Merge tag 'mmc-v5.7' of git://git.kernel.org/pub/scm/linux/kernel/git/ulfh/mmc

Pull MMC updates from Ulf Hansson:
"MMC core:
- Add support for host software queue for (e)MMC/SD
- Throttle polling rate for CMD6
- Update CMD13 busy condition check for CMD6 commands
- Improve busy detect polling for erase/trim/discard/HPI
- Fixup support for HW busy detection for HPI commands
- Re-work and improve support for eMMC sanitize commands

MMC host:
- mmci:
* Add support for sdmmc variant revision 2.0
- mmci_sdmmc:
* Improve support for busyend detection
* Fixup support for signal voltage switch
* Add support for tuning with delay block
- mtk-sd:
* Fix another SDIO irq issue
- sdhci:
* Disable native card detect when GPIO based type exist
- sdhci:
* Add option to defer request completion
- sdhci_am654:
* Add support to set a tap value per speed mode
- sdhci-esdhc-imx:
* Add support for i.MX8MM based variant
* Fixup support for standard tuning on i.MX8 usdhc
* Optimize for strobe/clock dll settings
* Fixup support for system and runtime suspend/resume
- sdhci-iproc:
* Update regulator/bus-voltage management for bcm2711
- sdhci-msm:
* Prevent clock gating with PWRSAVE_DLL on broken variants
* Fix management of CQE during SDHCI reset
- sdhci-of-arasan:
* Add support for auto tuning on ZynqMP based platforms
- sdhci-omap:
* Add support for system suspend/resume
- sdhci-sprd:
* Add support for HW busy detection
* Enable support host software queue
- sdhci-tegra:
* Add support for HW busy detection
- tmio/renesas_sdhi:
* Enforce retune after runtime suspend
- renesas_sdhi:
* Use manual tap correction for HS400 on some variants
* Add support for manual correction of tap values for tunings"

* tag 'mmc-v5.7' of git://git.kernel.org/pub/scm/linux/kernel/git/ulfh/mmc: (86 commits)
mmc: cavium-octeon: remove nonsense variable coercion
mmc: mediatek: fix SDIO irq issue
mmc: mmci_sdmmc: Fix clear busyd0end irq flag
dt-bindings: mmc: Fix node name in an example
mmc: core: Re-work the code for eMMC sanitize
mmc: sdhci: use FIELD_GET for preset value bit masks
mmc: sdhci-of-at91: Display clock changes for debug purpose only
mmc: sdhci: iproc: Add custom set_power() callback for bcm2711
mmc: sdhci: am654: Use sdhci_set_power_and_voltage()
mmc: sdhci: at91: Use sdhci_set_power_and_voltage()
mmc: sdhci: milbeaut: Use sdhci_set_power_and_voltage()
mmc: sdhci: arasan: Use sdhci_set_power_and_voltage()
mmc: sdhci: Introduce sdhci_set_power_and_bus_voltage()
mmc: vub300: Use scnprintf() for avoiding potential buffer overflow
dt-bindings: mmc: synopsys-dw-mshc: fix clock-freq-min-max in example
sdhci: tegra: Enable MMC_CAP_WAIT_WHILE_BUSY host capability
sdhci: tegra: Implement Tegra specific set_timeout callback
mmc: sdhci-omap: Add Support for Suspend/Resume
mmc: renesas_sdhi: simplify execute_tuning
mmc: renesas_sdhi: Use BITS_PER_LONG helper
...

Documentation/devicetree/bindings/mmc/fsl-imx-esdhc.txt

···
   This property allows user to change the tuning step to more than one delay
   cells which is useful for some special boards or cards when the default
   tuning step can't find the proper delay window within limited tuning retries.
+- fsl,strobe-dll-delay-target: Specify the strobe dll control slave delay target.
+  This delay target programming host controller loopback read clock, and this
+  property allows user to change the delay target for the strobe input read clock.
+  If not use this property, driver default set the delay target to value 7.
+  Only eMMC HS400 mode need to take care of this property.
 
 Examples:
 

Documentation/devicetree/bindings/mmc/mmc-controller.yaml

···
 
 examples:
   - |
-    sdhci@ab000000 {
+    mmc@ab000000 {
         compatible = "sdhci";
         reg = <0xab000000 0x200>;
         interrupts = <23>;

Documentation/devicetree/bindings/mmc/mmci.txt

···
 - st,sig-pin-fbclk       : feedback clock signal pin used.
 
 specific for sdmmc variant:
+- reg			 : a second base register may be defined if a delay
+                           block is present and used for tuning.
 - st,sig-dir             : signal direction polarity used for cmd, dat0 dat123.
 - st,neg-edge            : data & command phase relation, generated on
                            sd clock falling edge.

Documentation/devicetree/bindings/mmc/sdhci-am654.txt

···
 	- clocks: Handles to the clock inputs.
 	- clock-names: Tuple including "clk_xin" and "clk_ahb"
 	- interrupts: Interrupt specifiers
-	- ti,otap-del-sel: Output Tap Delay select
+	Output tap delay for each speed mode:
+	- ti,otap-del-sel-legacy
+	- ti,otap-del-sel-mmc-hs
+	- ti,otap-del-sel-sd-hs
+	- ti,otap-del-sel-sdr12
+	- ti,otap-del-sel-sdr25
+	- ti,otap-del-sel-sdr50
+	- ti,otap-del-sel-sdr104
+	- ti,otap-del-sel-ddr50
+	- ti,otap-del-sel-ddr52
+	- ti,otap-del-sel-hs200
+	- ti,otap-del-sel-hs400
+	  These bindings must be provided otherwise the driver will disable the
+	  corresponding speed mode (i.e. all nodes must provide at least -legacy)
 
 Optional Properties (Required for ti,am654-sdhci-5.1 and ti,j721e-sdhci-8bit):
 	- ti,trm-icp: DLL trim select
···
 		interrupts = <GIC_SPI 136 IRQ_TYPE_LEVEL_HIGH>;
 		sdhci-caps-mask = <0x80000007 0x0>;
 		mmc-ddr-1_8v;
-		ti,otap-del-sel = <0x2>;
+		ti,otap-del-sel-legacy = <0x0>;
+		ti,otap-del-sel-mmc-hs = <0x0>;
+		ti,otap-del-sel-ddr52 = <0x5>;
+		ti,otap-del-sel-hs200 = <0x5>;
+		ti,otap-del-sel-hs400 = <0x0>;
 		ti,trm-icp = <0x8>;
 	};

Documentation/devicetree/bindings/mmc/sdhci-msm.txt

···
 
 - reg: Base address and length of the register in the following order:
 	- Host controller register map (required)
-	- SD Core register map (required for msm-v4 and below)
+	- SD Core register map (required for controllers earlier than msm-v5)
+	- CQE register map (Optional, CQE support is present on SDHC instance meant
+	                    for eMMC and version v4.2 and above)
+- reg-names: When CQE register map is supplied, below reg-names are required
+	- "hc" for Host controller register map
+	- "core" for SD core register map
+	- "cqhci" for CQE register map
 - interrupts: Should contain an interrupt-specifiers for the interrupts:
 	- Host controller interrupt (required)
 - pinctrl-names: Should contain only one value - "default".

Documentation/devicetree/bindings/mmc/synopsys-dw-mshc.yaml

···
       cap-mmc-highspeed;
       cap-sd-highspeed;
       card-detect-delay = <200>;
-      clock-freq-min-max = <400000 200000000>;
+      max-frequency = <200000000>;
       clock-frequency = <400000000>;
       data-addr = <0x200>;
       fifo-depth = <0x80>;

drivers/firmware/xilinx/zynqmp.c

···
 static inline int zynqmp_is_valid_ioctl(u32 ioctl_id)
 {
 	switch (ioctl_id) {
+	case IOCTL_SD_DLL_RESET:
+	case IOCTL_SET_SD_TAPDELAY:
 	case IOCTL_SET_PLL_FRAC_MODE:
 	case IOCTL_GET_PLL_FRAC_MODE:
 	case IOCTL_SET_PLL_FRAC_DATA:

drivers/mmc/core/block.c

···
  * ample.
  */
 #define MMC_BLK_TIMEOUT_MS  (10 * 1000)
-#define MMC_SANITIZE_REQ_TIMEOUT 240000
 #define MMC_EXTRACT_INDEX_FROM_ARG(x) ((x & 0x00FF0000) >> 16)
 #define MMC_EXTRACT_VALUE_FROM_ARG(x) ((x & 0x0000FF00) >> 8)
 
···
 
 static inline int mmc_blk_part_switch(struct mmc_card *card,
 				      unsigned int part_type);
+static void mmc_blk_rw_rq_prep(struct mmc_queue_req *mqrq,
+			       struct mmc_card *card,
+			       int disable_multi,
+			       struct mmc_queue *mq);
+static void mmc_blk_hsq_req_done(struct mmc_request *mrq);
 
 static struct mmc_blk_data *mmc_blk_get(struct gendisk *disk)
 {
···
 	return 0;
 }
 
-static int ioctl_do_sanitize(struct mmc_card *card)
-{
-	int err;
-
-	if (!mmc_can_sanitize(card)) {
-			pr_warn("%s: %s - SANITIZE is not supported\n",
-				mmc_hostname(card->host), __func__);
-			err = -EOPNOTSUPP;
-			goto out;
-	}
-
-	pr_debug("%s: %s - SANITIZE IN PROGRESS...\n",
-		mmc_hostname(card->host), __func__);
-
-	err = mmc_switch(card, EXT_CSD_CMD_SET_NORMAL,
-					EXT_CSD_SANITIZE_START, 1,
-					MMC_SANITIZE_REQ_TIMEOUT);
-
-	if (err)
-		pr_err("%s: %s - EXT_CSD_SANITIZE_START failed. err=%d\n",
-		       mmc_hostname(card->host), __func__, err);
-
-	pr_debug("%s: %s - SANITIZE COMPLETED\n", mmc_hostname(card->host),
-					     __func__);
-out:
-	return err;
-}
-
-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);
-}
-
 static int card_busy_detect(struct mmc_card *card, unsigned int timeout_ms,
 			    u32 *resp_errs)
 {
···
 				 __func__, status);
 			return -ETIMEDOUT;
 		}
-
-		/*
-		 * Some cards mishandle the status bits,
-		 * so make sure to check both the busy
-		 * indication and the card state.
-		 */
-	} while (!mmc_blk_in_tran_state(status));
+	} while (!mmc_ready_for_data(status));
 
 	return err;
 }
···
 	}
 
 	if ((MMC_EXTRACT_INDEX_FROM_ARG(cmd.arg) == EXT_CSD_SANITIZE_START) &&
-	    (cmd.opcode == MMC_SWITCH)) {
-		err = ioctl_do_sanitize(card);
-
-		if (err)
-			pr_err("%s: ioctl_do_sanitize() failed. err = %d",
-			       __func__, err);
-
-		return err;
-	}
+	    (cmd.opcode == MMC_SWITCH))
+		return mmc_sanitize(card);
 
 	mmc_wait_for_req(card->host, &mrq);
 
···
 	return mmc_blk_cqe_start_req(mq->card->host, mrq);
 }
 
+static int mmc_blk_hsq_issue_rw_rq(struct mmc_queue *mq, struct request *req)
+{
+	struct mmc_queue_req *mqrq = req_to_mmc_queue_req(req);
+	struct mmc_host *host = mq->card->host;
+	int err;
+
+	mmc_blk_rw_rq_prep(mqrq, mq->card, 0, mq);
+	mqrq->brq.mrq.done = mmc_blk_hsq_req_done;
+	mmc_pre_req(host, &mqrq->brq.mrq);
+
+	err = mmc_cqe_start_req(host, &mqrq->brq.mrq);
+	if (err)
+		mmc_post_req(host, &mqrq->brq.mrq, err);
+
+	return err;
+}
+
 static int mmc_blk_cqe_issue_rw_rq(struct mmc_queue *mq, struct request *req)
 {
 	struct mmc_queue_req *mqrq = req_to_mmc_queue_req(req);
+	struct mmc_host *host = mq->card->host;
+
+	if (host->hsq_enabled)
+		return mmc_blk_hsq_issue_rw_rq(mq, req);
 
 	mmc_blk_data_prep(mq, mqrq, 0, NULL, NULL);
 
···
 			goto error_exit;
 
 		if (!mmc_host_is_spi(host) &&
-		    !mmc_blk_in_tran_state(status)) {
+		    !mmc_ready_for_data(status)) {
 			err = mmc_blk_fix_state(card, req);
 			if (err)
 				goto error_exit;
···
 	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));
+	       (rq_data_dir(req) == WRITE && !mmc_ready_for_data(status));
 }
 
 static inline bool mmc_blk_cmd_started(struct mmc_blk_request *brq)
···
 
 	/* Try to get back to "tran" state */
 	if (!mmc_host_is_spi(mq->card->host) &&
-	    (err || !mmc_blk_in_tran_state(status)))
+	    (err || !mmc_ready_for_data(status)))
 		err = mmc_blk_fix_state(mq->card, req);
 
 	/*
···
 {
 	if (mmc_blk_urgent_bkops_needed(mq, mqrq))
 		mmc_run_bkops(mq->card);
+}
+
+static void mmc_blk_hsq_req_done(struct mmc_request *mrq)
+{
+	struct mmc_queue_req *mqrq =
+		container_of(mrq, struct mmc_queue_req, brq.mrq);
+	struct request *req = mmc_queue_req_to_req(mqrq);
+	struct request_queue *q = req->q;
+	struct mmc_queue *mq = q->queuedata;
+	struct mmc_host *host = mq->card->host;
+	unsigned long flags;
+
+	if (mmc_blk_rq_error(&mqrq->brq) ||
+	    mmc_blk_urgent_bkops_needed(mq, mqrq)) {
+		spin_lock_irqsave(&mq->lock, flags);
+		mq->recovery_needed = true;
+		mq->recovery_req = req;
+		spin_unlock_irqrestore(&mq->lock, flags);
+
+		host->cqe_ops->cqe_recovery_start(host);
+
+		schedule_work(&mq->recovery_work);
+		return;
+	}
+
+	mmc_blk_rw_reset_success(mq, req);
+
+	/*
+	 * Block layer timeouts race with completions which means the normal
+	 * completion path cannot be used during recovery.
+	 */
+	if (mq->in_recovery)
+		mmc_blk_cqe_complete_rq(mq, req);
+	else
+		blk_mq_complete_request(req);
 }
 
 void mmc_blk_mq_complete(struct request *req)

drivers/mmc/core/core.c

···
 
 		cmd = mrq->cmd;
 
-		/*
-		 * If host has timed out waiting for the sanitize
-		 * to complete, card might be still in programming state
-		 * so let's try to bring the card out of programming
-		 * state.
-		 */
-		if (cmd->sanitize_busy && cmd->error == -ETIMEDOUT) {
-			if (!mmc_interrupt_hpi(host->card)) {
-				pr_warn("%s: %s: Interrupted sanitize\n",
-					mmc_hostname(host), __func__);
-				cmd->error = 0;
-				break;
-			} else {
-				pr_err("%s: %s: Failed to interrupt sanitize\n",
-				       mmc_hostname(host), __func__);
-			}
-		}
 		if (!cmd->error || !cmd->retries ||
 		    mmc_card_removed(host->card))
 			break;
···
 	struct mmc_command cmd = {};
 	unsigned int qty = 0, busy_timeout = 0;
 	bool use_r1b_resp = false;
-	unsigned long timeout;
-	int loop_udelay=64, udelay_max=32768;
 	int err;
 
 	mmc_retune_hold(card->host);
···
 	if ((card->host->caps & MMC_CAP_WAIT_WHILE_BUSY) && use_r1b_resp)
 		goto out;
 
-	timeout = jiffies + msecs_to_jiffies(busy_timeout);
-	do {
-		memset(&cmd, 0, sizeof(struct mmc_command));
-		cmd.opcode = MMC_SEND_STATUS;
-		cmd.arg = card->rca << 16;
-		cmd.flags = MMC_RSP_R1 | MMC_CMD_AC;
-		/* Do not retry else we can't see errors */
-		err = mmc_wait_for_cmd(card->host, &cmd, 0);
-		if (err || R1_STATUS(cmd.resp[0])) {
-			pr_err("error %d requesting status %#x\n",
-				err, cmd.resp[0]);
-			err = -EIO;
-			goto out;
-		}
-
-		/* Timeout if the device never becomes ready for data and
-		 * never leaves the program state.
-		 */
-		if (time_after(jiffies, timeout)) {
-			pr_err("%s: Card stuck in programming state! %s\n",
-				mmc_hostname(card->host), __func__);
-			err =  -EIO;
-			goto out;
-		}
-		if ((cmd.resp[0] & R1_READY_FOR_DATA) &&
-		    R1_CURRENT_STATE(cmd.resp[0]) != R1_STATE_PRG)
-			break;
-
-		usleep_range(loop_udelay, loop_udelay*2);
-		if (loop_udelay < udelay_max)
-			loop_udelay *= 2;
-	} while (1);
+	/* Let's poll to find out when the erase operation completes. */
+	err = mmc_poll_for_busy(card, busy_timeout, MMC_BUSY_ERASE);
 
 out:
 	mmc_retune_release(card->host);
···
 		return 1;
 	return 0;
 }
-EXPORT_SYMBOL(mmc_can_sanitize);
 
 int mmc_can_secure_erase_trim(struct mmc_card *card)
 {

drivers/mmc/core/mmc.c

···
 	err = __mmc_switch(card, EXT_CSD_CMD_SET_NORMAL,
 			   EXT_CSD_HS_TIMING, EXT_CSD_TIMING_HS,
 			   card->ext_csd.generic_cmd6_time, MMC_TIMING_MMC_HS,
-			   true, true, true);
+			   true, true);
 	if (err)
 		pr_warn("%s: switch to high-speed failed, err:%d\n",
 			mmc_hostname(card->host), err);
···
 			   ext_csd_bits,
 			   card->ext_csd.generic_cmd6_time,
 			   MMC_TIMING_MMC_DDR52,
-			   true, true, true);
+			   true, true);
 	if (err) {
 		pr_err("%s: switch to bus width %d ddr failed\n",
 			mmc_hostname(host), 1 << bus_width);
···
 	err = __mmc_switch(card, EXT_CSD_CMD_SET_NORMAL,
 			   EXT_CSD_HS_TIMING, val,
 			   card->ext_csd.generic_cmd6_time, 0,
-			   true, false, true);
+			   false, true);
 	if (err) {
 		pr_err("%s: switch to high-speed from hs200 failed, err:%d\n",
 			mmc_hostname(host), err);
···
 	max_dtr = card->ext_csd.hs_max_dtr;
 	mmc_set_clock(host, max_dtr);
 
-	err = mmc_switch_status(card);
+	err = mmc_switch_status(card, true);
 	if (err)
 		goto out_err;
 
···
 	err = __mmc_switch(card, EXT_CSD_CMD_SET_NORMAL,
 			   EXT_CSD_HS_TIMING, val,
 			   card->ext_csd.generic_cmd6_time, 0,
-			   true, false, true);
+			   false, true);
 	if (err) {
 		pr_err("%s: switch to hs400 failed, err:%d\n",
 			 mmc_hostname(host), err);
···
 	if (host->ops->hs400_complete)
 		host->ops->hs400_complete(host);
 
-	err = mmc_switch_status(card);
+	err = mmc_switch_status(card, true);
 	if (err)
 		goto out_err;
 
···
 	val = EXT_CSD_TIMING_HS;
 	err = __mmc_switch(card, EXT_CSD_CMD_SET_NORMAL, EXT_CSD_HS_TIMING,
 			   val, card->ext_csd.generic_cmd6_time, 0,
-			   true, false, true);
+			   false, true);
 	if (err)
 		goto out_err;
 
 	mmc_set_timing(host, MMC_TIMING_MMC_DDR52);
 
-	err = mmc_switch_status(card);
+	err = mmc_switch_status(card, true);
 	if (err)
 		goto out_err;
 
 	/* Switch HS DDR to HS */
 	err = __mmc_switch(card, EXT_CSD_CMD_SET_NORMAL, EXT_CSD_BUS_WIDTH,
 			   EXT_CSD_BUS_WIDTH_8, card->ext_csd.generic_cmd6_time,
-			   0, true, false, true);
+			   0, false, true);
 	if (err)
 		goto out_err;
 
···
 	if (host->ops->hs400_downgrade)
 		host->ops->hs400_downgrade(host);
 
-	err = mmc_switch_status(card);
+	err = mmc_switch_status(card, true);
 	if (err)
 		goto out_err;
 
···
 	      card->drive_strength << EXT_CSD_DRV_STR_SHIFT;
 	err = __mmc_switch(card, EXT_CSD_CMD_SET_NORMAL, EXT_CSD_HS_TIMING,
 			   val, card->ext_csd.generic_cmd6_time, 0,
-			   true, false, true);
+			   false, true);
 	if (err)
 		goto out_err;
 
···
 	 * failed. If there really is a problem, we would expect tuning will
 	 * fail and the result ends up the same.
 	 */
-	err = __mmc_switch_status(card, false);
+	err = mmc_switch_status(card, false);
 	if (err)
 		goto out_err;
 
···
 	err = __mmc_switch(card, EXT_CSD_CMD_SET_NORMAL,
 			   EXT_CSD_HS_TIMING, EXT_CSD_TIMING_HS,
 			   card->ext_csd.generic_cmd6_time, 0,
-			   true, false, true);
+			   false, true);
 	if (err) {
 		pr_err("%s: switch to hs for hs400es failed, err:%d\n",
 			mmc_hostname(host), err);
···
 	}
 
 	mmc_set_timing(host, MMC_TIMING_MMC_HS);
-	err = mmc_switch_status(card);
+	err = mmc_switch_status(card, true);
 	if (err)
 		goto out_err;
 
···
 	err = __mmc_switch(card, EXT_CSD_CMD_SET_NORMAL,
 			   EXT_CSD_HS_TIMING, val,
 			   card->ext_csd.generic_cmd6_time, 0,
-			   true, false, true);
+			   false, true);
 	if (err) {
 		pr_err("%s: switch to hs400es failed, err:%d\n",
 			mmc_hostname(host), err);
···
 	if (host->ops->hs400_enhanced_strobe)
 		host->ops->hs400_enhanced_strobe(host, &host->ios);
 
-	err = mmc_switch_status(card);
+	err = mmc_switch_status(card, true);
 	if (err)
 		goto out_err;
 
···
 		err = __mmc_switch(card, EXT_CSD_CMD_SET_NORMAL,
 				   EXT_CSD_HS_TIMING, val,
 				   card->ext_csd.generic_cmd6_time, 0,
-				   true, false, true);
+				   false, true);
 		if (err)
 			goto err;
 		old_timing = host->ios.timing;
···
 		 * switch failed. If there really is a problem, we would expect
 		 * tuning will fail and the result ends up the same.
 		 */
-		err = __mmc_switch_status(card, false);
+		err = mmc_switch_status(card, false);
 
 		/*
 		 * mmc_select_timing() assumes timing has not changed if
···
 	 */
 	card->reenable_cmdq = card->ext_csd.cmdq_en;
 
-	if (card->ext_csd.cmdq_en && !host->cqe_enabled) {
+	if (host->cqe_ops && !host->cqe_enabled) {
 		err = host->cqe_ops->cqe_enable(host, card);
-		if (err) {
-			pr_err("%s: Failed to enable CQE, error %d\n",
-				mmc_hostname(host), err);
-		} else {
+		if (!err) {
 			host->cqe_enabled = true;
-			pr_info("%s: Command Queue Engine enabled\n",
-				mmc_hostname(host));
+
+			if (card->ext_csd.cmdq_en) {
+				pr_info("%s: Command Queue Engine enabled\n",
+					mmc_hostname(host));
+			} else {
+				host->hsq_enabled = true;
+				pr_info("%s: Host Software Queue enabled\n",
+					mmc_hostname(host));
+			}
 		}
 	}
 
···
 
 	err = __mmc_switch(card, EXT_CSD_CMD_SET_NORMAL,
 			EXT_CSD_POWER_OFF_NOTIFICATION,
-			notify_type, timeout, 0, true, false, false);
+			notify_type, timeout, 0, false, false);
 	if (err)
 		pr_err("%s: Power Off Notification timed out, %u\n",
 		       mmc_hostname(card->host), timeout);

drivers/mmc/core/mmc_ops.c

···
 #include "host.h"
 #include "mmc_ops.h"
 
-#define MMC_OPS_TIMEOUT_MS		(10 * 60 * 1000) /* 10min*/
 #define MMC_BKOPS_TIMEOUT_MS		(120 * 1000) /* 120s */
 #define MMC_CACHE_FLUSH_TIMEOUT_MS	(30 * 1000) /* 30s */
+#define MMC_SANITIZE_TIMEOUT_MS		(240 * 1000) /* 240s */
 
 static const u8 tuning_blk_pattern_4bit[] = {
 	0xff, 0x0f, 0xff, 0x00, 0xff, 0xcc, 0xc3, 0xcc,
···
 }
 
 /* Caller must hold re-tuning */
-int __mmc_switch_status(struct mmc_card *card, bool crc_err_fatal)
+int mmc_switch_status(struct mmc_card *card, bool crc_err_fatal)
 {
 	u32 status;
 	int err;
···
 	return mmc_switch_status_error(card->host, status);
 }
 
-int mmc_switch_status(struct mmc_card *card)
+static int mmc_busy_status(struct mmc_card *card, bool retry_crc_err,
+			   enum mmc_busy_cmd busy_cmd, bool *busy)
 {
-	return __mmc_switch_status(card, true);
+	struct mmc_host *host = card->host;
+	u32 status = 0;
+	int err;
+
+	if (host->ops->card_busy) {
+		*busy = host->ops->card_busy(host);
+		return 0;
+	}
+
+	err = mmc_send_status(card, &status);
+	if (retry_crc_err && err == -EILSEQ) {
+		*busy = true;
+		return 0;
+	}
+	if (err)
+		return err;
+
+	switch (busy_cmd) {
+	case MMC_BUSY_CMD6:
+		err = mmc_switch_status_error(card->host, status);
+		break;
+	case MMC_BUSY_ERASE:
+		err = R1_STATUS(status) ? -EIO : 0;
+		break;
+	case MMC_BUSY_HPI:
+		break;
+	default:
+		err = -EINVAL;
+	}
+
+	if (err)
+		return err;
+
+	*busy = !mmc_ready_for_data(status);
+	return 0;
 }
 
-static int mmc_poll_for_busy(struct mmc_card *card, unsigned int timeout_ms,
-			bool send_status, bool retry_crc_err)
+static int __mmc_poll_for_busy(struct mmc_card *card, unsigned int timeout_ms,
+			       bool send_status, bool retry_crc_err,
+			       enum mmc_busy_cmd busy_cmd)
 {
 	struct mmc_host *host = card->host;
 	int err;
 	unsigned long timeout;
-	u32 status = 0;
+	unsigned int udelay = 32, udelay_max = 32768;
 	bool expired = false;
 	bool busy = false;
 
···
 		 */
 		expired = time_after(jiffies, timeout);
 
-		if (host->ops->card_busy) {
-			busy = host->ops->card_busy(host);
-		} else {
-			err = mmc_send_status(card, &status);
-			if (retry_crc_err && err == -EILSEQ) {
-				busy = true;
-			} else if (err) {
-				return err;
-			} else {
-				err = mmc_switch_status_error(host, status);
-				if (err)
-					return err;
-				busy = R1_CURRENT_STATE(status) == R1_STATE_PRG;
-			}
-		}
+		err = mmc_busy_status(card, retry_crc_err, busy_cmd, &busy);
+		if (err)
+			return err;
 
 		/* Timeout if the device still remains busy. */
 		if (expired && busy) {
···
 				mmc_hostname(host), __func__);
 			return -ETIMEDOUT;
 		}
+
+		/* Throttle the polling rate to avoid hogging the CPU. */
+		if (busy) {
+			usleep_range(udelay, udelay * 2);
+			if (udelay < udelay_max)
+				udelay *= 2;
+		}
 	} while (busy);
 
 	return 0;
+}
+
+int mmc_poll_for_busy(struct mmc_card *card, unsigned int timeout_ms,
+		      enum mmc_busy_cmd busy_cmd)
+{
+	return __mmc_poll_for_busy(card, timeout_ms, true, false, busy_cmd);
 }
 
 /**
···
  *	@timeout_ms: timeout (ms) for operation performed by register write,
  *                   timeout of zero implies maximum possible timeout
  *	@timing: new timing to change to
- *	@use_busy_signal: use the busy signal as response type
  *	@send_status: send status cmd to poll for busy
  *	@retry_crc_err: retry when CRC errors when polling with CMD13 for busy
  *
···
  */
 int __mmc_switch(struct mmc_card *card, u8 set, u8 index, u8 value,
 		unsigned int timeout_ms, unsigned char timing,
-		bool use_busy_signal, bool send_status,	bool retry_crc_err)
+		bool send_status, bool retry_crc_err)
 {
 	struct mmc_host *host = card->host;
 	int err;
 	struct mmc_command cmd = {};
-	bool use_r1b_resp = use_busy_signal;
+	bool use_r1b_resp = true;
 	unsigned char old_timing = host->ios.timing;
 
 	mmc_retune_hold(host);
···
 		cmd.flags |= MMC_RSP_SPI_R1 | MMC_RSP_R1;
 	}
 
-	if (index == EXT_CSD_SANITIZE_START)
-		cmd.sanitize_busy = true;
-
 	err = mmc_wait_for_cmd(host, &cmd, MMC_CMD_RETRIES);
 	if (err)
-		goto out;
-
-	/* No need to check card status in case of unblocking command */
-	if (!use_busy_signal)
 		goto out;
 
 	/*If SPI or used HW busy detection above, then we don't need to poll. */
···
 		goto out_tim;
 
 	/* Let's try to poll to find out when the command is completed. */
-	err = mmc_poll_for_busy(card, timeout_ms, send_status, retry_crc_err);
+	err = __mmc_poll_for_busy(card, timeout_ms, send_status, retry_crc_err,
+				  MMC_BUSY_CMD6);
 	if (err)
 		goto out;
 
···
 		mmc_set_timing(host, timing);
 
 	if (send_status) {
-		err = mmc_switch_status(card);
+		err = mmc_switch_status(card, true);
 		if (err && timing)
 			mmc_set_timing(host, old_timing);
 	}
···
 		unsigned int timeout_ms)
 {
 	return __mmc_switch(card, set, index, value, timeout_ms, 0,
-			true, true, false);
+			    true, false);
 }
 EXPORT_SYMBOL_GPL(mmc_switch);
 
···
 	return mmc_send_bus_test(card, card->host, MMC_BUS_TEST_R, width);
 }
 
-static int mmc_send_hpi_cmd(struct mmc_card *card, u32 *status)
+static int mmc_send_hpi_cmd(struct mmc_card *card)
 {
+	unsigned int busy_timeout_ms = card->ext_csd.out_of_int_time;
+	struct mmc_host *host = card->host;
+	bool use_r1b_resp = true;
 	struct mmc_command cmd = {};
-	unsigned int opcode;
 	int err;
 
-	opcode = card->ext_csd.hpi_cmd;
-	if (opcode == MMC_STOP_TRANSMISSION)
-		cmd.flags = MMC_RSP_R1B | MMC_CMD_AC;
-	else if (opcode == MMC_SEND_STATUS)
-		cmd.flags = MMC_RSP_R1 | MMC_CMD_AC;
-
-	cmd.opcode = opcode;
+	cmd.opcode = card->ext_csd.hpi_cmd;
 	cmd.arg = card->rca << 16 | 1;
 
-	err = mmc_wait_for_cmd(card->host, &cmd, 0);
+	/*
+	 * Make sure the host's max_busy_timeout fit the needed timeout for HPI.
+	 * In case it doesn't, let's instruct the host to avoid HW busy
+	 * detection, by using a R1 response instead of R1B.
+	 */
+	if (host->max_busy_timeout && busy_timeout_ms > host->max_busy_timeout)
+		use_r1b_resp = false;
+
+	if (cmd.opcode == MMC_STOP_TRANSMISSION && use_r1b_resp) {
+		cmd.flags = MMC_RSP_R1B | MMC_CMD_AC;
+		cmd.busy_timeout = busy_timeout_ms;
+	} else {
+		cmd.flags = MMC_RSP_R1 | MMC_CMD_AC;
+		use_r1b_resp = false;
+	}
+
+	err = mmc_wait_for_cmd(host, &cmd, 0);
 	if (err) {
-		pr_warn("%s: error %d interrupting operation. "
-			"HPI command response %#x\n", mmc_hostname(card->host),
-			err, cmd.resp[0]);
+		pr_warn("%s: HPI error %d. Command response %#x\n",
+			mmc_hostname(host), err, cmd.resp[0]);
 		return err;
 	}
-	if (status)
-		*status = cmd.resp[0];
 
-	return 0;
+	/* No need to poll when using HW busy detection. */
+	if (host->caps & MMC_CAP_WAIT_WHILE_BUSY && use_r1b_resp)
+		return 0;
+
+	/* Let's poll to find out when the HPI request completes. */
+	return mmc_poll_for_busy(card, busy_timeout_ms, MMC_BUSY_HPI);
 }
 
 /**
···
 {
 	int err;
 	u32 status;
-	unsigned long prg_wait;
 
 	if (!card->ext_csd.hpi_en) {
 		pr_info("%s: HPI enable bit unset\n", mmc_hostname(card->host));
···
 		goto out;
 	}
 
-	err = mmc_send_hpi_cmd(card, &status);
-	if (err)
-		goto out;
-
-	prg_wait = jiffies + msecs_to_jiffies(card->ext_csd.out_of_int_time);
-	do {
-		err = mmc_send_status(card, &status);
-
-		if (!err && R1_CURRENT_STATE(status) == R1_STATE_TRAN)
-			break;
-		if (time_after(jiffies, prg_wait))
-			err = -ETIMEDOUT;
-	} while (!err);
-
+	err = mmc_send_hpi_cmd(card);
 out:
 	return err;
 }
···
 	return mmc_cmdq_switch(card, false);
 }
 EXPORT_SYMBOL_GPL(mmc_cmdq_disable);
+
+int mmc_sanitize(struct mmc_card *card)
+{
+	struct mmc_host *host = card->host;
+	int err;
+
+	if (!mmc_can_sanitize(card)) {
+		pr_warn("%s: Sanitize not supported\n", mmc_hostname(host));
+		return -EOPNOTSUPP;
+	}
+
+	pr_debug("%s: Sanitize in progress...\n", mmc_hostname(host));
+
+	mmc_retune_hold(host);
+
+	err = mmc_switch(card, EXT_CSD_CMD_SET_NORMAL, EXT_CSD_SANITIZE_START,
+			 1, MMC_SANITIZE_TIMEOUT_MS);
+	if (err)
+		pr_err("%s: Sanitize failed err=%d\n", mmc_hostname(host), err);
+
+	/*
+	 * If the sanitize operation timed out, the card is probably still busy
+	 * in the R1_STATE_PRG. Rather than continue to wait, let's try to abort
+	 * it with a HPI command to get back into R1_STATE_TRAN.
+	 */
+	if (err == -ETIMEDOUT && !mmc_interrupt_hpi(card))
+		pr_warn("%s: Sanitize aborted\n", mmc_hostname(host));
+
+	mmc_retune_release(host);
+
+	pr_debug("%s: Sanitize completed\n", mmc_hostname(host));
+	return err;
+}
+EXPORT_SYMBOL_GPL(mmc_sanitize);

drivers/mmc/core/mmc_ops.h

···
 
 #include <linux/types.h>
 
+enum mmc_busy_cmd {
+	MMC_BUSY_CMD6,
+	MMC_BUSY_ERASE,
+	MMC_BUSY_HPI,
+};
+
 struct mmc_host;
 struct mmc_card;
 
···
 int mmc_spi_read_ocr(struct mmc_host *host, int highcap, u32 *ocrp);
 int mmc_spi_set_crc(struct mmc_host *host, int use_crc);
 int mmc_bus_test(struct mmc_card *card, u8 bus_width);
-int mmc_interrupt_hpi(struct mmc_card *card);
 int mmc_can_ext_csd(struct mmc_card *card);
 int mmc_get_ext_csd(struct mmc_card *card, u8 **new_ext_csd);
-int mmc_switch_status(struct mmc_card *card);
-int __mmc_switch_status(struct mmc_card *card, bool crc_err_fatal);
+int mmc_switch_status(struct mmc_card *card, bool crc_err_fatal);
+int mmc_poll_for_busy(struct mmc_card *card, unsigned int timeout_ms,
+		      enum mmc_busy_cmd busy_cmd);
 int __mmc_switch(struct mmc_card *card, u8 set, u8 index, u8 value,
 		unsigned int timeout_ms, unsigned char timing,
-		bool use_busy_signal, bool send_status,	bool retry_crc_err);
+		bool send_status, bool retry_crc_err);
 int mmc_switch(struct mmc_card *card, u8 set, u8 index, u8 value,
 		unsigned int timeout_ms);
 void mmc_run_bkops(struct mmc_card *card);
 int mmc_flush_cache(struct mmc_card *card);
 int mmc_cmdq_enable(struct mmc_card *card);
 int mmc_cmdq_disable(struct mmc_card *card);
+int mmc_sanitize(struct mmc_card *card);
 
 #endif
 

drivers/mmc/core/mmc_test.c

···
  * @sg_len: length of currently mapped scatterlist @sg
  * @mem: allocated memory
  * @sg: scatterlist
+ * @sg_areq: scatterlist for non-blocking request
  */
 struct mmc_test_area {
 	unsigned long max_sz;
···
 	unsigned int sg_len;
 	struct mmc_test_mem *mem;
 	struct scatterlist *sg;
+	struct scatterlist *sg_areq;
 };
 
 /**
···
 }
 
 static int mmc_test_nonblock_transfer(struct mmc_test_card *test,
-				      struct scatterlist *sg, unsigned sg_len,
-				      unsigned dev_addr, unsigned blocks,
-				      unsigned blksz, int write, int count)
+				      unsigned int dev_addr, int write,
+				      int count)
 {
 	struct mmc_test_req *rq1, *rq2;
 	struct mmc_request *mrq, *prev_mrq;
 	int i;
 	int ret = RESULT_OK;
+	struct mmc_test_area *t = &test->area;
+	struct scatterlist *sg = t->sg;
+	struct scatterlist *sg_areq = t->sg_areq;
 
 	rq1 = mmc_test_req_alloc();
 	rq2 = mmc_test_req_alloc();
···
 
 	for (i = 0; i < count; i++) {
 		mmc_test_req_reset(container_of(mrq, struct mmc_test_req, mrq));
-		mmc_test_prepare_mrq(test, mrq, sg, sg_len, dev_addr, blocks,
-				     blksz, write);
+		mmc_test_prepare_mrq(test, mrq, sg, t->sg_len, dev_addr,
+				     t->blocks, 512, write);
 		ret = mmc_test_start_areq(test, mrq, prev_mrq);
 		if (ret)
 			goto err;
···
 			prev_mrq = &rq2->mrq;
 
 		swap(mrq, prev_mrq);
-		dev_addr += blocks;
+		swap(sg, sg_areq);
+		dev_addr += t->blocks;
 	}
 
 	ret = mmc_test_start_areq(test, NULL, prev_mrq);
···
  * Map sz bytes so that it can be transferred.
  */
 static int mmc_test_area_map(struct mmc_test_card *test, unsigned long sz,
-			     int max_scatter, int min_sg_len)
+			     int max_scatter, int min_sg_len, bool nonblock)
 {
 	struct mmc_test_area *t = &test->area;
 	int err;
+	unsigned int sg_len = 0;
 
 	t->blocks = sz >> 9;
 
···
 		err = mmc_test_map_sg(t->mem, sz, t->sg, 1, t->max_segs,
 				      t->max_seg_sz, &t->sg_len, min_sg_len);
 	}
+
+	if (err || !nonblock)
+		goto err;
+
+	if (max_scatter) {
+		err = mmc_test_map_sg_max_scatter(t->mem, sz, t->sg_areq,
+						  t->max_segs, t->max_seg_sz,
+						  &sg_len);
+	} else {
+		err = mmc_test_map_sg(t->mem, sz, t->sg_areq, 1, t->max_segs,
+				      t->max_seg_sz, &sg_len, min_sg_len);
+	}
+	if (!err && sg_len != t->sg_len)
+		err = -EINVAL;
+
+err:
 	if (err)
 		pr_info("%s: Failed to map sg list\n",
 		       mmc_hostname(test->card->host));
···
 	struct timespec64 ts1, ts2;
 	int ret = 0;
 	int i;
-	struct mmc_test_area *t = &test->area;
 
 	/*
 	 * In the case of a maximally scattered transfer, the maximum transfer
···
 			sz = max_tfr;
 	}
 
-	ret = mmc_test_area_map(test, sz, max_scatter, min_sg_len);
+	ret = mmc_test_area_map(test, sz, max_scatter, min_sg_len, nonblock);
 	if (ret)
 		return ret;
 
 	if (timed)
 		ktime_get_ts64(&ts1);
 	if (nonblock)
-		ret = mmc_test_nonblock_transfer(test, t->sg, t->sg_len,
-				 dev_addr, t->blocks, 512, write, count);
+		ret = mmc_test_nonblock_transfer(test, dev_addr, write, count);
 	else
 		for (i = 0; i < count && ret == 0; i++) {
 			ret = mmc_test_area_transfer(test, dev_addr, write);
···
 	struct mmc_test_area *t = &test->area;
 
 	kfree(t->sg);
+	kfree(t->sg_areq);
 	mmc_test_free_mem(t->mem);
 
 	return 0;
···
 
 	t->sg = kmalloc_array(t->max_segs, sizeof(*t->sg), GFP_KERNEL);
 	if (!t->sg) {
+		ret = -ENOMEM;
+		goto out_free;
+	}
+
+	t->sg_areq = kmalloc_array(t->max_segs, sizeof(*t->sg_areq),
+				   GFP_KERNEL);
+	if (!t->sg_areq) {
 		ret = -ENOMEM;
 		goto out_free;
 	}
···
 	if (!(test->card->host->caps & MMC_CAP_CMD_DURING_TFR))
 		return RESULT_UNSUP_HOST;
 
-	ret = mmc_test_area_map(test, sz, 0, 0);
+	ret = mmc_test_area_map(test, sz, 0, 0, use_areq);
 	if (ret)
 		return ret;
 

drivers/mmc/core/queue.c

···
 {
 	struct mmc_host *host = mq->card->host;
 
-	if (mq->use_cqe)
+	if (mq->use_cqe && !host->hsq_enabled)
 		return mmc_cqe_issue_type(host, req);
 
 	if (req_op(req) == REQ_OP_READ || req_op(req) == REQ_OP_WRITE)
···
 {
 	struct request_queue *q = req->q;
 	struct mmc_queue *mq = q->queuedata;
+	struct mmc_card *card = mq->card;
+	struct mmc_host *host = card->host;
 	unsigned long flags;
 	int ret;
 
 	spin_lock_irqsave(&mq->lock, flags);
 
-	if (mq->recovery_needed || !mq->use_cqe)
+	if (mq->recovery_needed || !mq->use_cqe || host->hsq_enabled)
 		ret = BLK_EH_RESET_TIMER;
 	else
 		ret = mmc_cqe_timed_out(req);
···
 	struct mmc_queue *mq = container_of(work, struct mmc_queue,
 					    recovery_work);
 	struct request_queue *q = mq->queue;
+	struct mmc_host *host = mq->card->host;
 
 	mmc_get_card(mq->card, &mq->ctx);
 
 	mq->in_recovery = true;
 
-	if (mq->use_cqe)
+	if (mq->use_cqe && !host->hsq_enabled)
 		mmc_blk_cqe_recovery(mq);
 	else
 		mmc_blk_mq_recovery(mq);
···
 	spin_lock_irq(&mq->lock);
 	mq->recovery_needed = false;
 	spin_unlock_irq(&mq->lock);
+
+	if (host->hsq_enabled)
+		host->cqe_ops->cqe_recovery_finish(host);
 
 	mmc_put_card(mq->card, &mq->ctx);
 
···
 		}
 		break;
 	case MMC_ISSUE_ASYNC:
+		/*
+		 * For MMC host software queue, we only allow 2 requests in
+		 * flight to avoid a long latency.
+		 */
+		if (host->hsq_enabled && mq->in_flight[issue_type] > 2) {
+			spin_unlock_irq(&mq->lock);
+			return BLK_STS_RESOURCE;
+		}
 		break;
 	default:
 		/*
···
 	 * The queue depth for CQE must match the hardware because the request
 	 * tag is used to index the hardware queue.
 	 */
-	if (mq->use_cqe)
+	if (mq->use_cqe && !host->hsq_enabled)
 		mq->tag_set.queue_depth =
 			min_t(int, card->ext_csd.cmdq_depth, host->cqe_qdepth);
 	else

drivers/mmc/core/sd.c

···
 		}
 	}
 
+	if (host->cqe_ops && !host->cqe_enabled) {
+		err = host->cqe_ops->cqe_enable(host, card);
+		if (!err) {
+			host->cqe_enabled = true;
+			host->hsq_enabled = true;
+			pr_info("%s: Host Software Queue enabled\n",
+				mmc_hostname(host));
+		}
+	}
+
 	if (host->caps2 & MMC_CAP2_AVOID_3_3V &&
 	    host->ios.signal_voltage == MMC_SIGNAL_VOLTAGE_330) {
 		pr_err("%s: Host failed to negotiate down from 3.3V\n",

drivers/mmc/core/sdio_irq.c

···
 
 	card->sdio_single_irq = NULL;
 	if ((card->host->caps & MMC_CAP_SDIO_IRQ) &&
-	    card->host->sdio_irqs == 1)
+	    card->host->sdio_irqs == 1) {
 		for (i = 0; i < card->sdio_funcs; i++) {
-		       func = card->sdio_func[i];
-		       if (func && func->irq_handler) {
-			       card->sdio_single_irq = func;
-			       break;
-		       }
-	       }
+			func = card->sdio_func[i];
+			if (func && func->irq_handler) {
+				card->sdio_single_irq = func;
+				break;
+			}
+		}
+	}
 }
 
 /**

drivers/mmc/host/Kconfig

···
 	depends on ARCH_SPRD
 	depends on MMC_SDHCI_PLTFM
 	select MMC_SDHCI_IO_ACCESSORS
+	select MMC_HSQ
 	help
 	  This selects the SDIO Host Controller in Spreadtrum
 	  SoCs, this driver supports R11(IP version: R11P0).
···
 	  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_HSQ
+	tristate "MMC Host Software Queue support"
+	help
+	  This selects the MMC Host Software Queue support. This may increase
+	  performance, if the host controller and its driver supports it.
+
+	  If you have a controller/driver supporting this interface, say Y or M
+	  here.
 
 	  If unsure, say N.
 

drivers/mmc/host/Makefile

···
 obj-$(CONFIG_MMC_SDHCI_OMAP)		+= sdhci-omap.o
 obj-$(CONFIG_MMC_SDHCI_SPRD)		+= sdhci-sprd.o
 obj-$(CONFIG_MMC_CQHCI)			+= cqhci.o
+obj-$(CONFIG_MMC_HSQ)			+= mmc_hsq.o
 
 ifeq ($(CONFIG_CB710_DEBUG),y)
 	CFLAGS-cb710-mmc	+= -DDEBUG

drivers/mmc/host/cavium-octeon.c

···
 	base = devm_platform_ioremap_resource(pdev, 0);
 	if (IS_ERR(base))
 		return PTR_ERR(base);
-	host->base = (void __iomem *)base;
+	host->base = base;
 	host->reg_off = 0;
 
 	base = devm_platform_ioremap_resource(pdev, 1);
 	if (IS_ERR(base))
 		return PTR_ERR(base);
-	host->dma_base = (void __iomem *)base;
+	host->dma_base = base;
 	/*
 	 * To keep the register addresses shared we intentionaly use
 	 * a negative offset here, first register used on Octeon therefore

drivers/mmc/host/cqhci.c

···
 	cq_host->activated = false;
 }
 
-int cqhci_suspend(struct mmc_host *mmc)
+int cqhci_deactivate(struct mmc_host *mmc)
 {
 	struct cqhci_host *cq_host = mmc->cqe_private;
 
-	if (cq_host->enabled)
+	if (cq_host->enabled && cq_host->activated)
 		__cqhci_disable(cq_host);
 
 	return 0;
 }
-EXPORT_SYMBOL(cqhci_suspend);
+EXPORT_SYMBOL(cqhci_deactivate);
 
 int cqhci_resume(struct mmc_host *mmc)
 {
···
 	struct cqhci_host *cq_host = mmc->cqe_private;
 	int err;
 
+	if (!card->ext_csd.cmdq_en)
+		return -EINVAL;
+
 	if (cq_host->enabled)
 		return 0;
 
 	cq_host->rca = card->rca;
 
 	err = cqhci_host_alloc_tdl(cq_host);
-	if (err)
+	if (err) {
+		pr_err("%s: Failed to enable CQE, error %d\n",
+		       mmc_hostname(mmc), err);
 		return err;
+	}
 
 	__cqhci_enable(cq_host);
 
···
 
 	/* check and setup CMDQ interface */
 	cqhci_memres = platform_get_resource_byname(pdev, IORESOURCE_MEM,
-						   "cqhci_mem");
+						   "cqhci");
 	if (!cqhci_memres) {
 		dev_dbg(&pdev->dev, "CMDQ not supported\n");
 		return ERR_PTR(-EINVAL);

drivers/mmc/host/cqhci.h

···
 		      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_deactivate(struct mmc_host *mmc);
+static inline int cqhci_suspend(struct mmc_host *mmc)
+{
+	return cqhci_deactivate(mmc);
+}
 int cqhci_resume(struct mmc_host *mmc);
 
 #endif

drivers/mmc/host/mmc_hsq.c

···
+// SPDX-License-Identifier: GPL-2.0
+/*
+ *
+ * MMC software queue support based on command queue interfaces
+ *
+ * Copyright (C) 2019 Linaro, Inc.
+ * Author: Baolin Wang <[e-mail]>
+ */
+
+#include <linux/mmc/card.h>
+#include <linux/mmc/host.h>
+#include <linux/module.h>
+
+#include "mmc_hsq.h"
+
+#define HSQ_NUM_SLOTS	64
+#define HSQ_INVALID_TAG	HSQ_NUM_SLOTS
+
+static void mmc_hsq_pump_requests(struct mmc_hsq *hsq)
+{
+	struct mmc_host *mmc = hsq->mmc;
+	struct hsq_slot *slot;
+	unsigned long flags;
+
+	spin_lock_irqsave(&hsq->lock, flags);
+
+	/* Make sure we are not already running a request now */
+	if (hsq->mrq) {
+		spin_unlock_irqrestore(&hsq->lock, flags);
+		return;
+	}
+
+	/* Make sure there are remain requests need to pump */
+	if (!hsq->qcnt || !hsq->enabled) {
+		spin_unlock_irqrestore(&hsq->lock, flags);
+		return;
+	}
+
+	slot = &hsq->slot[hsq->next_tag];
+	hsq->mrq = slot->mrq;
+	hsq->qcnt--;
+
+	spin_unlock_irqrestore(&hsq->lock, flags);
+
+	mmc->ops->request(mmc, hsq->mrq);
+}
+
+static void mmc_hsq_update_next_tag(struct mmc_hsq *hsq, int remains)
+{
+	struct hsq_slot *slot;
+	int tag;
+
+	/*
+	 * If there are no remain requests in software queue, then set a invalid
+	 * tag.
+	 */
+	if (!remains) {
+		hsq->next_tag = HSQ_INVALID_TAG;
+		return;
+	}
+
+	/*
+	 * Increasing the next tag and check if the corresponding request is
+	 * available, if yes, then we found a candidate request.
+	 */
+	if (++hsq->next_tag != HSQ_INVALID_TAG) {
+		slot = &hsq->slot[hsq->next_tag];
+		if (slot->mrq)
+			return;
+	}
+
+	/* Othersie we should iterate all slots to find a available tag. */
+	for (tag = 0; tag < HSQ_NUM_SLOTS; tag++) {
+		slot = &hsq->slot[tag];
+		if (slot->mrq)
+			break;
+	}
+
+	if (tag == HSQ_NUM_SLOTS)
+		tag = HSQ_INVALID_TAG;
+
+	hsq->next_tag = tag;
+}
+
+static void mmc_hsq_post_request(struct mmc_hsq *hsq)
+{
+	unsigned long flags;
+	int remains;
+
+	spin_lock_irqsave(&hsq->lock, flags);
+
+	remains = hsq->qcnt;
+	hsq->mrq = NULL;
+
+	/* Update the next available tag to be queued. */
+	mmc_hsq_update_next_tag(hsq, remains);
+
+	if (hsq->waiting_for_idle && !remains) {
+		hsq->waiting_for_idle = false;
+		wake_up(&hsq->wait_queue);
+	}
+
+	/* Do not pump new request in recovery mode. */
+	if (hsq->recovery_halt) {
+		spin_unlock_irqrestore(&hsq->lock, flags);
+		return;
+	}
+
+	spin_unlock_irqrestore(&hsq->lock, flags);
+
+	 /*
+	  * Try to pump new request to host controller as fast as possible,
+	  * after completing previous request.
+	  */
+	if (remains > 0)
+		mmc_hsq_pump_requests(hsq);
+}
+
+/**
+ * mmc_hsq_finalize_request - finalize one request if the request is done
+ * @mmc: the host controller
+ * @mrq: the request need to be finalized
+ *
+ * Return true if we finalized the corresponding request in software queue,
+ * otherwise return false.
+ */
+bool mmc_hsq_finalize_request(struct mmc_host *mmc, struct mmc_request *mrq)
+{
+	struct mmc_hsq *hsq = mmc->cqe_private;
+	unsigned long flags;
+
+	spin_lock_irqsave(&hsq->lock, flags);
+
+	if (!hsq->enabled || !hsq->mrq || hsq->mrq != mrq) {
+		spin_unlock_irqrestore(&hsq->lock, flags);
+		return false;
+	}
+
+	/*
+	 * Clear current completed slot request to make a room for new request.
+	 */
+	hsq->slot[hsq->next_tag].mrq = NULL;
+
+	spin_unlock_irqrestore(&hsq->lock, flags);
+
+	mmc_cqe_request_done(mmc, hsq->mrq);
+
+	mmc_hsq_post_request(hsq);
+
+	return true;
+}
+EXPORT_SYMBOL_GPL(mmc_hsq_finalize_request);
+
+static void mmc_hsq_recovery_start(struct mmc_host *mmc)
+{
+	struct mmc_hsq *hsq = mmc->cqe_private;
+	unsigned long flags;
+
+	spin_lock_irqsave(&hsq->lock, flags);
+
+	hsq->recovery_halt = true;
+
+	spin_unlock_irqrestore(&hsq->lock, flags);
+}
+
+static void mmc_hsq_recovery_finish(struct mmc_host *mmc)
+{
+	struct mmc_hsq *hsq = mmc->cqe_private;
+	int remains;
+
+	spin_lock_irq(&hsq->lock);
+
+	hsq->recovery_halt = false;
+	remains = hsq->qcnt;
+
+	spin_unlock_irq(&hsq->lock);
+
+	/*
+	 * Try to pump new request if there are request pending in software
+	 * queue after finishing recovery.
+	 */
+	if (remains > 0)
+		mmc_hsq_pump_requests(hsq);
+}
+
+static int mmc_hsq_request(struct mmc_host *mmc, struct mmc_request *mrq)
+{
+	struct mmc_hsq *hsq = mmc->cqe_private;
+	int tag = mrq->tag;
+
+	spin_lock_irq(&hsq->lock);
+
+	if (!hsq->enabled) {
+		spin_unlock_irq(&hsq->lock);
+		return -ESHUTDOWN;
+	}
+
+	/* Do not queue any new requests in recovery mode. */
+	if (hsq->recovery_halt) {
+		spin_unlock_irq(&hsq->lock);
+		return -EBUSY;
+	}
+
+	hsq->slot[tag].mrq = mrq;
+
+	/*
+	 * Set the next tag as current request tag if no available
+	 * next tag.
+	 */
+	if (hsq->next_tag == HSQ_INVALID_TAG)
+		hsq->next_tag = tag;
+
+	hsq->qcnt++;
+
+	spin_unlock_irq(&hsq->lock);
+
+	mmc_hsq_pump_requests(hsq);
+
+	return 0;
+}
+
+static void mmc_hsq_post_req(struct mmc_host *mmc, struct mmc_request *mrq)
+{
+	if (mmc->ops->post_req)
+		mmc->ops->post_req(mmc, mrq, 0);
+}
+
+static bool mmc_hsq_queue_is_idle(struct mmc_hsq *hsq, int *ret)
+{
+	bool is_idle;
+
+	spin_lock_irq(&hsq->lock);
+
+	is_idle = (!hsq->mrq && !hsq->qcnt) ||
+		hsq->recovery_halt;
+
+	*ret = hsq->recovery_halt ? -EBUSY : 0;
+	hsq->waiting_for_idle = !is_idle;
+
+	spin_unlock_irq(&hsq->lock);
+
+	return is_idle;
+}
+
+static int mmc_hsq_wait_for_idle(struct mmc_host *mmc)
+{
+	struct mmc_hsq *hsq = mmc->cqe_private;
+	int ret;
+
+	wait_event(hsq->wait_queue,
+		   mmc_hsq_queue_is_idle(hsq, &ret));
+
+	return ret;
+}
+
+static void mmc_hsq_disable(struct mmc_host *mmc)
+{
+	struct mmc_hsq *hsq = mmc->cqe_private;
+	u32 timeout = 500;
+	int ret;
+
+	spin_lock_irq(&hsq->lock);
+
+	if (!hsq->enabled) {
+		spin_unlock_irq(&hsq->lock);
+		return;
+	}
+
+	spin_unlock_irq(&hsq->lock);
+
+	ret = wait_event_timeout(hsq->wait_queue,
+				 mmc_hsq_queue_is_idle(hsq, &ret),
+				 msecs_to_jiffies(timeout));
+	if (ret == 0) {
+		pr_warn("could not stop mmc software queue\n");
+		return;
+	}
+
+	spin_lock_irq(&hsq->lock);
+
+	hsq->enabled = false;
+
+	spin_unlock_irq(&hsq->lock);
+}
+
+static int mmc_hsq_enable(struct mmc_host *mmc, struct mmc_card *card)
+{
+	struct mmc_hsq *hsq = mmc->cqe_private;
+
+	spin_lock_irq(&hsq->lock);
+
+	if (hsq->enabled) {
+		spin_unlock_irq(&hsq->lock);
+		return -EBUSY;
+	}
+
+	hsq->enabled = true;
+
+	spin_unlock_irq(&hsq->lock);
+
+	return 0;
+}
+
+static const struct mmc_cqe_ops mmc_hsq_ops = {
+	.cqe_enable = mmc_hsq_enable,
+	.cqe_disable = mmc_hsq_disable,
+	.cqe_request = mmc_hsq_request,
+	.cqe_post_req = mmc_hsq_post_req,
+	.cqe_wait_for_idle = mmc_hsq_wait_for_idle,
+	.cqe_recovery_start = mmc_hsq_recovery_start,
+	.cqe_recovery_finish = mmc_hsq_recovery_finish,
+};
+
+int mmc_hsq_init(struct mmc_hsq *hsq, struct mmc_host *mmc)
+{
+	hsq->num_slots = HSQ_NUM_SLOTS;
+	hsq->next_tag = HSQ_INVALID_TAG;
+
+	hsq->slot = devm_kcalloc(mmc_dev(mmc), hsq->num_slots,
+				 sizeof(struct hsq_slot), GFP_KERNEL);
+	if (!hsq->slot)
+		return -ENOMEM;
+
+	hsq->mmc = mmc;
+	hsq->mmc->cqe_private = hsq;
+	mmc->cqe_ops = &mmc_hsq_ops;
+
+	spin_lock_init(&hsq->lock);
+	init_waitqueue_head(&hsq->wait_queue);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(mmc_hsq_init);
+
+void mmc_hsq_suspend(struct mmc_host *mmc)
+{
+	mmc_hsq_disable(mmc);
+}
+EXPORT_SYMBOL_GPL(mmc_hsq_suspend);
+
+int mmc_hsq_resume(struct mmc_host *mmc)
+{
+	return mmc_hsq_enable(mmc, NULL);
+}
+EXPORT_SYMBOL_GPL(mmc_hsq_resume);
+
+MODULE_DESCRIPTION("MMC Host Software Queue support");
+MODULE_LICENSE("GPL v2");

drivers/mmc/host/mmc_hsq.h

···
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef LINUX_MMC_HSQ_H
+#define LINUX_MMC_HSQ_H
+
+struct hsq_slot {
+	struct mmc_request *mrq;
+};
+
+struct mmc_hsq {
+	struct mmc_host *mmc;
+	struct mmc_request *mrq;
+	wait_queue_head_t wait_queue;
+	struct hsq_slot *slot;
+	spinlock_t lock;
+
+	int next_tag;
+	int num_slots;
+	int qcnt;
+
+	bool enabled;
+	bool waiting_for_idle;
+	bool recovery_halt;
+};
+
+int mmc_hsq_init(struct mmc_hsq *hsq, struct mmc_host *mmc);
+void mmc_hsq_suspend(struct mmc_host *mmc);
+int mmc_hsq_resume(struct mmc_host *mmc);
+bool mmc_hsq_finalize_request(struct mmc_host *mmc, struct mmc_request *mrq);
+
+#endif

drivers/mmc/host/mmci.c

···
 #include <linux/mmc/pm.h>
 #include <linux/mmc/host.h>
 #include <linux/mmc/card.h>
+#include <linux/mmc/sd.h>
 #include <linux/mmc/slot-gpio.h>
 #include <linux/amba/bus.h>
 #include <linux/clk.h>
···
 	.datactrl_blocksz	= 14,
 	.datactrl_any_blocksz	= true,
 	.stm32_idmabsize_mask	= GENMASK(12, 5),
+	.busy_timeout		= true,
+	.busy_detect		= true,
+	.busy_detect_flag	= MCI_STM32_BUSYD0,
+	.busy_detect_mask	= MCI_STM32_BUSYD0ENDMASK,
+	.init			= sdmmc_variant_init,
+};
+
+static struct variant_data variant_stm32_sdmmcv2 = {
+	.fifosize		= 16 * 4,
+	.fifohalfsize		= 8 * 4,
+	.f_max			= 208000000,
+	.stm32_clkdiv		= true,
+	.cmdreg_cpsm_enable	= MCI_CPSM_STM32_ENABLE,
+	.cmdreg_lrsp_crc	= MCI_CPSM_STM32_LRSP_CRC,
+	.cmdreg_srsp_crc	= MCI_CPSM_STM32_SRSP_CRC,
+	.cmdreg_srsp		= MCI_CPSM_STM32_SRSP,
+	.cmdreg_stop		= MCI_CPSM_STM32_CMDSTOP,
+	.data_cmd_enable	= MCI_CPSM_STM32_CMDTRANS,
+	.irq_pio_mask		= MCI_IRQ_PIO_STM32_MASK,
+	.datactrl_first		= true,
+	.datacnt_useless	= true,
+	.datalength_bits	= 25,
+	.datactrl_blocksz	= 14,
+	.datactrl_any_blocksz	= true,
+	.stm32_idmabsize_mask	= GENMASK(16, 5),
+	.dma_lli		= true,
 	.busy_timeout		= true,
 	.busy_detect		= true,
 	.busy_detect_flag	= MCI_STM32_BUSYD0,
···
 		writel_relaxed(clks, host->base + MMCIDATATIMER);
 	}
 
+	if (host->ops->pre_sig_volt_switch && cmd->opcode == SD_SWITCH_VOLTAGE)
+		host->ops->pre_sig_volt_switch(host);
+
 	if (/*interrupt*/0)
 		c |= MCI_CPSM_INTERRUPT;
 
···
 
 static int mmci_sig_volt_switch(struct mmc_host *mmc, struct mmc_ios *ios)
 {
+	struct mmci_host *host = mmc_priv(mmc);
 	int ret = 0;
 
 	if (!IS_ERR(mmc->supply.vqmmc)) {
···
 						1100000, 1300000);
 			break;
 		}
+
+		if (!ret && host->ops && host->ops->post_sig_volt_switch)
+			ret = host->ops->post_sig_volt_switch(host, ios);
 
 		if (ret)
 			dev_warn(mmc_dev(mmc), "Voltage switch failed\n");
···
 
 	host = mmc_priv(mmc);
 	host->mmc = mmc;
+	host->mmc_ops = &mmci_ops;
+	mmc->ops = &mmci_ops;
 
 	/*
 	 * Some variant (STM32) doesn't have opendrain bit, nevertheless
···
 	host->stop_abort.opcode = MMC_STOP_TRANSMISSION;
 	host->stop_abort.arg = 0;
 	host->stop_abort.flags = MMC_RSP_R1B | MMC_CMD_AC;
-
-	mmc->ops = &mmci_ops;
 
 	/* We support these PM capabilities. */
 	mmc->pm_caps |= MMC_PM_KEEP_POWER;
···
 		.id     = 0x10153180,
 		.mask	= 0xf0ffffff,
 		.data	= &variant_stm32_sdmmc,
+	},
+	{
+		.id     = 0x00253180,
+		.mask	= 0xf0ffffff,
+		.data	= &variant_stm32_sdmmcv2,
 	},
 	/* Qualcomm variants */
 	{

drivers/mmc/host/mmci.h

···
 /* Extended status bits for the STM32 variants */
 #define MCI_STM32_BUSYD0	BIT(20)
 #define MCI_STM32_BUSYD0END	BIT(21)
+#define MCI_STM32_VSWEND	BIT(25)
 
 #define MMCICLEAR		0x038
 #define MCI_CMDCRCFAILCLR	(1 << 0)
···
 #define MCI_ST_SDIOITC		(1 << 22)
 #define MCI_ST_CEATAENDC	(1 << 23)
 #define MCI_ST_BUSYENDC		(1 << 24)
+/* Extended clear bits for the STM32 variants */
+#define MCI_STM32_VSWENDC	BIT(25)
+#define MCI_STM32_CKSTOPC	BIT(26)
 
 #define MMCIMASK0		0x03c
 #define MCI_CMDCRCFAILMASK	(1 << 0)
···
 	void (*set_clkreg)(struct mmci_host *host, unsigned int desired);
 	void (*set_pwrreg)(struct mmci_host *host, unsigned int pwr);
 	bool (*busy_complete)(struct mmci_host *host, u32 status, u32 err_msk);
+	void (*pre_sig_volt_switch)(struct mmci_host *host);
+	int (*post_sig_volt_switch)(struct mmci_host *host, struct mmc_ios *ios);
 };
 
 struct mmci_host {
···
 	u32			mask1_reg;
 	u8			vqmmc_enabled:1;
 	struct mmci_platform_data *plat;
+	struct mmc_host_ops	*mmc_ops;
 	struct mmci_host_ops	*ops;
 	struct variant_data	*variant;
+	void			*variant_priv;
 	struct pinctrl		*pinctrl;
 	struct pinctrl_state	*pins_opendrain;
 

drivers/mmc/host/mmci_stm32_sdmmc.c

···
  * Copyright (C) STMicroelectronics 2018 - All Rights Reserved
  * Author: [e-mail] for STMicroelectronics.
  */
+#include <linux/bitfield.h>
 #include <linux/delay.h>
 #include <linux/dma-mapping.h>
+#include <linux/iopoll.h>
 #include <linux/mmc/host.h>
 #include <linux/mmc/card.h>
+#include <linux/of_address.h>
 #include <linux/reset.h>
 #include <linux/scatterlist.h>
 #include "mmci.h"
···
 #define SDMMC_LLI_BUF_LEN	PAGE_SIZE
 #define SDMMC_IDMA_BURST	BIT(MMCI_STM32_IDMABNDT_SHIFT)
 
+#define DLYB_CR			0x0
+#define DLYB_CR_DEN		BIT(0)
+#define DLYB_CR_SEN		BIT(1)
+
+#define DLYB_CFGR		0x4
+#define DLYB_CFGR_SEL_MASK	GENMASK(3, 0)
+#define DLYB_CFGR_UNIT_MASK	GENMASK(14, 8)
+#define DLYB_CFGR_LNG_MASK	GENMASK(27, 16)
+#define DLYB_CFGR_LNGF		BIT(31)
+
+#define DLYB_NB_DELAY		11
+#define DLYB_CFGR_SEL_MAX	(DLYB_NB_DELAY + 1)
+#define DLYB_CFGR_UNIT_MAX	127
+
+#define DLYB_LNG_TIMEOUT_US	1000
+#define SDMMC_VSWEND_TIMEOUT_US 10000
+
 struct sdmmc_lli_desc {
 	u32 idmalar;
 	u32 idmabase;
 	u32 idmasize;
 };
 
-struct sdmmc_priv {
+struct sdmmc_idma {
 	dma_addr_t sg_dma;
 	void *sg_cpu;
+};
+
+struct sdmmc_dlyb {
+	void __iomem *base;
+	u32 unit;
+	u32 max;
 };
 
 static int sdmmc_idma_validate_data(struct mmci_host *host,
···
 	 * excepted the last element which has no constraint on idmasize
 	 */
 	for_each_sg(data->sg, sg, data->sg_len - 1, i) {
-		if (!IS_ALIGNED(sg_dma_address(data->sg), sizeof(u32)) ||
-		    !IS_ALIGNED(sg_dma_len(data->sg), SDMMC_IDMA_BURST)) {
+		if (!IS_ALIGNED(data->sg->offset, sizeof(u32)) ||
+		    !IS_ALIGNED(data->sg->length, SDMMC_IDMA_BURST)) {
 			dev_err(mmc_dev(host->mmc),
 				"unaligned scatterlist: ofst:%x length:%d\n",
 				data->sg->offset, data->sg->length);
···
 		}
 	}
 
-	if (!IS_ALIGNED(sg_dma_address(data->sg), sizeof(u32))) {
+	if (!IS_ALIGNED(data->sg->offset, sizeof(u32))) {
 		dev_err(mmc_dev(host->mmc),
 			"unaligned last scatterlist: ofst:%x length:%d\n",
 			data->sg->offset, data->sg->length);
···
 
 static int sdmmc_idma_setup(struct mmci_host *host)
 {
-	struct sdmmc_priv *idma;
+	struct sdmmc_idma *idma;
 
 	idma = devm_kzalloc(mmc_dev(host->mmc), sizeof(*idma), GFP_KERNEL);
 	if (!idma)
···
 static int sdmmc_idma_start(struct mmci_host *host, unsigned int *datactrl)
 
 {
-	struct sdmmc_priv *idma = host->dma_priv;
+	struct sdmmc_idma *idma = host->dma_priv;
 	struct sdmmc_lli_desc *desc = (struct sdmmc_lli_desc *)idma->sg_cpu;
 	struct mmc_data *data = host->data;
 	struct scatterlist *sg;
···
 	mmci_write_clkreg(host, clk);
 }
 
+static void sdmmc_dlyb_input_ck(struct sdmmc_dlyb *dlyb)
+{
+	if (!dlyb || !dlyb->base)
+		return;
+
+	/* Output clock = Input clock */
+	writel_relaxed(0, dlyb->base + DLYB_CR);
+}
+
 static void mmci_sdmmc_set_pwrreg(struct mmci_host *host, unsigned int pwr)
 {
 	struct mmc_ios ios = host->mmc->ios;
+	struct sdmmc_dlyb *dlyb = host->variant_priv;
 
+	/* adds OF options */
 	pwr = host->pwr_reg_add;
+
+	sdmmc_dlyb_input_ck(dlyb);
 
 	if (ios.power_mode == MMC_POWER_OFF) {
 		/* Only a reset could power-off sdmmc */
···
 		 */
 		writel(MCI_IRQENABLE | host->variant->start_err,
 		       host->base + MMCIMASK0);
+
+		/* preserves voltage switch bits */
+		pwr |= host->pwr_reg & (MCI_STM32_VSWITCHEN |
+					MCI_STM32_VSWITCH);
 
 		/*
 		 * After a power-cycle state, we must set the SDMMC in
···
 	if (host->busy_status) {
 		writel_relaxed(mask & ~host->variant->busy_detect_mask,
 			       base + MMCIMASK0);
-		writel_relaxed(host->variant->busy_detect_mask,
-			       base + MMCICLEAR);
 		host->busy_status = 0;
 	}
 
+	writel_relaxed(host->variant->busy_detect_mask, base + MMCICLEAR);
+
 	return true;
+}
+
+static void sdmmc_dlyb_set_cfgr(struct sdmmc_dlyb *dlyb,
+				int unit, int phase, bool sampler)
+{
+	u32 cfgr;
+
+	writel_relaxed(DLYB_CR_SEN | DLYB_CR_DEN, dlyb->base + DLYB_CR);
+
+	cfgr = FIELD_PREP(DLYB_CFGR_UNIT_MASK, unit) |
+	       FIELD_PREP(DLYB_CFGR_SEL_MASK, phase);
+	writel_relaxed(cfgr, dlyb->base + DLYB_CFGR);
+
+	if (!sampler)
+		writel_relaxed(DLYB_CR_DEN, dlyb->base + DLYB_CR);
+}
+
+static int sdmmc_dlyb_lng_tuning(struct mmci_host *host)
+{
+	struct sdmmc_dlyb *dlyb = host->variant_priv;
+	u32 cfgr;
+	int i, lng, ret;
+
+	for (i = 0; i <= DLYB_CFGR_UNIT_MAX; i++) {
+		sdmmc_dlyb_set_cfgr(dlyb, i, DLYB_CFGR_SEL_MAX, true);
+
+		ret = readl_relaxed_poll_timeout(dlyb->base + DLYB_CFGR, cfgr,
+						 (cfgr & DLYB_CFGR_LNGF),
+						 1, DLYB_LNG_TIMEOUT_US);
+		if (ret) {
+			dev_warn(mmc_dev(host->mmc),
+				 "delay line cfg timeout unit:%d cfgr:%d\n",
+				 i, cfgr);
+			continue;
+		}
+
+		lng = FIELD_GET(DLYB_CFGR_LNG_MASK, cfgr);
+		if (lng < BIT(DLYB_NB_DELAY) && lng > 0)
+			break;
+	}
+
+	if (i > DLYB_CFGR_UNIT_MAX)
+		return -EINVAL;
+
+	dlyb->unit = i;
+	dlyb->max = __fls(lng);
+
+	return 0;
+}
+
+static int sdmmc_dlyb_phase_tuning(struct mmci_host *host, u32 opcode)
+{
+	struct sdmmc_dlyb *dlyb = host->variant_priv;
+	int cur_len = 0, max_len = 0, end_of_len = 0;
+	int phase;
+
+	for (phase = 0; phase <= dlyb->max; phase++) {
+		sdmmc_dlyb_set_cfgr(dlyb, dlyb->unit, phase, false);
+
+		if (mmc_send_tuning(host->mmc, opcode, NULL)) {
+			cur_len = 0;
+		} else {
+			cur_len++;
+			if (cur_len > max_len) {
+				max_len = cur_len;
+				end_of_len = phase;
+			}
+		}
+	}
+
+	if (!max_len) {
+		dev_err(mmc_dev(host->mmc), "no tuning point found\n");
+		return -EINVAL;
+	}
+
+	phase = end_of_len - max_len / 2;
+	sdmmc_dlyb_set_cfgr(dlyb, dlyb->unit, phase, false);
+
+	dev_dbg(mmc_dev(host->mmc), "unit:%d max_dly:%d phase:%d\n",
+		dlyb->unit, dlyb->max, phase);
+
+	return 0;
+}
+
+static int sdmmc_execute_tuning(struct mmc_host *mmc, u32 opcode)
+{
+	struct mmci_host *host = mmc_priv(mmc);
+	struct sdmmc_dlyb *dlyb = host->variant_priv;
+
+	if (!dlyb || !dlyb->base)
+		return -EINVAL;
+
+	if (sdmmc_dlyb_lng_tuning(host))
+		return -EINVAL;
+
+	return sdmmc_dlyb_phase_tuning(host, opcode);
+}
+
+static void sdmmc_pre_sig_volt_vswitch(struct mmci_host *host)
+{
+	/* clear the voltage switch completion flag */
+	writel_relaxed(MCI_STM32_VSWENDC, host->base + MMCICLEAR);
+	/* enable Voltage switch procedure */
+	mmci_write_pwrreg(host, host->pwr_reg | MCI_STM32_VSWITCHEN);
+}
+
+static int sdmmc_post_sig_volt_switch(struct mmci_host *host,
+				      struct mmc_ios *ios)
+{
+	unsigned long flags;
+	u32 status;
+	int ret = 0;
+
+	if (ios->signal_voltage == MMC_SIGNAL_VOLTAGE_180) {
+		spin_lock_irqsave(&host->lock, flags);
+		mmci_write_pwrreg(host, host->pwr_reg | MCI_STM32_VSWITCH);
+		spin_unlock_irqrestore(&host->lock, flags);
+
+		/* wait voltage switch completion while 10ms */
+		ret = readl_relaxed_poll_timeout(host->base + MMCISTATUS,
+						 status,
+						 (status & MCI_STM32_VSWEND),
+						 10, SDMMC_VSWEND_TIMEOUT_US);
+
+		writel_relaxed(MCI_STM32_VSWENDC | MCI_STM32_CKSTOPC,
+			       host->base + MMCICLEAR);
+		mmci_write_pwrreg(host, host->pwr_reg &
+				  ~(MCI_STM32_VSWITCHEN | MCI_STM32_VSWITCH));
+	}
+
+	return ret;
 }
 
 static struct mmci_host_ops sdmmc_variant_ops = {
···
 	.set_clkreg = mmci_sdmmc_set_clkreg,
 	.set_pwrreg = mmci_sdmmc_set_pwrreg,
 	.busy_complete = sdmmc_busy_complete,
+	.pre_sig_volt_switch = sdmmc_pre_sig_volt_vswitch,
+	.post_sig_volt_switch = sdmmc_post_sig_volt_switch,
 };
 
 void sdmmc_variant_init(struct mmci_host *host)
 {
+	struct device_node *np = host->mmc->parent->of_node;
+	void __iomem *base_dlyb;
+	struct sdmmc_dlyb *dlyb;
+
 	host->ops = &sdmmc_variant_ops;
+
+	base_dlyb = devm_of_iomap(mmc_dev(host->mmc), np, 1, NULL);
+	if (IS_ERR(base_dlyb))
+		return;
+
+	dlyb = devm_kzalloc(mmc_dev(host->mmc), sizeof(*dlyb), GFP_KERNEL);
+	if (!dlyb)
+		return;
+
+	dlyb->base = base_dlyb;
+	host->variant_priv = dlyb;
+	host->mmc_ops->execute_tuning = sdmmc_execute_tuning;
 }

drivers/mmc/host/mtk-sd.c

···
 #define MSDC_PS_CDSTS           (0x1 << 1)	/* R  */
 #define MSDC_PS_CDDEBOUNCE      (0xf << 12)	/* RW */
 #define MSDC_PS_DAT             (0xff << 16)	/* R  */
+#define MSDC_PS_DATA1           (0x1 << 17)	/* R  */
 #define MSDC_PS_CMD             (0x1 << 24)	/* R  */
 #define MSDC_PS_WP              (0x1 << 31)	/* R  */
 
···
 
 struct mtk_mmc_compatible {
 	u8 clk_div_bits;
+	bool recheck_sdio_irq;
 	bool hs400_tune; /* only used for MT8173 */
 	u32 pad_tune_reg;
 	bool async_fifo;
···
 
 static const struct mtk_mmc_compatible mt8135_compat = {
 	.clk_div_bits = 8,
+	.recheck_sdio_irq = false,
 	.hs400_tune = false,
 	.pad_tune_reg = MSDC_PAD_TUNE,
 	.async_fifo = false,
···
 
 static const struct mtk_mmc_compatible mt8173_compat = {
 	.clk_div_bits = 8,
+	.recheck_sdio_irq = true,
 	.hs400_tune = true,
 	.pad_tune_reg = MSDC_PAD_TUNE,
 	.async_fifo = false,
···
 
 static const struct mtk_mmc_compatible mt8183_compat = {
 	.clk_div_bits = 12,
+	.recheck_sdio_irq = false,
 	.hs400_tune = false,
 	.pad_tune_reg = MSDC_PAD_TUNE0,
 	.async_fifo = true,
···
 
 static const struct mtk_mmc_compatible mt2701_compat = {
 	.clk_div_bits = 12,
+	.recheck_sdio_irq = false,
 	.hs400_tune = false,
 	.pad_tune_reg = MSDC_PAD_TUNE0,
 	.async_fifo = true,
···
 
 static const struct mtk_mmc_compatible mt2712_compat = {
 	.clk_div_bits = 12,
+	.recheck_sdio_irq = false,
 	.hs400_tune = false,
 	.pad_tune_reg = MSDC_PAD_TUNE0,
 	.async_fifo = true,
···
 
 static const struct mtk_mmc_compatible mt7622_compat = {
 	.clk_div_bits = 12,
+	.recheck_sdio_irq = false,
 	.hs400_tune = false,
 	.pad_tune_reg = MSDC_PAD_TUNE0,
 	.async_fifo = true,
···
 
 static const struct mtk_mmc_compatible mt8516_compat = {
 	.clk_div_bits = 12,
+	.recheck_sdio_irq = false,
 	.hs400_tune = false,
 	.pad_tune_reg = MSDC_PAD_TUNE0,
 	.async_fifo = true,
···
 
 static const struct mtk_mmc_compatible mt7620_compat = {
 	.clk_div_bits = 8,
+	.recheck_sdio_irq = false,
 	.hs400_tune = false,
 	.pad_tune_reg = MSDC_PAD_TUNE,
 	.async_fifo = false,
···
 
 static void msdc_cmd_next(struct msdc_host *host,
 		struct mmc_request *mrq, struct mmc_command *cmd);
+static void __msdc_enable_sdio_irq(struct msdc_host *host, int enb);
 
 static const u32 cmd_ints_mask = MSDC_INTEN_CMDRDY | MSDC_INTEN_RSPCRCERR |
 			MSDC_INTEN_CMDTMO | MSDC_INTEN_ACMDRDY |
···
 	return cmd->error;
 }
 
+/**
+ * msdc_recheck_sdio_irq - recheck whether the SDIO irq is lost
+ *
+ * Host controller may lost interrupt in some special case.
+ * Add SDIO irq recheck mechanism to make sure all interrupts
+ * can be processed immediately
+ *
+ */
+static void msdc_recheck_sdio_irq(struct msdc_host *host)
+{
+	u32 reg_int, reg_inten, reg_ps;
+
+	if (host->mmc->caps & MMC_CAP_SDIO_IRQ) {
+		reg_inten = readl(host->base + MSDC_INTEN);
+		if (reg_inten & MSDC_INTEN_SDIOIRQ) {
+			reg_int = readl(host->base + MSDC_INT);
+			reg_ps = readl(host->base + MSDC_PS);
+			if (!(reg_int & MSDC_INT_SDIOIRQ ||
+			      reg_ps & MSDC_PS_DATA1)) {
+				__msdc_enable_sdio_irq(host, 0);
+				sdio_signal_irq(host->mmc);
+			}
+		}
+	}
+}
+
 static void msdc_track_cmd_data(struct msdc_host *host,
 				struct mmc_command *cmd, struct mmc_data *data)
 {
···
 	if (host->error)
 		msdc_reset_hw(host);
 	mmc_request_done(host->mmc, mrq);
+	if (host->dev_comp->recheck_sdio_irq)
+		msdc_recheck_sdio_irq(host);
 }
 
 /* returns true if command is fully handled; returns false otherwise */
···
 	if (enb) {
 		sdr_set_bits(host->base + MSDC_INTEN, MSDC_INTEN_SDIOIRQ);
 		sdr_set_bits(host->base + SDC_CFG, SDC_CFG_SDIOIDE);
+		if (host->dev_comp->recheck_sdio_irq)
+			msdc_recheck_sdio_irq(host);
 	} else {
 		sdr_clr_bits(host->base + MSDC_INTEN, MSDC_INTEN_SDIOIRQ);
 		sdr_clr_bits(host->base + SDC_CFG, SDC_CFG_SDIOIDE);

drivers/mmc/host/renesas_sdhi.h

···
 	void __iomem *scc_ctl;
 	u32 scc_tappos;
 	u32 scc_tappos_hs400;
+	bool doing_tune;
+
+	/* Tuning values: 1 for success, 0 for failure */
+	DECLARE_BITMAP(taps, BITS_PER_LONG);
+	unsigned int tap_num;
+	unsigned long tap_set;
 };
 
 #define host_to_priv(host) \

drivers/mmc/host/renesas_sdhi_core.c

···
 #define SH_MOBILE_SDHI_SCC_CKSEL	0x006
 #define SH_MOBILE_SDHI_SCC_RVSCNTL	0x008
 #define SH_MOBILE_SDHI_SCC_RVSREQ	0x00A
+#define SH_MOBILE_SDHI_SCC_SMPCMP       0x00C
 #define SH_MOBILE_SDHI_SCC_TMPPORT2	0x00E
 
-/* Definitions for values the SH_MOBILE_SDHI_SCC_DTCNTL register */
 #define SH_MOBILE_SDHI_SCC_DTCNTL_TAPEN		BIT(0)
 #define SH_MOBILE_SDHI_SCC_DTCNTL_TAPNUM_SHIFT	16
 #define SH_MOBILE_SDHI_SCC_DTCNTL_TAPNUM_MASK	0xff
 
-/* Definitions for values the SH_MOBILE_SDHI_SCC_CKSEL register */
 #define SH_MOBILE_SDHI_SCC_CKSEL_DTSEL		BIT(0)
-/* Definitions for values the SH_MOBILE_SDHI_SCC_RVSCNTL register */
+
 #define SH_MOBILE_SDHI_SCC_RVSCNTL_RVSEN	BIT(0)
-/* Definitions for values the SH_MOBILE_SDHI_SCC_RVSREQ register */
+
+#define SH_MOBILE_SDHI_SCC_RVSREQ_REQTAPDOWN	BIT(0)
+#define SH_MOBILE_SDHI_SCC_RVSREQ_REQTAPUP	BIT(1)
 #define SH_MOBILE_SDHI_SCC_RVSREQ_RVSERR	BIT(2)
-/* Definitions for values the SH_MOBILE_SDHI_SCC_TMPPORT2 register */
+
+#define SH_MOBILE_SDHI_SCC_SMPCMP_CMD_REQDOWN	BIT(8)
+#define SH_MOBILE_SDHI_SCC_SMPCMP_CMD_REQUP	BIT(24)
+#define SH_MOBILE_SDHI_SCC_SMPCMP_CMD_ERR	(BIT(8) | BIT(24))
+
 #define SH_MOBILE_SDHI_SCC_TMPPORT2_HS400OSEL	BIT(4)
 #define SH_MOBILE_SDHI_SCC_TMPPORT2_HS400EN	BIT(31)
 
···
 		SH_MOBILE_SDHI_SCC_DTCNTL_TAPNUM_MASK;
 }
 
-static void renesas_sdhi_prepare_tuning(struct tmio_mmc_host *host,
-					unsigned long tap)
+static void renesas_sdhi_hs400_complete(struct mmc_host *mmc)
 {
-	struct renesas_sdhi *priv = host_to_priv(host);
-
-	/* Set sampling clock position */
-	sd_scc_write32(host, priv, SH_MOBILE_SDHI_SCC_TAPSET, tap);
-}
-
-static void renesas_sdhi_hs400_complete(struct tmio_mmc_host *host)
-{
+	struct tmio_mmc_host *host = mmc_priv(mmc);
 	struct renesas_sdhi *priv = host_to_priv(host);
 
 	sd_ctrl_write16(host, CTL_SD_CARD_CLK_CTL, ~CLK_CTL_SCLKEN &
···
 
 	sd_scc_write32(host, priv, SH_MOBILE_SDHI_SCC_DT2FF,
 		       priv->scc_tappos_hs400);
+
+	/* Gen3 can't do automatic tap correction with HS400, so disable it */
+	if (sd_ctrl_read16(host, CTL_VERSION) == SDHI_VER_GEN3_SDMMC)
+		sd_scc_write32(host, priv, SH_MOBILE_SDHI_SCC_RVSCNTL,
+			       ~SH_MOBILE_SDHI_SCC_RVSCNTL_RVSEN &
+			       sd_scc_read32(host, priv, SH_MOBILE_SDHI_SCC_RVSCNTL));
 
 	sd_scc_write32(host, priv, SH_MOBILE_SDHI_SCC_TMPPORT2,
 		       (SH_MOBILE_SDHI_SCC_TMPPORT2_HS400EN |
···
 
 	if (priv->quirks && priv->quirks->hs400_4taps)
 		sd_scc_write32(host, priv, SH_MOBILE_SDHI_SCC_TAPSET,
-			       host->tap_set / 2);
+			       priv->tap_set / 2);
 
 	sd_scc_write32(host, priv, SH_MOBILE_SDHI_SCC_CKSEL,
 		       SH_MOBILE_SDHI_SCC_CKSEL_DTSEL |
···
 				     SH_MOBILE_SDHI_SCC_CKSEL));
 }
 
-static void renesas_sdhi_disable_scc(struct tmio_mmc_host *host)
+static void renesas_sdhi_disable_scc(struct mmc_host *mmc)
 {
+	struct tmio_mmc_host *host = mmc_priv(mmc);
 	struct renesas_sdhi *priv = host_to_priv(host);
 
 	renesas_sdhi_reset_scc(host, priv);
···
 			sd_ctrl_read16(host, CTL_SD_CARD_CLK_CTL));
 }
 
-static void renesas_sdhi_prepare_hs400_tuning(struct tmio_mmc_host *host)
+static int renesas_sdhi_prepare_hs400_tuning(struct mmc_host *mmc, struct mmc_ios *ios)
 {
+	struct tmio_mmc_host *host = mmc_priv(mmc);
+
 	renesas_sdhi_reset_hs400_mode(host, host_to_priv(host));
+	return 0;
 }
 
 #define SH_MOBILE_SDHI_MAX_TAP 3
···
 	unsigned long ntap;     /* temporary counter of tuning success */
 	unsigned long i;
 
+	priv->doing_tune = false;
+
 	/* Clear SCC_RVSREQ */
 	sd_scc_write32(host, priv, SH_MOBILE_SDHI_SCC_RVSREQ, 0);
 
···
 	 * result requiring the tap to be good in both runs before
 	 * considering it for tuning selection.
 	 */
-	for (i = 0; i < host->tap_num * 2; i++) {
-		int offset = host->tap_num * (i < host->tap_num ? 1 : -1);
+	for (i = 0; i < priv->tap_num * 2; i++) {
+		int offset = priv->tap_num * (i < priv->tap_num ? 1 : -1);
 
-		if (!test_bit(i, host->taps))
-			clear_bit(i + offset, host->taps);
+		if (!test_bit(i, priv->taps))
+			clear_bit(i + offset, priv->taps);
 	}
 
 	/*
···
 	ntap = 0;
 	tap_start = 0;
 	tap_end = 0;
-	for (i = 0; i < host->tap_num * 2; i++) {
-		if (test_bit(i, host->taps)) {
+	for (i = 0; i < priv->tap_num * 2; i++) {
+		if (test_bit(i, priv->taps)) {
 			ntap++;
 		} else {
 			if (ntap > tap_cnt) {
···
 	}
 
 	if (tap_cnt >= SH_MOBILE_SDHI_MAX_TAP)
-		host->tap_set = (tap_start + tap_end) / 2 % host->tap_num;
+		priv->tap_set = (tap_start + tap_end) / 2 % priv->tap_num;
 	else
 		return -EIO;
 
 	/* Set SCC */
-	sd_scc_write32(host, priv, SH_MOBILE_SDHI_SCC_TAPSET, host->tap_set);
+	sd_scc_write32(host, priv, SH_MOBILE_SDHI_SCC_TAPSET, priv->tap_set);
 
 	/* Enable auto re-tuning */
 	sd_scc_write32(host, priv, SH_MOBILE_SDHI_SCC_RVSCNTL,
···
 		       sd_scc_read32(host, priv, SH_MOBILE_SDHI_SCC_RVSCNTL));
 
 	return 0;
+}
+
+static int renesas_sdhi_execute_tuning(struct tmio_mmc_host *host, u32 opcode)
+{
+	struct renesas_sdhi *priv = host_to_priv(host);
+	int i;
+
+	priv->tap_num = renesas_sdhi_init_tuning(host);
+	if (!priv->tap_num)
+		return 0; /* Tuning is not supported */
+
+	if (priv->tap_num * 2 >= sizeof(priv->taps) * BITS_PER_BYTE) {
+		dev_err(&host->pdev->dev,
+			"Too many taps, please update 'taps' in tmio_mmc_host!\n");
+		return -EINVAL;
+	}
+
+	priv->doing_tune = true;
+	bitmap_zero(priv->taps, priv->tap_num * 2);
+
+	/* Issue CMD19 twice for each tap */
+	for (i = 0; i < 2 * priv->tap_num; i++) {
+		/* Set sampling clock position */
+		sd_scc_write32(host, priv, SH_MOBILE_SDHI_SCC_TAPSET, i % priv->tap_num);
+
+		if (mmc_send_tuning(host->mmc, opcode, NULL) == 0)
+			set_bit(i, priv->taps);
+	}
+
+	return renesas_sdhi_select_tuning(host);
+}
+
+static bool renesas_sdhi_manual_correction(struct tmio_mmc_host *host, bool use_4tap)
+{
+	struct renesas_sdhi *priv = host_to_priv(host);
+	unsigned long new_tap = priv->tap_set;
+	u32 val;
+
+	val = sd_scc_read32(host, priv, SH_MOBILE_SDHI_SCC_RVSREQ);
+	if (!val)
+		return false;
+
+	sd_scc_write32(host, priv, SH_MOBILE_SDHI_SCC_RVSREQ, 0);
+
+	/* Change TAP position according to correction status */
+	if (sd_ctrl_read16(host, CTL_VERSION) == SDHI_VER_GEN3_SDMMC &&
+	    host->mmc->ios.timing == MMC_TIMING_MMC_HS400) {
+		/*
+		 * With HS400, the DAT signal is based on DS, not CLK.
+		 * Therefore, use only CMD status.
+		 */
+		u32 smpcmp = sd_scc_read32(host, priv, SH_MOBILE_SDHI_SCC_SMPCMP) &
+					   SH_MOBILE_SDHI_SCC_SMPCMP_CMD_ERR;
+		if (!smpcmp)
+			return false;	/* no error in CMD signal */
+		else if (smpcmp == SH_MOBILE_SDHI_SCC_SMPCMP_CMD_REQUP)
+			new_tap++;
+		else if (smpcmp == SH_MOBILE_SDHI_SCC_SMPCMP_CMD_REQDOWN)
+			new_tap--;
+		else
+			return true;	/* need retune */
+	} else {
+		if (val & SH_MOBILE_SDHI_SCC_RVSREQ_RVSERR)
+			return true;    /* need retune */
+		else if (val & SH_MOBILE_SDHI_SCC_RVSREQ_REQTAPUP)
+			new_tap++;
+		else if (val & SH_MOBILE_SDHI_SCC_RVSREQ_REQTAPDOWN)
+			new_tap--;
+		else
+			return false;
+	}
+
+	priv->tap_set = (new_tap % priv->tap_num);
+	sd_scc_write32(host, priv, SH_MOBILE_SDHI_SCC_TAPSET,
+		       priv->tap_set / (use_4tap ? 2 : 1));
+
+	return false;
+}
+
+static bool renesas_sdhi_auto_correction(struct tmio_mmc_host *host)
+{
+	struct renesas_sdhi *priv = host_to_priv(host);
+
+	/* Check SCC error */
+	if (sd_scc_read32(host, priv, SH_MOBILE_SDHI_SCC_RVSREQ) &
+	    SH_MOBILE_SDHI_SCC_RVSREQ_RVSERR) {
+		sd_scc_write32(host, priv, SH_MOBILE_SDHI_SCC_RVSREQ, 0);
+		return true;
+	}
+
+	return false;
 }
 
 static bool renesas_sdhi_check_scc_error(struct tmio_mmc_host *host)
···
 	    !(host->mmc->ios.timing == MMC_TIMING_MMC_HS400 && !use_4tap))
 		return false;
 
-	if (mmc_doing_retune(host->mmc))
+	if (mmc_doing_retune(host->mmc) || priv->doing_tune)
 		return false;
 
-	/* Check SCC error */
 	if (sd_scc_read32(host, priv, SH_MOBILE_SDHI_SCC_RVSCNTL) &
-	    SH_MOBILE_SDHI_SCC_RVSCNTL_RVSEN &&
-	    sd_scc_read32(host, priv, SH_MOBILE_SDHI_SCC_RVSREQ) &
-	    SH_MOBILE_SDHI_SCC_RVSREQ_RVSERR) {
-		/* Clear SCC error */
-		sd_scc_write32(host, priv, SH_MOBILE_SDHI_SCC_RVSREQ, 0);
-		return true;
-	}
+	    SH_MOBILE_SDHI_SCC_RVSCNTL_RVSEN)
+		return renesas_sdhi_auto_correction(host);
 
-	return false;
+	return renesas_sdhi_manual_correction(host, use_4tap);
 }
 
 static void renesas_sdhi_hw_reset(struct tmio_mmc_host *host)
···
 
 	sd_ctrl_write16(host, CTL_SD_CARD_CLK_CTL, CLK_CTL_SCLKEN |
 			sd_ctrl_read16(host, CTL_SD_CARD_CLK_CTL));
-
-	sd_scc_write32(host, priv, SH_MOBILE_SDHI_SCC_RVSCNTL,
-		       ~SH_MOBILE_SDHI_SCC_RVSCNTL_RVSEN &
-		       sd_scc_read32(host, priv, SH_MOBILE_SDHI_SCC_RVSCNTL));
 
 	sd_scc_write32(host, priv, SH_MOBILE_SDHI_SCC_RVSCNTL,
 		       ~SH_MOBILE_SDHI_SCC_RVSCNTL_RVSEN &
···
 		if (!hit)
 			dev_warn(&host->pdev->dev, "Unknown clock rate for tuning\n");
 
-		host->init_tuning = renesas_sdhi_init_tuning;
-		host->prepare_tuning = renesas_sdhi_prepare_tuning;
-		host->select_tuning = renesas_sdhi_select_tuning;
-		host->check_scc_error = renesas_sdhi_check_scc_error;
-		host->prepare_hs400_tuning =
-			renesas_sdhi_prepare_hs400_tuning;
-		host->hs400_downgrade = renesas_sdhi_disable_scc;
-		host->hs400_complete = renesas_sdhi_hs400_complete;
+		host->execute_tuning = renesas_sdhi_execute_tuning;
+		host->check_retune = renesas_sdhi_check_scc_error;
+		host->ops.prepare_hs400_tuning = renesas_sdhi_prepare_hs400_tuning;
+		host->ops.hs400_downgrade = renesas_sdhi_disable_scc;
+		host->ops.hs400_complete = renesas_sdhi_hs400_complete;
 	}
 
 	num_irqs = platform_irq_count(pdev);

drivers/mmc/host/sdhci-acpi.c

···
 	bool				use_runtime_pm;
 	bool				is_intel;
 	bool				reset_signal_volt_on_suspend;
-	unsigned long			private[0] ____cacheline_aligned;
+	unsigned long			private[] ____cacheline_aligned;
 };
 
 enum {

drivers/mmc/host/sdhci-cadence.c

···
 	void __iomem *hrs_addr;
 	bool enhanced_strobe;
 	unsigned int nr_phy_params;
-	struct sdhci_cdns_phy_param phy_params[0];
+	struct sdhci_cdns_phy_param phy_params[];
 };
 
 struct sdhci_cdns_phy_cfg {

drivers/mmc/host/sdhci-esdhc-imx.c

···
  */
 
 #include <linux/io.h>
+#include <linux/iopoll.h>
 #include <linux/delay.h>
 #include <linux/err.h>
 #include <linux/clk.h>
···
 #define ESDHC_STROBE_DLL_CTRL		0x70
 #define ESDHC_STROBE_DLL_CTRL_ENABLE	(1 << 0)
 #define ESDHC_STROBE_DLL_CTRL_RESET	(1 << 1)
+#define ESDHC_STROBE_DLL_CTRL_SLV_DLY_TARGET_DEFAULT	0x7
 #define ESDHC_STROBE_DLL_CTRL_SLV_DLY_TARGET_SHIFT	3
 #define ESDHC_STROBE_DLL_CTRL_SLV_UPDATE_INT_DEFAULT	(4 << 20)
 
···
 #define ESDHC_FLAG_CQHCI		BIT(12)
 /* need request pmqos during low power */
 #define ESDHC_FLAG_PMQOS		BIT(13)
+/* The IP state got lost in low power mode */
+#define ESDHC_FLAG_STATE_LOST_IN_LPMODE		BIT(14)
+/* The IP lost clock rate in PM_RUNTIME */
+#define ESDHC_FLAG_CLK_RATE_LOST_IN_PM_RUNTIME	BIT(15)
+/*
+ * The IP do not support the ACMD23 feature completely when use ADMA mode.
+ * In ADMA mode, it only use the 16 bit block count of the register 0x4
+ * (BLOCK_ATT) as the CMD23's argument for ACMD23 mode, which means it will
+ * ignore the upper 16 bit of the CMD23's argument. This will block the reliable
+ * write operation in RPMB, because RPMB reliable write need to set the bit31
+ * of the CMD23's argument.
+ * imx6qpdl/imx6sx/imx6sl/imx7d has this limitation only for ADMA mode, SDMA
+ * do not has this limitation. so when these SoC use ADMA mode, it need to
+ * disable the ACMD23 feature.
+ */
+#define ESDHC_FLAG_BROKEN_AUTO_CMD23	BIT(16)
 
 struct esdhc_soc_data {
 	u32 flags;
···
 };
 
 static const struct esdhc_soc_data usdhc_imx6q_data = {
-	.flags = ESDHC_FLAG_USDHC | ESDHC_FLAG_MAN_TUNING,
+	.flags = ESDHC_FLAG_USDHC | ESDHC_FLAG_MAN_TUNING
+			| ESDHC_FLAG_BROKEN_AUTO_CMD23,
 };
 
 static const struct esdhc_soc_data usdhc_imx6sl_data = {
 	.flags = ESDHC_FLAG_USDHC | ESDHC_FLAG_STD_TUNING
 			| ESDHC_FLAG_HAVE_CAP1 | ESDHC_FLAG_ERR004536
-			| ESDHC_FLAG_HS200,
+			| ESDHC_FLAG_HS200
+			| ESDHC_FLAG_BROKEN_AUTO_CMD23,
+};
+
+static const struct esdhc_soc_data usdhc_imx6sll_data = {
+	.flags = ESDHC_FLAG_USDHC | ESDHC_FLAG_STD_TUNING
+			| ESDHC_FLAG_HAVE_CAP1 | ESDHC_FLAG_HS200
+			| ESDHC_FLAG_STATE_LOST_IN_LPMODE,
 };
 
 static const struct esdhc_soc_data usdhc_imx6sx_data = {
 	.flags = ESDHC_FLAG_USDHC | ESDHC_FLAG_STD_TUNING
-			| ESDHC_FLAG_HAVE_CAP1 | ESDHC_FLAG_HS200,
+			| ESDHC_FLAG_HAVE_CAP1 | ESDHC_FLAG_HS200
+			| ESDHC_FLAG_STATE_LOST_IN_LPMODE
+			| ESDHC_FLAG_BROKEN_AUTO_CMD23,
 };
 
 static const struct esdhc_soc_data usdhc_imx6ull_data = {
 	.flags = ESDHC_FLAG_USDHC | ESDHC_FLAG_STD_TUNING
 			| ESDHC_FLAG_HAVE_CAP1 | ESDHC_FLAG_HS200
-			| ESDHC_FLAG_ERR010450,
+			| ESDHC_FLAG_ERR010450
+			| ESDHC_FLAG_STATE_LOST_IN_LPMODE,
 };
 
 static const struct esdhc_soc_data usdhc_imx7d_data = {
 	.flags = ESDHC_FLAG_USDHC | ESDHC_FLAG_STD_TUNING
 			| ESDHC_FLAG_HAVE_CAP1 | ESDHC_FLAG_HS200
-			| ESDHC_FLAG_HS400,
+			| ESDHC_FLAG_HS400
+			| ESDHC_FLAG_STATE_LOST_IN_LPMODE
+			| ESDHC_FLAG_BROKEN_AUTO_CMD23,
 };
 
 static struct esdhc_soc_data usdhc_imx7ulp_data = {
 	.flags = ESDHC_FLAG_USDHC | ESDHC_FLAG_STD_TUNING
 			| ESDHC_FLAG_HAVE_CAP1 | ESDHC_FLAG_HS200
-			| ESDHC_FLAG_PMQOS | ESDHC_FLAG_HS400,
+			| ESDHC_FLAG_PMQOS | ESDHC_FLAG_HS400
+			| ESDHC_FLAG_STATE_LOST_IN_LPMODE,
 };
 
 static struct esdhc_soc_data usdhc_imx8qxp_data = {
 	.flags = ESDHC_FLAG_USDHC | ESDHC_FLAG_STD_TUNING
 			| ESDHC_FLAG_HAVE_CAP1 | ESDHC_FLAG_HS200
 			| ESDHC_FLAG_HS400 | ESDHC_FLAG_HS400_ES
-			| ESDHC_FLAG_CQHCI,
+			| ESDHC_FLAG_CQHCI
+			| ESDHC_FLAG_STATE_LOST_IN_LPMODE
+			| ESDHC_FLAG_CLK_RATE_LOST_IN_PM_RUNTIME,
+};
+
+static struct esdhc_soc_data usdhc_imx8mm_data = {
+	.flags = ESDHC_FLAG_USDHC | ESDHC_FLAG_STD_TUNING
+			| ESDHC_FLAG_HAVE_CAP1 | ESDHC_FLAG_HS200
+			| ESDHC_FLAG_HS400 | ESDHC_FLAG_HS400_ES
+			| ESDHC_FLAG_CQHCI
+			| ESDHC_FLAG_STATE_LOST_IN_LPMODE,
 };
 
 struct pltfm_imx_data {
···
 	{ .compatible = "fsl,imx53-esdhc", .data = &esdhc_imx53_data, },
 	{ .compatible = "fsl,imx6sx-usdhc", .data = &usdhc_imx6sx_data, },
 	{ .compatible = "fsl,imx6sl-usdhc", .data = &usdhc_imx6sl_data, },
+	{ .compatible = "fsl,imx6sll-usdhc", .data = &usdhc_imx6sll_data, },
 	{ .compatible = "fsl,imx6q-usdhc", .data = &usdhc_imx6q_data, },
 	{ .compatible = "fsl,imx6ull-usdhc", .data = &usdhc_imx6ull_data, },
 	{ .compatible = "fsl,imx7d-usdhc", .data = &usdhc_imx7d_data, },
 	{ .compatible = "fsl,imx7ulp-usdhc", .data = &usdhc_imx7ulp_data, },
 	{ .compatible = "fsl,imx8qxp-usdhc", .data = &usdhc_imx8qxp_data, },
+	{ .compatible = "fsl,imx8mm-usdhc", .data = &usdhc_imx8mm_data, },
 	{ /* sentinel */ }
 };
 MODULE_DEVICE_TABLE(of, imx_esdhc_dt_ids);
···
 	u32 shift = (reg & 0x3) * 8;
 
 	writel(((readl(base) & ~(mask << shift)) | (val << shift)), base);
+}
+
+static inline void esdhc_wait_for_card_clock_gate_off(struct sdhci_host *host)
+{
+	u32 present_state;
+	int ret;
+
+	ret = readl_poll_timeout(host->ioaddr + ESDHC_PRSSTAT, present_state,
+				(present_state & ESDHC_CLOCK_GATE_OFF), 2, 100);
+	if (ret == -ETIMEDOUT)
+		dev_warn(mmc_dev(host->mmc), "%s: card clock still not gate off in 100us!.\n", __func__);
 }
 
 static u32 esdhc_readl_le(struct sdhci_host *host, int reg)
···
 		else
 			new_val &= ~ESDHC_VENDOR_SPEC_FRC_SDCLK_ON;
 		writel(new_val, host->ioaddr + ESDHC_VENDOR_SPEC);
+		if (!(new_val & ESDHC_VENDOR_SPEC_FRC_SDCLK_ON))
+			esdhc_wait_for_card_clock_gate_off(host);
 		return;
 	case SDHCI_HOST_CONTROL2:
 		new_val = readl(host->ioaddr + ESDHC_VENDOR_SPEC);
···
 			 * For DMA access restore the levels to default value.
 			 */
 			m = readl(host->ioaddr + ESDHC_WTMK_LVL);
-			if (val & SDHCI_TRNS_DMA)
+			if (val & SDHCI_TRNS_DMA) {
 				wml = ESDHC_WTMK_LVL_WML_VAL_DEF;
-			else
+			} else {
+				u8 ctrl;
 				wml = ESDHC_WTMK_LVL_WML_VAL_MAX;
+
+				/*
+				 * Since already disable DMA mode, so also need
+				 * to clear the DMASEL. Otherwise, for standard
+				 * tuning, when send tuning command, usdhc will
+				 * still prefetch the ADMA script from wrong
+				 * DMA address, then we will see IOMMU report
+				 * some error which show lack of TLB mapping.
+				 */
+				ctrl = sdhci_readb(host, SDHCI_HOST_CONTROL);
+				ctrl &= ~SDHCI_CTRL_DMA_MASK;
+				sdhci_writeb(host, ctrl, SDHCI_HOST_CONTROL);
+			}
 			m &= ~(ESDHC_WTMK_LVL_RD_WML_MASK |
 			       ESDHC_WTMK_LVL_WR_WML_MASK);
 			m |= (wml << ESDHC_WTMK_LVL_RD_WML_SHIFT) |
···
 	int ddr_pre_div = imx_data->is_ddr ? 2 : 1;
 	int pre_div = 1;
 	int div = 1;
+	int ret;
 	u32 temp, val;
 
 	if (esdhc_is_usdhc(imx_data)) {
 		val = readl(host->ioaddr + ESDHC_VENDOR_SPEC);
 		writel(val & ~ESDHC_VENDOR_SPEC_FRC_SDCLK_ON,
 			host->ioaddr + ESDHC_VENDOR_SPEC);
+		esdhc_wait_for_card_clock_gate_off(host);
 	}
 
 	if (clock == 0) {
···
 		| (pre_div << ESDHC_PREDIV_SHIFT));
 	sdhci_writel(host, temp, ESDHC_SYSTEM_CONTROL);
 
+	/* need to wait the bit 3 of the PRSSTAT to be set, make sure card clock is stable */
+	ret = readl_poll_timeout(host->ioaddr + ESDHC_PRSSTAT, temp,
+				(temp & ESDHC_CLOCK_STABLE), 2, 100);
+	if (ret == -ETIMEDOUT)
+		dev_warn(mmc_dev(host->mmc), "card clock still not stable in 100us!.\n");
+
 	if (esdhc_is_usdhc(imx_data)) {
 		val = readl(host->ioaddr + ESDHC_VENDOR_SPEC);
 		writel(val | ESDHC_VENDOR_SPEC_FRC_SDCLK_ON,
 			host->ioaddr + ESDHC_VENDOR_SPEC);
 	}
 
-	mdelay(1);
 }
 
 static unsigned int esdhc_pltfm_get_ro(struct sdhci_host *host)
···
  */
 static void esdhc_set_strobe_dll(struct sdhci_host *host)
 {
+	struct sdhci_pltfm_host *pltfm_host = sdhci_priv(host);
+	struct pltfm_imx_data *imx_data = sdhci_pltfm_priv(pltfm_host);
+	u32 strobe_delay;
 	u32 v;
+	int ret;
 
 	/* disable clock before enabling strobe dll */
 	writel(readl(host->ioaddr + ESDHC_VENDOR_SPEC) &
 		~ESDHC_VENDOR_SPEC_FRC_SDCLK_ON,
 		host->ioaddr + ESDHC_VENDOR_SPEC);
+	esdhc_wait_for_card_clock_gate_off(host);
 
 	/* force a reset on strobe dll */
 	writel(ESDHC_STROBE_DLL_CTRL_RESET,
···
 	 * enable strobe dll ctrl and adjust the delay target
 	 * for the uSDHC loopback read clock
 	 */
+	if (imx_data->boarddata.strobe_dll_delay_target)
+		strobe_delay = imx_data->boarddata.strobe_dll_delay_target;
+	else
+		strobe_delay = ESDHC_STROBE_DLL_CTRL_SLV_DLY_TARGET_DEFAULT;
 	v = ESDHC_STROBE_DLL_CTRL_ENABLE |
 		ESDHC_STROBE_DLL_CTRL_SLV_UPDATE_INT_DEFAULT |
-		(7 << ESDHC_STROBE_DLL_CTRL_SLV_DLY_TARGET_SHIFT);
+		(strobe_delay << ESDHC_STROBE_DLL_CTRL_SLV_DLY_TARGET_SHIFT);
 	writel(v, host->ioaddr + ESDHC_STROBE_DLL_CTRL);
-	/* wait 5us to make sure strobe dll status register stable */
-	udelay(5);
-	v = readl(host->ioaddr + ESDHC_STROBE_DLL_STATUS);
-	if (!(v & ESDHC_STROBE_DLL_STS_REF_LOCK))
+
+	/* wait max 50us to get the REF/SLV lock */
+	ret = readl_poll_timeout(host->ioaddr + ESDHC_STROBE_DLL_STATUS, v,
+		((v & ESDHC_STROBE_DLL_STS_REF_LOCK) && (v & ESDHC_STROBE_DLL_STS_SLV_LOCK)), 1, 50);
+	if (ret == -ETIMEDOUT)
 		dev_warn(mmc_dev(host->mmc),
-		"warning! HS400 strobe DLL status REF not lock!\n");
-	if (!(v & ESDHC_STROBE_DLL_STS_SLV_LOCK))
-		dev_warn(mmc_dev(host->mmc),
-		"warning! HS400 strobe DLL status SLV not lock!\n");
+		"warning! HS400 strobe DLL status REF/SLV not lock in 50us, STROBE DLL status is %x!\n", v);
 }
 
 static void esdhc_reset_tuning(struct sdhci_host *host)
···
 {
 	struct sdhci_pltfm_host *pltfm_host = sdhci_priv(host);
 	struct pltfm_imx_data *imx_data = sdhci_pltfm_priv(pltfm_host);
+	struct cqhci_host *cq_host = host->mmc->cqe_private;
 	int tmp;
 
 	if (esdhc_is_usdhc(imx_data)) {
···
 			tmp = readl(host->ioaddr + ESDHC_TUNING_CTRL);
 			tmp &= ~ESDHC_STD_TUNING_EN;
 			writel(tmp, host->ioaddr + ESDHC_TUNING_CTRL);
+		}
+
+		/*
+		 * On i.MX8MM, we are running Dual Linux OS, with 1st Linux using SD Card
+		 * as rootfs storage, 2nd Linux using eMMC as rootfs storage. We let the
+		 * the 1st linux configure power/clock for the 2nd Linux.
+		 *
+		 * When the 2nd Linux is booting into rootfs stage, we let the 1st Linux
+		 * to destroy the 2nd linux, then restart the 2nd linux, we met SDHCI dump.
+		 * After we clear the pending interrupt and halt CQCTL, issue gone.
+		 */
+		if (cq_host) {
+			tmp = cqhci_readl(cq_host, CQHCI_IS);
+			cqhci_writel(cq_host, tmp, CQHCI_IS);
+			cqhci_writel(cq_host, CQHCI_HALT, CQHCI_CTL);
 		}
 	}
 }
···
 	of_property_read_u32(np, "fsl,tuning-start-tap",
 			     &boarddata->tuning_start_tap);
 
+	of_property_read_u32(np, "fsl,strobe-dll-delay-target",
+				&boarddata->strobe_dll_delay_target);
 	if (of_find_property(np, "no-1-8-v", NULL))
 		host->quirks2 |= SDHCI_QUIRK2_NO_1_8_V;
 
···
 	imx_data->pinctrl = devm_pinctrl_get(&pdev->dev);
 	if (IS_ERR(imx_data->pinctrl)) {
 		err = PTR_ERR(imx_data->pinctrl);
-		goto disable_ahb_clk;
+		dev_warn(mmc_dev(host->mmc), "could not get pinctrl\n");
 	}
 
 	if (esdhc_is_usdhc(imx_data)) {
···
 
 	if (imx_data->socdata->flags & ESDHC_FLAG_HS400)
 		host->quirks2 |= SDHCI_QUIRK2_CAPS_BIT63_FOR_HS400;
+
+	if (imx_data->socdata->flags & ESDHC_FLAG_BROKEN_AUTO_CMD23)
+		host->quirks2 |= SDHCI_QUIRK2_ACMD23_BROKEN;
 
 	if (imx_data->socdata->flags & ESDHC_FLAG_HS400_ES) {
 		host->mmc->caps2 |= MMC_CAP2_HS400_ES;
···
 static int sdhci_esdhc_suspend(struct device *dev)
 {
 	struct sdhci_host *host = dev_get_drvdata(dev);
+	struct sdhci_pltfm_host *pltfm_host = sdhci_priv(host);
+	struct pltfm_imx_data *imx_data = sdhci_pltfm_priv(pltfm_host);
 	int ret;
 
 	if (host->mmc->caps2 & MMC_CAP2_CQE) {
···
 			return ret;
 	}
 
+	if ((imx_data->socdata->flags & ESDHC_FLAG_STATE_LOST_IN_LPMODE) &&
+		(host->tuning_mode != SDHCI_TUNING_MODE_1)) {
+		mmc_retune_timer_stop(host->mmc);
+		mmc_retune_needed(host->mmc);
+	}
+
 	if (host->tuning_mode != SDHCI_TUNING_MODE_3)
 		mmc_retune_needed(host->mmc);
 
-	return sdhci_suspend_host(host);
+	ret = sdhci_suspend_host(host);
+	if (!ret)
+		return pinctrl_pm_select_sleep_state(dev);
+
+	return ret;
 }
 
 static int sdhci_esdhc_resume(struct device *dev)
 {
 	struct sdhci_host *host = dev_get_drvdata(dev);
 	int ret;
+
+	ret = pinctrl_pm_select_default_state(dev);
+	if (ret)
+		return ret;
 
 	/* re-initialize hw state in case it's lost in low power mode */
 	sdhci_esdhc_imx_hwinit(host);
···
 
 	if (imx_data->socdata->flags & ESDHC_FLAG_PMQOS)
 		cpu_latency_qos_add_request(&imx_data->pm_qos_req, 0);
+
+	if (imx_data->socdata->flags & ESDHC_FLAG_CLK_RATE_LOST_IN_PM_RUNTIME)
+		clk_set_rate(imx_data->clk_per, pltfm_host->clock);
 
 	err = clk_prepare_enable(imx_data->clk_ahb);
 	if (err)

drivers/mmc/host/sdhci-esdhc.h

···
 
 /* Present State Register */
 #define ESDHC_PRSSTAT			0x24
+#define ESDHC_CLOCK_GATE_OFF		0x00000080
 #define ESDHC_CLOCK_STABLE		0x00000008
 
 /* Protocol Control Register */

drivers/mmc/host/sdhci-iproc.c

···
 	.mmc_caps = 0x00000000,
 };
 
+static const struct sdhci_ops sdhci_iproc_bcm2711_ops = {
+	.read_l = sdhci_iproc_readl,
+	.read_w = sdhci_iproc_readw,
+	.read_b = sdhci_iproc_readb,
+	.write_l = sdhci_iproc_writel,
+	.write_w = sdhci_iproc_writew,
+	.write_b = sdhci_iproc_writeb,
+	.set_clock = sdhci_set_clock,
+	.set_power = sdhci_set_power_and_bus_voltage,
+	.get_max_clock = sdhci_iproc_get_max_clock,
+	.set_bus_width = sdhci_set_bus_width,
+	.reset = sdhci_reset,
+	.set_uhs_signaling = sdhci_set_uhs_signaling,
+};
+
 static const struct sdhci_pltfm_data sdhci_bcm2711_pltfm_data = {
 	.quirks = SDHCI_QUIRK_MULTIBLOCK_READ_ACMD12,
-	.ops = &sdhci_iproc_32only_ops,
+	.ops = &sdhci_iproc_bcm2711_ops,
 };
 
 static const struct sdhci_iproc_data bcm2711_data = {

drivers/mmc/host/sdhci-milbeaut.c

···
 	}
 }
 
-static void sdhci_milbeaut_set_power(struct sdhci_host *host,
-			unsigned char mode, unsigned short vdd)
-{
-	if (!IS_ERR(host->mmc->supply.vmmc)) {
-		struct mmc_host *mmc = host->mmc;
-
-		mmc_regulator_set_ocr(mmc, mmc->supply.vmmc, vdd);
-	}
-	sdhci_set_power_noreg(host, mode, vdd);
-}
-
 static const struct sdhci_ops sdhci_milbeaut_ops = {
 	.voltage_switch = sdhci_milbeaut_soft_voltage_switch,
 	.get_min_clock = sdhci_milbeaut_get_min_clock,
···
 	.set_clock = sdhci_set_clock,
 	.set_bus_width = sdhci_set_bus_width,
 	.set_uhs_signaling = sdhci_set_uhs_signaling,
-	.set_power = sdhci_milbeaut_set_power,
+	.set_power = sdhci_set_power_and_bus_voltage,
 };
 
 static void sdhci_milbeaut_bridge_reset(struct sdhci_host *host,

drivers/mmc/host/sdhci-msm.c

···
 		goto out;
 	}
 
-	config = readl_relaxed(host->ioaddr + msm_offset->core_vendor_spec3);
-	config |= CORE_PWRSAVE_DLL;
-	writel_relaxed(config, host->ioaddr + msm_offset->core_vendor_spec3);
+	/*
+	 * Set CORE_PWRSAVE_DLL bit in CORE_VENDOR_SPEC3.
+	 * When MCLK is gated OFF, it is not gated for less than 0.5us
+	 * and MCLK must be switched on for at-least 1us before DATA
+	 * starts coming. Controllers with 14lpp and later tech DLL cannot
+	 * guarantee above requirement. So PWRSAVE_DLL should not be
+	 * turned on for host controllers using this DLL.
+	 */
+	if (!msm_host->use_14lpp_dll_reset) {
+		config = readl_relaxed(host->ioaddr +
+				msm_offset->core_vendor_spec3);
+		config |= CORE_PWRSAVE_DLL;
+		writel_relaxed(config, host->ioaddr +
+				msm_offset->core_vendor_spec3);
+	}
 
 	/*
 	 * Drain writebuffer to ensure above DLL calibration
···
 	pr_debug("%s: supported caps: 0x%08x\n", mmc_hostname(mmc), caps);
 }
 
+static void sdhci_msm_reset(struct sdhci_host *host, u8 mask)
+{
+	if ((host->mmc->caps2 & MMC_CAP2_CQE) && (mask & SDHCI_RESET_ALL))
+		cqhci_deactivate(host->mmc);
+	sdhci_reset(host, mask);
+}
+
 static const struct sdhci_msm_variant_ops mci_var_ops = {
 	.msm_readl_relaxed = sdhci_msm_mci_variant_readl_relaxed,
 	.msm_writel_relaxed = sdhci_msm_mci_variant_writel_relaxed,
···
 MODULE_DEVICE_TABLE(of, sdhci_msm_dt_match);
 
 static const struct sdhci_ops sdhci_msm_ops = {
-	.reset = sdhci_reset,
+	.reset = sdhci_msm_reset,
 	.set_clock = sdhci_msm_set_clock,
 	.get_min_clock = sdhci_msm_get_min_clock,
 	.get_max_clock = sdhci_msm_get_max_clock,

drivers/mmc/host/sdhci-of-arasan.c

···
 	return -EINVAL;
 }
 
-static void sdhci_arasan_set_power(struct sdhci_host *host, unsigned char mode,
-		     unsigned short vdd)
-{
-	if (!IS_ERR(host->mmc->supply.vmmc)) {
-		struct mmc_host *mmc = host->mmc;
-
-		mmc_regulator_set_ocr(mmc, mmc->supply.vmmc, vdd);
-	}
-	sdhci_set_power_noreg(host, mode, vdd);
-}
-
 static const struct sdhci_ops sdhci_arasan_ops = {
 	.set_clock = sdhci_arasan_set_clock,
 	.get_max_clock = sdhci_pltfm_clk_get_max_clock,
···
 	.set_bus_width = sdhci_set_bus_width,
 	.reset = sdhci_arasan_reset,
 	.set_uhs_signaling = sdhci_set_uhs_signaling,
-	.set_power = sdhci_arasan_set_power,
+	.set_power = sdhci_set_power_and_bus_voltage,
 };
 
 static const struct sdhci_pltfm_data sdhci_arasan_pdata = {
···
 
 static struct sdhci_arasan_of_data sdhci_arasan_data = {
 	.pdata = &sdhci_arasan_pdata,
+};
+
+static const struct sdhci_pltfm_data sdhci_arasan_zynqmp_pdata = {
+	.ops = &sdhci_arasan_ops,
+	.quirks2 = SDHCI_QUIRK2_PRESET_VALUE_BROKEN |
+			SDHCI_QUIRK2_CLOCK_DIV_ZERO_BROKEN |
+			SDHCI_QUIRK2_STOP_WITH_TC,
+};
+
+static struct sdhci_arasan_of_data sdhci_arasan_zynqmp_data = {
+	.pdata = &sdhci_arasan_zynqmp_pdata,
 };
 
 static u32 sdhci_arasan_cqhci_irq(struct sdhci_host *host, u32 intmask)
···
 	.set_bus_width = sdhci_set_bus_width,
 	.reset = sdhci_arasan_reset,
 	.set_uhs_signaling = sdhci_set_uhs_signaling,
-	.set_power = sdhci_arasan_set_power,
+	.set_power = sdhci_set_power_and_bus_voltage,
 	.irq = sdhci_arasan_cqhci_irq,
 };
 
···
 	},
 	{
 		.compatible = "xlnx,zynqmp-8.9a",
-		.data = &sdhci_arasan_data,
+		.data = &sdhci_arasan_zynqmp_data,
 	},
 	{ /* sentinel */ }
 };
···
 	.recalc_rate = sdhci_arasan_sampleclk_recalc_rate,
 	.set_phase = sdhci_zynqmp_sampleclk_set_phase,
 };
+
+static void arasan_zynqmp_dll_reset(struct sdhci_host *host, u32 deviceid)
+{
+	struct sdhci_pltfm_host *pltfm_host = sdhci_priv(host);
+	struct sdhci_arasan_data *sdhci_arasan = sdhci_pltfm_priv(pltfm_host);
+	struct sdhci_arasan_zynqmp_clk_data *zynqmp_clk_data =
+		sdhci_arasan->clk_data.clk_of_data;
+	const struct zynqmp_eemi_ops *eemi_ops = zynqmp_clk_data->eemi_ops;
+	u16 clk;
+
+	clk = sdhci_readw(host, SDHCI_CLOCK_CONTROL);
+	clk &= ~(SDHCI_CLOCK_CARD_EN | SDHCI_CLOCK_INT_EN);
+	sdhci_writew(host, clk, SDHCI_CLOCK_CONTROL);
+
+	/* Issue DLL Reset */
+	eemi_ops->ioctl(deviceid, IOCTL_SD_DLL_RESET,
+			PM_DLL_RESET_PULSE, 0, NULL);
+
+	clk = sdhci_readw(host, SDHCI_CLOCK_CONTROL);
+
+	sdhci_enable_clk(host, clk);
+}
+
+static int arasan_zynqmp_execute_tuning(struct mmc_host *mmc, u32 opcode)
+{
+	struct sdhci_host *host = mmc_priv(mmc);
+	struct sdhci_pltfm_host *pltfm_host = sdhci_priv(host);
+	struct sdhci_arasan_data *sdhci_arasan = sdhci_pltfm_priv(pltfm_host);
+	struct clk_hw *hw = &sdhci_arasan->clk_data.sdcardclk_hw;
+	const char *clk_name = clk_hw_get_name(hw);
+	u32 device_id = !strcmp(clk_name, "clk_out_sd0") ? NODE_SD_0 :
+							   NODE_SD_1;
+	int err;
+
+	arasan_zynqmp_dll_reset(host, device_id);
+
+	err = sdhci_execute_tuning(mmc, opcode);
+	if (err)
+		return err;
+
+	arasan_zynqmp_dll_reset(host, device_id);
+
+	return 0;
+}
 
 /**
  * sdhci_arasan_update_clockmultiplier - Set corecfg_clockmultiplier
···
 
 		zynqmp_clk_data->eemi_ops = eemi_ops;
 		sdhci_arasan->clk_data.clk_of_data = zynqmp_clk_data;
+		host->mmc_host_ops.execute_tuning =
+			arasan_zynqmp_execute_tuning;
 	}
 
 	arasan_dt_parse_clk_phases(&pdev->dev, &sdhci_arasan->clk_data);

drivers/mmc/host/sdhci-of-at91.c

···
 	sdhci_writew(host, clk, SDHCI_CLOCK_CONTROL);
 }
 
-/*
- * In this specific implementation of the SDHCI controller, the power register
- * needs to have a valid voltage set even when the power supply is managed by
- * an external regulator.
- */
-static void sdhci_at91_set_power(struct sdhci_host *host, unsigned char mode,
-		     unsigned short vdd)
-{
-	if (!IS_ERR(host->mmc->supply.vmmc)) {
-		struct mmc_host *mmc = host->mmc;
-
-		mmc_regulator_set_ocr(mmc, mmc->supply.vmmc, vdd);
-	}
-	sdhci_set_power_noreg(host, mode, vdd);
-}
-
 static void sdhci_at91_set_uhs_signaling(struct sdhci_host *host,
 					 unsigned int timing)
 {
···
 	.set_bus_width		= sdhci_set_bus_width,
 	.reset			= sdhci_at91_reset,
 	.set_uhs_signaling	= sdhci_at91_set_uhs_signaling,
-	.set_power		= sdhci_at91_set_power,
+	.set_power		= sdhci_set_power_and_bus_voltage,
 };
 
 static const struct sdhci_pltfm_data sdhci_sama5d2_pdata = {
···
 	/* Set capabilities in ro mode. */
 	writel(0, host->ioaddr + SDMMC_CACR);
 
-	dev_info(dev, "update clk mul to %u as gck rate is %u Hz and clk base is %u Hz\n",
-		 clk_mul, gck_rate, clk_base_rate);
+	dev_dbg(dev, "update clk mul to %u as gck rate is %u Hz and clk base is %u Hz\n",
+		clk_mul, gck_rate, clk_base_rate);
 
 	/*
 	 * We have to set preset values because it depends on the clk_mul

drivers/mmc/host/sdhci-omap.c

···
 	struct pinctrl		*pinctrl;
 	struct pinctrl_state	**pinctrl_state;
 	bool			is_tuning;
+	/* Omap specific context save */
+	u32			con;
+	u32			hctl;
+	u32			sysctl;
+	u32			capa;
 };
 
 static void sdhci_omap_start_clock(struct sdhci_omap_host *omap_host);
···
 
 	return 0;
 }
+#ifdef CONFIG_PM_SLEEP
+static void sdhci_omap_context_save(struct sdhci_omap_host *omap_host)
+{
+	omap_host->con = sdhci_omap_readl(omap_host, SDHCI_OMAP_CON);
+	omap_host->hctl = sdhci_omap_readl(omap_host, SDHCI_OMAP_HCTL);
+	omap_host->capa = sdhci_omap_readl(omap_host, SDHCI_OMAP_CAPA);
+}
+
+static void sdhci_omap_context_restore(struct sdhci_omap_host *omap_host)
+{
+	sdhci_omap_writel(omap_host, SDHCI_OMAP_CON, omap_host->con);
+	sdhci_omap_writel(omap_host, SDHCI_OMAP_HCTL, omap_host->hctl);
+	sdhci_omap_writel(omap_host, SDHCI_OMAP_CAPA, omap_host->capa);
+}
+
+static int __maybe_unused sdhci_omap_suspend(struct device *dev)
+{
+	struct sdhci_host *host = dev_get_drvdata(dev);
+	struct sdhci_pltfm_host *pltfm_host = sdhci_priv(host);
+	struct sdhci_omap_host *omap_host = sdhci_pltfm_priv(pltfm_host);
+
+	sdhci_suspend_host(host);
+
+	sdhci_omap_context_save(omap_host);
+
+	pinctrl_pm_select_idle_state(dev);
+
+	pm_runtime_force_suspend(dev);
+
+	return 0;
+}
+
+static int __maybe_unused sdhci_omap_resume(struct device *dev)
+{
+	struct sdhci_host *host = dev_get_drvdata(dev);
+	struct sdhci_pltfm_host *pltfm_host = sdhci_priv(host);
+	struct sdhci_omap_host *omap_host = sdhci_pltfm_priv(pltfm_host);
+
+	pm_runtime_force_resume(dev);
+
+	pinctrl_pm_select_default_state(dev);
+
+	sdhci_omap_context_restore(omap_host);
+
+	sdhci_resume_host(host);
+
+	return 0;
+}
+#endif
+static SIMPLE_DEV_PM_OPS(sdhci_omap_dev_pm_ops, sdhci_omap_suspend,
+			 sdhci_omap_resume);
 
 static struct platform_driver sdhci_omap_driver = {
 	.probe = sdhci_omap_probe,
 	.remove = sdhci_omap_remove,
 	.driver = {
 		   .name = "sdhci-omap",
+		   .pm = &sdhci_omap_dev_pm_ops,
 		   .of_match_table = omap_sdhci_match,
 		  },
 };

drivers/mmc/host/sdhci-pci.h

···
 	bool			cd_override_level;
 
 	void (*hw_reset)(struct sdhci_host *host);
-	unsigned long		private[0] ____cacheline_aligned;
+	unsigned long		private[] ____cacheline_aligned;
 };
 
 struct sdhci_pci_chip {

drivers/mmc/host/sdhci-pltfm.h

···
 	unsigned int clock;
 	u16 xfer_mode_shadow;
 
-	unsigned long private[0] ____cacheline_aligned;
+	unsigned long private[] ____cacheline_aligned;
 };
 
 #ifdef CONFIG_MMC_SDHCI_BIG_ENDIAN_32BIT_BYTE_SWAPPER

drivers/mmc/host/sdhci-sprd.c

···
 #include <linux/slab.h>
 
 #include "sdhci-pltfm.h"
+#include "mmc_hsq.h"
 
 /* SDHCI_ARGUMENT2 register high 16bit */
 #define SDHCI_SPRD_ARG2_STUFF		GENMASK(31, 16)
···
 	return 0;
 }
 
+static void sdhci_sprd_request_done(struct sdhci_host *host,
+				    struct mmc_request *mrq)
+{
+	/* Validate if the request was from software queue firstly. */
+	if (mmc_hsq_finalize_request(host->mmc, mrq))
+		return;
+
+	 mmc_request_done(host->mmc, mrq);
+}
+
 static struct sdhci_ops sdhci_sprd_ops = {
 	.read_l = sdhci_sprd_readl,
 	.write_l = sdhci_sprd_writel,
···
 	.hw_reset = sdhci_sprd_hw_reset,
 	.get_max_timeout_count = sdhci_sprd_get_max_timeout_count,
 	.get_ro = sdhci_sprd_get_ro,
+	.request_done = sdhci_sprd_request_done,
 };
 
 static void sdhci_sprd_request(struct mmc_host *mmc, struct mmc_request *mrq)
···
 {
 	struct sdhci_host *host;
 	struct sdhci_sprd_host *sprd_host;
+	struct mmc_hsq *hsq;
 	struct clk *clk;
 	int ret = 0;
 
···
 		sdhci_sprd_voltage_switch;
 
 	host->mmc->caps = MMC_CAP_SD_HIGHSPEED | MMC_CAP_MMC_HIGHSPEED |
-		MMC_CAP_ERASE | MMC_CAP_CMD23;
+		MMC_CAP_ERASE | MMC_CAP_CMD23 | MMC_CAP_WAIT_WHILE_BUSY;
 	ret = mmc_of_parse(host->mmc);
 	if (ret)
 		goto pltfm_free;
···
 
 	sprd_host->flags = host->flags;
 
+	hsq = devm_kzalloc(&pdev->dev, sizeof(*hsq), GFP_KERNEL);
+	if (!hsq) {
+		ret = -ENOMEM;
+		goto err_cleanup_host;
+	}
+
+	ret = mmc_hsq_init(hsq, host->mmc);
+	if (ret)
+		goto err_cleanup_host;
+
+	host->always_defer_done = true;
+
 	ret = __sdhci_add_host(host);
 	if (ret)
 		goto err_cleanup_host;
···
 	struct sdhci_host *host = dev_get_drvdata(dev);
 	struct sdhci_sprd_host *sprd_host = TO_SPRD_HOST(host);
 
+	mmc_hsq_suspend(host->mmc);
 	sdhci_runtime_suspend_host(host);
 
 	clk_disable_unprepare(sprd_host->clk_sdio);
···
 		goto clk_disable;
 
 	sdhci_runtime_resume_host(host, 1);
+	mmc_hsq_resume(host->mmc);
+
 	return 0;
 
 clk_disable:

drivers/mmc/host/sdhci-tegra.c

···
 #define SDHCI_TEGRA_CAP_OVERRIDES_DQS_TRIM_SHIFT	8
 
 #define SDHCI_TEGRA_VENDOR_MISC_CTRL			0x120
+#define SDHCI_MISC_CTRL_ERASE_TIMEOUT_LIMIT		BIT(0)
 #define SDHCI_MISC_CTRL_ENABLE_SDR104			0x8
 #define SDHCI_MISC_CTRL_ENABLE_SDR50			0x10
 #define SDHCI_MISC_CTRL_ENABLE_SDHCI_SPEC_300		0x20
···
 	return 0;
 }
 
+static void tegra_sdhci_set_timeout(struct sdhci_host *host,
+				    struct mmc_command *cmd)
+{
+	u32 val;
+
+	/*
+	 * HW busy detection timeout is based on programmed data timeout
+	 * counter and maximum supported timeout is 11s which may not be
+	 * enough for long operations like cache flush, sleep awake, erase.
+	 *
+	 * ERASE_TIMEOUT_LIMIT bit of VENDOR_MISC_CTRL register allows
+	 * host controller to wait for busy state until the card is busy
+	 * without HW timeout.
+	 *
+	 * So, use infinite busy wait mode for operations that may take
+	 * more than maximum HW busy timeout of 11s otherwise use finite
+	 * busy wait mode.
+	 */
+	val = sdhci_readl(host, SDHCI_TEGRA_VENDOR_MISC_CTRL);
+	if (cmd && cmd->busy_timeout >= 11 * HZ)
+		val |= SDHCI_MISC_CTRL_ERASE_TIMEOUT_LIMIT;
+	else
+		val &= ~SDHCI_MISC_CTRL_ERASE_TIMEOUT_LIMIT;
+	sdhci_writel(host, val, SDHCI_TEGRA_VENDOR_MISC_CTRL);
+
+	__sdhci_set_timeout(host, cmd);
+}
+
 static const struct cqhci_host_ops sdhci_tegra_cqhci_ops = {
 	.write_l    = tegra_cqhci_writel,
 	.enable	= sdhci_tegra_cqe_enable,
···
 	.set_uhs_signaling = tegra_sdhci_set_uhs_signaling,
 	.voltage_switch = tegra_sdhci_voltage_switch,
 	.get_max_clock = tegra_sdhci_get_max_clock,
+	.set_timeout = tegra_sdhci_set_timeout,
 };
 
 static const struct sdhci_pltfm_data sdhci_tegra210_pdata = {
···
 	.voltage_switch = tegra_sdhci_voltage_switch,
 	.get_max_clock = tegra_sdhci_get_max_clock,
 	.irq = sdhci_tegra_cqhci_irq,
+	.set_timeout = tegra_sdhci_set_timeout,
 };
 
 static const struct sdhci_pltfm_data sdhci_tegra186_pdata = {
···
 	if (tegra_host->soc_data->nvquirks & NVQUIRK_ENABLE_DDR50)
 		host->mmc->caps |= MMC_CAP_1_8V_DDR;
 
-	/* R1B responses is required to properly manage HW busy detection. */
-	host->mmc->caps |= MMC_CAP_NEED_RSP_BUSY;
+	/* HW busy detection is supported, but R1B responses are required. */
+	host->mmc->caps |= MMC_CAP_WAIT_WHILE_BUSY | MMC_CAP_NEED_RSP_BUSY;
 
 	tegra_sdhci_parse_dt(host);
 

drivers/mmc/host/sdhci.c

···
  *     - JMicron (hardware and technical support)
  */
 
+#include <linux/bitfield.h>
 #include <linux/delay.h>
 #include <linux/dmaengine.h>
 #include <linux/ktime.h>
···
 	u32 present;
 
 	if ((host->quirks & SDHCI_QUIRK_BROKEN_CARD_DETECTION) ||
-	    !mmc_card_is_removable(host->mmc))
+	    !mmc_card_is_removable(host->mmc) || mmc_can_gpio_cd(host->mmc))
 		return;
 
 	if (enable) {
···
 
 			clk = sdhci_readw(host, SDHCI_CLOCK_CONTROL);
 			pre_val = sdhci_get_preset_value(host);
-			div = (pre_val & SDHCI_PRESET_SDCLK_FREQ_MASK)
-				>> SDHCI_PRESET_SDCLK_FREQ_SHIFT;
+			div = FIELD_GET(SDHCI_PRESET_SDCLK_FREQ_MASK, pre_val);
 			if (host->clk_mul &&
-				(pre_val & SDHCI_PRESET_CLKGEN_SEL_MASK)) {
+				(pre_val & SDHCI_PRESET_CLKGEN_SEL)) {
 				clk = SDHCI_PROG_CLOCK_MODE;
 				real_div = div + 1;
 				clk_mul = host->clk_mul;
···
 }
 EXPORT_SYMBOL_GPL(sdhci_set_power);
 
+/*
+ * Some controllers need to configure a valid bus voltage on their power
+ * register regardless of whether an external regulator is taking care of power
+ * supply. This helper function takes care of it if set as the controller's
+ * sdhci_ops.set_power callback.
+ */
+void sdhci_set_power_and_bus_voltage(struct sdhci_host *host,
+				     unsigned char mode,
+				     unsigned short vdd)
+{
+	if (!IS_ERR(host->mmc->supply.vmmc)) {
+		struct mmc_host *mmc = host->mmc;
+
+		mmc_regulator_set_ocr(mmc, mmc->supply.vmmc, vdd);
+	}
+	sdhci_set_power_noreg(host, mode, vdd);
+}
+EXPORT_SYMBOL_GPL(sdhci_set_power_and_bus_voltage);
+
 /*****************************************************************************\
  *                                                                           *
  * MMC callbacks                                                             *
···
 
 			sdhci_enable_preset_value(host, true);
 			preset = sdhci_get_preset_value(host);
-			ios->drv_type = (preset & SDHCI_PRESET_DRV_MASK)
-				>> SDHCI_PRESET_DRV_SHIFT;
+			ios->drv_type = FIELD_GET(SDHCI_PRESET_DRV_MASK,
+						  preset);
 		}
 
 		/* Re-enable SD Clock */
···
 
 	spin_unlock_irqrestore(&host->lock, flags);
 
-	mmc_request_done(host->mmc, mrq);
+	if (host->ops->request_done)
+		host->ops->request_done(host, mrq);
+	else
+		mmc_request_done(host->mmc, mrq);
 
 	return false;
 }
···
 {
 	struct mmc_data *data = mrq->data;
 
-	return host->pending_reset ||
+	return host->pending_reset || host->always_defer_done ||
 	       ((host->flags & SDHCI_REQ_USE_DMA) && data &&
 		data->host_cookie == COOKIE_MAPPED);
 }
···
 
 	/* Process mrqs ready for immediate completion */
 	for (i = 0; i < SDHCI_MAX_MRQS; i++) {
-		if (mrqs_done[i])
+		if (!mrqs_done[i])
+			continue;
+
+		if (host->ops->request_done)
+			host->ops->request_done(host, mrqs_done[i]);
+		else
 			mmc_request_done(host->mmc, mrqs_done[i]);
 	}
 

drivers/mmc/host/sdhci.h

···
 #ifndef __SDHCI_HW_H
 #define __SDHCI_HW_H
 
+#include <linux/bits.h>
 #include <linux/scatterlist.h>
 #include <linux/compiler.h>
 #include <linux/types.h>
···
 #define SDHCI_PRESET_FOR_SDR104        0x6C
 #define SDHCI_PRESET_FOR_DDR50 0x6E
 #define SDHCI_PRESET_FOR_HS400 0x74 /* Non-standard */
-#define SDHCI_PRESET_DRV_MASK  0xC000
-#define SDHCI_PRESET_DRV_SHIFT  14
-#define SDHCI_PRESET_CLKGEN_SEL_MASK   0x400
-#define SDHCI_PRESET_CLKGEN_SEL_SHIFT	10
-#define SDHCI_PRESET_SDCLK_FREQ_MASK   0x3FF
-#define SDHCI_PRESET_SDCLK_FREQ_SHIFT	0
+#define SDHCI_PRESET_DRV_MASK		GENMASK(15, 14)
+#define SDHCI_PRESET_CLKGEN_SEL		BIT(10)
+#define SDHCI_PRESET_SDCLK_FREQ_MASK	GENMASK(9, 0)
 
 #define SDHCI_SLOT_INT_STATUS	0xFC
 
···
 	bool irq_wake_enabled;	/* IRQ wakeup is enabled */
 	bool v4_mode;		/* Host Version 4 Enable */
 	bool use_external_dma;	/* Host selects to use external DMA */
+	bool always_defer_done;	/* Always defer to complete requests */
 
 	struct mmc_request *mrqs_done[SDHCI_MAX_MRQS];	/* Requests done */
 	struct mmc_command *cmd;	/* Current command */
···
 
 	u64			data_timeout;
 
-	unsigned long private[0] ____cacheline_aligned;
+	unsigned long private[] ____cacheline_aligned;
 };
 
 struct sdhci_ops {
···
 	void	(*voltage_switch)(struct sdhci_host *host);
 	void	(*adma_write_desc)(struct sdhci_host *host, void **desc,
 				   dma_addr_t addr, int len, unsigned int cmd);
+	void	(*request_done)(struct sdhci_host *host,
+				struct mmc_request *mrq);
 };
 
 #ifdef CONFIG_MMC_SDHCI_IO_ACCESSORS
···
 void sdhci_enable_clk(struct sdhci_host *host, u16 clk);
 void sdhci_set_power(struct sdhci_host *host, unsigned char mode,
 		     unsigned short vdd);
+void sdhci_set_power_and_bus_voltage(struct sdhci_host *host,
+				     unsigned char mode,
+				     unsigned short vdd);
 void sdhci_set_power_noreg(struct sdhci_host *host, unsigned char mode,
 			   unsigned short vdd);
 void sdhci_request(struct mmc_host *mmc, struct mmc_request *mrq);

drivers/mmc/host/sdhci_am654.c

···
 
 struct sdhci_am654_data {
 	struct regmap *base;
-	int otap_del_sel;
+	bool legacy_otapdly;
+	int otap_del_sel[11];
 	int trm_icp;
 	int drv_strength;
 	bool dll_on;
···
 #define DLL_PRESENT	(1 << 3)
 };
 
-static void sdhci_am654_set_clock(struct sdhci_host *host, unsigned int clock)
+struct timing_data {
+	const char *binding;
+	u32 capability;
+};
+
+static const struct timing_data td[] = {
+	[MMC_TIMING_LEGACY] = {"ti,otap-del-sel-legacy", 0},
+	[MMC_TIMING_MMC_HS] = {"ti,otap-del-sel-mmc-hs", MMC_CAP_MMC_HIGHSPEED},
+	[MMC_TIMING_SD_HS]  = {"ti,otap-del-sel-sd-hs", MMC_CAP_SD_HIGHSPEED},
+	[MMC_TIMING_UHS_SDR12] = {"ti,otap-del-sel-sdr12", MMC_CAP_UHS_SDR12},
+	[MMC_TIMING_UHS_SDR25] = {"ti,otap-del-sel-sdr25", MMC_CAP_UHS_SDR25},
+	[MMC_TIMING_UHS_SDR50] = {"ti,otap-del-sel-sdr50", MMC_CAP_UHS_SDR50},
+	[MMC_TIMING_UHS_SDR104] = {"ti,otap-del-sel-sdr104",
+				   MMC_CAP_UHS_SDR104},
+	[MMC_TIMING_UHS_DDR50] = {"ti,otap-del-sel-ddr50", MMC_CAP_UHS_DDR50},
+	[MMC_TIMING_MMC_DDR52] = {"ti,otap-del-sel-ddr52", MMC_CAP_DDR},
+	[MMC_TIMING_MMC_HS200] = {"ti,otap-del-sel-hs200", MMC_CAP2_HS200},
+	[MMC_TIMING_MMC_HS400] = {"ti,otap-del-sel-hs400", MMC_CAP2_HS400},
+};
+
+static void sdhci_am654_setup_dll(struct sdhci_host *host, unsigned int clock)
 {
 	struct sdhci_pltfm_host *pltfm_host = sdhci_priv(host);
 	struct sdhci_am654_data *sdhci_am654 = sdhci_pltfm_priv(pltfm_host);
 	int sel50, sel100, freqsel;
 	u32 mask, val;
 	int ret;
+
+	if (sdhci_am654->flags & FREQSEL_2_BIT) {
+		switch (clock) {
+		case 200000000:
+			sel50 = 0;
+			sel100 = 0;
+			break;
+		case 100000000:
+			sel50 = 0;
+			sel100 = 1;
+			break;
+		default:
+			sel50 = 1;
+			sel100 = 0;
+		}
+
+		/* Configure PHY DLL frequency */
+		mask = SEL50_MASK | SEL100_MASK;
+		val = (sel50 << SEL50_SHIFT) | (sel100 << SEL100_SHIFT);
+		regmap_update_bits(sdhci_am654->base, PHY_CTRL5, mask, val);
+
+	} else {
+		switch (clock) {
+		case 200000000:
+			freqsel = 0x0;
+			break;
+		default:
+			freqsel = 0x4;
+		}
+
+		regmap_update_bits(sdhci_am654->base, PHY_CTRL5, FREQSEL_MASK,
+				   freqsel << FREQSEL_SHIFT);
+	}
+	/* Configure DLL TRIM */
+	mask = DLL_TRIM_ICP_MASK;
+	val = sdhci_am654->trm_icp << DLL_TRIM_ICP_SHIFT;
+
+	/* Configure DLL driver strength */
+	mask |= DR_TY_MASK;
+	val |= sdhci_am654->drv_strength << DR_TY_SHIFT;
+	regmap_update_bits(sdhci_am654->base, PHY_CTRL1, mask, val);
+
+	/* Enable DLL */
+	regmap_update_bits(sdhci_am654->base, PHY_CTRL1, ENDLL_MASK,
+			   0x1 << ENDLL_SHIFT);
+	/*
+	 * Poll for DLL ready. Use a one second timeout.
+	 * Works in all experiments done so far
+	 */
+	ret = regmap_read_poll_timeout(sdhci_am654->base, PHY_STAT1, val,
+				       val & DLLRDY_MASK, 1000, 1000000);
+	if (ret) {
+		dev_err(mmc_dev(host->mmc), "DLL failed to relock\n");
+		return;
+	}
+
+	sdhci_am654->dll_on = true;
+}
+
+static void sdhci_am654_set_clock(struct sdhci_host *host, unsigned int clock)
+{
+	struct sdhci_pltfm_host *pltfm_host = sdhci_priv(host);
+	struct sdhci_am654_data *sdhci_am654 = sdhci_pltfm_priv(pltfm_host);
+	unsigned char timing = host->mmc->ios.timing;
+	u32 otap_del_sel;
+	u32 otap_del_ena;
+	u32 mask, val;
 
 	if (sdhci_am654->dll_on) {
 		regmap_update_bits(sdhci_am654->base, PHY_CTRL1, ENDLL_MASK, 0);
···
 
 	if (clock > CLOCK_TOO_SLOW_HZ) {
 		/* Setup DLL Output TAP delay */
+		if (sdhci_am654->legacy_otapdly)
+			otap_del_sel = sdhci_am654->otap_del_sel[0];
+		else
+			otap_del_sel = sdhci_am654->otap_del_sel[timing];
+
+		otap_del_ena = (timing > MMC_TIMING_UHS_SDR25) ? 1 : 0;
+
 		mask = OTAPDLYENA_MASK | OTAPDLYSEL_MASK;
-		val = (1 << OTAPDLYENA_SHIFT) |
-		      (sdhci_am654->otap_del_sel << OTAPDLYSEL_SHIFT);
-		regmap_update_bits(sdhci_am654->base, PHY_CTRL4, mask, val);
+		val = (otap_del_ena << OTAPDLYENA_SHIFT) |
+		      (otap_del_sel << OTAPDLYSEL_SHIFT);
+
 		/* Write to STRBSEL for HS400 speed mode */
-		if (host->mmc->ios.timing == MMC_TIMING_MMC_HS400) {
+		if (timing == MMC_TIMING_MMC_HS400) {
 			if (sdhci_am654->flags & STRBSEL_4_BIT)
-				mask = STRBSEL_4BIT_MASK;
+				mask |= STRBSEL_4BIT_MASK;
 			else
-				mask = STRBSEL_8BIT_MASK;
+				mask |= STRBSEL_8BIT_MASK;
 
-			regmap_update_bits(sdhci_am654->base, PHY_CTRL4, mask,
-					   sdhci_am654->strb_sel <<
-					   STRBSEL_SHIFT);
+			val |= sdhci_am654->strb_sel << STRBSEL_SHIFT;
 		}
 
-		if (sdhci_am654->flags & FREQSEL_2_BIT) {
-			switch (clock) {
-			case 200000000:
-				sel50 = 0;
-				sel100 = 0;
-				break;
-			case 100000000:
-				sel50 = 0;
-				sel100 = 1;
-				break;
-			default:
-				sel50 = 1;
-				sel100 = 0;
-			}
+		regmap_update_bits(sdhci_am654->base, PHY_CTRL4, mask, val);
 
-			/* Configure PHY DLL frequency */
-			mask = SEL50_MASK | SEL100_MASK;
-			val = (sel50 << SEL50_SHIFT) | (sel100 << SEL100_SHIFT);
-			regmap_update_bits(sdhci_am654->base, PHY_CTRL5, mask,
-					   val);
-		} else {
-			switch (clock) {
-			case 200000000:
-				freqsel = 0x0;
-				break;
-			default:
-				freqsel = 0x4;
-			}
-
-			regmap_update_bits(sdhci_am654->base, PHY_CTRL5,
-					   FREQSEL_MASK,
-					   freqsel << FREQSEL_SHIFT);
-		}
-
-		/* Configure DLL TRIM */
-		mask = DLL_TRIM_ICP_MASK;
-		val = sdhci_am654->trm_icp << DLL_TRIM_ICP_SHIFT;
-
-		/* Configure DLL driver strength */
-		mask |= DR_TY_MASK;
-		val |= sdhci_am654->drv_strength << DR_TY_SHIFT;
-		regmap_update_bits(sdhci_am654->base, PHY_CTRL1, mask, val);
-		/* Enable DLL */
-		regmap_update_bits(sdhci_am654->base, PHY_CTRL1, ENDLL_MASK,
-				   0x1 << ENDLL_SHIFT);
-		/*
-		 * Poll for DLL ready. Use a one second timeout.
-		 * Works in all experiments done so far
-		 */
-		ret = regmap_read_poll_timeout(sdhci_am654->base, PHY_STAT1,
-					       val, val & DLLRDY_MASK, 1000,
-					       1000000);
-		if (ret) {
-			dev_err(mmc_dev(host->mmc), "DLL failed to relock\n");
-			return;
-		}
-
-		sdhci_am654->dll_on = true;
+		if (timing > MMC_TIMING_UHS_SDR25)
+			sdhci_am654_setup_dll(host, clock);
 	}
 }
 
···
 {
 	struct sdhci_pltfm_host *pltfm_host = sdhci_priv(host);
 	struct sdhci_am654_data *sdhci_am654 = sdhci_pltfm_priv(pltfm_host);
-	int val, mask;
+	unsigned char timing = host->mmc->ios.timing;
+	u32 otap_del_sel;
+	u32 mask, val;
+
+	/* Setup DLL Output TAP delay */
+	if (sdhci_am654->legacy_otapdly)
+		otap_del_sel = sdhci_am654->otap_del_sel[0];
+	else
+		otap_del_sel = sdhci_am654->otap_del_sel[timing];
 
 	mask = OTAPDLYENA_MASK | OTAPDLYSEL_MASK;
-	val = (1 << OTAPDLYENA_SHIFT) |
-	      (sdhci_am654->otap_del_sel << OTAPDLYSEL_SHIFT);
+	val = (0x1 << OTAPDLYENA_SHIFT) |
+	      (otap_del_sel << OTAPDLYSEL_SHIFT);
 	regmap_update_bits(sdhci_am654->base, PHY_CTRL4, mask, val);
 
 	sdhci_set_clock(host, clock);
-}
-
-static void sdhci_am654_set_power(struct sdhci_host *host, unsigned char mode,
-				  unsigned short vdd)
-{
-	if (!IS_ERR(host->mmc->supply.vmmc)) {
-		struct mmc_host *mmc = host->mmc;
-
-		mmc_regulator_set_ocr(mmc, mmc->supply.vmmc, vdd);
-	}
-	sdhci_set_power_noreg(host, mode, vdd);
 }
 
 static void sdhci_am654_write_b(struct sdhci_host *host, u8 val, int reg)
···
 	.get_timeout_clock = sdhci_pltfm_clk_get_max_clock,
 	.set_uhs_signaling = sdhci_set_uhs_signaling,
 	.set_bus_width = sdhci_set_bus_width,
-	.set_power = sdhci_am654_set_power,
+	.set_power = sdhci_set_power_and_bus_voltage,
 	.set_clock = sdhci_am654_set_clock,
 	.write_b = sdhci_am654_write_b,
 	.irq = sdhci_am654_cqhci_irq,
···
 	.get_timeout_clock = sdhci_pltfm_clk_get_max_clock,
 	.set_uhs_signaling = sdhci_set_uhs_signaling,
 	.set_bus_width = sdhci_set_bus_width,
-	.set_power = sdhci_am654_set_power,
+	.set_power = sdhci_set_power_and_bus_voltage,
 	.set_clock = sdhci_am654_set_clock,
 	.write_b = sdhci_am654_write_b,
 	.irq = sdhci_am654_cqhci_irq,
···
 	.get_timeout_clock = sdhci_pltfm_clk_get_max_clock,
 	.set_uhs_signaling = sdhci_set_uhs_signaling,
 	.set_bus_width = sdhci_set_bus_width,
-	.set_power = sdhci_am654_set_power,
+	.set_power = sdhci_set_power_and_bus_voltage,
 	.set_clock = sdhci_j721e_4bit_set_clock,
 	.write_b = sdhci_am654_write_b,
 	.irq = sdhci_am654_cqhci_irq,
···
 	return ret;
 }
 
+static int sdhci_am654_get_otap_delay(struct sdhci_host *host,
+				      struct sdhci_am654_data *sdhci_am654)
+{
+	struct device *dev = mmc_dev(host->mmc);
+	int i;
+	int ret;
+
+	ret = device_property_read_u32(dev, td[MMC_TIMING_LEGACY].binding,
+				 &sdhci_am654->otap_del_sel[MMC_TIMING_LEGACY]);
+	if (ret) {
+		/*
+		 * ti,otap-del-sel-legacy is mandatory, look for old binding
+		 * if not found.
+		 */
+		ret = device_property_read_u32(dev, "ti,otap-del-sel",
+					       &sdhci_am654->otap_del_sel[0]);
+		if (ret) {
+			dev_err(dev, "Couldn't find otap-del-sel\n");
+
+			return ret;
+		}
+
+		dev_info(dev, "Using legacy binding ti,otap-del-sel\n");
+		sdhci_am654->legacy_otapdly = true;
+
+		return 0;
+	}
+
+	for (i = MMC_TIMING_MMC_HS; i <= MMC_TIMING_MMC_HS400; i++) {
+
+		ret = device_property_read_u32(dev, td[i].binding,
+					       &sdhci_am654->otap_del_sel[i]);
+		if (ret) {
+			dev_dbg(dev, "Couldn't find %s\n",
+				td[i].binding);
+			/*
+			 * Remove the corresponding capability
+			 * if an otap-del-sel value is not found
+			 */
+			if (i <= MMC_TIMING_MMC_DDR52)
+				host->mmc->caps &= ~td[i].capability;
+			else
+				host->mmc->caps2 &= ~td[i].capability;
+		}
+	}
+
+	return 0;
+}
+
 static int sdhci_am654_init(struct sdhci_host *host)
 {
 	struct sdhci_pltfm_host *pltfm_host = sdhci_priv(host);
···
 	if (ret)
 		goto err_cleanup_host;
 
+	ret = sdhci_am654_get_otap_delay(host, sdhci_am654);
+	if (ret)
+		goto err_cleanup_host;
+
 	ret = __sdhci_add_host(host);
 	if (ret)
 		goto err_cleanup_host;
···
 	struct device *dev = &pdev->dev;
 	int drv_strength;
 	int ret;
-
-	ret = device_property_read_u32(dev, "ti,otap-del-sel",
-				       &sdhci_am654->otap_del_sel);
-	if (ret)
-		return ret;
 
 	if (sdhci_am654->flags & DLL_PRESENT) {
 		ret = device_property_read_u32(dev, "ti,trm-icp",

drivers/mmc/host/tmio_mmc.h

···
 	int (*write16_hook)(struct tmio_mmc_host *host, int addr);
 	void (*reset)(struct tmio_mmc_host *host);
 	void (*hw_reset)(struct tmio_mmc_host *host);
-	void (*prepare_tuning)(struct tmio_mmc_host *host, unsigned long tap);
-	bool (*check_scc_error)(struct tmio_mmc_host *host);
+	bool (*check_retune)(struct tmio_mmc_host *host);
 
 	/*
 	 * Mandatory callback for tuning to occur which is optional for SDR50
 	 * and mandatory for SDR104.
 	 */
-	unsigned int (*init_tuning)(struct tmio_mmc_host *host);
-	int (*select_tuning)(struct tmio_mmc_host *host);
-
-	/* Tuning values: 1 for success, 0 for failure */
-	DECLARE_BITMAP(taps, BITS_PER_BYTE * sizeof(long));
-	unsigned int tap_num;
-	unsigned long tap_set;
+	int (*execute_tuning)(struct tmio_mmc_host *host, u32 opcode);
 
 	void (*prepare_hs400_tuning)(struct tmio_mmc_host *host);
 	void (*hs400_downgrade)(struct tmio_mmc_host *host);

drivers/mmc/host/tmio_mmc_core.c

···
 static int tmio_mmc_execute_tuning(struct mmc_host *mmc, u32 opcode)
 {
 	struct tmio_mmc_host *host = mmc_priv(mmc);
-	int i, ret = 0;
+	int ret;
 
-	if (!host->init_tuning || !host->select_tuning)
-		/* Tuning is not supported */
-		goto out;
+	if (!host->execute_tuning)
+		return 0;
 
-	host->tap_num = host->init_tuning(host);
-	if (!host->tap_num)
-		/* Tuning is not supported */
-		goto out;
+	ret = host->execute_tuning(host, opcode);
 
-	if (host->tap_num * 2 >= sizeof(host->taps) * BITS_PER_BYTE) {
-		dev_warn_once(&host->pdev->dev,
-			"Too many taps, skipping tuning. Please consider updating size of taps field of tmio_mmc_host\n");
-		goto out;
-	}
-
-	bitmap_zero(host->taps, host->tap_num * 2);
-
-	/* Issue CMD19 twice for each tap */
-	for (i = 0; i < 2 * host->tap_num; i++) {
-		if (host->prepare_tuning)
-			host->prepare_tuning(host, i % host->tap_num);
-
-		ret = mmc_send_tuning(mmc, opcode, NULL);
-		if (ret == 0)
-			set_bit(i, host->taps);
-	}
-
-	ret = host->select_tuning(host);
-
-out:
 	if (ret < 0) {
 		dev_warn(&host->pdev->dev, "Tuning procedure failed\n");
 		tmio_mmc_hw_reset(mmc);
···
 	if (mrq->cmd->error || (mrq->data && mrq->data->error))
 		tmio_mmc_abort_dma(host);
 
-	/* SCC error means retune, but executed command was still successful */
-	if (host->check_scc_error && host->check_scc_error(host))
+	/* Error means retune, but executed command was still successful */
+	if (host->check_retune && host->check_retune(host))
 		mmc_retune_needed(host->mmc);
 
 	/* If SET_BLOCK_COUNT, continue with main command */
···
 	return blk_size;
 }
 
-static int tmio_mmc_prepare_hs400_tuning(struct mmc_host *mmc,
-					 struct mmc_ios *ios)
-{
-	struct tmio_mmc_host *host = mmc_priv(mmc);
-
-	if (host->prepare_hs400_tuning)
-		host->prepare_hs400_tuning(host);
-
-	return 0;
-}
-
-static void tmio_mmc_hs400_downgrade(struct mmc_host *mmc)
-{
-	struct tmio_mmc_host *host = mmc_priv(mmc);
-
-	if (host->hs400_downgrade)
-		host->hs400_downgrade(host);
-}
-
-static void tmio_mmc_hs400_complete(struct mmc_host *mmc)
-{
-	struct tmio_mmc_host *host = mmc_priv(mmc);
-
-	if (host->hs400_complete)
-		host->hs400_complete(host);
-}
-
-static const struct mmc_host_ops tmio_mmc_ops = {
+static struct mmc_host_ops tmio_mmc_ops = {
 	.request	= tmio_mmc_request,
 	.set_ios	= tmio_mmc_set_ios,
 	.get_ro         = tmio_mmc_get_ro,
···
 	.multi_io_quirk	= tmio_multi_io_quirk,
 	.hw_reset	= tmio_mmc_hw_reset,
 	.execute_tuning = tmio_mmc_execute_tuning,
-	.prepare_hs400_tuning = tmio_mmc_prepare_hs400_tuning,
-	.hs400_downgrade = tmio_mmc_hs400_downgrade,
-	.hs400_complete	= tmio_mmc_hs400_complete,
 };
 
 static int tmio_mmc_init_ocr(struct tmio_mmc_host *host)
···
 }
 EXPORT_SYMBOL_GPL(tmio_mmc_host_runtime_suspend);
 
-static bool tmio_mmc_can_retune(struct tmio_mmc_host *host)
-{
-	return host->tap_num && mmc_can_retune(host->mmc);
-}
-
 int tmio_mmc_host_runtime_resume(struct device *dev)
 {
 	struct tmio_mmc_host *host = dev_get_drvdata(dev);
···
 
 	tmio_mmc_enable_dma(host, true);
 
-	if (tmio_mmc_can_retune(host) && host->select_tuning(host))
-		dev_warn(&host->pdev->dev, "Tuning selection failed\n");
+	mmc_retune_needed(host->mmc);
 
 	return 0;
 }

drivers/mmc/host/vub300.c

···
 	u8 port_number;
 	u8 command_type;
 	u8 command_index;
-	u8 command_response[0];
+	u8 command_response[];
 } __packed;
 
 struct sd_status_header {
···
 	int retval;
 	for (n = 0; n < sdio_funcs; n++) {
 		struct sdio_func *sf = card->sdio_func[n];
-		l += snprintf(vub300->vub_name + l,
+		l += scnprintf(vub300->vub_name + l,
 			      sizeof(vub300->vub_name) - l, "_%04X%04X",
 			      sf->vendor, sf->device);
 	}

include/linux/firmware/xlnx-zynqmp.h

···
 };
 
 enum pm_ioctl_id {
-	IOCTL_SET_SD_TAPDELAY = 7,
+	IOCTL_SD_DLL_RESET = 6,
+	IOCTL_SET_SD_TAPDELAY,
 	IOCTL_SET_PLL_FRAC_MODE,
 	IOCTL_GET_PLL_FRAC_MODE,
 	IOCTL_SET_PLL_FRAC_DATA,
···
 enum tap_delay_type {
 	PM_TAPDELAY_INPUT = 0,
 	PM_TAPDELAY_OUTPUT,
+};
+
+enum dll_reset_type {
+	PM_DLL_RESET_ASSERT,
+	PM_DLL_RESET_RELEASE,
+	PM_DLL_RESET_PULSE,
 };
 
 /**

include/linux/mmc/core.h

···
  */
 
 	unsigned int		busy_timeout;	/* busy detect timeout in ms */
-	/* Set this flag only for blocking sanitize request */
-	bool			sanitize_busy;
-
 	struct mmc_data		*data;		/* data segment associated with cmd */
 	struct mmc_request	*mrq;		/* associated request */
 };

include/linux/mmc/host.h

···
 #define MMC_CAP_3_3V_DDR	(1 << 11)	/* Host supports eMMC DDR 3.3V */
 #define MMC_CAP_1_8V_DDR	(1 << 12)	/* Host supports eMMC DDR 1.8V */
 #define MMC_CAP_1_2V_DDR	(1 << 13)	/* Host supports eMMC DDR 1.2V */
+#define MMC_CAP_DDR		(MMC_CAP_3_3V_DDR | MMC_CAP_1_8V_DDR | \
+				 MMC_CAP_1_2V_DDR)
 #define MMC_CAP_POWER_OFF_CARD	(1 << 14)	/* Can power off after boot */
 #define MMC_CAP_BUS_WIDTH_TEST	(1 << 15)	/* CMD14/CMD19 bus width ok */
 #define MMC_CAP_UHS_SDR12	(1 << 16)	/* Host supports UHS SDR12 mode */
···
 	bool			cqe_enabled;
 	bool			cqe_on;
 
-	unsigned long		private[0] ____cacheline_aligned;
+	/* Host Software Queue support */
+	bool			hsq_enabled;
+
+	unsigned long		private[] ____cacheline_aligned;
 };
 
 struct device_node;

include/linux/mmc/mmc.h

···
 #define R1_STATE_PRG	7
 #define R1_STATE_DIS	8
 
+static inline bool mmc_ready_for_data(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;
+}
+
 /*
  * MMC/SD in SPI mode reports R1 status always, and R2 for SEND_STATUS
  * R1 is the low order byte; R2 is the next highest byte, when present.

include/linux/mmc/sdio_func.h

···
 	struct sdio_func_tuple *next;
 	unsigned char code;
 	unsigned char size;
-	unsigned char data[0];
+	unsigned char data[];
 };
 
 /*

include/linux/platform_data/mmc-esdhc-imx.h

···
 	unsigned int delay_line;
 	unsigned int tuning_step;       /* The delay cell steps in tuning procedure */
 	unsigned int tuning_start_tap;	/* The start delay cell point in tuning procedure */
+	unsigned int strobe_dll_delay_target;	/* The delay cell for strobe pad (read clock) */
 };
 #endif /* __ASM_ARCH_IMX_ESDHC_H */

include/uapi/linux/mmc/ioctl.h

···
  */
 struct mmc_ioc_multi_cmd {
 	__u64 num_of_cmds;
-	struct mmc_ioc_cmd cmds[0];
+	struct mmc_ioc_cmd cmds[];
 };
 
 #define MMC_IOC_CMD _IOWR(MMC_BLOCK_MAJOR, 0, struct mmc_ioc_cmd)