Skip to content

L
linux
Switch branch/tag
  1. 07 Oct, 2021 1 commit
  3. 30 Sep, 2021 1 commit
  5. 27 Sep, 2021 2 commits
  7. 21 Sep, 2021 2 commits
  9. 17 Sep, 2021 2 commits
  11. 16 Sep, 2021 2 commits
  13. 15 Sep, 2021 3 commits
  15. 10 Sep, 2021 7 commits
  17. 09 Sep, 2021 3 commits
  19. 03 Sep, 2021 3 commits
  21. 02 Sep, 2021 1 commit
  23. 01 Sep, 2021 7 commits
  25. 30 Aug, 2021 3 commits
  27. 26 Aug, 2021 3 commits
    • Mark Brown's avatar
      Merge series "ASoC: wcd9335: Firx some resources leak in the probe and remove... · 52c64e5f
      Mark Brown authored 
      Merge series "ASoC: wcd9335: Firx some resources leak in the probe and remove function" from Christophe JAILLET <christophe.jaillet@wanadoo.fr>:

The first 2 patches are sraightforward and look logical to me.

However, the 3rd one in purely speculative. It is based on the fact that a
comment states that we enable some irqs on some slave ports. That said, it writes
0xFF in some registers.

So, I guess that we should disable these irqs when the driver is removed. That
said, writing 0x00 at the same place looks logical to me.

This cis untested and NOT based on any documentation. Just a blind fix.
Review with care.
You'll be warned :)

Christophe JAILLET (3):
  ASoC: wcd9335: Fix a double irq free in the remove function
  ASoC: wcd9335: Fix a memory leak in the error handling path of the
    probe function
  ASoC: wcd9335: Disable irq on slave ports in the remove function

 sound/soc/codecs/wcd9335.c | 23 +++++++++++++++++++++--
 1 file changed, 21 insertions(+), 2 deletions(-)

--
2.30.2
      52c64e5f
    • Pierre-Louis Bossart's avatar
      ASoC: soc-pcm: test refcount before triggering · 6479f758
      Pierre-Louis Bossart authored 
      On start/pause_release/resume, when more than one FE is connected to
the same BE, it's possible that the trigger is sent more than
once. This is not desirable, we only want to trigger a BE once, which
is straightforward to implement with a refcount.

For stop/pause/suspend, the problem is more complicated: the check
implemented in snd_soc_dpcm_can_be_free_stop() may fail due to a
conceptual deadlock when we trigger the BE before the FE. In this
case, the FE states have not yet changed, so there are corner cases
where the TRIGGER_STOP is never sent - the dual case of start where
multiple triggers might be sent.

This patch suggests an unconditional trigger in all cases, without
checking the FE states, using a refcount protected by a spinlock.
Signed-off-by: default avatarPierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Message-Id: <20210817164054.250028-3-pierre-louis.bossart@linux.intel.com>
Signed-off-by: default avatarMark Brown <broonie@kernel.org>
      6479f758
    • Pierre-Louis Bossart's avatar
      ASoC: soc-pcm: protect BE dailink state changes in trigger · 0c75fc71
      Pierre-Louis Bossart authored 
      When more than one FE is connected to a BE, e.g. in a mixing use case,
the BE can be triggered multiple times when the FE are opened/started
concurrently. This race condition is problematic in the case of
SoundWire BE dailinks, and this is not desirable in a general
case. The code carefully checks when the BE can be stopped or
hw_free'ed, but the trigger code does not use any mutual exclusion.

Fix by using the same spinlock already used to check FE states, and
set the state before the trigger. In case of errors,  the initial
state will be restored.

This patch does not change how the triggers are handled, it only makes
sure the states are handled in critical sections.
Signed-off-by: default avatarPierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Message-Id: <20210817164054.250028-2-pierre-louis.bossart@linux.intel.com>
Signed-off-by: default avatarMark Brown <broonie@kernel.org>
      0c75fc71
GitLab Nexedi Edition | About GitLab | About Nexedi | 沪ICP备2021021310号-2 | 沪ICP备2021021310号-7