A failed system resume can leave the card unusable until reboot.
usb_audio_resume() jumps to err_out when snd_usb_pcm_resume() or
snd_usb_mixer_resume() fails. The error path skips the out: block, which
restores D0 and decrements chip->num_suspended_intf.
The card stays in SNDRV_CTL_POWER_D3hot, so later control access blocks in
snd_power_ref_and_wait(). USB core logs an interface resume callback error.
It does not retry that callback, so a later callback cannot complete the
skipped cleanup.
usb_audio_suspend() increments num_suspended_intf before returning success.
A system-resume callback must consume the system-suspend count even if a
component resume fails. Otherwise, the stranded count skews later suspend
and resume cycles.
Do not apply this cleanup to runtime-resume errors. Runtime PM can retry
-EAGAIN or -EBUSY without another suspend callback. The count must continue
to describe that suspended interface. Other runtime-resume errors latch
runtime_error in the PM core and do not cause an immediate callback retry.
Both parts of the system-resume error path are longstanding. Commit
88a8516a2128a ("ALSA: usbaudio: implement USB autosuspend") introduced
err_out past the D0 restore. Commit 862b2509d157c ("ALSA: usb-audio: Fix
inconsistent card PM state after resume") later moved
num_suspended_intf-- into the out: block. The error path now skips both
operations.
No third-party code is needed to reach the error path.
snd_usb_mixer_resume() ends in snd_usb_mixer_activate(), which returns the
result of usb_submit_urb() for devices that have a mixer status URB. Its
mixer->private_resume hook can also fail through scarlett2_init_notify().
snd_usb_pcm_resume() issues a SET_CUR request to a UAC3 power domain. It
can return -EPIPE or -EIO when the device stalls the request.
Route a component error through out: only when system_suspend is nonzero.
Continue to return runtime-resume errors through err_out. Later component
resume stages remain skipped. The original error still reaches USB core.
A later transfer can fail if the device did not recover.
I reproduced the system-resume failure on an Audient iD14 MkI with an
out-of-tree diagnostic mixer resume hook. An injected -EIO on the unpatched
core left control readers in uninterruptible sleep in
snd_power_ref_and_wait() until a reboot. With this patch, the same failure
restored control access. A second system suspend and resume also succeeded
after I disabled fault injection.
Assisted-by: Claude:claude-opus-5
Assisted-by: Antigravity:gemini-3.1-pro-high
Assisted-by: Codex:gpt-5.6-sol
Fixes: 88a8516a2128a ("ALSA: usbaudio: implement USB autosuspend")
Fixes: 862b2509d157c ("ALSA: usb-audio: Fix inconsistent card PM state after resume")
Cc: <stable@vger.kernel.org>
Signed-off-by: Will Porter <mrwillporter@gmail.com>
Link: https://patch.msgid.link/20260824225757.26749-1-mrwillporter@gmail.com
Signed-off-by: Takashi Iwai <tiwai@suse.de>
The M-Audio Venom (0763:2084) does not answer any GET_CUR request of its feature units, hanging up the USB microcontroller and "responding" with timeouts. So the mixer building lasts around 47 seconds, and the device does not stream at all. The same GET_CUR requests issued through libusb (with no kernel driver bound) complete correctly and blazingly fast. So it seems to only happen during the initial probe. I defined an explicit composite quirk to bypass the mixer creation, as it is not needed (the synth already comes with volume controls). With both this and the device flag applied, the synth works flawlessly. Signed-off-by: Federico Valentín Andrade <fandrade@frba.utn.edu.ar> Link: https://patch.msgid.link/20260824140211.17003-3-fandrade@frba.utn.edu.ar Signed-off-by: Takashi Iwai <tiwai@suse.de>
The M-Audio Venom (0763:2084) is an USB Audio Class 1 compliant synth with an integrated audio interface, it does not implement GET_CUR on SAMPLING_FREQ_CONTROL, hanging up if requested on endpoint 0. The first class request issued by the driver after enumeration is a GET_CUR on endpoint 0x02, freezing the device's USB microcontroller. Timing out on every transfer afterwards with -ETIMEDOUT, such as SET_INTERFACE, so neither the mixer nor any streaming interface is set up. Analyzing a USBPcap capture of the Windows driver I found it never requests the sampling frequency, as the synth exposes a single discrete of 44100Hz on both streaming interfaces, thus asking for it is unnecessary. So I applied get_sample_rate to avoid this check, and disable_autosuspend because the synth doesn't come back from being suspended. Signed-off-by: Federico Valentín Andrade <fandrade@frba.utn.edu.ar> Link: https://patch.msgid.link/20260824140211.17003-2-fandrade@frba.utn.edu.ar Signed-off-by: Takashi Iwai <tiwai@suse.de>
snd_usbmidi_novation_output() lays out a two-byte header at
transfer_buffer[0..1] and passes &transfer_buffer[2] together with a
length of ep->max_transfer - 2 to snd_rawmidi_transmit():
count = snd_rawmidi_transmit(ep->ports[0].substream,
&transfer_buffer[2],
ep->max_transfer - 2);
ep->max_transfer comes from the output endpoint's wMaxPacketSize via
usb_maxpacket(). A malformed or malicious device can advertise a bulk
OUT endpoint with a wMaxPacketSize of 1 - the USB core only clamps this
value downwards - so ep->max_transfer becomes 1 and the count argument
becomes -1.
snd_rawmidi_transmit() passes the negative count on to
__snd_rawmidi_transmit_peek(), where "if (count1 > count) count1 = count"
leaves count1 negative; get_aligned_size() keeps it negative for a
byte-stream substream, so the following memcpy(buffer, ..., count1) runs
with a (size_t)-1 length and writes far past the transfer buffer, which
was allocated with usb_alloc_coherent(ep->max_transfer).
This is the same class of bug that was fixed for snd_usbmidi_akai_output()
in commit 0970274613fb ("ALSA: usb-audio: fix OOB write in
snd_usbmidi_akai_output()"); the novation output routine was left
unguarded. Bail out when the endpoint cannot hold the two-byte header
plus at least one payload byte.
Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
Cc: stable@vger.kernel.org
Signed-off-by: Marouane El Moufid <eun0us@espilon.net>
Link: https://patch.msgid.link/178749334830.543645.13722252148340572274@espilon.net
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Add the iD14 MkI USB ID to the existing audient_id14_map. This names feature unit 12 Monitor Mix Playback Volume. Live meter A/B testing on MkI USB ID 2708:0002 reproduced the same asymmetric attenuation seen on MkII. At ALSA value 104, the right channel was 23.07 dB below the left. At 127, both channels were equal. The misleading Speaker Playback Volume name causes PipeWire to adopt FU12 as the stream master. Reuse the existing map to keep the control available while preventing that selection. Hardware-tested on Bazzite 7.2.0-ogc4.1.fc44.x86_64 with this change backported. ALSA exposed Monitor Mix Playback Volume after module load and USB reconnect. PipeWire volume changes from 100% to 35% and back left all four FU12 values at 127. Channel-isolated left/right playback remained balanced after reconnect. Build-tested with CONFIG_SND_USB_AUDIO=m using the focused sound/usb target. Assisted-by: Codex:gpt-5 Signed-off-by: Will Porter <mrwillporter@gmail.com> Link: https://patch.msgid.link/20260823015353.686080-1-mrwillporter@gmail.com Signed-off-by: Takashi Iwai <tiwai@suse.de>
The Logitech PRO X Wireless is a wireless headset with a hotpluggable microphone. Its Playback mixer's GET_CUR somehow becomes broken when the microphone is detached, so set QUIRK_FLAG_MIXER_GET_CUR_OK to prevent the mixer behavior from depending on whether the microphone is attached. Meanwhile, the Playback mixer's minimum value doesn't work properly, thus set QUIRK_FLAG_MIXER_PLAYBACK_MIN_MUTE to prevent userspace audio stack from using the minimum mixer value to tune volume (setting it when muted is OK). Reported-by: Alexander Niemeyer <adventureFAN@gmx.de> Closes: https://msgid.link/6262cbbd-d1f2-4c9d-a1c7-9c5d12636f4b@gmx.de Closes: https://msgid.link/7984832b-86f6-4934-bfc0-1ed70218973a@gmx.de Signed-off-by: Rong Zhang <i@rong.moe> Signed-off-by: Takashi Iwai <tiwai@suse.de> Link: https://patch.msgid.link/20260823-uac-demote-sticky-check-v1-4-a8cbe572edff@rong.moe
The default behavior of sticky check now becomes what QUIRK_FLAG_MIXER_GET_CUR_BROKEN originally does, so the quirk flag is no longer needed. On some devices, whether their GET_CUR being sticky depends on whether hotpluggable components are present. When the hotpluggable components are missing on probe, their GET_CUR behavior is classified as broken. Therefore, reverse QUIRK_FLAG_MIXER_GET_CUR_BROKEN as QUIRK_FLAG_MIXER_GET_CUR_OK, so that it can be set to prevent the heuristics from gating GET_CUR. Note that even if the quirk flag is set, init_cur_mix_raw() should still initialize the mixer value to cval->min, otherwise restoring the bogus saved value on the first channel could lead to unbalanced channels. Signed-off-by: Rong Zhang <i@rong.moe> Signed-off-by: Takashi Iwai <tiwai@suse.de> Link: https://patch.msgid.link/20260823-uac-demote-sticky-check-v1-3-a8cbe572edff@rong.moe
Currently, a mixer is disabled when its GET_CUR is sticky, causing userspace to fall back to soft mixers, unless QUIRK_FLAG_MIXER_GET_CUR_BROKEN is set. This leads to issues on some wireless headphones with broken GET_CUR but effective SET_CUR, which use poorly-performed lossy codecs and are prone to audible distortion at low volume. They have to set the quirk flag to reeanble the mixer. Considering that users can always opt into soft mixers if they need it, i.e., when SET_CUR is stubbed, demote the severity of sticky GET_CUR by marking GET_CUR as broken and only provide mixer values from the cache. The mixer itself is still registered. Signed-off-by: Rong Zhang <i@rong.moe> Signed-off-by: Takashi Iwai <tiwai@suse.de> Link: https://patch.msgid.link/20260823-uac-demote-sticky-check-v1-2-a8cbe572edff@rong.moe
QUIRK_FLAG_ALWAYS_SET_RATE was introduced into usb-audio before without
appropriate documentation, so add it.
There is an odd colon in its comments, so remove it too.
As this is often forgotten, also add a reminder to the end of the
enumeration.
Fixes: 786f91da8535 ("ALSA: usb-audio: add QUIRK_FLAG_ALWAYS_SET_RATE for Mackie DLZ Creator XS")
Signed-off-by: Rong Zhang <i@rong.moe>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Link: https://patch.msgid.link/20260823-uac-demote-sticky-check-v1-1-a8cbe572edff@rong.moe
The SMSL USB DAC with USB ID 152a:85dd produces an audible pop when snd-usb-audio performs its redundant probe-time interface setup. It also pops when returning from runtime suspend. Skip the probe-time interface setup and disable runtime autosuspend for this device. Both flags have been verified through the quirk_flags module parameter. Keep QUIRK_FLAG_DSD_RAW in the device entry because the exact match takes precedence over the generic 0x152a vendor entry from which the device currently inherits that flag. Signed-off-by: Edward Blair <edward.blair@gmail.com> Link: https://patch.msgid.link/20260822114110.1069541-1-edward.blair@gmail.com Signed-off-by: Takashi Iwai <tiwai@suse.de>