| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <202303172150163e1de7e3@mail.local>
Date: Fri, 17 Mar 2023 22:50:16 +0100
From: Alexandre Belloni <alexandre.belloni@...tlin.com>
To: Ibrahim Tilki <Ibrahim.Tilki@...log.com>
Cc: a.zummo@...ertech.it, jdelvare@...e.com, linux@...ck-us.net,
robh+dt@...nel.org, krzysztof.kozlowski+dt@...aro.org,
linux-rtc@...r.kernel.org, linux-kernel@...r.kernel.org,
linux-hwmon@...r.kernel.org, devicetree@...r.kernel.org,
Zeynep Arslanbenzer <Zeynep.Arslanbenzer@...log.com>
Subject: Re: [PATCH v4 1/2] drivers: rtc: add max313xx series rtc driver
On 15/03/2023 19:16:25+0300, Ibrahim Tilki wrote:
> +static int max313xx_trickle_charger_setup(struct device *dev)
> +{
> + struct max313xx *rtc = dev_get_drvdata(dev);
> + bool trickle_diode_en;
> + u32 charger_en = 0;
> + int index, reg;
> + u32 ohms = 0;
> + int ret;
> +
> + device_property_read_u32(dev, "aux-voltage-chargeable", &charger_en);
> + if (charger_en > 1)
> + return dev_err_probe(dev, -EINVAL,
> + "invalid value %d for property aux-voltage-chargeable\n",
> + charger_en);
This is too verbose, you can simply return -EINVAL here. dtbs_check will
report the invalid value.
> +
> + if (!charger_en)
> + return 0;
> +
> + trickle_diode_en = device_property_read_bool(dev, "adi,trickle-diode-enable");
There is no point having this property, simply use aux-voltage-chargeable.
> + device_property_read_u32(dev, "trickle-resistor-ohms", &ohms);
> +
> + if (!rtc->chip->trickle_reg) {
> + dev_warn(dev, "device does not have trickle charger\n");
Simply return 0, no string.
> + return 0;
> + }
> +
> + index = find_closest(ohms, max313xx_trickle_ohms,
> + ARRAY_SIZE(max313xx_trickle_ohms)) + 1;
> +
> + switch (rtc->id) {
> + case ID_MAX31329:
> + reg = FIELD_PREP(MAX31329_TRICKLE_ENABLE_BIT, 1) |
> + FIELD_PREP(MAX31329_43_TRICKLE_RES_MASK, index) |
> + FIELD_PREP(MAX31329_43_TRICKLE_DIODE_EN, trickle_diode_en);
> + break;
> + case ID_MAX31331:
> + case ID_MAX31334:
> + reg = FIELD_PREP(MAX3133X_TRICKLE_ENABLE_BIT, 1) |
> + FIELD_PREP(MAX3133X_TRICKLE_RES_MASK, index) |
> + FIELD_PREP(MAX3133X_TRICKLE_DIODE_EN, trickle_diode_en);
> + break;
> + case ID_MAX31341:
> + if (index == 1)
> + index = 0;
> +
> + reg = FIELD_PREP(MAX31341_TRICKLE_ENABLE_BIT, 1) |
> + FIELD_PREP(MAX31341_TRICKLE_RES_MASK, index) |
> + FIELD_PREP(MAX31341_TRICKLE_DIODE_EN, trickle_diode_en);
> +
> + ret = regmap_set_bits(rtc->regmap, MAX31341_POWER_MGMT_REG,
> + MAX31341_POWER_MGMT_TRICKLE_BIT);
> + if (ret)
> + return ret;
> +
> + break;
> + case ID_MAX31343:
> + reg = FIELD_PREP(MAX31329_43_TRICKLE_RES_MASK, index) |
> + FIELD_PREP(MAX31329_43_TRICKLE_DIODE_EN, trickle_diode_en) |
> + FIELD_PREP(MAX31343_TRICKLE_ENABLE_MASK,
> + MAX31343_TRICKLE_ENABLE_CODE);
> + break;
> + default:
> + return -EOPNOTSUPP;
> + }
> +
> + return regmap_write(rtc->regmap, rtc->chip->trickle_reg, reg);
> +}
> +
> +static int max313xx_find_clkin_freq_index(struct clk *clk)
> +{
> + unsigned long rate = clk_get_rate(clk);
> + int freq;
> + int i;
> +
> + i = find_closest(rate, max313xx_clkin_freq,
> + ARRAY_SIZE(max313xx_clkin_freq));
> + if (max313xx_clkin_freq[i] == rate)
> + return i;
> +
> + for (i = ARRAY_SIZE(max313xx_clkin_freq) - 1; i >= 0; i--) {
> + freq = max313xx_clkin_freq[i];
> + if (freq == clk_round_rate(clk, freq))
> + return i;
> + }
> +
> + /* supplied clock cannot produce one of desired frequency rate */
> + return -ENODEV;
> +}
> +
> +static int max313xx_clkin_init(struct device *dev)
> +{
> + struct max313xx *rtc = dev_get_drvdata(dev);
> + int rate;
> + int ret;
> +
> + rtc->clkin = devm_clk_get_enabled(dev, NULL);
> + if (IS_ERR(rtc->clkin)) {
> + if (PTR_ERR(rtc->clkin) == -ENOENT)
> + rtc->clkin = NULL;
> + else
> + return dev_err_probe(dev, PTR_ERR(rtc->clkin),
> + "error while clkin setup\n");
> + }
> +
> + if (!rtc->clkin) {
> + switch (rtc->id) {
> + case ID_MAX31329:
> + return regmap_clear_bits(rtc->regmap,
> + MAX31329_CONFIG2_REG,
> + MAX31329_CONFIG2_CLKIN_EN);
> + case ID_MAX31341:
> + case ID_MAX31342:
> + return regmap_clear_bits(rtc->regmap,
> + MAX31341_42_CONFIG1_REG,
> + MAX31341_42_CONFIG1_CLKIN_EN);
> + default:
> + return 0;
> + }
> + }
> +
> + rate = max313xx_find_clkin_freq_index(rtc->clkin);
> + if (rate < 0)
> + return dev_err_probe(dev, rate,
> + "clkin cannot produce required frequency\n");
> +
> + ret = clk_set_rate(rtc->clkin, max313xx_clkin_freq[rate]);
> + if (ret)
> + return ret;
> +
> + switch (rtc->id) {
> + case ID_MAX31329:
> + ret = regmap_update_bits(rtc->regmap, MAX31329_CONFIG2_REG,
> + MAX31329_CONFIG2_CLKIN_FREQ, rate);
> + if (ret)
> + return ret;
> +
> + return regmap_set_bits(rtc->regmap, MAX31329_CONFIG2_REG,
> + MAX31329_CONFIG2_CLKIN_EN);
> + case ID_MAX31341:
> + case ID_MAX31342:
> + ret = regmap_update_bits(rtc->regmap, MAX31341_42_CONFIG1_REG,
> + MAX31341_42_CONFIG1_CLKIN_FREQ,
> + FIELD_PREP(MAX31341_42_CONFIG1_CLKIN_FREQ, rate));
> + if (ret)
> + return ret;
> +
> + return regmap_set_bits(rtc->regmap, MAX31341_42_CONFIG1_REG,
> + MAX31341_42_CONFIG1_CLKIN_EN);
> + default:
> + rtc->clkin = NULL;
> + dev_warn(dev, "device does not have clock input\n");
> + return 0;
> + }
> +}
> +
> +static int max313xx_irq_init(struct device *dev, const char *devname)
> +{
> + struct max313xx *rtc = dev_get_drvdata(dev);
> + int ret;
> +
> + switch (rtc->id) {
> + case ID_MAX31328:
> + /* max31328 sqw and interrupt pin is shared */
> + if (rtc->irq > 0 && rtc->clkout.clk)
> + return dev_err_probe(dev, -EOPNOTSUPP,
> + "cannot have both sqw clock output and irq enabled");
This still doesn't work, having a clock registered doesn't mean it is
used or muxed. Also, you want to always register it so it gets disabled
when it is not used.
> +
> + break;
> + case ID_MAX31331:
> + case ID_MAX31334:
> + if (rtc->clkout.clk) {
> + /*
> + * interrupt muxing depends on clkout so enable clkout
> + * if configured before requesting interrupt
> + */
> + ret = clk_prepare_enable(rtc->clkout.clk);
> + if (ret)
> + return dev_err_probe(dev, ret,
> + "cannot enable clkout\n");
> + }
> + break;
> + default:
> + if (rtc->clkin) {
> + if (rtc->clkout.clk && rtc->irq > 0)
> + return dev_err_probe(dev, -EOPNOTSUPP,
> + "irq not possible when both clkin and clkout are configured\n");
> +
> + if (rtc->irq <= 0)
> + break;
> +
> + /* clkout needs to be disabled for interrupt */
> + if (rtc->chip->clkout->en_invert)
> + ret = regmap_set_bits(rtc->regmap,
> + rtc->chip->clkout->reg,
> + rtc->chip->clkout->en_bit);
> + else
> + ret = regmap_clear_bits(rtc->regmap,
> + rtc->chip->clkout->reg,
> + rtc->chip->clkout->en_bit);
> +
What if you want alarm1 on INTA without any clkin? This doesn't seem to
be configured.
> + if (ret)
> + return ret;
> + }
> + break;
> + }
> +
> + if (rtc->irq > 0) {
> + return devm_request_threaded_irq(dev, rtc->irq, NULL,
> + &max313xx_irq, IRQF_ONESHOT,
> + devname, rtc);
> + }
You still want to parse wakeup-source in case INTA/INTB is connected
directly to a PMIC for exemple.
> +
> + clear_bit(RTC_FEATURE_ALARM, rtc->rtc->features);
> +
> + return 0;
> +}
> +
> +static int max313xx_probe(struct i2c_client *client,
> + const struct i2c_device_id *id)
> +{
> + struct device *dev = &client->dev;
> + struct max313xx *max313xx;
> + struct device *hwmon;
> + const void *match;
> + int ret;
> +
> + max313xx->irq = client->irq;
> +
> + ret = max313xx_clkout_register(dev);
> + if (ret)
> + return ret;
> +
> + ret = max313xx_clkin_init(dev);
> + if (ret)
> + return ret;
> +
> + /* IRQ wiring depends on the clock configuration so parse them first */
> + ret = max313xx_irq_init(dev, client->name);
> + if (ret)
> + return ret;
> +
> + if (max313xx->chip->ram_size) {
> + max313xx_nvmem_cfg.size = max313xx->chip->ram_size;
> + max313xx_nvmem_cfg.priv = max313xx;
> +
> + ret = devm_rtc_nvmem_register(max313xx->rtc, &max313xx_nvmem_cfg);
> + if (ret)
> + dev_warn(dev, "cannot register rtc nvmem\n");
> + }
> +
> + if (max313xx->chip->temp_reg) {
> + hwmon = devm_hwmon_device_register_with_info(dev, client->name,
> + max313xx,
> + &max313xx_chip_info,
> + NULL);
> + if (IS_ERR(hwmon))
> + dev_warn(dev, "cannot register hwmon device: %li\n",
> + PTR_ERR(hwmon));
> + }
> +
> + return max313xx_trickle_charger_setup(dev);
> +}
> +
> +static const struct of_device_id max313xx_of_id[] = {
> + { .compatible = "adi,max31328", .data = &chip[ID_MAX31328] },
> + { .compatible = "adi,max31329", .data = &chip[ID_MAX31329] },
> + { .compatible = "adi,max31331", .data = &chip[ID_MAX31331] },
> + { .compatible = "adi,max31334", .data = &chip[ID_MAX31334] },
> + { .compatible = "adi,max31341", .data = &chip[ID_MAX31341] },
> + { .compatible = "adi,max31342", .data = &chip[ID_MAX31342] },
> + { .compatible = "adi,max31343", .data = &chip[ID_MAX31343] },
> + { }
> +};
> +MODULE_DEVICE_TABLE(of, max313xx_of_id);
> +
> +static const struct i2c_device_id max313xx_id[] = {
> + { "max31328", (kernel_ulong_t)&chip[ID_MAX31328] },
> + { "max31329", (kernel_ulong_t)&chip[ID_MAX31329] },
> + { "max31331", (kernel_ulong_t)&chip[ID_MAX31331] },
> + { "max31334", (kernel_ulong_t)&chip[ID_MAX31334] },
> + { "max31341", (kernel_ulong_t)&chip[ID_MAX31341] },
> + { "max31342", (kernel_ulong_t)&chip[ID_MAX31342] },
> + { "max31343", (kernel_ulong_t)&chip[ID_MAX31343] },
> + { }
> +};
> +MODULE_DEVICE_TABLE(i2c, max313xx_id);
> +
> +static struct i2c_driver max313xx_driver = {
> + .driver = {
> + .name = "rtc-max313xx",
> + .of_match_table = max313xx_of_id,
> + },
> + .probe = max313xx_probe,
> + .id_table = max313xx_id,
> +};
> +module_i2c_driver(max313xx_driver);
> +
> +MODULE_DESCRIPTION("Analog Devices MAX313XX RTCs");
> +MODULE_AUTHOR("Zeynep Arslanbenzer <Zeynep.Arslanbenzer@...log.com>");
> +MODULE_AUTHOR("Ibrahim Tilki <Ibrahim.Tilki@...log.com>");
> +MODULE_SOFTDEP("pre: regmap-i2c");
> +MODULE_LICENSE("GPL");
> +MODULE_VERSION("1.0");
> --
> 2.25.1
>
--
Alexandre Belloni, co-owner and COO, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com
Powered by blists - more mailing lists