[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <34c8c9d5-fad0-52ce-d1e6-798546005bfd@starfivetech.com>
Date: Thu, 3 Aug 2023 16:37:50 +0800
From: Hal Feng <hal.feng@...rfivetech.com>
To: Mark Brown <broonie@...nel.org>
CC: Liam Girdwood <lgirdwood@...il.com>,
Jaroslav Kysela <perex@...ex.cz>,
Takashi Iwai <tiwai@...e.com>,
Claudiu Beznea <claudiu.beznea@...rochip.com>,
Rob Herring <robh+dt@...nel.org>,
Krzysztof Kozlowski <krzysztof.kozlowski+dt@...aro.org>,
Conor Dooley <conor+dt@...nel.org>,
Walker Chen <walker.chen@...rfivetech.com>,
Xingyu Wu <xingyu.wu@...rfivetech.com>,
Emil Renner Berthing <emil.renner.berthing@...onical.com>,
<alsa-devel@...a-project.org>, <devicetree@...r.kernel.org>,
<linux-riscv@...ts.infradead.org>, <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH v2 2/3] ASoC: starfive: Add JH7110 PWM-DAC driver
On Mon, 31 Jul 2023 20:14:38 +0100, Mark Brown wrote:
> On Mon, Jul 31, 2023 at 11:28:28AM +0800, Hal Feng wrote:
>
>> +static const struct jh7110_ct_pwmdac pwmdac_ct_data_shift[] = {
>> + { .name = "left 0 bit", .vals = PWMDAC_DATA_LEFT_SHIFT_BIT_0 },
>> + { .name = "left 1 bit", .vals = PWMDAC_DATA_LEFT_SHIFT_BIT_1 },
>> + { .name = "left 2 bit", .vals = PWMDAC_DATA_LEFT_SHIFT_BIT_2 },
>> + { .name = "left 3 bit", .vals = PWMDAC_DATA_LEFT_SHIFT_BIT_3 },
>> + { .name = "left 4 bit", .vals = PWMDAC_DATA_LEFT_SHIFT_BIT_4 },
>> + { .name = "left 5 bit", .vals = PWMDAC_DATA_LEFT_SHIFT_BIT_5 },
>> + { .name = "left 6 bit", .vals = PWMDAC_DATA_LEFT_SHIFT_BIT_6 },
>> + { .name = "left 7 bit", .vals = PWMDAC_DATA_LEFT_SHIFT_BIT_7 }
>> +};
>
> It's not clear to me why this is user selectable - what does the control
> do?
It's convenient to change the register values in user space with the
control. But actually using fixed register configuration is enough.
I will drop all these control in the next version.
>
>> +static int jh7110_pwmdac_put(struct snd_kcontrol *kcontrol,
>> + struct snd_ctl_elem_value *ucontrol,
>> + int pwmdac_ct)
>> +{
>> + struct snd_soc_component *component = snd_kcontrol_chip(kcontrol);
>> + struct jh7110_pwmdac_dev *dev = snd_soc_component_get_drvdata(component);
>> + int sel = ucontrol->value.enumerated.item[0];
>> + unsigned int items;
>
>> + if (pwmdac_ct == PWMDAC_CT_SHIFT)
>> + dev->shift = pwmdac_ct_shift[sel].vals;
>> + else if (pwmdac_ct == PWMDAC_CT_DUTY_CYCLE)
>> + dev->duty_cycle = pwmdac_ct_duty_cycle[sel].vals;
>> + else if (pwmdac_ct == PWMDAC_CT_DATA_CHANGE)
>> + dev->data_change = pwmdac_ct_data_change[sel].vals;
>> + else if (pwmdac_ct == PWMDAC_CT_DATA_MODE)
>> + dev->data_mode = pwmdac_ct_data_mode[sel].vals;
>> + else if (pwmdac_ct == PWMDAC_CT_DATA_SHIFT)
>> + dev->data_shift = pwmdac_ct_data_shift[sel].vals;
>> +
>> + return 0;
>> +}
>
> _put() operations should return 1 if the control value changes so event
> generation works - please test a card using this driver with the
> mixer-test selftest, it'll identify this and a bunch of other potential
> issues.
Thanks for the correct, but it doesn't matter as I will drop these control
later.
>
>> +static int jh7110_pwmdac_component_probe(struct snd_soc_component *component)
>> +{
>> + snd_soc_add_component_controls(component, jh7110_pwmdac_snd_controls,
>> + ARRAY_SIZE(jh7110_pwmdac_snd_controls));
>> + return 0;
>> +}
>
> The driver can just point to the controls from the _driver struct and
> skip having a probe() function.
Thanks for your advice, but I would like to drop these control later.
Thanks again for your review.
Best regards,
Hal
Powered by blists - more mailing lists