kernel/linux.git/sound/usb/qcom, branch v6.18.22

usb: host: xhci-sideband: delegate offload_usage tracking to class drivers

2026-04-11T12:26:42+00:00

commit 5abbe6ecc6203355c770bf232ade88e29c960049 upstream. Remove usb_offload_get() and usb_offload_put() from the xHCI sideband interrupter creation and removal paths. The responsibility of manipulating offload_usage now lies entirely with the USB class drivers. They have the precise context of when an offload data stream actually starts and stops, ensuring a much more accurate representation of offload activity for power management. Cc: stable Fixes: ef82a4803aab ("xhci: sideband: add api to trace sideband usage") Signed-off-by: Guan-Yu Lin Tested-by: Hailong Liu Tested-by: hailong.liu@oppo.com Acked-by: Mathias Nyman Link: https://patch.msgid.link/20260401123238.3790062-3-guanyulin@google.com Signed-off-by: Greg Kroah-Hartman Signed-off-by: Greg Kroah-Hartman

ALSA: usb: qcom: Correct parameter comment for uaudio_transfer_buffer_setup()

2026-03-12T11:09:21+00:00

[ Upstream commit 1d6452a0ce78cd3f4e48943b5ba21d273a658298 ] At fixing the memory leak of xfer buffer, we forgot to update the corresponding comment, too. This resulted in a kernel-doc warning with W=1. Let's correct it. Fixes: 5c7ef5001292 ("ALSA: qc_audio_offload: avoid leaking xfer_buf allocation") Link: https://patch.msgid.link/20260226154414.1081568-4-tiwai@suse.de Signed-off-by: Takashi Iwai Signed-off-by: Sasha Levin

Merge branch 'for-linus' into for-next

2025-09-28T06:34:58+00:00

Pull 6.17-devel branch for applying further changes cleanly. Signed-off-by: Taksahi Iwai

ALSA: usb: qcom: Fix false-positive address space check

2025-09-17T13:10:58+00:00

The sanity check previously added to uaudio_transfer_buffer_setup() assumed the allocated buffer being linear-mapped. But the buffer allocated via usb_alloc_coherent() isn't always so, rather to be used with (SG-)DMA API. This leaded to a false-positive warning and the driver failed to work. Actually uaudio_transfer_buffer_setup() deals only with the DMA-API addresses for MEM_XFER_BUF type, while other callers of uaudio_iommu_map() are with pages with physical addresses for MEM_EVENT_RING and MEM_XFER_RING types. So this patch splits the mapping helper function to two different ones, uaudio_iommu_map() for the DMA pages and uaudio_iommu_map_pa() for the latter, in order to handle mapping differently for each type. Along with it, the unnecessary address check that caused probe error is dropped, too. Fixes: 3335a1bbd624 ("ALSA: qc_audio_offload: try to reduce address space confusion") Suggested-by: Arnd Bergmann Acked-by: Arnd Bergmann Reported-and-tested-by: Luca Weiss Closes: https://lore.kernel.org/DBR2363A95M1.L9XBNC003490@fairphone.com Signed-off-by: Takashi Iwai

ALSA: usb: qcom: Use guard() for mutex locks

2025-09-01T11:54:06+00:00

Replace the manual mutex lock/unlock pairs with guard() for code simplification. The manual mutex lock/unlock are still left in handle_uaudio_stream_req() and its callee as they have a bit complex locking patterns. Only code refactoring, and no behavior change. Signed-off-by: Takashi Iwai Link: https://patch.msgid.link/20250829150724.6886-8-tiwai@suse.de

ALSA: usb-audio: Use auto-cleanup for shutdown locks

2025-08-13T15:39:28+00:00

Introduce an auto-cleanup macro for the temporary shutdown locks for USB-audio, and replace the manual lock/unlock pairs with it. Namely, the former err = snd_usb_lock_shutdown(chip); if (err < 0) return err; .... snd_usb_unlock_shutdown(chip); is replaced with CLASS(snd_usb_lock, pm)(chip); if (pm.err < 0) return pm.err; .... with the automatic unlocking. Link: https://patch.msgid.link/20250811101647.8637-1-tiwai@suse.de Signed-off-by: Takashi Iwai

ALSA: usb-audio: qcom: Adjust mutex unlock order

2025-07-22T09:39:40+00:00

The mutexes qdev_mutex and chip->mutex are acquired in that order throughout the driver. To preserve proper lock hierarchy and avoid potential deadlocks, they must be released in the reverse order of acquisition. This change reorders the unlock sequence to first release chip->mutex followed by qdev_mutex, ensuring consistency with the locking pattern. [ fixed the code indentations and Fixes tag by tiwai ] Fixes: 326bbc348298a ("ALSA: usb-audio: qcom: Introduce QC USB SND offloading support") Signed-off-by: Erick Karanja Link: https://patch.msgid.link/20250721114554.1666104-1-karanja99erick@gmail.com Signed-off-by: Takashi Iwai

ALSA: usb: qcom: fix NULL pointer dereference in qmi_stop_session

2025-06-25T06:28:16+00:00

The find_substream() call may return NULL, but the error path dereferenced 'subs' unconditionally via dev_err(&subs->dev->dev, ...), causing a NULL pointer dereference when subs is NULL. Fix by switching to &uadev[idx].udev->dev which is always valid in this context. Signed-off-by: Pei Xiao Link: https://patch.msgid.link/86ac2939273ac853535049e60391c09d7688714e.1750755508.git.xiaopei01@kylinos.cn Signed-off-by: Takashi Iwai

ALSA: qc_audio_offload: Fix missing error code in prepare_qmi_response()

2025-06-24T08:11:02+00:00

When snd_soc_usb_find_priv_data() fails, return failure instead of success. While we are at it also use direct returns at first few error paths where there is no additional cleanup needed. Reported-by: Dan Carpenter Closes: https://lore.kernel.org/all/Z_40qL4JnyjR4j0O@stanley.mountain/ Fixes: 326bbc348298 ("ALSA: usb-audio: qcom: Introduce QC USB SND offloading support") Signed-off-by: Harshit Mogalapalli Link: https://patch.msgid.link/20250623142639.2938056-1-harshit.m.mogalapalli@oracle.com Signed-off-by: Takashi Iwai

ALSA: qc_audio_offload: try to reduce address space confusion

2025-05-21T12:34:10+00:00

uaudio_transfer_buffer_setup() allocates a buffer for the subs->dev device, and the returned address for the buffer is a CPU local virtual address that may or may not be in the linear mapping, as well as a DMA address token that is accessible by the USB device, and this in turn may or may not correspond to the physical address. The use in the driver however assumes that these addresses are the linear map and the CPU physical address, respectively. Both are nonportable here, but in the end only the virtual address gets used by converting it to a physical address that gets mapped into a second iommu. Make this more explicit by pulling the conversion out first and warning if it is not part of the linear map, and using the actual physical address to map into the iommu in place of the dma address that may already be iommu-mapped into the usb host. Signed-off-by: Arnd Bergmann Link: https://lore.kernel.org/r/20250513123442.159936-4-arnd@kernel.org Signed-off-by: Greg Kroah-Hartman