Re: [PATCH 3/3] ARM: AT91: IIO: fix missing Sample and Hold time

[Maxime Ripard <maxime.ripard@free-electrons.com>]

Hi,

Le 19/12/2012 19:37, Jean-Christophe PLAGNIOL-VILLARD a écrit :
> On the at91_adc a minimal Sample and Hold Time is necessary for the ADC to
> guarantee the best converted final value between two channels selection.
> This time has to be programmed through the bitfield SHTIM in the
> Mode Register ADC_MR.
> 
> Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
> Cc: linux-iio@vger.kernel.org
> Cc: Nicolas Ferre <nicolas.ferre@atmel.com>
> Cc: Ludovic Desroches <ludovic.desroches@atmel.com>
> ---
>  Documentation/devicetree/bindings/arm/atmel-adc.txt |    1 +
>  drivers/iio/adc/at91_adc.c                          |   15 ++++++++++++++-
>  2 files changed, 15 insertions(+), 1 deletion(-)
> 
> diff --git a/Documentation/devicetree/bindings/arm/atmel-adc.txt b/Documentation/devicetree/bindings/arm/atmel-adc.txt
> index efb6f02..dd2ca90 100644
> --- a/Documentation/devicetree/bindings/arm/atmel-adc.txt
> +++ b/Documentation/devicetree/bindings/arm/atmel-adc.txt
> @@ -26,6 +26,7 @@ Optional properties:
>  		       atmel,adc-res-names property. If not specified, the highest
>  		       resolution will be used.
>    - atmel,atmel,adc-sleep-mode: Boolean to enable of sleep mode when no conversion
> +  - atmel,adc-sample-hold-time: Sample and Hold Time

In which unit? nanoseconds? number of clock ticks?

>   
>  Optional trigger Nodes:
>    - Required properties:
> diff --git a/drivers/iio/adc/at91_adc.c b/drivers/iio/adc/at91_adc.c
> index c563488..9481375 100644
> --- a/drivers/iio/adc/at91_adc.c
> +++ b/drivers/iio/adc/at91_adc.c
> @@ -52,6 +52,7 @@ struct at91_adc_state {
>  	void __iomem		*reg_base;
>  	struct at91_adc_reg_desc *registers;
>  	u8			startup_time;
> +	u8			sample_hold_time;
>  	bool			sleep_mode;
>  	struct iio_trigger	**trig;
>  	struct at91_adc_trigger	*trigger_list;
> @@ -465,6 +466,9 @@ static int at91_adc_probe_dt(struct at91_adc_state *st,
>  	}
>  	st->startup_time = prop;
>  
> +	prop = 0;
> +	of_property_read_u32(node, "atmel,adc-sample-hold-time", &prop);
> +	st->sample_hold_time = prop;
>  
>  	if (of_property_read_u32(node, "atmel,adc-vref", &prop)) {
>  		dev_err(&idev->dev, "Missing adc-vref property in the DT.\n");
> @@ -578,7 +582,7 @@ static const struct iio_info at91_adc_info = {
>  
>  static int __devinit at91_adc_probe(struct platform_device *pdev)
>  {
> -	unsigned int prsc, mstrclk, ticks, adc_clk;
> +	unsigned int prsc, mstrclk, ticks, adc_clk, shtim;
>  	int ret;
>  	struct iio_dev *idev;
>  	struct at91_adc_state *st;
> @@ -691,12 +695,21 @@ static int __devinit at91_adc_probe(struct platform_device *pdev)
>  	 */
>  	ticks = round_up((st->startup_time * adc_clk /
>  			  1000000) - 1, 8) / 8;
> +	/*
> +	 * a minimal Sample and Hold Time is necessary for the ADC to guarantee
> +	 * the best converted final value between two channels selection
> +	 * The formula thus is : Sample and Hold Time = (shtim + 1) / ADCClock
> +	 */
> +	shtim = round_up((st->sample_hold_time * adc_clk /
> +			  1000000) - 1, 1);
> +

Since, from all the datasheet I have (G20, G45 and X5), the minimum time
is always 500ns, do we really need to add a dt property for that?

Maxime

-- 
Maxime Ripard, Free Electrons
Kernel, drivers, real-time and embedded Linux
development, consulting, training and support.
http://free-electrons.com