance degradation from 3% to 1%.
>>>>
>>>> While we should look at changing blk-mq to give better workqueue performance,
>>>> a bigger gain is likely to be made by adding a new host API to enable the
>>>> next already-prepared request to be issued directly from within ->done()
>>>> callback of the current request.
>>>
>>> Adrian, I am reviewing this series, however let me comment on each
>>> change individually.
>>>
>>> I have also run some test on my ux500 board and enabling the blkmq
>>> path via the new MMC Kconfig option. My idea was to run some iozone
>>> comparisons between the legacy path and the new blkmq path, but I just
>>> couldn't get to that point because of the following errors.
>>>
>>> I am using a Kingston 4GB SDHC card, which is detected and mounted
>>> nicely. However, when I decide to do some writes to the card I get the
>>> following errors.
>>>
>>> root@ME:/mnt/sdcard dd if=/dev/zero of=testfile bs=8192 count=5000 conv=fsync
>>> [  463.714294] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>> [  464.722656] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>> [  466.081481] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>> [  467.111236] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>> [  468.669647] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>> [  469.685699] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>> [  471.043334] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>> [  472.052337] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>> [  473.342651] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>> [  474.323760] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>> [  475.544769] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>> [  476.539031] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>> [  477.748474] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>> [  478.724182] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>
>>> I haven't yet got the point of investigating this any further, and
>>> unfortunate I have a busy schedule with traveling next week. I will do
>>> my best to look into this as soon as I can.
>>>
>>> Perhaps you have some ideas?
>>
>> The behaviour depends on whether you have MMC_CAP_WAIT_WHILE_BUSY. Try
>> changing that and see if it makes a difference.
> 
> Yes, it does! I disabled MMC_CAP_WAIT_WHILE_BUSY (and its
> corresponding code in mmci.c) and the errors goes away.
> 
> When I use MMC_CAP_WAIT_WHILE_BUSY I get these problems:
> 
> [  223.820983] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
> [  224.815795] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
> [  226.034881] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
> [  227.112884] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
> [  227.220275] mmc0: Card stuck in wrong state! mmcblk0 mmc_blk_card_stuck
> [  228.686798] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
> [  229.892150] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
> [  231.031890] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
> [  232.239013] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
> 5000+0 records in
> 5000+0 records out
> root@ME:/mnt/sdcard
> 
> I looked at the new blkmq code from patch v10 13/15. It seems like the
> MMC_CAP_WAIT_WHILE_BUSY is used to determine whether the async request
> mechanism should be used or not. Perhaps I didn't looked close enough,
> but maybe you could elaborate on why this seems to be the case!?

MMC_CAP_WAIT_WHILE_BUSY is necessary because it means that a data transfer
request has finished when the host controller calls mmc_request_done(). i.e.
polling the card is not necessary.

Have you tried V9 or V10.  There was a fix in V9 related to calling
->post_req() which could mess up DMA.

The other thing that could go wrong with DMA is if it cannot accept
->post_req() being called from mmc_request_done().

Re: [PATCH V8 00/14] mmc: Add Command Queue support

[Ulf Hansson]

[...]

>>>>
>>>> I have also run some test on my ux500 board and enabling the blkmq
>>>> path via the new MMC Kconfig option. My idea was to run some iozone
>>>> comparisons between the legacy path and the new blkmq path, but I just
>>>> couldn't get to that point because of the following errors.
>>>>
>>>> I am using a Kingston 4GB SDHC card, which is detected and mounted
>>>> nicely. However, when I decide to do some writes to the card I get the
>>>> following errors.
>>>>
>>>> root@ME:/mnt/sdcard dd if=/dev/zero of=testfile bs=8192 count=5000 conv=fsync
>>>> [  463.714294] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>> [  464.722656] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>> [  466.081481] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>> [  467.111236] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>> [  468.669647] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>> [  469.685699] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>> [  471.043334] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>> [  472.052337] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>> [  473.342651] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>> [  474.323760] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>> [  475.544769] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>> [  476.539031] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>> [  477.748474] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>> [  478.724182] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>
>>>> I haven't yet got the point of investigating this any further, and
>>>> unfortunate I have a busy schedule with traveling next week. I will do
>>>> my best to look into this as soon as I can.
>>>>
>>>> Perhaps you have some ideas?
>>>
>>> The behaviour depends on whether you have MMC_CAP_WAIT_WHILE_BUSY. Try
>>> changing that and see if it makes a difference.
>>
>> Yes, it does! I disabled MMC_CAP_WAIT_WHILE_BUSY (and its
>> corresponding code in mmci.c) and the errors goes away.
>>
>> When I use MMC_CAP_WAIT_WHILE_BUSY I get these problems:
>>
>> [  223.820983] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>> [  224.815795] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>> [  226.034881] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>> [  227.112884] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>> [  227.220275] mmc0: Card stuck in wrong state! mmcblk0 mmc_blk_card_stuck
>> [  228.686798] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>> [  229.892150] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>> [  231.031890] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>> [  232.239013] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>> 5000+0 records in
>> 5000+0 records out
>> root@ME:/mnt/sdcard
>>
>> I looked at the new blkmq code from patch v10 13/15. It seems like the
>> MMC_CAP_WAIT_WHILE_BUSY is used to determine whether the async request
>> mechanism should be used or not. Perhaps I didn't looked close enough,
>> but maybe you could elaborate on why this seems to be the case!?
>
> MMC_CAP_WAIT_WHILE_BUSY is necessary because it means that a data transfer
> request has finished when the host controller calls mmc_request_done(). i.e.
> polling the card is not necessary.

Well, that is a rather big change on its own. Earlier we polled with
CMD13 to verify that the card has moved back to the transfer state, in
case it was a write. And that was no matter of MMC_CAP_WAIT_WHILE_BUSY
was set or not. Right!?

I am not sure it's a good idea to bypass that validation, it seems
fragile to rely only on the busy detection on DAT line for writes.

>
> Have you tried V9 or V10.  There was a fix in V9 related to calling
> ->post_req() which could mess up DMA.

I have used V10.

>
> The other thing that could go wrong with DMA is if it cannot accept
> ->post_req() being called from mmc_request_done().

I don't think mmci has a problem with that, however why do you want to
do this? Wouldn't that defeat some of the benefits with the async
request mechanism?

Kind regards
Uffe

Re: [PATCH V8 00/14] mmc: Add Command Queue support

[Adrian Hunter]

On 10/10/17 16:08, Ulf Hansson wrote:
> [...]
> 
>>>>>
>>>>> I have also run some test on my ux500 board and enabling the blkmq
>>>>> path via the new MMC Kconfig option. My idea was to run some iozone
>>>>> comparisons between the legacy path and the new blkmq path, but I just
>>>>> couldn't get to that point because of the following errors.
>>>>>
>>>>> I am using a Kingston 4GB SDHC card, which is detected and mounted
>>>>> nicely. However, when I decide to do some writes to the card I get the
>>>>> following errors.
>>>>>
>>>>> root@ME:/mnt/sdcard dd if=/dev/zero of=testfile bs=8192 count=5000 conv=fsync
>>>>> [  463.714294] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>> [  464.722656] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>> [  466.081481] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>> [  467.111236] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>> [  468.669647] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>> [  469.685699] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>> [  471.043334] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>> [  472.052337] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>> [  473.342651] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>> [  474.323760] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>> [  475.544769] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>> [  476.539031] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>> [  477.748474] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>> [  478.724182] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>
>>>>> I haven't yet got the point of investigating this any further, and
>>>>> unfortunate I have a busy schedule with traveling next week. I will do
>>>>> my best to look into this as soon as I can.
>>>>>
>>>>> Perhaps you have some ideas?
>>>>
>>>> The behaviour depends on whether you have MMC_CAP_WAIT_WHILE_BUSY. Try
>>>> changing that and see if it makes a difference.
>>>
>>> Yes, it does! I disabled MMC_CAP_WAIT_WHILE_BUSY (and its
>>> corresponding code in mmci.c) and the errors goes away.
>>>
>>> When I use MMC_CAP_WAIT_WHILE_BUSY I get these problems:
>>>
>>> [  223.820983] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>> [  224.815795] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>> [  226.034881] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>> [  227.112884] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>> [  227.220275] mmc0: Card stuck in wrong state! mmcblk0 mmc_blk_card_stuck
>>> [  228.686798] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>> [  229.892150] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>> [  231.031890] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>> [  232.239013] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>> 5000+0 records in
>>> 5000+0 records out
>>> root@ME:/mnt/sdcard
>>>
>>> I looked at the new blkmq code from patch v10 13/15. It seems like the
>>> MMC_CAP_WAIT_WHILE_BUSY is used to determine whether the async request
>>> mechanism should be used or not. Perhaps I didn't looked close enough,
>>> but maybe you could elaborate on why this seems to be the case!?
>>
>> MMC_CAP_WAIT_WHILE_BUSY is necessary because it means that a data transfer
>> request has finished when the host controller calls mmc_request_done(). i.e.
>> polling the card is not necessary.
> 
> Well, that is a rather big change on its own. Earlier we polled with
> CMD13 to verify that the card has moved back to the transfer state, in
> case it was a write. And that was no matter of MMC_CAP_WAIT_WHILE_BUSY
> was set or not. Right!?

Yes

> 
> I am not sure it's a good idea to bypass that validation, it seems
> fragile to rely only on the busy detection on DAT line for writes.

Can you cite something from the specifications that backs that up, because I
couldn't find anything to suggest that CMD13 polling was expected.

> 
>>
>> Have you tried V9 or V10.  There was a fix in V9 related to calling
>> ->post_req() which could mess up DMA.
> 
> I have used V10.
> 
>>
>> The other thing that could go wrong with DMA is if it cannot accept
>> ->post_req() being called from mmc_request_done().
> 
> I don't think mmci has a problem with that, however why do you want to
> do this? Wouldn't that defeat some of the benefits with the async
> request mechanism?

Perhaps - but it would need to be tested.  If there are more requests
waiting, one optimization could be to defer ->post_req() until after the
next request is started.

Re: [PATCH V8 00/14] mmc: Add Command Queue support

[Ulf Hansson]

On 10 October 2017 at 15:31, Adrian Hunter <adrian.hunter@intel.com> wrote:
> On 10/10/17 16:08, Ulf Hansson wrote:
>> [...]
>>
>>>>>>
>>>>>> I have also run some test on my ux500 board and enabling the blkmq
>>>>>> path via the new MMC Kconfig option. My idea was to run some iozone
>>>>>> comparisons between the legacy path and the new blkmq path, but I just
>>>>>> couldn't get to that point because of the following errors.
>>>>>>
>>>>>> I am using a Kingston 4GB SDHC card, which is detected and mounted
>>>>>> nicely. However, when I decide to do some writes to the card I get the
>>>>>> following errors.
>>>>>>
>>>>>> root@ME:/mnt/sdcard dd if=/dev/zero of=testfile bs=8192 count=5000 conv=fsync
>>>>>> [  463.714294] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>> [  464.722656] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>> [  466.081481] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>> [  467.111236] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>> [  468.669647] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>> [  469.685699] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>> [  471.043334] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>> [  472.052337] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>> [  473.342651] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>> [  474.323760] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>> [  475.544769] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>> [  476.539031] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>> [  477.748474] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>> [  478.724182] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>
>>>>>> I haven't yet got the point of investigating this any further, and
>>>>>> unfortunate I have a busy schedule with traveling next week. I will do
>>>>>> my best to look into this as soon as I can.
>>>>>>
>>>>>> Perhaps you have some ideas?
>>>>>
>>>>> The behaviour depends on whether you have MMC_CAP_WAIT_WHILE_BUSY. Try
>>>>> changing that and see if it makes a difference.
>>>>
>>>> Yes, it does! I disabled MMC_CAP_WAIT_WHILE_BUSY (and its
>>>> corresponding code in mmci.c) and the errors goes away.
>>>>
>>>> When I use MMC_CAP_WAIT_WHILE_BUSY I get these problems:
>>>>
>>>> [  223.820983] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>> [  224.815795] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>> [  226.034881] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>> [  227.112884] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>> [  227.220275] mmc0: Card stuck in wrong state! mmcblk0 mmc_blk_card_stuck
>>>> [  228.686798] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>> [  229.892150] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>> [  231.031890] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>> [  232.239013] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>> 5000+0 records in
>>>> 5000+0 records out
>>>> root@ME:/mnt/sdcard
>>>>
>>>> I looked at the new blkmq code from patch v10 13/15. It seems like the
>>>> MMC_CAP_WAIT_WHILE_BUSY is used to determine whether the async request
>>>> mechanism should be used or not. Perhaps I didn't looked close enough,
>>>> but maybe you could elaborate on why this seems to be the case!?
>>>
>>> MMC_CAP_WAIT_WHILE_BUSY is necessary because it means that a data transfer
>>> request has finished when the host controller calls mmc_request_done(). i.e.
>>> polling the card is not necessary.
>>
>> Well, that is a rather big change on its own. Earlier we polled with
>> CMD13 to verify that the card has moved back to the transfer state, in
>> case it was a write. And that was no matter of MMC_CAP_WAIT_WHILE_BUSY
>> was set or not. Right!?
>
> Yes
>
>>
>> I am not sure it's a good idea to bypass that validation, it seems
>> fragile to rely only on the busy detection on DAT line for writes.
>
> Can you cite something from the specifications that backs that up, because I
> couldn't find anything to suggest that CMD13 polling was expected.

No I can't, but I don't see why that matters.

My point is, if we want to go down that road by avoiding the CMD13
polling, that needs to be a separate change, which we can test and
confirm on its own.

>
>>
>>>
>>> Have you tried V9 or V10.  There was a fix in V9 related to calling
>>> ->post_req() which could mess up DMA.
>>
>> I have used V10.
>>
>>>
>>> The other thing that could go wrong with DMA is if it cannot accept
>>> ->post_req() being called from mmc_request_done().
>>
>> I don't think mmci has a problem with that, however why do you want to
>> do this? Wouldn't that defeat some of the benefits with the async
>> request mechanism?
>
> Perhaps - but it would need to be tested.  If there are more requests
> waiting, one optimization could be to defer ->post_req() until after the
> next request is started.

This is already proven, because this how the existing mmc async
request mechanism works.

In ->post_req() callbacks, host drivers may do dma_unmap_sg(), which
is something that could be costly and therefore it's better to start a
new request before, such these things can go on in parallel.

Kind regards
Uffe

Re: [PATCH V8 00/14] mmc: Add Command Queue support

[Adrian Hunter]

On 11/10/17 15:13, Ulf Hansson wrote:
> On 10 October 2017 at 15:31, Adrian Hunter <adrian.hunter@intel.com> wrote:
>> On 10/10/17 16:08, Ulf Hansson wrote:
>>> [...]
>>>
>>>>>>>
>>>>>>> I have also run some test on my ux500 board and enabling the blkmq
>>>>>>> path via the new MMC Kconfig option. My idea was to run some iozone
>>>>>>> comparisons between the legacy path and the new blkmq path, but I just
>>>>>>> couldn't get to that point because of the following errors.
>>>>>>>
>>>>>>> I am using a Kingston 4GB SDHC card, which is detected and mounted
>>>>>>> nicely. However, when I decide to do some writes to the card I get the
>>>>>>> following errors.
>>>>>>>
>>>>>>> root@ME:/mnt/sdcard dd if=/dev/zero of=testfile bs=8192 count=5000 conv=fsync
>>>>>>> [  463.714294] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>> [  464.722656] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>> [  466.081481] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>> [  467.111236] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>> [  468.669647] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>> [  469.685699] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>> [  471.043334] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>> [  472.052337] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>> [  473.342651] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>> [  474.323760] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>> [  475.544769] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>> [  476.539031] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>> [  477.748474] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>> [  478.724182] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>
>>>>>>> I haven't yet got the point of investigating this any further, and
>>>>>>> unfortunate I have a busy schedule with traveling next week. I will do
>>>>>>> my best to look into this as soon as I can.
>>>>>>>
>>>>>>> Perhaps you have some ideas?
>>>>>>
>>>>>> The behaviour depends on whether you have MMC_CAP_WAIT_WHILE_BUSY. Try
>>>>>> changing that and see if it makes a difference.
>>>>>
>>>>> Yes, it does! I disabled MMC_CAP_WAIT_WHILE_BUSY (and its
>>>>> corresponding code in mmci.c) and the errors goes away.
>>>>>
>>>>> When I use MMC_CAP_WAIT_WHILE_BUSY I get these problems:
>>>>>
>>>>> [  223.820983] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>> [  224.815795] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>> [  226.034881] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>> [  227.112884] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>> [  227.220275] mmc0: Card stuck in wrong state! mmcblk0 mmc_blk_card_stuck
>>>>> [  228.686798] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>> [  229.892150] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>> [  231.031890] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>> [  232.239013] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>> 5000+0 records in
>>>>> 5000+0 records out
>>>>> root@ME:/mnt/sdcard
>>>>>
>>>>> I looked at the new blkmq code from patch v10 13/15. It seems like the
>>>>> MMC_CAP_WAIT_WHILE_BUSY is used to determine whether the async request
>>>>> mechanism should be used or not. Perhaps I didn't looked close enough,
>>>>> but maybe you could elaborate on why this seems to be the case!?
>>>>
>>>> MMC_CAP_WAIT_WHILE_BUSY is necessary because it means that a data transfer
>>>> request has finished when the host controller calls mmc_request_done(). i.e.
>>>> polling the card is not necessary.
>>>
>>> Well, that is a rather big change on its own. Earlier we polled with
>>> CMD13 to verify that the card has moved back to the transfer state, in
>>> case it was a write. And that was no matter of MMC_CAP_WAIT_WHILE_BUSY
>>> was set or not. Right!?
>>
>> Yes
>>
>>>
>>> I am not sure it's a good idea to bypass that validation, it seems
>>> fragile to rely only on the busy detection on DAT line for writes.
>>
>> Can you cite something from the specifications that backs that up, because I
>> couldn't find anything to suggest that CMD13 polling was expected.
> 
> No I can't, but I don't see why that matters.
> 
> My point is, if we want to go down that road by avoiding the CMD13
> polling, that needs to be a separate change, which we can test and
> confirm on its own.
> 
>>
>>>
>>>>
>>>> Have you tried V9 or V10.  There was a fix in V9 related to calling
>>>> ->post_req() which could mess up DMA.
>>>
>>> I have used V10.
>>>
>>>>
>>>> The other thing that could go wrong with DMA is if it cannot accept
>>>> ->post_req() being called from mmc_request_done().
>>>
>>> I don't think mmci has a problem with that, however why do you want to
>>> do this? Wouldn't that defeat some of the benefits with the async
>>> request mechanism?
>>
>> Perhaps - but it would need to be tested.  If there are more requests
>> waiting, one optimization could be to defer ->post_req() until after the
>> next request is started.
> 
> This is already proven, because this how the existing mmc async
> request mechanism works.
> 
> In ->post_req() callbacks, host drivers may do dma_unmap_sg(), which
> is something that could be costly and therefore it's better to start a
> new request before, such these things can go on in parallel.

OK I will make a patch that takes care of both issues.  That will also mean
the request is not completed in the ->done() callback because ->post_req()
must precede block layer completion.

Re: [PATCH V8 00/14] mmc: Add Command Queue support

[Ulf Hansson]

On 11 October 2017 at 14:58, Adrian Hunter <adrian.hunter@intel.com> wrote:
> On 11/10/17 15:13, Ulf Hansson wrote:
>> On 10 October 2017 at 15:31, Adrian Hunter <adrian.hunter@intel.com> wrote:
>>> On 10/10/17 16:08, Ulf Hansson wrote:
>>>> [...]
>>>>
>>>>>>>>
>>>>>>>> I have also run some test on my ux500 board and enabling the blkmq
>>>>>>>> path via the new MMC Kconfig option. My idea was to run some iozone
>>>>>>>> comparisons between the legacy path and the new blkmq path, but I just
>>>>>>>> couldn't get to that point because of the following errors.
>>>>>>>>
>>>>>>>> I am using a Kingston 4GB SDHC card, which is detected and mounted
>>>>>>>> nicely. However, when I decide to do some writes to the card I get the
>>>>>>>> following errors.
>>>>>>>>
>>>>>>>> root@ME:/mnt/sdcard dd if=/dev/zero of=testfile bs=8192 count=5000 conv=fsync
>>>>>>>> [  463.714294] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>> [  464.722656] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>> [  466.081481] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>> [  467.111236] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>> [  468.669647] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>> [  469.685699] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>> [  471.043334] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>> [  472.052337] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>> [  473.342651] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>> [  474.323760] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>> [  475.544769] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>> [  476.539031] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>> [  477.748474] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>> [  478.724182] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>
>>>>>>>> I haven't yet got the point of investigating this any further, and
>>>>>>>> unfortunate I have a busy schedule with traveling next week. I will do
>>>>>>>> my best to look into this as soon as I can.
>>>>>>>>
>>>>>>>> Perhaps you have some ideas?
>>>>>>>
>>>>>>> The behaviour depends on whether you have MMC_CAP_WAIT_WHILE_BUSY. Try
>>>>>>> changing that and see if it makes a difference.
>>>>>>
>>>>>> Yes, it does! I disabled MMC_CAP_WAIT_WHILE_BUSY (and its
>>>>>> corresponding code in mmci.c) and the errors goes away.
>>>>>>
>>>>>> When I use MMC_CAP_WAIT_WHILE_BUSY I get these problems:
>>>>>>
>>>>>> [  223.820983] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>> [  224.815795] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>> [  226.034881] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>> [  227.112884] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>> [  227.220275] mmc0: Card stuck in wrong state! mmcblk0 mmc_blk_card_stuck
>>>>>> [  228.686798] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>> [  229.892150] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>> [  231.031890] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>> [  232.239013] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>> 5000+0 records in
>>>>>> 5000+0 records out
>>>>>> root@ME:/mnt/sdcard
>>>>>>
>>>>>> I looked at the new blkmq code from patch v10 13/15. It seems like the
>>>>>> MMC_CAP_WAIT_WHILE_BUSY is used to determine whether the async request
>>>>>> mechanism should be used or not. Perhaps I didn't looked close enough,
>>>>>> but maybe you could elaborate on why this seems to be the case!?
>>>>>
>>>>> MMC_CAP_WAIT_WHILE_BUSY is necessary because it means that a data transfer
>>>>> request has finished when the host controller calls mmc_request_done(). i.e.
>>>>> polling the card is not necessary.
>>>>
>>>> Well, that is a rather big change on its own. Earlier we polled with
>>>> CMD13 to verify that the card has moved back to the transfer state, in
>>>> case it was a write. And that was no matter of MMC_CAP_WAIT_WHILE_BUSY
>>>> was set or not. Right!?
>>>
>>> Yes
>>>
>>>>
>>>> I am not sure it's a good idea to bypass that validation, it seems
>>>> fragile to rely only on the busy detection on DAT line for writes.
>>>
>>> Can you cite something from the specifications that backs that up, because I
>>> couldn't find anything to suggest that CMD13 polling was expected.
>>
>> No I can't, but I don't see why that matters.
>>
>> My point is, if we want to go down that road by avoiding the CMD13
>> polling, that needs to be a separate change, which we can test and
>> confirm on its own.
>>
>>>
>>>>
>>>>>
>>>>> Have you tried V9 or V10.  There was a fix in V9 related to calling
>>>>> ->post_req() which could mess up DMA.
>>>>
>>>> I have used V10.
>>>>
>>>>>
>>>>> The other thing that could go wrong with DMA is if it cannot accept
>>>>> ->post_req() being called from mmc_request_done().
>>>>
>>>> I don't think mmci has a problem with that, however why do you want to
>>>> do this? Wouldn't that defeat some of the benefits with the async
>>>> request mechanism?
>>>
>>> Perhaps - but it would need to be tested.  If there are more requests
>>> waiting, one optimization could be to defer ->post_req() until after the
>>> next request is started.
>>
>> This is already proven, because this how the existing mmc async
>> request mechanism works.
>>
>> In ->post_req() callbacks, host drivers may do dma_unmap_sg(), which
>> is something that could be costly and therefore it's better to start a
>> new request before, such these things can go on in parallel.
>
> OK I will make a patch that takes care of both issues.  That will also mean
> the request is not completed in the ->done() callback because ->post_req()
> must precede block layer completion.

Right.

Actually completing the request in the ->done callback, may still be
possible, because in principle it only needs to inform the other
prepared request that it may start, before it continues to post
process/completes the current one.

However, by looking at for example how mmci.c works, it actually holds
its spinlock while it calls mmc_request_done(). The same spinlock is
taken in the ->request() function, but not in the ->post_req()
function. In other words, completing the request in the ->done()
callback, would make mmci to keep the spinlock held throughout the
post processing cycle, which then prevents the next request from being
started.

So my conclusion is, let's start a as you suggested, by not completing
the request in ->done() as to maintain existing behavior. Then we can
address optimizations on top, which very likely will involve doing
changes to host drivers as well.

Kind regards
Uffe

Re: [PATCH V8 00/14] mmc: Add Command Queue support

[Linus Walleij]

On Wed, Oct 11, 2017 at 3:58 PM, Ulf Hansson <ulf.hansson@linaro.org> wrote:

> Actually completing the request in the ->done callback, may still be
> possible, because in principle it only needs to inform the other
> prepared request that it may start, before it continues to post
> process/completes the current one.

I try to do that in this patch:
https://marc.info/?l=linux-mmc&m=148665460925587&w=2

I'm trying to rebase this work now.

> However, by looking at for example how mmci.c works, it actually holds
> its spinlock while it calls mmc_request_done(). The same spinlock is
> taken in the ->request() function, but not in the ->post_req()
> function. In other words, completing the request in the ->done()
> callback, would make mmci to keep the spinlock held throughout the
> post processing cycle, which then prevents the next request from being
> started.

It seems my patch would not deliver in some drivers until we look
into the locking semantics in the drivers.

Yours,
Linus Walleij

Re: [PATCH V8 00/14] mmc: Add Command Queue support

[Ulf Hansson]

On 12 October 2017 at 10:08, Linus Walleij <linus.walleij@linaro.org> wrote:
> On Wed, Oct 11, 2017 at 3:58 PM, Ulf Hansson <ulf.hansson@linaro.org> wrote:
>
>> Actually completing the request in the ->done callback, may still be
>> possible, because in principle it only needs to inform the other
>> prepared request that it may start, before it continues to post
>> process/completes the current one.
>
> I try to do that in this patch:
> https://marc.info/?l=linux-mmc&m=148665460925587&w=2

Yeah, something like that!

>
> I'm trying to rebase this work now.

Great!

>
>> However, by looking at for example how mmci.c works, it actually holds
>> its spinlock while it calls mmc_request_done(). The same spinlock is
>> taken in the ->request() function, but not in the ->post_req()
>> function. In other words, completing the request in the ->done()
>> callback, would make mmci to keep the spinlock held throughout the
>> post processing cycle, which then prevents the next request from being
>> started.
>
> It seems my patch would not deliver in some drivers until we look
> into the locking semantics in the drivers.

Yes!

As a matter of fact the above change may actually introduce
performance regressions, in case the behavior of the driver doesn't
follow the expectations from the core. For that reason, either we have
to invent a new MMC cap to allow drivers to opt-in using the
"shortcut" or fix host drivers first.

Kind regards
Uffe

Re: [PATCH V8 00/14] mmc: Add Command Queue support

[Adrian Hunter]

On 11/10/17 16:58, Ulf Hansson wrote:
> On 11 October 2017 at 14:58, Adrian Hunter <adrian.hunter@intel.com> wrote:
>> On 11/10/17 15:13, Ulf Hansson wrote:
>>> On 10 October 2017 at 15:31, Adrian Hunter <adrian.hunter@intel.com> wrote:
>>>> On 10/10/17 16:08, Ulf Hansson wrote:
>>>>> [...]
>>>>>
>>>>>>>>>
>>>>>>>>> I have also run some test on my ux500 board and enabling the blkmq
>>>>>>>>> path via the new MMC Kconfig option. My idea was to run some iozone
>>>>>>>>> comparisons between the legacy path and the new blkmq path, but I just
>>>>>>>>> couldn't get to that point because of the following errors.
>>>>>>>>>
>>>>>>>>> I am using a Kingston 4GB SDHC card, which is detected and mounted
>>>>>>>>> nicely. However, when I decide to do some writes to the card I get the
>>>>>>>>> following errors.
>>>>>>>>>
>>>>>>>>> root@ME:/mnt/sdcard dd if=/dev/zero of=testfile bs=8192 count=5000 conv=fsync
>>>>>>>>> [  463.714294] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>> [  464.722656] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>> [  466.081481] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>> [  467.111236] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>> [  468.669647] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>> [  469.685699] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>> [  471.043334] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>> [  472.052337] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>> [  473.342651] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>> [  474.323760] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>> [  475.544769] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>> [  476.539031] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>> [  477.748474] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>> [  478.724182] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>
>>>>>>>>> I haven't yet got the point of investigating this any further, and
>>>>>>>>> unfortunate I have a busy schedule with traveling next week. I will do
>>>>>>>>> my best to look into this as soon as I can.
>>>>>>>>>
>>>>>>>>> Perhaps you have some ideas?
>>>>>>>>
>>>>>>>> The behaviour depends on whether you have MMC_CAP_WAIT_WHILE_BUSY. Try
>>>>>>>> changing that and see if it makes a difference.
>>>>>>>
>>>>>>> Yes, it does! I disabled MMC_CAP_WAIT_WHILE_BUSY (and its
>>>>>>> corresponding code in mmci.c) and the errors goes away.
>>>>>>>
>>>>>>> When I use MMC_CAP_WAIT_WHILE_BUSY I get these problems:
>>>>>>>
>>>>>>> [  223.820983] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>> [  224.815795] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>> [  226.034881] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>> [  227.112884] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>> [  227.220275] mmc0: Card stuck in wrong state! mmcblk0 mmc_blk_card_stuck
>>>>>>> [  228.686798] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>> [  229.892150] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>> [  231.031890] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>> [  232.239013] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>> 5000+0 records in
>>>>>>> 5000+0 records out
>>>>>>> root@ME:/mnt/sdcard
>>>>>>>
>>>>>>> I looked at the new blkmq code from patch v10 13/15. It seems like the
>>>>>>> MMC_CAP_WAIT_WHILE_BUSY is used to determine whether the async request
>>>>>>> mechanism should be used or not. Perhaps I didn't looked close enough,
>>>>>>> but maybe you could elaborate on why this seems to be the case!?
>>>>>>
>>>>>> MMC_CAP_WAIT_WHILE_BUSY is necessary because it means that a data transfer
>>>>>> request has finished when the host controller calls mmc_request_done(). i.e.
>>>>>> polling the card is not necessary.
>>>>>
>>>>> Well, that is a rather big change on its own. Earlier we polled with
>>>>> CMD13 to verify that the card has moved back to the transfer state, in
>>>>> case it was a write. And that was no matter of MMC_CAP_WAIT_WHILE_BUSY
>>>>> was set or not. Right!?
>>>>
>>>> Yes
>>>>
>>>>>
>>>>> I am not sure it's a good idea to bypass that validation, it seems
>>>>> fragile to rely only on the busy detection on DAT line for writes.
>>>>
>>>> Can you cite something from the specifications that backs that up, because I
>>>> couldn't find anything to suggest that CMD13 polling was expected.
>>>
>>> No I can't, but I don't see why that matters.
>>>
>>> My point is, if we want to go down that road by avoiding the CMD13
>>> polling, that needs to be a separate change, which we can test and
>>> confirm on its own.
>>>
>>>>
>>>>>
>>>>>>
>>>>>> Have you tried V9 or V10.  There was a fix in V9 related to calling
>>>>>> ->post_req() which could mess up DMA.
>>>>>
>>>>> I have used V10.
>>>>>
>>>>>>
>>>>>> The other thing that could go wrong with DMA is if it cannot accept
>>>>>> ->post_req() being called from mmc_request_done().
>>>>>
>>>>> I don't think mmci has a problem with that, however why do you want to
>>>>> do this? Wouldn't that defeat some of the benefits with the async
>>>>> request mechanism?
>>>>
>>>> Perhaps - but it would need to be tested.  If there are more requests
>>>> waiting, one optimization could be to defer ->post_req() until after the
>>>> next request is started.
>>>
>>> This is already proven, because this how the existing mmc async
>>> request mechanism works.
>>>
>>> In ->post_req() callbacks, host drivers may do dma_unmap_sg(), which
>>> is something that could be costly and therefore it's better to start a
>>> new request before, such these things can go on in parallel.
>>
>> OK I will make a patch that takes care of both issues.  That will also mean
>> the request is not completed in the ->done() callback because ->post_req()
>> must precede block layer completion.
> 
> Right.
> 
> Actually completing the request in the ->done callback, may still be
> possible, because in principle it only needs to inform the other
> prepared request that it may start, before it continues to post
> process/completes the current one.

It already does that.

> 
> However, by looking at for example how mmci.c works, it actually holds
> its spinlock while it calls mmc_request_done(). The same spinlock is
> taken in the ->request() function, but not in the ->post_req()
> function. In other words, completing the request in the ->done()
> callback, would make mmci to keep the spinlock held throughout the
> post processing cycle, which then prevents the next request from being
> started.

It is in interrupt context also. So if the block layer work is on the same
CPU it will anyway wait.

> 
> So my conclusion is, let's start a as you suggested, by not completing
> the request in ->done() as to maintain existing behavior. Then we can
> address optimizations on top, which very likely will involve doing
> changes to host drivers as well.

Ideally it should be possible to start the next already prepared request
from the ->done() callback of the current request.  Perhaps add a new host API.

Re: [PATCH V8 00/14] mmc: Add Command Queue support

[Adrian Hunter]

On 11/10/17 16:58, Ulf Hansson wrote:
> On 11 October 2017 at 14:58, Adrian Hunter <adrian.hunter@intel.com> wrote:
>> On 11/10/17 15:13, Ulf Hansson wrote:
>>> On 10 October 2017 at 15:31, Adrian Hunter <adrian.hunter@intel.com> wrote:
>>>> On 10/10/17 16:08, Ulf Hansson wrote:
>>>>> [...]
>>>>>
>>>>>>>>>
>>>>>>>>> I have also run some test on my ux500 board and enabling the blkmq
>>>>>>>>> path via the new MMC Kconfig option. My idea was to run some iozone
>>>>>>>>> comparisons between the legacy path and the new blkmq path, but I just
>>>>>>>>> couldn't get to that point because of the following errors.
>>>>>>>>>
>>>>>>>>> I am using a Kingston 4GB SDHC card, which is detected and mounted
>>>>>>>>> nicely. However, when I decide to do some writes to the card I get the
>>>>>>>>> following errors.
>>>>>>>>>
>>>>>>>>> root@ME:/mnt/sdcard dd if=/dev/zero of=testfile bs=8192 count=5000 conv=fsync
>>>>>>>>> [  463.714294] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>> [  464.722656] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>> [  466.081481] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>> [  467.111236] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>> [  468.669647] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>> [  469.685699] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>> [  471.043334] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>> [  472.052337] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>> [  473.342651] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>> [  474.323760] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>> [  475.544769] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>> [  476.539031] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>> [  477.748474] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>> [  478.724182] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>
>>>>>>>>> I haven't yet got the point of investigating this any further, and
>>>>>>>>> unfortunate I have a busy schedule with traveling next week. I will do
>>>>>>>>> my best to look into this as soon as I can.
>>>>>>>>>
>>>>>>>>> Perhaps you have some ideas?
>>>>>>>>
>>>>>>>> The behaviour depends on whether you have MMC_CAP_WAIT_WHILE_BUSY. Try
>>>>>>>> changing that and see if it makes a difference.
>>>>>>>
>>>>>>> Yes, it does! I disabled MMC_CAP_WAIT_WHILE_BUSY (and its
>>>>>>> corresponding code in mmci.c) and the errors goes away.
>>>>>>>
>>>>>>> When I use MMC_CAP_WAIT_WHILE_BUSY I get these problems:
>>>>>>>
>>>>>>> [  223.820983] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>> [  224.815795] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>> [  226.034881] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>> [  227.112884] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>> [  227.220275] mmc0: Card stuck in wrong state! mmcblk0 mmc_blk_card_stuck
>>>>>>> [  228.686798] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>> [  229.892150] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>> [  231.031890] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>> [  232.239013] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>> 5000+0 records in
>>>>>>> 5000+0 records out
>>>>>>> root@ME:/mnt/sdcard
>>>>>>>
>>>>>>> I looked at the new blkmq code from patch v10 13/15. It seems like the
>>>>>>> MMC_CAP_WAIT_WHILE_BUSY is used to determine whether the async request
>>>>>>> mechanism should be used or not. Perhaps I didn't looked close enough,
>>>>>>> but maybe you could elaborate on why this seems to be the case!?
>>>>>>
>>>>>> MMC_CAP_WAIT_WHILE_BUSY is necessary because it means that a data transfer
>>>>>> request has finished when the host controller calls mmc_request_done(). i.e.
>>>>>> polling the card is not necessary.
>>>>>
>>>>> Well, that is a rather big change on its own. Earlier we polled with
>>>>> CMD13 to verify that the card has moved back to the transfer state, in
>>>>> case it was a write. And that was no matter of MMC_CAP_WAIT_WHILE_BUSY
>>>>> was set or not. Right!?
>>>>
>>>> Yes
>>>>
>>>>>
>>>>> I am not sure it's a good idea to bypass that validation, it seems
>>>>> fragile to rely only on the busy detection on DAT line for writes.
>>>>
>>>> Can you cite something from the specifications that backs that up, because I
>>>> couldn't find anything to suggest that CMD13 polling was expected.
>>>
>>> No I can't, but I don't see why that matters.
>>>
>>> My point is, if we want to go down that road by avoiding the CMD13
>>> polling, that needs to be a separate change, which we can test and
>>> confirm on its own.
>>>
>>>>
>>>>>
>>>>>>
>>>>>> Have you tried V9 or V10.  There was a fix in V9 related to calling
>>>>>> ->post_req() which could mess up DMA.
>>>>>
>>>>> I have used V10.
>>>>>
>>>>>>
>>>>>> The other thing that could go wrong with DMA is if it cannot accept
>>>>>> ->post_req() being called from mmc_request_done().
>>>>>
>>>>> I don't think mmci has a problem with that, however why do you want to
>>>>> do this? Wouldn't that defeat some of the benefits with the async
>>>>> request mechanism?
>>>>
>>>> Perhaps - but it would need to be tested.  If there are more requests
>>>> waiting, one optimization could be to defer ->post_req() until after the
>>>> next request is started.
>>>
>>> This is already proven, because this how the existing mmc async
>>> request mechanism works.
>>>
>>> In ->post_req() callbacks, host drivers may do dma_unmap_sg(), which
>>> is something that could be costly and therefore it's better to start a
>>> new request before, such these things can go on in parallel.
>>
>> OK I will make a patch that takes care of both issues.  That will also mean
>> the request is not completed in the ->done() callback because ->post_req()
>> must precede block layer completion.
> 
> Right.
> 
> Actually completing the request in the ->done callback, may still be
> possible, because in principle it only needs to inform the other
> prepared request that it may start, before it continues to post
> process/completes the current one.
> 
> However, by looking at for example how mmci.c works, it actually holds
> its spinlock while it calls mmc_request_done(). The same spinlock is
> taken in the ->request() function, but not in the ->post_req()
> function. In other words, completing the request in the ->done()
> callback, would make mmci to keep the spinlock held throughout the
> post processing cycle, which then prevents the next request from being
> started.
> 
> So my conclusion is, let's start a as you suggested, by not completing
> the request in ->done() as to maintain existing behavior. Then we can
> address optimizations on top, which very likely will involve doing
> changes to host drivers as well.

Have you tested the latest version now?

Re: [PATCH V8 00/14] mmc: Add Command Queue support

[Adrian Hunter]

On 18/10/17 09:16, Adrian Hunter wrote:
> On 11/10/17 16:58, Ulf Hansson wrote:
>> On 11 October 2017 at 14:58, Adrian Hunter <adrian.hunter@intel.com> wrote:
>>> On 11/10/17 15:13, Ulf Hansson wrote:
>>>> On 10 October 2017 at 15:31, Adrian Hunter <adrian.hunter@intel.com> wrote:
>>>>> On 10/10/17 16:08, Ulf Hansson wrote:
>>>>>> [...]
>>>>>>
>>>>>>>>>>
>>>>>>>>>> I have also run some test on my ux500 board and enabling the blkmq
>>>>>>>>>> path via the new MMC Kconfig option. My idea was to run some iozone
>>>>>>>>>> comparisons between the legacy path and the new blkmq path, but I just
>>>>>>>>>> couldn't get to that point because of the following errors.
>>>>>>>>>>
>>>>>>>>>> I am using a Kingston 4GB SDHC card, which is detected and mounted
>>>>>>>>>> nicely. However, when I decide to do some writes to the card I get the
>>>>>>>>>> following errors.
>>>>>>>>>>
>>>>>>>>>> root@ME:/mnt/sdcard dd if=/dev/zero of=testfile bs=8192 count=5000 conv=fsync
>>>>>>>>>> [  463.714294] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>> [  464.722656] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>> [  466.081481] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>> [  467.111236] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>> [  468.669647] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>> [  469.685699] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>> [  471.043334] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>> [  472.052337] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>> [  473.342651] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>> [  474.323760] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>> [  475.544769] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>> [  476.539031] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>> [  477.748474] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>> [  478.724182] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>>
>>>>>>>>>> I haven't yet got the point of investigating this any further, and
>>>>>>>>>> unfortunate I have a busy schedule with traveling next week. I will do
>>>>>>>>>> my best to look into this as soon as I can.
>>>>>>>>>>
>>>>>>>>>> Perhaps you have some ideas?
>>>>>>>>>
>>>>>>>>> The behaviour depends on whether you have MMC_CAP_WAIT_WHILE_BUSY. Try
>>>>>>>>> changing that and see if it makes a difference.
>>>>>>>>
>>>>>>>> Yes, it does! I disabled MMC_CAP_WAIT_WHILE_BUSY (and its
>>>>>>>> corresponding code in mmci.c) and the errors goes away.
>>>>>>>>
>>>>>>>> When I use MMC_CAP_WAIT_WHILE_BUSY I get these problems:
>>>>>>>>
>>>>>>>> [  223.820983] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>> [  224.815795] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>> [  226.034881] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>> [  227.112884] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>> [  227.220275] mmc0: Card stuck in wrong state! mmcblk0 mmc_blk_card_stuck
>>>>>>>> [  228.686798] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>> [  229.892150] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>> [  231.031890] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>> [  232.239013] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>> 5000+0 records in
>>>>>>>> 5000+0 records out
>>>>>>>> root@ME:/mnt/sdcard
>>>>>>>>
>>>>>>>> I looked at the new blkmq code from patch v10 13/15. It seems like the
>>>>>>>> MMC_CAP_WAIT_WHILE_BUSY is used to determine whether the async request
>>>>>>>> mechanism should be used or not. Perhaps I didn't looked close enough,
>>>>>>>> but maybe you could elaborate on why this seems to be the case!?
>>>>>>>
>>>>>>> MMC_CAP_WAIT_WHILE_BUSY is necessary because it means that a data transfer
>>>>>>> request has finished when the host controller calls mmc_request_done(). i.e.
>>>>>>> polling the card is not necessary.
>>>>>>
>>>>>> Well, that is a rather big change on its own. Earlier we polled with
>>>>>> CMD13 to verify that the card has moved back to the transfer state, in
>>>>>> case it was a write. And that was no matter of MMC_CAP_WAIT_WHILE_BUSY
>>>>>> was set or not. Right!?
>>>>>
>>>>> Yes
>>>>>
>>>>>>
>>>>>> I am not sure it's a good idea to bypass that validation, it seems
>>>>>> fragile to rely only on the busy detection on DAT line for writes.
>>>>>
>>>>> Can you cite something from the specifications that backs that up, because I
>>>>> couldn't find anything to suggest that CMD13 polling was expected.
>>>>
>>>> No I can't, but I don't see why that matters.
>>>>
>>>> My point is, if we want to go down that road by avoiding the CMD13
>>>> polling, that needs to be a separate change, which we can test and
>>>> confirm on its own.
>>>>
>>>>>
>>>>>>
>>>>>>>
>>>>>>> Have you tried V9 or V10.  There was a fix in V9 related to calling
>>>>>>> ->post_req() which could mess up DMA.
>>>>>>
>>>>>> I have used V10.
>>>>>>
>>>>>>>
>>>>>>> The other thing that could go wrong with DMA is if it cannot accept
>>>>>>> ->post_req() being called from mmc_request_done().
>>>>>>
>>>>>> I don't think mmci has a problem with that, however why do you want to
>>>>>> do this? Wouldn't that defeat some of the benefits with the async
>>>>>> request mechanism?
>>>>>
>>>>> Perhaps - but it would need to be tested.  If there are more requests
>>>>> waiting, one optimization could be to defer ->post_req() until after the
>>>>> next request is started.
>>>>
>>>> This is already proven, because this how the existing mmc async
>>>> request mechanism works.
>>>>
>>>> In ->post_req() callbacks, host drivers may do dma_unmap_sg(), which
>>>> is something that could be costly and therefore it's better to start a
>>>> new request before, such these things can go on in parallel.
>>>
>>> OK I will make a patch that takes care of both issues.  That will also mean
>>> the request is not completed in the ->done() callback because ->post_req()
>>> must precede block layer completion.
>>
>> Right.
>>
>> Actually completing the request in the ->done callback, may still be
>> possible, because in principle it only needs to inform the other
>> prepared request that it may start, before it continues to post
>> process/completes the current one.
>>
>> However, by looking at for example how mmci.c works, it actually holds
>> its spinlock while it calls mmc_request_done(). The same spinlock is
>> taken in the ->request() function, but not in the ->post_req()
>> function. In other words, completing the request in the ->done()
>> callback, would make mmci to keep the spinlock held throughout the
>> post processing cycle, which then prevents the next request from being
>> started.
>>
>> So my conclusion is, let's start a as you suggested, by not completing
>> the request in ->done() as to maintain existing behavior. Then we can
>> address optimizations on top, which very likely will involve doing
>> changes to host drivers as well.
> 
> Have you tested the latest version now?
> 

Ping?

Re: [PATCH V8 00/14] mmc: Add Command Queue support

[Adrian Hunter]

On 19/10/17 14:44, Adrian Hunter wrote:
> On 18/10/17 09:16, Adrian Hunter wrote:
>> On 11/10/17 16:58, Ulf Hansson wrote:
>>> On 11 October 2017 at 14:58, Adrian Hunter <adrian.hunter@intel.com> wrote:
>>>> On 11/10/17 15:13, Ulf Hansson wrote:
>>>>> On 10 October 2017 at 15:31, Adrian Hunter <adrian.hunter@intel.com> wrote:
>>>>>> On 10/10/17 16:08, Ulf Hansson wrote:
>>>>>>> [...]
>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> I have also run some test on my ux500 board and enabling the blkmq
>>>>>>>>>>> path via the new MMC Kconfig option. My idea was to run some iozone
>>>>>>>>>>> comparisons between the legacy path and the new blkmq path, but I just
>>>>>>>>>>> couldn't get to that point because of the following errors.
>>>>>>>>>>>
>>>>>>>>>>> I am using a Kingston 4GB SDHC card, which is detected and mounted
>>>>>>>>>>> nicely. However, when I decide to do some writes to the card I get the
>>>>>>>>>>> following errors.
>>>>>>>>>>>
>>>>>>>>>>> root@ME:/mnt/sdcard dd if=/dev/zero of=testfile bs=8192 count=5000 conv=fsync
>>>>>>>>>>> [  463.714294] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>>> [  464.722656] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>>> [  466.081481] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>>> [  467.111236] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>>> [  468.669647] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>>> [  469.685699] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>>> [  471.043334] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>>> [  472.052337] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>>> [  473.342651] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>>> [  474.323760] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>>> [  475.544769] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>>> [  476.539031] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>>> [  477.748474] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>>> [  478.724182] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>>>
>>>>>>>>>>> I haven't yet got the point of investigating this any further, and
>>>>>>>>>>> unfortunate I have a busy schedule with traveling next week. I will do
>>>>>>>>>>> my best to look into this as soon as I can.
>>>>>>>>>>>
>>>>>>>>>>> Perhaps you have some ideas?
>>>>>>>>>>
>>>>>>>>>> The behaviour depends on whether you have MMC_CAP_WAIT_WHILE_BUSY. Try
>>>>>>>>>> changing that and see if it makes a difference.
>>>>>>>>>
>>>>>>>>> Yes, it does! I disabled MMC_CAP_WAIT_WHILE_BUSY (and its
>>>>>>>>> corresponding code in mmci.c) and the errors goes away.
>>>>>>>>>
>>>>>>>>> When I use MMC_CAP_WAIT_WHILE_BUSY I get these problems:
>>>>>>>>>
>>>>>>>>> [  223.820983] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>> [  224.815795] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>> [  226.034881] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>> [  227.112884] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>> [  227.220275] mmc0: Card stuck in wrong state! mmcblk0 mmc_blk_card_stuck
>>>>>>>>> [  228.686798] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>> [  229.892150] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>> [  231.031890] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>> [  232.239013] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>> 5000+0 records in
>>>>>>>>> 5000+0 records out
>>>>>>>>> root@ME:/mnt/sdcard
>>>>>>>>>
>>>>>>>>> I looked at the new blkmq code from patch v10 13/15. It seems like the
>>>>>>>>> MMC_CAP_WAIT_WHILE_BUSY is used to determine whether the async request
>>>>>>>>> mechanism should be used or not. Perhaps I didn't looked close enough,
>>>>>>>>> but maybe you could elaborate on why this seems to be the case!?
>>>>>>>>
>>>>>>>> MMC_CAP_WAIT_WHILE_BUSY is necessary because it means that a data transfer
>>>>>>>> request has finished when the host controller calls mmc_request_done(). i.e.
>>>>>>>> polling the card is not necessary.
>>>>>>>
>>>>>>> Well, that is a rather big change on its own. Earlier we polled with
>>>>>>> CMD13 to verify that the card has moved back to the transfer state, in
>>>>>>> case it was a write. And that was no matter of MMC_CAP_WAIT_WHILE_BUSY
>>>>>>> was set or not. Right!?
>>>>>>
>>>>>> Yes
>>>>>>
>>>>>>>
>>>>>>> I am not sure it's a good idea to bypass that validation, it seems
>>>>>>> fragile to rely only on the busy detection on DAT line for writes.
>>>>>>
>>>>>> Can you cite something from the specifications that backs that up, because I
>>>>>> couldn't find anything to suggest that CMD13 polling was expected.
>>>>>
>>>>> No I can't, but I don't see why that matters.
>>>>>
>>>>> My point is, if we want to go down that road by avoiding the CMD13
>>>>> polling, that needs to be a separate change, which we can test and
>>>>> confirm on its own.
>>>>>
>>>>>>
>>>>>>>
>>>>>>>>
>>>>>>>> Have you tried V9 or V10.  There was a fix in V9 related to calling
>>>>>>>> ->post_req() which could mess up DMA.
>>>>>>>
>>>>>>> I have used V10.
>>>>>>>
>>>>>>>>
>>>>>>>> The other thing that could go wrong with DMA is if it cannot accept
>>>>>>>> ->post_req() being called from mmc_request_done().
>>>>>>>
>>>>>>> I don't think mmci has a problem with that, however why do you want to
>>>>>>> do this? Wouldn't that defeat some of the benefits with the async
>>>>>>> request mechanism?
>>>>>>
>>>>>> Perhaps - but it would need to be tested.  If there are more requests
>>>>>> waiting, one optimization could be to defer ->post_req() until after the
>>>>>> next request is started.
>>>>>
>>>>> This is already proven, because this how the existing mmc async
>>>>> request mechanism works.
>>>>>
>>>>> In ->post_req() callbacks, host drivers may do dma_unmap_sg(), which
>>>>> is something that could be costly and therefore it's better to start a
>>>>> new request before, such these things can go on in parallel.
>>>>
>>>> OK I will make a patch that takes care of both issues.  That will also mean
>>>> the request is not completed in the ->done() callback because ->post_req()
>>>> must precede block layer completion.
>>>
>>> Right.
>>>
>>> Actually completing the request in the ->done callback, may still be
>>> possible, because in principle it only needs to inform the other
>>> prepared request that it may start, before it continues to post
>>> process/completes the current one.
>>>
>>> However, by looking at for example how mmci.c works, it actually holds
>>> its spinlock while it calls mmc_request_done(). The same spinlock is
>>> taken in the ->request() function, but not in the ->post_req()
>>> function. In other words, completing the request in the ->done()
>>> callback, would make mmci to keep the spinlock held throughout the
>>> post processing cycle, which then prevents the next request from being
>>> started.
>>>
>>> So my conclusion is, let's start a as you suggested, by not completing
>>> the request in ->done() as to maintain existing behavior. Then we can
>>> address optimizations on top, which very likely will involve doing
>>> changes to host drivers as well.
>>
>> Have you tested the latest version now?
>>
> 
> Ping?

Still ping?

Re: [PATCH V8 00/14] mmc: Add Command Queue support

[Adrian Hunter]

On 20/10/17 15:30, Adrian Hunter wrote:
> On 19/10/17 14:44, Adrian Hunter wrote:
>> On 18/10/17 09:16, Adrian Hunter wrote:
>>> On 11/10/17 16:58, Ulf Hansson wrote:
>>>> On 11 October 2017 at 14:58, Adrian Hunter <adrian.hunter@intel.com> wrote:
>>>>> On 11/10/17 15:13, Ulf Hansson wrote:
>>>>>> On 10 October 2017 at 15:31, Adrian Hunter <adrian.hunter@intel.com> wrote:
>>>>>>> On 10/10/17 16:08, Ulf Hansson wrote:
>>>>>>>> [...]
>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>>> I have also run some test on my ux500 board and enabling the blkmq
>>>>>>>>>>>> path via the new MMC Kconfig option. My idea was to run some iozone
>>>>>>>>>>>> comparisons between the legacy path and the new blkmq path, but I just
>>>>>>>>>>>> couldn't get to that point because of the following errors.
>>>>>>>>>>>>
>>>>>>>>>>>> I am using a Kingston 4GB SDHC card, which is detected and mounted
>>>>>>>>>>>> nicely. However, when I decide to do some writes to the card I get the
>>>>>>>>>>>> following errors.
>>>>>>>>>>>>
>>>>>>>>>>>> root@ME:/mnt/sdcard dd if=/dev/zero of=testfile bs=8192 count=5000 conv=fsync
>>>>>>>>>>>> [  463.714294] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>>>> [  464.722656] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>>>> [  466.081481] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>>>> [  467.111236] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>>>> [  468.669647] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>>>> [  469.685699] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>>>> [  471.043334] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>>>> [  472.052337] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>>>> [  473.342651] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>>>> [  474.323760] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>>>> [  475.544769] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>>>> [  476.539031] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>>>> [  477.748474] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>>>> [  478.724182] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>>>>
>>>>>>>>>>>> I haven't yet got the point of investigating this any further, and
>>>>>>>>>>>> unfortunate I have a busy schedule with traveling next week. I will do
>>>>>>>>>>>> my best to look into this as soon as I can.
>>>>>>>>>>>>
>>>>>>>>>>>> Perhaps you have some ideas?
>>>>>>>>>>>
>>>>>>>>>>> The behaviour depends on whether you have MMC_CAP_WAIT_WHILE_BUSY. Try
>>>>>>>>>>> changing that and see if it makes a difference.
>>>>>>>>>>
>>>>>>>>>> Yes, it does! I disabled MMC_CAP_WAIT_WHILE_BUSY (and its
>>>>>>>>>> corresponding code in mmci.c) and the errors goes away.
>>>>>>>>>>
>>>>>>>>>> When I use MMC_CAP_WAIT_WHILE_BUSY I get these problems:
>>>>>>>>>>
>>>>>>>>>> [  223.820983] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>> [  224.815795] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>> [  226.034881] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>> [  227.112884] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>> [  227.220275] mmc0: Card stuck in wrong state! mmcblk0 mmc_blk_card_stuck
>>>>>>>>>> [  228.686798] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>> [  229.892150] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>> [  231.031890] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>> [  232.239013] mmci-pl18x 80126000.sdi0_per1: error during DMA transfer!
>>>>>>>>>> 5000+0 records in
>>>>>>>>>> 5000+0 records out
>>>>>>>>>> root@ME:/mnt/sdcard
>>>>>>>>>>
>>>>>>>>>> I looked at the new blkmq code from patch v10 13/15. It seems like the
>>>>>>>>>> MMC_CAP_WAIT_WHILE_BUSY is used to determine whether the async request
>>>>>>>>>> mechanism should be used or not. Perhaps I didn't looked close enough,
>>>>>>>>>> but maybe you could elaborate on why this seems to be the case!?
>>>>>>>>>
>>>>>>>>> MMC_CAP_WAIT_WHILE_BUSY is necessary because it means that a data transfer
>>>>>>>>> request has finished when the host controller calls mmc_request_done(). i.e.
>>>>>>>>> polling the card is not necessary.
>>>>>>>>
>>>>>>>> Well, that is a rather big change on its own. Earlier we polled with
>>>>>>>> CMD13 to verify that the card has moved back to the transfer state, in
>>>>>>>> case it was a write. And that was no matter of MMC_CAP_WAIT_WHILE_BUSY
>>>>>>>> was set or not. Right!?
>>>>>>>
>>>>>>> Yes
>>>>>>>
>>>>>>>>
>>>>>>>> I am not sure it's a good idea to bypass that validation, it seems
>>>>>>>> fragile to rely only on the busy detection on DAT line for writes.
>>>>>>>
>>>>>>> Can you cite something from the specifications that backs that up, because I
>>>>>>> couldn't find anything to suggest that CMD13 polling was expected.
>>>>>>
>>>>>> No I can't, but I don't see why that matters.
>>>>>>
>>>>>> My point is, if we want to go down that road by avoiding the CMD13
>>>>>> polling, that needs to be a separate change, which we can test and
>>>>>> confirm on its own.
>>>>>>
>>>>>>>
>>>>>>>>
>>>>>>>>>
>>>>>>>>> Have you tried V9 or V10.  There was a fix in V9 related to calling
>>>>>>>>> ->post_req() which could mess up DMA.
>>>>>>>>
>>>>>>>> I have used V10.
>>>>>>>>
>>>>>>>>>
>>>>>>>>> The other thing that could go wrong with DMA is if it cannot accept
>>>>>>>>> ->post_req() being called from mmc_request_done().
>>>>>>>>
>>>>>>>> I don't think mmci has a problem with that, however why do you want to
>>>>>>>> do this? Wouldn't that defeat some of the benefits with the async
>>>>>>>> request mechanism?
>>>>>>>
>>>>>>> Perhaps - but it would need to be tested.  If there are more requests
>>>>>>> waiting, one optimization could be to defer ->post_req() until after the
>>>>>>> next request is started.
>>>>>>
>>>>>> This is already proven, because this how the existing mmc async
>>>>>> request mechanism works.
>>>>>>
>>>>>> In ->post_req() callbacks, host drivers may do dma_unmap_sg(), which
>>>>>> is something that could be costly and therefore it's better to start a
>>>>>> new request before, such these things can go on in parallel.
>>>>>
>>>>> OK I will make a patch that takes care of both issues.  That will also mean
>>>>> the request is not completed in the ->done() callback because ->post_req()
>>>>> must precede block layer completion.
>>>>
>>>> Right.
>>>>
>>>> Actually completing the request in the ->done callback, may still be
>>>> possible, because in principle it only needs to inform the other
>>>> prepared request that it may start, before it continues to post
>>>> process/completes the current one.
>>>>
>>>> However, by looking at for example how mmci.c works, it actually holds
>>>> its spinlock while it calls mmc_request_done(). The same spinlock is
>>>> taken in the ->request() function, but not in the ->post_req()
>>>> function. In other words, completing the request in the ->done()
>>>> callback, would make mmci to keep the spinlock held throughout the
>>>> post processing cycle, which then prevents the next request from being
>>>> started.
>>>>
>>>> So my conclusion is, let's start a as you suggested, by not completing
>>>> the request in ->done() as to maintain existing behavior. Then we can
>>>> address optimizations on top, which very likely will involve doing
>>>> changes to host drivers as well.
>>>
>>> Have you tested the latest version now?
>>>
>>
>> Ping?
> 
> Still ping?

How is your silence in any way an acceptable way to execute your
responsibilities as maintainer!

Re: [PATCH V8 00/14] mmc: Add Command Queue support

[Ulf Hansson]

+ Bartlomiej

[...]

>>>>> So my conclusion is, let's start a as you suggested, by not completing
>>>>> the request in ->done() as to maintain existing behavior. Then we can
>>>>> address optimizations on top, which very likely will involve doing
>>>>> changes to host drivers as well.
>>>>
>>>> Have you tested the latest version now?
>>>>
>>>
>>> Ping?
>>
>> Still ping?
>
> How is your silence in any way an acceptable way to execute your
> responsibilities as maintainer!

Seriously? You posted the new version Oct 13.

I had to make some late minute travel decisions, so unfortunate I
won't be able to test this on HW from this Friday.

However, you have completely ignored mine, Linus and Bartlomiej's
comments about that we want the blkmq port being a separate patch(es)
and then make the CMDQ patches on top. This worries me, because it
seems like our messages don't reach you.

Kind regards
Uffe

Re: [PATCH V8 00/14] mmc: Add Command Queue support

[Adrian Hunter]

On 24/10/17 08:37, Ulf Hansson wrote:
> + Bartlomiej
> 
> [...]
> 
>>>>>> So my conclusion is, let's start a as you suggested, by not completing
>>>>>> the request in ->done() as to maintain existing behavior. Then we can
>>>>>> address optimizations on top, which very likely will involve doing
>>>>>> changes to host drivers as well.
>>>>>
>>>>> Have you tested the latest version now?
>>>>>
>>>>
>>>> Ping?
>>>
>>> Still ping?
>>
>> How is your silence in any way an acceptable way to execute your
>> responsibilities as maintainer!
> 
> Seriously? You posted the new version Oct 13.

When you constantly go silent it looks like you are deliberately delaying
the patches - which you have a long history of doing.  Are you deliberately
delaying again?

What are you plans for the patches?

Why do you think mmc work is so unimportant it can be left to whenever you
deign to get around to it?

> 
> I had to make some late minute travel decisions, so unfortunate I
> won't be able to test this on HW from this Friday.
> 
> However, you have completely ignored mine, Linus and Bartlomiej's
> comments about that we want the blkmq port being a separate patch(es)
> and then make the CMDQ patches on top. This worries me, because it
> seems like our messages don't reach you.

Rubbish!  I gave a very good reason for keeping the CQE code in - it is
designed to work together.  I also pointed out that it is trivial to see the
CQE code and that it is all '+' lines anyway.

But not one question in response!  Where is a single example of why it is
difficult like it is.  Where are the questions!  Not even a request for
documentation!  How I am supposed to know what you do or don't understand if
you don't ask any questions! There is no evidence that you guys have read a
single line!

So, what are your plans for the patches?  What don't you understand?

Re: [PATCH V8 00/14] mmc: Add Command Queue support

[Ulf Hansson]

[...]

>> However, you have completely ignored mine, Linus and Bartlomiej's
>> comments about that we want the blkmq port being a separate patch(es)
>> and then make the CMDQ patches on top. This worries me, because it
>> seems like our messages don't reach you.
>
> Rubbish!  I gave a very good reason for keeping the CQE code in - it is
> designed to work together.  I also pointed out that it is trivial to see the
> CQE code and that it is all '+' lines anyway.

You gave reasons, but none of us bought them.

>
> But not one question in response!  Where is a single example of why it is
> difficult like it is.  Where are the questions!  Not even a request for
> documentation!  How I am supposed to know what you do or don't understand if
> you don't ask any questions! There is no evidence that you guys have read a
> single line!

I have and I have also tested it, finding it not working. As reported.

However, I have also told you that I am having a *hard time* to review
it, because it implements both blkmq and CMDQ in the same patch to
code changes get complex.

>
> So, what are your plans for the patches?  What don't you understand?

I have told you this several time, so has Linus and Bartlomiej.

If you can split it up such the blkmq support comes first, then I can
review/test and pick it up.

Kind regards
Uffe

Re: [PATCH V8 00/14] mmc: Add Command Queue support

[Adrian Hunter]

On 24/10/17 10:39, Ulf Hansson wrote:
> [...]
> 
>>> However, you have completely ignored mine, Linus and Bartlomiej's
>>> comments about that we want the blkmq port being a separate patch(es)
>>> and then make the CMDQ patches on top. This worries me, because it
>>> seems like our messages don't reach you.
>>
>> Rubbish!  I gave a very good reason for keeping the CQE code in - it is
>> designed to work together.  I also pointed out that it is trivial to see the
>> CQE code and that it is all '+' lines anyway.
> 
> You gave reasons, but none of us bought them.
> 
>>
>> But not one question in response!  Where is a single example of why it is
>> difficult like it is.  Where are the questions!  Not even a request for
>> documentation!  How I am supposed to know what you do or don't understand if
>> you don't ask any questions! There is no evidence that you guys have read a
>> single line!
> 
> I have and I have also tested it, finding it not working. As reported.
> 
> However, I have also told you that I am having a *hard time* to review
> it, because it implements both blkmq and CMDQ in the same patch to
> code changes get complex.
> 
>>
>> So, what are your plans for the patches?  What don't you understand?
> 
> I have told you this several time, so has Linus and Bartlomiej.
> 
> If you can split it up such the blkmq support comes first, then I can
> review/test and pick it up.

I have done the split, but of course the code is just the same.