RE: [PATCH 1/2] arm64: dts: renesas: r9a07g044: Fix external clk node names

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

 



Hi Krzysztof Kozlowski,

Thanks for feedback.


> Subject: Re: [PATCH 1/2] arm64: dts: renesas: r9a07g044: Fix external clk
> node names
> 
> On 25/04/2022 17:28, Biju Das wrote:
> >>>> My suggestion was to move the [12] part into the first part, so the
> >>>> suffix "clk" stays consistent:
> >>>> audio1-clk
> >>>> audio2-clk
> >>>
> >>> From HW perspective,  there are 2 audio clocks, audio
> >>> clock1(multiple and sub multiple of 44.1 Khz) and audio clk
> >>> 2(Multiple and submultiple
> >> of 48Khz) connected to a single audio Codec.
> >>>
> >>> Based on the sampling rate, through clock generator driver we can
> >>> switch the clock source for audio mclock along with audio clock for
> >>> SSI and we can support both these rates
> >>>
> >>> Since there is a single audio codec, I am not sure, audio1-clk and
> >> audio2-clk is a good choise.
> >>
> >> The name of the clock is not "audio clock" but "audio", because you
> >> do not call a car "Ford Mustang car", but just "Ford Mustang".
> Therefore "clock"
> >> is not part of the name, but just description of a type.
> >
> > The hardware mention the name as AUDIO_CLK1 and AUDIO_CLK2.
> 
> The hardware document might call it "AUDIO_CLK_REAL_CLK_CLK" and it won't
> be an argument to call device node that way in DTS.

Got it.

> > > There are 2 Clock availables for audio interface.
> > In that case if you term it as audio1-clk and audio2-clk it may not be correct, But as you
> > said clk-1-audio and clk-2-audio will it be correct?
> 
> If you change all other clocks to follow same principle - generic name
> followed by specific suffix - then yes. Then you should have "clk-extal",
> "clk-can" etc.

Ok.

> 
> >
> >>
> >>>
> >>> What about like
> >>>
> >>> audio_clk1: audio-clk-1 ?
> >>> audio_clk2: audio-clk-2 ?
> >>>
> >>> Which is consistent with naming used for cpu and opp-tables?
> >>
> >>
> >> It's not consistent with clk naming. Nodes should have generic names,
> >> so the generic part is "clk". You add specific audio/audio-X prefix
> >> or suffix
> >> - it's fine, but not both.
> >>
> >> This is exactly the trouble when you start using specific names and
> >> Devicetree spec explicitly asks for generic names. So maybe go with
> >> the spec and call of these "clk-[0-9]" and problem is gone.
> >
> > Ok Will change like
> >
> > "audio_clk1: clk-1-audio"
> >
> 
> What do you mean "ok"? I said "clk-[0-9]", so "clk-0", "clk-1", "clk-2"
> and so on. No specific prefix.

Ah ok.

As you mentioned above "generic part is "clk". You add specific audio/audio-X prefix
or suffix"

So based on that, I thought "clk" is generic part and "-1-audio" as suffix, "clk-1-audio" 
should be acceptable.

For eg:- If I plan to match the label name with the hardware manual(audio_clk1),

then, as per the discussion we have, the node names should be

either

"audio_clk1: clk-0"

or 

"audio_clk1: clk-1"

Or

"audio_clk1: audio1-clk"

Correct?

Regards,
Biju
> 
> > Label name matches with hardware manual and node names as per Device
> tree spec.
> 
> 
> Best regards,
> Krzysztof




[Index of Archives]     [Linux Samsung SOC]     [Linux Wireless]     [Linux Kernel]     [ATH6KL]     [Linux Bluetooth]     [Linux Netdev]     [Kernel Newbies]     [IDE]     [Security]     [Git]     [Netfilter]     [Bugtraq]     [Yosemite News]     [MIPS Linux]     [ARM Linux]     [Linux Security]     [Linux RAID]     [Linux ATA RAID]     [Samba]     [Device Mapper]

  Powered by Linux