On 04/03/2021 1:49 pm, Takashi Iwai wrote:

On Wed, 03 Mar 2021 19:29:59 +0100, Vitaly Rodionov wrote:

...

+static int cs8409_cs42l42_volume_get(struct snd_kcontrol *kcontrol,

• 		 struct snd_ctl_elem_value *ucontrol)
  

+{

• struct hda_codec *codec = snd_kcontrol_chip(kcontrol);
• hda_nid_t nid = get_amp_nid(kcontrol);
• int chs = get_amp_channels(kcontrol);
• long *valp = ucontrol->value.integer.value;
• char vol = 0;
• codec_dbg(codec, "%s() nid: %d\n", __func__, nid);
• snd_hda_power_up(codec);
• switch (nid) {
• case CS8409_CS42L42_HP_PIN_NID:

• mutex_lock(&cs8409_i2c_mux);
  

• if (chs & 1) {
  

• 	vol = -(cs8409_i2c_read(codec, CS42L42_I2C_ADDR,
  

• 		CS8409_CS42L42_REG_HS_VOLUME_CHA, 1));
  

Better to cache the values instead of i2c read at each time? Then the unnecessary power up/down sequence can be avoided, too.

Yes, agree. Will be fixed in next version of patch.

thanks,

Takashi

Thank you,

Vitaly