On Tue, 11 Aug 2020 13:55:25 +0200, Hui Wang wrote:
On 2020/8/11 下午5:22, Takashi Iwai wrote:
On Tue, 11 Aug 2020 11:03:55 +0200, Hui Wang wrote:
On 2020/8/11 下午4:56, Kai-Heng Feng wrote:
Hi,
On Aug 10, 2020, at 14:51, Takashi Iwai tiwai@suse.de wrote:
On Mon, 10 Aug 2020 08:34:36 +0200, Hui Wang wrote:
On 2020/8/10 下午2:30, Takashi Iwai wrote: > On Mon, 10 Aug 2020 04:16:59 +0200, > Hui Wang wrote: >> After installing the Ubuntu Linux, the micmute led status is not >> correct. Users expect that the led is on if the capture is disabled, >> but with the current kernel, the led is off with the capture disabled. >> >> We tried the old linux kernel like linux-4.15, there is no this issue. >> It looks like we introduced this issue when switching to the led_cdev. > The behavior can be controlled via "Mic Mute-LED Mode" enum kcontrol. > Which value does it have now? If it's "Follow Capture", that's the > correct behavior. OTOH, if it's "Follow Mute", the LED polarity is > indeed wrong. It is "Follow Mute", if I change it to "Follow Capture" manually, the led status becomes correct.
OK, thanks for confirmation. Applied now.
I wonder if it's because how brightness value passed to gpio_mute_led_set() and micmute_led_set():
static int gpio_mute_led_set(struct led_classdev *led_cdev, enum led_brightness brightness) { struct hda_codec *codec = dev_to_hda_codec(led_cdev->dev->parent); struct alc_spec *spec = codec->spec;
alc_update_gpio_led(codec, spec->gpio_mute_led_mask, spec->mute_led_polarity, !brightness); return 0;}
static int micmute_led_set(struct led_classdev *led_cdev, enum led_brightness brightness) { struct hda_codec *codec = dev_to_hda_codec(led_cdev->dev->parent); struct alc_spec *spec = codec->spec;
alc_update_gpio_led(codec, spec->gpio_mic_led_mask, spec->micmute_led_polarity, !!brightness); return 0;}
Maybe we should let micmute_led_set() match gpio_mute_led_set() so the micmute polarity can be removed?
This will impact many many machines, I don't know if the current code could work correctly or not on other machines.
True. But we should rather fix this, the current flag is illogical. I forgot about this problem while I also noticed during working on the led cdev conversion.
I guess most of configurations can be verified with hda-emu or such. Let's see...
I just checked, the commit 87dc36482cab (ALSA: hda/realtek - Add LED class support for micmute LED) introduced this issue. In the micmute_led_set(), it set the led with the !!mic_mute.led_value, while before this commit, the driver sets the led with !mic_mute.led_value.
Add with this commit 7cdf8c49b1df (ALSA: hda: generic: Add a helper for mic-mute LED with LED classdev), all micmute led updating will call micmute_led_set(). It impacts alc269_fixup_hp_gpio_led(), alc285_fixup_hp_gpio_led(), alc286_fixup_hp_gpio_led(), alc280_fixup_hp_gpio2_mic_hotkey() and alc233_fixup_lenovo_line2_mic_hotkey()
I think it is safe to change the micmute_led_set() and remove all spec->micmute_led_polarity = 1.
Let me write a patch.
Great, thanks for spotting out.
Takashi