From: jonghwa3.lee@samsung.com (jonghwa3.lee at samsung.com) Date: Wed, 23 May 2012 13:40:13 +0900 Subject: [PATCH v3 2/2] regulator: Add support for MAX77686. In-Reply-To: References: <4fbb2ada.63de440a.589d.78e2@mx.google.com> <4FBC3FF0.5080408@samsung.com> Message-ID: <4FBC6A2D.6040408@samsung.com> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org On 2012? 05? 23? 13:16, Yadwinder Singh Brar wrote: >>> + max77686_update_reg(i2c, MAX77686_REG_BUCK2CTRL1, >>> + max77686->ramp_delay << 6, RAMP_MASK); >>> + max77686_update_reg(i2c, MAX77686_REG_BUCK3CTRL1, >>> + max77686->ramp_delay << 6, RAMP_MASK); >>> + max77686_update_reg(i2c, MAX77686_REG_BUCK4CTRL1, >>> + max77686->ramp_delay << 6, RAMP_MASK); >>> + >> >> >> Why do you use i2c client still? If you registered regmap you can use >> its API. I recommend you to use regmap_update_bits() directly. >> >> > > Yes, we are using regmap_update_bits(). max77686_update_reg() is just > a wrapper over it. > Yes, i know what you mean. However it doesn't need max77686_update_reg() any more since it uses regmap API. Why don't you just pass iodev->regmap to regmap_update_bits(). It is clear that there is no reason for using i2c client as a medium. Please check regulator and mfd driver of my previous patch. Thanks.