| Serial Peripheral Interface (SPI) | |
| Type: | Serial communication bus |
| Designer: | Motorola |
| Data Style: | Full-duplex serial |
| Daisy Chain: | Depends on devices |
| Manufacturer: | various |
| Data Devices: | Multidrop limited by chip selects. Daisy chaining unlimited. |
| Data Bit Width: | 1 bit (bidirectional) |
| High Freq: | yes |
| Pin Custom1 Name: | MOSI |
| Pin Name Custom1: | Master Out Slave In |
| Pin Name Custom2: | Master In Slave Out |
| Pin Custom2 Name: | MISO |
| Pin Custom3 Name: | SCLK |
| Pin Name Custom3: | Serial Clock |
| Pin Name Custom4: | Chip Select (one or more) |
| Pinout Notes: | (pins may have alternative names) |
| Maximum Voltage: | Unspecified |
| Maximum Current: | Unspecified |
| Physical Connector: | Unspecified |
| Design Date: | Around early 1980s |
Serial Peripheral Interface (SPI) is a de facto standard (with many variants) for synchronous serial communication, used primarily in embedded systems for short-distance wired communication between integrated circuits.
SPI uses a master–slave architecture, described here with the terms "main" and "sub", [1] where one main device orchestrates communication with some number of peripheral (sub) devices by driving the clock signal and chip select signal(s).
Motorola's original specification (early 1980s) uses four wires to perform full duplex communication. It is sometimes called a four-wire serial bus to contrast with three-wire variants which are half duplex, and with the two-wire I²C and 1-Wire serial buses.
Typical applications include interfacing microcontrollers with peripheral chips for Secure Digital cards, liquid crystal displays, analog-to-digital and digital-to-analog converters, flash and EEPROM memory, and various communication chips.
SPI may be accurately described as a synchronous serial interface,[2] but it is different from the Synchronous Serial Interface (SSI) protocol.
(Note: Variations section describes operation of non-standard variants.)
SPI has four logic signals (which may have alternative names):
| Short Name | Long Name | Description (historical terms in parens) | |
|---|---|---|---|
| Active-low chip select signal from main (master) to enable communication with a specific sub (slave) device. | |||
| Clock signal from main (master). | |||
| Serial data from main (master), most-significant bit first. | |||
| Serial data from sub (slave), most-significant bit first. |
MOSI on a main outputs to MOSI on a sub. MISO on a sub outputs to MISO on a main.
SPI operates with a single device acting as main and with one or more sub devices.
Sub devices should use tri-state outputs so their MISO signal becomes high impedance (electrically disconnected) when the device is not selected. Subs without tri-state outputs cannot share a MISO wire with other subs without using an external tri-state buffer.
To begin communication, the SPI main first selects a sub device by pulling its low. (Note: the bar above indicates it is an active low signal, so a low voltage means "selected", while a high voltage means "not selected")
If a waiting period is required, such as for an analog-to-digital conversion, the main must wait for at least that period of time before issuing clock cycles.
During each SPI clock cycle, full-duplex transmission of a single bit occurs. The main sends a bit on the MOSI line while the sub sends a bit on the MISO line, and then each reads their corresponding incoming bit. This sequence is maintained even when only one-directional data transfer is intended.
Transmission using a single sub (Figure 1) involves one shift register in the main and one shift register in the sub, both of some given word size (e.g. 8 bits), connected in a virtual ring topology. Data is usually shifted out with the most-significant bit (MSB) first. On the clock edge, both main and sub shift out a bit to its counterpart. On the next clock edge, each receiver samples the transmitted bit and stores it in the shift register as the new least-significant bit. After all bits have been shifted out and in, the main and sub have exchanged register values. If more data needs to be exchanged, the shift registers are reloaded and the process repeats. Transmission may continue for any number of clock cycles. When complete, the main stops toggling the clock signal, and typically deselects the sub.
If a single sub device is used, its pin may be fixed to logic low if the sub permits it. With multiple sub devices, a multidrop configuration requires an independent signal from the main for each sub device, while a daisy-chain configuration only requires one signal.
Every sub on the bus that has not been selected should disregard the input clock and MOSI signals. And to prevent contention on MISO, non-selected subs must use tristate output. Subs that aren't already tristate will need external tristate buffers to ensure this.
In addition to setting the clock frequency, the main must also configure the clock polarity and phase with respect to the data. Motorola[3] named these two options as CPOL and CPHA (for clock polarity and clock phase) respectively, a convention most vendors have also adopted.
The SPI timing diagram shown is further described below:
The combinations of polarity and phases are referred to by these "SPI mode" numbers with CPOL as the high order bit and CPHA as the low order bit:
| SPI mode | Clock polarity (CPOL) | Clock phase (CPHA) | Data is shifted out on | Data is sampled on | |
|---|---|---|---|---|---|
| 0 | 0 | 0 | falling SCLK, and when activates | rising SCLK | |
| 1 | 0 | 1 | rising SCLK | falling SCLK | |
| 2 | 1 | 0 | rising SCLK, and when activates | falling SCLK | |
| 3 | 1 | 1 | falling SCLK | rising SCLK |
Notes:
Some sub devices are designed to ignore any SPI communications in which the number of clock pulses is greater than specified. Others do not care, ignoring extra inputs and continuing to shift the same output bit. It is common for different devices to use SPI communications with different lengths, as, for example, when SPI is used to access an IC's scan chain by issuing a command word of one size (perhaps 32 bits) and then getting a response of a different size (perhaps 153 bits, one for each pin in that scan chain).
Interrupts are outside the scope of SPI; their usage is neither forbidden nor specified, and so may optionally be implemented.
Microcontrollers configured as sub devices may have hardware support for generating interrupt signals to themselves when data words are received or overflow occurs in a receive FIFO buffer,[4] and may also set up an interrupt routine when their chip select input line is pulled low or high.
SPI subs sometimes use an out-of-band signal (another wire) to send an interrupt signal to a main. Examples include pen-down interrupts from touchscreen sensors, thermal limit alerts from temperature sensors, alarms issued by real-time clock chips, SDIO and audio jack insertions for an audio codec. Interrupts to main may also be faked by using polling (similarly to USB 1.1 and 2.0).
SPI lends itself to a "bus driver" software design. Software for attached devices is written to call a "bus driver" that handles the actual low-level SPI hardware. This permits the driver code for attached devices to port easily to other hardware or a bit-banging software implementation.
The pseudocode below outlines a software-implementation ("bit-banging") of SPI's protocol as a main with simultaneous output and input. This pseudocode is for CPHA=0 and CPOL=0, thus SCLK is pulled low before is activated and bits are inputted on SCLK's rising edge while bits are outputted on SCLK's falling edge.
Bit-banging a sub's protocol is similar but different from above. An implementation might involve busy waiting for to fall or triggering an interrupt routine when falls, and then shifting in and out bits when the received SCLK changes appropriately for however long the transfer size is.
Though the previous operation section focused on a basic interface with a single sub, SPI can instead communicate with multiple subs using multidrop, daisy chain, or expander configurations.
In the multidrop bus configuration, each sub has its own, and the main selects only one at a time. MISO, SCLK, and MOSI are each shared by all devices. This is the way SPI is normally used.
Since the MISO pins of the subs are connected together, they are required to be tri-state pins (high, low or high-impedance), where the high-impedance output must be applied when the sub is not selected. Sub devices not supporting tri-state may be used in multidrop configuration by adding a tri-state buffer chip controlled by its signal.[5] (Since only a single signal line needs to be tristated per sub, one typical standard logic chip that contains four tristate buffers with independent gate inputs can be used to interface up to four sub devices to an SPI bus)
Caveat: All signals should start high (to indicate no chips are selected) before sending initialization messages to any sub, so other uninitialized subs ignore messages not addressed to them. This is a concern if the main uses general-purpose input/output (GPIO) pins (which may default to an undefined state) for and if the main uses separate software libraries to initialize each device. One solution is to configure all GPIOs used for to output a high voltage for all subs before running initialization code from any of those software libraries. Another solution is to add a pull-up resistor on each, to ensure that all signals are initially high.
Some products that implement SPI may be connected in a daisy chain configuration, where the first sub's output is connected to the second sub's input, and so on with subsequent subs, until the final sub, whose output is connected back to the main's input. This effectively merges the individual communication shift registers of each sub to form a single larger combined shift register that shifts data through the chain. This configuration only requires a single line from the main, rather than a separate line for each sub.[6]
In addition to using SPI-specific subs, daisy-chained SPI can include discrete shift registers for more pins of inputs (e.g. using the parallel-in serial-out 74
Expander configurations use SPI-controlled addressing units (e.g. binary decoders, demultiplexers, or shift registers) to add chip selects.
For example, one can be used for transmitting to a SPI-controlled demultiplexer an index number controlling its select signals, while another is routed through that demultiplexer according to that index to select the desired sub.[9]
SPI is used to talk to a variety of peripherals, such as
Board real estate and wiring savings compared to a parallel bus are significant, and have earned SPI a solid role in embedded systems. That is true for most system-on-a-chip processors, both with higher-end 32-bit processors such as those using ARM, MIPS, or PowerPC and with lower-end microcontrollers such as the AVR, PIC, and MSP430. These chips usually include SPI controllers capable of running in either main or sub mode. In-system programmable AVR controllers (including blank ones) can be programmed using SPI.[10]
Chip or FPGA based designs sometimes use SPI to communicate between internal components; on-chip real estate can be as costly as its on-board cousin. And for high-performance systems, FPGAs sometimes use SPI to interface as a sub to a host, as a main to sensors, or for flash memory used to bootstrap if they are SRAM-based.
The full-duplex capability makes SPI very simple and efficient for single main/single sub applications. Some devices use the full-duplex mode to implement an efficient, swift data stream for applications such as digital audio, digital signal processing, or telecommunications channels, but most off-the-shelf chips stick to half-duplex request/response protocols.
SPI is a de facto standard. However, the lack of a formal standard is reflected in a wide variety of protocol options. Some devices are transmit-only; others are receive-only. Chip selects are sometimes active-high rather than active-low. Some devices send the least-significant bit first. Signal levels depend entirely on the chips involved. And while the baseline SPI protocol has no command codes, every device may define its own protocol of command codes. Some variations are minor or informal, while others have an official defining document and may be considered to be separate but related protocols.
Motorola in 1983 listed[11] three 6805 8-bit microcomputers that have an integrated "Serial Peripheral Interface", whose functionality is described in a 1984 manual.[12]
Motorola's 1987 Application Node AN991 "Using the Serial Peripheral Interface to Communicate Between Multiple Microcomputers"[13] (now under NXP, last revised 2002[14]) informally serves as the "official" defining document for SPI.
Some devices have timing variations from Motorola's CPOL/CPHA modes. Sending data from sub to main may use the opposite clock edge as main to sub. Devices often require extra clock idle time before the first clock or after the last one, or between a command and its response.
Some devices have two clocks, one to read data, and another to transmit it into the device. Many of the read clocks run from the chip select line.
Different transmission word sizes are common. Many SPI chips only support messages that are multiples of 8 bits. Such chips can not interoperate with the JTAG or SGPIO protocols, or any other protocol that requires messages that are not multiples of 8 bits.
Some devices don't use chip select, and instead manage protocol state machine entry/exit using other methods.
Anyone needing an external connector for SPI defines their own or uses another standard connection such as: UEXT, Pmod, various JTAG connectors, Secure Digital card socket, etc.
Some devices require an additional flow control signal from sub to main, indicating when data is ready. This leads to a 5-wire protocol instead of the usual 4. Such a ready or enable signal is often active-low, and needs to be enabled at key points such as after commands or between words. Without such a signal, data transfer rates may need to be slowed down significantly, or protocols may need to have dummy bytes inserted, to accommodate the worst case for the sub response time. Examples include initiating an ADC conversion, addressing the right page of flash memory, and processing enough of a command that device firmware can load the first word of the response. (Many SPI mains do not support that signal directly, and instead rely on fixed delays.)
SafeSPI[15] is an industry standard for SPI in automotive applications. Its main focus is the transmission of sensor data between different devices.
In electrically noisy environments, since SPI has few signals, it can be economical to reduce the effects of common mode noise by adapting SPI to use low-voltage differential signaling.[16] Another advantage is that the controlled devices can be designed to loop-back to test signal integrity.[17]
A Queued Serial Peripheral Interface (QSPI; different to but has same abbreviation as Quad SPI described in) is a type of SPI controller that uses a data queue to transfer data across an SPI bus.[18] It has a wrap-around mode allowing continuous transfers to and from the queue with only intermittent attention from the CPU. Consequently, the peripherals appear to the CPU as memory-mapped parallel devices. This feature is useful in applications such as control of an A/D converter. Other programmable features in Queued SPI are chip selects and transfer length/delay.
SPI controllers from different vendors support different feature sets; such direct memory access (DMA) queues are not uncommon, although they may be associated with separate DMA engines rather than the SPI controller itself, such as used by Multichannel Buffered Serial Port (MCBSP). Most SPI main controllers integrate support for up to four chip selects, although some require chip selects to be managed separately through GPIO lines.
Note that Queued SPI is different from Quad SPI, and some processors even confusingly allow a single "QSPI" interface to operate in either quad or queued mode![19]
Microwire,[20] often spelled μWire, is essentially a predecessor of SPI and a trademark of National Semiconductor. It's a strict subset of SPI: half-duplex, and using SPI mode 0. Microwire chips tend to need slower clock rates than newer SPI versions; perhaps 2 MHz vs. 20 MHz. Some Microwire chips also support a three-wire mode.
Microwire/Plus[21] is an enhancement of Microwire and features full-duplex communication and support for SPI modes 0 and 1. There was no specified improvement in serial clock speed.
Three-wire variants of SPI restricted to a half-duplex mode use a single bidirectional data line called SISO (sub out/sub in) or MOMI (main out/main in) instead of SPI's two unidirectional lines (MOSI and MISO). Three-wire tends to be used for lower-performance parts, such as small EEPROMs used only during system startup, certain sensors, and Microwire. Few SPI controllers support this mode, although it can be easily bit-banged in software.
For instances where the full-duplex nature of SPI is not used, an extension uses both data pins in a half-duplex configuration to send two bits per clock cycle. Typically a command byte is sent requesting a response in dual mode, after which the MOSI line becomes SIO0 (serial I/O 0) and carries even bits, while the MISO line becomes SIO1 and carries odd bits. Data is still transmitted most-significant bit first, but SIO1 carries bits 7, 5, 3 and 1 of each byte, while SIO0 carries bits 6, 4, 2 and 0.
This is particularly popular among SPI ROMs, which have to send a large amount of data, and comes in two variants:[22] [23]
Quad SPI (QSPI; different to but has same abbreviation as Queued-SPI described in) goes beyond dual SPI, adding two more I/O lines (SIO2 and SIO3) and sends 4 data bits per clock cycle. Again, it is requested by special commands, which enable quad mode after the command itself is sent in single mode.
Further extending quad SPI, some devices support a "quad everything" mode where all communication takes place over 4 data lines, including commands.[24] This is variously called "QPI" (not to be confused with Intel QuickPath Interconnect) or "serial quad I/O" (SQI)[25]
This requires programming a configuration bit in the device and requires care after reset to establish communication.
In addition to using multiple lines for I/O, some devices increase the transfer rate by using double data rate transmission.[26] [27]
See main article: articles and JTAG. Although there are some similarities between SPI and the JTAG (IEEE 1149.1-2013) protocol, they are not interchangeable. JTAG is specifically intended to provide reliable test access to the I/O pins from an off-board controller with less precise signal delay and skew parameters, while SPI has many varied applications. While not strictly a level sensitive interface, the JTAG protocol supports the recovery of both setup and hold violations between JTAG devices by reducing the clock rate or changing the clock's duty cycles. Consequently, the JTAG interface is not intended to support extremely high data rates.[28]
See main article: articles and SGPIO. SGPIO is essentially another (incompatible) application stack for SPI designed for particular backplane management activities. SGPIO uses 3-bit messages.
Intel has developed a successor to its Low Pin Count (LPC) bus that it calls the Enhanced Serial Peripheral Interface (eSPI) bus. Intel aims to reduce the number of pins required on motherboards and increase throughput compared to LPC, reduce the working voltage to 1.8 volts to facilitate smaller chip manufacturing processes, allow eSPI peripherals to share SPI flash devices with the host (the LPC bus did not allow firmware hubs to be used by LPC peripherals), tunnel previous out-of-band pins through eSPI, and allow system designers to trade off cost and performance.[29] [30]
An eSPI bus can either be shared with SPI devices to save pins or be separate from an SPI bus to allow more performance, especially when eSPI devices need to use SPI flash devices.
This standard defines an Alert# signal that is used by an eSPI sub to request service from the main. In a performance-oriented design or a design with only one eSPI sub, each eSPI sub will have its Alert# pin connected to an Alert# pin on the eSPI main that is dedicated to each sub, allowing the eSPI main to grant low-latency service, because the eSPI main will know which eSPI sub needs service and will not need to poll all of the subs to determine which device needs service. In a budget design with more than one eSPI sub, all of the Alert# pins of the subs are connected to one Alert# pin on the eSPI main in a wired-OR connection, which requires the main to poll all the subs to determine which ones need service when the Alert# signal is pulled low by one or more peripherals that need service. Only after all of the devices are serviced will the Alert# signal be pulled high due to none of the eSPI subs needing service and therefore pulling the Alert# signal low.
This standard allows designers to use 1-bit, 2-bit, or 4-bit communications at speeds from 20 to 66 MHz to further allow designers to trade off performance and cost.
All communications that were out-of-band of LPC like general-purpose input/output (GPIO) and System Management Bus (SMBus) are tunneled through eSPI via virtual wire cycles and out-of-band message cycles respectively in order to remove those pins from motherboard designs using eSPI.
This standard supports standard memory cycles with lengths of 1 byte to 4 kilobytes of data, short memory cycles with lengths of 1, 2, or 4 bytes that have much less overhead compared to standard memory cycles, and I/O cycles with lengths of 1, 2, or 4 bytes of data which are low overhead as well. This significantly reduces overhead compared to the LPC bus, where all cycles except for the 128-byte firmware hub read cycle spends more than one-half of all of the bus's throughput and time in overhead. The standard memory cycle allows a length of anywhere from 1 byte to 4 kilobytes in order to allow its larger overhead to be amortised over a large transaction. eSPI subs are allowed to initiate bus master versions of all of the memory cycles. Bus master I/O cycles, which were introduced by the LPC bus specification, and ISA-style DMA including the 32-bit variant introduced by the LPC bus specification, are not present in eSPI. Therefore, bus master memory cycles are the only allowed DMA in this standard.
eSPI subs are allowed to use the eSPI main as a proxy to perform flash operations on a standard SPI flash memory sub on behalf of the requesting eSPI sub.
64-bit memory addressing is also added, but is only permitted when there is no equivalent 32-bit address.
The Intel Z170 chipset can be configured to implement either this bus or a variant of the LPC bus that is missing its ISA-style DMA capability and is underclocked to 24 MHz instead of the standard 33 MHz.[31]
Single-board computers may provide pin access to SPI hardware units. For instance, the Raspberry Pi's J8 header exposes at least two SPI units that can be used via Linux drivers or python.
There are a number of USB adapters that allow a desktop PC or smartphone with USB to communicate with SPI chips (e.g. FT221xs[32]). They are used for embedded systems, chips (FPGA, ASIC, and SoC) and peripheral testing, programming and debugging. Many of them also provide scripting or programming capabilities (e.g. Visual Basic, C/C++, VHDL).
The key SPI parameters are: the maximum supported frequency for the serial interface, command-to-command latency, and the maximum length for SPI commands. It is possible to find SPI adapters on the market today that support up to 100 MHz serial interfaces, with virtually unlimited access length.
SPI protocol being a de facto standard, some SPI host adapters also have the ability of supporting other protocols beyond the traditional 4-wire SPI (for example, support of quad-SPI protocol or other custom serial protocol that derive from SPI[33]).
See main article: articles and Logic analyzer. Logic analyzers are tools which collect, timestamp, analyze, decode, store, and view the high-speed waveforms, to help debug and develop. Most logic analyzers have the capability to decode SPI bus signals into high-level protocol data with human-readable labels.
SPI waveforms can be seen on analog channels (and/or via digital channels in mixed-signal oscilloscopes).[34] Most oscilloscope vendors offer optional support for SPI protocol analysis (both 2-, 3-, and 4-wire SPI) with triggering.
The term "master" is commonly used to identify the main device and "slave" for peripheral (sub) devices. These terms reflect how the main device is responsible for driving the serial clock and initiating communication: peripheral devices are only able to communicate when the main device is driving the clock.
Various more contemporary alternative names for each of the four signals have been proposed:
Microchip uses "Host" and "Client" though keeps the abbreviation MOSI and MISO.[38]