Date:   Thu, 18 Nov 2021 09:01:38 +0100
Message-ID: <s5hpmqxki3h.wl-tiwai@suse.de>
From:   Takashi Iwai <tiwai@suse.de>
To:     "Kirill A. Shutemov" <kirill@shutemov.name>
Cc:     Linus Torvalds <torvalds@linux-foundation.org>,
        Mark Brown <broonie@kernel.org>,
        Liam Girdwood <lgirdwood@gmail.com>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>
Subject: Re: [GIT PULL] sound updates for 5.16-rc1
In-Reply-To: <s5hfsrulkxk.wl-tiwai@suse.de>
References: <s5hwnlpmt9x.wl-tiwai@suse.de>
        <20211117161855.m45mxcqszkfcetai@box.shutemov.name>
        <s5hmtm2lphf.wl-tiwai@suse.de>
        <20211117174826.23eakoivl33tawb2@box.shutemov.name>
        <s5hfsrulkxk.wl-tiwai@suse.de>
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, 17 Nov 2021 19:02:47 +0100,
Takashi Iwai wrote:
> 
> On Wed, 17 Nov 2021 18:48:26 +0100,
> Kirill A. Shutemov wrote:
> > 
> > On Wed, Nov 17, 2021 at 05:24:28PM +0100, Takashi Iwai wrote:
> > > On Wed, 17 Nov 2021 17:18:55 +0100,
> > > Kirill A. Shutemov wrote:
> > > > 
> > > > On Wed, Nov 03, 2021 at 11:24:10AM +0100, Takashi Iwai wrote:
> > > > > Linus,
> > > > > 
> > > > > please pull sound updates for v5.16-rc1 from:
> > > > > 
> > > > >   git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound.git tags/sound-5.16-rc1
> > > > > 
> > > > > The topmost commit is df0380b9539b04c1ae8854a984098da06d5f1e67
> > > > > 
> > > > > ----------------------------------------------------------------
> > > > > 
> > > > > sound updates for 5.16-rc1
> > > > > 
> > > > > Lots of code development have been see in ASoC side as usual, while
> > > > > the continued development on memalloc helper and USB-audio low-
> > > > > latency support are found in the rest.  Note that a few changes in the
> > > > > unusual places like arch/sh are included, which are a part of ASoC DAI
> > > > > format cleanups.
> > > > > 
> > > > > ALSA core:
> > > > > - Continued memallloc helper updates and cleanups, now supporting
> > > > >   non-coherent and non-contiguous pages
> > > > > - Fixes for races in mixer OSS layer
> > > > > 
> > > > > ASoC:
> > > > > - A new version of the audio graph card which supports a wider range
> > > > >   of systems
> > > > > - Several conversions to YAML DT bindings
> > > > > - Continuing cleanups to the SOF and Intel code
> > > > > - Move of the Cirrus DSP framework into drivers/firmware to allow for
> > > > >   future use by non-audio DSPs
> > > > > - An overhaul of the cs42l42 driver, correcting many problems
> > > > > - DAI format terminology conversions over many drivers for cleanups
> > > > > - Support for AMD Vangogh and Yelow Cap, Cirrus CS35L41, Maxim
> > > > >   MAX98520 and MAX98360A, Mediatek MT8195, Nuvoton NAU8821, nVidia
> > > > >   Tegra210, NXP i.MX8ULP, Qualcomm AudioReach, Realtek ALC5682I-VS,
> > > > >   RT5682S, and RT9120 and Rockchip RV1126 and RK3568
> > > > > 
> > > > > USB-audio:
> > > > > - Continued improvements on low-latency playback
> > > > > - Quirks for Pioneer devices, Line6 HX-Stomp XL, Audient iD14
> > > > > 
> > > > > HD-audio:
> > > > > - Reduce excessive udelay() calls on Intel platforms; this should
> > > > >   reduce the CPU load with PulseAudio
> > > > > - Quirks for HP and Clevo laptops
> > > > > 
> > > > > FireWire:
> > > > > - Support for meter information on MOTU
> > > > > 
> > > > 
> > > > ...
> > > > 
> > > > > Takashi Iwai (29):
> > > > >       ALSA: hda: Reduce udelay() at SKL+ position reporting
> > > > >       ALSA: hda: Use position buffer for SKL+ again
> > > > >       ALSA: usb-audio: Restrict rates for the shared clocks
> > > > >       ALSA: usb-audio: Fix possible race at sync of urb completions
> > > > >       ALSA: usb-audio: Rename early_playback_start flag with lowlatency_playback
> > > > >       ALSA: usb-audio: Disable low-latency playback for free-wheel mode
> > > > >       ALSA: usb-audio: Disable low-latency mode for implicit feedback sync
> > > > >       ALSA: usb-audio: Check available frames for the next packet size
> > > > >       ALSA: usb-audio: Add spinlock to stop_urbs()
> > > > >       ALSA: usb-audio: Improved lowlatency playback support
> > > > 
> > > > This commit breaks audio playback for me. The sample plays for fraction of
> > > > a second than stops.
> > > > 
> > > > I use SMSL SA300. Let me know what ifo is needed to track it down.
> > > > 
> > > > Also I can test potential fixups if needed.
> > > 
> > > Could you give alsa-info.sh output with the device?  Run the script
> > > with --no-upload option, and attach the output.
> > 
> > Attached. I run it on kernel without the commit. I hope it is fine.
> > 
> > One note: I use monolitic kernel without CONFIG_MODULES. The script
> > complained about missing /proc/asound/modules and other modules related
> > stuff.
> 
> It's a known problem I forgot, but this doesn't matter much.
> 
> > > Also /proc/asound/card*/stream* file for that usb device, too,
> > > please.
> > 
> > Attached.
> 
> Thanks!
> 
> > > And, which environment are you testing, more exactly?  Which platform,
> > > and with PulseAudio, pipewire, JACK, or direct ALSA device use?
> > 
> > Nothing fancy. Direct ALSA. I tested with 'speaker-test -c2 -t wav' during
> > bisect.
> 
> Did you notice the problem with other applications, too?
> 
> When you wrote "The sample plays for fraction of a second than
> stops.", it means that the speaker-test program stalls?  Or it's
> shorter playbacks?
> 
> Also, please test booting with snd_usb_audio.lowlatency=0 boot
> option.  Does it behave like before again?
> 
> > > For checking the behavior, we'd need to check hw_params in
> > > /proc/asound/card*/pcm*/sub*/* for the corresponding device during
> > > playback/capture operations.
> > 
> > You need this to be collected when the problem triggers, right?
> > I will follow up with this.
> 
> I'll need to test my device at first.  Maybe tomorrow.

I checked mine, and all worked fine.  So this doesn't look like a
generic breakage but somehow specific to certain devices or system.

Could you verify the lowlatency flag off?

Also, the simple aplay invocation (e.g. with a 2-channel 16bit sample)
also goes to the same problem, I suppose?


Takashi