Wolfram Sang wrote:
>>>>
>>>> -static void __devinit mpc_i2c_setclock_52xx(struct device_node *node,
>>>> - struct mpc_i2c *i2c,
>>>> - u32 clock, u32 prescaler)
>>>> +static void __devinit mpc_i2c_setup_52xx(struct device_node *node,
>>>> + struct mpc_i2c *i2c,
>>>> + u32 clock, u32 prescaler)
>>>> {
>>>> int ret, fdr;
>>>>
>>>> + if (clock == -1) {
>>> Could we use 0 for 'no_clock'? This would make the above statement simply
>> "0" is already used to maintain backward compatibility setting a safe
>> divider.
>
> Ah, now I see:
>
> 'clock == -1' means 'preserve clocks' (and is checked here in
> mpc_i2c_setup_52xx())
Yes, this is now necessary because "setup" does not just do clock settings.
> 'clock == 0' means 'safe divider' (and is checked in mpc_i2c_get_fdr_52xx())
This is for compatibility with old DTS files and last time it was tricky
to get that right and therefore...
> This is not a beauty ;)
>
> What about adding a flags variable to the setup-functions?
.. I hesitate to make bigger changes to the code flow, which the
introduction of a flags variable would required. Also it seems to be
overkill to me. I will have a closer look, though. At a minimum I will
replace "-1" with "MPC_I2C_PRESERVE_CLOCK".
Wolfgang.
_______________________________________________
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev
