[PATCH v2 0/3] (Qualcomm) UFS device reset support

[PATCH v2 0/3] (Qualcomm) UFS device reset support

[Bjorn Andersson]

This series exposes the ufs_reset line as a gpio, adds support for ufshcd to
acquire and toggle this and then adds this to SDM845 MTP.

Bjorn Andersson (3):
  pinctrl: qcom: sdm845: Expose ufs_reset as gpio
  scsi: ufs: Allow resetting the UFS device
  arm64: dts: qcom: sdm845-mtp: Specify UFS device-reset GPIO

 .../bindings/pinctrl/qcom,sdm845-pinctrl.txt  |  2 +-
 .../devicetree/bindings/ufs/ufshcd-pltfrm.txt |  2 +
 arch/arm64/boot/dts/qcom/sdm845-mtp.dts       |  2 +
 drivers/pinctrl/qcom/pinctrl-sdm845.c         | 12 ++---
 drivers/scsi/ufs/ufshcd.c                     | 44 +++++++++++++++++++
 drivers/scsi/ufs/ufshcd.h                     |  4 ++
 6 files changed, 59 insertions(+), 7 deletions(-)

-- 
2.18.0

[PATCH v2 1/3] pinctrl: qcom: sdm845: Expose ufs_reset as gpio

[Bjorn Andersson]

The ufs_reset pin is expected to be wired to the reset pin of the
primary UFS memory but is pretty much just a general purpose output pinr

Reorder the pins and expose it as gpio 150, so that the UFS driver can
toggle it.

Tested-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
---

Changes since v1:
- None

 .../bindings/pinctrl/qcom,sdm845-pinctrl.txt         |  2 +-
 drivers/pinctrl/qcom/pinctrl-sdm845.c                | 12 ++++++------
 2 files changed, 7 insertions(+), 7 deletions(-)

diff --git a/Documentation/devicetree/bindings/pinctrl/qcom,sdm845-pinctrl.txt b/Documentation/devicetree/bindings/pinctrl/qcom,sdm845-pinctrl.txt
index 321bdb9be0d2..7462e3743c68 100644
--- a/Documentation/devicetree/bindings/pinctrl/qcom,sdm845-pinctrl.txt
+++ b/Documentation/devicetree/bindings/pinctrl/qcom,sdm845-pinctrl.txt
@@ -79,7 +79,7 @@ to specify in a pin configuration subnode:
 		      gpio0-gpio149
 		        Supports mux, bias and drive-strength
 
-		      sdc2_clk, sdc2_cmd, sdc2_data
+		      sdc2_clk, sdc2_cmd, sdc2_data, ufs_reset
 		        Supports bias and drive-strength
 
 - function:
diff --git a/drivers/pinctrl/qcom/pinctrl-sdm845.c b/drivers/pinctrl/qcom/pinctrl-sdm845.c
index c97f20fca5fd..e4e5acade086 100644
--- a/drivers/pinctrl/qcom/pinctrl-sdm845.c
+++ b/drivers/pinctrl/qcom/pinctrl-sdm845.c
@@ -420,10 +420,10 @@ DECLARE_MSM_GPIO_PINS(147);
 DECLARE_MSM_GPIO_PINS(148);
 DECLARE_MSM_GPIO_PINS(149);
 
-static const unsigned int sdc2_clk_pins[] = { 150 };
-static const unsigned int sdc2_cmd_pins[] = { 151 };
-static const unsigned int sdc2_data_pins[] = { 152 };
-static const unsigned int ufs_reset_pins[] = { 153 };
+static const unsigned int ufs_reset_pins[] = { 150 };
+static const unsigned int sdc2_clk_pins[] = { 151 };
+static const unsigned int sdc2_cmd_pins[] = { 152 };
+static const unsigned int sdc2_data_pins[] = { 153 };
 
 enum sdm845_functions {
 	msm_mux_gpio,
@@ -1271,10 +1271,10 @@ static const struct msm_pingroup sdm845_groups[] = {
 	PINGROUP(147, NORTH, _, _, _, _, _, _, _, _, _, _),
 	PINGROUP(148, NORTH, _, _, _, _, _, _, _, _, _, _),
 	PINGROUP(149, NORTH, _, _, _, _, _, _, _, _, _, _),
+	UFS_RESET(ufs_reset, 0x99f000),
 	SDC_QDSD_PINGROUP(sdc2_clk, 0x99a000, 14, 6),
 	SDC_QDSD_PINGROUP(sdc2_cmd, 0x99a000, 11, 3),
 	SDC_QDSD_PINGROUP(sdc2_data, 0x99a000, 9, 0),
-	UFS_RESET(ufs_reset, 0x99f000),
 };
 
 static const struct msm_pinctrl_soc_data sdm845_pinctrl = {
@@ -1284,7 +1284,7 @@ static const struct msm_pinctrl_soc_data sdm845_pinctrl = {
 	.nfunctions = ARRAY_SIZE(sdm845_functions),
 	.groups = sdm845_groups,
 	.ngroups = ARRAY_SIZE(sdm845_groups),
-	.ngpios = 150,
+	.ngpios = 151,
 };
 
 static int sdm845_pinctrl_probe(struct platform_device *pdev)
-- 
2.18.0

[PATCH v2 2/3] scsi: ufs: Allow resetting the UFS device

[Bjorn Andersson]

Acquire the device-reset GPIO and toggle this to reset the UFS device
during initialization and host reset.

Based on downstream support implemented by Subhash Jadavani
<subhashj@codeaurora.org>.

Tested-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
---

Changes since v1:
- Added gpio to DT binding document
- Fixed spelling of UFS

 .../devicetree/bindings/ufs/ufshcd-pltfrm.txt |  2 +
 drivers/scsi/ufs/ufshcd.c                     | 44 +++++++++++++++++++
 drivers/scsi/ufs/ufshcd.h                     |  4 ++
 3 files changed, 50 insertions(+)

diff --git a/Documentation/devicetree/bindings/ufs/ufshcd-pltfrm.txt b/Documentation/devicetree/bindings/ufs/ufshcd-pltfrm.txt
index a74720486ee2..d562d8b4919c 100644
--- a/Documentation/devicetree/bindings/ufs/ufshcd-pltfrm.txt
+++ b/Documentation/devicetree/bindings/ufs/ufshcd-pltfrm.txt
@@ -54,6 +54,8 @@ Optional properties:
 			  PHY reset from the UFS controller.
 - resets            : reset node register
 - reset-names       : describe reset node register, the "rst" corresponds to reset the whole UFS IP.
+- device-reset-gpios	: A phandle and gpio specifier denoting the GPIO connected
+			  to the RESET pin of the UFS memory device.
 
 Note: If above properties are not defined it can be assumed that the supply
 regulators or clocks are always on.
diff --git a/drivers/scsi/ufs/ufshcd.c b/drivers/scsi/ufs/ufshcd.c
index a208589426b1..eaa384d67684 100644
--- a/drivers/scsi/ufs/ufshcd.c
+++ b/drivers/scsi/ufs/ufshcd.c
@@ -42,6 +42,7 @@
 #include <linux/nls.h>
 #include <linux/of.h>
 #include <linux/bitfield.h>
+#include <linux/gpio/consumer.h>
 #include "ufshcd.h"
 #include "ufs_quirks.h"
 #include "unipro.h"
@@ -6135,6 +6136,25 @@ static int ufshcd_abort(struct scsi_cmnd *cmd)
 	return err;
 }
 
+/**
+ * ufshcd_device_reset() - toggle the (optional) device reset line
+ * @hba: per-adapter instance
+ *
+ * Toggles the (optional) reset line to reset the attached device.
+ */
+static void ufshcd_device_reset(struct ufs_hba *hba)
+{
+	/*
+	 * The UFS device shall detect reset pulses of 1us, sleep for 10us to
+	 * be on the safe side.
+	 */
+	gpiod_set_value_cansleep(hba->device_reset, 1);
+	usleep_range(10, 15);
+
+	gpiod_set_value_cansleep(hba->device_reset, 0);
+	usleep_range(10, 15);
+}
+
 /**
  * ufshcd_host_reset_and_restore - reset and restore host controller
  * @hba: per-adapter instance
@@ -6190,6 +6210,9 @@ static int ufshcd_reset_and_restore(struct ufs_hba *hba)
 	int retries = MAX_HOST_RESET_RETRIES;
 
 	do {
+		/* Reset the attached device */
+		ufshcd_device_reset(hba);
+
 		err = ufshcd_host_reset_and_restore(hba);
 	} while (err && --retries);
 
@@ -7386,6 +7409,18 @@ static void ufshcd_variant_hba_exit(struct ufs_hba *hba)
 	ufshcd_vops_exit(hba);
 }
 
+static int ufshcd_init_device_reset(struct ufs_hba *hba)
+{
+	hba->device_reset = devm_gpiod_get_optional(hba->dev, "device-reset",
+						    GPIOD_OUT_HIGH);
+	if (IS_ERR(hba->device_reset)) {
+		dev_err(hba->dev, "failed to acquire reset gpio: %ld\n",
+			PTR_ERR(hba->device_reset));
+	}
+
+	return PTR_ERR_OR_ZERO(hba->device_reset);
+}
+
 static int ufshcd_hba_init(struct ufs_hba *hba)
 {
 	int err;
@@ -7425,9 +7460,15 @@ static int ufshcd_hba_init(struct ufs_hba *hba)
 	if (err)
 		goto out_disable_vreg;
 
+	err = ufshcd_init_device_reset(hba);
+	if (err)
+		goto out_disable_variant;
+
 	hba->is_powered = true;
 	goto out;
 
+out_disable_variant:
+	ufshcd_vops_setup_regulators(hba, false);
 out_disable_vreg:
 	ufshcd_setup_vreg(hba, false);
 out_disable_clks:
@@ -8321,6 +8362,9 @@ int ufshcd_init(struct ufs_hba *hba, void __iomem *mmio_base, unsigned int irq)
 		goto exit_gating;
 	}
 
+	/* Reset the attached device */
+	ufshcd_device_reset(hba);
+
 	/* Host controller enable */
 	err = ufshcd_hba_enable(hba);
 	if (err) {
diff --git a/drivers/scsi/ufs/ufshcd.h b/drivers/scsi/ufs/ufshcd.h
index 994d73d03207..4443928ba7d2 100644
--- a/drivers/scsi/ufs/ufshcd.h
+++ b/drivers/scsi/ufs/ufshcd.h
@@ -72,6 +72,8 @@
 #define UFSHCD "ufshcd"
 #define UFSHCD_DRIVER_VERSION "0.2"
 
+struct gpio_desc;
+
 struct ufs_hba;
 
 enum dev_cmd_type {
@@ -706,6 +708,8 @@ struct ufs_hba {
 
 	struct device		bsg_dev;
 	struct request_queue	*bsg_queue;
+
+	struct gpio_desc *device_reset;
 };
 
 /* Returns true if clocks can be gated. Otherwise false */
-- 
2.18.0

RE: [PATCH v2 2/3] scsi: ufs: Allow resetting the UFS device

[Avri Altman]

>  static int ufshcd_hba_init(struct ufs_hba *hba)
>  {
>  	int err;
> @@ -7425,9 +7460,15 @@ static int ufshcd_hba_init(struct ufs_hba *hba)
>  	if (err)
>  		goto out_disable_vreg;
> 
> +	err = ufshcd_init_device_reset(hba);
> +	if (err)
> +		goto out_disable_variant;
> +
>  	hba->is_powered = true;
>  	goto out;
> 
> +out_disable_variant:
> +	ufshcd_vops_setup_regulators(hba, false);
Is this necessary?
ufshcd_vops_setup_regulators() was just called as part of ufshcd_variant_hba_init

Thanks,
Avri

Re: [PATCH v2 2/3] scsi: ufs: Allow resetting the UFS device

[Bjorn Andersson]

On Wed 05 Jun 23:36 PDT 2019, Avri Altman wrote:

> 
> >  static int ufshcd_hba_init(struct ufs_hba *hba)
> >  {
> >  	int err;
> > @@ -7425,9 +7460,15 @@ static int ufshcd_hba_init(struct ufs_hba *hba)
> >  	if (err)
> >  		goto out_disable_vreg;
> > 
> > +	err = ufshcd_init_device_reset(hba);
> > +	if (err)
> > +		goto out_disable_variant;
> > +
> >  	hba->is_powered = true;
> >  	goto out;
> > 
> > +out_disable_variant:
> > +	ufshcd_vops_setup_regulators(hba, false);
> Is this necessary?
> ufshcd_vops_setup_regulators() was just called as part of ufshcd_variant_hba_init
> 

Yes, so my attempt here is to reverse the enablement of the vops
regulators (hence passing false). But looking at it again I see that we
should also do ufshcd_vops_exit(), so the right thing to call here is
ufshcd_variant_hba_exit().

PS. This initialization sequence should really be rewritten to first
acquire all resources and then turn them on. This mixes init/setup
sequence is really hard to reason about.

Regards,
Bjorn

Re: [PATCH v2 2/3] scsi: ufs: Allow resetting the UFS device

[Rob Herring]

On Wed, Jun 05, 2019 at 06:02:48PM -0700, Bjorn Andersson wrote:
> Acquire the device-reset GPIO and toggle this to reset the UFS device
> during initialization and host reset.
> 
> Based on downstream support implemented by Subhash Jadavani
> <subhashj@codeaurora.org>.
> 
> Tested-by: John Stultz <john.stultz@linaro.org>
> Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
> ---
> 
> Changes since v1:
> - Added gpio to DT binding document
> - Fixed spelling of UFS
> 
>  .../devicetree/bindings/ufs/ufshcd-pltfrm.txt |  2 +
>  drivers/scsi/ufs/ufshcd.c                     | 44 +++++++++++++++++++
>  drivers/scsi/ufs/ufshcd.h                     |  4 ++
>  3 files changed, 50 insertions(+)
> 
> diff --git a/Documentation/devicetree/bindings/ufs/ufshcd-pltfrm.txt b/Documentation/devicetree/bindings/ufs/ufshcd-pltfrm.txt
> index a74720486ee2..d562d8b4919c 100644
> --- a/Documentation/devicetree/bindings/ufs/ufshcd-pltfrm.txt
> +++ b/Documentation/devicetree/bindings/ufs/ufshcd-pltfrm.txt
> @@ -54,6 +54,8 @@ Optional properties:
>  			  PHY reset from the UFS controller.
>  - resets            : reset node register
>  - reset-names       : describe reset node register, the "rst" corresponds to reset the whole UFS IP.
> +- device-reset-gpios	: A phandle and gpio specifier denoting the GPIO connected
> +			  to the RESET pin of the UFS memory device.

A sign we should have a child node for the device...

Doesn't using 'reset-gpios' work as I doubt one would have a GPIO reset 
for the host controller.

Rob

Re: [PATCH v2 2/3] scsi: ufs: Allow resetting the UFS device

[Linus Walleij]

On Thu, Jun 6, 2019 at 3:02 AM Bjorn Andersson
<bjorn.andersson@linaro.org> wrote:

> Acquire the device-reset GPIO and toggle this to reset the UFS device
> during initialization and host reset.
>
> Based on downstream support implemented by Subhash Jadavani
> <subhashj@codeaurora.org>.
>
> Tested-by: John Stultz <john.stultz@linaro.org>
> Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>

I agree with Rob that just "reset-gpios" looks better.
Otherwise:
Reviewed-by: Linus Walleij <linus.walleij@linaro.org>

Yours,
Linus Walleij

[PATCH v2 3/3] arm64: dts: qcom: sdm845-mtp: Specify UFS device-reset GPIO

[Bjorn Andersson]

Specify the UFS device-reset gpio, so that the controller will issue a
reset of the UFS device.

Tested-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
---

Changes since v1:
- None

 arch/arm64/boot/dts/qcom/sdm845-mtp.dts | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/arch/arm64/boot/dts/qcom/sdm845-mtp.dts b/arch/arm64/boot/dts/qcom/sdm845-mtp.dts
index 2e78638eb73b..d116a0956a9c 100644
--- a/arch/arm64/boot/dts/qcom/sdm845-mtp.dts
+++ b/arch/arm64/boot/dts/qcom/sdm845-mtp.dts
@@ -388,6 +388,8 @@
 &ufs_mem_hc {
 	status = "okay";
 
+	device-reset-gpios = <&tlmm 150 GPIO_ACTIVE_LOW>;
+
 	vcc-supply = <&vreg_l20a_2p95>;
 	vcc-max-microamp = <600000>;
 };
-- 
2.18.0