kernel/linux.git/drivers/soundwire/slave.c, branch v5.10.257

soundwire: sysfs: add slave status and device number before probe

2020-09-28T05:47:43+00:00

The MIPI DisCo device properties that are read by the driver from platform firmware, or hard-coded in the driver, should only be provided as sysfs entries when a driver probes successfully. However the device status and device number is updated even when there is no driver present, and hence can be updated when a Slave device is detected on the bus without being described in platform firmware and without any driver registered/probed. As suggested by GregKH, the attribute group for Slave status and device number is is added by default upon device registration. Credits to Vinod Koul for the status_show() function, shared in a separate patch and used as is here. The status table was modified to remove an unnecessary enum and status_show() is handled in a different group attribute than what was suggested by Vinod. Tested-by: Srinivas Kandgatla Signed-off-by: Pierre-Louis Bossart Co-developed-by: Vinod Koul Reviewed-by: Greg Kroah-Hartman Link: https://lore.kernel.org/r/20200924194430.121058-3-pierre-louis.bossart@linux.intel.com Signed-off-by: Vinod Koul

soundwire: bus: add enumerated Slave device to device list

2020-09-28T05:47:40+00:00

Currently Slave devices are only added on startup, either from Device Tree or ACPI entries. However Slave devices that are physically present on the bus, but not described in platform firmware, will never be added to the device list. The user/integrator can only know the list of devices by looking a dynamic debug logs. This patch suggests adding a Slave device even if there is no matching DT or ACPI entry, so that we can see this in sysfs entry. Initial code from Srinivas. Comments, fixes for ACPI probe and edits of commit message by Pierre. Signed-off-by: Srinivas Kandagatla Co-developed-by: Pierre-Louis Bossart Signed-off-by: Pierre-Louis Bossart Link: https://lore.kernel.org/r/20200924194430.121058-2-pierre-louis.bossart@linux.intel.com Signed-off-by: Vinod Koul

soundwire: slave: add first_interrupt_done status

2020-09-10T05:51:04+00:00

Some Slaves report incorrect information in their interrupt status registers after a master/bus reset, track the initial interrupt handling so that quirks can be introduced to filter out incorrect information while keeping interrupts enabled in steady state. Signed-off-by: Pierre-Louis Bossart Signed-off-by: Bard Liao Reviewed-by: Kai Vehmanen Reviewed-by: Guennadi Liakhovetski Link: https://lore.kernel.org/r/20200908134521.6781-4-yung-chuan.liao@linux.intel.com Signed-off-by: Vinod Koul

soundwire: slave: use SDW_DISCO_LINK_ID()

2020-09-04T09:16:42+00:00

use SDW_DISCO_LINK_ID() in slave code to extract field values instead of open coding masks and shift operations to extract link_id Signed-off-by: Vinod Koul Tested-by: Bard Liao Link: https://lore.kernel.org/r/20200903114504.1202143-4-vkoul@kernel.org Signed-off-by: Vinod Koul

soundwire: fix port_ready[] dynamic allocation in mipi_disco

2020-09-03T10:32:29+00:00

The existing code allocates memory for the total number of ports. This only works if the ports are contiguous, but will break if e.g. a Devices uses port0, 1, and 14. The port_ready[] array would contain 3 elements, which would lead to an out-of-bounds access. Conversely in other cases, the wrong port index would be used leading to timeouts on prepare. This can be fixed by allocating for the worst-case of 15 ports (DP0..DP14). In addition since the number is now fixed, we can use an array instead of a dynamic allocation. Signed-off-by: Pierre-Louis Bossart Signed-off-by: Bard Liao Reviewed-by: Rander Wang Reviewed-by: Guennadi Liakhovetski Link: https://lore.kernel.org/r/20200831134318.11443-4-yung-chuan.liao@linux.intel.com Signed-off-by: Vinod Koul

soundwire: bus_type: introduce sdw_slave_type and sdw_master_type

2020-05-19T07:14:34+00:00

this is a preparatory patch before the introduction of the sdw_master_type. The SoundWire slave support is slightly modified with the use of a sdw_slave_type, and the uevent handling move to slave.c (since it's not necessary for the master). No functionality change other than moving code around. Signed-off-by: Pierre-Louis Bossart Signed-off-by: Bard Liao Acked-by: Jaroslav Kysela Link: https://lore.kernel.org/r/20200518174322.31561-3-yung-chuan.liao@linux.intel.com Signed-off-by: Vinod Koul

soundwire: slave: don't init debugfs on device registration error

2020-05-05T03:21:44+00:00

The error handling flow seems incorrect, there is no reason to try and add debugfs support if the device registration did not succeed. Return on error. Signed-off-by: Pierre-Louis Bossart Signed-off-by: Bard Liao Reviewed-by: Ranjani Sridharan Reviewed-by: Guennadi Liakhovetski Link: https://lore.kernel.org/r/20200419185117.4233-2-yung-chuan.liao@linux.intel.com Signed-off-by: Vinod Koul

soundwire: bus: fix race condition with initialization_complete signaling

2020-02-25T10:27:02+00:00

Waiting for the enumeration to be complete may not be enough for a Slave driver, there is a possible race condition between resume operations and initializations handled in an interrupt thread, which can results in settings not being fully restored after system or pm_runtime resume. This patch builds on the changes added for enumeration_complete, init_completion() is called when the Slave device becomes UNATTACHED, as done with enumeration_complete. The difference with the enumeration_complete case is that complete() is signaled after the Slave device is fully initialized after the .update_status() callback is called. A Slave device driver can decide to wait on either of the two complete() cases, depending on its initialization code and requirements. Signed-off-by: Rander Wang Signed-off-by: Pierre-Louis Bossart Link: https://lore.kernel.org/r/20200115000844.14695-4-pierre-louis.bossart@linux.intel.com Signed-off-by: Vinod Koul

soundwire: bus: fix race condition with enumeration_complete signaling

2020-02-25T10:27:02+00:00

This patch adds the signaling needed for Slave drivers to wait until the enumeration completes so that race conditions when issuing read/write commands are avoided. The calls for wait_for_completion() will be added in codec drivers in follow-up patches. The order between init_completion() and complete() is deterministic, the Slave is marked as UNATTACHED either during a Master-initiated HardReset, or when the hardware detects the Slave no longer reports as ATTACHED. Signed-off-by: Pierre-Louis Bossart Link: https://lore.kernel.org/r/20200115000844.14695-3-pierre-louis.bossart@linux.intel.com Signed-off-by: Vinod Koul

soundwire: bus: fix race condition with probe_complete signaling

2020-02-25T10:27:01+00:00

The driver probe takes care of basic initialization and is invoked when a Slave becomes attached, after a match between the Slave DevID registers and ACPI/DT entries. The update_status callback is invoked when a Slave state changes, e.g. when it is assigned a non-zero Device Number and it reports with an ATTACHED/ALERT state. The state change detection is usually hardware-based and based on the SoundWire frame rate (e.g. double-digit microseconds) while the probe is a pure software operation, which may involve a kernel module load. In corner cases, it's possible that the state changes before the probe completes. This patch suggests the use of wait_for_completion to avoid races on startup, so that the update_status callback does not rely on invalid pointers/data structures. Signed-off-by: Rander Wang Signed-off-by: Pierre-Louis Bossart Link: https://lore.kernel.org/r/20200115000844.14695-2-pierre-louis.bossart@linux.intel.com Signed-off-by: Vinod Koul