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;
MIME-Version: 1.0
References: <20221127-snd-freeze-v5-0-4ededeb08ba0@chromium.org> <alpine.DEB.2.22.394.2211281629120.3532114@eliteleevi.tm.intel.com>
In-Reply-To: <alpine.DEB.2.22.394.2211281629120.3532114@eliteleevi.tm.intel.com>
From:   Ricardo Ribalda <ribalda@chromium.org>
Date:   Mon, 28 Nov 2022 18:10:41 +0100
Message-ID: <CANiDSCtd52UX1WQDj4PLTuTogurr63ZdMXVWmYqJHCMo_kEoMQ@mail.gmail.com>
Subject: Re: [PATCH v5] ASoC: SOF: Fix deadlock when shutdown a frozen userspace
To:     Kai Vehmanen <kai.vehmanen@linux.intel.com>
Cc:     Liam Girdwood <lgirdwood@gmail.com>,
        Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>,
        Mark Brown <broonie@kernel.org>,
        Daniel Baluta <daniel.baluta@nxp.com>,
        Peter Ujfalusi <peter.ujfalusi@linux.intel.com>,
        Jaroslav Kysela <perex@perex.cz>,
        Takashi Iwai <tiwai@suse.com>,
        Bard Liao <yung-chuan.liao@linux.intel.com>,
        Ranjani Sridharan <ranjani.sridharan@linux.intel.com>,
        Alsa-devel <alsa-devel@alsa-project.org>,
        linux-kernel@vger.kernel.org, stable@vger.kernel.org,
        sound-open-firmware@alsa-project.org
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

Hi Kay

Thanks for your review

On Mon, 28 Nov 2022 at 15:41, Kai Vehmanen <kai.vehmanen@linux.intel.com> wrote:
>
> Hi,
>
> On Mon, 28 Nov 2022, Ricardo Ribalda wrote:
>
> > During kexec(), the userspace is frozen. Therefore we cannot wait for it
> > to complete.
> >
> > Avoid running snd_sof_machine_unregister during shutdown.
> [...]
> >       /*
> > -      * make sure clients and machine driver(s) are unregistered to force
> > -      * all userspace devices to be closed prior to the DSP shutdown sequence
> > +      * make sure clients are unregistered prior to the DSP shutdown
> > +      * sequence.
> >        */
> >       sof_unregister_clients(sdev);
> >
> > -     snd_sof_machine_unregister(sdev, pdata);
> > -
> >       if (sdev->fw_state == SOF_FW_BOOT_COMPLETE)
>
> this is problematic as removing that machine_unregister() call will (at
> least) bring back an issue on Intel platforms (rare problem hitting S5 on
> Chromebooks).

Do you know which devices were affected or how to trigger the issue?

I have access to the ChromeOS lab, so I can test on a big variety of devices

Thanks!


>
> Not sure how to solve this, but if that call needs to be removed
> (unsafe to call at shutdown), then we need to rework how SOF
> does the cleanup.
>
> Br, Kai


-- 
Ricardo Ribalda