Audio Driver Streaming Interface

This document describes the audio streaming interface exposed by audio drivers in Zircon. It is meant to serve as a reference for both users and driver-authors, and to unambiguously define the interface contract which drivers must implement and users must follow.

Overview

Audio streams are device nodes published by driver services intended to be used by applications in order to capture or render audio on a Zircon device, or both. Each stream in the system (input or output) represents a stream of digital audio information which may be either received or transmitted by device. Streams are dynamic and may created or destroyed by the system at any time. Which streams exist at any given point in time, and what controls their lifecycles are considered to be issues of audio policy and codec management and are not discussed in this document. Additionally, the information present in audio outputs streams is exclusive to the application owner of the stream. Mixing of audio is not a service provided by the audio stream interface.

TODO: extend this interface to support the concept of low-latency hardware mixers.

Basic Vocabulary

Term Definition
Sample A representation of the sound rendered by a single speaker, or captured by a single microphone, at a single instant in time.
LPCM Linear pulse code modulation. The specific representation of audio samples present in all Zircon uncompressed audio streams. LPCM audio samples are representations of the amplitude of the audio signal at an instant in time where the numeric values of the encoded audio are linearly distributed across the amplitude levels of the rendering or capture device. This is in contrast to A-law and μ-law encodings which have non-linear mappings from numeric value to amplitude level.
Channel Within an audio stream, the subset of information which will be rendered by a single speaker, or which was captured by a single microphone in a stream.
Frame A set of audio samples for every channel of a audio stream captured/rendered at a single instant in time.
Frame Rate a.k.a. "Sample Rate". The rate (in Hz) at which audio frames are produced or consumed. Common sample rates include 44.1 KHz, 48 KHz, 96 KHz, and so on.
Client or User or Application These terms are used interchangeably in this document. They refer to modules that use these interfaces to communicate with an audio driver/device.

TODO: do we need to extend this interface to support non-linear audio sample encodings? This may be important for telephony oriented microphones which deliver μ-law encoded samples.

Basic Operation

Communication with an audio stream device is performed using messages sent over a channel. Applications open the device node for a stream and obtain a channel by issuing a FIDL request. After obtaining the channel, the device node may be closed. All subsequent communication with the stream occurs using channels.

The stream channel is used for most command and control tasks, including:

  • Capability interrogation
  • Format negotiation
  • Hardware gain control
  • Determining outboard latency
  • Plug detection notification
  • Access control capability detection and signalling

TODO: Should plug/unplug detection be done by sending notifications over the stream channel (as it is today), or by publishing/unpublishing the device nodes (and closing all channels in the case of unpublished channels)?

In order to actually send or receive audio information on the stream, the specific format to be used must first be set. The response to a successful SetFormat operation will contain a new "ring-buffer" channel. The ring-buffer channel may be used to request a shared buffer from the stream (delivered in the form of a VMO) which may be mapped into the address space of the application and used to send or receive audio data as appropriate. Generally, the operations conducted over the ring buffer channel include:

  • Requesting a shared buffer
  • Starting and Stopping stream playback/capture
  • Receiving notifications of playback/capture progress
  • Receiving notifications of error conditions such as HW FIFO under/overflow, bus transaction failure, etc.
  • Receiving clock recovery information in the case that the audio output clock is based on a different oscillator than the oscillator which backs ZX_CLOCK_MONOTONIC

Operational Details

Protocol definition

In order to use the C API definitions of the audio protocol, applications and drivers simply say

#include <device/audio.h>

Device nodes

Audio stream device nodes must be published by drivers using the protocol preprocessor symbol given in the table below. This will cause stream device nodes to be published in the locations given in the table. Applications can monitor these directories in order to discover new streams as they are published by the drivers.

Stream Type Protocol Location
Input ZX_PROTOCOL_AUDIO_INPUT /dev/class/audio-input
Output ZX_PROTOCOL_AUDIO_OUTPUT /dev/class/audio-output

Establishing the stream channel

After opening the device node, client applications may obtain a stream channel for subsequent communication using the fuchsia.hardware.audio.Device/GetChannel FIDL message. For example:

zx_handle_t OpenStream(const char* dev_node_path) {
    zx_handle_t local, remote;
    zx_status_t status = zx_channel_create(0, &local, &remote);
    if (status != ZX_OK) {
      return ZX_HANDLE_INVALID;
    }
    status = fdio_service_connect(dev_node_path, remote);
    if (status != ZX_OK) {
      LOG("Failed to open \"%s\" (res %d)\n", dev_node_path, status);
      zx_handle_close(local);
      return ZX_HANDLE_INVALID;
    }

    zx_handle_t audio_channel;
    status = fuchsia_hardware_audio_DeviceGetChannel(local, &audio_channel);
    if (status != ZX_OK) {
      printf("Failed to obtain channel (res %d)\n", status);
      return ZX_HANDLE_INVALID;
    }
    zx_handle_close(local);
    return audio_channel;
}

Client side termination of the stream channel

Clients may terminate the connection to the stream at any time simply by calling zx_handle_close(...) on the stream channel. Drivers must close any active ring-buffer channels established using this stream channel and must make every attempt to gracefully quiesce any on-going streaming operations in the process.

Sending and receiving messages on the stream and ring-buffer channels

All of the messages and message payloads which may be sent or received over stream and ring buffer channels are defined in the audio protocol header. Messages may be sent to the driver using the zx_channel_write(...) syscall. If a response is expected, it may be read using the zx_channel_read(...) syscall. Best practice, however, is to queue packets for your channel(s) port using the zx_port_queue(...) syscall, and use the zx_port_wait(...) syscall to determine when your set of channels have messages (either expected responses or asynchronous notifications) to be read.

All messages either sent or received over stream and ring buffer channels are prefaced with an audio_cmd_hdr_t structure which contains a 32-bit transaction ID and an audio_cmd_hdr_t enumeration value indicating the specific command being requested by the application, the specific command being responded to by the driver, or the asynchronous notification being delivered by the driver to the application.

When sending a command to the driver, applications must place a transaction ID in the header's transaction_id field which is not equal to AUDIO_INVALID_TRANSACTION_ID. If a response to a command needs to be sent by the driver to the application, the driver must use the transaction ID and audio_cmd_t values sent by the client during the request. When sending asynchronous notification to the application, the driver must use AUDIO_INVALID_TRANSACTION_ID as the transaction ID for the message. Transaction IDs may be used by clients for whatever purpose they desire, however if the IDs are kept unique across all transactions in-flight, the zx_channel_call(...) may be used to implement a simple synchronous calling interface.

Validation requirements

All drivers must validate requests and enforce the protocol described above. In case of any violation, drivers should immediately quiesce their hardware and must close the channel, terminating any operations which happen to be in flight at the time. Additionally, they may log a message to a central logging service to assist in application developers in debugging the cause of the protocol violation. Examples of protocol violation include:

  • Using AUDIO_INVALID_TRANSACTION_ID as the value of message.hdr.transaction_id
  • Using a value not present in the audio_cmd_t enumeration as the value of message.hdr.cmd
  • Supplying a payload whose size does not match the size of the request payload for a given command.

Format Negotiation

Sample Formats

Sample formats are described using the audio_sample_format_t type. It is a bitfield style enumeration which describes either the numeric encoding of the uncompressed LPCM audio samples as they reside in memory, or indicating that the audio stream consists of a compressed bitstream instead of uncompressed LPCM samples. Refer to the audio protocol header for exact symbol definitions.

The formats described by audio_sample_format_t have the following properties:

  • With the exception of FORMAT_BITSTREAM, samples are always assumed to use linear PCM encoding. BITSTREAM is used for transporting compressed audio encodings (such as AC3, DTS, and so on) over a digital interconnect to a decoder device somewhere outside of the system.
  • By default, multi-byte sample formats are assumed to use host-endianness. If the INVERT_ENDIAN flag is set on the format, the format uses the opposite of host endianness. eg. A 16 bit little endian PCM audio format would have the INVERT_ENDIAN flag set on it in a when used on a big endian host. The INVERT_ENDIAN flag has no effect on COMPRESSED, 8BIT or FLOAT encodings.
  • The 32BIT_FLOAT encoding uses specifically the IEEE 754 floating point representation.
  • By default, non-floating point PCM encodings are assumed expressed using two's complement signed integers. eg. the bit values for a 16 bit PCM sample format would range from [0x8000, 0x7FFF] with 0x0000 representing zero speaker deflection. If the UNSIGNED flag is set on the format, the bit values would range from [0x0000, 0xFFFF] with 0x8000 representing zero deflection.
  • When used to set formats, exactly one non-flag bit must be set.
  • When used to describe supported formats, any number of non-flag bits may be set. Flags (when present) apply to all of the relevant non-flag bits in the bitfield. eg. If a stream supports BITSTREAM, 16BIT and 32BIT_FLOAT, and the UNSIGNED bit is set, it applies only to the 16BIT format.
  • When encoding a smaller sample size in a larger container (eg 20 or 24bit in 32), the most significant bits of the 32 bit container are used while the least significant bits should be zero. eg. a 20 bit sample would be mapped onto the range [12,32] of the 32 bit container.

TODO: can we make the claim that the LSBs will be ignored, or do we have to require that they be zero?

TODO: describe what 20-bit packed audio looks like in memory. Does it need to have an even number of channels in the overall format? Should we strike it from this list if we cannot find a piece of hardware which demands this format in memory?

Enumeration of supported formats

In order to determine the formats supported by a given audio stream, applications send an AUDIO_STREAM_CMD_GET_FORMATS message over the stream channel. No additional parameters are required. Drivers must respond to this request using one or more audio_stream_cmd_get_formats_resp_t messages, even if only to report that there are no formats currently supported.

Range structures

Drivers indicate support for formats by sending messages containing zero or more audio_stream_format_range_t structures. Each structure contains fields which describe:

  • A bitmask of supported sample formats.
  • A minimum and maximum number of channels.
  • A set of frame rates.

A single range structure indicates support for each of the combinations of the three different sets of values (sample formats, channel counts, and frame rates). For example, if a range structure indicated support for:

  • 16 bit signed LPCM samples
  • 48000, and 44100 Hz frame rates
  • 1 and 2 channels

Then the fully expanded set of supported formats indicated by the range structure would be:

  • Stereo 16-bit 48 KHz audio
  • Stereo 16-bit 44.1 KHz audio
  • Mono 16-bit 48 KHz audio
  • Mono 16-bit 44.1 KHz audio

See the Sample Formats section (above) for a description of how sample formats are encoded in the sample_formats member of a range structure.

Supported channel counts are indicated using a pair of min/max channels fields which indicate an exclusive range of channel counts which apply to this range. For example, a min/max channels range of [1, 4] would indicate that this audio stream supports 1, 2, 3 or 4 channels. A range of [2, 2] would indicate that this audio stream supports only stereo audio.

Supported frame rates are signalled similarly to channel counts using a pair of min/max frame per second fields along with a flags field. While the min/max values provide an inclusive range of frame rates, the flags determine how to interpret this range. Currently defined flags include:

Flag Definition
ASF_RANGE_FLAG_FPS_CONTINUOUS The frame rate range is continuous. All frame rates in the range [min, max] are valid.
ASF_RANGE_FLAG_FPS_48000_FAMILY The frame rate range includes the members of the 48 KHz family which exist in the range [min, max]
ASF_RANGE_FLAG_FPS_44100_FAMILY The frame rate range includes the members of the 44.1 KHz family which exist in the range [min, max]

So, conceptually, the valid frame rates are the union of the sets produced by applying each of the flags which are set to the inclusive [min, max] range. For example, if both the 48 KHz and 44.1 KHz were set, and the range given was [16000, 47999], then the supported frame rates for this range would be

  • 16000 Hz
  • 22050 Hz
  • 32000 Hz
  • 44100 Hz

The official members of the 48 KHz and 44.1 KHz families are

Family Frame Rates
ASF_RANGE_FLAG_FPS_48000_FAMILY 8000, 16000, 32000, 48000, 96000, 192000, 384000, 768000
ASF_RANGE_FLAG_FPS_44100_FAMILY 11025, 22050, 44100, 88200, 176400

Drivers must set at least one of the flags, or else the set of supported frame rates is empty and this range structure is not allowed. Also note that the set of valid frame rates is the union of the frame rates produce by applying each of the set flags. If the ASF_RANGE_FLAG_FPS_CONTINUOUS flag is set the other flags have no effect. While it is legal to do so, drivers should avoid this behavior.

Transporting range structures

Range structures are transmitted from drivers to applications within the audio_stream_cmd_get_formats_resp_t message. If a large number of formats are supported by a stream, drivers may need to send multiple messages to enumerate all available modes. Messages include the following fields:

  • A standard audio_cmd_hdr_t header. All messages involved in the response to an AUDIO_STREAM_CMD_GET_FORMATS request must use the transaction ID of the original request, and the cmd field in the header must be AUDIO_STREAM_CMD_GET_FORMATS.
  • A format_range_count field. This indicates the total number of format range structures which will be sent in this response to the application. This number must be present in all messages involved in the response, and must not change from message to message.
  • A first_format_range_ndx field indicating the zero-based index of the first format range being specified in this particular message. See below for details.
  • An array of audio_stream_cmd_get_formats_resp_t structures which is at most AUDIO_STREAM_CMD_GET_FORMATS_MAX_RANGES_PER_RESPONSE elements long.

Drivers must:

  • Always transmit all of the available audio format ranges.
  • Always transmit the available audio format ranges in ascending index order.
  • Always pack as many ranges as possible in the fixed size message structure.
  • Never overlap index regions or leave gaps.

Given these requirements, if the maximum number of ranges per response were 15, and a driver needed to send 35 ranges in response to an application's request, then 3 messages in total would be needed, and the format_range_count and first_format_range_ndx fields for each message would be as follows.

Msg # format_range_count first_format_range_ndx
1 35 0
2 35 15
3 35 30

first_format_range_ndx must never be greater than format_range_count, however format_range_count may be zero if an audio stream currently supports no formats. The total number of audio_stream_format_range_t structures in an audio_stream_cmd_get_formats_resp_t message is given by the formula

valid_ranges = MIN(AUDIO_STREAM_CMD_GET_FORMATS_MAX_RANGES_PER_RESPONSE,
                   msg.format_range_count - msg.first_format_range_ndx);

Drivers may choose to always send an entire audio_stream_cmd_get_formats_resp_t message, or to send a truncated message which ends after the last valid range structure in the format_ranges array. Applications must be prepared to receive up to sizeof(audio_stream_cmd_get_formats_resp_t) bytes for each message, but also accept messages as short as offsetof(audio_stream_cmd_get_formats_resp_t, format_ranges)

TODO: how do devices signal a change of supported formats (e.g., HDMI hot-plug event)? Are such devices required to simply remove and republish the device?

TODO: define how to enumerate supported compressed bitstream formats.

Setting the desired stream format

In order to select a stream format, applications send an AUDIO_STREAM_CMD_SET_FORMAT message over the stream channel. In the message, for uncompressed audio streams, the application specifies:

  • The frame rate of the stream in Hz using the frames_per_second field (in the case of an uncompressed audio stream).
  • The number of channels packed into each frame using the channels field.
  • The format of the samples in the frame using the sample_format field (see Sample Formats, above)

Success or failure, drivers must respond to a request to set format using a audio_stream_cmd_set_format_resp_t.

In the case of success, drivers must set the result field of the response to ZX_OK and must return a new ring buffer channel over which streaming operations will be conducted. If a previous ring buffer channel had been established and was still active, the driver must close this channel and make every attempt to gracefully quiesce any on-going streaming operations in the process.

In the case of failure, drivers must indicate the cause of failure using the result field of the message and must not simply close the stream channel as is done for a generic protocol violation. Additionally, they may choose to preserve a pre-existing ring-buffer channel, or to simply close such a channel as is mandated for a successful operation.

TODO: specify how compressed bitstream formats will be set

Determining external latency

The external latency of an audio stream is defined as the amount of time it takes outbound audio to travel from the system's interconnect to the speakers themselves, or inbound audio to travel from the microphone to the system's interconnect. As an example, consider an external codec connected to the system using a TDM interconnect: if this interconnect introduces a 4 frame delay between the reception of a TDM frame and the rendering of that frame at the speakers themselves, then the external delay of this audio path is the time duration equivalent to 4 audio frames.

External delay is reported in the external_delay_nsec field of a successful AUDIO_STREAM_CMD_SET_FORMAT response as a non-negative number of nanoseconds. Drivers should make their best attempt to accurately report the total of all of the sources of delay the driver knows about. Information about this delay can frequently be found in codec data sheets, dynamically reported as properties of codecs using protocols such as Intel HDA or the USB Audio specifications, or reported by down stream devices using mechanisms such as EDID when using HDMI or DisplayPort interconnects.

Hardware Gain Control

Hardware gain control capability reporting

In order to determine a stream's gain control capabilities, applications send an AUDIO_STREAM_CMD_GET_GAIN message over the stream channel. No parameters need to be supplied with this message. All stream drivers must respond to this message, regardless of whether or not the stream hardware is capable of any gain control. All gain values are expressed using 32 bit floating point numbers expressed in dB.

Drivers respond to this message with values which indicate the current gain settings of the stream, as well as the stream's gain control capabilities. Current gain settings are expressed using a bool/float tuple indicating if the stream is currently muted or not along with the current dB gain of the stream. Gain capabilities consist of bool and 3 floats. The bool indicates whether or not the stream can be muted. The floats give the minimum and maximum gain settings, along with the gain step size. The gain step size indicates the smallest increment with which the gain can be controlled counting from the minimum gain value.

For example, an amplifier which has 5 gain steps of 7.5 dB each and a maximum 0 dB gain would indicate a range of (-30.0, 0.0) and a step size of 7.5. Amplifiers capable of functionally continuous gain control may encode their gain step size as 0.0.

Regardless of mute capabilities, drivers for fixed gain streams must report their min/max gain as (0.0, 0.0). The gain step size is meaningless in this situation, but drivers should report their step size as 0.0.

Setting hardware gain control levels

In order to change a stream's current gain settings, applications send an AUDIO_STREAM_CMD_SET_GAIN message over the stream channel. Two parameters are supplied with this message, a set of flags which control the request, and a float indicating the dB gain which should be applied to the stream.

Three valid flags are currently defined:

  • AUDIO_SGF_MUTE_VALID. Set when the application wishes to set the muted/un-muted state of the stream. Clear if the application wishes to preserve the current muted/un-muted state.
  • AUDIO_SGF_GAIN_VALID. Set when the application wishes to set the dB gain state of the stream. Clear if the application wishes to preserve the current gain state.
  • AUDIO_SGF_MUTE. Indicates the application's desired mute/un-mute state for the stream. Significant only if AUDIO_SGF_MUTE_VALID is also set.

Drivers must fail the request with an ZX_ERR_INVALID_ARGS result if the application's request is incompatible with the stream's capabilities. Incompatible requests include:

  • The requested gain is less than the minimum support gain for the stream.
  • The requested gain is more than the maximum support gain for the stream.
  • Mute was requested, but the stream does not support an explicit mute.

Presuming that the request is valid, drivers should round the request to the nearest supported gain step size. For example, if a stream can control its gain on the range from -60.0 to 0.0 dB, using a gain step size of 0.5 dB, then a request to set the gain to -33.3 dB should result in a gain of -33.5 being applied. A request to that same stream for a gain of -33.2 dB should result in a gain of -33.0 being applied.

Applications may choose not to receive an acknowledgement of a SET_GAIN command by setting the AUDIO_FLAG_NO_ACK flag on their command. No response message will be sent to the application, regardless of the success or failure of the command. If an acknowledgement was requested by the application, drivers respond with a message indicating the success or failure of the operation as well as the current gain/mute status of the system (regardless of whether the request was a success).

Plug Detection

In addition to streams being published/unpublished in response to being connected or disconnected to/from their bus, streams may have the ability to be plugged or unplugged at any given point in time. For example, a set of USB headphones may publish a new output stream when connected to USB, but choose to be "hardwired" from a plug detection standpoint. A different USB audio adapter with a standard 3.5mm phono jack might publish an output stream when connected via USB, but choose to change its plugged/unplugged state as the user plugs and unplugs an analog device via the 3.5mm jack.

The ability to query the currently plugged or unplugged state of a stream, and to register for asynchonous notifications of plug state changes (if supported) is handled via plug detection messages.

AUDIO_STREAM_CMD_PLUG_DETECT

In order to determine a stream's plug detection capabilities and current plug state, and to enable or disable for asynchronous plug detection notifications, applications send a AUDIO_STREAM_CMD_PLUG_DETECT command over the stream channel. Drivers respond with a set of audio_pd_notify_flags_t, along with a timestamp referenced from ZX_CLOCK_MONOTONIC indicating the last time the plug state changed.

Three valid plug-detect notification flags (PDNF) are currently defined:

  • AUDIO_PDNF_HARDWIRED is set when the stream hardware is considered to be "hardwired". In other words, the stream is considered to be connected as long as the device is published. Examples include a set of built-in speakers, a pair of USB headphones, or a pluggable audio device with no plug detection functionality.
  • AUDIO_PDNF_CAN_NOTIFY is set when the stream hardware is capable of both asynchronously detecting that a device's plug state has changed, and sending a notification message if the client has requested these notifications.
  • AUDIO_PDNF_PLUGGED is set when the stream hardware considers the stream to be currently in the "plugged-in" state.

When responding to the PLUG_DETECT message, drivers for "hardwired" streams must not set the CAN_NOTIFY flag, and must set the PLUGGED flag. Additionally, these drivers should always set the plug state time to the time at which the stream device was published by the driver.

Applications may choose not to receive an acknowledgement of a PLUG_DETECT command by setting the AUDIO_FLAG_NO_ACK flag on their command. No response message will be sent to the application, regardless of the success or failure of the command. The most common use for this would be when an application wanted to disable asynchronous plug state detection messages and was not actually interested in the current plugged/unplugged state of the stream.

AUDIO_STREAM_PLUG_DETECT_NOTIFY

Applications may request that streams send them asynchronous notifications of plug state changes, using the flags field of the AUDIO_STREAM_CMD_PLUG_DETECT command.

Two valid flags are currently defined:

  • AUDIO_PDF_ENABLE NOTIFICATIONS is set by clients in order to request that the stream proactively generate AUDIO_STREAM_PLUG_DETECT_NOTIFY messages when its plug state changes, if the stream has this capability.
  • AUDIO_PDF_DISABLE_NOTIFICATIONS is set by clients in order to request that NO subsequent AUDIO_STREAM_PLUG_DETECT_NOTIFY messages should be sent, regardless of the stream's ability to generate them.

In order to request the current plug state without altering the current notification behavior, clients simply set neither ENABLE nor DISABLE -- passing either 0, or the value AUDIO_PDF_NONE. Clients should not set both flags at the same time. If they do, drivers must interpret this to mean that the final state of the system should be disabled.

Clients which request asynchronous notifications of plug state changes should always check the CAN_NOTIFY flag in the driver response. Streams may be capable of plug detection (i.e. if HARDWIRED is not set), yet be incapable of detecting plug state changes asynchronously. Clients may still learn of plug state changes, but only by periodically polling the state with PLUG_DETECT commands. Drivers for streams which do not set the CAN_NOTIFY flag are free to ignore enable/disable notification requests from applications, and must not ever send an AUDIO_STREAM_PLUG_DETECT_NOTIFY message. Note that even such a driver must always respond to a AUDIO_STREAM_CMD_PLUG_DETECT message.

Access control capability detection and signaling

TODO: specify how this works. In particular, specify how drivers indicate to applications support for various digital access control mechanisms such as S/PDIF control words and HDCP.

Stream purpose and association

TODO: specify how drivers can indicate the general "purpose" of an audio stream in the system (if known), as well as its relationship to other streams (if known). For example, an embedded target like a phone or a tablet needs to indicate which output stream is the built-in speaker vs. which is the headset jack output. In addition, it needs to make clear which input stream is the microphone associated with the headset output vs. the builtin speaker.

Ring-Buffer Channels

Overview

Once an application has successfully set the format of a stream, it receives in the response a new channel representing its connection to the stream's ring-buffer. Clients use the ring-buffer channel to establish a shared memory buffer and start/stop playback/capture of audio stream data.

Once started, stream consumption/production is assumed to proceed at the nominal rate from the point in time given in a successful response to the start command, allowing clients to operate without the need to receive any periodic notifications about consumption/production position from the ring buffer itself. Note that the ring-buffer will almost certainly have some form of FIFO buffer between the memory bus and the audio hardware which causes it to either read-ahead in the stream (in the case of playback), or potentially hold onto data (in the case of capturing). In the case of open-loop operation, it is important for clients to query the size of this buffer before beginning operation so they know how far ahead/behind the stream's nominal inferred read/write position they need to stay in order to prevent audio glitching.

Also note that because of the shared buffer nature of the system, and the fact that drivers are likely to be DMA-ing directly from this buffer to hardware, it is important for clients running on architectures which are not automatically cache coherent to be sure that they have properly written-back their cache after writing playback data to the buffer, or invalidated their cache before reading captured data.

Determining the FIFO depth

Applications determine a stream's FIFO depth using the AUDIO_RB_CMD_GET_FIFO_DEPTH command. Drivers must return their FIFO depth, expressed in bytes, in the fifo_depth field of the response. To ensure proper playback or capture of audio, applications and drivers must be careful to respect this value. Drivers must not read beyond the nominal playback position of the stream plus this number of bytes when playing audio stream data. Applications must stay this number of bytes behind the nominal capture point of the stream when capturing audio stream data.

Once the format of a stream is set and a ring-buffer channel has been opened, the driver must not change this value. From an application's point of view, it is a constant property of the ring-buffer channel.

Obtaining a shared buffer

To send or receive audio, the application must first establish a shared memory buffer. This is done by sending an AUDIO_RB_CMD_GET_BUFFER request over the ring-buffer channel. This may only be done while the ring-buffer is stopped. Applications must specify two parameters when requesting a ring buffer: min_ring_buffer_frames and notifications_per_ring.

min_ring_buffer_frames

The minimum number of frames of audio the client needs allocated for the ring buffer. Drivers may make this buffer larger to meet hardware requirements. Clients must use the returned VMOs size (in bytes) to determine the actual size of the ring buffer. Clients may not assume that the size of the buffer (as determined by the driver) is exactly the size they requested. Drivers must ensure that the size of the ring buffer is an integral number of audio frames.

TODO : Is it reasonable to require that drivers produce buffers which are an integral number of audio frames in length? It certainly makes the audio client's life easier (client code never needs to split or re-assemble a frame before processing), but it might make it difficult for some audio hardware to meet its requirements without making the buffer significantly larger than the client asked for.

notifications_per_ring

The number of position update notifications (audio_rb_position_notify_t) the client would like the driver to send per cycle through the ring buffer. Drivers should attempt to space notifications uniformly throughout the ring. Clients may not rely on perfectly uniform spacing of the update notifications. Clients are not required to request any notifications and may use only start time and FIFO depth information to determine the driver's playout or capture position.

Success or failure, drivers must respond to a GET_BUFFER request using an audio_rb_cmd_get_buffer_resp_t message. If the driver fails the request because a buffer has already been established and the ring-buffer has already been started, it must not either stop the ring-buffer, or discard the existing shared memory. If the application requests a new buffer after having already established a buffer while the ring buffer is stopped, it must consider the existing buffer is has to be invalid. Success or failure, the old buffer is now gone.

If the request succeeds, the driver must return a handle to a VMO with permissions which allow applications to map the VMO into their address space using zx_vmar_map, and to read/write data in the buffer in the case of playback, or simply to read the data in the buffer in the case of capture. Additionally, the driver must report the actual number of frames of audio it will use in the buffer via the num_ring_buffer_frames field of the audio_rb_cmd_get_buffer_resp_t message. The size of the VMO returned (as reported by zx_vmo_get_size()) must not be larger than this number of frames (when converted to bytes). This number may be larger than the min_ring_buffer_frames request from the client but must not be smaller than this number.

Starting and Stopping the ring-buffer

Clients may request that a ring-buffer start or stop using the AUDIO_RB_CMD_START and AUDIO_RB_CMD_STOP commands. Success or failure, drivers must send a response to these requests. Attempting to start a stream which is already started must be considered a failure. Attempting to stop a stream which is already stopped should be considered a success. Ring-buffers cannot be either stopped or started until after a shared buffer has been established using the GET_BUFFER operation.

Upon successfully starting a stream, drivers must provide their best estimate of the time at which their hardware began to transmit or capture the stream in the start_time field of the response. This time stamp must be taken from the clock exposed via the ZX_CLOCK_MONOTONIC syscall. Along with the FIFO depth property of the ring buffer, this timestamp allows applications to send or receive stream data without the need for periodic position updates from the driver. Along with the outboard latency estimate provided by the stream channel, this timestamp allows applications to synchronize presentation of audio information across multiple streams, or even multiple devices (provided that an external time synchronization protocol is used to synchronize the ZX_CLOCK_MONOTONIC timelines across the cohort of synchronized devices).

TODO: Redefine start_time to allow it to be an arbitrary 'audio stream clock' instead of the ZX_CLOCK_MONOTONIC clock. If the stream clock is made to count in audio frames since start, then this start_time can be replaced with the terms for a segment of a piecewise linear transformation which can be subsequently updated via notifications sent by the driver in the case that the audio hardware clock is rooted in a different oscillator from the system's tick counter. Clients can then use this transformation either to control the rate of consumption of input streams, or to determine where to sample in the input stream to effect clock correction.

Upon successfully starting a stream, drivers must guarantee that no position notifications will be sent before the start response has been enqueued into the ring-buffer channel.

Upon successfully stopping a stream, drivers must guarantee that no position notifications will be enqueued into the ring-buffer channel after the stop response has been enqueued.

Position notifications

If requested by the client during the GET_BUFFER operation, the driver will periodically send updates to the client informing it of its current production or consumption position in the buffer. This position is expressed in bytes in the ring_buffer_pos field of the audio_rb_position_notify_t message. The message also includes a monotonic_time field that contains the time (as zx_time_t) that this byte position was valid. AUDIO_RB_POSITION_NOTIFY messages must only be sent while the ring-buffer is started. Note, these position notifications indicate where in the buffer the driver has consumed or produced data, not the nominal playback or capture position (sometimes called the "write cursor" or "read cursor" respectively). The timing of their arrival is not guaranteed to be perfectly uniform and should not be used to effect clock recovery. However, the correspondence pair (monotonic_time, ring_buffer_pos) values themselves ARE intended to be used to recover the clock for the audio stream. If a client discovers that a driver has consumed past the point in the ring buffer where that client has written playback data, audio presentation is undefined. Clients should increase their clock lead time and be certain to stay ahead of this point in the stream in the future. Likewise, clients which capture audio should not attempt to read beyond the point in the ring buffer indicated by the most recent position notification sent by the driver.

Driver playback/capture position must always begin at ring buffer byte 0, immediately following a successful AUDIO_RB_CMD_START command. When the ring buffer position reaches the end of the VMO (as indicated by zx_vmo_get_size(...)), the ring buffer position wraps back to zero. Drivers are not required to consume or produce data in integral numbers of audio frames. Clients whose notion of stream position depends on position notifications should take care to request that a sufficient number of notifications per ring be sent (minimum 2) and to process them quickly enough that aliasing does not occur.

Clock recovery

TODO: rewrite this section to include how clock recovery occurs, and how this is exposed to clients. Also, detail how slewable oscillators are discovered and controlled. We may need rate-change notifications to clients of slewable clocks.

Previous content: TODO: define a way that clock recovery information can be sent to clients in the case that the audio output oscillator is not derived from the ZX_CLOCK_MONOTONIC oscillator. In addition, if the oscillator is slew-able in hardware, provide the ability to discover this capability and control the slew rate. Given the fact that this oscillator is likely to be shared by multiple streams, it might be best to return some form of system wide clock identifier and provide the ability to obtain a channel on which clock recovery notifications can be delivered to clients and HW slewing command can be sent from clients to the clock.

Error notifications

TODO: define these and what driver behavior should be, if/when they occur.

Unexpected client termination

If the client side of a ring buffer control channel is closed for any reason, drivers must immediately close the control channel and shut down the ring buffer, such that no further audio is emitted nor captured. While drivers are encouraged to do so in a way which produces a graceful transition to silence, they must ensure that the audio stream goes silent instead of looping. Once the transition to silence is complete, resources associated with playback or capture may be released and reused by the driver.

This way, if a playback client teminates unexpectedly, the system will close the client channels, causing audio playback to stop instead of continuing to loop.