[PATCH v2 5/6] mfd: qcom-pm8008: Add support for PM8010 PMIC
Lee Jones
lee at kernel.org
Tue Sep 22 02:33:27 PDT 2026
On Mon, 07 Sep 2026, Nihal Kumar Gupta wrote:
> From: Jishnu Prakash <jishnu.prakash at oss.qualcomm.com>
>
> The PM8010 is a variant of the PM8008 PMIC with a slightly different
> IRQ layout and MFD cells. Introduce per-variant match data (IRQ chip
> descriptor and MFD cells) selected via the new "qcom,pm8010-i2c"
> compatible string, and support probing without an interrupt line by
> falling back to a reduced set of MFD cells when the client has no IRQ
> assigned.
>
> Signed-off-by: Jishnu Prakash <jishnu.prakash at oss.qualcomm.com>
> Signed-off-by: Dhruvin Rajpura <drajpura at qti.qualcomm.com>
> Signed-off-by: Konrad Dybcio <konrad.dybcio at oss.qualcomm.com>
> Signed-off-by: Nihal Kumar Gupta <nihal.gupta at oss.qualcomm.com>
Who are all of these people?
Did they all work on this? If so, where are their Co-developed-bys?
> ---
> drivers/mfd/qcom-pm8008.c | 140 ++++++++++++++++++++++++++++++++++++----------
> 1 file changed, 112 insertions(+), 28 deletions(-)
>
> diff --git a/drivers/mfd/qcom-pm8008.c b/drivers/mfd/qcom-pm8008.c
> index 60204cc9a2dc60cc1fe1b64030f5d803100e7874..28532cef7908ef3409a3dfc0779aae3b1a3885b9 100644
> --- a/drivers/mfd/qcom-pm8008.c
> +++ b/drivers/mfd/qcom-pm8008.c
> @@ -34,6 +34,7 @@ enum {
> PM8008_GPIO1,
> PM8008_GPIO2,
> PM8008_NUM_PERIPHS,
> + PM8010_NUM_PERIPHS = 2,
> };
>
> #define PM8008_PERIPH_0_BASE 0x900
> @@ -55,6 +56,10 @@ enum {
> #define PM8008_IRQ_GPIO1 6
> #define PM8008_IRQ_GPIO2 7
>
> +#define PM8010_IRQ_MISC_MBG_FAULT 0
> +/* 1-3 are unused */
This is not required.
> +#define PM8010_IRQ_MISC_LDO_OCP 4
> +
> enum {
> SET_TYPE_INDEX,
> POLARITY_HI_INDEX,
> @@ -88,6 +93,12 @@ static const struct regmap_irq pm8008_irqs[] = {
> _IRQ(PM8008_IRQ_GPIO2, PM8008_GPIO2, BIT(0), IRQ_TYPE_SENSE_MASK),
> };
>
> +static const struct regmap_irq pm8010_irqs[] = {
> + _IRQ(PM8010_IRQ_MISC_MBG_FAULT, PM8008_MISC, BIT(0), IRQ_TYPE_EDGE_RISING),
> + _IRQ(PM8010_IRQ_MISC_LDO_OCP, PM8008_MISC, BIT(4), IRQ_TYPE_EDGE_RISING),
> + _IRQ(PM8008_IRQ_TEMP_ALARM, PM8008_TEMP_ALARM, BIT(0), IRQ_TYPE_SENSE_MASK),
> +};
> +
> static const unsigned int pm8008_periph_base[] = {
> PM8008_PERIPH_0_BASE,
> PM8008_PERIPH_1_BASE,
> @@ -158,6 +169,25 @@ static const struct regmap_irq_chip pm8008_irq_chip = {
> .get_irq_reg = pm8008_get_irq_reg,
> };
>
> +static const struct regmap_irq_chip pm8010_irq_chip = {
> + .name = "pm8010",
> + .main_status = I2C_INTR_STATUS_BASE,
> + .num_main_regs = 1,
> + .irqs = pm8010_irqs,
> + .num_irqs = ARRAY_SIZE(pm8010_irqs),
> + .num_regs = PM8010_NUM_PERIPHS,
> + .status_base = INT_LATCHED_STS_OFFSET,
> + .mask_base = INT_EN_CLR_OFFSET,
> + .unmask_base = INT_EN_SET_OFFSET,
> + .mask_unmask_non_inverted = true,
> + .ack_base = INT_LATCHED_CLR_OFFSET,
> + .config_base = pm8008_config_regs,
> + .num_config_bases = ARRAY_SIZE(pm8008_config_regs),
> + .num_config_regs = PM8010_NUM_PERIPHS,
> + .set_type_config = pm8008_set_type_config,
> + .get_irq_reg = pm8008_get_irq_reg,
> +};
> +
> static const struct regmap_config qcom_mfd_regmap_cfg = {
> .name = "primary",
> .reg_bits = 16,
> @@ -179,10 +209,32 @@ static const struct resource pm8008_temp_res[] = {
>
> static const struct mfd_cell pm8008_cells[] = {
> MFD_CELL_NAME("pm8008-regulator"),
> - MFD_CELL_RES("qpnp-temp-alarm", pm8008_temp_res),
> + MFD_CELL_RES("spmi-temp-alarm", pm8008_temp_res),
Why is this cell being renamed from 'qpnp-temp-alarm' to 'spmi-temp-alarm'?
There is no mention of this change in the commit log, and if intentional, it
should be submitted in its own separate patch.
> MFD_CELL_NAME("pm8008-gpio"),
> };
>
> +static const struct mfd_cell pm8010_cells[] = {
> + MFD_CELL_NAME("pm8010-regulator"),
> + MFD_CELL_RES("spmi-temp-alarm", pm8008_temp_res),
> + MFD_CELL_NAME("pm8008-gpio"),
> +};
> +
> +static const struct mfd_cell pm8008_no_irq_cells[] = {
> + MFD_CELL_NAME("pm8008-regulator"),
> +};
> +
> +static const struct mfd_cell pm8010_no_irq_cells[] = {
> + MFD_CELL_NAME("pm8010-regulator"),
> +};
> +
> +struct pm8008_match_data {
> + const struct regmap_irq_chip *irq_chip_desc;
> + const struct mfd_cell *mfd_cells;
> + const struct mfd_cell *no_irq_mfd_cells;
> + int num_mfd_cells;
> + int no_irq_num_mfd_cells;
> +};
> +
> static void devm_irq_domain_fwnode_release(void *data)
> {
> struct fwnode_handle *fwnode = data;
> @@ -192,15 +244,22 @@ static void devm_irq_domain_fwnode_release(void *data)
>
> static int pm8008_probe(struct i2c_client *client)
> {
> - struct regmap_irq_chip_data *irq_data;
> + struct regmap_irq_chip_data *irq_data = NULL;
> + const struct pm8008_match_data *data;
> struct device *dev = &client->dev;
> struct regmap *regmap, *regmap2;
> struct fwnode_handle *fwnode;
> + const struct mfd_cell *cells;
> struct i2c_client *dummy;
> struct gpio_desc *reset;
> + int num_cells;
> char *name;
> int ret;
>
> + data = device_get_match_data(dev);
> + if (!data)
> + return dev_err_probe(dev, -ENODATA, "Missing driver match data\n");
> +
> dummy = devm_i2c_new_dummy_device(dev, client->adapter, client->addr + 1);
> if (IS_ERR(dummy)) {
> ret = PTR_ERR(dummy);
> @@ -231,37 +290,62 @@ static int pm8008_probe(struct i2c_client *client)
> */
> usleep_range(1000, 2000);
>
> - name = devm_kasprintf(dev, GFP_KERNEL, "%pOF-internal", dev->of_node);
> - if (!name)
> - return -ENOMEM;
> -
> - name = strreplace(name, '/', ':');
> -
> - fwnode = irq_domain_alloc_named_fwnode(name);
> - if (!fwnode)
> - return -ENOMEM;
> -
> - ret = devm_add_action_or_reset(dev, devm_irq_domain_fwnode_release, fwnode);
> - if (ret)
> - return ret;
> -
> - ret = devm_regmap_add_irq_chip_fwnode(dev, fwnode, regmap, client->irq,
> - IRQF_SHARED, 0, &pm8008_irq_chip, &irq_data);
> - if (ret) {
> - dev_err(dev, "failed to add IRQ chip: %d\n", ret);
> - return ret;
> + if (client->irq) {
Turn this massive if clause into a function.
> + name = devm_kasprintf(dev, GFP_KERNEL, "%pOF-internal", dev->of_node);
> + if (!name)
> + return -ENOMEM;
> +
> + name = strreplace(name, '/', ':');
> +
> + fwnode = irq_domain_alloc_named_fwnode(name);
> + if (!fwnode)
> + return -ENOMEM;
> +
> + ret = devm_add_action_or_reset(dev, devm_irq_domain_fwnode_release, fwnode);
> + if (ret)
> + return ret;
> +
> + ret = devm_regmap_add_irq_chip_fwnode(dev, fwnode, regmap, client->irq,
> + IRQF_SHARED, 0, data->irq_chip_desc,
> + &irq_data);
> + if (ret) {
> + dev_err(dev, "failed to add IRQ chip: %d\n", ret);
> + return ret;
> + }
> +
> + /* Needed by GPIO driver. */
> + dev_set_drvdata(dev, regmap_irq_get_domain(irq_data));
> + cells = data->mfd_cells;
> + num_cells = data->num_mfd_cells;
> + } else {
> + cells = data->no_irq_mfd_cells;
> + num_cells = data->no_irq_num_mfd_cells;
> }
>
> - /* Needed by GPIO driver. */
> - dev_set_drvdata(dev, regmap_irq_get_domain(irq_data));
> -
> - return devm_mfd_add_devices(dev, PLATFORM_DEVID_AUTO, pm8008_cells,
> - ARRAY_SIZE(pm8008_cells), NULL, 0,
> - regmap_irq_get_domain(irq_data));
> + return devm_mfd_add_devices(dev, PLATFORM_DEVID_AUTO, cells,
> + num_cells, NULL, 0,
> + regmap_irq_get_domain(irq_data));
> }
>
> +static const struct pm8008_match_data pm8008_data = {
> + .irq_chip_desc = &pm8008_irq_chip,
> + .mfd_cells = pm8008_cells,
> + .no_irq_mfd_cells = pm8008_no_irq_cells,
> + .num_mfd_cells = ARRAY_SIZE(pm8008_cells),
> + .no_irq_num_mfd_cells = ARRAY_SIZE(pm8008_no_irq_cells),
> +};
> +
> +static const struct pm8008_match_data pm8010_data = {
> + .irq_chip_desc = &pm8010_irq_chip,
> + .mfd_cells = pm8010_cells,
> + .no_irq_mfd_cells = pm8010_no_irq_cells,
> + .num_mfd_cells = ARRAY_SIZE(pm8010_cells),
> + .no_irq_num_mfd_cells = ARRAY_SIZE(pm8010_no_irq_cells),
> +};
> +
> static const struct of_device_id pm8008_match[] = {
> - { .compatible = "qcom,pm8008", },
> + { .compatible = "qcom,pm8008", .data = &pm8008_data },
> + { .compatible = "qcom,pm8010-i2c", .data = &pm8010_data },
We do not allow data from one registration mechanism (MFD) to be piped through
another (OF). Please pass through an enum identifier instead and select the
cells and IRQ chip in a 'switch()' statement? Also, why does this compatible
carry an '-i2c' suffix when the existing device is just "qcom,pm8008"?
> { },
> };
> MODULE_DEVICE_TABLE(of, pm8008_match);
>
> --
> 2.34.1
>
--
Lee Jones
More information about the linux-phy
mailing list