Re: [PATCH v1 2/6] i2c: npcm: Fix timeout calculation

["Jonathan Neuschäfer" <j.neuschaefer@gmx.net>]

[-- Attachment #1: Type: text/plain, Size: 3931 bytes --]

Hello,

On Mon, Feb 07, 2022 at 02:33:34PM +0800, Tyrone Ting wrote:
> From: Tali Perry <tali.perry1@gmail.com>
> 
> Use adap.timeout for timeout calculation instead of hard-coded
> value of 35ms.


> Use syscon to access gcr, instead of "compatible".

Please put the GCR/syscon change into a separate patch, because it is
not obvious from the commit title that such a change would happen in
this patch.

> 
> Fixes: 56a1485b102e ("i2c: npcm7xx: Add Nuvoton NPCM I2C controller driver")
> Signed-off-by: Tali Perry <tali.perry1@gmail.com>
> Signed-off-by: Tyrone Ting <kfting@nuvoton.com>
> ---
>  drivers/i2c/busses/i2c-npcm7xx.c | 18 +++++++-----------
>  1 file changed, 7 insertions(+), 11 deletions(-)
> 
> diff --git a/drivers/i2c/busses/i2c-npcm7xx.c b/drivers/i2c/busses/i2c-npcm7xx.c
> index 2ad166355ec9..ddeee6f53621 100644
> --- a/drivers/i2c/busses/i2c-npcm7xx.c
> +++ b/drivers/i2c/busses/i2c-npcm7xx.c
> @@ -2047,7 +2047,7 @@ static int npcm_i2c_master_xfer(struct i2c_adapter *adap, struct i2c_msg *msgs,
>  	u16 nwrite, nread;
>  	u8 *write_data, *read_data;
>  	u8 slave_addr;
> -	int timeout;
> +	unsigned long timeout;
>  	int ret = 0;
>  	bool read_block = false;
>  	bool read_PEC = false;
> @@ -2099,13 +2099,13 @@ static int npcm_i2c_master_xfer(struct i2c_adapter *adap, struct i2c_msg *msgs,
>  	 * 9: bits per transaction (including the ack/nack)
>  	 */
>  	timeout_usec = (2 * 9 * USEC_PER_SEC / bus->bus_freq) * (2 + nread + nwrite);
> -	timeout = max(msecs_to_jiffies(35), usecs_to_jiffies(timeout_usec));
> +	timeout = max(bus->adap.timeout, usecs_to_jiffies(timeout_usec));
>  	if (nwrite >= 32 * 1024 || nread >= 32 * 1024) {
>  		dev_err(bus->dev, "i2c%d buffer too big\n", bus->num);
>  		return -EINVAL;
>  	}
>  
> -	time_left = jiffies + msecs_to_jiffies(DEFAULT_STALL_COUNT) + 1;
> +	time_left = jiffies + timeout + 1;
>  	do {
>  		/*
>  		 * we must clear slave address immediately when the bus is not
> @@ -2131,7 +2131,7 @@ static int npcm_i2c_master_xfer(struct i2c_adapter *adap, struct i2c_msg *msgs,
>  	}
>  
>  	npcm_i2c_init_params(bus);
> -	bus->dest_addr = slave_addr;
> +	bus->dest_addr = slave_addr << 1;

This seems unrelated to timeout calculation.

>  	bus->msgs = msgs;
>  	bus->msgs_num = num;
>  	bus->cmd_err = 0;
> @@ -2233,9 +2233,9 @@ static int npcm_i2c_probe_bus(struct platform_device *pdev)
>  	struct i2c_adapter *adap;
>  	struct clk *i2c_clk;
>  	static struct regmap *gcr_regmap;
> -	static struct regmap *clk_regmap;
>  	int irq;
>  	int ret;
> +	struct device_node *np = pdev->dev.of_node;
>  
>  	bus = devm_kzalloc(&pdev->dev, sizeof(*bus), GFP_KERNEL);
>  	if (!bus)
> @@ -2250,15 +2250,11 @@ static int npcm_i2c_probe_bus(struct platform_device *pdev)
>  		return PTR_ERR(i2c_clk);
>  	bus->apb_clk = clk_get_rate(i2c_clk);
>  
> -	gcr_regmap = syscon_regmap_lookup_by_compatible("nuvoton,npcm750-gcr");
> +	gcr_regmap = syscon_regmap_lookup_by_phandle(np, "syscon");
>  	if (IS_ERR(gcr_regmap))
>  		return PTR_ERR(gcr_regmap);
>  	regmap_write(gcr_regmap, NPCM_I2CSEGCTL, NPCM_I2CSEGCTL_INIT_VAL);
>  
> -	clk_regmap = syscon_regmap_lookup_by_compatible("nuvoton,npcm750-clk");
> -	if (IS_ERR(clk_regmap))
> -		return PTR_ERR(clk_regmap);

I agree that clk_regmap can be removed, but I'd rather see it in a
separate patch, because it's unrelated to the timeout calculation.

> -
>  	bus->reg = devm_platform_ioremap_resource(pdev, 0);
>  	if (IS_ERR(bus->reg))
>  		return PTR_ERR(bus->reg);
> @@ -2269,7 +2265,7 @@ static int npcm_i2c_probe_bus(struct platform_device *pdev)
>  	adap = &bus->adap;
>  	adap->owner = THIS_MODULE;
>  	adap->retries = 3;
> -	adap->timeout = HZ;
> +	adap->timeout = msecs_to_jiffies(35);
>  	adap->algo = &npcm_i2c_algo;
>  	adap->quirks = &npcm_i2c_quirks;
>  	adap->algo_data = bus;


Best regards,
Jonathan

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]