linux.git/sound/core, branch v5.12.5

ALSA: core: remove redundant spin_lock pair in snd_card_disconnect

2021-05-14T08:52:47+00:00

[ Upstream commit abc21649b3e5c34b143bf86f0c78e33d5815e250 ]

modification in commit 2a3f7221acdd ("ALSA: core: Fix card races between
register and disconnect") resulting in this problem.

Fixes: 2a3f7221acdd ("ALSA: core: Fix card races between register and disconnect")
Signed-off-by: Jia Zhou 
Signed-off-by: Yi Wang 
Link: https://lore.kernel.org/r/1616989007-34429-1-git-send-email-wang.yi59@zte.com.cn
Signed-off-by: Takashi Iwai 
Signed-off-by: Sasha Levin

Merge tag 'asoc-v5.12' of https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound into for-linus

2021-02-17T20:16:27+00:00

ASoC: Updates for v5.12

Another quiet release in terms of features, though several of the
drivers got quite a bit of work and there were a lot of general changes
resulting from Morimoto-san's ongoing cleanup work.

 - As ever, lots of hard work by Morimoto-san cleaning up the code and
   making it more consistent.
 - Many improvements in the Intel drivers including a wide range of
   quirks and bug fixes.
 - A KUnit testsuite for the topology code.
 - Support for Ingenic JZ4760(B), Intel AlderLake-P, DT configured
   nVidia cards, Qualcomm lpass-rx-macro and lpass-tx-macro
 - Removal of obsolete SIRF prima/atlas, Txx9 and ZTE zx drivers.

Merge remote-tracking branch 'asoc/for-5.12' into asoc-linus

2021-02-17T18:52:26+00:00

ALSA: core - add missing compress device type to /proc/asound/devices

2021-02-10T09:31:53+00:00

Signed-off-by: Jaroslav Kysela 
Acked-by: Vinod Koul 
Link: https://lore.kernel.org/r/20210210083713.1034201-1-perex@perex.cz
Signed-off-by: Takashi Iwai

ALSA: pcm: Use for_each_pcm_substream() macro

2021-02-08T15:01:01+00:00

There are a few places doing the same loop iterating all PCM
substreams belonging to the PCM object.  Introduce a local helper
macro, for_each_pcm_substream(), to simplify the code.

Link: https://lore.kernel.org/r/20210206203656.15959-5-tiwai@suse.de
Signed-off-by: Takashi Iwai

ALSA: pcm: Don't call sync_stop if it hasn't been stopped

2021-02-08T15:00:49+00:00

The PCM stop operation sets the stop_operating flag for indicating the
sync_stop post-process.  This flag is, however, set unconditionally
even if the PCM trigger weren't issued.  This may lead to
inconsistency in the driver side.

Correct the code to set stop_operating flag only after the trigger
STOP is actually called.

Fixes: 1e850beea278 ("ALSA: pcm: Add the support for sync-stop operation")
Cc: 
Link: https://lore.kernel.org/r/20210206203656.15959-4-tiwai@suse.de
Signed-off-by: Takashi Iwai

ALSA: pcm: Assure sync with the pending stop operation at suspend

2021-02-08T15:00:39+00:00

The current PCM code calls the sync_stop at the resume action due to
the analogy to the PCM prepare call pattern.  But, it makes little
sense, as the sync should have been done rather at the suspend time,
not at the resume time.

This patch corrects the sync_stop call at suspend/resume to assure the
sync before finishing the suspend.

Fixes: 1e850beea278 ("ALSA: pcm: Add the support for sync-stop operation")
Cc: 
Link: https://lore.kernel.org/r/20210206203656.15959-3-tiwai@suse.de
Signed-off-by: Takashi Iwai

ALSA: pcm: Call sync_stop at disconnection

2021-02-08T15:00:28+00:00

The PCM core should perform the sync for the pending stop operations
at disconnection.  Otherwise it may lead to unexpected access.

Currently the old user of sync_stop, USB-audio driver, has its own
sync, so this isn't needed, but it's better to guarantee the sync in
the PCM core level.

This patch adds the missing sync_stop call at PCM disconnection
callback.  It also assures the IRQ sync if it's specified in the
card.  snd_pcm_sync_stop() is slightly modified to be called also for
any PCM substream object now.

Fixes: 1e850beea278 ("ALSA: pcm: Add the support for sync-stop operation")
Cc: 
Link: https://lore.kernel.org/r/20210206203656.15959-2-tiwai@suse.de
Signed-off-by: Takashi Iwai

ASoC: dmaengine_pcm: add peripheral configuration

2021-02-05T17:16:41+00:00

The commit e7bbb7acabf4 ("dmaengine: add peripheral configuration")
adds peripheral configuration for dma_slave_config.

This configuration is useful for some audio peripherals, for
example, the peripheral supports multi fifos, we can
let the DMA know which fifos are selected. So also add
this configuration for snd_dmaengine_dai_dma_data.

Signed-off-by: Shengjiu Wang 
Link: https://lore.kernel.org/r/1612509985-11063-1-git-send-email-shengjiu.wang@nxp.com
Signed-off-by: Mark Brown

ALSA: core: Fix the debugfs removal at snd_card_free()

2021-02-02T22:57:30+00:00

The commit 2d670ea2bd53 ("ALSA: jack: implement software jack
injection via debugfs") introduced a debugfs root for each sound card
object.  The debugfs entry gets removed at the card removal, too, but
it turned out that debugfs_remove() is called at a wrong place; it's
after the card object gets freed, hence it leads to use-after-free.

Fix it by moving the debugfs_remove() at the right place, the
destructor of the card device.

Fixes: 2d670ea2bd53 ("ALSA: jack: implement software jack injection via debugfs")
Reported-and-tested-by: Chris Wilson 
Link: https://lore.kernel.org/r/161228343605.1150.8862281636043446562@build.alporthouse.com
Link: https://lore.kernel.org/r/20210202225629.1965-1-tiwai@suse.de
Signed-off-by: Takashi Iwai