Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Wed, 03 Jul 2019 09:49:44 +0200
Message-ID: <s5h36jno8vb.wl-tiwai@suse.de>
From:   Takashi Iwai <tiwai@suse.de>
To:     nick83ola <nick83ola@gmail.com>
Cc:     "Wasko, Michal" <michal.wasko@linux.intel.com>,
        alsa-devel@alsa-project.org, info@jensverwiebe.de,
        Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
        linux-kernel@vger.kernel.org, Jussi Laako <jussi@sonarnerd.net>,
        Richard Fontana <rfontana@redhat.com>,
        Thomas Gleixner <tglx@linutronix.de>,
        Allison Randal <allison@lohutok.net>
Subject: Re: [alsa-devel] [PATCH] ALSA: usb-audio: fix Line6 Helix audio format rates
In-Reply-To: <CABPh3UMttE1s7oNt0-JLNm2N3wxb-JXBTLzMYZM42ENiy9NC9Q@mail.gmail.com>
References: <20190702004439.30131-1-nick83ola@gmail.com>
        <s5hlfxg4i4r.wl-tiwai@suse.de>
        <4181a467-5332-c256-5124-513a0343ec70@linux.intel.com>
        <s5ha7dw4egd.wl-tiwai@suse.de>
        <CABPh3UMttE1s7oNt0-JLNm2N3wxb-JXBTLzMYZM42ENiy9NC9Q@mail.gmail.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
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Wed, 03 Jul 2019 09:41:00 +0200,
nick83ola wrote:
> 
> On Tue 2 Jul 2019, 16:57 Takashi Iwai, <tiwai@suse.de> wrote:
> 
>     On Tue, 02 Jul 2019 17:52:01 +0200,
>     Wasko, Michal wrote:
>     >
>     > On 7/2/2019 4:37 PM, Takashi Iwai wrote:
>     > > On Tue, 02 Jul 2019 02:43:14 +0200,
>     > > Nicola Lunghi wrote:
>     > >> Line6 Helix and HX stomp don't support retrieving
>     > >> the number of clock sample rate.
>     > >>
>     > >> Add a quirk to return the default value of 48Khz.
>     > >>
>     > >> Signed-off-by: Nicola Lunghi <nick83ola@gmail.com>
>     > > It's not particularly good place to put a quirk, but there seems no
>     > > other better place, unfortunately. 
> 
> If you prefer I can add a function to quirk.c

It's OK to place in format.c as a start.

>     Is this specific to certain unit
>     > > or all I/Os on the device suffer from this problem?
> 
> This is specific to the helix line of line6
> There are currently 4 devices in that line that I think shared their firmware
> more or less.
> Unfortunately I have only one device to test here (but maybe someone else can
> add the others)

Ah sorry I wasn't clear enough.  What I meant as "unit" is the USB
descriptor unit (aka widget), a feature unit, a terminal, whatever.
Does this error appear only once at parsing, or does it hit multiple
times per device?

>     > >
>     > > In anyway, if the behavior is expected, we don't need to use
>     > > dev_warn() to annoy users unnecessarily.  Replace it with dev_info().
>     > >
> 
> Ok
> 
>     > > Also, the code that creates a single 48k entry would be better to be
>     > > put into a function for readability.
>     > >
>     > > Could you resubmit with that change?
>     > >
> 
> Yes (in quirks.c or here?)
> 
> Also I suspect that there' s a way to get the clock rates because windows
> driver supports getting the rates and switching (and in Mac the device has the
> same problem without a custom driver only supports 48khz)

It might be some fixed list without inquiries?


>     > >
>     > > Thanks!
>     > >
>     > > Takashi
>     >
> 
> Thank you
> 
>     > If the listed USB devices do not support sample rate format retrieval
>     > then maybe it would be a better idea to perform below check before
>     > sending message?
>    
>     Yes, if we know that it always fails, we don't need to query.
> 
> Sorry I don't understand this :-)

Well, the error happens always on your device in the very same way,
right?  Then we don't have to send the specific command that leads to
an error, but takes the fixed 48k from the beginning.


> My idea is that if line6 in the future fixes their code (they are quite active
> on the helix line) the call will not fail and we get a proper device without
> quirks.
> But If the driver fail to get the clock this settings works as a "failsafe"
> and get the device working.
> I also tried to contact their support but they don't care a lot about Linux
> for now :-(

The fallback might be good, but I'm not sure whether there are many
devices that hit the same problem.  If we encounter more, let's try to
make it as a fallback.


thanks,

Takashi

> 
> I will clean this better and resubmit
> Thank you
> Nick
> 
>     > Have you also considered new function or macro that check device
>     > support? This would separate formatfunctionality code from routine
>     > that identifies applicable devices- in case if in future more devices
>     > will require quirk.
>    
>     The split can be done later.  It's always hard to know what kind of
>     quirk would be needed in future.  If any more devices show the same
>     problem, we can reorganize the quirk in a saner way.
> 
>     thanks,
>    
>     Takashi
> 
>     >
>     > Michal W.
>     >
>     > >> ---
>     > >>   sound/usb/format.c | 28 +++++++++++++++++++++++++---
>     > >>   1 file changed, 25 insertions(+), 3 deletions(-)
>     > >>
>     > >> diff --git a/sound/usb/format.c b/sound/usb/format.c
>     > >> index c02b51a82775..05442f6ada62 100644
>     > >> --- a/sound/usb/format.c
>     > >> +++ b/sound/usb/format.c
>     > >> @@ -313,10 +313,32 @@ static int parse_audio_format_rates_v2v3(struct
>     snd_usb_audio *chip,
>     > >>                          tmp, sizeof(tmp));
>     > >>            if (ret < 0) {
>     > >> -          dev_err(&dev->dev,
>     > >> -                  "%s(): unable to retrieve number of sample rates
>     (clock %d)\n",
>     > >> +          switch (chip->usb_id) {
>     > >> +          /* LINE 6 HX pedals don't support getting the clock sample
>     rate.
>     > >> +           * Set the framerate to 48khz by default
>     > >> +           */
>     > >> +          case USB_ID(0x0E41, 0x4244): /* HELIX */
>     > >> +          case USB_ID(0x0E41, 0x4246): /* HX STOMP */
>     > >> +                  dev_warn(&dev->dev,
>     > >> +                          "%s(): line6 helix: unable to retrieve
>     number of sample rates. Set it to default value (clock %d).\n",
>     > >>                            __func__, clock);
>     > >> -          goto err;
>     > >> +                  fp->nr_rates = 1;
>     > >> +                  fp->rate_min = 48000;
>     > >> +                  fp->rate_max = 48000;
>     > >> +                  fp->rates = SNDRV_PCM_RATE_48000;
>     > >> +                  fp->rate_table = kmalloc(sizeof(int), GFP_KERNEL);
>     > >> +                  if (!fp->rate_table) {
>     > >> +                          ret = -ENOMEM;
>     > >> +                          goto err_free;
>     > >> +                  }
>     > >> +                  fp->rate_table[0] = 48000;
>     > >> +                  return 0;
>     > >> +          default:
>     > >> +                  dev_err(&dev->dev,
>     > >> +                          "%s(): unable to retrieve number of sample
>     rates (clock %d)\n",
>     > >> +                                  __func__, clock);
>     > >> +                  goto err;
>     > >> +          }
>     > >>    }
>     > >>            nr_triplets = (tmp[1] << 8) | tmp[0];
>     > >> --
>     > >> 2.19.1
>     > >>
>     > >>
>     > > _______________________________________________
>     > > Alsa-devel mailing list
>     > > Alsa-devel@alsa-project.org
>     > > https://mailman.alsa-project.org/mailman/listinfo/alsa-devel
>     > >
>     >
> 
>