Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Date:   Wed, 11 May 2022 16:09:16 +0200
Message-ID: <s5hv8ucf9f7.wl-tiwai@suse.de>
From:   Takashi Iwai <tiwai@suse.de>
To:     Werner Sembach <wse@tuxedocomputers.com>
Cc:     alsa-devel@alsa-project.org, kai.vehmanen@linux.intel.com,
        linux-kernel@vger.kernel.org, imre.deak@intel.com,
        pierre-louis.bossart@linux.intel.com, tiwai@suse.com
Subject: Re: [PATCH] ALSA: hda/intel: Add quirk for TongFang devices with pop noise
In-Reply-To: <71fadca3-7750-2fc0-c04a-f05ab4e85112@tuxedocomputers.com>
References: <20220511133828.13724-1-wse@tuxedocomputers.com>
        <s5h35hggpao.wl-tiwai@suse.de>
        <71fadca3-7750-2fc0-c04a-f05ab4e85112@tuxedocomputers.com>
User-Agent: Wanderlust/2.15.9 (Almost Unreal) SEMI/1.14.6 (Maruoka)
 FLIM/1.14.9 (=?UTF-8?B?R29qxY0=?=) APEL/10.8 Emacs/25.3
 (x86_64-suse-linux-gnu) MULE/6.0 (HANACHIRUSATO)
MIME-Version: 1.0 (generated by SEMI 1.14.6 - "Maruoka")
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Precedence: bulk

On Wed, 11 May 2022 15:58:03 +0200,
Werner Sembach wrote:
> 
> Am 11.05.22 um 15:41 schrieb Takashi Iwai:
> > On Wed, 11 May 2022 15:38:28 +0200,
> > Werner Sembach wrote:
> >> When audio stops playing and sometimes when it starts playing, there is an
> >> audible "pop" noise when using headphones on most Tongfang GMxMxxx,
> >> GKxNxxx, GMxZxxx, GMxTxxx, and GMxAxxx devices.
> >>
> >> Disabling power saving for the Realtek codec fixes this noise. Presumably
> >> it is triggered on some power event in the audio circuit.
> >>
> >> Signed-off-by: Werner Sembach <wse@tuxedocomputers.com>
> >> Cc: stable@vger.kernel.org
> > Usually this denylist is a last resort if any other methods never
> > helped.  Is it the case?
> 
> I also tried setting codec->power_save_node = 0; in a patch_realtek.c quirk.
> 
> static void alc274_fixup_pop_noise(struct hda_codec *codec, const struct
> hda_fixup *fix, int action)
> {
>        codec->power_save_node = 0;
> }
> 
> That initially seemed to work, but when I tested it again a day later
> after a fresh install, it didn't anymore. I don't know what is
> different. On the install before, I did some debugging with boot
> parameters and other stuff, cant reproduce anymore what exactly.

power_save_node is already 0 as default for patch_alc269(), so it must
be irrelevant.

> I took the line from alc274_fixup_bind_dacs which fixes the pop noise
> when applied to the devices, but does a lot of random other stuff too
> not meant for the device.

It's only one thing, there are tons of different fixes :)
alc274_fixup_bind_dacs() rather specifies the routing so that the
speaker is connected to the preferred DAC.

Many click noises come from the default pin shut-up behavior.  You
can disable it or change it in other way.  Also, setting
auto_mute_via_amp may influence on such behavior (that is included in
alc274_fixup_bind_dacs()).

Note that many quirks can be enabled even without compiling but via
the codec patch loading (using hints).


thanks,

Takashi