Re: [PATCH] Fixed the schema binding according to test

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



> > > > +
> > > > +  i2c-retry:
> > > > +    description: number of retries for I2C regmap.
> > >
> > > Why do you need this? Why this fits the purpose of DT (or IOW why
> > > this differs between boards)?
> >
> > When working with drivers on mulitiple platforms, there were cases
> > where I2C did not work properly dpending on the AP or setting.
> > So I made it possible to set a few retry settings, and then check or
> > do other actions. Retry is performed only when I2C fails.
> >
> > And each device may have a different pull-up resisor or strength, so
> > there may be differences between boards.
> 
> None of I2C drivers need it (except SBS battery), so it should not be
> needed here. If you have wrong pin setup, this one should be corrected
> instead.

Okay, I agree. It doesn't seem necessary.
I'll remove this property. Thanks!

> > > > +  tdm-slot-rx:
> > > > +    description: set the tdm rx start slot.
> > >
> > > Aren't you now re-writing dai-tdm-slot-rx-mask property? Same for tx
> below.
> > >
> >
> > It can be the same as audio DAI's tdm slot, I think but there are
> > cases where it is set differently, so I omitted it separately.
> Unfortunately I still do not understand why do you need it. Use generic
> DAI/TDM properties.

Looking back, it seems to be the same as the dai-tdm-slot-rx-mask.
It seems to be correct to map the amp function in the driver.

However, I looked for the ones used in other amplifier drivers,
Most of them were added to the user control using mixer.

Come to think of it, there are many cases where the TDM slot is
changed after the driver is connected.

> > > > +  sys-clk-id:
> > > > +    description: select the using system clock.
> > >
> > > What does it mean? Why do you need such property instead of clocks?
> >
> > This can receive an external clock, but it can use internal clock.
> > Should I write all the clock descriptions in case?
> 
> How do you configure and enable external clock with this property? I don't
> see it. If the device has clock input, this should be "clocks". If it is
> omitted, then internal clock is used.
> 

Basically, this value is set with set_sysclk in the dai operations.
So, I also get the clk_id from this function and set it.
>From the point of view of the codec driver, there are case where the machine
driver does not give this value(clk_id).

So this is a property that can set an initial value.
It's not requied value.

All three properties mentioned above are optional.
So, if it need to be removed, it's okay.
But I think it seems correct to register the second item(tdm slot) as a mixer.

Thank you for your good feeback!

Best regards,
Kiseok Jo




[Index of Archives]     [Device Tree Compilter]     [Device Tree Spec]     [Linux Driver Backports]     [Video for Linux]     [Linux USB Devel]     [Linux PCI Devel]     [Linux Audio Users]     [Linux Kernel]     [Linux SCSI]     [XFree86]     [Yosemite Backpacking]


  Powered by Linux