Re: [PATCH v3 4/6] ASoC: mediatek: mt7986: add machine driver with wm8960

["Maso Huang (黃加竹)" <Maso.Huang@mediatek.com>]

On Fri, 2023-07-28 at 11:52 +0200, AngeloGioacchino Del Regno wrote:
> Il 28/07/23 11:49, AngeloGioacchino Del Regno ha scritto:
> > Il 28/07/23 11:08, Maso Huang ha scritto:
> > > Add support for mt7986 board with wm8960.
> > > 
> > > Signed-off-by: Maso Huang <maso.huang@mediatek.com>
> > > ---
> > >   sound/soc/mediatek/Kconfig                |  10 ++
> > >   sound/soc/mediatek/mt7986/Makefile        |   1 +
> > >   sound/soc/mediatek/mt7986/mt7986-wm8960.c | 184
> > > ++++++++++++++++++++++
> > >   3 files changed, 195 insertions(+)
> > >   create mode 100644 sound/soc/mediatek/mt7986/mt7986-wm8960.c
> > > 
> > > diff --git a/sound/soc/mediatek/Kconfig
> > > b/sound/soc/mediatek/Kconfig
> > > index 558827755a8d..8d1bc8814486 100644
> > > --- a/sound/soc/mediatek/Kconfig
> > > +++ b/sound/soc/mediatek/Kconfig
> > > @@ -64,6 +64,16 @@ config SND_SOC_MT7986
> > >         Select Y if you have such device.
> > >         If unsure select "N".
> > > +config SND_SOC_MT7986_WM8960
> > > +    tristate "ASoc Audio driver for MT7986 with WM8960 codec"
> > > +    depends on SND_SOC_MT7986 && I2C
> > > +    select SND_SOC_WM8960
> > > +    help
> > > +      This adds support for ASoC machine driver for MediaTek
> > > MT7986
> > > +      boards with the WM8960 codecs.
> > > +      Select Y if you have such device.
> > > +      If unsure select "N".
> > > +
> > >   config SND_SOC_MT8173
> > >       tristate "ASoC support for Mediatek MT8173 chip"
> > >       depends on ARCH_MEDIATEK
> > > diff --git a/sound/soc/mediatek/mt7986/Makefile
> > > b/sound/soc/mediatek/mt7986/Makefile
> > > index de0742a67cae..fc4c82559b29 100644
> > > --- a/sound/soc/mediatek/mt7986/Makefile
> > > +++ b/sound/soc/mediatek/mt7986/Makefile
> > > @@ -6,3 +6,4 @@ snd-soc-mt7986-afe-objs := \
> > >       mt7986-dai-etdm.o
> > >   obj-$(CONFIG_SND_SOC_MT7986) += snd-soc-mt7986-afe.o
> > > +obj-$(CONFIG_SND_SOC_MT7986_WM8960) += mt7986-wm8960.o
> > > diff --git a/sound/soc/mediatek/mt7986/mt7986-wm8960.c 
> > > b/sound/soc/mediatek/mt7986/mt7986-wm8960.c
> > > new file mode 100644
> > > index 000000000000..a880fcb8662e
> > > --- /dev/null
> > > +++ b/sound/soc/mediatek/mt7986/mt7986-wm8960.c
> > 
> > ..snip..
> > 
> > > +static int mt7986_wm8960_machine_probe(struct platform_device
> > > *pdev)
> > > +{
> > > +    struct snd_soc_card *card = &mt7986_wm8960_card;
> > > +    struct snd_soc_dai_link *dai_link;
> > > +    struct mt7986_wm8960_priv *priv;
> > > +    int ret, i;
> > > +
> > > +    priv = devm_kzalloc(&pdev->dev, sizeof(*priv), GFP_KERNEL);
> > > +    if (!priv)
> > > +        return -ENOMEM;
> > > +
> > > +    priv->platform_node = of_parse_phandle(pdev->dev.of_node,
> > > +                           "mediatek,platform", 0);
> > > +    if (!priv->platform_node) {
> > > +        dev_err(&pdev->dev, "Property 'platform' missing or
> > > invalid\n");
> > > +        return -EINVAL;
> > > +    }
> > > +
> > > +    for_each_card_prelinks(card, i, dai_link) {
> > > +        if (dai_link->platforms->name)
> > > +            continue;
> > > +        dai_link->platforms->of_node = priv->platform_node;
> > > +    }
> > > +
> > > +    card->dev = &pdev->dev;
> > > +
> > > +    priv->codec_node = of_parse_phandle(pdev->dev.of_node,
> > > +                        "mediatek,audio-codec", 0);
> > > +    if (!priv->codec_node) {
> > > +        dev_err(&pdev->dev,
> > > +            "Property 'audio-codec' missing or invalid\n");
> > > +        of_node_put(priv->platform_node);
> > > +        return -EINVAL;
> > > +    }
> > > +
> > > +    for_each_card_prelinks(card, i, dai_link) {
> > > +        if (dai_link->codecs->name)
> > > +            continue;
> > > +        dai_link->codecs->of_node = priv->codec_node;
> > > +    }
> > > +
> > > +    ret = snd_soc_of_parse_audio_routing(card, "audio-routing");
> > > +    if (ret) {
> > > +        dev_err(&pdev->dev, "failed to parse audio-routing:
> > > %d\n", ret);
> > > +        goto err_of_node_put;
> > > +    }
> > > +
> > > +    ret = devm_snd_soc_register_card(&pdev->dev, card);
> > > +    if (ret) {
> > > +        dev_err(&pdev->dev, "%s snd_soc_register_card fail
> > > %d\n",
> > > +            __func__, ret);
> > > +        goto err_of_node_put;
> > > +    }
> > > +
> > > +err_of_node_put:
> > > +    of_node_put(priv->codec_node);
> > > +    of_node_put(priv->platform_node);
> > > +    return ret;
> > > +}
> > > +
> > > +static void mt7986_wm8960_machine_remove(struct platform_device
> > > *pdev)
> > > +{
> > > +    struct snd_soc_card *card = platform_get_drvdata(pdev);
> > > +    struct mt7986_wm8960_priv *priv =
> > > snd_soc_card_get_drvdata(card);
> > > +
> > > +    of_node_put(priv->codec_node);
> > > +    of_node_put(priv->platform_node);
> > > +}
> > > +
> > > +#ifdef CONFIG_OF
> > 
> > Your probe function *relies on* devicetree, and you're adding an
> > ifdef for
> > CONFIG_OF? That wouldn't make sense, would it? ;-)
> > 

Hi Angelo,

Thanks for your review.

It seems we don't need CONFIG_OF here, so I'll remove it in v4 patch.


> > > +static const struct of_device_id
> > > mt7986_wm8960_machine_dt_match[] = {
> > > +    {.compatible = "mediatek,mt7986-wm8960-machine",},
> > 
> > please...
> > 
> > { .compatible = "mediatek,mt7986-wm8960-machine" },
> 
> Actually, I noticed that just after sending the review.
> 
> Can you also please change the compatible, as "machine" doesn't
> really
> mean "sound"? :-)
> 
> .compatible = "mediatek,mt7986-wm8960-sound"
> 
> Thanks!
> 

For compatible suffix, I'll discuss with you in PATCH v3[5/6] mail, and
will refine this with [5/6] mail conclusion :)


> > 
> > > +    { /* sentinel */ }
> > > +};
> > > +MODULE_DEVICE_TABLE(of, mt7986_wm8960_machine_dt_match);
> > > +#endif
> > > +
> > > +static struct platform_driver mt7986_wm8960_machine = {
> > > +    .driver = {
> > > +        .name = "mt7986-wm8960",
> > > +#ifdef CONFIG_OF
> > 
> > Check `struct device_driver`: const struct of_device_id
> > *of_match_table is
> > always present, there's no ifdef.... and that's done in order to
> > avoid seeing
> > a bunch of ifdefs in drivers...
> > 
> > ...so, why is this callback enclosed in an ifdef?
> > 
> > Please drop all those ifdefs.
> > 
> > 
> > After addressing those last comments, you can get my
> > 
> > Reviewed-by: AngeloGioacchino Del Regno <
> > angelogioacchino.delregno@collabora.com>
> > 
> > Regards,
> > Angelo
> 
> 

No more needed for the ifdef, so I'll drop them in v4 patch :)

Best regards,
Maso