[alsa-devel] sun4i-i2s: Wrong quirks for Allwinner A64 ?

[Clément Péron]

Hi Vasily,

On Mon, 22 Apr 2019 at 04:35, Vasily Khoruzhick <anarsoul at gmail.com> wrote:
>
> On Sun, Apr 21, 2019 at 5:35 AM Clément Péron <peron.clem at gmail.com> wrote:
> >
> > Hi,
> >
> > I'm looking to port the sun4i-i2s drivers for Allwinner H6, but
> > regarding the quirks I found strange that H3 and A64 was so different.
> >
> > I have analyzed the datasheet and using human reading comparison, I
> > have found only 3 minor differences in the registers.
> >
> > H3 reference i have used : H3 Datasheet(Revision 1.0)
> > For A64 : A64 User Manual(Revision 1.1)
> >
> > H3 is identical to A64 except H3 has new :
> > LRCKR_OUT bit(16) in PCM_CONTROL_REG
> > SDI_sync_sel Bit(31) in PCM_FMT0_REG
> > LRCKR_PERIOD Mask(29:20) in PCM_FMT0_REG
> >
> > Regarding the quirks struct in the drivers and the commit log
> > sha1: 66ecce3325383c8304063e7d5a30f4374ef5a33e
> > ASoC: sun4i-i2s: Add compatibility with A64 codec I2S
>
> Note that it mentions *codec* I2S. A64 has 4 I2S, 3 are pretty much
> like H3, but the other (one that is used with built-in audio codec) is
> totally different.

Indeed... Not sure why my brain skip this word when I read the commit log.
But its well explained and obvious when looking at the device-tree.

Thanks for the reply,
Clement

>
> See 3.20.4 in A64 user manual, regs 0x00-0x3c are codec I2S.
>
> >
> >     The I2S block used for the audio codec in the A64 differs from other 3
> >     I2S modules in A64 and isn't compatible with H3. But it is very similar
> >     to what is found in A10(sun4i). However, its TX FIFO is
> >     located at a different address.
> >
> > It said that A64 is really different which I disagree, for me the H3
> > use the same IP as A64 with a little improvement.
> > Which seems logical because "The A64 is basically an Allwinner H3 with
> > the Cortex-A7 cores replaced with Cortex-A53 cores" (source
> > linux-sunxi.org)
> >
> > I'm okay to send a patch to fix that but I don't have an H3 nor A64 to
> > test the modification.
> >
> > Thanks,
> > Clement