[PATCH v4 04/10] memory: ti-aemif: Create aemif_check_cs_timings()
Krzysztof Kozlowski
krzk at kernel.org
Thu Nov 21 02:28:01 PST 2024
On 15/11/2024 14:26, Bastien Curutchet wrote:
> aemif_calc_rate() check the validity of a new computed timing against a
> 'max' value given as input. This isn't convenient if we want to check
> the CS timing configuration somewhere else in the code.
>
> Wrap the verification of all the chip select's timing configuration into a
> single function to ease its exportation in upcoming patches.
> Remove the 'max' input from aemif_calc_rate() as it's no longer used.
>
> Signed-off-by: Bastien Curutchet <bastien.curutchet at bootlin.com>
> ---
> drivers/memory/ti-aemif.c | 107 +++++++++++++++++---------------------
> 1 file changed, 49 insertions(+), 58 deletions(-)
>
> diff --git a/drivers/memory/ti-aemif.c b/drivers/memory/ti-aemif.c
> index aec6d6464efa..5c1c6f95185f 100644
> --- a/drivers/memory/ti-aemif.c
> +++ b/drivers/memory/ti-aemif.c
> @@ -132,18 +132,48 @@ struct aemif_device {
> struct aemif_cs_data cs_data[NUM_CS];
> };
>
> +/**
> + * aemif_check_cs_timings - Check the validity of a CS timing configuration.
> + * @timings: timings configuration
> + *
> + * @return: 0 if the timing configuration is valid, negative error number otherwise.
> + */
> +static int aemif_check_cs_timings(struct aemif_cs_timings *timings)
> +{
> + if (timings->ta > TA_MAX)
> + return -EINVAL;
> +
> + if (timings->rhold > RHOLD_MAX)
> + return -EINVAL;
> +
> + if (timings->rstrobe > RSTROBE_MAX)
> + return -EINVAL;
> +
> + if (timings->rsetup > RSETUP_MAX)
> + return -EINVAL;
> +
> + if (timings->whold > WHOLD_MAX)
> + return -EINVAL;
> +
> + if (timings->wstrobe > WSTROBE_MAX)
> + return -EINVAL;
> +
> + if (timings->wsetup > WSETUP_MAX)
> + return -EINVAL;
> +
> + return 0;
> +}
> +
> /**
> * aemif_calc_rate - calculate timing data.
> * @pdev: platform device to calculate for
> * @wanted: The cycle time needed in nanoseconds.
> * @clk: The input clock rate in kHz.
> - * @max: The maximum divider value that can be programmed.
> *
> * On success, returns the calculated timing value minus 1 for easy
> * programming into AEMIF timing registers, else negative errno.
> */
> -static int aemif_calc_rate(struct platform_device *pdev, int wanted,
> - unsigned long clk, int max)
> +static int aemif_calc_rate(struct platform_device *pdev, int wanted, unsigned long clk)
> {
> int result;
>
> @@ -156,10 +186,6 @@ static int aemif_calc_rate(struct platform_device *pdev, int wanted,
> if (result < 0)
> result = 0;
>
> - /* ... But configuring tighter timings is not an option. */
> - else if (result > max)
> - result = -EINVAL;
> -
> return result;
> }
>
> @@ -249,7 +275,6 @@ static int of_aemif_parse_abus_config(struct platform_device *pdev,
> struct aemif_device *aemif = platform_get_drvdata(pdev);
> unsigned long clk_rate = aemif->clk_rate;
> struct aemif_cs_data *data;
> - int ret;
> u32 cs;
> u32 val;
>
> @@ -275,68 +300,34 @@ static int of_aemif_parse_abus_config(struct platform_device *pdev,
> aemif_get_hw_params(pdev, aemif->num_cs++);
>
> /* override the values from device node */
> - if (!of_property_read_u32(np, "ti,cs-min-turnaround-ns", &val)) {
> - ret = aemif_calc_rate(pdev, val, clk_rate, TA_MAX);
> - if (ret < 0)
> - return ret;
> -
> - data->timings.ta = ret;
> - }
> + if (!of_property_read_u32(np, "ti,cs-min-turnaround-ns", &val))
> + data->timings.ta = aemif_calc_rate(pdev, val, clk_rate);
>
You just changed these lines in patch #1. Basically this is partial
revert of #1.
Best regards,
Krzysztof
More information about the linux-mtd
mailing list