diff options
Diffstat (limited to 'Documentation/driver-api/soundwire')
| -rw-r--r-- | Documentation/driver-api/soundwire/bra.rst | 336 | ||||
| -rw-r--r-- | Documentation/driver-api/soundwire/bra_cadence.rst | 66 | ||||
| -rw-r--r-- | Documentation/driver-api/soundwire/error_handling.rst | 65 | ||||
| -rw-r--r-- | Documentation/driver-api/soundwire/index.rst | 13 | ||||
| -rw-r--r-- | Documentation/driver-api/soundwire/locking.rst | 108 | ||||
| -rw-r--r-- | Documentation/driver-api/soundwire/stream.rst | 527 | ||||
| -rw-r--r-- | Documentation/driver-api/soundwire/summary.rst | 200 |
7 files changed, 1315 insertions, 0 deletions
diff --git a/Documentation/driver-api/soundwire/bra.rst b/Documentation/driver-api/soundwire/bra.rst new file mode 100644 index 000000000..c08ab2591 --- /dev/null +++ b/Documentation/driver-api/soundwire/bra.rst @@ -0,0 +1,336 @@ +========================== +Bulk Register Access (BRA) +========================== + +Conventions +----------- + +Capitalized words used in this documentation are intentional and refer +to concepts of the SoundWire 1.x specification. + +Introduction +------------ + +The SoundWire 1.x specification provides a mechanism to speed-up +command/control transfers by reclaiming parts of the audio +bandwidth. The Bulk Register Access (BRA) protocol is a standard +solution based on the Bulk Payload Transport (BPT) definitions. + +The regular control channel uses Column 0 and can only send/retrieve +one byte per frame with write/read commands. With a typical 48kHz +frame rate, only 48kB/s can be transferred. + +The optional Bulk Register Access capability can transmit up to 12 +Mbits/s and reduce transfer times by several orders of magnitude, but +has multiple design constraints: + + (1) Each frame can only support a read or a write transfer, with a + 10-byte overhead per frame (header and footer response). + + (2) The read/writes SHALL be from/to contiguous register addresses + in the same frame. A fragmented register space decreases the + efficiency of the protocol by requiring multiple BRA transfers + scheduled in different frames. + + (3) The targeted Peripheral device SHALL support the optional Data + Port 0, and likewise the Manager SHALL expose audio-like Ports + to insert BRA packets in the audio payload using the concepts of + Sample Interval, HSTART, HSTOP, etc. + + (4) The BRA transport efficiency depends on the available + bandwidth. If there are no on-going audio transfers, the entire + frame minus Column 0 can be reclaimed for BRA. The frame shape + also impacts efficiency: since Column0 cannot be used for + BTP/BRA, the frame should rely on a large number of columns and + minimize the number of rows. The bus clock should be as high as + possible. + + (5) The number of bits transferred per frame SHALL be a multiple of + 8 bits. Padding bits SHALL be inserted if necessary at the end + of the data. + + (6) The regular read/write commands can be issued in parallel with + BRA transfers. This is convenient to e.g. deal with alerts, jack + detection or change the volume during firmware download, but + accessing the same address with two independent protocols has to + be avoided to avoid undefined behavior. + + (7) Some implementations may not be capable of handling the + bandwidth of the BRA protocol, e.g. in the case of a slow I2C + bus behind the SoundWire IP. In this case, the transfers may + need to be spaced in time or flow-controlled. + + (8) Each BRA packet SHALL be marked as 'Active' when valid data is + to be transmitted. This allows for software to allocate a BRA + stream but not transmit/discard data while processing the + results or preparing the next batch of data, or allowing the + peripheral to deal with the previous transfer. In addition BRA + transfer can be started early on without data being ready. + + (9) Up to 470 bytes may be transmitted per frame. + + (10) The address is represented with 32 bits and does not rely on + the paging registers used for the regular command/control + protocol in Column 0. + + +Error checking +-------------- + +Firmware download is one of the key usages of the Bulk Register Access +protocol. To make sure the binary data integrity is not compromised by +transmission or programming errors, each BRA packet provides: + + (1) A CRC on the 7-byte header. This CRC helps the Peripheral Device + check if it is addressed and set the start address and number of + bytes. The Peripheral Device provides a response in Byte 7. + + (2) A CRC on the data block (header excluded). This CRC is + transmitted as the last-but-one byte in the packet, prior to the + footer response. + +The header response can be one of: + (a) Ack + (b) Nak + (c) Not Ready + +The footer response can be one of: + (1) Ack + (2) Nak (CRC failure) + (3) Good (operation completed) + (4) Bad (operation failed) + +Example frame +------------- + +The example below is not to scale and makes simplifying assumptions +for clarity. The different chunks in the BRA packets are not required +to start on a new SoundWire Row, and the scale of data may vary. + + :: + + +---+--------------------------------------------+ + + | | + + | BRA HEADER | + + | | + + +--------------------------------------------+ + + C | HEADER CRC | + + O +--------------------------------------------+ + + M | HEADER RESPONSE | + + M +--------------------------------------------+ + + A | | + + N | | + + D | DATA | + + | | + + | | + + | | + + +--------------------------------------------+ + + | DATA CRC | + + +--------------------------------------------+ + + | FOOTER RESPONSE | + +---+--------------------------------------------+ + + +Assuming the frame uses N columns, the configuration shown above can +be programmed by setting the DP0 registers as: + + - HSTART = 1 + - HSTOP = N - 1 + - Sampling Interval = N + - WordLength = N - 1 + +Addressing restrictions +----------------------- + +The Device Number specified in the Header follows the SoundWire +definitions, and broadcast and group addressing are permitted. For now +the Linux implementation only allows for a single BPT transfer to a +single device at a time. This might be revisited at a later point as +an optimization to send the same firmware to multiple devices, but +this would only be beneficial for single-link solutions. + +In the case of multiple Peripheral devices attached to different +Managers, the broadcast and group addressing is not supported by the +SoundWire specification. Each device must be handled with separate BRA +streams, possibly in parallel - the links are really independent. + +Unsupported features +-------------------- + +The Bulk Register Access specification provides a number of +capabilities that are not supported in known implementations, such as: + + (1) Transfers initiated by a Peripheral Device. The BRA Initiator is + always the Manager Device. + + (2) Flow-control capabilities and retransmission based on the + 'NotReady' header response require extra buffering in the + SoundWire IP and are not implemented. + +Bi-directional handling +----------------------- + +The BRA protocol can handle writes as well as reads, and in each +packet the header and footer response are provided by the Peripheral +Target device. On the Peripheral device, the BRA protocol is handled +by a single DP0 data port, and at the low-level the bus ownership can +will change for header/footer response as well as the data transmitted +during a read. + +On the host side, most implementations rely on a Port-like concept, +with two FIFOs consuming/generating data transfers in parallel +(Host->Peripheral and Peripheral->Host). The amount of data +consumed/produced by these FIFOs is not symmetrical, as a result +hardware typically inserts markers to help software and hardware +interpret raw data + +Each packet will typically have: + + (1) a 'Start of Packet' indicator. + + (2) an 'End of Packet' indicator. + + (3) a packet identifier to correlate the data requested and + transmitted, and the error status for each frame + +Hardware implementations can check errors at the frame level, and +retry a transfer in case of errors. However, as for the flow-control +case, this requires extra buffering and intelligence in the +hardware. The Linux support assumes that the entire transfer is +cancelled if a single error is detected in one of the responses. + +Abstraction required +~~~~~~~~~~~~~~~~~~~~ + +There are no standard registers or mandatory implementation at the +Manager level, so the low-level BPT/BRA details must be hidden in +Manager-specific code. For example the Cadence IP format above is not +known to the codec drivers. + +Likewise, codec drivers should not have to know the frame size. The +computation of CRC and handling of responses is handled in helpers and +Manager-specific code. + +The host BRA driver may also have restrictions on pages allocated for +DMA, or other host-DSP communication protocols. The codec driver +should not be aware of any of these restrictions, since it might be +reused in combination with different implementations of Manager IPs. + +Concurrency between BRA and regular read/write +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + +The existing 'nread/nwrite' API already relies on a notion of start +address and number of bytes, so it would be possible to extend this +API with a 'hint' requesting BPT/BRA be used. + +However BRA transfers could be quite long, and the use of a single +mutex for regular read/write and BRA is a show-stopper. Independent +operation of the control/command and BRA transfers is a fundamental +requirement, e.g. to change the volume level with the existing regmap +interface while downloading firmware. The integration must however +ensure that there are no concurrent access to the same address with +the command/control protocol and the BRA protocol. + +In addition, the 'sdw_msg' structure hard-codes support for 16-bit +addresses and paging registers which are irrelevant for BPT/BRA +support based on native 32-bit addresses. A separate API with +'sdw_bpt_msg' makes more sense. + +One possible strategy to speed-up all initialization tasks would be to +start a BRA transfer for firmware download, then deal with all the +"regular" read/writes in parallel with the command channel, and last +to wait for the BRA transfers to complete. This would allow for a +degree of overlap instead of a purely sequential solution. As such, +the BRA API must support async transfers and expose a separate wait +function. + + +Peripheral/bus interface +------------------------ + +The bus interface for BPT/BRA is made of two functions: + + - sdw_bpt_send_async(bpt_message) + + This function sends the data using the Manager + implementation-defined capabilities (typically DMA or IPC + protocol). + + Queueing is currently not supported, the caller + needs to wait for completion of the requested transfer. + + - sdw_bpt_wait() + + This function waits for the entire message provided by the + codec driver in the 'send_async' stage. Intermediate status for + smaller chunks will not be provided back to the codec driver, + only a return code will be provided. + +Regmap use +~~~~~~~~~~ + +Existing codec drivers rely on regmap to download firmware to +Peripherals. regmap exposes an async interface similar to the +send/wait API suggested above, so at a high-level it would seem +natural to combine BRA and regmap. The regmap layer could check if BRA +is available or not, and use a regular read-write command channel in +the latter case. + +The regmap integration will be handled in a second step. + +BRA stream model +---------------- + +For regular audio transfers, the machine driver exposes a dailink +connecting CPU DAI(s) and Codec DAI(s). + +This model is not required BRA support: + + (1) The SoundWire DAIs are mainly wrappers for SoundWire Data + Ports, with possibly some analog or audio conversion + capabilities bolted behind the Data Port. In the context of + BRA, the DP0 is the destination. DP0 registers are standard and + can be programmed blindly without knowing what Peripheral is + connected to each link. In addition, if there are multiple + Peripherals on a link and some of them do not support DP0, the + write commands to program DP0 registers will generate harmless + COMMAND_IGNORED responses that will be wired-ORed with + responses from Peripherals which support DP0. In other words, + the DP0 programming can be done with broadcast commands, and + the information on the Target device can be added only in the + BRA Header. + + (2) At the CPU level, the DAI concept is not useful for BRA; the + machine driver will not create a dailink relying on DP0. The + only concept that is needed is the notion of port. + + (3) The stream concept relies on a set of master_rt and slave_rt + concepts. All of these entities represent ports and not DAIs. + + (4) With the assumption that a single BRA stream is used per link, + that stream can connect master ports as well as all peripheral + DP0 ports. + + (5) BRA transfers only make sense in the context of one + Manager/Link, so the BRA stream handling does not rely on the + concept of multi-link aggregation allowed by regular DAI links. + +Audio DMA support +----------------- + +Some DMAs, such as HDaudio, require an audio format field to be +set. This format is in turn used to define acceptable bursts. BPT/BRA +support is not fully compatible with these definitions in that the +format and bandwidth may vary between read and write commands. + +In addition, on Intel HDaudio Intel platforms the DMAs need to be +programmed with a PCM format matching the bandwidth of the BPT/BRA +transfer. The format is based on 192kHz 32-bit samples, and the number +of channels varies to adjust the bandwidth. The notion of channel is +completely notional since the data is not typical audio +PCM. Programming such channels helps reserve enough bandwidth and adjust +FIFO sizes to avoid xruns. + +Alignment requirements are currently not enforced at the core level +but at the platform-level, e.g. for Intel the data sizes must be +equal to or larger than 16 bytes. diff --git a/Documentation/driver-api/soundwire/bra_cadence.rst b/Documentation/driver-api/soundwire/bra_cadence.rst new file mode 100644 index 000000000..4560752e8 --- /dev/null +++ b/Documentation/driver-api/soundwire/bra_cadence.rst @@ -0,0 +1,66 @@ +Cadence IP BRA support +---------------------- + +Format requirements +~~~~~~~~~~~~~~~~~~~ + +The Cadence IP relies on PDI0 for TX and PDI1 for RX. The data needs +to be formatted with the following conventions: + + (1) all Data is stored in bits 15..0 of the 32-bit PDI FIFOs. + + (2) the start of packet is BIT(31). + + (3) the end of packet is BIT(30). + + (4) A packet ID is stored in bits 19..16. This packet ID is + determined by software and is typically a rolling counter. + + (5) Padding shall be inserted as needed so that the Header CRC, + Header response, Footer CRC, Footer response are always in + Byte0. Padding is inserted by software for writes, and on reads + software shall discard the padding added by the hardware. + +Example format +~~~~~~~~~~~~~~ + +The following table represents the sequence provided to PDI0 for a +write command followed by a read command. + +:: + + +---+---+--------+---------------+---------------+ + + 1 | 0 | ID = 0 | WR HDR[1] | WR HDR[0] | + + | | | WR HDR[3] | WR HDR[2] | + + | | | WR HDR[5] | WR HDR[4] | + + | | | pad | WR HDR CRC | + + | | | WR Data[1] | WR Data[0] | + + | | | WR Data[3] | WR Data[2] | + + | | | WR Data[n-2] | WR Data[n-3] | + + | | | pad | WR Data[n-1] | + + 0 | 1 | | pad | WR Data CRC | + +---+---+--------+---------------+---------------+ + + 1 | 0 | ID = 1 | RD HDR[1] | RD HDR[0] | + + | | | RD HDR[3] | RD HDR[2] | + + | | | RD HDR[5] | RD HDR[4] | + + 0 | 1 | | pad | RD HDR CRC | + +---+---+--------+---------------+---------------+ + + +The table below represents the data received on PDI1 for the same +write command followed by a read command. + +:: + + +---+---+--------+---------------+---------------+ + + 1 | 0 | ID = 0 | pad | WR Hdr Rsp | + + 0 | 1 | | pad | WR Ftr Rsp | + +---+---+--------+---------------+---------------+ + + 1 | 0 | ID = 0 | pad | Rd Hdr Rsp | + + | | | RD Data[1] | RD Data[0] | + + | | | RD Data[3] | RD Data[2] | + + | | | RD HDR[n-2] | RD Data[n-3] | + + | | | pad | RD Data[n-1] | + + | | | pad | RD Data CRC | + + 0 | 1 | | pad | RD Ftr Rsp | + +---+---+--------+---------------+---------------+ diff --git a/Documentation/driver-api/soundwire/error_handling.rst b/Documentation/driver-api/soundwire/error_handling.rst new file mode 100644 index 000000000..aa3a0a23a --- /dev/null +++ b/Documentation/driver-api/soundwire/error_handling.rst @@ -0,0 +1,65 @@ +======================== +SoundWire Error Handling +======================== + +The SoundWire PHY was designed with care and errors on the bus are going to +be very unlikely, and if they happen it should be limited to single bit +errors. Examples of this design can be found in the synchronization +mechanism (sync loss after two errors) and short CRCs used for the Bulk +Register Access. + +The errors can be detected with multiple mechanisms: + +1. Bus clash or parity errors: This mechanism relies on low-level detectors + that are independent of the payload and usages, and they cover both control + and audio data. The current implementation only logs such errors. + Improvements could be invalidating an entire programming sequence and + restarting from a known position. In the case of such errors outside of a + control/command sequence, there is no concealment or recovery for audio + data enabled by the SoundWire protocol, the location of the error will also + impact its audibility (most-significant bits will be more impacted in PCM), + and after a number of such errors are detected the bus might be reset. Note + that bus clashes due to programming errors (two streams using the same bit + slots) or electrical issues during the transmit/receive transition cannot + be distinguished, although a recurring bus clash when audio is enabled is a + indication of a bus allocation issue. The interrupt mechanism can also help + identify Slaves which detected a Bus Clash or a Parity Error, but they may + not be responsible for the errors so resetting them individually is not a + viable recovery strategy. + +2. Command status: Each command is associated with a status, which only + covers transmission of the data between devices. The ACK status indicates + that the command was received and will be executed by the end of the + current frame. A NAK indicates that the command was in error and will not + be applied. In case of a bad programming (command sent to non-existent + Slave or to a non-implemented register) or electrical issue, no response + signals the command was ignored. Some Master implementations allow for a + command to be retransmitted several times. If the retransmission fails, + backtracking and restarting the entire programming sequence might be a + solution. Alternatively some implementations might directly issue a bus + reset and re-enumerate all devices. + +3. Timeouts: In a number of cases such as ChannelPrepare or + ClockStopPrepare, the bus driver is supposed to poll a register field until + it transitions to a NotFinished value of zero. The MIPI SoundWire spec 1.1 + does not define timeouts but the MIPI SoundWire DisCo document adds + recommendation on timeouts. If such configurations do not complete, the + driver will return a -ETIMEOUT. Such timeouts are symptoms of a faulty + Slave device and are likely impossible to recover from. + +Errors during global reconfiguration sequences are extremely difficult to +handle: + +1. BankSwitch: An error during the last command issuing a BankSwitch is + difficult to backtrack from. Retransmitting the Bank Switch command may be + possible in a single segment setup, but this can lead to synchronization + problems when enabling multiple bus segments (a command with side effects + such as frame reconfiguration would be handled at different times). A global + hard-reset might be the best solution. + +Note that SoundWire does not provide a mechanism to detect illegal values +written in valid registers. In a number of cases the standard even mentions +that the Slave might behave in implementation-defined ways. The bus +implementation does not provide a recovery mechanism for such errors, Slave +or Master driver implementers are responsible for writing valid values in +valid registers and implement additional range checking if needed. diff --git a/Documentation/driver-api/soundwire/index.rst b/Documentation/driver-api/soundwire/index.rst new file mode 100644 index 000000000..f7abf4a95 --- /dev/null +++ b/Documentation/driver-api/soundwire/index.rst @@ -0,0 +1,13 @@ +======================= +SoundWire Documentation +======================= + +.. toctree:: + :maxdepth: 1 + + summary + stream + error_handling + locking + bra + bra_cadence diff --git a/Documentation/driver-api/soundwire/locking.rst b/Documentation/driver-api/soundwire/locking.rst new file mode 100644 index 000000000..3a7ffb3d8 --- /dev/null +++ b/Documentation/driver-api/soundwire/locking.rst @@ -0,0 +1,108 @@ +================= +SoundWire Locking +================= + +This document explains locking mechanism of the SoundWire Bus. Bus uses +following locks in order to avoid race conditions in Bus operations on +shared resources. + + - Bus lock + + - Message lock + +Bus lock +======== + +SoundWire Bus lock is a mutex and is part of Bus data structure +(sdw_bus) which is used for every Bus instance. This lock is used to +serialize each of the following operations(s) within SoundWire Bus instance. + + - Addition and removal of Slave(s), changing Slave status. + + - Prepare, Enable, Disable and De-prepare stream operations. + + - Access of Stream data structure. + +Message lock +============ + +SoundWire message transfer lock. This mutex is part of +Bus data structure (sdw_bus). This lock is used to serialize the message +transfers (read/write) within a SoundWire Bus instance. + +Below examples show how locks are acquired. + +Example 1 +--------- + +Message transfer. + + 1. For every message transfer + + a. Acquire Message lock. + + b. Transfer message (Read/Write) to Slave1 or broadcast message on + Bus in case of bank switch. + + c. Release Message lock + + :: + + +----------+ +---------+ + | | | | + | Bus | | Master | + | | | Driver | + | | | | + +----+-----+ +----+----+ + | | + | bus->ops->xfer_msg() | + <-------------------------------+ a. Acquire Message lock + | | b. Transfer message + | | + +-------------------------------> c. Release Message lock + | return success/error | d. Return success/error + | | + + + + +Example 2 +--------- + +Prepare operation. + + 1. Acquire lock for Bus instance associated with Master 1. + + 2. For every message transfer in Prepare operation + + a. Acquire Message lock. + + b. Transfer message (Read/Write) to Slave1 or broadcast message on + Bus in case of bank switch. + + c. Release Message lock. + + 3. Release lock for Bus instance associated with Master 1 :: + + +----------+ +---------+ + | | | | + | Bus | | Master | + | | | Driver | + | | | | + +----+-----+ +----+----+ + | | + | sdw_prepare_stream() | + <-------------------------------+ 1. Acquire bus lock + | | 2. Perform stream prepare + | | + | | + | bus->ops->xfer_msg() | + <-------------------------------+ a. Acquire Message lock + | | b. Transfer message + | | + +-------------------------------> c. Release Message lock + | return success/error | d. Return success/error + | | + | | + | return success/error | 3. Release bus lock + +-------------------------------> 4. Return success/error + | | + + + diff --git a/Documentation/driver-api/soundwire/stream.rst b/Documentation/driver-api/soundwire/stream.rst new file mode 100644 index 000000000..d66201299 --- /dev/null +++ b/Documentation/driver-api/soundwire/stream.rst @@ -0,0 +1,527 @@ +========================= +Audio Stream in SoundWire +========================= + +An audio stream is a logical or virtual connection created between + + (1) System memory buffer(s) and Codec(s) + + (2) DSP memory buffer(s) and Codec(s) + + (3) FIFO(s) and Codec(s) + + (4) Codec(s) and Codec(s) + +which is typically driven by a DMA(s) channel through the data link. An +audio stream contains one or more channels of data. All channels within +stream must have same sample rate and same sample size. + +Assume a stream with two channels (Left & Right) is opened using SoundWire +interface. Below are some ways a stream can be represented in SoundWire. + +Stream Sample in memory (System memory, DSP memory or FIFOs) :: + + ------------------------- + | L | R | L | R | L | R | + ------------------------- + +Example 1: Stereo Stream with L and R channels is rendered from Master to +Slave. Both Master and Slave is using single port. :: + + +---------------+ Clock Signal +---------------+ + | Master +----------------------------------+ Slave | + | Interface | | Interface | + | | | 1 | + | | Data Signal | | + | L + R +----------------------------------+ L + R | + | (Data) | Data Direction | (Data) | + +---------------+ +-----------------------> +---------------+ + + +Example 2: Stereo Stream with L and R channels is captured from Slave to +Master. Both Master and Slave is using single port. :: + + + +---------------+ Clock Signal +---------------+ + | Master +----------------------------------+ Slave | + | Interface | | Interface | + | | | 1 | + | | Data Signal | | + | L + R +----------------------------------+ L + R | + | (Data) | Data Direction | (Data) | + +---------------+ <-----------------------+ +---------------+ + + +Example 3: Stereo Stream with L and R channels is rendered by Master. Each +of the L and R channel is received by two different Slaves. Master and both +Slaves are using single port. :: + + +---------------+ Clock Signal +---------------+ + | Master +---------+------------------------+ Slave | + | Interface | | | Interface | + | | | | 1 | + | | | Data Signal | | + | L + R +---+------------------------------+ L | + | (Data) | | | Data Direction | (Data) | + +---------------+ | | +-------------> +---------------+ + | | + | | + | | +---------------+ + | +----------------------> | Slave | + | | Interface | + | | 2 | + | | | + +----------------------------> | R | + | (Data) | + +---------------+ + +Example 4: Stereo Stream with L and R channels is rendered by +Master. Both of the L and R channels are received by two different +Slaves. Master and both Slaves are using single port handling +L+R. Each Slave device processes the L + R data locally, typically +based on static configuration or dynamic orientation, and may drive +one or more speakers. :: + + +---------------+ Clock Signal +---------------+ + | Master +---------+------------------------+ Slave | + | Interface | | | Interface | + | | | | 1 | + | | | Data Signal | | + | L + R +---+------------------------------+ L + R | + | (Data) | | | Data Direction | (Data) | + +---------------+ | | +-------------> +---------------+ + | | + | | + | | +---------------+ + | +----------------------> | Slave | + | | Interface | + | | 2 | + | | | + +----------------------------> | L + R | + | (Data) | + +---------------+ + +Example 5: Stereo Stream with L and R channel is rendered by two different +Ports of the Master and is received by only single Port of the Slave +interface. :: + + +--------------------+ + | | + | +--------------+ +----------------+ + | | || | | + | | Data Port || L Channel | | + | | 1 |------------+ | | + | | L Channel || | +-----+----+ | + | | (Data) || | L + R Channel || Data | | + | Master +----------+ | +---+---------> || Port | | + | Interface | | || 1 | | + | +--------------+ | || | | + | | || | +----------+ | + | | Data Port |------------+ | | + | | 2 || R Channel | Slave | + | | R Channel || | Interface | + | | (Data) || | 1 | + | +--------------+ Clock Signal | L + R | + | +---------------------------> | (Data) | + +--------------------+ | | + +----------------+ + +Example 6: Stereo Stream with L and R channel is rendered by 2 Masters, each +rendering one channel, and is received by two different Slaves, each +receiving one channel. Both Masters and both Slaves are using single port. :: + + +---------------+ Clock Signal +---------------+ + | Master +----------------------------------+ Slave | + | Interface | | Interface | + | 1 | | 1 | + | | Data Signal | | + | L +----------------------------------+ L | + | (Data) | Data Direction | (Data) | + +---------------+ +-----------------------> +---------------+ + + +---------------+ Clock Signal +---------------+ + | Master +----------------------------------+ Slave | + | Interface | | Interface | + | 2 | | 2 | + | | Data Signal | | + | R +----------------------------------+ R | + | (Data) | Data Direction | (Data) | + +---------------+ +-----------------------> +---------------+ + +Example 7: Stereo Stream with L and R channel is rendered by 2 +Masters, each rendering both channels. Each Slave receives L + R. This +is the same application as Example 4 but with Slaves placed on +separate links. :: + + +---------------+ Clock Signal +---------------+ + | Master +----------------------------------+ Slave | + | Interface | | Interface | + | 1 | | 1 | + | | Data Signal | | + | L + R +----------------------------------+ L + R | + | (Data) | Data Direction | (Data) | + +---------------+ +-----------------------> +---------------+ + + +---------------+ Clock Signal +---------------+ + | Master +----------------------------------+ Slave | + | Interface | | Interface | + | 2 | | 2 | + | | Data Signal | | + | L + R +----------------------------------+ L + R | + | (Data) | Data Direction | (Data) | + +---------------+ +-----------------------> +---------------+ + +Example 8: 4-channel Stream is rendered by 2 Masters, each rendering a +2 channels. Each Slave receives 2 channels. :: + + +---------------+ Clock Signal +---------------+ + | Master +----------------------------------+ Slave | + | Interface | | Interface | + | 1 | | 1 | + | | Data Signal | | + | L1 + R1 +----------------------------------+ L1 + R1 | + | (Data) | Data Direction | (Data) | + +---------------+ +-----------------------> +---------------+ + + +---------------+ Clock Signal +---------------+ + | Master +----------------------------------+ Slave | + | Interface | | Interface | + | 2 | | 2 | + | | Data Signal | | + | L2 + R2 +----------------------------------+ L2 + R2 | + | (Data) | Data Direction | (Data) | + +---------------+ +-----------------------> +---------------+ + +Note1: In multi-link cases like above, to lock, one would acquire a global +lock and then go on locking bus instances. But, in this case the caller +framework(ASoC DPCM) guarantees that stream operations on a card are +always serialized. So, there is no race condition and hence no need for +global lock. + +Note2: A Slave device may be configured to receive all channels +transmitted on a link for a given Stream (Example 4) or just a subset +of the data (Example 3). The configuration of the Slave device is not +handled by a SoundWire subsystem API, but instead by the +snd_soc_dai_set_tdm_slot() API. The platform or machine driver will +typically configure which of the slots are used. For Example 4, the +same slots would be used by all Devices, while for Example 3 the Slave +Device1 would use e.g. Slot 0 and Slave device2 slot 1. + +Note3: Multiple Sink ports can extract the same information for the +same bitSlots in the SoundWire frame, however multiple Source ports +shall be configured with different bitSlot configurations. This is the +same limitation as with I2S/PCM TDM usages. + +SoundWire Stream Management flow +================================ + +Stream definitions +------------------ + + (1) Current stream: This is classified as the stream on which operation has + to be performed like prepare, enable, disable, de-prepare etc. + + (2) Active stream: This is classified as the stream which is already active + on Bus other than current stream. There can be multiple active streams + on the Bus. + +SoundWire Bus manages stream operations for each stream getting +rendered/captured on the SoundWire Bus. This section explains Bus operations +done for each of the stream allocated/released on Bus. Following are the +stream states maintained by the Bus for each of the audio stream. + + +SoundWire stream states +----------------------- + +Below shows the SoundWire stream states and state transition diagram. :: + + +-----------+ +------------+ +----------+ +----------+ + | ALLOCATED +---->| CONFIGURED +---->| PREPARED +---->| ENABLED | + | STATE | | STATE | | STATE | | STATE | + +-----------+ +------------+ +---+--+---+ +----+-----+ + ^ ^ ^ + | | | + __| |___________ | + | | | + v | v + +----------+ +-----+------+ +-+--+-----+ + | RELEASED |<----------+ DEPREPARED |<-------+ DISABLED | + | STATE | | STATE | | STATE | + +----------+ +------------+ +----------+ + +NOTE: State transitions between ``SDW_STREAM_ENABLED`` and +``SDW_STREAM_DISABLED`` are only relevant when then INFO_PAUSE flag is +supported at the ALSA/ASoC level. Likewise the transition between +``SDW_DISABLED_STATE`` and ``SDW_PREPARED_STATE`` depends on the +INFO_RESUME flag. + +NOTE2: The framework implements basic state transition checks, but +does not e.g. check if a transition from DISABLED to ENABLED is valid +on a specific platform. Such tests need to be added at the ALSA/ASoC +level. + +Stream State Operations +----------------------- + +Below section explains the operations done by the Bus on Master(s) and +Slave(s) as part of stream state transitions. + +SDW_STREAM_ALLOCATED +~~~~~~~~~~~~~~~~~~~~ + +Allocation state for stream. This is the entry state +of the stream. Operations performed before entering in this state: + + (1) A stream runtime is allocated for the stream. This stream + runtime is used as a reference for all the operations performed + on the stream. + + (2) The resources required for holding stream runtime information are + allocated and initialized. This holds all stream related information + such as stream type (PCM/PDM) and parameters, Master and Slave + interface associated with the stream, stream state etc. + +After all above operations are successful, stream state is set to +``SDW_STREAM_ALLOCATED``. + +Bus implements below API for allocate a stream which needs to be called once +per stream. From ASoC DPCM framework, this stream state maybe linked to +.startup() operation. + +.. code-block:: c + + int sdw_alloc_stream(char * stream_name, enum sdw_stream_type type); + +The SoundWire core provides a sdw_startup_stream() helper function, +typically called during a dailink .startup() callback, which performs +stream allocation and sets the stream pointer for all DAIs +connected to a stream. + +SDW_STREAM_CONFIGURED +~~~~~~~~~~~~~~~~~~~~~ + +Configuration state of stream. Operations performed before entering in +this state: + + (1) The resources allocated for stream information in SDW_STREAM_ALLOCATED + state are updated here. This includes stream parameters, Master(s) + and Slave(s) runtime information associated with current stream. + + (2) All the Master(s) and Slave(s) associated with current stream provide + the port information to Bus which includes port numbers allocated by + Master(s) and Slave(s) for current stream and their channel mask. + +After all above operations are successful, stream state is set to +``SDW_STREAM_CONFIGURED``. + +Bus implements below APIs for CONFIG state which needs to be called by +the respective Master(s) and Slave(s) associated with stream. These APIs can +only be invoked once by respective Master(s) and Slave(s). From ASoC DPCM +framework, this stream state is linked to .hw_params() operation. + +.. code-block:: c + + int sdw_stream_add_master(struct sdw_bus * bus, + struct sdw_stream_config * stream_config, + const struct sdw_ports_config * ports_config, + struct sdw_stream_runtime * stream); + + int sdw_stream_add_slave(struct sdw_slave * slave, + struct sdw_stream_config * stream_config, + const struct sdw_ports_config * ports_config, + struct sdw_stream_runtime * stream); + + +SDW_STREAM_PREPARED +~~~~~~~~~~~~~~~~~~~ + +Prepare state of stream. Operations performed before entering in this state: + + (0) Steps 1 and 2 are omitted in the case of a resume operation, + where the bus bandwidth is known. + + (1) Bus parameters such as bandwidth, frame shape, clock frequency, + are computed based on current stream as well as already active + stream(s) on Bus. Re-computation is required to accommodate current + stream on the Bus. + + (2) Transport and port parameters of all Master(s) and Slave(s) port(s) are + computed for the current as well as already active stream based on frame + shape and clock frequency computed in step 1. + + (3) Computed Bus and transport parameters are programmed in Master(s) and + Slave(s) registers. The banked registers programming is done on the + alternate bank (bank currently unused). Port(s) are enabled for the + already active stream(s) on the alternate bank (bank currently unused). + This is done in order to not disrupt already active stream(s). + + (4) Once all the values are programmed, Bus initiates switch to alternate + bank where all new values programmed gets into effect. + + (5) Ports of Master(s) and Slave(s) for current stream are prepared by + programming PrepareCtrl register. + +After all above operations are successful, stream state is set to +``SDW_STREAM_PREPARED``. + +Bus implements below API for PREPARE state which needs to be called +once per stream. From ASoC DPCM framework, this stream state is linked +to .prepare() operation. Since the .trigger() operations may not +follow the .prepare(), a direct transition from +``SDW_STREAM_PREPARED`` to ``SDW_STREAM_DEPREPARED`` is allowed. + +.. code-block:: c + + int sdw_prepare_stream(struct sdw_stream_runtime * stream); + + +SDW_STREAM_ENABLED +~~~~~~~~~~~~~~~~~~ + +Enable state of stream. The data port(s) are enabled upon entering this state. +Operations performed before entering in this state: + + (1) All the values computed in SDW_STREAM_PREPARED state are programmed + in alternate bank (bank currently unused). It includes programming of + already active stream(s) as well. + + (2) All the Master(s) and Slave(s) port(s) for the current stream are + enabled on alternate bank (bank currently unused) by programming + ChannelEn register. + + (3) Once all the values are programmed, Bus initiates switch to alternate + bank where all new values programmed gets into effect and port(s) + associated with current stream are enabled. + +After all above operations are successful, stream state is set to +``SDW_STREAM_ENABLED``. + +Bus implements below API for ENABLE state which needs to be called once per +stream. From ASoC DPCM framework, this stream state is linked to +.trigger() start operation. + +.. code-block:: c + + int sdw_enable_stream(struct sdw_stream_runtime * stream); + +SDW_STREAM_DISABLED +~~~~~~~~~~~~~~~~~~~ + +Disable state of stream. The data port(s) are disabled upon exiting this state. +Operations performed before entering in this state: + + (1) All the Master(s) and Slave(s) port(s) for the current stream are + disabled on alternate bank (bank currently unused) by programming + ChannelEn register. + + (2) All the current configuration of Bus and active stream(s) are programmed + into alternate bank (bank currently unused). + + (3) Once all the values are programmed, Bus initiates switch to alternate + bank where all new values programmed gets into effect and port(s) associated + with current stream are disabled. + +After all above operations are successful, stream state is set to +``SDW_STREAM_DISABLED``. + +Bus implements below API for DISABLED state which needs to be called once +per stream. From ASoC DPCM framework, this stream state is linked to +.trigger() stop operation. + +When the INFO_PAUSE flag is supported, a direct transition to +``SDW_STREAM_ENABLED`` is allowed. + +For resume operations where ASoC will use the .prepare() callback, the +stream can transition from ``SDW_STREAM_DISABLED`` to +``SDW_STREAM_PREPARED``, with all required settings restored but +without updating the bandwidth and bit allocation. + +.. code-block:: c + + int sdw_disable_stream(struct sdw_stream_runtime * stream); + + +SDW_STREAM_DEPREPARED +~~~~~~~~~~~~~~~~~~~~~ + +De-prepare state of stream. Operations performed before entering in this +state: + + (1) All the port(s) of Master(s) and Slave(s) for current stream are + de-prepared by programming PrepareCtrl register. + + (2) The payload bandwidth of current stream is reduced from the total + bandwidth requirement of bus and new parameters calculated and + applied by performing bank switch etc. + +After all above operations are successful, stream state is set to +``SDW_STREAM_DEPREPARED``. + +Bus implements below API for DEPREPARED state which needs to be called +once per stream. ALSA/ASoC do not have a concept of 'deprepare', and +the mapping from this stream state to ALSA/ASoC operation may be +implementation specific. + +When the INFO_PAUSE flag is supported, the stream state is linked to +the .hw_free() operation - the stream is not deprepared on a +TRIGGER_STOP. + +Other implementations may transition to the ``SDW_STREAM_DEPREPARED`` +state on TRIGGER_STOP, should they require a transition through the +``SDW_STREAM_PREPARED`` state. + +.. code-block:: c + + int sdw_deprepare_stream(struct sdw_stream_runtime * stream); + + +SDW_STREAM_RELEASED +~~~~~~~~~~~~~~~~~~~ + +Release state of stream. Operations performed before entering in this state: + + (1) Release port resources for all Master(s) and Slave(s) port(s) + associated with current stream. + + (2) Release Master(s) and Slave(s) runtime resources associated with + current stream. + + (3) Release stream runtime resources associated with current stream. + +After all above operations are successful, stream state is set to +``SDW_STREAM_RELEASED``. + +Bus implements below APIs for RELEASE state which needs to be called by +all the Master(s) and Slave(s) associated with stream. From ASoC DPCM +framework, this stream state is linked to .hw_free() operation. + +.. code-block:: c + + int sdw_stream_remove_master(struct sdw_bus * bus, + struct sdw_stream_runtime * stream); + int sdw_stream_remove_slave(struct sdw_slave * slave, + struct sdw_stream_runtime * stream); + + +The .shutdown() ASoC DPCM operation calls below Bus API to release +stream assigned as part of ALLOCATED state. + +In .shutdown() the data structure maintaining stream state are freed up. + +.. code-block:: c + + void sdw_release_stream(struct sdw_stream_runtime * stream); + +The SoundWire core provides a sdw_shutdown_stream() helper function, +typically called during a dailink .shutdown() callback, which clears +the stream pointer for all DAIS connected to a stream and releases the +memory allocated for the stream. + +Not Supported +============= + +1. A single port with multiple channels supported cannot be used between two + streams or across stream. For example a port with 4 channels cannot be used + to handle 2 independent stereo streams even though it's possible in theory + in SoundWire. diff --git a/Documentation/driver-api/soundwire/summary.rst b/Documentation/driver-api/soundwire/summary.rst new file mode 100644 index 000000000..df7805374 --- /dev/null +++ b/Documentation/driver-api/soundwire/summary.rst @@ -0,0 +1,200 @@ +=========================== +SoundWire Subsystem Summary +=========================== + +SoundWire is a new interface ratified in 2015 by the MIPI Alliance. +SoundWire is used for transporting data typically related to audio +functions. SoundWire interface is optimized to integrate audio devices in +mobile or mobile inspired systems. + +SoundWire is a 2-pin multi-drop interface with data and clock line. It +facilitates development of low cost, efficient, high performance systems. +Broad level key features of SoundWire interface include: + + (1) Transporting all of payload data channels, control information, and setup + commands over a single two-pin interface. + + (2) Lower clock frequency, and hence lower power consumption, by use of DDR + (Dual Data Rate) data transmission. + + (3) Clock scaling and optional multiple data lanes to give wide flexibility + in data rate to match system requirements. + + (4) Device status monitoring, including interrupt-style alerts to the Master. + +The SoundWire protocol supports up to eleven Slave interfaces. All the +interfaces share the common Bus containing data and clock line. Each of the +Slaves can support up to 14 Data Ports. 13 Data Ports are dedicated to audio +transport. Data Port0 is dedicated to transport of Bulk control information, +each of the audio Data Ports (1..14) can support up to 8 Channels in +transmit or receiving mode (typically fixed direction but configurable +direction is enabled by the specification). Bandwidth restrictions to +~19.2..24.576Mbits/s don't however allow for 11*13*8 channels to be +transmitted simultaneously. + +Below figure shows an example of connectivity between a SoundWire Master and +two Slave devices. :: + + +---------------+ +---------------+ + | | Clock Signal | | + | Master |-------+-------------------------------| Slave | + | Interface | | Data Signal | Interface 1 | + | |-------|-------+-----------------------| | + +---------------+ | | +---------------+ + | | + | | + | | + +--+-------+--+ + | | + | Slave | + | Interface 2 | + | | + +-------------+ + + +Terminology +=========== + +The MIPI SoundWire specification uses the term 'device' to refer to a Master +or Slave interface, which of course can be confusing. In this summary and +code we use the term interface only to refer to the hardware. We follow the +Linux device model by mapping each Slave interface connected on the bus as a +device managed by a specific driver. The Linux SoundWire subsystem provides +a framework to implement a SoundWire Slave driver with an API allowing +3rd-party vendors to enable implementation-defined functionality while +common setup/configuration tasks are handled by the bus. + +Bus: +Implements SoundWire Linux Bus which handles the SoundWire protocol. +Programs all the MIPI-defined Slave registers. Represents a SoundWire +Master. Multiple instances of Bus may be present in a system. + +Slave: +Registers as SoundWire Slave device (Linux Device). Multiple Slave devices +can register to a Bus instance. + +Slave driver: +Driver controlling the Slave device. MIPI-specified registers are controlled +directly by the Bus (and transmitted through the Master driver/interface). +Any implementation-defined Slave register is controlled by Slave driver. In +practice, it is expected that the Slave driver relies on regmap and does not +request direct register access. + +Programming interfaces (SoundWire Master interface Driver) +========================================================== + +SoundWire Bus supports programming interfaces for the SoundWire Master +implementation and SoundWire Slave devices. All the code uses the "sdw" +prefix commonly used by SoC designers and 3rd party vendors. + +Each of the SoundWire Master interfaces needs to be registered to the Bus. +Bus implements API to read standard Master MIPI properties and also provides +callback in Master ops for Master driver to implement its own functions that +provides capabilities information. DT support is not implemented at this +time but should be trivial to add since capabilities are enabled with the +``device_property_`` API. + +The Master interface along with the Master interface capabilities are +registered based on board file, DT or ACPI. + +Following is the Bus API to register the SoundWire Bus: + +.. code-block:: c + + int sdw_bus_master_add(struct sdw_bus *bus, + struct device *parent, + struct fwnode_handle) + { + sdw_master_device_add(bus, parent, fwnode); + + mutex_init(&bus->lock); + INIT_LIST_HEAD(&bus->slaves); + + /* Check ACPI for Slave devices */ + sdw_acpi_find_slaves(bus); + + /* Check DT for Slave devices */ + sdw_of_find_slaves(bus); + + return 0; + } + +This will initialize sdw_bus object for Master device. "sdw_master_ops" and +"sdw_master_port_ops" callback functions are provided to the Bus. + +"sdw_master_ops" is used by Bus to control the Bus in the hardware specific +way. It includes Bus control functions such as sending the SoundWire +read/write messages on Bus, setting up clock frequency & Stream +Synchronization Point (SSP). The "sdw_master_ops" structure abstracts the +hardware details of the Master from the Bus. + +"sdw_master_port_ops" is used by Bus to setup the Port parameters of the +Master interface Port. Master interface Port register map is not defined by +MIPI specification, so Bus calls the "sdw_master_port_ops" callback +function to do Port operations like "Port Prepare", "Port Transport params +set", "Port enable and disable". The implementation of the Master driver can +then perform hardware-specific configurations. + +Programming interfaces (SoundWire Slave Driver) +=============================================== + +The MIPI specification requires each Slave interface to expose a unique +48-bit identifier, stored in 6 read-only dev_id registers. This dev_id +identifier contains vendor and part information, as well as a field enabling +to differentiate between identical components. An additional class field is +currently unused. Slave driver is written for a specific vendor and part +identifier, Bus enumerates the Slave device based on these two ids. +Slave device and driver match is done based on these two ids . Probe +of the Slave driver is called by Bus on successful match between device and +driver id. A parent/child relationship is enforced between Master and Slave +devices (the logical representation is aligned with the physical +connectivity). + +The information on Master/Slave dependencies is stored in platform data, +board-file, ACPI or DT. The MIPI Software specification defines additional +link_id parameters for controllers that have multiple Master interfaces. The +dev_id registers are only unique in the scope of a link, and the link_id +unique in the scope of a controller. Both dev_id and link_id are not +necessarily unique at the system level but the parent/child information is +used to avoid ambiguity. + +.. code-block:: c + + static const struct sdw_device_id slave_id[] = { + SDW_SLAVE_ENTRY(0x025d, 0x700, 0), + {}, + }; + MODULE_DEVICE_TABLE(sdw, slave_id); + + static struct sdw_driver slave_sdw_driver = { + .driver = { + .name = "slave_xxx", + .pm = &slave_runtime_pm, + }, + .probe = slave_sdw_probe, + .remove = slave_sdw_remove, + .ops = &slave_slave_ops, + .id_table = slave_id, + }; + + +For capabilities, Bus implements API to read standard Slave MIPI properties +and also provides callback in Slave ops for Slave driver to implement own +function that provides capabilities information. Bus needs to know a set of +Slave capabilities to program Slave registers and to control the Bus +reconfigurations. + +Links +===== + +SoundWire MIPI specification 1.1 is available at: +https://members.mipi.org/wg/All-Members/document/70290 + +SoundWire MIPI DisCo (Discovery and Configuration) specification is +available at: +https://www.mipi.org/specifications/mipi-disco-soundwire + +(publicly accessible with registration or directly accessible to MIPI +members) + +MIPI Alliance Manufacturer ID Page: mid.mipi.org |