[U-Boot-Users] [PATCH 1/1] Fix I2C for m547x/548x and m5445x processors.
Wolfgang Denk
wd at denx.de
Mon Jul 7 16:59:37 CEST 2008
In message <1215439715-28742-2-git-send-email-luigi.mantellini at idf-hit.com> you wrote:
> ---
> drivers/i2c/fsl_i2c.c | 68 +++++++++++++++++++++++++++++++++++++++++++++++++
> 1 files changed, 68 insertions(+), 0 deletions(-)
>
> diff --git a/drivers/i2c/fsl_i2c.c b/drivers/i2c/fsl_i2c.c
> index 0690340..8fd73c0 100644
> --- a/drivers/i2c/fsl_i2c.c
> +++ b/drivers/i2c/fsl_i2c.c
> @@ -78,6 +78,73 @@ static const struct fsl_i2c *i2c_dev[2] = {
> * register. See the application note AN2919 "Determining the I2C Frequency
> * Divider Ratio for SCL"
> */
> +#if defined(CONFIG_MCF547x_8x) || defined(CONFIG_MCF5445x)
> +static const struct {
> + u8 fdr;
> + unsigned short divider;
> +} fsl_i2c_speed_map[] = {
> + {.fdr=0x20, .divider= 20}, {.fdr=0x21, .divider= 22}, {.fdr=0x22, .divider= 24},
> + {.fdr=0x23, .divider= 26}, {.fdr=0x00, .divider= 28}, {.fdr=0x24, .divider= 28},
> + {.fdr=0x01, .divider= 30}, {.fdr=0x25, .divider= 32}, {.fdr=0x02, .divider= 34},
> + {.fdr=0x26, .divider= 36}, {.fdr=0x03, .divider= 40}, {.fdr=0x27, .divider= 40},
> + {.fdr=0x04, .divider= 44}, {.fdr=0x05, .divider= 48}, /*{.fdr=0x28, .divider= 48},*/
> + {.fdr=0x06, .divider= 56}, /*{.fdr=0x29, .divider= 56},*/ {.fdr=0x2A, .divider= 64},
> + {.fdr=0x07, .divider= 68}, {.fdr=0x2B, .divider= 72}, {.fdr=0x08, .divider= 80},
> + /*{.fdr=0x2C, .divider= 80},*/ {.fdr=0x09, .divider= 88}, {.fdr=0x2D, .divider= 96},
> + {.fdr=0x0A, .divider= 104}, {.fdr=0x2E, .divider= 112}, {.fdr=0x0B, .divider= 128},
> + /*{.fdr=0x2F, .divider= 128},*/ {.fdr=0x0C, .divider= 144}, {.fdr=0x0D, .divider= 160},
> + /*{.fdr=0x30, .divider= 160},*/ {.fdr=0x0E, .divider= 192}, /*{.fdr=0x31, .divider= 192},*/
> + {.fdr=0x32, .divider= 224}, {.fdr=0x0F, .divider= 240}, {.fdr=0x33, .divider= 256},
> + {.fdr=0x10, .divider= 288}, {.fdr=0x11, .divider= 320}, /*{.fdr=0x34, .divider= 320},*/
> + {.fdr=0x12, .divider= 384}, /*{.fdr=0x35, .divider= 384},*/ {.fdr=0x36, .divider= 448},
> + /*{.fdr=0x13, .divider= 480},*/ {.fdr=0x37, .divider= 512}, {.fdr=0x14, .divider= 576},
> + {.fdr=0x15, .divider= 640}, /*{.fdr=0x38, .divider= 640},*/ {.fdr=0x16, .divider= 768},
> + /*{.fdr=0x39, .divider= 768},*/ {.fdr=0x3A, .divider= 896}, {.fdr=0x17, .divider= 960},
> + {.fdr=0x3B, .divider=1024}, {.fdr=0x18, .divider=1152}, {.fdr=0x19, .divider=1280},
> + /*{.fdr=0x3C, .divider=1280},*/ {.fdr=0x1A, .divider=1536}, /*{.fdr=0x3D, .divider=1536},*/
> + {.fdr=0x3E, .divider=1792}, {.fdr=0x1B, .divider=1920}, {.fdr=0x3F, .divider=2048},
> + {.fdr=0x1C, .divider=2304}, {.fdr=0x1D, .divider=2560}, {.fdr=0x1E, .divider=3072},
> + {.fdr=0x1F, .divider=3840}, {.fdr=0x00, .divider=-1}
> +};
> +
> +/**
> + * Set the I2C bus speed for a given I2C device
> + *
> + * @param dev: the I2C device
> + * @i2c_clk: I2C bus clock frequency
> + * @speed: the desired speed of the bus
> + *
> + * The I2C device must be stopped before calling this function.
> + *
> + * The return value is the actual bus speed that is set.
> + */
> +static unsigned int set_i2c_bus_speed(const struct fsl_i2c *dev,
> + unsigned int i2c_clk, unsigned int speed)
> +{
> + unsigned short divider = min(i2c_clk / speed, (unsigned short) -1);
> + unsigned int i;
> + u8 fdr = 0;
> +
> + /*
> + * We want to choose an FDR/DFSR that generates an I2C bus speed that
> + * is equal to or lower than the requested speed. That means that we
> + * want the first divider that is equal to or greater than the
> + * calculated divider.
> + */
> +
> + for (i = 0; i < ARRAY_SIZE(fsl_i2c_speed_map); i++)
> + if (fsl_i2c_speed_map[i].divider >= divider) {
> + fdr = fsl_i2c_speed_map[i].fdr;
> + speed = i2c_clk / fsl_i2c_speed_map[i].divider;
> + break;
> + }
> +
> + writeb(fdr, &dev->fdr); /* set bus speed */
> +
> + return speed;
> +}
> +#else /* ! (defined(CONFIG_MCF547x_8x) || defined(CONFIG_MCF5445x) */
> +
What a nightmare of data and code. So many magic numbers, so much
potential for errors. And for what? Just for I2C support in U-Boot?
Are you really sure you don't simply want to use the bit-banging
driver?
Best regards,
Wolfgang Denk
--
DENX Software Engineering GmbH, MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd at denx.de
Committee, n.: A group of men who individually can do nothing but as
a group decide that nothing can be done. - Fred Allen
More information about the U-Boot
mailing list