Date:   Wed, 12 Jan 2022 11:13:44 +0100
Message-ID: <s5hilup9s87.wl-tiwai@suse.de>
From:   Takashi Iwai <tiwai@suse.de>
To:     Alexander Sergeyev <sergeev917@gmail.com>
Cc:     Jeremy Szu <jeremy.szu@canonical.com>, tiwai@suse.com,
        "moderated list:SOUND" <alsa-devel@alsa-project.org>,
        Kailang Yang <kailang@realtek.com>,
        open list <linux-kernel@vger.kernel.org>,
        Huacai Chen <chenhuacai@kernel.org>,
        Jian-Hong Pan <jhp@endlessos.org>,
        Hui Wang <hui.wang@canonical.com>,
        PeiSen Hou <pshou@realtek.com>
Subject: Re: [PATCH 1/4] ALSA: hda/realtek: fix mute/micmute LEDs for HP 855 G8
In-Reply-To: <20220112101249.ya73jvpmqmeh4ggg@localhost.localdomain>
References: <20210519170357.58410-1-jeremy.szu@canonical.com>
        <20220111195229.a77wrpjclqwrx4bx@localhost.localdomain>
        <s5ho84h9tit.wl-tiwai@suse.de>
        <20220112101249.ya73jvpmqmeh4ggg@localhost.localdomain>
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=US-ASCII
Precedence: bulk

On Wed, 12 Jan 2022 11:12:49 +0100,
Alexander Sergeyev wrote:
> 
> On Wed, Jan 12, 2022 at 10:45:46AM +0100, Takashi Iwai wrote:
> > The problem is about the built-in drivers, or do you see the very
> > same problem even with modules?
> 
> The problem is definitely there for the built-in drivers which I've
> tested quite a lot. It's the primary usecase for me, as I tend to
> build minimal device-specific and self-contained kernels in Gentoo.
> 
> For builds with modules things are not very consistent. Live Ubuntu
> with an older (and probably vendor-patched) kernel works just fine,
> but when I pull Ubuntu kernel sources and build it with the mostly
> same config (including modules) it boots with no sound in
> Gentoo. Mostly same -- because I need nvme drivers to be built-in as I
> don't use initrd.

Sounds like some timing issue, then.  It's pretty hard to debug,
unfortunately.

You may try to get the codec proc dump with COEF by passing
snd_hda_codec.dump_coef=1 module option for both working and
non-working cases.  Check the difference of the COEF and apply the
difference with hda-verb manually.


> > AFAIK, quite a few AMD platforms tend to have some issues with
> > various devices showing initialization problems at the early
> > boot. Just reloading / rebinding the device later often helps.
> 
> Is it possible to do with the built-in drivers?

You can unbind and re-bind the PCI (HD-audio controller) device via
sysfs.


Takashi