[PATCH] ALSA: usb-audio: Add implicit feedback quirk for Zoom UAC-2
Hi folks,
We're using a Zoom UAC-2 USB audio interface, which provides an async playback endpoint ("1 OUT (ASYNC)") and an async capture endpoint ("2 IN (ASYNC)"), both with 2-channel S32_LE in 44100, 48000, 88200, 96000, 176400, or 192000 samples per second. For whatever reason, the explicit feedback on the playback endpoint seems a bit unstable and also biased a little high. alsaloop -t 1000 gets repeated playback underruns and starts trying to resample the audio to adjust the rate (even though it's just looping from capture back to playback on the same device!).
I've found that using implicit feedback produces much more stable results and doesn't seem to hurt anything else (patch below). I wasn't sure if the other "implicit feedback quirk" devices in sound/usb/pcm.c are also ASYNC devices that happen to perform better using implicit feedback instead (in which case adding the Zoom UAC-2 to the list seems pretty routine), or if the others are SYNC devices that only need help identifying the right endpoint and ifnum -- i.e., how unorthodox is it to use implicit feedback on an ASYNC device? Hopefully not very.
Please let me know if you'd like any additional information or testing.
Best regards, Keith
From 2c8d691d39ab2824c344ac2562a1e8605b3384ba Mon Sep 17 00:00:00 2001
From: Keith Winstein keithw@cs.stanford.edu Date: Sun, 25 Oct 2020 22:05:47 -0700 Subject: [PATCH] ALSA: usb-audio: Add implicit feedback quirk for Zoom UAC-2
Signed-off-by: Keith Winstein keithw@cs.stanford.edu Tested-by: Keith Winstein keithw@cs.stanford.edu --- sound/usb/pcm.c | 4 ++++ 1 file changed, 4 insertions(+)
diff --git a/sound/usb/pcm.c b/sound/usb/pcm.c index b401ee894e1b..5723e8e69cf1 100644 --- a/sound/usb/pcm.c +++ b/sound/usb/pcm.c @@ -352,6 +352,10 @@ static int set_sync_ep_implicit_fb_quirk(struct snd_usb_substream *subs, ep = 0x81; ifnum = 2; goto add_sync_ep_from_ifnum; + case USB_ID(0x1686, 0xf029): /* ZOOM UAC-2 */ + ep = 0x82; + ifnum = 2; + goto add_sync_ep_from_ifnum; case USB_ID(0x1397, 0x0001): /* Behringer UFX1604 */ case USB_ID(0x1397, 0x0002): /* Behringer UFX1204 */ ep = 0x81;
Here is the patch again without line-wrapping.
From 2c8d691d39ab2824c344ac2562a1e8605b3384ba Mon Sep 17 00:00:00 2001
From: Keith Winstein keithw@cs.stanford.edu Date: Sun, 25 Oct 2020 22:05:47 -0700 Subject: [PATCH] ALSA: usb-audio: Add implicit feedback quirk for Zoom UAC-2
Signed-off-by: Keith Winstein keithw@cs.stanford.edu Tested-by: Keith Winstein keithw@cs.stanford.edu --- sound/usb/pcm.c | 4 ++++ 1 file changed, 4 insertions(+)
diff --git a/sound/usb/pcm.c b/sound/usb/pcm.c index b401ee894e1b..5723e8e69cf1 100644 --- a/sound/usb/pcm.c +++ b/sound/usb/pcm.c @@ -352,6 +352,10 @@ static int set_sync_ep_implicit_fb_quirk(struct snd_usb_substream *subs, ep = 0x81; ifnum = 2; goto add_sync_ep_from_ifnum; + case USB_ID(0x1686, 0xf029): /* ZOOM UAC-2 */ + ep = 0x82; + ifnum = 2; + goto add_sync_ep_from_ifnum; case USB_ID(0x1397, 0x0001): /* Behringer UFX1604 */ case USB_ID(0x1397, 0x0002): /* Behringer UFX1204 */ ep = 0x81;
On Mon, Oct 26, 2020 at 02:52:28PM +0100, Takashi Iwai wrote:
On Mon, 26 Oct 2020 09:01:43 +0100, Keith Winstein wrote:
Here is the patch again without line-wrapping.
Could you put the patch description in the patch itself? Basically I'm fine to merge this as is, just missing information for git.
Sure, here you go:
From 51fac91fa3a5da656ca6a49ca6cc98be7664efd7 Mon Sep 17 00:00:00 2001
From: Keith Winstein keithw@cs.stanford.edu Date: Sun, 25 Oct 2020 22:05:47 -0700 Subject: [PATCH] ALSA: usb-audio: Add implicit feedback quirk for Zoom UAC-2
The Zoom UAC-2 USB audio interface provides an async playback endpoint ("1 OUT (ASYNC)") and capture endpoint ("2 IN (ASYNC)"), both with 2-channel S32_LE in 44.1, 48, 88.2, 96, 176.4, or 192 kilosamples/s. The device provides explicit feedback to adjust the host's playback rate, but the feedback appears unstable and biased relative to the device's capture rate. "alsaloop -t 1000" experiences playback underruns and tries to resample the captured audio to match the varying playback rate. Forcing the kernel to use implicit feedback appears to produce more stable results. This causes the host to transmit one playback sample for each capture sample received. (Zoom North America has been notified of this change.)
Signed-off-by: Keith Winstein keithw@cs.stanford.edu Tested-by: Keith Winstein keithw@cs.stanford.edu --- sound/usb/pcm.c | 4 ++++ 1 file changed, 4 insertions(+)
diff --git a/sound/usb/pcm.c b/sound/usb/pcm.c index b401ee894e1b..31b1edf53c41 100644 --- a/sound/usb/pcm.c +++ b/sound/usb/pcm.c @@ -352,6 +352,10 @@ static int set_sync_ep_implicit_fb_quirk(struct snd_usb_substream *subs, ep = 0x81; ifnum = 2; goto add_sync_ep_from_ifnum; + case USB_ID(0x1686, 0xf029): /* Zoom UAC-2 */ + ep = 0x82; + ifnum = 2; + goto add_sync_ep_from_ifnum; case USB_ID(0x1397, 0x0001): /* Behringer UFX1604 */ case USB_ID(0x1397, 0x0002): /* Behringer UFX1204 */ ep = 0x81;
On Tue, 27 Oct 2020 08:18:41 +0100, Keith Winstein wrote:
On Mon, Oct 26, 2020 at 02:52:28PM +0100, Takashi Iwai wrote:
On Mon, 26 Oct 2020 09:01:43 +0100, Keith Winstein wrote:
Here is the patch again without line-wrapping.
Could you put the patch description in the patch itself? Basically I'm fine to merge this as is, just missing information for git.
Sure, here you go:
From 51fac91fa3a5da656ca6a49ca6cc98be7664efd7 Mon Sep 17 00:00:00 2001
From: Keith Winstein keithw@cs.stanford.edu Date: Sun, 25 Oct 2020 22:05:47 -0700 Subject: [PATCH] ALSA: usb-audio: Add implicit feedback quirk for Zoom UAC-2
The Zoom UAC-2 USB audio interface provides an async playback endpoint ("1 OUT (ASYNC)") and capture endpoint ("2 IN (ASYNC)"), both with 2-channel S32_LE in 44.1, 48, 88.2, 96, 176.4, or 192 kilosamples/s. The device provides explicit feedback to adjust the host's playback rate, but the feedback appears unstable and biased relative to the device's capture rate. "alsaloop -t 1000" experiences playback underruns and tries to resample the captured audio to match the varying playback rate. Forcing the kernel to use implicit feedback appears to produce more stable results. This causes the host to transmit one playback sample for each capture sample received. (Zoom North America has been notified of this change.)
Signed-off-by: Keith Winstein keithw@cs.stanford.edu Tested-by: Keith Winstein keithw@cs.stanford.edu
Applied now. Thanks.
Takashi
participants (2)
-
Keith Winstein
-
Takashi Iwai