A preliminary change for the later big changes.  This is a minor code
refactoring to drop the unnecessary arguments that can be retrieved in
a different way.
Tested-by: NKeith Milner <kamilner@superlative.org>
Tested-by: NDylan Robinson <dylan_robinson@motu.com>
Link: https://lore.kernel.org/r/20201123085347.19667-20-tiwai@suse.deSigned-off-by: NTakashi Iwai <tiwai@suse.de>

A preliminary patch for the later big change.  Just a minor code
refactoring.
Tested-by: NKeith Milner <kamilner@superlative.org>
Tested-by: NDylan Robinson <dylan_robinson@motu.com>
Link: https://lore.kernel.org/r/20201123085347.19667-19-tiwai@suse.deSigned-off-by: NTakashi Iwai <tiwai@suse.de>

Add a helper function to retrieve the usb_host_interface object from
the given interface and altsetting number pair, which is a commonly
used procedure in the driver code.

No functional changes, just minor code refactoring.
Tested-by: NKeith Milner <kamilner@superlative.org>
Tested-by: NDylan Robinson <dylan_robinson@motu.com>
Link: https://lore.kernel.org/r/20201123085347.19667-17-tiwai@suse.deSigned-off-by: NTakashi Iwai <tiwai@suse.de>

This behavior turned out to be invalid from the USB spec POV and
shouldn't be applied.  As it's an optional flag that is set only via
an card control element that must be hardly used, let's drop it
again.
Tested-by: NKeith Milner <kamilner@superlative.org>
Tested-by: NDylan Robinson <dylan_robinson@motu.com>
Link: https://lore.kernel.org/r/20201123085347.19667-16-tiwai@suse.deSigned-off-by: NTakashi Iwai <tiwai@suse.de>

Currently snd_usb_endpoint objects are created at first when the
substream is opened and tries to assign the endpoints corresponding to
the matching audioformat.  But since basically the all endpoints have
been already parsed and the information have been obtained, we may
create the endpoint objects statically at the init phase.  It's easier
to manage for the implicit fb case, for example.

This patch changes the endpoint object management and lets the parser
to create the all endpoint objects.

This change shouldn't bring any functional changes.
Tested-by: NKeith Milner <kamilner@superlative.org>
Tested-by: NDylan Robinson <dylan_robinson@motu.com>
Link: https://lore.kernel.org/r/20201123085347.19667-15-tiwai@suse.deSigned-off-by: NTakashi Iwai <tiwai@suse.de>

The implicit feedback mode initializes both the main data stream and
the sync data stream.  When a sync stream was already opened, this
would result in the doubly initialization and might screw up things.

Add the check of already opened sync streams and skip the unnecessary
initialization.
Tested-by: NKeith Milner <kamilner@superlative.org>
Tested-by: NDylan Robinson <dylan_robinson@motu.com>
Link: https://lore.kernel.org/r/20201123085347.19667-14-tiwai@suse.deSigned-off-by: NTakashi Iwai <tiwai@suse.de>

The file debug.h contains a simple macro for debug prints, and it's
used only in two places, the format parser and the hw_params rules.
The former actually should print a more informative message instead,
so the only users are the hw_parmas rules.

This patch moves the contents of debug.h into the hw_params rules
local code and remove the unneeded includes.  Also, the debug print in
the format parser is replaced with the information print with more
useful information, and the raw printk() call is replaced with
pr_debug().
Tested-by: NKeith Milner <kamilner@superlative.org>
Tested-by: NDylan Robinson <dylan_robinson@motu.com>
Link: https://lore.kernel.org/r/20201123085347.19667-13-tiwai@suse.deSigned-off-by: NTakashi Iwai <tiwai@suse.de>

Several hw_params functions narrows the interval via min/max rule in
the very similar way, so factor out those into a helper function and
use commonly.

No functional changes, just minor code refactoring.
Tested-by: NKeith Milner <kamilner@superlative.org>
Tested-by: NDylan Robinson <dylan_robinson@motu.com>
Link: https://lore.kernel.org/r/20201123085347.19667-12-tiwai@suse.deSigned-off-by: NTakashi Iwai <tiwai@suse.de>

In the current code, there is no check at the stream open time whether
the endpoint is being already used by others.  In the normal
operations, this shouldn't happen, but in the case of the implicit
feedback mode, it's a common problem with the full duplex operation,
because the capture stream is always opened by the playback stream as
an implicit sync source.

Although we recently introduced the check of such a conflict of
parameters at the PCM hw_params time, it doesn't give any hint at the
hw_params itself and just gives the error.  This isn't quite
comfortable, and it caused problems on many applications.

This patch attempts to make the parameter handling easier by
introducing the strict hw constraint matching with the counterpart
stream that is being used.  That said, when an implicit feedback
playback stream is running before a capture stream is opened, the
capture stream carries the PCM hw-constraint to allow only the same
sample rate, format, periods and period frames as the running playback
stream.  If not opened or there is no conflict of endpoints, the
behavior remains as same as before.

Note that this kind of "weak link" should work for most cases, but
this is no concrete solution; e.g. if an application changes the hw
params multiple times while another stream is opened, this would lead
to inconsistencies.
Tested-by: NKeith Milner <kamilner@superlative.org>
Tested-by: NDylan Robinson <dylan_robinson@motu.com>
Link: https://lore.kernel.org/r/20201123085347.19667-11-tiwai@suse.deSigned-off-by: NTakashi Iwai <tiwai@suse.de>

This is a preliminary work for the upcoming hw-constraint change for
the implicit feedback mode.

Currently snd_usb_autoresume() is called at the end of
setup_hwinfo().  It's a bit confusing; because of this implicit
refcount usage, the caller side needs to call snd_usb_autosuspend()
later in the error path although it's not seen inside the function.
Instead, it's clearer to call both snd_usb_autoresume() and suspend()
in the very same function.

It's only refactoring and no functional changes.
Tested-by: NKeith Milner <kamilner@superlative.org>
Tested-by: NDylan Robinson <dylan_robinson@motu.com>
Link: https://lore.kernel.org/r/20201123085347.19667-10-tiwai@suse.deSigned-off-by: NTakashi Iwai <tiwai@suse.de>

Instead of parsing and evaluating the sync endpoint and the implicit
feedback mode at each time the audio stream is opened, let's parse it
once at the probe time, as the all needed information can be obtained
statically from the descriptor or from the quirk.

This patch extends audioformat struct to record the sync endpoint,
interface and altsetting as well as the implicit feedback flag, which
are filled at parsing the streams.  Then, set_sync_endpoint() is much
simplified just to follow the already parsed data.
Tested-by: NKeith Milner <kamilner@superlative.org>
Tested-by: NDylan Robinson <dylan_robinson@motu.com>
Link: https://lore.kernel.org/r/20201123085347.19667-9-tiwai@suse.deSigned-off-by: NTakashi Iwai <tiwai@suse.de>

There are a few rooms for improvements wrt the debug prints:
- The EP debug print is shown only at starting, not at stopping
- The EP debug print contains useless object addresses
- Some helpers show the urb and the EP object addresses, too

This patch addresses those shortcomings.
Tested-by: NKeith Milner <kamilner@superlative.org>
Tested-by: NDylan Robinson <dylan_robinson@motu.com>
Link: https://lore.kernel.org/r/20201123085347.19667-8-tiwai@suse.deSigned-off-by: NTakashi Iwai <tiwai@suse.de>

The sync EP setup isn't cleared at stopping the stream but expected to
be cleared at the next stream start.  This may leave the sync link
setup stale and can spoof wrongly when full duplex streams were
running in the implicit fb sync.  Let's initialize them properly at
start and end of the stream.
Tested-by: NKeith Milner <kamilner@superlative.org>
Tested-by: NDylan Robinson <dylan_robinson@motu.com>
Link: https://lore.kernel.org/r/20201123085347.19667-7-tiwai@suse.deSigned-off-by: NTakashi Iwai <tiwai@suse.de>

It seems that many UAC2 devices are with the implicit feedback, but
they couldn't be probed properly because the assumption the driver
takes currently isn't applied: they have the single endpoint for both
data and implicit-fb streams, while we checked only the classical sync
endpoints assigned to the next altsetting in the same interface.

This patch extends the search to match with those typical cases where
the implicit fb stream is found in the next interface number.

While we're at it, slightly refactor the code, not returning 0/-ERROR
but use the standard bool to success/failur, which is more intuitive
in this particular case.
Reported-by: NDylan Robinson <dylan_robinson@motu.com>
Tested-by: NKeith Milner <kamilner@superlative.org>
Tested-by: NDylan Robinson <dylan_robinson@motu.com>
Link: https://lore.kernel.org/r/20201123085347.19667-5-tiwai@suse.deSigned-off-by: NTakashi Iwai <tiwai@suse.de>

The PCM trigger callback is atomic, hence we must not call a function
like usb_set_interface() there.  Calling it from there would lead to a
kernel Oops.

Fix it by moving the usb_set_interface() call to set_sync_endpoint().

Also, apply the snd_usb_set_interface_quirk() for consistency, too.
Tested-by: NKeith Milner <kamilner@superlative.org>
Tested-by: NDylan Robinson <dylan_robinson@motu.com>
Link: https://lore.kernel.org/r/20201123085347.19667-3-tiwai@suse.deSigned-off-by: NTakashi Iwai <tiwai@suse.de>

In the current code, when the device provides the discrete sample rate
tables with unusual sample rates, the driver tries to gather the whole
values from the audioformat entries and create a hw-constraint rule to
restrict with this single rate list.  This is rather inefficient and
may overlook the rates that are associated only with the certain
audioformat entries.

This patch improves the hw constraint setup by rewriting the existing
hw_rule_rate().  The discrete sample rates (identified by rate_table
and nr_rates of format entry) are checked in the existing
hw_rule_rate() instead of extra rules; in the case of discrete rates,
the function compares with each rate table entry and calculates the
min/max values from there.  For the contiguous rates, the behavior
doesn't change.

Along with it, snd_usb_pcm_check_knot() and snb_usb_substream
rate_list field become superfluous, thus those are dropped.
Tested-by: NKeith Milner <kamilner@superlative.org>
Tested-by: NDylan Robinson <dylan_robinson@motu.com>
Link: https://lore.kernel.org/r/20201123085347.19667-2-tiwai@suse.deSigned-off-by: NTakashi Iwai <tiwai@suse.de>

This patch fixes audio distortion on playback for the Allen&Heath
Qu-16.
Signed-off-by: NGeoffrey D. Bennett <g@b4.vu>
Cc: <stable@vger.kernel.org>
Link: https://lore.kernel.org/r/20201104115717.GA19046@b4.vuSigned-off-by: NTakashi Iwai <tiwai@suse.de>

This patch fixes audio distortion on playback for the Yamaha MODX.
Signed-off-by: NGeoffrey D. Bennett <g@b4.vu>
Tested-by: NFrank Slotta <frank.slotta@posteo.de>
Cc: <stable@vger.kernel.org>
Link: https://lore.kernel.org/r/20201104120705.GA19126@b4.vuSigned-off-by: NTakashi Iwai <tiwai@suse.de>

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: NKeith Winstein <keithw@cs.stanford.edu>
Tested-by: NKeith Winstein <keithw@cs.stanford.edu>
Cc: <stable@vger.kernel.org>
BugLink: https://lore.kernel.org/r/20201027071841.GA164525@trolley.csail.mit.eduSigned-off-by: NTakashi Iwai <tiwai@suse.de>

This patch extends support for DJM-250MK2 and allows recording.
However, DVS is not possible yet (see the comment in code).
Signed-off-by: NFrantišek Kučera <franta-linux@frantovo.cz>
Cc: <stable@vger.kernel.org>
Link: https://lore.kernel.org/r/20200825153113.6352-1-konference@frantovo.czSigned-off-by: NTakashi Iwai <tiwai@suse.de>

This uses the same quirk as the Motu and SSL2 devices.
Tested on the UR22C.

Fixes bug 208851.
Signed-off-by: NJoshua Sivec <sivec@posteo.net>
BugLink: https://bugzilla.kernel.org/show_bug.cgi?id=208851
Cc: <stable@vger.kernel.org>
Link: https://lore.kernel.org/r/20200825165515.8239-1-sivec@posteo.netSigned-off-by: NTakashi Iwai <tiwai@suse.de>

Further investigation of the L-R swap problem on the MS2109 reveals that
the problem isn't that the channels are swapped, but rather that they
are swapped and also out of phase by one sample. In other words, the
issue is actually that the very first frame that comes from the hardware
is a half-frame containing only the right channel, and after that
everything becomes offset.

So introduce a new quirk field to drop the very first 2 bytes that come
in after the format is configured and a capture stream starts. This puts
the channels in phase and in the correct order.

Cc: stable@vger.kernel.org
Signed-off-by: NHector Martin <marcan@marcan.st>
Link: https://lore.kernel.org/r/20200810082400.225858-1-marcan@marcan.stSigned-off-by: NTakashi Iwai <tiwai@suse.de>

As expected, this requires the same quirk as the SSL2+ in order for the
clock to sync. This was suggested by, and tested on an SSL2, by Dmitry.
Suggested-by: NDmitry <dpavlushko@gmail.com>
Signed-off-by: NLaurence Tratt <laurie@tratt.net>
Cc: <stable@vger.kernel.org>
Link: https://lore.kernel.org/r/20200621075005.52mjjfc6dtdjnr3h@overdrive.tratt.netSigned-off-by: NTakashi Iwai <tiwai@suse.de>

Replace the existing /* fall through */ comments and its variants with
the new pseudo-keyword macro fallthrough[1]. Also, remove unnecessary
fall-through markings when it is the case.

[1] https://www.kernel.org/doc/html/latest/process/deprecated.html?highlight=fallthrough#implicit-switch-case-fall-throughSigned-off-by: NGustavo A. R. Silva <gustavoars@kernel.org>
Link: https://lore.kernel.org/r/20200708203236.GA5112@embeddedorSigned-off-by: NTakashi Iwai <tiwai@suse.de>

USB Audio analyzer RTX6001 uses the same implicit feedback quirk
as other XMOS-based devices.
Signed-off-by: NPavel Hofman <pavel.hofman@ivitera.com>
Tested-by: NPavel Hofman <pavel.hofman@ivitera.com>
Cc: <stable@vger.kernel.org>
Link: https://lore.kernel.org/r/822f0f20-1886-6884-a6b2-d11c685cbafa@ivitera.comSigned-off-by: NTakashi Iwai <tiwai@suse.de>

Fix the following typos in comments and in the code:
 - KHz -> kHz
 - procssed -> processed
Signed-off-by: NAlexander Tsoy <alexander@tsoy.me>
Link: https://lore.kernel.org/r/20200629032607.255419-1-alexander@tsoy.meSigned-off-by: NTakashi Iwai <tiwai@suse.de>

With the recent full-duplex support of implicit feedback streams, an
endpoint can be still running after closing the capture stream as long
as the playback stream with the sync-endpoint is running.  In such a
state, the URBs are still be handled and they may call retire_data_urb
callback, which tries to transfer the data from the PCM buffer.  Since
the PCM stream gets closed, this may lead to use-after-free.

This patch adds the proper clearance of the callback at stopping the
capture stream for addressing the possible UAF above.

Fixes: 10ce77e4 ("ALSA: usb-audio: Add duplex sound support for USB devices using implicit feedback")
Link: https://lore.kernel.org/r/20200616120921.12249-1-tiwai@suse.deSigned-off-by: NTakashi Iwai <tiwai@suse.de>

This uses the same quirk as the Motu M2 and M4 to ensure the driver uses the
audio interface's clock. Tested on an SSL2+.
Signed-off-by: NLaurence Tratt <laurie@tratt.net>
Cc: <stable@vger.kernel.org>
Link: https://lore.kernel.org/r/20200612111807.dgnig6rwhmsl2bod@overdrive.tratt.netSigned-off-by: NTakashi Iwai <tiwai@suse.de>

Pioneer DJ DJM-900NXS2 is a widely used DJ mixer with 2 audio USB
interfaces. Both have a MIDI controller, 10 playback and 12 capture
channels. Audio endpoints are vendor-specific and 3 files need to be
patched. All playback and capture channels work fine with all supported
sample rates (44.1k, 48k, 96k). Patches are attached.
Signed-off-by: NDmitry Panchenko <dmitry@d-systems.ee>
Link: https://lore.kernel.org/r/48ab19ff-3303-9bf8-ed0e-bcb31d8537eb@d-systems.eeSigned-off-by: NTakashi Iwai <tiwai@suse.de>

For USB sound devices using implicit feedback the endpoint used for
this feedback should be able to be opened twice, once for required
feedback and second time for audio data. This way these devices can be
put in duplex audio mode. Since this only works if the settings of the
endpoint don't change a check is included for this.

This fixes bug 207023 ("MOTU M2 regression on duplex audio") and
should also fix bug 103751 ("M-Audio Fast Track Ultra usb audio device
will not operate full-duplex")

Fixes: c2491779 ("ALSA: usb-audio: add implicit fb quirk for MOTU M Series")
Signed-off-by: NErwin Burema <e.burema@gmail.com>
BugLink: https://bugzilla.kernel.org/show_bug.cgi?id=207023
BugLink: https://bugzilla.kernel.org/show_bug.cgi?id=103751
Link: https://lore.kernel.org/r/2410739.SCZni40SNb@alpha-wolfSigned-off-by: NTakashi Iwai <tiwai@suse.de>

For computation of the the next frame size current value of fs/fps and
accumulated fractional parts of fs/fps are used, where values are stored
in Q16.16 format. This is quite natural for computing frame size for
asynchronous endpoints driven by explicit feedback, since in this case
fs/fps is a value provided by the feedback endpoint and it's already in
the Q format. If an error is accumulated over time, the device can
adjust fs/fps value to prevent buffer overruns/underruns.

But for synchronous endpoints the accuracy provided by these computations
is not enough. Due to accumulated error the driver periodically produces
frames with incorrect size (+/- 1 audio sample).

This patch fixes this issue by implementing a different algorithm for
frame size computation. It is based on accumulating of the remainders
from division fs/fps and it doesn't accumulate errors over time. This
new method is enabled for synchronous and adaptive playback endpoints.
Signed-off-by: NAlexander Tsoy <alexander@tsoy.me>
Link: https://lore.kernel.org/r/20200424022449.14972-1-alexander@tsoy.meSigned-off-by: NTakashi Iwai <tiwai@suse.de>

MicroBook IIc operates in UAC2 mode by default. This patch addresses
several issues with it:

- MicroBook II and IIc shares the same USB ID. We can distinguish them
  by interface class.
- MaxPacketsOnly attribute is erroneously set in endpoint descriptors.
  As a result this card produces noise with all sample rates other than
  96 KHz. This also causes issues like IOMMU page faults and other
  problems with host controller.
- Sample rate changes takes more than 2 seconds for this device. Clock
  validity request returns false during that period, so the clock validity
  quirk is required.
Signed-off-by: NAlexander Tsoy <alexander@tsoy.me>
Link: https://lore.kernel.org/r/20200229151815.14199-1-alexander@tsoy.meSigned-off-by: NTakashi Iwai <tiwai@suse.de>

This fixes crackling sound during playback.

Further note: MOTU is known for reusing Product IDs for different
devices or different generations of the device (e.g. MicroBook
I/II/IIc shares a single Product ID). This patch was only tested with
M4 audio interface, but the same Product ID is also used by M2. Hope
it will work for M2 as well.
Signed-off-by: NAlexander Tsoy <alexander@tsoy.me>
Link: https://lore.kernel.org/r/20200115151358.56672-1-alexander@tsoy.meSigned-off-by: NTakashi Iwai <tiwai@suse.de>

The altsetting sanity check in set_sync_ep_implicit_fb_quirk() was
checking for there to be at least one altsetting but then went on to
access the second one, which may not exist.

This could lead to random slab data being used to initialise the sync
endpoint in snd_usb_add_endpoint().

Fixes: c75a8a7a ("ALSA: snd-usb: add support for implicit feedback")
Fixes: ca10a7eb ("ALSA: usb-audio: FT C400 sync playback EP to capture EP")
Fixes: 5e35dc03 ("ALSA: usb-audio: add implicit fb quirk for Behringer UFX1204")
Fixes: 17f08b0d ("ALSA: usb-audio: add implicit fb quirk for Axe-Fx II")
Fixes: 103e9625 ("ALSA: usb-audio: simplify set_sync_ep_implicit_fb_quirk")
Cc: stable <stable@vger.kernel.org>     # 3.5
Signed-off-by: NJohan Hovold <johan@kernel.org>
Link: https://lore.kernel.org/r/20200114083953.1106-1-johan@kernel.orgSigned-off-by: NTakashi Iwai <tiwai@suse.de>

Make sure to check the return value of usb_altnum_to_altsetting() to
avoid dereferencing a NULL pointer when the requested alternate settings
is missing.

The format altsetting number may come from a quirk table and there does
not seem to be any other validation of it (the corresponding index is
checked however).

Fixes: b099b969 ("ALSA: usb-audio: Avoid superfluous usb_set_interface() calls")
Cc: stable <stable@vger.kernel.org>     # 4.18
Signed-off-by: NJohan Hovold <johan@kernel.org>
Link: https://lore.kernel.org/r/20191220093134.1248-1-johan@kernel.orgSigned-off-by: NTakashi Iwai <tiwai@suse.de>

Recently we found the headset-mic on the Dell Dock WD19 doesn't work
anymore after s3 (s2i or deep), this problem could be workarounded by
closing (pcm_close) the app and then reopening (pcm_open) the app, so
this bug is not easy to be detected by users.

When problem happens, retire_capture_urb() could still be called
periodically, but the size of captured data is always 0, it could be
a firmware bug on the dock. Anyway I found after resuming, the
snd_usb_pcm_prepare() will be called, and if we forcibly run
set_format() to set the interface and its endpoint, the capture
size will be normal again. This problem and workaound also apply to
playback.

To fix it in the kernel, add a quirk to let set_format() run
forcibly once after resume.
Signed-off-by: NHui Wang <hui.wang@canonical.com>
Cc: <stable@vger.kernel.org>
Link: https://lore.kernel.org/r/20191218132650.6303-1-hui.wang@canonical.comSigned-off-by: NTakashi Iwai <tiwai@suse.de>

USB-audio driver had some implementation of its own sync-stop
mechanism.  This patch moved a part of it to the common PCM sync_stop
ops.

Link: https://lore.kernel.org/r/20191210063454.31603-56-tiwai@suse.deSigned-off-by: NTakashi Iwai <tiwai@suse.de>

PCM core deals the empty ioctl field now as default(*).
Let's kill the redundant lines.

(*) commit fc033cbf ("ALSA: pcm: Allow NULL ioctl ops")

Link: https://lore.kernel.org/r/20191210061145.24641-22-tiwai@suse.deSigned-off-by: NTakashi Iwai <tiwai@suse.de>

Clean up the driver with the new managed buffer allocation API.
The superfluous snd_pcm_lib_malloc_pages() and
snd_pcm_lib_free_pages() calls are dropped.

Link: https://lore.kernel.org/r/20191209094943.14984-71-tiwai@suse.deSigned-off-by: NTakashi Iwai <tiwai@suse.de>

The recent change (*) in the ALSA memalloc core allows us to drop the
special vmalloc-specific allocation and page handling.  This patch
coverts to the common code.
(*) 1fe7f397: ALSA: memalloc: Add vmalloc buffer allocation
                  support
    7e8edae3: ALSA: pcm: Handle special page mapping in the
                  default mmap handler

Also, since the SG-buffer-specific PCM ops becomes identical with the
normal PCM ops, unify them again to the single ops, too.

Link: https://lore.kernel.org/r/20191105151856.10785-9-tiwai@suse.deSigned-off-by: NTakashi Iwai <tiwai@suse.de>