[PATCH v2] powerpc: i2c-mpc: make I2C bus speed configurable

Grant Likely grant.likely at secretlab.ca
Thu Apr 2 00:44:59 EST 2009


On Wed, Apr 1, 2009 at 7:13 AM, Wolfgang Grandegger <wg at grandegger.com> wrote:
> This patch makes the I2C bus speed configurable by using the I2C node
> property "clock-frequency". If the property is not defined, the old
> fixed clock settings will be used for backward compatibility.
>
> The generic I2C clock properties, especially the CPU-specific source
> clock pre-scaler are defined via the OF match table:
>
>  static const struct of_device_id mpc_i2c_of_match[] = {
>        ...
>        {.compatible = "fsl,mpc8543-i2c",
>         .data = &(struct fsl_i2c_match_data) {
>                        .setclock = mpc_i2c_setclock_8xxx,
>                        .prescaler = 2,
>                },
>        },
>
> The "data" field defines the relevant I2C setclock function and the
> relevant prescaler for the I2C source clock frequency.
>
> It uses arch-specific tables and functions to determine resonable
> Freqency Divider Register (fdr) values for MPC83xx, MPC85xx, MPC86xx,
> MPC5200 and MPC5200B.
>
> The i2c->flags field and the corresponding FSL_I2C_DEV_* definitions
> have been removed as they are obsolete.
>
> Furthermore dev_dbg() and dev_info() are now used to profit from a more
> comprehensive output.
>
> Signed-off-by: Wolfgang Grandegger <wg at grandegger.com>

>  arch/powerpc/include/asm/mpc52xx.h             |   11 +
>  arch/powerpc/platforms/52xx/mpc52xx_common.c   |   57 ++++++++++
>  arch/powerpc/sysdev/fsl_soc.c                  |   90 ++++++++++++++++
>  arch/powerpc/sysdev/fsl_soc.h                  |   12 ++
   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

I mostly like this patch.  However, as I mentioned in my other reply,
I'm going to NAK it because the divisor tables are not common code and
do not belong in common code files.

Other than that, I like this version of the patch and I think it is
almost finished.  More comments below.

> -static void mpc_i2c_setclock(struct mpc_i2c *i2c)
> -{
> -       /* Set clock and filters */
> -       if (i2c->flags & FSL_I2C_DEV_SEPARATE_DFSRR) {
> -               writeb(0x31, i2c->base + MPC_I2C_FDR);
> -               writeb(0x10, i2c->base + MPC_I2C_DFSRR);
> -       } else if (i2c->flags & FSL_I2C_DEV_CLOCK_5200)
> -               writeb(0x3f, i2c->base + MPC_I2C_FDR);
> -       else
> -               writel(0x1031, i2c->base + MPC_I2C_FDR);
> +static void mpc_i2c_setclock_52xx(struct device_node *node,
> +                                 struct mpc_i2c *i2c,
> +                                 u32 clock, u32 prescaler)
> +{
> +       int fdr = mpc52xx_i2c_get_fdr(node, clock, prescaler);
> +
> +       if (fdr < 0)
> +               fdr = 0x3f; /* backward compatibility */
> +       writeb(fdr & 0xff, i2c->base + MPC_I2C_FDR);
> +       dev_info(i2c->dev, "clock %d Hz (fdr=%d)\n", clock, fdr);
> +}
> +
> +static void mpc_i2c_setclock_8xxx(struct device_node *node,
> +                                 struct mpc_i2c *i2c,
> +                                 u32 clock, u32 prescaler)
> +{
> +       int fdr = fsl_i2c_get_fdr(node, clock, prescaler);
> +
> +       if (fdr < 0)
> +               fdr = 0x1031; /* backward compatibility */
> +       writeb(fdr & 0xff, i2c->base + MPC_I2C_FDR);
> +       writeb((fdr >> 8) & 0xff, i2c->base + MPC_I2C_DFSRR);
> +       dev_info(i2c->dev, "clock %d Hz (dfsrr=%d fdr=%d)\n",
> +                clock, fdr >> 8, fdr & 0xff);
>  }

Very neat and tidy indeed.  I like this.

> Index: linux-2.6/arch/powerpc/sysdev/fsl_soc.c
> ===================================================================
> --- linux-2.6.orig/arch/powerpc/sysdev/fsl_soc.c        2009-03-31 21:27:52.000000000 +0200
> +++ linux-2.6/arch/powerpc/sysdev/fsl_soc.c     2009-04-01 12:09:18.796719327 +0200
> @@ -39,7 +39,7 @@
>  #include <sysdev/fsl_soc.h>
>  #include <mm/mmu_decl.h>
>  #include <asm/cpm2.h>
> -
> +#define DEBUG
  ^^^^^^^^^^^^^^^^^^^^^^^  oops?

g.

-- 
Grant Likely, B.Sc., P.Eng.
Secret Lab Technologies Ltd.



More information about the devicetree-discuss mailing list