mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Larisa Ileana Grigore <larisa.grigore@oss.nxp.com>
To: Bough Chen <haibo.chen@oss.nxp.com>
Cc: NXP S32 Linux Team <s32@nxp.com>,
	Alim Akhtar <alim.akhtar@samsung.com>,
	Avri Altman <avri.altman@sandisk.com>,
	Bart Van Assche <bvanassche@acm.org>,
	Rob Herring <robh@kernel.org>,
	Krzysztof Kozlowski <krzk+dt@kernel.org>,
	Conor Dooley <conor+dt@kernel.org>, Frank Li <Frank.Li@nxp.com>,
	Sascha Hauer <s.hauer@pengutronix.de>,
	Pengutronix Kernel Team <kernel@pengutronix.de>,
	Fabio Estevam <festevam@gmail.com>,
	"James E . J . Bottomley" <James.Bottomley@HansenPartnership.com>,
	"Martin K . Petersen" <martin.petersen@oracle.com>,
	Sai Krishna Potthuri <sai.krishna.potthuri@amd.com>,
	Ajay Neeli <ajay.neeli@amd.com>,
	Manivannan Sadhasivam <mani@kernel.org>,
	Pedro Sousa <pedrom.sousa@synopsys.com>,
	clizzi@redhat.com, aruizrui@redhat.com, eballetb@redhat.com,
	echanude@redhat.com, linux-scsi@vger.kernel.org,
	devicetree@vger.kernel.org, linux-kernel@vger.kernel.org,
	imx@lists.linux.dev, linux-arm-kernel@lists.infradead.org,
	linux-arm-msm@vger.kernel.org
Subject: Re: [PATCH 07/10] scsi: ufs: Add NXP S32N79 UFS host controller driver
Date: Mon, 14 Sep 2026 15:20:17 +0300	[thread overview]
Message-ID: <cb5e8a0e-b371-4a77-b023-689c6818b5ae@oss.nxp.com> (raw)
In-Reply-To: <20260828092724.m46ftm7daavufuxy@shlinux89>

On 8/28/2026 12:27 PM, Bough Chen wrote:
> On Wed, Aug 26, 2026 at 03:40:45PM +0200, Larisa Grigore wrote:
>> Add support for the UFS host controller on the NXP S32N79 SoC, built on
>> the Synopsys DesignWare (DWC) UFS architecture, using the UFSHCD DWC and
>> UFSHCD platform glue.
>>
>> This controller requires S32N79-specific initialization prior to
>> UIC_LINKSTARTUP, including:
>>    - programming SCM.ONE_US_TICK and HC.HCLKDIV based on the core clock,
>>    - applying the vendor-defined M-PHY calibration sequence, and
>>    - performing post-link calibration steps needed for HS operation.
>>
>> The M-PHY boot mode is selected via the 'nxp,mphy-boot-mode' device tree
>> property. If the property is omitted, M-PHY initialization is assumed to
>> be handled by external firmware or by a simulation environment.
>>
>> Signed-off-by: Larisa Grigore <larisa.grigore@oss.nxp.com>
>> ---
>>   drivers/ufs/host/Kconfig         |  12 +
>>   drivers/ufs/host/Makefile        |   1 +
>>   drivers/ufs/host/ufs-nxp-s32n7.c | 464 +++++++++++++++++++++++++++++++
>>   drivers/ufs/host/ufshcd-dwc.h    |   9 +-
>>   4 files changed, 484 insertions(+), 2 deletions(-)
>>   create mode 100644 drivers/ufs/host/ufs-nxp-s32n7.c
>>
>> diff --git a/drivers/ufs/host/Kconfig b/drivers/ufs/host/Kconfig
>> index ff170c0b6da0..e8ee7dd05f1a 100644
>> --- a/drivers/ufs/host/Kconfig
>> +++ b/drivers/ufs/host/Kconfig
>> @@ -168,3 +168,15 @@ config SCSI_UFS_AMD_VERSAL2
>>   
>>   	  Select this if you have UFS controller on AMD Versal Gen 2 SoC.
>>   	  If unsure, say N.
>> +
>> +config SCSI_UFS_S32N7
>> +	tristate "NXP S32N7 platform driver"
>> +	depends on SCSI_UFSHCD_PLATFORM && (ARCH_S32 || COMPILE_TEST)
>> +	help
>> +	  This selects the S32N7 specific additions on top of the UFSHCD DWC
>> +	  and UFSHCD platform driver. UFS host on S32N79 needs some vendor
>> +	  specific configurations like PHY and vendor specific register accesses
>> +	  before accessing the hardware.
>> +
>> +	  Select this if you have UFS controller on an S32N7 based board.
>> +	  If unsure, say N.
>> diff --git a/drivers/ufs/host/Makefile b/drivers/ufs/host/Makefile
>> index 65d8bb23ab7b..0381750a35d3 100644
>> --- a/drivers/ufs/host/Makefile
>> +++ b/drivers/ufs/host/Makefile
>> @@ -2,6 +2,7 @@
>>   
>>   obj-$(CONFIG_SCSI_UFS_DWC_TC_PCI) += tc-dwc-g210-pci.o ufshcd-dwc.o tc-dwc-g210.o
>>   obj-$(CONFIG_SCSI_UFS_DWC_TC_PLATFORM) += tc-dwc-g210-pltfrm.o ufshcd-dwc.o tc-dwc-g210.o
>> +obj-$(CONFIG_SCSI_UFS_S32N7) += ufs-nxp-s32n7.o ufshcd-dwc.o
>>   obj-$(CONFIG_SCSI_UFS_CDNS_PLATFORM) += cdns-pltfrm.o
>>   obj-$(CONFIG_SCSI_UFS_QCOM) += ufs-qcom.o
>>   obj-$(CONFIG_SCSI_UFS_EXYNOS) += ufs-exynos.o
>> diff --git a/drivers/ufs/host/ufs-nxp-s32n7.c b/drivers/ufs/host/ufs-nxp-s32n7.c
>> new file mode 100644
>> index 000000000000..8a125105536b
>> --- /dev/null
>> +++ b/drivers/ufs/host/ufs-nxp-s32n7.c
>> @@ -0,0 +1,464 @@
>> +// SPDX-License-Identifier: (GPL-2.0-or-later OR BSD-3-Clause)
>> +/*
>> + * Copyright 2026 NXP
>> + *
>> + */
>> +
>> +#include <linux/clk.h>
>> +#include <linux/iopoll.h>
>> +#include <linux/kernel.h>
>> +#include <linux/module.h>
>> +#include <linux/of.h>
>> +#include <linux/platform_device.h>
>> +
>> +#include "ufshcd-pltfrm.h"
>> +#include "ufshcd-dwc.h"
>> +#include "ufshci-dwc.h"
>> +
>> +/* SCM Register Offsets. */
>> +#define SCM_ONE_US_TICK			0x2CU
>> +#define SCM_STATUS			0x30
>> +
>> +/* SCM_ONE_US_TICK Register. */
>> +#define SCM_ONE_US_TICK_MASK		GENMASK(8, 0)
>> +
>> +/* SCM_MPHY_RAM_CONFIG_STATUS Register Fields. */
>> +#define SCM_STATUS_PHY_RESET_MASK		BIT(0)
>> +#define SCM_STATUS_SRAM_BYPASS_MASK		BIT(1)
>> +#define SCM_STATUS_SRAM_INIT_DONE_MASK		BIT(16)
>> +
>> +#define SCM_STATUS_PHY_RESET(val)		\
>> +	FIELD_PREP(SCM_STATUS_PHY_RESET_MASK, val)
>> +
>> +#define SCM_STATUS_SRAM_BYPASS(val)		\
>> +	FIELD_PREP(SCM_STATUS_SRAM_BYPASS_MASK, val)
>> +
>> +/* Timeout values. */
>> +#define CFG_MPHY_INIT_TIMEOUT_VALUE_US	3000000
>> +
>> +/* Clock validation limits. */
>> +#define MAX_VALID_ONE_US_TICK		SCM_ONE_US_TICK_MASK
>> +
>> +/* CPort definitions */
>> +#define CPORT_0				0
>> +
>> +/* Hibern8 state poll timeout. */
>> +#define HBRN8_POLL_TOUT_MS      1000
>> +
>> +/**
>> + * enum mphy_boot_mode - MPHY boot mode options
>> + * @MPHY_BOOT_NONE: Skip MPHY initialization
>> + * @MPHY_BOOT_ROM: Running MPHY from internal FW ROM
>> + */
>> +enum mphy_boot_mode {
>> +	MPHY_BOOT_NONE = 0,
>> +	MPHY_BOOT_ROM = 1,
>> +};
>> +
>> +/**
>> + * struct s32n_ufs - S32N7 UFS host controller data
>> + * @hba: UFS host controller instance
>> + * @reg_scm: SCM register base address
>> + * @core_clk: core reference clock used to derive ONE_US_TICK/HCLKDIV
>> + * @mphy_boot_mode: MPHY boot mode configuration
>> + */
>> +struct s32n_ufs {
>> +	struct ufs_hba *hba;
>> +	void __iomem *reg_scm;
>> +	struct clk *core_clk;
>> +	enum mphy_boot_mode mphy_boot_mode;
>> +};
>> +
>> +/**
>> + * struct phy_reg_cfg - PHY register configuration entry
>> + * @reg: Register offset/address
>> + * @val: Value to write
>> + */
>> +struct phy_reg_cfg {
>> +	u32 reg;
>> +	u32 val;
>> +};
>> +
>> +/**
>> + * ufs_s32n_phy_write_sequence - Write a sequence of PHY registers
>> + * @hba: UFS host controller instance
>> + * @cfg: Array of register configurations
>> + * @count: Number of entries in the array
>> + *
>> + * Return: 0 on success, error code on failure
>> + */
>> +static int ufs_s32n_phy_write_sequence(struct ufs_hba *hba,
>> +				       const struct phy_reg_cfg *cfg,
>> +				       size_t count)
>> +{
>> +	int ret;
>> +	size_t i;
>> +
>> +	for (i = 0; i < count; i++) {
>> +		ret = ufshcd_dwc_phy_reg_write(hba, cfg[i].reg, cfg[i].val);
>> +		if (ret) {
>> +			dev_err(hba->dev,
>> +				"Failed to write PHY reg 0x%x = 0x%x (step %zu)\n",
>> +				cfg[i].reg, cfg[i].val, i);
>> +			return ret;
>> +		}
>> +	}
>> +
>> +	return 0;
>> +}
>> +
>> +static unsigned long ufs_s32n_calculate_us_tick(unsigned long clk_rate)
>> +{
>> +	return clk_rate / USEC_PER_SEC;
>> +}
>> +
>> +static int ufs_s32n_configure_clocks(struct ufs_hba *hba)
>> +{
>> +	struct s32n_ufs *ufs = ufshcd_get_variant(hba);
>> +	struct device *dev = hba->dev;
>> +	unsigned long clk_rate = 0;
>> +	unsigned long one_us_tick;
>> +
>> +	if (!ufs->core_clk) {
>> +		dev_err(dev, "Invalid core_clk.\n");
>> +		return -EINVAL;
>> +	}
>> +
>> +	clk_rate = clk_get_rate(ufs->core_clk);
>> +	if (!clk_rate) {
>> +		dev_err(dev, "Failed to get valid clock rate.\n");
>> +		return -EINVAL;
>> +	}
>> +
>> +	one_us_tick = ufs_s32n_calculate_us_tick(clk_rate);
>> +	if (one_us_tick == 0 || one_us_tick > MAX_VALID_ONE_US_TICK) {
>> +		dev_err(dev, "Invalid one_us_tick value: %lu (clk_rate: %lu Hz).\n",
>> +			one_us_tick, clk_rate);
>> +		return -EINVAL;
>> +	}
>> +
>> +	dev_dbg(dev, "Core clock rate: %lu Hz, one_us_tick = %lu.\n",
>> +		 clk_rate, one_us_tick);
>> +
>> +	/*
>> +	 * Configure the micro-second tick rate generator based on core
>> +	 * clock rate.
>> +	 */
>> +	writel(one_us_tick, ufs->reg_scm + SCM_ONE_US_TICK);
>> +	ufshcd_dwc_program_clk_div(hba, one_us_tick);
>> +
>> +	return 0;
>> +}
>> +
>> +static int ufs_s32n_phy_initial_calib(struct ufs_hba *hba)
>> +{
>> +	static const struct phy_reg_cfg initial_calib[] = {
>> +		{ FAST_FLAGS(0), 0x6 },
>> +		{ FAST_FLAGS(1), 0x6 },
>> +		{ RX_DAC_CTRL_OVRD(0), 0x1 },
>> +		{ RX_DAC_CTRL_OVRD(1), 0x1 },
>> +		{ RX_DAC_CTRL(0), 0x8E },
>> +		{ RX_DAC_CTRL(1), 0x91 },
>> +		{ RX_DAC_CTRL_SEL(0), 0x1 },
>> +		{ RX_DAC_CTRL_SEL(1), 0x1 },
>> +		{ RX_DAC_CTRL_EN(0), 0x1 },
>> +		{ RX_DAC_CTRL_EN(1), 0x1 },
>> +		{ RX_DAC_CTRL_OVRD(0), 0x0 },
>> +		{ RX_DAC_CTRL_OVRD(1), 0x0 },
>> +		{ RX_DAC_CTRL_OVRD(0), 0x1 },
>> +		{ RX_DAC_CTRL_OVRD(1), 0x1 },
>> +		{ RX_DAC_CTRL(0), 0x71 },
>> +		{ RX_DAC_CTRL(1), 0x7F },
>> +		{ RX_DAC_CTRL_SEL(0), 0x2 },
>> +		{ RX_DAC_CTRL_SEL(1), 0x2 },
>> +		{ RX_DAC_CTRL_EN(0), 0x1 },
>> +		{ RX_DAC_CTRL_EN(1), 0x1 },
>> +		{ RX_DAC_CTRL_OVRD(0), 0x0 },
>> +		{ RX_DAC_CTRL_OVRD(1), 0x0 },
>> +		{ FW_CALIB_CCFG(0), 0x100 },
>> +		{ FW_CALIB_CCFG(1), 0x100 },
>> +	};
>> +	int ret;
>> +
>> +	/* Clock Control */
>> +	ret = ufshcd_dme_set(hba, UIC_ARG_MIB_SEL(CBREFCLKCTRL2, 0), 0x80);
>> +	if (ret)
>> +		return ret;
>> +
>> +	ret = ufshcd_dme_set(hba, UIC_ARG_MIB_SEL(VS_MPHYCFGUPDT, 0), 0x1);
>> +	if (ret)
>> +		return ret;
>> +
>> +	/* Initial calibration sequence */
>> +	return ufs_s32n_phy_write_sequence(hba, initial_calib,
>> +					   ARRAY_SIZE(initial_calib));
>> +}
>> +
>> +static int ufs_s32n_link_startup_pre_change(struct ufs_hba *hba)
>> +{
>> +	struct s32n_ufs *ufs = ufshcd_get_variant(hba);
>> +	struct device *dev = hba->dev;
>> +	int ret;
>> +	u32 reg;
>> +
>> +	/*
>> +	 * There are two supported MPHY boot options implemented:
>> +	 * - Running MPHY from internal FW ROM:
>> +	 *  MPHY_RAM_CONFIG_STATUS.SRAM_BYPASS = 1
>> +	 *  MPHY_RAM_CONFIG_STATUS.SRAM_EXT_LD_DONE = 0
>> +	 * - Skip M-PHY initialization when it is handled externally.
>> +	 */
>> +	if (ufs->mphy_boot_mode == MPHY_BOOT_NONE) {
>> +		dev_dbg(dev, "Skipping UFS MPHY init.\n");
>> +		return 0;
>> +	}
>> +
>> +	ret = ufs_s32n_configure_clocks(hba);
>> +	if (ret)
>> +		return ret;
>> +
>> +	/* Reset SCM.MPHY_RAM_CONFIG_STATUS to default value; keep MPHY in reset. */
>> +	writel(SCM_STATUS_PHY_RESET(1), ufs->reg_scm + SCM_STATUS);
>> +	if (ufs->mphy_boot_mode == MPHY_BOOT_ROM) {
>> +		/* ROM Mode. */
>> +		writel(SCM_STATUS_PHY_RESET(1) | SCM_STATUS_SRAM_BYPASS(1),
>> +		       ufs->reg_scm + SCM_STATUS);
>> +	}
>> +
>> +	ret = ufshcd_dme_set(hba, UIC_ARG_MIB_SEL(CBCRCTRL, 0), 0x1);
>> +	if (ret)
>> +		return ret;
>> +
>> +	ret = ufshcd_dme_set(hba, UIC_ARG_MIB_SEL(VS_MPHYCFGUPDT, 0), 0x1);
>> +	if (ret)
>> +		return ret;
>> +
>> +	/* Release PHY_RESET. */
>> +	writel(readl(ufs->reg_scm + SCM_STATUS) & ~SCM_STATUS_PHY_RESET_MASK,
>> +	       ufs->reg_scm + SCM_STATUS);
>> +
>> +	/* Wait until SRAM_INIT_DONE = 1. */
>> +	ret = readl_poll_timeout(ufs->reg_scm + SCM_STATUS, reg,
>> +			reg & SCM_STATUS_SRAM_INIT_DONE_MASK,
>> +			1000, CFG_MPHY_INIT_TIMEOUT_VALUE_US);
>> +	if (ret) {
>> +		dev_err(dev, "UFS MPHY init not done!\n");
>> +		return ret;
>> +	}
>> +
>> +	/* Start of initial calibration */
>> +	ret = ufs_s32n_phy_initial_calib(hba);
>> +	if (ret) {
>> +		dev_err(dev, "UFS MPHY initial calibration failed!\n");
>> +		return ret;
>> +	}
>> +
>> +	ret = ufshcd_dme_set(hba, UIC_ARG_MIB_SEL(VS_MPHYDISABLE, 0), 0x0);
>> +	if (ret)
>> +		return ret;
>> +
>> +	ret = ufshcd_dme_set(hba, UIC_ARG_MIB_SEL(VS_MPHYCFGUPDT, 0), 0x1);
>> +	if (ret)
>> +		return ret;
>> +
>> +	/* Set Local DeviceID. */
>> +	ret = ufshcd_dme_set(hba, UIC_ARG_MIB_SEL(N_DEVICEID, 0), 0x0);
>> +	if (ret)
>> +		return ret;
> 
> N_DEVICEID belong to network layer of Unipro, set DEVICEID in pre_change seems
> too early. link startup still not done, so unipro link still not up.
> I suggest to move this DEVICEID to post_change, and just before CPort, so that
> the whole config sequence is: Mphy -> PA -> DL -> N -> T, from down to high.
>
Correct. Thank you for the review!

>> +
>> +	/* End of Gear1 settings */
>> +
>> +	return ufshcd_check_hibern8(hba, hba->lanes_per_direction,
>> +				    HBRN8_POLL_TOUT_MS);
>> +}
>> +
>> +static int ufs_s32n_link_startup_post_change(struct ufs_hba *hba)
>> +{
>> +	static const struct ufshcd_dme_attr_val cport_setup[] = {
>> +		{ UIC_ARG_MIB_SEL(T_CONNECTIONSTATE, CPORT_0),
>> +				CPORT_IDLE, DME_LOCAL },
>> +		{ UIC_ARG_MIB_SEL(T_CPORTFLAGS, CPORT_0),
>> +				CPORT_DEF_FLAGS, DME_LOCAL },
>> +		{ UIC_ARG_MIB_SEL(T_CONNECTIONSTATE, CPORT_0),
>> +				CPORT_CONNECTED, DME_LOCAL },
>> +	};
>> +	static const struct phy_reg_cfg post_calib[] = {
>> +		{ RX_OVRD_IN_1(0), 0xc },
>> +		{ RX_OVRD_IN_1(1), 0xc },
>> +		{ RX_OVRD_IN_1(0), 0x8 },
>> +		{ RX_OVRD_IN_1(1), 0x8 },
>> +		{ RX_OVRD_IN_1(0), 0x0 },
>> +		{ RX_OVRD_IN_1(1), 0x0 },
>> +	};
>> +	struct s32n_ufs *ufs = ufshcd_get_variant(hba);
>> +	unsigned int data = 0;
>> +	int ret;
>> +
>> +	if (ufs->mphy_boot_mode == MPHY_BOOT_NONE)
>> +		return 0;
>> +
>> +	/* Set Connection State to IDLE (it allows CPort Attributes to be set). */
>> +	ret = ufshcd_dwc_dme_set_attrs(hba, cport_setup, ARRAY_SIZE(cport_setup));
>> +	if (ret)
>> +		return ret;
>> +
>> +	ret = ufshcd_dme_get(hba, UIC_ARG_MIB_SEL(T_CONNECTIONSTATE, CPORT_0), &data);
>> +	if (ret)
>> +		return ret;
>> +
>> +	if (data != CPORT_CONNECTED)
>> +		return -EIO;
>> +
>> +	/* Post Link Startup Calibration sequence */
>> +	ret = ufs_s32n_phy_write_sequence(hba, post_calib,
>> +					   ARRAY_SIZE(post_calib));
>> +	if (ret)
>> +		return ret;
>> +
>> +	/*
>> +	 * Performing MPHY configuration for rate change:
>> +	 * CB rate selection: 0 - rate A, 1 - rate B;
>> +	 */
>> +	ret = ufshcd_dme_set(hba, UIC_ARG_MIB_SEL(CBRATESEL, 0), 0x1);
>> +	if (ret)
>> +		return ret;
>> +
>> +	ret = ufshcd_dme_set(hba, UIC_ARG_MIB_SEL(VS_MPHYCFGUPDT, 0), 0x1);
>> +	if (ret)
>> +		return ret;
> 
> Should the rate change operation be moved before post link startup calibration sequence?
> in post link startup calibration sequence, it reset the RX CDR, but thie CDR reset should be
> based on the target rate B, right? If do the RX CDR reset first, it is based on rate A, then
> when swithc to rate B, seems need this RX CDR reset again.
> 
Indeed you are right. I will move this operation accordingly. Thank you!

> Another question is why need to switch to rate B here in link_startup_post_change?
> after link_startup_post_change, common code will call
> ufshcd_dme_change_power_mode {
> ...
> PA_HSSERIES = PA_HS_MODE_B;
> PA_PWRMode = FAST_MODE << 4 | FAST_MODE;
> ...
> }
> 
> In the standard DesignWare (DW) UFS IP design, when the PA_HSSERIES value in the PA layer
> changes, the M-PHY rate selection register (CBRATESEL) is automatically synchronized by
> hardware. No manual software intervention is required.
> 
> Does this rate B change here really necessary? or there are some hardware limitation on
> S32N79?

I also verified whether the explicit CBRATESEL programming can be 
removed. On S32N79, setting PA_HSSERIES to Rate B and completing the 
power-mode change updates PA_HSSERIES and the TX/RX lane rate attributes 
to Rate B, but CBRATESEL remains at Rate A. The link later becomes 
non-functional.

>> +
>> +	/* End of Gear4 calibration */
>> +	return 0;
>> +}
>> +
>> +static int ufs_s32n_link_startup_notify(struct ufs_hba *hba,
>> +			     enum ufs_notify_change_status status)
>> +{
>> +	int err;
>> +
>> +	if (status == PRE_CHANGE) {
>> +		err = ufs_s32n_link_startup_pre_change(hba);
>> +		if (err) {
>> +			dev_err(hba->dev, "MPHY setup failed (%d).\n", err);
>> +			return err;
>> +		}
>> +		return 0;
>> +	}
>> +
>> +	/* POST_CHANGE */
>> +	err = ufshcd_dwc_link_is_up(hba);
>> +	if (err) {
>> +		dev_err(hba->dev, "Link is not up.\n");
>> +		return err;
>> +	}
>> +
>> +	err = ufs_s32n_link_startup_post_change(hba);
>> +	if (err)
>> +		dev_err(hba->dev, "Connection setup failed (%d).\n", err);
>> +
>> +	return err;
>> +}
>> +
>> +static int ufs_s32n_init(struct ufs_hba *hba)
>> +{
>> +	struct device *dev = hba->dev;
>> +	struct platform_device *pdev = to_platform_device(dev);
>> +	struct device_node *np = dev->of_node;
>> +	const char *mphy_boot_mode;
>> +	struct s32n_ufs *ufs;
>> +	int ret;
>> +
>> +	ufs = devm_kzalloc(dev, sizeof(*ufs), GFP_KERNEL);
>> +	if (!ufs)
>> +		return -ENOMEM;
>> +
>> +	ufs->hba = hba;
>> +	ufs->mphy_boot_mode = MPHY_BOOT_NONE;
>> +
>> +	hba->quirks |= UFSHCD_QUIRK_BROKEN_AUTO_HIBERN8;
> 
> why broken auto hibernate here? or meet any issue when enable auto hibernate?
> Better to add a comment here.

At the time these patches were created, Auto-Hibern8 was not functional 
on S32N79. We have since resolved the issue and verified that 
Auto-Hibern8 works correctly. Therefore, this quirk will be removed in 
the next revision of the patch series.

>> +	hba->quirks |= UFSHCD_QUIRK_PERFORM_LINK_STARTUP_ONCE;
>> +	hba->spm_lvl = UFS_PM_LVL_5;
> 
> Any reason why not support LVL_6? Does this chip do not support reset pin?

The controller is completely powered off during system suspend and is 
fully reinitialized during resume. Therefore, we use UFS_PM_LVL_5, which 
is sufficient for this suspend/resume sequence.

> Regards
> Haibo Chen
>> +
>> +	ret = of_property_read_string(np, "nxp,mphy-boot-mode", &mphy_boot_mode);
>> +	if (ret || !mphy_boot_mode) {
>> +		dev_dbg(dev,
>> +			"nxp,mphy-boot-mode property not found. UFS MPHY init will be skipped.\n");
>> +		goto init_out;
>> +	}
>> +
>> +	if (!strcmp(mphy_boot_mode, "rom")) {
>> +		ufs->mphy_boot_mode = MPHY_BOOT_ROM;
>> +	} else {
>> +		return dev_err_probe(dev, -EINVAL,
>> +				     "Unrecognized nxp,mphy-boot-mode property. UFS MPHY init will be skipped.\n");
>> +	}
>> +
>> +	ufs->reg_scm = devm_platform_ioremap_resource_byname(pdev, "scm");
>> +	if (IS_ERR(ufs->reg_scm))
>> +		return dev_err_probe(dev, PTR_ERR(ufs->reg_scm),
>> +				     "ioremap failed for SCM registers.\n");
>> +
>> +	ufs->core_clk = devm_clk_get(dev, "core_clk");
>> +	if (IS_ERR(ufs->core_clk))
>> +		return dev_err_probe(dev, PTR_ERR(ufs->core_clk),
>> +				     "Failed to get core clock.\n");
>> +
>> +init_out:
>> +	ufshcd_set_variant(hba, ufs);
>> +
>> +	return 0;
>> +}
>> +
>> +static const struct ufs_hba_variant_ops ufs_hba_s32n79_vops = {
>> +	.name = "s32n79",
>> +	.init = ufs_s32n_init,
>> +	.link_startup_notify = ufs_s32n_link_startup_notify,
>> +};
>> +
>> +static const struct of_device_id ufs_s32n7_match[] = {
>> +	{
>> +		.compatible = "nxp,s32n79-ufshc",
>> +		.data = &ufs_hba_s32n79_vops,
>> +	},
>> +	{ /* sentinel */ }
>> +};
>> +MODULE_DEVICE_TABLE(of, ufs_s32n7_match);
>> +
>> +static int ufs_s32n7_probe(struct platform_device *pdev)
>> +{
>> +	const struct ufs_hba_variant_ops *vops;
>> +
>> +	vops = device_get_match_data(&pdev->dev);
>> +	if (!vops)
>> +		return -ENODEV;
>> +
>> +	return ufshcd_pltfrm_init(pdev, vops);
>> +}
>> +
>> +static void ufs_s32n7_remove(struct platform_device *pdev)
>> +{
>> +	ufshcd_pltfrm_remove(pdev);
>> +}
>> +
>> +static const struct dev_pm_ops ufs_s32n7_pm_ops = {
>> +	SET_SYSTEM_SLEEP_PM_OPS(ufshcd_system_suspend, ufshcd_system_resume)
>> +	SET_RUNTIME_PM_OPS(ufshcd_runtime_suspend, ufshcd_runtime_resume, NULL)
>> +	.prepare	= ufshcd_suspend_prepare,
>> +	.complete	= ufshcd_resume_complete,
>> +};
>> +
>> +static struct platform_driver ufs_s32n7_driver = {
>> +	.probe = ufs_s32n7_probe,
>> +	.remove = ufs_s32n7_remove,
>> +	.driver	= {
>> +		.name	= "ufs-s32n7",
>> +		.pm	= &ufs_s32n7_pm_ops,
>> +		.of_match_table	= ufs_s32n7_match,
>> +	},
>> +};
>> +
>> +module_platform_driver(ufs_s32n7_driver);
>> +
>> +MODULE_AUTHOR("Larisa Grigore <larisa.grigore@oss.nxp.com>");
>> +MODULE_DESCRIPTION("NXP S32N7 UFS Host Controller platform driver");
>> +MODULE_LICENSE("Dual BSD/GPL");
>> diff --git a/drivers/ufs/host/ufshcd-dwc.h b/drivers/ufs/host/ufshcd-dwc.h
>> index 53b523ef59df..747383cfb839 100644
>> --- a/drivers/ufs/host/ufshcd-dwc.h
>> +++ b/drivers/ufs/host/ufshcd-dwc.h
>> @@ -13,8 +13,7 @@
>>   #include <ufs/ufshcd.h>
>>   
>>   /* RMMI Attributes */
>> -#define CBREFCLKCTRL2		0x8132
>> -#define CBCRCTRL		0x811F
>> +#define RXSQCONTROL		0x8009
>>   #define CBC10DIRECTCONF2	0x810E
>>   #define CBRATESEL		0x8114
>>   #define CBCREGADDRLSB		0x8116
>> @@ -24,10 +23,16 @@
>>   #define CBCREGRDLSB		0x811A
>>   #define CBCREGRDMSB		0x811B
>>   #define CBCREGRDWRSEL		0x811C
>> +#define CBCRCTRL		0x811F
>> +#define CBREFCLKCTRL2		0x8132
>>   
>>   #define CBREFREFCLK_GATE_OVR_EN		BIT(7)
>>   
>>   /* M-PHY registers */
>> +#define RX_DAC_CTRL(n)		(0x10AF + ((n) * 0x100))
>> +#define RX_DAC_CTRL_OVRD(n)	(0x10B0 + ((n) * 0x100))
>> +#define RX_DAC_CTRL_SEL(n)	(0x10B1 + ((n) * 0x100))
>> +#define RX_DAC_CTRL_EN(n)	(0x10B8 + ((n) * 0x100))
>>   #define RX_OVRD_IN_1(n)		(0x3006 + ((n) * 0x100))
>>   #define RX_PCS_OUT(n)		(0x300F + ((n) * 0x100))
>>   #define FAST_FLAGS(n)		(0x401C + ((n) * 0x100))
>> -- 
>> 2.43.0
>>

Best regards,
Larisa


  reply	other threads:[~2026-09-14 12:20 UTC|newest]

Thread overview: 42+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-08-26 13:40 [PATCH 00/10] Add S32N79RDB UFS support Larisa Grigore
2026-08-26 13:40 ` [PATCH 01/10] ufs: unipro: Move Tx/Rx FSM state definitions to unipro.h Larisa Grigore
2026-08-26 15:05   ` Frank Li
2026-08-27 13:59     ` Larisa Ileana Grigore
2026-08-26 13:40 ` [PATCH 02/10] ufs: unipro: Add TX/RX FSM state attributes Larisa Grigore
2026-08-26 15:09   ` Frank Li
2026-08-26 13:40 ` [PATCH 03/10] ufs: host: Add common Hibern8 TX FSM polling helper Larisa Grigore
2026-08-26 15:17   ` Frank Li
2026-08-26 15:33   ` Bart Van Assche
2026-08-27 14:03     ` Larisa Ileana Grigore
2026-08-27  9:14   ` Bough Chen
2026-08-27 14:10     ` Larisa Ileana Grigore
2026-08-27 15:00   ` Stanley Jhu
2026-08-28 11:13     ` Larisa Ileana Grigore
2026-08-26 13:40 ` [PATCH 04/10] scsi: ufs: Move Versal2 M-PHY CREG access helpers into ufshcd-dwc Larisa Grigore
2026-08-26 15:20   ` Frank Li
2026-08-27 14:11     ` Larisa Ileana Grigore
2026-08-26 15:36   ` Bart Van Assche
2026-08-27 14:13     ` Larisa Ileana Grigore
2026-08-27 10:47   ` Bough Chen
2026-08-28  9:37     ` Larisa Ileana Grigore
2026-08-26 13:40 ` [PATCH 05/10] scsi: ufs: dwc: Export common clock divider and link status helpers Larisa Grigore
2026-08-26 15:22   ` Frank Li
2026-08-26 13:40 ` [PATCH 06/10] dt-bindings: ufs: Add compatibles for S32N79 Host Controller platform driver Larisa Grigore
2026-08-26 14:35   ` Frank Li
2026-08-26 15:57     ` Conor Dooley
2026-08-27 14:37       ` Larisa Ileana Grigore
2026-08-27 14:36     ` Larisa Ileana Grigore
2026-08-27 14:45       ` Frank Li
2026-08-27 15:09         ` Larisa Ileana Grigore
2026-08-28 10:28   ` Krzysztof Kozlowski
2026-09-14 12:05     ` Larisa Ileana Grigore
2026-08-28 10:29   ` Krzysztof Kozlowski
2026-09-14 12:05     ` Larisa Ileana Grigore
2026-08-26 13:40 ` [PATCH 07/10] scsi: ufs: Add NXP S32N79 UFS host controller driver Larisa Grigore
2026-08-26 15:49   ` Frank Li
2026-08-27 14:44     ` Larisa Ileana Grigore
2026-08-28  9:27   ` Bough Chen
2026-09-14 12:20     ` Larisa Ileana Grigore [this message]
2026-08-26 13:40 ` [PATCH 08/10] arm64: dts: freescale: s32n79: Add UFS host controller Larisa Grigore
2026-08-26 13:40 ` [PATCH 09/10] arm64: dts: freescale: s32n79-rdb: Enable UFS Larisa Grigore
2026-08-26 13:40 ` [PATCH 10/10] MAINTAINERS: Add NXP S32N7 UFS host controller entry Larisa Grigore

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=cb5e8a0e-b371-4a77-b023-689c6818b5ae@oss.nxp.com \
    --to=larisa.grigore@oss.nxp.com \
    --cc=Frank.Li@nxp.com \
    --cc=James.Bottomley@HansenPartnership.com \
    --cc=ajay.neeli@amd.com \
    --cc=alim.akhtar@samsung.com \
    --cc=aruizrui@redhat.com \
    --cc=avri.altman@sandisk.com \
    --cc=bvanassche@acm.org \
    --cc=clizzi@redhat.com \
    --cc=conor+dt@kernel.org \
    --cc=devicetree@vger.kernel.org \
    --cc=eballetb@redhat.com \
    --cc=echanude@redhat.com \
    --cc=festevam@gmail.com \
    --cc=haibo.chen@oss.nxp.com \
    --cc=imx@lists.linux.dev \
    --cc=kernel@pengutronix.de \
    --cc=krzk+dt@kernel.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-arm-msm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-scsi@vger.kernel.org \
    --cc=mani@kernel.org \
    --cc=martin.petersen@oracle.com \
    --cc=pedrom.sousa@synopsys.com \
    --cc=robh@kernel.org \
    --cc=s.hauer@pengutronix.de \
    --cc=s32@nxp.com \
    --cc=sai.krishna.potthuri@amd.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®