Intel Stratix 10 Configuration User Guide

RSS

UG-S10CONFIG | 2019.07.19

Intel Stratix 10 Configuration User Guide

Intel Stratix 10 Configuration Overview

All Intel^® Stratix^® 10 devices include a Secure Device Manager (SDM) to manage FPGA configuration and security. The SDM provides a failsafe, strongly authenticated, programmable security mode for device configuration. Previous FPGA families include a fixed state machine to manage device configuration.

The Intel^® Quartus^® Prime software also provides flexible and robust security features to protect sensitive data, intellectual property, and the device itself under both remote and physical attacks. Configuration bitstream authentication ensures that the firmware and configuration bitstream are from a trusted source. Encryption prevents theft of intellectual property. The Intel^® Quartus^® Prime software also compresses FPGA bitstreams, reducing memory utilization.

Intel describes configuration schemes from the point-of-view of the FPGA. Intel^® Stratix^® 10 devices support active and passive configuration schemes. In active configuration schemes the FPGA acts as the master and the external memory acts as a slave device. In passive configuration schemes an external host acts as the master and controls configuration. The FPGA acts as the slave device. All Intel^® Stratix^® 10 configuration schemes support design security, and partial reconfiguration. All Intel^® Stratix^® 10 active configuration schemes support remote system update (RSU) with quad SPI flash memory. To implement RSU in passive configuration schemes, an external controller must store and drive the configuration bitstream.

Intel^® Stratix^® 10 devices support the following configuration schemes:

Avalon^® Streaming ( Avalon^® -ST)
JTAG
Configuration via Protocol (CvP)
Active Serial (AS) normal and fast modes
Secure Digital and Multi Media Card (SD/MMC)

Table 1. Intel^® Stratix^® 10 Configuration Data Width, Clock Rates, and Data Rates
Configuration Scheme		Data Width (bits)	MSEL[2:0]
Passive	Avalon^® -ST	32	000
		16	101
		8	110
	JTAG	1	111
	Configuration via Protocol (CvP)	x1, x2, x4, x8, x16 lanes	001¹
Active	SD/MMC	4/8	100
	AS - fast mode	4	001
	AS - normal mode	4	011

Avalon-ST

The Avalon^® -ST configuration scheme is a passive configuration scheme. Avalon^® -ST is the fastest configuration scheme for Intel^® Stratix^® 10 devices. Avalon^® -ST configuration supports x8, x16, and x32 modes. The x16 and x32 bit modes use general-purpose I/Os (GPIOs) for configuration. The x8 bit mode uses dedicated SDM I/O pins.

Note: The AVST_data[15:0], AVST_data[31:0], AVST_clk, and AVST_valid use dual-purpose GPIOs. You can use these pins as regular I/Os after the device enters user mode.

Avalon^® -ST supports backpressure using the AVST_READY and AVST_VALID pins. Because the time to decompress the incoming bitstream varies, backpressure support is necessary to transfer data to the Intel^® Stratix^® 10 device. For more information about the Avalon^® -ST refer to the Avalon^® Interface Specifications.

JTAG

You can configure the Intel^® Stratix^® 10 device using the dedicated JTAG pins. The JTAG port provides seamless access to many useful tools and functions. In addition to configuring the Intel^® Stratix^® 10, you use the JTAG port for debugging with Signal Tap or the System Console tools.

The JTAG port has the highest priority and overrides the MSEL pin settings. Consequently, you can configure the Intel^® Stratix^® 10 device over JTAG even if the MSEL pin specify a different configuration scheme unless you disabled JTAG for security reasons.

CvP

CvP uses an external PCIe* host device as a Root Port to configure the Intel^® Stratix^® 10 device over the PCIe* link. You can specify up to a x16 PCIe* link. Typically, the bitstream compression ratio and the SDM input buffer data rate, not the PCIe* link, limit the configuration data rate Intel^® Stratix^® 10 devices support two CvP modes, CvP init and CvP update.

CvP initialization process includes the following two steps:

CvP configures the FPGA periphery image which includes I/O information and hard IP blocks, including the PCIe* IP. CvP uses quad SPI memory and AS x4 to configure the peripheral image. Because the PCIe* IP is in the periphery image, PCIe* link training establishes the PCIe* link of the CvP PCIe* IP before the core fabric configures.
The host device uses the CvP PCIe* link to configure your design in the core fabric.

CvP update mode updates the FPGA core image using the PCIe* link already established from a previous full chip configuration or CvP init configuration. After the Intel^® Stratix^® 10 enters user mode, you can use the CvP update mode to reconfigure the FPGA fabric. This mode has the following advantages:

Allows reprogramming of the core to run different algorithms.
Provides a mechanism for standard updates as a part of a release process.
Customizes core processing for different components that are part of a complex system.
For both CvP Init and CvP Update modes, the maximum data rate depends on the PCIe* generation and number of lanes.

For Intel^® Stratix^® 10 SoC devices, CvP is only supported in FPGA configuration first mode.

For more information refer to the Intel^® Stratix^® 10 Configuration via Protocol (CvP) Implementation User Guide.

AS Normal Mode

Active Serial x4 or AS x4 or Quad SPI is an active configuration scheme that supports flash memories capable of three- and four-byte addressing. Upon power up, the SDM boots from a boot ROM which uses three-byte addressing to load the configuration firmware from the Quad SPI flash. After the configuration firmware loads, the Quad SPI flash operates using four-byte addressing for the rest of the configuration process. This mode supports Intel's serial flash configuration memory solution for the following third-party flash devices:

Micron MT25QU128, MT25QU256, MT25QU512, MT25QU01G, MT25QU02G
Macronix MX25U128, MX25U256, MX25U512, MX66U512, MX66U1G, MX66U2G

Refer to the Supported Flash Devices for Intel^® Stratix^® 10 Devices for complete list of supported flash devices.

AS Fast Mode

The only difference between AS normal mode and fast mode is speed. Use AS fast mode when configuration timing is a concern. This mode does not delay for 10 ms before beginning configuration. Use this mode to meet the 100 ms of power up requirement for PCIe* or for other systems with strict timing requirements.

In AS fast mode, the power-on sequence must ensure that the AS device out out of reset before the SDM because the Intel^® Stratix^® 10 accesses flash memory immediately after exiting reset. The power supply must be able to provide an equally fast ramp up for the Intel^® Stratix^® 10 device and the external AS x4 flash devices. Failing to meet this requirement causes the SDM to assume the memory is missing . Consequently, configuration fails.

Refer to the Intel^® Stratix^® 10 Device Family Pin Connection Guidelines and AN692: Power Sequencing Considerations for Intel^® Cyclone^® 10 GX, Intel^® Arria^® 10, and Intel^® Stratix^® 10 Devices for additional details.

SD/MMC

SD/MMC is an active configuration scheme. The Intel^® Stratix^® 10 SDM can initiate configuration from SD, Secure Digital High Capacity (SDHC*). Secure Digital Extended Capacity (SDXC*), MMC cards, and eMMC devices. The advantages of this mode are cost, capacity, availability, portability, and compatibility. Because Intel^® Stratix^® 10 devices operate at 1.8 volt an intermediate voltage level shifter may be required to interface with the higher voltage I/Os in SD/MMC devices.

Note: The SD/MMC configuration scheme is not supported in the current release.

¹ Before you can use CvP you must configure either the periphery image or full image configuration via the AS scheme. Then you can configure the core image using CvP.

Configuration and Related Signals

The following figure shows the configuration interfaces and configuration-related device functions. Pins shown in dark blue use dedicated SDM I/Os. Pins shown in black use general purpose I/Os (GPIOs). Pins shown in red are dedicated JTAG I/Os.

You specify SDM I/O pin functions using the Device > Configuration > Device and Pin Options dialog box in the Intel^® Quartus^® Prime software.

Figure 1. Intel^® Stratix^® 10 Configuration Interfaces

This user guide discusses most of the interfaces shown in the figure. Refer to the separate Intel^® Stratix^® 10 Configuration via Protocol (CvP) Implementation User Guide and Intel^® Stratix^® 10 Power Management User Guide for more information about those features.

Intel Download Cables Supporting Configuration in Intel Stratix 10 Devices

Intel provides the following cables to download your design to the Intel^® Stratix^® 10 device on the PCB. Download cables support prototyping activity by providing detailed debug messages via Intel^® Quartus^® Prime Programmer. You must use Intel download cables for advanced debugging using the Signal Tap logic analyzer or the System Console tools.

Table 2. Intel^® Stratix^® 10-Supported Download Cable Capabilities
Download Cable	Protocol Support Intel^® Stratix^® 10 Device	Cable Connection to PCB
Intel^® FPGA Download Cable II (formerly the USB-Blaster II)	JTAG, AS	10-pin female plug 3M Part number: 2510-6002UB
Intel FPGA Ethernet Cable (formerly the Ethernet Blaster II)	JTAG, AS	10-pin female plug

The Intel FPGAs and Programmable Devices / Download Cables provides more information about the download cables and includes links to the user guides for all cables listed in the table above.

Intel Stratix 10 Configuration Architecture

The Secure Device Manager (SDM) is a triple-redundant processor-based module that manages configuration and the security features of Intel^® Stratix^® 10 devices. The SDM is available on all Intel^® Stratix^® 10 FPGAs and SoC devices.

The block diagram below provides an overview of the Intel^® Stratix^® 10 configuration architecture which includes the following blocks:

SDM: More information about the SDM is contained in later sections.
Configuration network: The SDM uses this dedicated, parallel configuration network to distribute the configuration bitstream to Local Sector Managers (LSMs). You cannot access this network.
LSMs: The LSM is a microprocessor. Each configuration sector includes an LSM. The LSM parses configuration bitstream and configures the logic elements for its sector. After configuration, the LSM performs the following functions:
- Monitors for single event upsets at the sector level
- Processes responses to single event upsets (SEUs)
- Performs hashing or integrity checks in real time
Specific blocks for Intel^® Stratix^® 10 variants:
- SX devices include the hard processor system (HPS) in addition to FPGA logic.
- MX devices include a High Bandwidth Memory (HBM2) in addition to FPGA logic.
- GX devices include FPGA logic and L- and H-Tile transceivers.
- TX devices include FPGA logic and E- and H-Tile transceivers.

Figure 2. Intel^® Stratix^® 10 Configuration Architecture Block Diagram

Secure Device Manager

The SDM comprises peripherals, cryptographic IP and sensors, boot ROM, triple-redundant lockstep processors, and other blocks shown the block diagram below. The SDM performs and manages the following security functions:

Configuration bitstream authentication: After power-on during startup, the SDM triple-redundant lockstep processors run code from the boot ROM. The boot ROM code authenticates the Intel-generated configuration firmware and configuration bitstream, ensuring that configuration bitstream is from a trusted source.
Encryption: Encryption protects the configuration bitstream or confidential data from unauthorized third-party access.
Side channel attack protection: Side channel attack protection guards AES Key and confidential data under non-intrusive attacks.
Integrity checking: Integrity checking verifies that an accidental event has not corrupted the configuration bitstream. This function is active, even if you do not enable authentication.

Note: These security features are available in Intel^® Stratix^® 10 devices that support advanced security. The ordering codes for Intel^® Stratix^® 10 devices that include advanced security features includes the AS (Advanced Security) suffix. Contact your Intel Programmable Solutions representative to get additional information about Intel^® Stratix^® 10 device security features.

Figure 3. SDM Block Diagram

Here is an overview of the additional functions the SDM controls:

The Power Management block consists of a voltage and temperature sensor which enables the SmartVID feature via an external PMBus voltage regulator when you select -V devices.
The AES/SHA and other Crypto Accelerator blocks implement secure configuration and boot.
The Key Vault provides volatile and non-volatile cryptographic key storage. To mitigate potential side-channel attacks, crypto functions that use keys require the keys a special hardware storage mechanism.
The AS and SD/MMC configuration flash controllers enable active configuration schemes via dedicated SDM pins.
The x8 Avalon^® -ST configuration scheme uses SDM I/O pins. The x16 and x32 Avalon^® -ST configuration schemes use dedicated SDM I/O pins and dual-purpose I/O pins. Refer to the SDM Pin Mapping for more information.
To reduce configuration file size and support smaller memory sizes, and enable faster configuration, the Intel^® Quartus^® Prime software compresses the configuration data. Intel^® Stratix^® 10 devices all compress the configuration bitstream. You cannot disable this feature. The decompression block in the SDM decompresses both encrypted and non-encrypted configuration files.
A specific PCIe* block included in the Intel^® Stratix^® 10 device supports CvP.

Updating the SDM Firmware

When you generate a configuration bitstream using the File > Programming File Generator menu item, the Intel^® Quartus^® Prime combines the SDM firmware with your .sof.

Depending on the configuration scheme you specify the resulting file can be in any of the following formats:

Raw Binary File, .rbf
Programmer Object File, .pof
JTAG Indirect Configuration, .jic
Raw Programming Data, .rpd
Jam*Standard Test and Programming Language (STAPL) STAPL, .jam
Jam* Byte Code, .jbc

Newer versions of the Intel^® Quartus^® Prime software typically include new or updated SDM features.

Intel recommends using the latest version of the Intel^® Quartus^® Prime software when regenerating your configuration bitstream. Even if your .sof has not changed, the Programming File Generator adds the latest SDM firmware which may address known problems or add new features.

Specifying Boot Order for Intel Stratix 10 SoC Devices

For Intel^® Stratix^® 10 SoC devices you can specify the configuration order, choosing either the FPGA First or the Hard Processor System (HPS) First options. When you select the FPGA First option, the SDM fully configures the FPGA, then configures the HPS SDRAM pins, loads the HPS first stage boot loader (FSBL) and takes the HPS out of reset. When you select the HPS First option, the SDM first configures the HPS SDRAM pins, loads the HPS FSBL and takes the HPS out of reset. Then the HPS configures the FPGA I/O and FPGA fabric at a later time. The HPS First option has the following advantages:

Minimizes the amount of SDM flash memory required
Minimizes the amount of time it takes for the HPS software to be up and running
Supports FPGA reconfiguration while the HPS is running

For more information about specifying configuration order refer to the FPGA Configuration First Mode and HPS Boot First Mode chapters in the Intel^® Stratix^® 10 SoC FPGA Boot User Guide.

Intel Stratix 10 Configuration Details

Intel Stratix 10 Configuration Timing Diagram

Initial Configuration Timing

Figure 4. Configuration, Reconfiguration, and Error Timing Diagram

The SDM drives Intel^® Stratix^® 10 device configuration.

The first section of the figure shows the expected timing for initial configuration after a normal power-on reset. The initial state of the nCONFIG and nSTATUS signals is low.

The numbers in the Initial Configuration part of the timing diagram mark the following events:

The SDM boots up and samples the MSEL signals to determine the specified FPGA configuration scheme. The SDM does not sample the MSEL pins again until the next power cycle.
With the nCONFIG signal low, the SDM enters Idle mode after booting.
When the external host drives nCONFIG signal high, the SDM initiates configuration. The SDM drives the nSTATUS signal high, signaling the beginning of FPGA configuration. The SDM receives the configuration bitstream on the interface that the MSEL bus specified in Step 1 The diagram shows AVST_READY and AVST_VALID continuously high. It is possible for AVST_READY to deassert which would require AVST_VALID to deassert within six cycles.
The SDM drives the CONF_DONE signal high, indicating the SDM received the bitstream successfully.
When the Intel^® Stratix^® 10 device asserts INIT_DONE the FPGA enters user mode. GPIO pins exit the high impedance state. The time between the assertion of CONF_DONE and INIT_DONE is variable. For FPGA First configuration, INIT_DONE asserts after initialization of the FPGA fabric, including registers and state machines. For HPS first configuration, the HPS application controls the time between CONF_DONE and INIT_DONE. INIT_DONE does not assert until after the software running on the HPS such as uboot or the operating system (OS) initiates the configuration.
The entire device does not enter user mode simultaneously. Intel requires you to include the Intel Stratix 10 Reset Release IP to hold your application logic in the reset state until the entire FPGA fabric is in user mode. Failure to include this IP in your design may result in intermittent application logic failures.

Reconfiguration Timing

The second event the timing diagram illustrates the Intel^® Stratix^® 10 device reconfiguration. If you change the MSEL setting after power-on, you must power-cycle the Intel^® Stratix^® 10. Power cycling forces the SDM to sample the MSEL pins before reconfiguring the device.

The numbers in the Reconfiguration part of the timing diagram mark the following events:

The external host drives nCONFIG signal low.
The SDM initiates device cleaning.
The SDM drives the nSTATUS signal low when device cleaning is complete.
The external host drives the nCONFIG signal high to initiate reconfiguration.
The SDM drives the nSTATUS signal high signaling the device is ready for reconfiguration.

Configuration Error

The numbers in the Configuration Error part of the timing diagram mark the following events:

The SDM drives nSTATUS signal low for 1 ms ±50% to indicate a configuration error. The Intel^® Stratix^® 10 devices does not assert CONF_DONE indicating that configuration did not complete successfully.
The SDM enters the error state.
The SDM enters the idle state. The external host deasserts nCONFIG. The device is ready for reconfiguration by driving a low to high transition on nCONFIG. You can also power cycle the device by following the device power down sequence.
Note: The nCONFIG signal can only change levels when it has the same value as nSTATUS. This restriction means that when nSTATUS = 1, nCONFIG can transition from 1 to 0. When nSTATUS = 0, nCONFIG can transition from 0 to 1. Apart from error reporting, nSTATUS only changes to follow nCONFIG.

Power Supply Status

The power-on reset (POR) holds the Intel^® Stratix^® 10 device in the reset state until the power supply outputs are within the recommended operating range. t_RAMP defines the maximum power supply ramp time. If POR does not meet the t_RAMP time, the Intel^® Stratix^® 10 device I/O pins and programming registers remain tri-stated.

For more information about POR refer to the Intel^® Stratix^® 10 Power Management User Guide. For more information about t_RAMP refer to the Intel^® Stratix^® 10 datasheet.

Configuration Flow Diagram

This topic describes the configuration flow for Intel^® Stratix^® 10 devices.

Figure 5. Intel^® Stratix^® 10 FPGA Configuration Flow

Power Up

The Intel^® Stratix^® 10 power supplies power following the guidelines in the Power-Up Sequence Requirements for Intel^® Stratix^® 10 Devices section of the Intel^® Stratix^® 10 Power Management User Guide.
A device-wide power-on reset (POR) asserts after the power supplies reach the correct operating voltages. The external power supply ramp must not be slower than the minimum ramping rate until the supplies reach the operating voltage.
The SDM_IO0, SDM_IO8, and SDM_IO16 pins remain low internally. Internal circuitry pulls the remaining SDM_IO pins to a weak high.

SDM Startup

The SDM samples the MSEL pins during power-on.
If MSEL is set to JTAG, the SDM remains in the Startup state.
The SDM runs firmware stored in the on-chip boot ROM and enters the Idle state until the host drives nCONFIG high. The host should not drive nCONFIG high before all clocks are stable.

Idle

The SDM remains in IDLE state until the external host initiates configuration by driving the nCONFIG pin from low to high. Alternatively, the SDM enters the idle state after it exits the error state.
The SDM also enters the Idle state after it exits the error state.

Configuration Start

After the SDM receives a configuration initiation request (nCONFIG = HIGH), the SDM signals the beginning of configuration by driving the nSTATUS pin high.
Upon receiving configuration data, the SDM performs authentication, decryption and decompression.
The nCONFIG pin remains high during configuration and in user mode. The host monitors the nSTATUS pin continuously for configuration errors.

Configuration Pass

The SDM drives the CONF_DONE pin high after successfully receiving full bitstream.
The CONF_DONE pin signals an external host that bitstream transfer is successful.

Configuration Error

A low pulse on the nSTATUS pin indicates a configuration error.
Errors usually require reconfiguration.
After a low pulse indicating an error, configuration stops. The nSTATUS pin remains high.
Following an error, the SDM drives nSTATUS low after the external host drives nCONFIG low.
The device enters Idle state after the nSTATUS pin recovers to initial pre-configuration low state.

User Mode

The SDM drives the INIT_DONE pin high after initializing internal registers and releases GPIO pins from the high impedance state. The device enters user mode. The entire device does not enter user mode at the same instant.
The nCONFIG pin should remain high in user mode.
You may re-configure the device by driving nCONFIG pin from low to high.

Device Clean

In the Device Clean state the design stops functioning.
Device cleaning zeros out all configuration data.
The Intel^® Stratix^® 10 device drives CONF_DONE and INIT_DONE low.
The SDM drives the nSTATUS pin low when device cleaning completes.

JTAG Configuration

Note: You can perform JTAG configuration anytime from any state except the power-on and SDM startup state. The Intel^® Stratix^® 10 device cancels the previous configuration and accepts the reconfiguration data from the JTAG interface. The nCONFIG signal must be held in a stable state during JTAG configuration. A falling edge on the nCONFIG signal cancels the JTAG configuration.

Note: The SDM only samples the MSEL pins at power-on and initiates bitstream configuration using the configuration scheme specified at power-on.

Intel Stratix 10 Reset Release IP

Intel requires that you either use the Intel^® Stratix^® 10 Reset Release IP or the INIT_DONE signal routed back in through a pin to hold your design in reset until configuration is complete.

The Reset Release IP is available in the Intel^® Quartus^® Prime Software. This IP consists of a single output signal, nINIT_DONE. The nINIT_DONE signal is the core version of the INIT_DONE pin and has the same function in both FPGA First and HPS First configuration modes. Intel recommends that you use the nINIT_DONE signal to hold your design in reset.

When you instantiate the Reset Release IP in your design, the SDM drives the nINIT_DONE signal. Consequently, the IP does not consume any FPGA fabric resources, but does require routing resources.

Figure 6. Intel^® Stratix^® 10 Reset Release IP nINIT_DONE Internal Connection

Figure 7. Intel^® Stratix^® 10 Reset Release IP INIT_DONE External Connection

Additional Clock Requirements for HPS, PCIe , eSRAM, and HBM2

The Intel^® Stratix^® 10 device has additional clock requirements for PCIe* , HPS EMIF, eSRAM, and the High Bandwidth Memory (HBM2) IP.

To avoid configuration failures, the Intel^® Stratix^® 10 device requires additional clocks for the PCIe* , HPS EMIF, eSRAM, and the HBM2 IP. You must provide a free-running, stable reference clock to these blocks before configuration begins. This reference clock is in addition to the configuration clock requirements for an internal or external oscillator described in OSC_CLK_1 Requirements. These blocks and their specific clock names are as listed below.

HBM2: pll_ref_clk and ext_core_clk
eSRAM: CLK_ESRAM_[0,1]p and CLK_ESRAM_[0,1]n
HPS EMIF: pll_ref_clk
L- and H-Tile PCIe* channels: REFCLK_GXB

Note: The transceiver power supplies must be a nominal levels for successful configuration. You can use the V_CC and V_CCP power supplies for limited transceiver channel testing. Designs that include many transceivers require an auxiliary power supply to operate reliably.

Intel Stratix 10 Configuration Pins

The Intel^® Stratix^® 10 device uses SDM_IO pins for device configuration. Here are the states that SDM I/O go through:

After power-on, SDM I/O pins 0, 8, and 16 have weak pull-downs. All other SDM I/O pins have weak pull-ups. (These initial voltage levels ensure correct operation during initialization. For example, for Avalon^® -ST configuration SDM_IO8 is the Avalon^® -ST ready signal which should not be asserted until the device reaches the FPGA Configuration state.)
The Boot ROM samples MSEL to determine the configuration scheme you specified and drives pins required for that configuration scheme. SDM I/O pins not required for the your configuration scheme remain weakly pulled up.
In approximately 10 ms the SDM I/O pins take on the state that your design specifies.
After device cleaning, the SDM reads pin information from firmware and restores the pin states that your design specifies. If you reconfigure the device, the SDM uses the updated pin information when initializing the device.

SDM Pin Mapping

You can use SDM I/O pins for configuration and other functions such as power management and SEU detection. You specify SDM I/O pin functions using the Device > Configuration > Device and Pin Options dialog box in the Intel^® Quartus^® Prime software. All SDM input signals include Schmitt triggers. All SDM outputs are open collector.

Fixed SDM I/O Pin Assignments for Avalon^® -ST x8 and AS x4

The Avalon^® -ST x8 and AS x4 configuration schemes use the dedicated SDM I/O pin assignments listed in in the table below. Use the assignments in this table for MSEL and AVSTx8_DATA0 to AVSTx8_DATA8 and AS x4.

Table 3. SDM Pin Mapping for Avalon^® -ST x8 and AS x4
SDM Pins	MSEL Function	Configuration Source Function
SDM Pins	MSEL Function	Avalon^® -ST x8	AS x4
`SDM_IO0`	—	—	—
`SDM_IO1`	—	`AVSTx8_DATA2`	`AS_DATA1`
`SDM_IO2`	—	`AVSTx8_DATA0`	`AS_CLK`
`SDM_IO3`	—	`AVSTx8_DATA3`	`AS_DATA2`
`SDM_IO4`	—	`AVSTx8_DATA1`	`AS_DATA0`
`SDM_IO5`	`MSEL0`	—	`AS_nCSO`
`SDM_IO6`	—	`AVSTx8_DATA4`	`AS_DATA3`
`SDM_IO7`	`MSEL1`	—	`AS_nCS2`
`SDM_IO8`	—	`AVST_READY`	`AS_nCS3`
`SDM_IO9`	`MSEL2`	—	`AS_nCS1`
`SDM_IO10`	—	`AVSTx8_DATA7`	—
`SDM_IO11`	—	`AVSTx8_VALID`	—
`SDM_IO12`	—	—	—
`SDM_IO13`	—	`AVSTx8_DATA5`	—
`SDM_IO14`	—	`AVSTx8_CLK`	—
`SDM_IO15`	—	`AVSTx8_DATA6`	—
`SDM_IO16`	—	—	—

MSEL Settings

After power-on MSEL[2:0] pins specify the configuration scheme for Intel^® Stratix^® 10 devices. Use 4.7-kΩ resistors to pull the MSEL[2:0] pins up to V_{CCIO_SDM} or down to ground as required by the MSEL[2:0] setting for your configuration scheme.

Figure 8. MSEL Pull-Up and Pull-Down Circuit Diagram

Table 4. MSEL Settings for Each Configuration Scheme of Intel^® Stratix^® 10 Devices
Configuration Scheme	MSEL[2:0]
Avalon-ST (x32)	000
Avalon-ST (x16)	101
Avalon-ST (x8)	110
AS (Fast mode – for CvP)²	001
AS (Normal mode)	011
SD/MMC x4/x8	100
JTAG only³	111

You must also specify the configuration scheme on the Configuration page of the Device and Pin Options dialog box in the Intel^® Quartus^® Prime Software.

Figure 9. Specify Configuration Scheme to Specify MSEL Value

² If you use AS Fast mode and are not concerned about 100 ms PCIe* linkup, you must still ramp the V_{CCIO_SDM} supply within 18 ms. This ramp-up requirement ensures that the AS x4 device is within its operating voltage range when the Intel^® Stratix^® 10 device begins to access it.

³ JTAG configuration works with any valid MSEL settings, unless disabled for security.

Device Configuration Pins for Optional Configuration Signals

All configuration schemes use the same dedicated pins for the standard control signals shown in the Intel^® Stratix^® 10 Configuration Timing Diagram. Many other optional configuration signals do not have dedicated pin assignments.

Device Configuration Pins without Fixed Assignments

Note: Although the CONF_DONE and INIT_DONE configuration signals are not required, Intel recommends that you use these signals. The SDM drives the CONF_DONE signal high after successfully receiving full bitstream. The SDM drives the INIT_DONE signal high to indicate the device is fully in user mode. These signals are important when debugging configuration.

Table 5. Available SDM I/O Pin Assignments for Configuration Signals that Do Not Use Dedicated SDM I/O Pins
Signal Names	Configuration Scheme
	Avalon^® -ST			AS x4
	x8	x16	x32	AS x4
`PWRMGT_SCL`	`SDM_IO0`	`SDM_IO0` `SDM_IO14`	`SDM_IO0` `SDM_IO14`	`SDM_IO0` `SDM_IO14`
`PWRMGT_SDA`	`SDM_IO12` `SDM_IO16`	`SDM_IO11` `SDM_IO12` `SDM_IO16`	`SDM_IO11` `SDM_IO12` `SDM_IO16`	`SDM_IO11` `SDM_IO12` `SDM_IO16`
`PWRMGT_ALERT`	`SDM_IO0` `SDM_IO9` `SDM_IO12`	`SDM_IO0` `SDM_IO9` `SDM_IO12`	`SDM_IO0` `SDM_IO12`	`SDM_IO0` `SDM_IO12`
`CONF_DONE`	`SDM_IO0` `SDM_IO5` `SDM_IO12` `SDM_IO16`	`SDM_IO0` `SDM_IO1` `SDM_IO2` `SDM_IO3` `SDM_IO4` `SDM_IO5` `SDM_IO6` `SDM_IO7` `SDM_IO10` `SDM_IO11` `SDM_IO12` `SDM_IO13` `SDM_IO14` `SDM_IO15` `SDM_IO16`	`SDM_IO0` `SDM_IO1` `SDM_IO2` `SDM_IO3` `SDM_IO4` `SDM_IO5` `SDM_IO6` `SDM_IO7` `SDM_IO9` `SDM_IO10` `SDM_IO11` `SDM_IO12` `SDM_IO13` `SDM_IO14` `SDM_IO15` `SDM_IO16`	`SDM_IO0` `SDM_IO10` `SDM_IO11` `SDM_IO12` `SDM_IO13` `SDM_IO14` `SDM_IO15` `SDM_IO16`
`INIT_DONE`	`SDM_IO0` `SDM_IO5` `SDM_IO12` `SDM_IO16`	`SDM_IO0` `SDM_IO1` `SDM_IO2` `SDM_IO3` `SDM_IO4` `SDM_IO5` `SDM_IO6` `SDM_IO7` `SDM_IO9` `SDM_IO10` `SDM_IO11` `SDM_IO12` `SDM_IO13` `SDM_IO14` `SDM_IO15` `SDM_IO16`	`SDM_IO0` `SDM_IO1` `SDM_IO2` `SDM_IO3` `SDM_IO4` `SDM_IO5` `SDM_IO6` `SDM_IO7` `SDM_IO9` `SDM_IO10` `SDM_IO11` `SDM_IO12` `SDM_IO13` `SDM_IO14` `SDM_IO15` `SDM_IO16`	`SDM_IO0` `SDM_IO10` `SDM_IO11` `SDM_IO12` `SDM_IO13` `SDM_IO14` `SDM_IO15` `SDM_IO16`
`CVP_CONFDONE`	Not supported	Not supported	Not supported	`SDM_IO0` `SDM_IO10` `SDM_IO11` `SDM_IO12` `SDM_IO13` `SDM_IO14` `SDM_IO15` `SDM_IO16`
`SEU_ERROR`	`SDM_IO0` `SDM_IO5` `SDM_IO7` `SDM_IO9` `SDM_IO12` `SDM_IO16`	`SDM_IO0` `SDM_IO1` `SDM_IO2` `SDM_IO3` `SDM_IO4` `SDM_IO5` `SDM_IO6` `SDM_IO7` `SDM_IO9` `SDM_IO10` `SDM_IO11` `SDM_IO12` `SDM_IO13` `SDM_IO14` `SDM_IO15` `SDM_IO16`	`SDM_IO0` `SDM_IO1` `SDM_IO2` `SDM_IO3` `SDM_IO4` `SDM_IO5` `SDM_IO6` `SDM_IO7` `SDM_IO9` `SDM_IO10` `SDM_IO11` `SDM_IO12` `SDM_IO13` `SDM_IO14` `SDM_IO15` `SDM_IO16`	`SDM_IO0` `SDM_IO10` `SDM_IO11` `SDM_IO12` `SDM_IO13` `SDM_IO14` `SDM_IO15` `SDM_IO16`
`HPS_COLD_nRESET`	`SDM_IO0` `SDM_IO5` `SDM_IO7` `SDM_IO9` `SDM_IO12` `SDM_IO16`	`SDM_IO0` `SDM_IO1` `SDM_IO2` `SDM_IO3` `SDM_IO4` `SDM_IO5` `SDM_IO6` `SDM_IO7` `SDM_IO9` `SDM_IO10` `SDM_IO11` `SDM_IO12` `SDM_IO13` `SDM_IO14` `SDM_IO15` `SDM_IO16`	`SDM_IO0` `SDM_IO1` `SDM_IO2` `SDM_IO3` `SDM_IO4` `SDM_IO5` `SDM_IO6` `SDM_IO7` `SDM_IO9` `SDM_IO10` `SDM_IO11` `SDM_IO12` `SDM_IO13` `SDM_IO14` `SDM_IO15` `SDM_IO16`	`SDM_IO0` `SDM_IO10` `SDM_IO11` `SDM_IO12` `SDM_IO13` `SDM_IO14` `SDM_IO15` `SDM_IO16`
`Direct to Factory Image`	Not applicable	Not applicable	Not applicable	`SDM_IO0` `SDM_IO10` `SDM_IO11` `SDM_IO12` `SDM_IO13` `SDM_IO14` `SDM_IO15` `SDM_IO16`
`DATA UNLOCK`	`SDM_IO0` `SDM_IO5` `SDM_IO7` `SDM_IO9` `SDM_IO12` `SDM_IO16`	`SDM_IO0` `SDM_IO1` `SDM_IO2` `SDM_IO3` `SDM_IO4` `SDM_IO5` `SDM_IO6` `SDM_IO7` `SDM_IO9` `SDM_IO10` `SDM_IO11` `SDM_IO12` `SDM_IO13` `SDM_IO14` `SDM_IO15` `SDM_IO16`	`SDM_IO0` `SDM_IO1` `SDM_IO2` `SDM_IO3` `SDM_IO4` `SDM_IO5` `SDM_IO6` `SDM_IO7` `SDM_IO9` `SDM_IO10` `SDM_IO11` `SDM_IO12` `SDM_IO13` `SDM_IO14` `SDM_IO15` `SDM_IO16`	`SDM_IO0` `SDM_IO10` `SDM_IO11` `SDM_IO12` `SDM_IO13` `SDM_IO14` `SDM_IO15` `SDM_IO16`

Note:

Intel recommends that you assign the CONF_DONE and INIT_DONE pins to SDM I/O pins 0, 8, or 16. These pins have weak internal pull-downs resistors. If you cannot use these pins, Intel^® recommends that you include external 4.7-kΩ pull-down resistors to avoid false signaling.

Specifying Optional Configuration Pins

You enable and assign the SDM I/O pins using the Intel^® Quartus^® Prime software.

Complete the following steps to assign these additional configuration pins:

On the Assignments menu, click Device.
In the Device and Pin Options dialog box, select the Configuration category and click Configuration Pins Options.
In the Configuration Pin window, enable and assign the configuration pin that you want to include in your design.
Click OK to confirm and close the Configuration Pin dialog box.

Enabling Dual-Purpose Pins

AVST_CLK, AVST_DATA[15:0], AVST_DATA[31:16], and AVST_VALID are dual-purpose pins. Once the device enters user mode these pins can function either as GPIOs or as tri-state inputs.

If you use these pins as GPIOs, make the following assignments:

Set V_CCIO of the I/O bank at 1.8 V
Assign the 1.8 V I/O standard to these pins

Complete the following steps to assign these settings to the dual-purpose pins:

On the Assignments menu, click Device.
In the Device and Pin Options dialog box, select the Dual-Purpose Pins category.
In the Dual-purpose pins table, set the pin functionality in the Value column.
Click OK to confirm and close the Device and Pin Options
Attention: When you use the Avalon^® -ST configuration scheme the dual-purpose Avalon^® -ST pins have the following restrictions:
- You cannot use the Avalon^® -ST interface for partial reconfiguration (PR).
- You cannot use the Avalon^® -ST pins in user mode in designs that include the HPS. This restriction means that you cannot use the Avalon^® -ST as dual-purpose I/Os in designs that include the HPS.

Configuration Pins I/O Standard, Drive Strength, and IBIS Model

Table 6. Intel^® Stratix^® 10 Configuration Pins I/O Standard, Drive Strength, and IBIS Model
Configuration Pin Function	Direction	I/O Standard	Drive Strength (mA)	IBIS Model
`TDO`	Output	1.8V LVCMOS	8	18_io_d8s1_sdm_lvl
`TMS`	Input	Schmitt Trigger Input	—	18_in_sdm_lv
`TCK`	Input	Schmitt Trigger Input	—	18_in_sdm_lv
`TDI`	Input	Schmitt Trigger Input	—	18_in_sdm_lv
`nSTATUS`	Output	1.8V LVCMOS	8	18_in_sdm_lv
`OSC_CLK_1`	Input	Schmitt Trigger Input	—	18_in_sdm_lv
`nCONFIG`	Input	Schmitt Trigger Input	—	18_in_sdm_lv
`SDM_IO[16:0]`	I/O	Schmitt Trigger Input or 1.8V LVCMOS	8	Input: 18_in_sdm_lv Output: 18_in_sdm_lv
`AVST_DATA[31:0]`, `AVST_CLK`, `AVST_VALID`	I/O	Schmitt Trigger Input or 1.8V LVCMOS	8	Input: 18_in_sdm_lv Output: 18_in_sdm_lv

You can download the IBIS models from the IBIS Models for Intel Devices web page. The Intel^® Quartus^® Prime software does not support IBIS model generation for configuration pins in the current release.

Unused SDM Pins

You can specify other functions on unused SDM pins in the Intel^® Quartus^® Prime software.

SDM I/O Pins for Power Management and SmartVID

SDM pins are also available for the SmartVID power management feature for -V devices.

Intel recommends that you use the Analog Devices LTM4677 Dual 18A or Single 36A μModule Regulator with Digital Power System Management to regulate the PMBus. The LTM4677 device is the default setting for the Device > Device and Pin Options > Power Management & VID > Slave device type parameter. If you are using a different PMBus regular change the default setting from LTM4677 to Other.

Figure 10. Specifying the Slave Device Type for Power Management and VID

Refer to the Intel^® Stratix^® 10 Power Management User Guide for more information about the pin assignments and PMBus setting.

Specifying Pins for Partial Reconfiguration (PR)

The partial reconfiguration signals use GPIO pins.

The following signals control partial reconfiguration in Intel^® Stratix^® 10 devices:

PR_REQUEST
PR_READY
PR_ERROR
PR_DONE

Connect these partial reconfiguration signals to the Partial Reconfiguration External Configuration Controller Intel^® Stratix^® 10 FPGA IP.

Configuration Clocks

Setting Configuration Clock Source

You must specify the configuration clock source by selecting either the internal oscillator or OSC_CLK_1 with the supported frequency. By default, the SDM uses the internal oscillator for device configuration. Specify an OSC_CLK_1 clock source for the fastest configuration time.

Complete the following steps to select the configuration clock source:

Specify an OSC_CLK_1 clock source for the fastest configuration time. On the Assignments menu, click Device.
In the Device and Pin Options dialog box, select the General category.
Specify the configuration clock source from the Configuration clock source drop down menu.
Click OK to confirm and close the Device and Pin Options.

OSC_CLK_1 Clock Input

OSC_CLK_1 Requirements

When you drive OSC_CLK_1 input clock with an external clock source and enable OSC_CLK_1 in the Intel^® Quartus^® Prime software, the device loads the majority of the configuration bitstream at 250 MHz. Intel^® Stratix^® 10 devices include an internal oscillator in addition to OSC_CLK_1 which runs the configuration process at a frequency between 170-230 MHz. Intel^® Stratix^® 10 devices always use this internal oscillator to load the first section of the bitstream, approximately 200 kilobyte (KB). The SDM can use either clock source for the remainder of device configuration. If you use the internal oscillator, you can leave the OSC_CLK_1 unconnected.

Note:

Device configuration may fail under the following conditions when you select the OSC_CLK_1 as the clock source for configuration:

You fail to drive the OSC_CLK_1 pin.
You drive the OSC_CLK_1 pin at an incorrect frequency. Select one of the following input reference clock frequencies to drive the OSC_CLK_1 pin:
- 25 MHz
- 100 MHz
- 125 MHz

The Intel^® Stratix^® 10 device multiplies the OSC_CLK_1 source clock frequency to generate a 250 MHz clock for configuration. Using an OSC_CLK_1 source enables the fastest possible configuration. Refer to Setting Configuration Clock Source for instructions setting this frequency using the Intel^® Quartus^® Prime Software.

You can also specify this frequency by editing your .qsf file. Here are the possible assignments:

# EXTERNAL OSCILLATOR CLOCK  VIA OSC_CLK_1 PIN 
set_global_assignment -name DEVICE_INITIALIZATION_CLOCK OSC_CLK_1_25MHZ
set_global_assignment -name DEVICE_INITIALIZATION_CLOCK OSC_CLK_1_100MHZ
set_global_assignment -name DEVICE_INITIALIZATION_CLOCK OSC_CLK_1_125MHZ

Intel Stratix 10 Configuration Schemes

Avalon-ST Configuration

The Avalon^® -ST configuration scheme replaces the FPP mode available in earlier device families. Avalon^® -ST is the fastest configuration scheme for Intel^® Stratix^® 10 devices. This scheme uses an external host, such as a microprocessor, MAX^® II, MAX^® V, or Intel^® MAX^® 10 device to drive configuration. The external host controls the transfer of configuration data from external storage such as flash memory to the FPGA. The logic that controls the configuration process resides in the external host. You can use the PFL II IP with a MAX^® II, MAX^® V, or Intel^® MAX^® 10 device as the host to read configuration data from the flash memory device and configure the Intel^® Stratix^® 10 device. The Avalon^® -ST configuration scheme is called passive because the external host, not the Intel^® Stratix^® 10 device, controls configuration.

Table 7. Avalon^® -ST Configuration Data Width, Clock Rates, and Data RatesMb is an abbreviation for Megabits per second.
Protocol	Data Width (bits)	Max Clock Rate	Max Data Rate	MSEL[2:0]
Avalon^® -ST	32	125 MHz	4000 Mb	000
	16	125 MHz	2000 Mb	101
	8	125 MHz	1000 Mb	110

Table 8. Required Configuration Signals for the Avalon^® -ST Configuration Scheme You can use an 8-, 16-, or 32-bit Avalon-ST configuration data bus. You specify SDM I/O pin functions using the Device > Configuration > Device and Pin Options dialog box in the Intel^® Quartus^® Prime software. For the Avalon-ST x16 and x32 configuration, you can reassign the GPIO, dual-purpose configuration pins for other functions in user mode using the Device > Configuration > Device and Pin Options > Dual-Purpose Pins dialog box.
Signal Name	Pin Type	Direction	Powered by
`nSTATUS`	SDM I/O	Output	V_{CCIO_SDM}
`nCONFIG`	SDM I/O	Input	V_{CCIO_SDM}
`MSEL[2:0]`	SDM I/O, Dual-Purpose	Input	V_{CCIO_SDM}
`CONF_DONE`	SDM I/O	Output	V_{CCIO_SDM}
`AVST_READY`	SDM I/O	Output	V_{CCIO_SDM}
`AVSTx8_DATA[7:0]`	SDM I/O	Input	V_{CCIO_SDM}
`AVSTx8_VALID`	SDM I/O	Input	V_{CCIO_SDM}
`AVSTx8_CLK`	SDM I/O	Input	V_{CCIO_SDM}
`AVST_DATA[31:0]`	GPIO, Dual-Purpose	Input	V_CCIO
`AVST_VALID`	GPIO, Dual-Purpose	Input	V_CCIO
`AVST_CLK`	GPIO, Dual-Purpose	Input	V_CCIO

Refer to the Intel^® Stratix^® 10 Data Sheet for configuration timing estimates.

Note: Although the CONF_DONE and INIT_DONE configuration signals are not required, Intel recommends that you use these signals. The SDM drives the CONF_DONE signal high after successfully receiving the full bitstream. The SDM drives the INIT_DONE signal high to indicate the device is fully in user mode. These signals are important when debugging configuration.

Note: If you create custom logic instead of using the PFL II IP to drive configuration, refer to the Avalon Streaming Interfaces in the Avalon Interface Specifications for protocol details.

Avalon -ST Configuration Scheme Hardware Components and File Types

You can use the following components to implement the Avalon^® -ST configuration scheme:

A CPLD with PFL II IP and common flash interface (CFI) flash or Quad SPI flash memory
A custom host, typically a microprocessor, with any external memory
The Intel^® FPGA Download Cable II to connect the Intel^® Quartus^® Prime Programmer to the PCB.

The following block diagram illustrates the components and design flow using the Avalon^® -ST configuration scheme.

Figure 11. Components and Design Flow for .pof Programming

Table 9. Output File Types
Programming File Type	Extension	Description
Programmer Object File	.pof	The .pof is a proprietary Intel^® FPGA file type. Use the PFL II IP core via a JTAG header to write the .pof to an external CFI flash or serial flash device.
Raw Binary File	.rbf	You can also use the .rbf with the Avalon^® -ST configuration scheme and an external host such as a CPU or microcontroller. You can program the configuration bitstreams or data in the .rbf file directly into flash via a third-party programmer. Then, you can use an external host to configure the device with the Avalon^® -ST configuration scheme.

Note: If you choose a third-party microprocessor for Avalon^® -ST configuration, refer to the Avalon^® Streaming Interfaces in the Avalon^® Interface Specifications for protocol details.

Enabling Avalon-ST Device Configuration

You enable the Avalon^® -ST device configuration scheme in the Intel^® Quartus^® Prime software.

Complete the following steps to specify an Avalon^® -ST interface for device configuration.

On the Assignments menu, click Device.
In the Device and Pin Options dialog box, select the Configuration category.
In the Configuration window, in the Configuration scheme dropdown list, select the appropriate Avalon^® -ST bus width.
Click OK to confirm and close the Device and Pin Options dialog box.

The AVST_READY Signal

Before beginning configuration, trigger device cleaning by toggling the nCONFIG pin from high to low to high. This nCONFIG transition also returns the device to the configuration state.

Figure 12. Monitoring the AVST_READY Signal and Responding to Backpressure

The configuration files for Intel^® Stratix^® 10 devices can be highly compressed. During configuration, the decompression of the bit stream inside the device requires the host to pause before sending more data. The Intel^® Stratix^® 10 device asserts the AVST_READY signal when the device is ready to accept data. The AVST_READY signal is only valid when the nSTATUS pin is high. In addition, the host must handle backpressure by monitoring the AVST_READY signal and may assert AVST_VALID signal any time after the assertion of AVST_READY signal. The host must monitor the AVST_READY signal throughout the configuration.

The AVST_READY signal sent by the Intel^® Stratix^® 10 device to the host is not synchronized with the AVSTx8_CLK or AVST_CLK. To configure the Intel^® Stratix^® 10 device successfully, the host must adhere to the following constraints:

The host must drive no more than six data words after the deassertion of the AVST_READY signal including the delay incurred by the 2-stage register synchronizer.
The host must synchronize the AVST_READY signal to the AVST_CLK signal using a 2-stage register synchronizer. Here is Register transfer level (RTL) example code for 2-stage register synchronizer:
```
always @(posedge avst_clk or negedge reset_n) 
	begin
		if (~reset_n)
		begin
			fpga_avst_ready_reg1 <= 0;
			fpga_avst_ready_reg2 <= 0;
		else 
			fpga_avst_ready_reg1 <= fpga_avst_ready;
			fpga_avst_ready_reg2 <= fpga_avst_ready_reg1;
		end
	end
```
Where:
- The AVST_CLK signal comes from either PFL II IP or your Avalon^® -ST controller logic.
- fpga_avst_ready is the AVST_READY signal from the Intel^® Stratix^® 10 device.
- fpga_avst_ready_reg2 signal is the AVST_READY signal that is synchronous to AVST_CLK.

You must properly constrain the AVST_CLK and AVST_DATA signals at the host. Perform timing analysis on both signals between the host and Intel^® Stratix^® 10 device to ensure the Avalon-ST configuration timing specifications are met. Refer to the Avalon-ST Configuration Timing section of the Intel^® Stratix^® 10 Device Data Sheet for information about the timing specifications.

Note: The AVST_CLK signal must run continuously during configuration. The AVST_READY signal cannot assert unless the clock is running.

Optionally, you can monitor the CONF_DONE signal to indicate the flash has sent all the data to FPGA or to indicate the configuration process is complete.

If you use the PFL II IP core as the configuration host, you can use the Intel^® Quartus^® Prime software to store the binary configuration data to the flash memory through the PFL II IP core.

If you use the Avalon-ST Adapter IP core as part of the configuration host, set the Source Ready Latency value between 1- 6.

Avalon-ST x8 configuration scheme uses the SDM pins only. Avalon-ST x16 and x32 configuration scheme only use dual-purpose I/O pins that you can use as general-purpose I/O pins after configuration.

RBF Configuration File Format

If you do not use the Parallel Flash Loader II Intel^® FPGA IP core to program the flash, you must generate the .rbf file.

The data in .rbf file are in little-endian format

Table 10. Writing 32-bit Data For a x32 data bus, the first byte in the file is the least significant byte of the configuration double word, and the fourth byte is the most significant byte.
Double Word = 01EE1B02
LSB: BYTE0 = 02	BYTE1 = 1B	BYTE2 = EE	MSB: BYTE3 = 01
D[7:0]	D[15:8]	D[23:16]	D[31:24]
0000 0010	0001 1011	1110 1110	0000 0001

Table 11. Writing 16-bit DataFor a x16 data bus, the first byte in the file is the least significant byte of the configuration word, and the second byte is the most significant byte of the configuration word.
WORD0 = 1B02		WORD1 = 01EE
LSB: BYTE0 = 02	MSB: BYTE1 = 1B	LSB: BYTE2 = EE	MSB: BYTE3 = 01
D[7:0]	D[15:8]	D[7:0]	D[15:8]
0000 0010	0001 1011	1110 1110	0000 0001

Avalon-ST Single-Device Configuration

Refer to the Intel^® Stratix^® 10 Device Family Pin Connection Guidelines for additional information about individual pin usage and requirements.

Figure 13. Connections for Avalon-ST x8 Single-Device Configuration

Figure 14. Connections for Avalon-ST x16 Single-Device Configuration

Figure 15. Connections for Avalon-ST x32 Single-Device Configuration

Notes for Figure:

Refer to MSEL Settings for the correct resistor pull-up and pull-down values for all configuration schemes.
The MSEL pins are dual-purpose. After power-on, you can reassign these pins to other functions. For more information, refer to Enabling Dual Purpose Pins
The synchronizers shown in all three figures can be internal if the host is an FPGA or CPLD. If the host is a microprocessor, you must use discrete synchronizers.

Debugging Guidelines for the Avalon -ST Configuration Scheme

The Avalon^® -ST configuration scheme replaces the previously available in fast passive parallel (FPP) modes. This configuration scheme retains similar functionality and performance. Here are the important differences:

The Avalon^® -ST configuration scheme requires you to monitor the flow control signal, AVST_READY. The AVST_READY signal indicates if the device can receive configuration data.
The AVST_CLK and AVSTx8_CLK clock signals cannot pause when configuration data is not being transferred. Data is not transferred when AVST_READY and AVST_VALID are low. The AVST_CLK and AVSTx8_CLK clock signals must run continuously until CONF_DONE asserts.

Debugging Suggestions

Review the general Configuration Debugging Checklist in the Debugging Guide chapter before considering these debugging tips that pertain to the Avalon^® -ST configuration scheme.

Only assert AVST_VALID after the SDM asserts AVST_READY.
Only assert AVST_VALID when the AVST_DATA is valid.
Ensure that the AVST_CLK clock signal is continuous and free running until configuration completes. The AVST_CLK can stop after CONF_DONE asserts. The initialization state does not require the AVST_CLK signal.
If using x8 mode, ensure that you use the dedicated SDM_IO pins for this interface (clock, data, valid and ready).
If using x16 or x32 mode, power the I/O bank containing the x16 or x32 pins (I/O Bank 3A) at 1.8 V.
Ensure you select the appropriate Avalon^® -ST configuration scheme in your Intel^® Quartus^® Prime Pro Edition project.
Ensure the MSEL pins reflect this mode on the PCB.
Verify that host device does not drive configuration pins before the Intel^® Stratix^® 10 device powers up.

QSF Assignments for Avalon-ST x8

You can specify many Intel^® Quartus^® Prime project settings using Intel^® Quartus^® Prime Software GUI or by editing the Intel^® Quartus^® Prime Settings File (.qsf). The following assignments in the .qsf show typical settings for a Intel^® Stratix^® 10 device using Avalon^® -ST x8 configuration.

These settings are for a Intel^® Stratix^® 10 SmartVID device operating in PMBus slave mode which requires most of the SDM_IO pins. Refer to the Intel^® Stratix^® 10 Power Management User Guide for the PMBus constraints in master mode.

# Fitter Assignments
# ==================
set_global_assignment -name CONFIGURATION_VCCIO_LEVEL 1.8V

# SDM IO Assignments
# ==================
set_global_assignment -name USE_PWRMGT_SCL SDM_IO0
set_global_assignment -name USE_PWRMGT_SDA SDM_IO12
set_global_assignment -name USE_PWRMGT_ALERT SDM_IO9
set_global_assignment -name USE_CONF_DONE SDM_IO16
set_global_assignment -name USE_SEU_ERROR SDM_IO5

# Configuration settings
# ======================
set_global_assignment -name STRATIXV_CONFIGURATION_SCHEME "AVST X8"
set_global_assignment -name USE_CONFIGURATION_DEVICE OFF
set_global_assignment -name ERROR_CHECK_FREQUENCY_DIVISOR 256
set_global_assignment -name GENERATE_PR_RBF_FILE ON
set_global_assignment -name ENABLE_ED_CRC_CHECK ON
set_global_assignment -name MINIMUM_SEU_INTERVAL 479

# SmartVID feature PMBus settings [Slave mode settings only]
# ==========================================================
set_global_assignment -name VID_OPERATION_MODE "PMBUS SLAVE"
set_global_assignment -name PWRMGT_DEVICE_ADDRESS_IN_PMBUS_SLAVE_MODE 3F

You can also set the SDM_IO configuration pins using the Assignments > Device > Device and Pin Options > Configuration > Configuration Pin Options.

Figure 16. Set SDM_IO Configuration Pins Using the Intel^® Quartus^® Prime Software

QSF Assignments for Avalon-ST x16

# Fitter Assignments
# ==================
set_global_assignment -name CONFIGURATION_VCCIO_LEVEL 1.8V

# SDM IO Assignments
# ==================
set_global_assignment -name USE_PWRMGT_SCL SDM_IO0
set_global_assignment -name USE_PWRMGT_SDA SDM_IO12
set_global_assignment -name USE_PWRMGT_ALERT SDM_IO9
set_global_assignment -name USE_CONF_DONE SDM_IO16
set_global_assignment -name USE_SEU_ERROR SDM_IO5

# Configuration settings
# ======================
set_global_assignment -name STRATIXV_CONFIGURATION_SCHEME "AVST X16"
set_global_assignment -name USE_CONFIGURATION_DEVICE OFF
set_global_assignment -name ERROR_CHECK_FREQUENCY_DIVISOR 256
set_global_assignment -name GENERATE_PR_RBF_FILE ON
set_global_assignment -name ENABLE_ED_CRC_CHECK ON
set_global_assignment -name MINIMUM_SEU_INTERVAL 479

# SmartVID feature PMBus settings [Slave mode settings only]
# ==========================================================
set_global_assignment -name VID_OPERATION_MODE "PMBUS SLAVE"
set_global_assignment -name PWRMGT_DEVICE_ADDRESS_IN_PMBUS_SLAVE_MODE 3F

You can also set the SDM_IO configuration pins using the Assignments > Device > Device and Pin Options > Configuration > Configuration Pin Options.

Figure 17. Set SDM_IO Configuration Pins Using the Intel^® Quartus^® Prime Software

QSF Assignments for Avalon-ST x32

# Fitter Assignments
# ==================
set_global_assignment -name CONFIGURATION_VCCIO_LEVEL 1.8V

# SDM IO Assignments
# ==================
set_global_assignment -name USE_PWRMGT_SCL SDM_IO14
set_global_assignment -name USE_PWRMGT_SDA SDM_IO16
set_global_assignment -name USE_PWRMGT_ALERT SDM_IO12
set_global_assignment -name USE_CONF_DONE SDM_IO5
set_global_assignment -name USE_INIT_DONE SDM_IO0
set_global_assignment -name USE_SEU_ERROR SDM_IO1

# Configuration settings
# ======================
set_global_assignment -name STRATIXV_CONFIGURATION_SCHEME "AVST X32"
set_global_assignment -name USE_CONFIGURATION_DEVICE OFF
set_global_assignment -name ERROR_CHECK_FREQUENCY_DIVISOR 256
set_global_assignment -name GENERATE_PR_RBF_FILE ON
set_global_assignment -name ENABLE_ED_CRC_CHECK ON
set_global_assignment -name MINIMUM_SEU_INTERVAL 479

# SmartVID feature PMBus settings [Slave mode settings only]
# ==========================================================
set_global_assignment -name VID_OPERATION_MODE "PMBUS SLAVE"
set_global_assignment -name PWRMGT_DEVICE_ADDRESS_IN_PMBUS_SLAVE_MODE 3F

You can also set the SDM_IO configuration pins using the Assignments > Device > Device and Pin Options > Configuration > Configuration Pin Options.

Figure 18. Set SDM_IO Configuration Pins Using the Intel^® Quartus^® Prime Software

IP for Use with the Avalon -ST Configuration Scheme: Intel FPGA Parallel Flash Loader II IP Core

Functional Description

You can use the Parallel Flash Loader II Intel FPGA IP (PFL II) with an external host, such as the MAX^® II, MAX^® V, or Intel^® MAX^® 10 devices to complete the following tasks:

Program configuration data into a flash memory device using JTAG interface.
Configure the Intel^® Stratix^® 10 device with the Avalon^® -ST configuration scheme from the flash memory device.

Note: Use the Parallel Flash Loader II IP Intel FPGA IP and not the earlier Parallel Flash Loader IP with the Avalon^® -ST configuration scheme in Intel^® Stratix^® 10 devices.

Generating and Programming a .pof into CFI Flash

The Intel^® Quartus^® Prime software generates the .sof when you compile your design. You use the .sof to generate the .pof.. This process includes the following steps:

Generating a .pof for the PFL II IP using the Intel^® Quartus^® Prime File > Programming File Generator.
Using the Intel^® Quartus^® Prime Programmer to write the Intel^® Stratix^® 10 device .pof to the flash device.

Figure 19. Programming the CFI Flash Memory with the JTAG Interface

The PFL II IP core supports dual flash memory devices in burst read mode to achieve faster configuration times. You can connect two MT29EW CFI flash memory devices to the host in parallel using the same data bus, clock, and control signals. During FPGA configuration, the AVST_CLK frequency is four times faster than the flash_clk frequency.

Figure 20. PFL II IP core with Dual CFI Flash Memory DevicesThe flash memory devices must have the same memory density from the same device family and manufacturer.

Controlling Avalon-ST Configuration with PFL II IP Core

The PFL II IP core in the host determines when to start the configuration process, read the data from the flash memory device, and configure the Intel^® Stratix^® 10 device using the Avalon-ST configuration scheme.

Figure 21. FPGA Configuration with Flash Memory Data

You can use the PFL II IP core to either program the flash memory devices, configure your FPGA, or both. To perform both functions, create separate PFL II functions if any of the following conditions apply to your design:

You modify the flash data infrequently.
You have JTAG or In-System Programming (ISP) access to the configuration host.
You want to program the flash memory device with non-Intel FPGA data, for example initialization storage for an ASSP. You can use the PFL II IP core to program the flash memory device for the following purposes:
- To write the initialization data
- To store your design source code to implement the read and initialization control with the host logic

Mapping PFL II IP Core and Flash Address

The address connections between the PFL II IP core and the flash memory device vary depending on the flash memory device vendor and data bus width.

Figure 22. Flash Memory in 8-Bit ModeThe address connection between the PFL II IP core and the flash memory device are the same.

Figure 23. Flash Memories in 16-Bit ModeThe flash memory addresses in 16-bit flash memory shift one bit down in comparison with the flash addresses in PFL II IP core. The flash address in the flash memory starts from bit 1 instead of bit 0.

Figure 24. Cypress and Micron M28, M29 Flash Memory in 8-Bit ModeThe flash memory addresses in Cypress 8-bit flash shifts one bit up. Address bit 0 of the PFL II IP core connects to data pin D15 of the flash memory.

Figure 25. Cypress and Micron M28, M29 Flash Memory in 16-Bit ModeThe address bit numbers in the PFL II IP core and the flash memory device are the same.

Implementing Multiple Pages in the Flash .pof

The PFL II IP core stores configuration data in a maximum of eight pages in a flash memory block.

The total number of pages and the size of each page depends on the flash density. Here are some guidelines for storing your designs to pages:

Always store designs for different FPGA chains on different pages.
You may choose store different designs for a FPGA chain on a single page or on multiple pages.
When you choose to store the designs for a FPGA chain on a single page, the design order must match the JTAG chain device order.

Use the generated .sof to create a flash memory device .pof. The following address modes are available for the .sof to .pof conversion:

Block mode—allows you to specify the start and end addresses for the page.
Start mode—allows you to specify only the start address. The start address for each page must be on an 8 KB boundary. If the first valid start address is 0×000000, the next valid start address is an increment of 0×2000.
Auto mode—allows the Intel^® Quartus^® Prime software to automatically determine the start address of the page. The Intel^® Quartus^® Prime software aligns the pages on a 128 KB boundary. If the first valid start address is 0x000000, the next valid start address is an multiple of 0x20000.

Storing Option Bits

In addition to design data, the flash memory stores the option bits. You must specify the address for the options bits in two places: the PFL II IP and in the option bits address of the flash memory device.

The option bits contain the following information:

The start address for each page.
The .pof version for flash programming. This value is the same for all pages.
The Page-Valid bits for each page. The Page-Valid bit is bit 0 of the start address. The PLF II IP core writes this bit after successfully programming the page.

You use the Programming File Generator dialog box to specify the Start address of the option bits. Specify your flash device using Add Device on the Configuration Tab of the Programming File Generator dialog box. Then click OPTIONS and EDIT to specify the Start address for the option bits. This Start address must match the address you specify for What is the byte address of the option bits, in hex? when specifying the PFL II IP parameters.

You set the option bits in the PFL II IP Intel FPGA IP using the parameter editor. By default the PFL II IP displays Flash Programming for the What operating mode will be used? parameter. In this default state, the FPGA Configuration tab is not visible. Select either FPGA Configuration or Flash Programming and FPGA Configuration for the What operating mode will be used parameter on the General tab. The following figure shows the FPGA Configuration option.

Figure 26. General Tab of the PFL II IP

Specify the options bits hex address for the What is the base address of the option bits, in hex? parameter on the FPGA Configuration tab.

Figure 27. FPGA Configuration Tab of the PFL II IP

The Intel^® Quartus^® Prime Programming File Generator generates the information for the .pof version when you convert the .sofs to .pofs. The value for the .pof version for Intel^® Stratix^® 10 is 0x05. The following table shows an example of page layout for a .pof using all eight pages. This example stores the .pof version at 0x80.

Table 12. Option Bits Sector Format
Sector Offset	Value
`0x00`–`0x03`	Page 0 start address
`0x04`–`0x07`	Page 0 end address
`0x08`–`0x0B`	Page 1 start address
`0x0C`–`0x0F`	Page 1 end address
`0x10`–`0x13`	Page 2 start address
`0x14`–`0x17`	Page 2 end address
`0x18`–`0x1B`	Page 3 start address
`0x1C`–`0x1F`	Page 3 end address
`0x20`–`0x23`	Page 4 start address
`0x24`–`0x27`	Page 4 end address
`0x28`–`0x2B`	Page 5 start address
`0x2C`–`0x2F`	Page 5 end address
`0x30`–`0x33`	Page 6 start address
`0x34`–`0x37`	Page 6 end address
`0x38`–`0x3B`	Page 7 start address
`0x3C`–`0x3F`	Page 7 end address
`0x40`–`0x7F`	Reserved
`0x80` ⁴	.pof version
`0x81`-`0xFF`	Reserved

CAUTION:

To prevent the PFL II IP core from malfunctioning, do not overwrite any information in the option bits sector. Always store the option bits in unused addresses in the flash memory device.

⁴ The .pof version occupies only one byte in the option bits sector.

Verifying the Option Bit Start and End Addresses

You can decode the start and end address that you specified for each of the SOF page when converting a .sof to .pof file from the 32-bit value of the sector offset address. If you encounter a configuration error you can verify that the generated bitstream addresses match the addresses you specified in the Intel^® Quartus^® Prime Software.

The following table shows the bit fields of the start address.

Table 13. Start Address Bit Content
Bit	Width	Description
31:11	21	Addressable start address
10:1	10	Reserved bits
0	1	Page valid bit 0=Valid 1=Error

Table 14. End Address Bit Content
Bit	Width	Description
31:0	32	Addressable end address

To restore the addresses:

Start address—append 13'b0000000000000 to the addressable start address
End address—append 2'b11 to the addressable end address

For a .pof that has two page addresses with the values shown in the following table.

Sector Offset	Value
`0x00` – `0x03`	`0x00004000`
`0x04` – `0x07`	`0x00196E30`
`0x08` – `0x0B`	`0x001C0000`
`0x0C` – `0x0F`	`0x00352E30`

For Page 0 if you append the start address bits[31:11] with 13'b0000000000000, the result is 32'b00000000000000010000000000000000 = 0x10000.

If you append the end address 0x00196E0 with 2'b11, the result is 26'b00011001011011100011000011 = 0x65B8C3.

For Page 1 if you append the start address with 13'b0000000000000, the result is 32'b0000000000011100000000000000000000 = 0x700000.

If you append end address 0x00352E30 with 2’b11, the result is 32'b00000000110101001011100011000011 = 0xD4B8C3.

The start and end address must be correlated with the start and end address for each page printed in the .map file.

Implementing Page Mode and Option Bits in the CFI Flash Memory Device

The following figure shows an sample layout of a .pof with three pages. The end addresses depend on the density of the flash memory device. For different density devices refer to the Byte Address Range for CFI Flash Memory Devices with Different Densities table below. The option bits follow the configuration data in memory.

Figure 28. Implementing Page Mode and Option Bits in the CFI Flash Memory Device

The following figure shows the layout of the option bits for a single page. Because the start address must be on an 8 KB boundary, bits 0-12 of the page start address are set to zero and are not stored in the option bits.

Figure 29. Page Start Address, End Address, and Page-Valid Bit Stored as Option BitsThe Page-Valid bits indicate whether each page is successfully programmed. The PFL II IP core sets the Page-Valid bits after successfully programming the pages.

Table 15. Byte Address Range for CFI Flash Memory Devices with Different Densities
CFI Device (Megabit)	Address Range
8	`0x0000000`–`0x00FFFFF`
16	`0x0000000`–`0x01FFFFF`
32	`0x0000000`–`0x03FFFFF`
64	`0x0000000`–`0x07FFFFF`
128	`0x0000000`–`0x0FFFFFF`
256	`0x0000000`–`0x1FFFFFF`
512	`0x0000000`–`0x3FFFFFF`
1024	`0x0000000`–`0x7FFFFFF`

Using the PFL II IP Core

Converting .sof to .pof File

You can use the Programming File Generator to convert the .sof file to a .pof. The Programming File Generator options change dynamically according to your device and configuration mode selection.

Click File > Programming File Generator.
For Device family select Intel^® Stratix^® 10.
For Configuration mode select Avalon^® -ST configuration scheme that you plan to use.
For Output directory, click Browse to select your output file directory.
For Name specify a name for your output file.
On the Output Files tab, enable the checkbox for generation of the file or files you want to generate.
Specify the Output directory and Name for the file or files you generate.

Figure 30. Programming File Generator Output Files Tab
To specify a .sof that contains the configuration bitstream, on the Input Files tab, click Add Bitstream.
To include raw data, click Add Raw Data and specify a Hexadecimal ( Intel^® -Format) Output File (.hex) or binary (.bin) file. This step is optional.
On the Configuration Device tab, click Add device. The Add Device dialog box appears. Select your flash device from the drop-down list of available parallel flash devices.
Click OPTIONS and then Edit. In the Edit Partition dialog box specify the Start address of the Options in flash memory. This address must match the address you specify for What is the byte address of the option bits, in hex? when specifying the PFL II IP parameters. Ensure that the option bits sector does not overlap with the configuration data pages and that the start address is on an 8 KB boundary.

Figure 31. Edit Partition: OPTIONS for Flash Device
With the flash device selected, click Add Partition to specify a partition in flash memory.

Figure 32. Add Flash Device and Partition
1. For Name select a Partition name.
2. For Input File specify the .sof.
3. From the Page dropdown list, select the page to write this .sof.
4. For Address mode select the addressing mode to use.
  The following modes are available:
  - Auto— For the tool to automatically allocates a block in the flash device to store the data.
  - Block—To specify the start and end address of the flash partition.
  - Start—To specify the start address of the partition. The tool assigns the end address of the partition based on the input data size.
5. For Block and Start options, specify the address information.

Creating Separate PFL II Functions

Follow these steps to create separate PFL II IP instantiations for programming and configuration control:

In the IP Catalog locate the Parallel Flash Loader II Intel FPGA IP.
On the General tab for What operating mode will be used, select Flash Programming Only.
Intel recommends that you turn on the Set flash bus pins to tri-state when not in use.
Specify the parameters on the Flash Interface Settings and Flash Programming tabs to match your design.
Compile and generate a .pof for the flash memory device. Ensure that you tri-state all unused I/O pins.
To create a second PFL II instantiation for FPGA configuration, on the General tab, for What operating mode will be used, select FPGA Configuration.
Use this Flash Programming Only instance of the PFL II IP to write data to the flash device.
Whenever you must program the flash memory device, program the CPLD with the flash memory device .pof and update the flash memory device contents.
Reprogram the host with the production design .pof that includes the configuration controller.

Note: By default, all unused pins are set to ground. When programming the configuration flash memory device through the host JTAG pins, you must tri-state the FPGA configuration pins common to the host and the configuration flash memory device. You can use the pfl_flash_access_request and pfl_flash_access_granted signals of the PFL II block to tri-state the correct FPGA configuration pins.

Programming CPLDs and Flash Memory Devices Sequentially

This procedure provides a single set of instructions for the Intel^® Quartus^® Prime Programmer to configure the CPLD and write the flash memory device.

Open the Programmer and click Add File to add the .pof for the CPLD.
Right-click the CPLD .pof and click Attach Flash Device.
In the Flash Device menu, select the appropriate density for the ﬂash memory device.
Right-click the ﬂash memory device density and click Change File.
Select the .pof generated for the ﬂash memory device. The Programmer appends the .pof for the ﬂash memory device to the .pof for the CPLD.
Repeat this process if your chain has additional devices.
Check all the boxes in the Program/Configure column for the new .pof and click Start to program the CPLD and ﬂash memory device.

Programming CPLDs and Flash Memory Devices Separately

Follow these instructions to program the CPLD and the ﬂash memory devices separately:

Open the Programmer and click Add File.
In the Select Programming File, add the targeted .pof, and click OK.
Check the boxes under the Program/Configure column of the .pof.
Click Start to program the CPLD.
After the programming progress bar reaches 100%, click Auto Detect.
For example, if you are using dual Micron or Macronix flash devices, the programmer window shows a dual chain in your setup. Alternatively, you can add the ﬂash memory device to the programmer manually. Right-click the CPLD .pof and click Attach Flash Device. In the Select Flash Device dialog box, select the device of your choice.
Right-click the ﬂash memory device density and click Change File.

Note: For designs with more than one flash device, you must select the density that is equivalent to the sum of the densities of all devices. For example, if the design includes two 512-Mb CFI ﬂash memory devices, select CFI 1 Gbit.
Select the .pof generated for the ﬂash memory device. The Programmer attaches the .pof for the ﬂash memory device to the .pof of the CPLD.
Check the boxes under the Program/Configure column for the added .pof and click Start to program the ﬂash memory devices.

Note: If your design includes the PFL II IP the Programmer allows you to program, verify, erase, blank-check, or examine the configuration data page, the user data page, and the option bits sector separately. The programmer erases the ﬂash memory device if you select the .pof of the ﬂash memory device before programming. To prevent the Programmer from erasing other sectors in the ﬂash memory device, select only the pages, .hex data, and option bits.

Defining New CFI Flash Memory Device

The PFL II IP core supports Intel^® - and AMD-compatible flash memory devices. In addition to the supported flash memory devices, you can define the new Intel^® - or AMD-compatible CFI flash memory device in the PFL II-supported flash database using the Define New CFI Flash Device function.

To add a new CFI flash memory device to the database or update a CFI flash memory in the database, follow these steps:

In the Programmer window, on the Edit menu, select Define New CFI Flash Device. The following table lists the three functions available in the Define CFI Flash Device window.

Table 16. Functions of the Define CFI Flash Device Feature
Function	Description
New	Add a new Intel^® - or AMD-compatible CFI flash memory device into the PFL II-supported flash database.
Edit	Edit the parameters of the newly added Intel^® - or AMD-compatible CFI flash memory device in the PFL II-supported flash database.
Remove	Remove the newly added Intel^® - or AMD-compatible CFI flash memory device from the PFL II-supported flash database.

To add a new CFI flash memory device or edit the parameters of the newly added CFI flash memory device, select New or Edit. The New CFI Flash Device dialog box appears.

In the New CFI Flash Device dialog box, specify or update the parameters of the new flash memory device. You can obtain the values for these parameters from the data sheet of the flash memory device manufacturer.

Figure 33. Using the Programmer Edit Menu to Define a New Flash Device

Table 17. Parameter Settings for New CFI Flash Device
Parameter	Description
CFI flash device name	Define the CFI flash name
CFI flash device ID	Specify the CFI flash identifier code
CFI flash manufacturer ID	Specify the CFI flash manufacturer identification number
CFI flash extended device ID	Specify the CFI flash extended device identifier, only applicable for AMD-compatible CFI flash memory device
Flash device is Intel^® compatible	Turn on the option if the CFI flash is Intel^® compatible
Typical word programming time	Typical word programming time value in µs unit
Maximum word programming time	Maximum word programming time value in µs unit
Typical buffer programming time	Typical buffer programming time value in µs unit
Maximum buffer programming time	Maximum buffer programming time value in µs unit

Note: You must specify either the word programming time parameters, buffer programming time parameters, or both. Do not leave both programming time parameters with the default value of zero.

Click OK to save the parameter settings.
After you add, update, or remove the new CFI flash memory device, click OK.

The Windows registry stores user flash information. Consequently, you must have system administrator privileges to store the parameters in the Define New CFI Flash Device window in the Intel^® Quartus^® Prime Pro Edition Programmer.

PFL II Parameters

Table 18. PFL II General Parameters
Options	Value	Description
What operating mode will be used?	Flash Programming FPGA Configuration Flash Programming and FPGA Configuration	Specifies the operating mode of flash programming and FPGA configuration control in one IP core or separate these functions into individual blocks and functionality.
What is the targeted flash?	CFI Parallel Flash Quad SPI Flash	Specifies the flash memory device connected to the PFL II IP core.
Set flash bus pins to tri-state when not in use	On Off	Allows the PFL II IP core to tri-state all pins interfacing with the flash memory device when the PFL II IP core does not require access to the flash memory.

Table 19. PFL II Flash Interface Setting Parameters
Options	Value	Description
How many flash devices will be used?	1–16	Specifies the number of flash memory devices connected to the PFL II IP core.
What's the largest flash device that will be used?	8 Mbit–4 Gbit	Specifies the density of the flash memory device to be programmed or used for FPGA configuration. If you have more than one flash memory device connected to the PFL II IP core, specify the largest flash memory device density. For dual CFI flash, select the density that is equivalent to the sum of the density of two flash memories. For example, if you use two 512-Mb CFI flashes, you must select CFI 1 Gbit.
What is the flash interface data width	8 16 32	Specifies the flash data width in bits. The flash data width depends on the flash memory device you use. For multiple flash memory device support, the data width must be the same for all connected flash memory devices. Select the flash data width that is equivalent to the sum of the data width of two flash memories. For example, if you are targeting dual solution, you must select 32 bits because each CFI flash data width is 16 bits.
Allow user to control FLASH_NRESET pin	On Off	Creates a `FLASH_NRESET` pin in the PFL II IP core to connect to the reset pin of the flash memory device. A low signal resets the flash memory device. In burst mode, this pin is available by default. When using a Cypress GL flash memory, connect this pin to the `RESET` pin of the flash memory.

Table 20. PFL II Flash Programming Parameters
Options	Value	Description
Flash programming IP optimization target	Area Speed	Specifies the flash programming IP optimization. If you optimize the PFL II IP core for Speed, the flash programming time is shorter, but the IP core uses more LEs. If you optimize the PFL II IP core for Area, the IP core uses fewer LEs, but the flash programming time is longer.
Flash programming IP FIFO size	16 32	Specifies the FIFO size if you select Speed for flash programming IP optimization. The PFL II IP core uses additional LEs to implement FIFO as temporary storage for programming data during flash programming. With a larger FIFO size, programming time is shorter.
Add Block-CRC verification acceleration support	On Off	Adds a block to accelerate verification.

Table 21. PFL II FPGA Configuration Parameters
Options	Value	Description
What is the external clock frequency?	Provide the frequency of your external clock.	Specifies the user-supplied clock frequency for the IP core to configure the FPGA. The clock frequency must not exceed two times the maximum clock (`AVST_CLK`) frequency the FPGA can use for configuration. The PFL II IP core can divide the frequency of the input clock maximum by two.
What is the flash access time?	Provide the access time from the flash data sheet.	Specifies the flash access time. This information is available from the flash datasheet. Intel recommends specifying a flash access time that is equal to or greater than the required time. For CFI parallel flash, the unit is in ns. For NAND flash, the unit is in μs. NAND flash uses pages instead of bytes and requires greater access time. This option is disabled for quad SPI flash.
What is the byte address of the option bits, in hex?	Provide the byte address of the option bits.	Specifies the option bits start address in flash memory. The start address must reside on an 8 KB boundary. This address must be the same as the bit sector address you specified when converting the .sof to a .pof. For more information refer to Storing Option Bits.
Which FPGA configuration scheme will be used?	Avalon^® -ST x8 Avalon^® -ST x16 Avalon^® -ST x32	Specifies the width of the Avalon^® -ST interface.
What should occur on configuration failure?	Halt Retry same page Retry from fixed address	Configuration behavior after configuration failure. If you select Halt, the FPGA configuration stops completely after failure. If you select Retry same page, after failure, the PFL II IP core reconfigures the FPGA with data from the page that failed. If you select Retry from fixed address, the PFL II IP core reconfigures the FPGA a fixed address.
What is the byte address to retry from failure	—	If you select Retry from fixed address for configuration failure option, this option specifies the flash address the PFL II IP core to reads from.
Include input to force reconfiguration	On Off	Includes the optional `pfl_nreconfigure` reconfiguration input pin to enable reconfiguration of the FPGA.
Enable watchdog timer on Remote System Update support	On Off	Enables a watchdog timer for remote system update support. Turning on this option enables the `pfl_reset_watchdog` input pin and `pfl_watchdog_error` output pin. This option also specifies the period before the watchdog timer times out. The watchdog timer runs at the `pfl_clk frequency`.
Time period before the watchdog timer times out	—	Specifies the time out period for the watchdog timer. The default time out period is 100 ms.
Use advance read mode?	Normal mode Intel Burst mode 16 byte page mode (GL only) 32 byte page mode (MT23EW) Micron Burst Mode (M58BW)	This option improves the overall flash access time for the read process during the FPGA configuration. Normal mode—applicable for all flash memory Intel Burst mode—Applicable for devices that support bursting. Reduces sequential read access time 16 byte page mode (GL only)—applicable for Cypress GL flash memory only 32 byte page mode (MT23EW)—applicable tor MT23EW only Micron Burst Mode (M58BW)—applicable for Micron M58BW flash memory only For more information about the read-access modes of the flash memory device, refer to the respective flash memory data sheet.
Latency count	3 4 5	Specifies the latency count for Intel Burst mode.

PFL II Signals

Table 22. PFL II Signals
Pin	Type	Weak Pull-Up	Function
`pfl_nreset`	Input	—	Asynchronous reset for the PFL II IP core. Pull high to enable FPGA configuration. To prevent FPGA configuration, pull low when you do not use the PFL II IP core. This pin does not affect the PFL II IP flash programming functionality.
`pfl_flash_access_granted`	Input	—	For system-level synchronization. A processor or any arbiter that controls access to the flash drives this input pin. To use the PFL II IP core function as the flash master pull this pin high. Driving the `pfl_flash_access_granted` pin low prevents the JTAG interface from accessing the flash and FPGA configuration.
`pfl_clk`	Input	—	User input clock for the device. This is the frequency you specify for the What is the external clock frequency? parameter on the Configuration tab of the PFL II IP. This frequency must not be higher than the maximum `DCLK` frequency you specify for FPGA during configuration. This pin is not available if you are only using the PFL II IP for flash programming.
`fpga_pgm[]`	Input	—	Determines the page for the configuration. This pin is not available if you are only using the PFL II IP for flash programming.
`fpga_conf_done`	Input	10 kΩ Pull-Up Resistor	Connects to the `CONF_DONE` pin of the FPGA. The FPGA releases the pin high if the configuration is successful. During FPGA configuration, this pin remains low. This pin is not available if you are only using the PFL II IP for flash programming.
`fpga_nstatus`	Input	10 kΩ Pull-Up Resistor	Connects to the `nSTATUS` pin of the FPGA. This pin is high before the FPGA configuration begins and must stay high during FPGA configuration. If a configuration error occurs, the FPGA pulls this pin low and the PFL II IP core stops reading the data from the flash memory device. This pin is not available if you are only using the PFL II IP for flash programming.
`pfl_nreconfigure`	Input	—	When low initiates FPGA reconfiguration. To implement manual control of reconfiguration connect this pin to a switch. You can use this input to write your own logic in a CPLD to trigger reconfiguration via the PFL II IP. When FPGA reconfiguration begins, the `fpga_nconfig` pin is pulled low to reset the FPGA device. The `pfl_clk` pin registers this signal. This pin is not available if you are only using the PFL II IP for flash programming.
`pfl_flash_access_request`	Output	—	For system-level synchronization. When necessary, this pin connects to a processor or an arbiter. The PFL II IP core drives this pin high when the JTAG interface accesses the flash or the PFL II IP configures the FPGA. This output pin works in conjunction with the `flash_noe` and `flash_nwe` pins.
`flash_addr[]`	Output	—	The flash memory address. The width of the address bus depends on the density of the flash memory device and the width of the `flash_data` bus. Intel recommends that you turn On the Set flash bus pins to tri-state when not in use option in the PFL II .
`flash_data[]`	Input or Output (bidirectional pin)	—	Bidirectional data bus to transmit or receive 8-, 16-, or 32-bit data. Intel recommends that you turn On the Set flash bus pins to tri-state when not in use option in the PFL II. ⁵
`flash_nce[]`	Output	—	Connects to the `nCE` pin of the flash memory device. A low signal enables the flash memory device. Use this bus for multiple flash memory device support. The `flash_nce` pin connects to each `nCE` pin of all the connected flash memory devices. The width of this port depends on the number of flash memory devices in the chain.
`flash_nwe`	Output	—	Connects to the `nWE` pin of the flash memory device. When low enables write operations to the flash memory device.
`flash_noe`	Output	—	Connects to the `nOE` pin of the flash memory device. When low enables the outputs of the flash memory device during a read operation.
`flash_clk`	Output	—	For burst mode. Connects to the `CLK` input pin of the flash memory device. The active edges of `CLK` increment the flash memory device internal address counter. The `flash_clk` frequency is half of the `pfl_clk` frequency in burst mode for a single CFI flash. In dual CFI flash solution, the `flash_clk` frequency runs at a quarter of the `pfl_clk` frequency. Use this pin for burst mode only. Do not connect these pins from the flash memory device to the host if you are not using burst mode.
`flash_nadv`	Output	—	For burst mode. Connects to the address valid input pin of the flash memory device. Use this signal to latch the start address. Use this pin for burst mode only. Do not connect these pins from the flash memory device to the host if you are not using burst mode.
`flash_nreset`	Output	—	Connects to the reset pin of the flash memory device. A low signal resets the flash memory device.
`fpga_nconfig`	Open Drain Output	10-kW Pull-Up Resistor	Connects to the `nCONFIG` pin of the FPGA. A low pulse resets the FPGA and initiates configuration. These pins are not available for the flash programming option in the PFL II IP core. ⁵
`pfl_reset_watchdog`	Input	—	A switch signal to reset the watchdog timer before the watchdog timer times out. To reset the watchdog timer hold the signal high or low for at least two `pfl_clk` clock cycles.
`pfl_watchdog_error`	Output	—	When high indicates an error condition to the watchdog timer.

⁵ Intel recommends that you do not insert logic between the PFL II pins and the host I/O pins, especially on the flash_data and fpga_nconfig pins.

AS Configuration

In AS configuration schemes, the SDM block in the Intel^® Stratix^® 10 device controls the configuration process and interfaces. The serial flash configuration device stores the configuration data. During AS Configuration, the SDM first powers on with the boot ROM. Then, the SDM loads the initial configuration firmware from AS x4 flash. After the configuration firmware loads, this firmware controls the remainder of the configuration process, including I/O configuration and FPGA core configuration. Designs including an HPS, can use the HPS to access serial flash memory after the initial configuration.

Note: The serial flash configuration device must be fully powered up at the same time or before ramping up V_{CCIO_SDM} of the Intel^® Stratix^® 10 device.

The AS configuration scheme supports AS x4 (4-bit data width) mode only.

Table 23. Intel^® Stratix^® 10 Configuration Data Width, Clock Rates, and Data Rates
Mode		Data Width (bits)	Max Clock Rate	Max Data Rate	MSEL[2:0]
Active	Active Serial (AS)	4	133 MHz	532 Mb	Fast mode - 001 Normal mode - 011

Table 24. Required Configuration Signals for the AS Configuration SchemeYou specify SDM I/O pin functions using the Device > Configuration > Device and Pin Options dialog box in the Intel^® Quartus^® Prime software. You can reassign the GPIO, dual-purpose configuration pins for other functions in user mode.
Configuration Function	Pin Type	Direction	Powered by
`nSTATUS`	SDM I/O	Output	V_{CCIO_SDM}
`nCONFIG`	SDM I/O	Input	V_{CCIO_SDM}
`MSEL[2:0]`	SDM I/O, Dual-Purpose	Input	V_{CCIO_SDM}
`CONF_DONE`	SDM I/O	Output	V_{CCIO_SDM}
`AS_nCSO[3:0]`	SDM I/O	Output	V_{CCIO_SDM}
`AS_DATA[3:0]`	SDM I/O	Bidirectional	V_{CCIO_SDM}
`AS_CLK`	SDM I/O	Output	V_{CCIO_SDM}

Note: Although the CONF_DONE and INIT_DONE configuration signals are not required, Intel recommends that you use these signals. The SDM drives the CONF_DONE signal high after successfully receiving full bitstream. The SDM drives the INIT_DONE signal high to indicate the device is fully in user mode.These signals are important when debugging configuration. You can reassign the GPIO, dual-purpose configuration pins for other functions in user mode.

MSEL Pin Function for the AS x4 Configuration Scheme

The SDM samples the MSEL pins immediately after power-on in the SDM Start state. After the SDM samples the MSEL pins, the MSEL pins become active-low chips selects. For AS x4 designs using one flash device, AS_nCSO asserts low. The remaining chip select pins, AS_nCS1 - AS_nCS3 deassert high.

AS Configuration Scheme Hardware Components and File Types

You use the following components to implement the AS configuration scheme:

Quad SPI flash memory
The Intel^® FPGA Download Cable II to connect the Intel^® Quartus^® Prime Programmer to the PCB.

The following block diagram illustrates the components and design flow using the AS configuration scheme.

Figure 34. Components and Design Flow for .jic Programming

In addition to AS programming using a .jic, the Programmer supports direct programming of the quad SPI flash using a .pof as shown in AS Programming Using Intel Quartus Prime or Third-Party Programmer.

Table 25. Output File Types
Programming File Type	Extension	Description
JTAG Indirect Configuration File	.jic	The .jic enables serial flash programming via Intel^® FPGA JTAG pins. This file type is available only for ASx4 configuration. A newly populated board using the ASx4 configuration scheme requires initial SDM firmware programming. The helper SOF image provides the required SDM firmware. You initially use the JTAG cable to load a SDM Helper SOF into the Intel^® Stratix^® 10 device. The SDM can then load the flash device with the Intel^® Stratix^® 10 design.

AS Single-Device Configuration

Refer to the Intel^® Stratix^® 10 Device Family Pin Connection Guidelines for additional information about individual pin usage and requirements.

Figure 35. Connections for AS x4 Single-Device Configuration

AS Using Multiple Serial Flash Devices

Intel^® Stratix^® 10 devices support one AS x4 flash memory device for AS configuration and up to three AS x4 flash memories for use with HPS data storage. The MSEL pins are dual-purpose and operate as MSEL only during POR state. After the FPGA device enters user mode, you can repurpose the MSEL pins as chip select pins. You must to ensure appropriate chip select pin connections to the configuration AS x4 flash memory and the HPS AS x4 flash memory. Each flash device has a dedicated AS_nCSO pin but shares other pins.

Refer to the Intel^® Stratix^® 10 Device Family Pin Connection Guidelines for additional information about individual pin usage and requirements.

Figure 36. Connections for AS Configuration with Multiple Serial Flash Devices

The following table shows the maximum supported AS_CLK frequency for a range of capacitance loading values when using multiple flash devices. The maximum AS_CLK frequency also depends on whether you use the OSC_CLK_1 or internal oscillator as the clock source.

Table 26. Maximum AS_CLK Frequency as a Function of Board Capacitance Loading and Clock Source
Capacitance Loading (pF)	Maximum Supported AS_CLK (MHz)
Capacitance Loading (pF)	OSC_CLK_1 (MHz)	Internal Oscillator (MHz)
10	133/125	115
19	108	115
30	100	77
37	80	77
80	50	58
140	25	25

AS Configuration Timing Parameters

Figure 37. AS Configuration Serial Output Timing Diagram

Figure 38. AS Configuration Serial Input Timing Diagram

Note: For more information about the timing parameters, refer to the Intel^® Stratix^® 10 Device Datasheet.

Maximum Allowable External AS_DATA Pin Skew Delay Guidelines

You must minimize the skew on the AS data pins.

Skew delay includes the following elements:

The delay due to the differences in board traces lengths on the PCB
The capacitance loading of the flash device

The table below lists the maximum allowable skew delay depending on the AS_CLK frequency. Intel recommends that you to perform IBIS simulations to ensure that the skew delay does not exceed the maximum delay specified in this table.

Table 27. Maximum Skew for AS Data Pins in Nanoseconds (ns)
Symbol	Description	Frequency	Min	Typical	Max
T_{ext_skew}	Skew delay for `AS_DATA` for the `AS_CLK` frequency specified	133 MHz	—	—	3.60
		125 MHz	—	—	4.00
		115 MHz	—	—	4.20
		108 MHz	—	—	4.60
		100 MHz	—	—	5.0
		<100 MHz	—	—	5.0

Programming Serial Flash Devices

You can program serial flash devices in-system using the Intel^® FPGA Download Cable II or Intel^® FPGA Ethernet Cable.

You have the following two in-system programming options:

Active Serial
JTAG

Programming Serial Flash Devices using the AS Interface

When you select AS programming the Intel^® Quartus^® Prime software or any supported third-party software programs the configuration data directly into the serial flash device.

You must set MSEL to JTAG. When MSEL is set to JTAG, the SDM tristates the following AS pins: AS_CLK, AS_DATA0-AS_DATA3, and AS_nCS0-AS_nCS3. The Intel^® Quartus^® Prime Programmer programs the flash memory devices via the AS header.

Figure 39. AS Programming Using Intel^® Quartus^® Prime or Third-Party Programmer

Programming Serial Flash Devices using the JTAG Interface

The Intel^® Quartus^® Prime Programmer interfaces to the SDM device through JTAG interface and programs the serial flash device. The SDM emulates AS programming.

Figure 40. Programming Your Serial Configuration Device Using JTAG and SDM Emulation of AS

Figure 41. Connections for Programming the Serial Flash Devices using the JTAG Interface

Intel recommends using the JTAG interface to prepare the Quad SPI flash device for later use in AS mode.

This configuration scheme includes the following steps:

In the Intel^® Quartus^® Prime Programmer, select the JTAG programming mode and initiate programming by clicking Start.
The Programmer drives .jic configuration data to the board using the JTAG header connection.
The programmer first configures the SDM with configuration firmware. Then, the SDM drives configuration data from the programmer to the AS x4 flash device using SDM_IOs.
To use the Intel^® Stratix^® 10 device in AS mode after successful programming of the flash device, set the MSEL pins to either AS fast or AS normal mode and power cycle the device.

The Intel^® Quartus^® Prime Programmer interfaces to the SDM device through JTAG interface and programs the serial flash device.

Serial Flash Memory Layout

Serial flash devices store the configuration data in sections.

The following diagram illustrates sections of a non-HPS Intel^® Stratix^® 10 configuration data mapping in a serial flash device. Refer to Intel^® Stratix^® 10 SoC FPGA Bitstream Sections of the HPS Technical Reference Manual for more information about flash memory layout for HPS devices.

Figure 42. Serial Flash Memory Layout Diagram

If you use a third-party programmer to program an .rpd, ensure that the configuration data is stored starting from address 0 of the serial flash device. If you use .jic or .pof files, the Intel^® Stratix^® 10 Programmer automatically programs the configuration data starting from address 0 of the serial flash device.

Intel currently support the following listed Supported Flash Devices for Intel^® Stratix^® 10 10

Understanding Quad SPI Flash Byte-Addressing

At power-on the SDM operates from boot ROM. The SDM loads configuration firmware from Quad SPI flash using 3-byte addressing. Once loaded, if the flash size is 256 Mb or larger, the SDM configures the Quad SPI flash to operate in 4-byte addressing mode and continues to load the rest of the bitstream until configuration completes.

Intel^® Stratix^® 10 devices support the following third-party flash devices operating at 1.8 V:

Macronix MX66U 512 Mb, 1 and 2 gigabits (Gb)
Macronix MX25U 128 Mb, 256 Mb, and 512 Mb
Micron MT25QU 128 Mb, 256 Mb, 512 Mb, 1 Gb, and 2 Gb

Micron and Macronix both offer Quad SPI memories a density range of 128Mb—2Gb.

AS_CLK

The Intel^® Stratix^® 10 device drives AS_CLK to the serial flash device. An internal oscillator or the external clock that drives the OSC_CLK_1 pin generates AS_CLK. Using an external clock source allows the AS_CLK to run at a higher frequency. If you provide a 25 MHz, 100 MHz, or 125 MHz clock to the OSC_CLK_1 pin, the AS_CLK can run up to 133 MHz.

Set the maximum required frequency for the AS_CLK pin in the Intel^® Quartus^® Prime software as described in Active Serial Configuration Software Settings. The AS_CLK pin runs at or below your selected frequency.

Table 28. Supported configuration clock source and `AS_CLK` Frequencies in Intel^® Stratix^® 10 Devices
Configuration Clock Source	`AS_CLK` Frequency (MHz)
Internal oscillator	25 58 77 115
`OSC_CLK_1`	25 50 71.5 80 100 108 125 133

Active Serial Configuration Software Settings

You must set the parameters in the Device and Pin Options of the Intel^® Quartus^® Prime software when using the AS configuration scheme.

To set the parameters for AS configuration scheme, complete the following steps:

On the Assignments menu, click Device.
In the Device and Pin Options select the Configuration category.
1. Select Active Serial x4 from the Configuration scheme drop down menu.
2. Select Auto or 1.8 V in the Configuration device I/O voltage drop-down list.
3. Select the AS clock frequency from the Active serial clock source drop-down list.
Click OK to confirm and close the Device and Pin Options.

Intel Quartus Prime Programming Steps

Generating Programming Files using the Programming File Generator

By default, the Intel^® Quartus^® Prime Compiler's Assembler module generates the primary files required for device programming at the end of full compilation. Primary programming files include the .sof, .pof, and .rpd. You can use the Programming File Generator to generate programming files for alternative device programming methods, such as the .jic for flash programming,.rbf for partial reconfiguration, or .rpd for third-party programmer configuration. The Programming File Generator supports Intel^® Stratix^® 10 and later devices. The legacy Convert Programming Files dialog box does not support some advanced programming features for Intel^® Stratix^® 10 and later devices.

Note: If you are generating an .rpd for remote system update (RSU), you must follow the instructions in Generating an Application Image in the Remote System Update chapter. This procedure generates flash programming files for Intel^® Stratix^® 10 devices.

Complete the following steps generate the programming file or files you require:

Click File Programming File Generator.
For Device Family select Intel^® Stratix^® 10
In the Configuration mode, select Active Serial x4.
Specify the Output directory and Name for the file you generate.
Under Output directory, select the appropriate file type for your design. The AS scheme supports the Programmer Object File (.pof), JTAG Indirect Configuration File (.jic), and Raw Programming Data File (.rpd) file types.

Figure 43. Programming File Generator Output Files
For the JTAG Indirect Configuration File (.jic) and Programmer Object File (.pof) you can turn on the Memory Map File (.map). This option describes flash memory address locations.
The Input Files tab is now available.
On the Input Files tab, click Add Bitstream and browse to your configuration bitstream.

Figure 44. Programming File Generator Input Files
On the Configuration Device tab, click Add Device. You can select your flash device from the from the Configuration Device list, or define a custom device using the available menu options. For more information about defining a custom configuration device, refer to the Configuration Device Tab Settings (Programming File Generator) in the Intel^® Quartus^® Prime Pro Edition User Guide: Programmer

Figure 45. Programming File Generator Input Files

Note: You do not need to specify the flash device for .rpd files because the .rpd format is independent of the flash device. In contrast, the .poc and .jic files include both programming data and additional data specific to the configuration device. The Intel^® Quartus^® Prime Programmer uses this additional data to establish communication with the configuration device and then write the programming data.
Click Generate to generate the programming file or files.

Programming .pof files into Serial Flash Device

To program the .pof into the serial flash device through the AS header, perform the following steps:

In the Programmer window, click Hardware Setup and select the desired download cable.
In the Mode list, select Active Serial Programming.
Click Auto Detect button on the left pane.
Select the device to be programmed and click Add File.
Select the .pof to be programmed to the selected device.
Click Start to start programming.

Programming .jic files into Serial Flash Device

To program the .jic into the serial flash device through the JTAG interface, perform the following steps:

In the Programmer window, click Hardware Setup and select the desired download cable.
In the Mode list, select JTAG.
Select the device to be programmed and click Add File.
Select the .jic to be programmed to the selected device.
Click Start to start programming.

Debugging Guidelines for the AS Configuration Scheme

The AS configuration scheme operation is like earlier device families. However, there is one significant difference. Intel^® Stratix^® 10 devices using AS mode, try to load a firmware section from addresses 0, 256k, 512k and 768k in the serial flash device connected to the CS0 pin.

If the configuration bitstream does not include a valid image, the SDM asserts an error by driving nSTATUS low. You can recover from the error by reconfiguring the FPGA over JTAG, or by driving nCONFIG low.

SDM tristates AS pins, AS_CLK, AS_DATA0-AS_DATA3, and AS_nCS0-AS_nCS3, only when the device powers on if you set MSEL to JTAG. If MSEL is either AS fast or normal, the SDM drives the AS pins until you power cycle the Intel^® Stratix^® 10 device. Unlike earlier device families, the AS pins are not tristated when the device enters user mode.

The AS configuration scheme has power-on requirements. If you use AS Fast mode and are not concerned about 100 ms PCIe* link training requirement, you must still ramp the V_{CCIO_SDM} supply within 18 ms. This ramp-up requirement ensures that the AS x4 device is within its operating voltage range when the Intel^® Stratix^® 10 device begins assessing the AS x4 device.

When using AS fast mode, all power supplies to the Intel^® Stratix^® 10 device must be fully ramped-up to the recommended operating conditions before the SDM releases from reset. To meet the PCIe* 100 ms power-up-to-active time requirement for CvP, the V_{CCIO_SDM} power to the Intel^® Stratix^® 10 device must be at the recommended operating range within 10 ms.

Debugging Suggestions

Here are some debugging tips for the AS configuration scheme:

Ensure that the boot address for your configuration image is correctly defined when generating the programming file for the flash. The boot address defaults to 0 for AS configuration.
Ensure that the design meets the power-supply ramp requirements for fast AS mode. If using fast mode, V_{CCIO_SDM} must ramp up within 18 ms.
Ensure that the flash is powered up and ready to be accessed when the Intel^® Stratix^® 10 device exits power-on reset.
If you are using an external clock source for configuration, ensure the OSC_CLK_1 pin is fed correctly, and the frequency matches the frequency you set for the OSC_CLK_1 in your Intel^® Quartus^® Prime Pro Edition project.
Ensure the MSEL pins reflect the correct AS configuration scheme.
If the AS configuration is failing due to a corrupt image inside the serial flash device and reprogramming the flash via JTAG does not work, change the MSEL pins to JTAG only mode, verify that configuration is successful over JTAG. Then, erase and reprogram the serial flash device.
If you are using AS x4 flash memories, ensure that you use AS Fast mode, if you are not concerned about 100 ms PCIe* linkup, you must still ramp the V_{CCIO_SDM} supply within 18 ms. This ramp-up requirement ensures that the AS x4 device is within its operating voltage range when the Intel^® Stratix^® 10 device begins to access it.
Check endianness of the .rpd if using a third-party programmer to program Quad SPI device. You should generate the .rpd as big endian.

QSF Assignments for AS

# Fitter Assignments
# ==================
set_global_assignment -name DEVICE 1SG280LU3F50E3VG
set_global_assignment -name CONFIGURATION_VCCIO_LEVEL 1.8V

# SDM IO Assignments
# ==================
set_global_assignment -name USE_PWRMGT_SCL SDM_IO14
set_global_assignment -name USE_PWRMGT_SDA SDM_IO11
set_global_assignment -name USE_PWRMGT_ALERT SDM_IO12
set_global_assignment -name USE_CONF_DONE SDM_IO16
set_global_assignment -name USE_INIT_DONE SDM_IO0
set_global_assignment -name USE_CVP_CONFDONE SDM_IO13
set_global_assignment -name USE_SEU_ERROR SDM_IO10
set_global_assignment -name SDM_DIRECT_TO_FACTORY_IMAGE SDM_IO15

# Configuration settings
# ======================
set_global_assignment -name STRATIXV_CONFIGURATION_SCHEME "ACTIVE SERIAL X4"
set_global_assignment -name USE_CONFIGURATION_DEVICE ON
set_global_assignment -name ERROR_CHECK_FREQUENCY_DIVISOR 256
set_global_assignment -name GENERATE_PR_RBF_FILE ON
set_global_assignment -name ENABLE_ED_CRC_CHECK ON
set_global_assignment -name MINIMUM_SEU_INTERVAL 479

# SmartVID feature PMBus settings [Slave mode settings only]
# ==========================================================
set_global_assignment -name VID_OPERATION_MODE "PMBUS SLAVE"
set_global_assignment -name PWRMGT_DEVICE_ADDRESS_IN_PMBUS_SLAVE_MODE 3F

You can also set the SDM_IO configuration pins using the Assignments > Device > Device and Pin Options > Configuration > Configuration Pin Options.

Figure 46. Set SDM_IO Configuration Pins Using the Intel^® Quartus^® Prime Software

SD/MMC Configuration

Note: Contact your Intel^® sales representative for information about SD/MMC support.

In the configuration scheme using SD memory cards or MMC, the memory cards store configuration data. The SDM uses the on-chip SD or MMC controller to interface to the memory cards. The SDM block reads the configuration data from the memory cards for the configuration process. The configuration from SD and MMC supports x4 SD memory cards and x8 MMC.

Table 29. Intel^® Stratix^® 10 Configuration Data Width, Clock Rates, and Data Rates
Mode		Data Width (bits)	Max Clock Rate	Max Data Rate	MSEL[2:0]
Active	SD/MMC	4 or 8	50 MHz	400 Mb	3'b100

Table 30. Required Configuration Signals for the SD/MMC Configuration SchemeYou specify SDM I/O pin functions using the Device > Configuration > Device and Pin Options dialog box in the Intel^® Quartus^® Prime software.
Configuration Function		Direction	Powered by
`nSTATUS`	SDM I/O	Output	V_{CCIO_SDM}
`nCONFIG`	SDM I/O	Input	V_{CCIO_SDM}
`MSEL[2:0]`	SDM I/O, Dual-Purpose	Input	V_{CCIO_SDM}
`SDMMC_CFG_CMD`	GPIO	Output	V_{CCIO_SDM}
`SDMMC_CFG_DATA[7:0]`	GPIO	Bidirectional	V_{CCIO_SDM}
`SDMMC_CFG_CCLK`	GPIO	Output	V_{CCIO_SDM}

Note: Although the CONF_DONE and INIT_DONE configuration signals are not required, Intel recommends that you use these signals. The SDM drives the CONF_DONE signal high after successfully receiving full bitstream. The SDM drives the INIT_DONE signal high to indicate the device is fully in user mode. You can reassign the GPIO, dual-purpose configuration pins for other functions in user mode.

SD/MMC Single-Device Configuration

Figure 47. Connections for SD/MMC Single-Device Configuration

Note: The External Level Shifter is not mandatory for embedded multimedia cards (eMMC).

JTAG Configuration

JTAG-chain device programming is ideal during development. JTAG-chain device configuration uses the JTAG pins to configure the Intel^® Stratix^® 10 FPGA directly with the .sof file. Configuration using the JTAG device chain is faster because it does not require you to program an external flash memory. You can also use JTAG to reprogram if image stored in memory is corrupt, preventing the device from configuring using the AS scheme.

The Intel^® Quartus^® Prime software generates a .sof containing the FPGA design information. You can use the .sof with a JTAG programmer to configure the Intel^® Stratix^® 10 device. The Intel^® FPGA Download Cable II and the Intel^® FPGA Ethernet Cable both can support the V_{CCIO_SDM} supply at 1.8 V. Alternatively, you can use the Jam*STAPL Format File (.jam) or Jam* Byte Code File (.jbc) for JTAG configuration.

Intel^® Stratix^® 10 devices automatically compress the configuration bitstream. You cannot disable compression in Intel^® Stratix^® 10 devices.

Table 31. Intel^® Stratix^® 10 Configuration Data Width, Clock Rates, and Data Rates
Mode		Data Width (bits)	Max Clock Rate	Max Data Rate	MSEL[2:0]
Passive	JTAG	1	30 MHz	30 Mb	3'b111

Note: The JTAG port has the highest priority and overrides the MSEL pin settings. Consequently, you can configure the Intel^® Stratix^® 10 device over JTAG even if the MSEL pin specify a different configuration scheme unless you disabled JTAG for security reasons.

Table 32. Power Rails for the Intel^® Stratix^® 10 Device Configuration PinsYou can view the pin assignments for fixed pins in the Pin-Out File for your device. You specify SDM I/O pin functions using the Device > Configuration > Device and Pin Options dialog box in the Intel^® Quartus^® Prime software.
Configuration Function	Pin Type	Direction	Powered by
`TCK`	Fixed	Input	V_{CCIO_SDM}
`TDI` ⁶	Fixed	Input	V_{CCIO_SDM}
`TMS` ⁶	Fixed	Input	V_{CCIO_SDM}
`TDO` ⁶	Fixed	Output	V_{CCIO_SDM}
`nSTATUS`	SDM I/O	Output	V_{CCIO_SDM}
`nCONFIG`	SDM I/O	Input	V_{CCIO_SDM}
`MSEL[2:0]`	SDM I/O, Dual-Purpose	Input	V_{CCIO_SDM}

Note: Pin-Out files are not yet available for Intel^® Agilex™ devices. Intel^® Agilex™

⁶ The JTAG pins can access the HPS JTAG chain in Intel^® Stratix^® 10 SoC devices.

JTAG Configuration Scheme Hardware Components and File Types

The following figure illustrates JTAG programming. This is the simplest device configuration scheme. You do not have to use the File > Programming File Generator to convert the .sof file to a .pof.

Figure 48. JTAG Configuration Scheme

Figure 49. Components and Design Flow for JTAG Programming

JTAG Device Configuration

To configure a single device in a JTAG chain, the programming software sets the other devices to bypass mode. A device in bypass mode transfers the programming data from the TDI pin to the TDO pin through a single bypass register. The configuration data is available on the TDO pin one clock cycle later.

You can configure the Intel^® Stratix^® 10 device through JTAG using a download cable or a microprocessor.

JTAG Single-Device Configuration using Download Cable Connections

Refer to the Intel^® Stratix^® 10 Device Family Pin Connection Guidelines for additional information about individual pin usage and requirements.

Figure 50. Connection Setup for JTAG Single-Device Configuration using Download Cable

JTAG Single-Device Configuration using a Microprocessor

Refer to the Intel^® Stratix^® 10 Device Family Pin Connection Guidelines for additional information about individual pin usage and requirements.

Figure 51. Connection Setup for JTAG Single-Device Configuration using a Microprocessor

JTAG Multi-Device Configuration

You can configure multiple devices in a JTAG chain. Observe the following pin connections and guidelines for this configuration setup:

One JTAG-compatible header connects to several devices in a JTAG chain. The drive capability of the download cable is the only limit on the number of devices in the JTAG chain.
If you have four or more devices in a JTAG chain, buffer the TCK, TDI, and TMS pins with an on-board buffer. You can also connect other Intel FPGA devices with JTAG support to the chain.

JTAG Multi-Device Configuration using Download Cable

Figure 52. Connection Setup for JTAG Multi Device Configuration using Download Cable

Debugging Guidelines for the JTAG Configuration Scheme

The JTAG configuration scheme overrides all other configuration schemes. The SDM is always ready to accept configuration over JTAG unless a security feature disables the JTAG interface. JTAG is particularly useful in recovering a device that may be in an unrecoverable state reached when trying to configure using a corrupted image.

An nCONFIG falling edge terminates any JTAG access and the device reverts to the MSEL-specified boot source. nCONFIG must be stable during JTAG configuration. nSTATUS follows nCONFIG during JTAG configuration. Consequently, nCONFIG also must be stable.

Unlike other configuration schemes, nSTATUS does not assert if an error occurs during JTAG configuration. You must monitor the error messages that the Intel^® Quartus^® Prime Pro Edition Programmer generates for error reporting.

Note: For Intel^® Stratix^® 10 SX devices when you choose to configure the FPGA fabric first, the JTAG chain has no mechanism to redeliver the HPS boot information following a cold reset. Consequently, you must reconfig the device with the .sof file or avoid cold resets to continue operation.

Debugging Suggestions

Here are some debugging tips for JTAG:

Verify that the JTAG pin connections are correct.
If JTAG configuration is failing, check that the FPGA has successfully powered up and exited POR. One strategy is to check the hand shaking behavior between nCONFIG and nSTATUS by driving nCONFIG low and ensuring that nSTATUS also goes low.
Verify that the nCONFIG pin remains high during JTAG configuration.
Another way to determine whether the device has exited the POR state is to use the Intel^® Quartus^® Prime Programmer to detect the device. If the programmer can detect the Intel^® Stratix^® 10 device, it has exited the POR state.
If you are using an Intel^® FPGA Download Cable II, reduce the cable clock speed to 6 MHz.
If you have multiple devices in the JTAG chain, try to disconnect other devices from the JTAG chain to isolate the Intel^® Stratix^® 10 device.
If you specify the OSC_CLK_1 as the clock source for configuration, ensure that OSC_CLK_1 is running at the frequency you specify in the Intel^® Quartus^® Prime software.
For designs including the High Bandwidth Memory (HBM2) IP or any IP using transceivers, you must provide a free running and stable reference clock to the device before device configuration begins. All transceiver power supplies must be at the required voltage before configuration begins.
When you use the JTAG interface for reconfiguration after an initial reconfiguration using AS or the Avalon^® -ST interface, the .sof must be in the file format you specified in the Intel^® Quartus^® Prime project. For example, if initially configure the MSEL pins for AS configuration and configure using the AS scheme, a subsequent JTAG reconfiguration using a .sof generated for Avalon^® -ST fails.

Remote System Update (RSU)

RSU implements device reconfiguration using dedicated RSU circuitry available in all Intel^® Stratix^® 10 devices. RSU has the following advantages:

Provides a mechanism to deliver feature enhancements and bug fixes without recalling your products
Reduces time-to-market
Extends product life

Using RSU and the Mailbox Client Intel^® Stratix^® 10 FPGA IP Intel^® FPGA IP core you can write configuration bitstreams to the AS x4 flash device. Then you can use the Mailbox Client Intel^® Stratix^® 10 FPGA IP core to instruct the SDM to restart from the updated image. You can store multiple application images and a single factory image in the configuration device. Your design manages remote updates of the application images in the configuration device.

A command to the Mailbox Client Intel^® Stratix^® 10 FPGA IP core initiates reconfiguration. The RSU performs configuration error detection during and after the reconfiguration process. If errors in the application image or images prevent reconfiguration, the configuration circuitry reverts to the factory image and provides error status information.

This chapter explains the remote system update implementation for active configuration schemes. The FPGA drives the RSU. For the Intel^® Stratix^® 10 SoC devices, HPS can drive the RSU process. For more information about using the HPS to drive RSU, refer to Intel^® Stratix^® 10 SoC Remote System Update (RSU) User Guide.

For passive configuration schemes, an external host implements remote system update rather than the Intel^® Stratix^® 10 device. To learn more about remote system update for passive configuration schemes, refer to Remote Update Intel^® FPGA IP User Guide for remote system update implementations in earlier device families.

The following figure shows functional diagrams for typical remote system update processes.

Figure 53. Typical Remote System Update Process

Remote System Update Functional Description

RSU Glossary

Table 33. RSU Glossary
Term	Meaning
Firmware	Firmware that runs on SDM. Implements many functions including the functions listed here: FPGA configuration Voltage regulator configuration Temperature measurement HPS software load HPS Reset RSU Read, erase, and program flash memory Device security, including authentication and encryption
Decision firmware	Firmware to identify and load the highest priority image. Previous versions of this user guide refer to decision firmware as static firmware. Starting in version 19.1 of the Intel^® Quartus^® Prime software, you can use RSU to update this firmware.
Decision firmware data	Decision firmware data structure containing the following information: The Direct to Factory Image pin assignment. PLL settings for the external clock source. This optional clock source drives `OSC_CLK_1`. For more information, refer to OSC_CLK_1 Clock Input. Quad SPI pins.
Configuration pointer block (CPB)	A list of application image addresses in order of priority. When you add an image, that image becomes the highest priority.
Sub-partition table (SPT)	Data structure to facilitate the management of the flash storage.
Application image	Configuration bitstream that implements your design. This image includes the SDM firmware.
Factory image	The fallback configuration bitstream that the RSU loads when all attempts to load an application image fail.
Initial RSU flash image	Contains the factory image, the application images, and the associated RSU data structures.
Factory update image	An image that updates the following RSU-releated items in flash: The factory image The decision firmware The decision firmware data

Remote System Update Using AS Configuration

Remote system update using AS configuration includes the following components:

Your remote system update host design. The host can be custom logic, the HPS, or a Nios^® II processor in the FPGA.
One factory image.
Flash memory for image storage.
At least one application image.
Designs that do not use the HPS as the remote system update host require an Intel^® Stratix^® 10 Mailbox Client FPGA IP core as shown in the figure below. The Mailbox Client sends and receives remote system update operation commands and responses, such as QSPI_READ and QSPI_WRITE. Optionally, you may use Intel^® Stratix^® 10 Serial Flash Mailbox Client IP to perform flash transactions, such as rewriting the application image to the serial flash.
Figure 54. Intel^® Stratix^® 10 Remote System Update Components

Attention:

Starting in version 19.2 of the Intel^® Quartus^® Prime software, a restriction applies to the following mailbox client IPs that access the SDM mailbox over an Avalon^® Memory-Mapped ( Avalon^® -MM) interface:

Temperature Sensor
Voltage Sensor
Chip ID
Serial Flash Mailbox Client
Mailbox Client IP
Advanced SEU Detection IP
Partial Reconfiguration IP

You cannot use a .sof created in version 19.2 of the Intel^® Quartus^® Prime or later that includes any of the listed SDM mailbox client IP for bitstream generation or for JTAG configuration in Intel^® Quartus^® Prime software version 19.1 or earlier. However, you can use a .sof including the listed SDM mailbox client IPs that you created in version 19.1 of the Intel^® Quartus^® Prime or earlier, in later versions of the Intel^® Quartus^® Prime software, including Intel^® Quartus^® Prime version 19.2.

Remote System Update Configuration Images

Intel^® Stratix^® 10 devices using remote system update require the following configuration images:

A Factory image—This image includes logic to implement the following functions:
- Your design-specific logic to obtain new application images
- Your design-specific logic to request reconfiguration using a specific application image
- Image storage in flash memory
Application image—contains logic to implement the custom application. The application image must also contain logic to obtain new application images and store the images in the flash memory.

Depending on the storage space of your flash memory, Intel^® Stratix^® 10 remote system update supports one factory image and up to 507 application images. The Quartus Programming File Generator only supports up to three remote system update images. However, you can add more images using the Mailbox Client IP or Serial Flash Mailbox Client IP with the device in user mode.

Remote System Update Configuration Sequence

Figure 55. Remote System Update Configuration Sequence

In the following figure the blue text are states shown in the Configuration Flow Diagram.

Reconfiguration includes the following steps:

After the device exits power-on-reset (POR), the boot ROM loads flash memory from the first valid decision firmware from one of the copies at addresses 0, 256k, 512k, or 768k to initialize the SDM. The same configuration firmware is present in each of these locations. This firmware is part of the configuration bitstream. ( Refer to Step 2 of Guidelines for Performing Remote System Update Functions for Non-HPS for step-by-step details for programming the firmware into the flash.)
The optional Direct to Factory pin controls whether the SDM firmware loads the factory or application image. You can assign the Direct to Factory input to any unused SDM pin. The SDM loads the application image if you do not assign this pin.
The configuration pointer block in the flash device maintains a list of pointers to the application images.
When loading an application image, the SDM traverses the pointer block in reverse order. The SDM loads the highest priority image. When image loading completes, the device enters user mode.
If loading the newest (highest priority) image is unsuccessful, the SDM tries the next application image from the list. If none of the application loads successfully, the SDM loads the factory image.
If loading the factory image fails, you can recover by reprogramming the quad SPI flash with the initial RSU flash image using the JTAG interface.

RSU Recovery from Corrupted Images

When an RSU fails, the Mailbox Client Intel^® Stratix^® 10 FPGA IP RSU_STATUS command provides information about the current configuration status, including the currently running image and most recent failing image. The rsu1.tcl script implements the RSU_STATUS commands. You can download the rsu1.tcl script from the following web page. Under Device Configuration Support Center, click Advanced Configuration Features, click the triangle next to Remote System Upgrade to expand this section, then click Example of Tcl Script.

The following example illustrates recovery from a corrupted image:

Multiple Corrupted Images

If the flash memory includes multiple corrupted images, the RSU_STATUS only reports status for the highest priority failing image. The following example illustrates this procedure.

The flash memory includes the following four images, in order of priority:
1. Application Image3 (highest priority)
2. Application Image2
3. Application image1
4. Application image0 (lowest priority)
Application Image3, Application Image2, and Application Image1 are corrupted.
RSU_STATUS includes the following information:
- Current_Image: Application Image0
- Highest priority failing image, State, Version, Error location, Error details: records information for Application Image3 which is the highest priority failing image.

Figure 56. Multiple Corrupt Images

Updates with the Factory Update Image

In rare instances you may need to update flash memory with a new factory image and the associated decision firmware and decision firmware data.

An update may be required for the following reasons:

If there are vulnerabilities in the firmware
If there are errors in the firmware or in the factory image

Intel provides a safe solution for you to update the factory image and the associated decision firmware and decision firmware data remotely. The update process stores multiple copies of critical data so that if power is lost or the update is disrupted, the device is still able to restart and continue the update. The update continues automatically when power is restored. Here are the steps to perform the update:

Generate the factory update image using the Programming File Generator. The image contains the new factory Image, decision firmware, and decision firmware data.
Program the factory update image, (*.rpd) to an empty partition slot starting from a new sector boundary in the flash device. If an empty slot is not available, program the update file to any application slot other than application image 1.
Trigger reconfiguration to load the update image from the starting address specified in step 2.
The updated image performs the following operations:
1. Erases and replaces the previous decision firmware and decision firmware data in the flash device.
2. Reprograms the new factory image in the flash device.
3. After the update completes, the updated image removes itself from the flash and loads the application image or the factory image if an application image is not available.
If the update process used an application slot, you must restore the application image by writing the application image .rpd to the application slot.

Guidelines for Performing Remote System Update Functions for Non-HPS

Figure 57. Intel^® Stratix^® 10 Modules and Interfaces to Implement RSU Using Images Stored in Flash Memory

Here are guidelines to follow when implementing remote system update:

The factory or application image must at least contain a remote system update host controller and the Mailbox Client Intel^® Stratix^® 10 FPGA IP.
- You can use either custom logic, the Nios^® II processor, or the JTAG to Avalon Master Bridge IP as a remote system update host controller.
- The remote system update host controller controls the remote system update function by sending commands to and receiving responses from the SDM via Mailbox Client Intel^® Stratix^® 10 FPGA IP. The Mailbox Client functions as the messenger between the remote system update host and SDM. It passes the commands to and responses from the SDM.
The pre-generated standard remote system update image file should include a factory image and at least one application image. The remote system update image must be programmed into the flash memory. You can use a dummy image to begin developing RSU functionality before the actual application image is complete. In user mode you can program additional application images.
- Refer to Generating Remote System Update Image Files Using the Programming File Generator for the step by step process to generate the standard and single remote system update image files using the programming file generator.
The remote system update requires you to use the AS x4 configuration scheme to configure the FPGA with the pre-generated remote system update image.
Once the device enters user mode with either the factory image or an application image, the remote system update host can perform the following remote system update operations:
1. Reconfiguring the device with an application or factory image:
  1. From factory image to an application image or vice versa
  2. From an application image to another application image
2. Erasing the application image
3. Adding an application image
4. Updating an application or factory image

Commands and Responses

The remote system update host communicates with the SDM using command and response packets via the Mailbox Client Intel^® Stratix^® 10 FPGA IP.

Block Diagram

The following figure illustrates the role of the Mailbox Client Intel^® Stratix^® 10 in a Intel^® Stratix^® 10 design. The Mailbox Client IP enables communication with the SDM to access quad SPI flash memory and system status.

Mailbox Client Role

Figure 58. Command and Response Header Format

Note: The LENGTH field in the command header must match the command length of corresponding command.

The following table describes the fields of the header command.

Table 34. Mailbox Client Intel^® Stratix^® 10 FPGA IP Command and Response Header Description
Header	Bit	Description
Reserved	[31:28]	Reserved.
`ID`	[27:24]	The command ID. The response header returns the ID specified in the command header. Set different IDs in each command to match responses with commands.
`0`	[23]	Reserved.
`Length`	[22:12]	Number of words of arguments following the header. The IP responds with an error if a wrong number of words of arguments is entered for a given command.
Reserved	[11]	Reserved. Must be set to 0.
`Command Code/Error Code`	[10:0]	`Command Code` specifies the command. The `Error Code` indicates whether the command succeeded or failed. In the command header, these bits represent command code. In the response header, these bits represent error code.

Operation Commands

Table 35. Mailbox Client Intel^® Stratix^® 10 FPGA IP Command List and Descriptions (RSU Functions for Non-HPS Variants)
Command	Code (Hex)	Command Length ⁷	Response Length ⁷	Description
`RSU_IMAGE_UPDATE`	5C	2	0	Triggers reconfiguration from data source or configuration data stored in AS x4 flash memory. This command takes a 64-bit argument to specify the reconfiguration address in flash memory. Bit [63:32]: Reserved (write as 0). Bit [31:0]: The start address of an application or factory image. Returns non-zero response if the device is already processing a configuration.
`CONFIG_STATUS`	4	0	6	Reports the status of the last reconfiguration. You can use this command to check the configuration status during and after configuration. The response contains the following fields:
				Word	Summary	Description
				0	State	Describes the most recent configuration related error. 0 when no configuration errors. The error field has 2 fields: Upper 16 bits: Major error code Lower 16 bits: Minor errors that do not contain meaningful data. Refer to the Table 36 and Table 37 for more information.
						Major Error Code	Error Type
						0xF001	`BITSTREAM_ERROR`
						0xF002	`HARDWARE_ACCESS_FAILURE`
						0xF003	`BITSTREAM_CORRUPTION`
						0xF004	`INTERNAL_ERROR`
						0xF005	`DEVICE_ERROR`
						0xF006	`HPS_WATCHDOG_TIMEOUT`
						0xF007	`INTERNAL_UNKNOWN_ERROR`
				1	Version	0 for this version.
				2	Pin status	Specifies the values of the following configuration control signals. Bit [31]: Current `nSTATUS` output value (active low). Bit [30]: Detected `nCONFIG` input value (active low). Bit [29:3]: Reserved. Bit [2:0]: The `MSEL` value at power up.
				3	Soft function status	Specifies the value of each of the soft functions, whether you assigned the function assigned to an SDM pin. Bit [31:4]: Reserved. Bit [3]: `SEU_ERROR`. Bit [2]: `CVP_DONE`. Bit [1]: `INIT_DONE`. Bit [0]: `CONF_DONE`.
				4	Error location	Contains the error location. Returns 0 for no error.
				5	Error details	Contains the error details. Returns 0 for no error.
`RSU_STATUS`	5B	0	8	Reports the current remote system update status. This command returns the following responses:
				Word	Summary	Description
				0-1	Current image	Flash offset of the currently running application image.
				2-3	First failing image	Flash offset of the first failing application image. A value of all 1s indicates no failing images. When there are no failing images, the following words do not contain meaningful data.
				4	State	Failure code of the last failing image. The error field has two parts: Upper 16 bits: Major error code. Lower 16 bits: Minor errors that do not contain meaningful data. The following major error codes are defined:
						Major Error Code	Error Type
						0xF001	`BITSTREAM_ERROR`
						0xF002	`HARDWARE_ACCESS_FAILURE`
						0xF003	`BITSTREAM_CORRUPTION`
						0xF004	`INTERNAL_ERROR`
						0xF005	`DEVICE_ERROR`
						0xF006	`HPS_WATCHDOG_TIMEOUT`
						0xF007	`INTERNAL_UNKNOWN_ERROR`
						Refer to the Table 36 and Table 37 for more information.
				5	Version	Contains the value of each of the soft functions, whether that function is on an SDM pin.
				6	Error location	Contains the error location of the last failing image. Returns 0 for no error.
				7	Error details	Contains the error details of the last failing image. Returns 0 for no error.
`GET_DESIGNHASH`	14	1	3	The GET_DESIGNHASH command is little-endian. This command returns a number that is the 3 least significant words of the hash value. Number format is 0x(Byte N-1)(Byte N-2)...(Byte2)(Byte1)(Byte0). In contrast, the Assembler reports a complete hash value as hex stream, with byte ordering from low to high, (Byte0)(Byte1)(Byte2)... (Byte N-2)(Byte N-1). For example, for the design hash value dbe4554730f6bb487e52eb31cc68b194 from the Assembler report (*.asm.rpt), the Mailbox Client `GET_DESIGNHASH` command returns 0x31eb527e48bbf6304755e4db.
`QSPI_OPEN`	32	0	0	Clients use this command to request exclusive access AS x4 interface. The SDM returns the appropriate response: If AS x4 is available, the SDM grants the request by responding OK. If the SDM is in the process of configuring the device or AS x4 is in use, the SDM returns the error response. Note: The SDM grants exclusive access only a single client using this mailbox. Other clients are not able to access AS x4 until the currently granted client sends the `QSPI_CLOSE` command.
`QSPI_CLOSE`	33	0	0	Closes exclusive access to the AS x4.
`QSPI_SET_CS`	34	1	0	Selects the flash memory using chip select lines. Bit [31:28]: Chip selects for flash memories 0-4 using one-hot encoding. Bit [27:0]: Reserved (write as 0). The `QSPI_SET_CS` is mandatory for all configuration schemes except AS x4. `QSPI_SET_CS` is not mandatory if you are using the AS x4 configuration scheme.
`QSPI_READ`	3A	2	N	Reads the attached quad SPI flash device. Defines the following 2 parameters: The flash address offset to start a read. The read must be word-aligned. Reads one word. Number of words to read. When successful returns the OK response code followed by the data read from the quad SPI flash device. When unsuccessful, returns 1 of the following responses: Returns an error code. In the case of multiple reads, if some of the reads fail returns read data for the successful reads and OK for the failure. Note: The maximum transfer size is 4 KB. The `QSPI_READ` command cannot run during configuration.
`QSPI_WRITE`	39	2+N	0	Writes data to the flash memory. Defines the following 3 parameters: The word start address in flash memory. The write must be word aligned. The size in words. The data. A successful write returns an OK response code. The client may need to issue `QSPI_ERASE` command before issuing this command to prepare the memory for writing. Note: The maximum transfer size is limited to 4 KB. The `QSPI_WRITE` command cannot run during configuration.
`QSPI_ERASE`	38	2	0	Erases a sector of the flash memory. Defines the following 2 parameters: The word start address in flash memory to begin the erasure. The address must be 4 KB aligned. The number of words to erase. The erasure size must be a multiple of 1024 words. A successful erase returns an OK response code.
`QSPI_READ_DEVICE_REG`	35	2	N	Reads registers from the flash memory. If the data is not a multiple of 4 bytes, it is padded with 0 bytes until the next word boundary. Takes 2 parameters: The opcode for the read command. The number of bytes to read. The maximum size is 8 bytes. A successful read returns an OK response code followed by the data read from the device.
`QSPI_WRITE_DEVICE_REG`	36	2+N	0	Writes to registers on the flash. Takes 3 arguments: The opcode for the write command. The number of bytes to write. The maximum size is 8 bytes. The data. The maximum write is 2 words, padded with 0 to the word boundary. You must specify the serial flash address in most significant byte (MSB) to least significant byte (LSB) order as illustrated by the example below. Example: Sector or Sub-Sector Erase or using the Mailbox Client FPGA IP. To perform a sector erase on Micron's 2 Gb flash at flash address 0x04FF0000 using the `QSPI_WRITE_DEVICE_REG` command, you write the flash address in MSB to LSB as shown: Header: 0x00003036 Opcode: 0x000000DC Number of bytes to write: 0x00000004 Flash address: 0x0000FF04
`QSPI_SEND_DEVICE_OP`	37	1	0	Sends a command opcode to flash memory. Takes 1 argument: The opcode to send the attached flash memory. A successful command returns an OK response code.

Table 36. CONFIG_STATUS and RSU_STATUS Major Error Code Descriptions
Major Error Code	Error Type	Description
0xF001	`BITSTREAM_ERROR`	Potential unsigned bitstream used. Ensure the bitstream is signed with the correct key.
0xF002	`HARDWARE_ACCESS_FAILURE`	Failure to communicate to PMBus-compliant voltage regulator. Check your power management and smart voltage identification (SmartVID) parameter settings and PMBus interface connections.
0xF003	`BITSTREAM_CORRUPTION`	Bitstream is corrupt. Ensure the bit stream in configuration device or flash is not corrupt.
0xF004	`INTERNAL_ERROR`	This major code may indicate the following error events: An error in the SDM Crypto IP task. An RSU operation error. Refer to Table 37 for more information.
0xF005	`DEVICE_ERROR`	Indicates an SDM internal device error. The following errors are possible: A device cleaning failure An HPS configuration failure Contact your local Field Applications Engineer (FAE). Alternatively, submit a Service Request on the My Intel support page to get the support on capturing the error log for further debug.
0xF006	`HPS_WATCHDOG_TIMEOUT`	HPS watchdog timeout failure. Ensure that your design resets the watchdog timer correctly.
0xF007	`INTERNAL_UNKNOWN_ERROR`	Indicates an internal device error due to an unknown task. You can contact your local Field Applications Engineer (FAE). Alternatively, submit a Service Request on the My Intel support page to get the support on capturing the error log for further debug.

Table 37. CONFIG_STATUS and RSU_STATUS Minor Error Code Descriptions
Minor Error Code	Error Type	Description
0xD001	`RSU_CMF_AUTH_ERR`	Authentication failure for the firmware.
0xD002	`RSU_USER_AUTH_ERR`	Authentication failure for the design.
0xD003	`RSU_CMF_DESC_SHA_MISMATCH`	The SHA does not match for the firmware descriptor.
0xD004	`RSU_POINTERS_NOT_FOUND_ERR`	Unable to read data from boot ROM on first boot after the device exits power-on reset (POR).
0xD005	`RSU_QSPI_REQ_CHANGE`	Unable to configure the quad SPI flash during RSU initialization.
0xD006	`RSU_FACTORY_IMAGE_FAILED`	Failed to load any image, including the factory image.
0xD007	`RSU_CMF_TYPE_ERR`	The firmware version does not match the version that was previously loaded.

⁷ This number does not include the command and response header.

Error Code Responses

Table 38. Mailbox Client Intel^® Stratix^® 10 FPGA IP Error Code Responses and Description
Value (Hex)	Error Code Response	Description
0	`OK`	Indicates that the command completed successfully. Depending on the command delivered to the Mailbox Client, the response error code may not be sufficient to ensure that the operation completed successfully.
1	`INVALID_COMMAND`	Indicates that the command is in an incorrect format.
2	`UNKNOWN_BR`	Indicates that the command code is not understood.
3	`UNKNOWN`	Indicates that the command code is not understood by the currently loaded firmware.
100	`NOT_CONFIGURED`	Indicates that the device is not configured.
1FF	`ALT_SDM_MBOX_RESP_DEVICE_BUSY`	Indicates that the device is busy.
2FF	`ALT_SDM_MBOX_RESP_NO_VALID_RESP_AVAILABLE`	Indicates that there is no valid response available.
3FF	`ALT_SDM_MBOX_RESP_ERROR`	General Error

QSPI Flash Layout

High Level Flash Layout

Standard (non-RSU) Flash Layout

In the standard (non-RSU) case, the flash contains four firmware images and the application image. To guard against possible corruption, there are four redundant copies of the firmware. The firmware contains a pointer to the location of the highest priority application image in flash. Typically the application image is immediately after the four firmware copies, but the Intel^® Quartus^® Prime Pro Edition tools do not require this location.

Figure 59. Flash Layout - Non-RSU

RSU Flash Layout – SDM Perspective

In the RSU case, decision firmware replaces the standard firmware. The decision firmware copies have pointers to the following structures in flash:

Decision data
One factory image
Two Pointer Blocks (CPBs)

Figure 60. RSU Flash Layout - SDM Perspective

The decision firmware data stores basic settings, including the following:

The clock and pins that connect to quad SPI flash memory
The Direct to Factory Image pin that forces the SDM to load the factory image. ( You can set this pin on the following menu: Assignments > Device > Device and Pin Options > Configuration > Configuration Pin Options).

The pointer blocks contain a list of application images to be tried until one of them is successful. If none are successful, the SDM loads the factory image. To ensure reliability, the pointer block includes a main and a backup copy in case an update operation fails.

Both the factory image and the application images start with firmware. First, the decision firmware loads the firmware. Then, the decision firmware loads the rest of the image. These implementation details are not shown in the figure above. For more information, refer to the Application Image Layout section.

RSU Flash Layout – Your Perspective

The sub-partition table (SPT) manages the quad SPI flash. In devices that include the HPS, the SDM uses the sub-partition table to report its position in flash to the HPS.

The Intel^® Quartus^® Prime Programming File Generator creates the SPT when creating the initial manufacturing image. To ensure reliable operation, the data structure creates two copies, SPT0 and SPT1.

The SPT typically contains the following partitions:

Table 39. Typical Sub-Partitions
Sub-partition Name	Contents
BOOT_INFO	Decision firmware and decision firmware data
FACTORY_IMAGE	Factory Image
SPT0	Sub-partition table 0
SPT1	Sub-partition table 1
CPB0	Pointer block 0
CPB1	Pointer block 1
A1 (you enter)	Application image 1
A2 (you enter)	Application image 2

Figure 61. RSU Flash Layout - Your Perspective

To summarize, your view of flash memory is different from SDM view in two ways:

You do not need to know the addresses of the decision firmware, decision firmware data, and factory image. This information is read-only in the Intel^® Quartus^® Prime Pro Edition Software.
You have access to the sub-partition tables. The sub-partition tables provide access to the data structures required for remote system update.

Detailed Quad SPI Flash Layout

RSU Sub-Partitions Layout

The Flash Sub-Partitions Layout table shows the layout of RSU flash images.

Table 40. Flash Sub-Partitions Layout
Flash Offset	Size (in bytes)	Contents	Sub-Partition Name
0k	256k	Decision firmware	BOOT_INFO
256k	256k	Decision firmware
512k	256k	Decision firmware
768k	256k	Decision firmware
1M	8k + 24 pad	Decision firmware data
1M+32k	32k	Reserved for SDM
1M+64k	varies	Factory image	FACTORY_IMAGE
Next	4k + 28k pad	Sub-partition table (copy 0)	SPT0
Next	4k + 28k pad	Sub-partition table (copy 1)	SPT1
Next	4k + 28k pad	Pointer block (copy 0)	CPB0
Next	4k + 28k pad	Pointer block (copy 1)	CPB1
Next	varies	Application image 1	You assign
Next	varies	Application image 2	You assign

The Intel^® Quartus^® Prime Programming File Generator allows you to create many user partitions. When the HPS initiates the RSU, these partitions can contain application images and other items such as the Second Stage Boot Loader (SSBL), Linux* kernel, or Linux* root file system.

When you create the initial flash image, you can create up to three partitions for application images. There are no limitations on creating empty partitions.

Sub-Partition Table Layout

The following table shows the structure of the sub-partition table. The Intel^® Quartus^® Prime software supports up to 126 partitions. Each sub-partition is 32 bytes.

Table 41. Sub-partition Table Layout
Offset	Size (in bytes)	Description
`0x000`	4	Magic number 0x57713427
`0x004`	4	Version number (0 for this document)
`0x008`	4	Number of entries
`0x00C`	20	Reserved (write as 0)
`0x020`	32	Sub-partition Descriptor 1
`0x040`	32	Sub-partition Descriptor 2
`0xFE0`	32	Sub-partition Descriptor 126

Each 32-byte sub-partition descriptor contains the following information:

Table 42. Sub-partition Descriptor Layout
Offset	Size	Description
`0x00`	16	Sub-partition name, including a null string terminator
`0x10`	8	Sub-partition start offset
`0x18`	4	Sub-partition length
`0x1C`	4	Sub-partition flags

Pointer Block Layout

The pointer block contains a list of application images. The SDM tries the images in sequence until one of them is successful or all fail. The structure contains the following information:

Table 43. Pointer Block Layout
Offset	Size (in bytes)	Description
`0x00`	4	Magic number 0x57789609
`0x04`	4	Size of pointer block header (0x18 for this document)
`0x08`	4	Size of pointer block (4096 for this document)
`0x0C`	4	Offset from primary to backup pointer block (sector size)
`0x10`	4	Offset to image pointers (IPTAB)
`0x14`	4	Number of image pointer slots (`NSLOTS`)
`0x18`	—	Reserved
IPTAB	8	First (lowest priority) image pointer slot
	8	Second (2nd lowest priority) image pointer slot
	8	…
	8	Last (highest priority) image pointer

The pointer block can contain up to 508 application image pointers. The actual number is listed as NSLOTS. A typical pointer block update procedure consists of adding a new pointer and potentially clearing an older pointer. Typically, the pointer block update uses one additional entry. Consequently, you can make 508 remote system updates before the pointer block must be erased. The erase procedure is called pointer block compression. This procedure is safe. There are two copies of pointer block. The copies are in different flash erase sectors. While one copy is being updated the other copy is still valid.

Modifying the List of Application Images

The SDM uses the pointer block to determine priority of application images.

The pointer block must operate taking into account the following characteristics of quad SPI flash memory:

On a sector erase, all the sector flash bits become 1’s
A program operation can only turn 1’s into 0’s

The pointer block contains an array of values which have the following meaning:

All 1’s – the entry is unused. The SDM can write a pointer to this entry. This is the state after a quad SPI erase operation on the pointer block.
All 0’s – the entry was previously used and then canceled.
A combination of 1's and 0's – a valid pointer to an application image.

Initially when the SDM erases a pointer block, the SDM marks all entries as unused. To add an application image to the list, the SDM finds the first unused location and writes the application image address to this location. To remove an application image from the list, the SDM finds the application image address in the pointer block list and writes this address to 0s.

For designs that drive RSUs from FPGA logic, if the pointer runs out of space for new application images, your logic must must compress the pointer block by completing the following actions:

Erasing the pointer block
Copying all previously valid entries
Adding the new image

You can implement pointer block compression many different ways, including Nios^® II code, a scripting language, or a state machine. Pointer block compression does not occur frequently because the pointer block has up to 508 available entries.

There are two pointer blocks: a primary (CPB0) and a backup (CPB1). Two pointers enable the list of application images to be protected if a power failure occurs just after erasing a pointer block. For more information, refer to the Pointer Block Layout topic. When compressing, the client should compress (erase and rewrite) the primary CPB completely. Once the primary CPB is valid, it is safe to modify the secondary CPB. When rewriting, the magic number at the start of a CPB must be the last word written in the CPB. (after this number is written only image pointer slot values can be changed)

The only RSU component which updates the list of application images is LIBRSU. The rest of the components have read access to the pointer blocks.

When the client writes the application image to flash it must ensure that the pointers within the main image pointer of its first signature block update to point to the correct locations in flash. For more information, refer to the Application Image Layout topic.

Application Image Layout

The application image comprises decision firmware and the application image. The application image includes up to four sections. The decision firmware contains pointers to those sections. The table below shows the location of the number of sections and the section pointers in a application image.

By default the first 16 bytes of the application image, *.rpd, starting at address offset 0x1FC0 are 0. However you can use these 16 bytes to store a Version ID to identify your application image. Providing this Version ID allows you to verify the images stored in flash memory at a later time.

Table 44. Application Image Section Pointers
Offset	Size (in bytes)	Description
0x1F00	4	Number of sections
	…
0x1F08	8	Address of 1st section
0x1F10	8	Address of 2nd section
0x1F18	8	Address of 3rd section
0x1F20	8	Address of 4th section
	…
0x1FFC	4	CRC32 of 0x1000 to 0x1FFB

The section pointers must match the actual location of the FPGA image in flash. Two options are available to meet this requirement:

You can generate the application image to match the actual location in quad SPI flash memory. Because different systems may have a different sets of updates this option may not be practical.
You can generate the application image as if it is located at address zero, then update the pointers to match the actual location.

Here is the procedure to update the pointers from a application image created for INITIAL_ADDRESS to NEW_ADDRESS:

Create the application image, targeting the INITIAL_ADDRESS.
Read the 32-bit value from offset 0xF100 of the application image to determine the number of sections.
For <s>= 1 to number_of_sections:
1. section_pointer = read the 64-bit section pointer from 0xF100 + (s * 8)
2. Subtract INITIAL_ADDRESS from section_pointer
3. Add NEW_ADDRESS to section_pointer
4. Store updated section_pointer
Recompute the CRC32 for addresses 0x1000 to 0x1FFB. Store the new value at offset 0x1FFC. The CRC32 value must be computed on a copy of the data using the following procedure:
1. Swap the bits of each byte so that the bits occur in reverse order and compute the CRC.
2. Swap the bytes of the computed CRC32 value to appear in reverse order.
3. Swap the bits in each byte of the CRC32 value.
4. Write the CRC32 value to flash.

Note: The factory update image has a different format. You cannot relocate the factory image using the instructions above. You must provide the correct address to the Intel^® Quartus^® Prime Programming when programming the flash device.

Generating Remote System Update Image Files Using the Programming File Generator

Use the Intel^® Quartus^® Prime Programming File Generator tool to generate the Intel^® Stratix^® 10 remote system update flash programming files.

Generating the Initial RSU Image

Note: Program file generation is not available for Intel^® Agilex™ devices in the current release.

Follow these steps to generate the initial RSU image:

On the File menu, click Programming File Generator.
Select Intel^® Stratix^® 10 from the Device family drop-down list.
Select the configuration scheme from the Configuration scheme drop-down list. The current Intel^® Quartus^® Prime only supports remote system update feature in Active Serial x4.
On the Output Files tab, assign the output directory and file name.
Select the output file type.
Select the following file types for AS x4 configuration mode:
- JTAG Indirect Configuration File (.jic)/Programmer Object File (.pof)
- Memory Map File (.map)
- Raw Programming File (.rpd)
On the Input Files tab, click Add Bitstream, select the factory and application image .sof files and click Open.
On the Configuration Device tab, click Add Device, select your flash memory and click OK. The Programming File Generator tool automatically populates the flash partitions.
Select the FACTORY_IMAGE partition and click Edit.
In the Edit Partition dialog box, select your factory image .sof file in the Input file drop-down list and click OK.

Note: You must assign Page 0 to Factory Image. Intel^® recommends that you let the Intel^® Quartus^® Prime software assign the Start address of the FACTORY_IMAGE automatically by retaining the default value for Address Mode which is Auto. From the Address Mode drop down list, select Block to set an End address value for the FACTORY_IMAGE. The Programming File Generator reserves and assigns the start and end flash addresses to store BOOT_INFO, SPT0, SPT1, CPB0, and CPB1.
Select the flash memory and click Add Partition.
In the Add Partition dialog box, select for application image .sof file from the Input file drop-down list, assign the page number.
Repeat this step for additional application images and click OK. You can add up to three partitions for three application images. The page 1 application image is the highest priority, and the page 3 image is the lowest priority.
For .jic files,
Click Select at the Flash loader, select your device family and device name, and click OK.
Click Generate to generate the remote system update programming files. After generating the programming file, you can proceed to program the flash memory.

Note:
The generated .jic file contains only the initial flash data. If a remote host updates the initial flash image and then the application performs a verify operation, the verify operation fails.

You can use the programmer to examine the flash content and compare it to the new flash image .rpd.

Note: If you plan to update the factory image, Intel recommends reserving an additional 64 KB space for possible expansion of the factory image. Complete the following steps to reserve extra space for updates to the factory image:
1. Identify the new end address by adding 64 KB to the existing END ADDRESS of the FACTORY_IMAGE. The end address is available in the .map file. For example, if the current end address is 0x00423FF, the new end address is 0x00523FF .
2. Repeat the steps to regenerate the new .jic file. On the Configuration Device tab, select the FACTORY_IMAGE partition and click Edit. In the Edit Partition dialog box, under the Address Mode drop down list, select Block to set the new End address value for the FACTORY_IMAGE.
3. You can optionally Click File > Save As .. to save the configuration parameters as a file with the .pfg extension. The .pfg file contains your settings for the Programming File Generator. After you save the .pfg, you can the use this file to regenerate the programming file by running the following command:
```
quartus_pfg -c <configuration_file>.pfg
```
The .pfg file is actually an XML file which you can edit to replace absolute file paths with relative file paths. You cannot edit the.pfg file for other reasons. You can open and edit the .pfg in the Programming File Generator.

Generating an Application Image

Note: Program file generation is not available for Intel^® Agilex™ devices in the current release.

You can generate the RSU image from the command line directly, by running the quartus_pfg with the following arguments:

quartus_pfg -c fpga.sof application.rpd -o mode=ASX4 -o start_address=<address> -o bitswap=ON

Alternatively, you can use the Intel^® Quartus^® Prime Pro Edition Programming File Generator to generate the .rpd image by completing the following procedure:

On the File menu, click Programming File Generator.
Select Intel^® Stratix^® 10 from the Device family drop-down list.
Select the configuration mode from the Configuration mode drop-down list. The current Intel^® Quartus^® Prime only supports remote system update feature in Active Serial x4.
On the Output Files tab, assign the output directory and file name.
Select the output file type.
Select the following file types for AS x4 configuration mode:
- Raw Programming File (.rpd)
Click the Edit… button and assign the Start address for the image in flash memory. This Start address must match the starting address of the target partition in flash memory.

Figure 62. Specifying Parameters for an Application .rpd Stored in Flash Memory
By default, the .rpd file type is little-endian, if you are using a third-party programmer that does not support the little-endian format. Set the Bit swap to On to generate the .rpd file in big endian format.
On the Input Files tab, click Add Bitstream. Change the Files of type to SRAM Object File (*.sof). Then, select application image .sof file and click Open.

Figure 63. Specify the .sof File
Click Generate to generate the remote system update programming files. You can now program the flash memory. You can save the configuration in a .pfg file for later use.

Generating a Factory Update Image

Note: Program file generation is not available for Intel^® Agilex™ devices in the current release.

You can generate the factory update image from the command line directly, by running the quartus_pfg with the following arguments:

quartus_pfg -c fpga.sof factory_update.rpd -o mode=ASX4 -o start_address=<address> -o bitswap=ON -o rsu_upgrade=ON

Alternatively, you can use the Intel^® Quartus^® Prime Pro Edition Programming File Generator to generate a factory update image (.rpd). You can use this image to update the decision firmware, decision firmware data, and the factory image.

Note: Starting with the Stratix 10 device family the .rpd to program flash memory includes firmware pointer information for image addresses as shown in . You must use the Programming File Generator to generate the .rpd for flash devices.

On the File menu, click Programming File Generator.
Select Intel^® Stratix^® 10 from the Device family drop-down list.
Select the configuration mode from the Configuration mode drop-down list. The current Intel^® Quartus^® Prime only supports the RSU feature in the Active Serial x4 configuration mode.
On the Output Files tab, assign the Output directory and Name.
Select the .rpd output file type.
Click the Edit… button and assign the Start address for the update image in flash memory. This Start address should be the sector boundary of unused space in flash memory.

Note: If unused space is not available, you can use an application image space other than application image 1. In this case after the update operation completes you must restore the application image by writing the associated application image (.rpd) to the application slot.

Figure 64. Specifying Parameters for Single .rpd Stored in Flash Memory
By default, the .rpd file type is little-endian. If you are using a third-party programmer that does not support the little-endian format, set Bit swap to On to generate the .rpd file in big endian format.
On the Input Files tab, click Add Bitstream. Change the Files of type to SRAM Object File (*.sof). Then, select factory image .sof file and click Open.

Figure 65. Specify the .sof File
Select the .sof and then click Properties. Turn On Remote System Firmware Update.

Figure 66. Turn On Remote System Firmware Upgrade
Click Generate to generate the RSU programming files. You can now update the Intel^® Stratix^® 10 firmware. You can save the configuration in a .pfg file for later use.

Command Sequence To Perform Quad SPI Operations

Here is the recommended command sequence to access quad SPI flash memory or perform an RSU update operation.

Refer to Table 35 for more information about these commands.

Request exclusive access to the AS x4 interface: QSPI_OPEN.
Specify a quad SPI flash chip: QSPI_SET_CS*. This command is optional for the AS x4 configuration scheme and mandatory for other configuration schemes.
Perform the desired operation or operations. The following operations are available: QSPI_READ, QSPI_WRITE, QSPI_ERASE, QSPI_READ_DEVICE_REG, QSPI_WRITE_DEVICE_REG, QSPI_SEND_DEVICE_OP, and RSU_IMAGE_UPDATE.
Close exclusive access to the AS x4 interface: QSPI_CLOSE.

Remote System Update from FPGA Core Example

This section presents a complete remote system update example, including the following steps:

Creating the initial remote system update image (.jic) containing the bitstreams for the factory image and one application image.
Programming the flash memory with the initial remote system update image that subsequently configures the device.
Reconfiguring the device with an application or factory image.
Creating a single remote system update (.rpd) containing the bitstreams to add an application image in user mode.
Adding an application image.
Removing an application image.

Prerequisites

To run this remote system update example, your system must meet the following hardware and software requirements:

You should be running the Intel^® Quartus^® Prime Pro Edition software version 19.1 or later.
You should create and download this example to the Intel^® Stratix^® 10 SoC Development Kit.
Your design should include the Mailbox Client Intel^® Stratix^® 10 FPGA IP that connects to a JTAG to Avalon^® Master Bridge as shown the Platform Designer system. The JTAG to Avalon^® Master Bridge acts as the remote system update host controller for your factory and application images.
Figure 67. Required Communication and Host Components for the Remote System Update Design Example

Creating Initial Flash Image Containing Bitstreams for Factory Image and One Application Image

On the File menu, click Programming File Generator.
Select Intel^® Stratix^® 10 from the Device family drop-down list.
Select the configuration mode from the Configuration mode drop-down list. The current Intel^® Quartus^® Prime Software only supports remote system update feature in Active Serial x4.
On the Output Files tab, assign the output directory and file name.
Select the output file type.
Select the following file types for the Active Serial (AS) x4 configuration mode:
- JTAG Indirect Configuration File (.jic)
- Memory Map File (.map)
- Raw Programming File (.rpd). It is optional to generate the .rpd file.
On the Input Files tab, click Add Bitstream, select the factory and application image .sof files and click Open.
1. Bitstream_1 is the bitstream for factory image.
2. Bitstream_2 is the bitstream for application image.
On the Configuration Device tab, click Add Device, select MT25QU02G flash memory and click OK. The Programming File Generator tool automatically populates the flash partitions.
Select the FACTORY_IMAGE partition and click Edit.
On the Edit Partition dialog box, select Bitstream_1 as the factory image .sof in the Input file drop-down list. Keep the default settings for Page 0 and Address Mode. Click OK.
Select the MT25QU02G flash memory and click Add Partition.
In the Add Partition dialog box, select Bitstream_2 for the application image .sof in the Input file drop-down list. Assign Page: 1.Keep the default settings for Address Mode. Click OK.
For Flash loader click Select. Select Intel^® Stratix^® 10 from Device family list. Select 1SX280LU3S2 for the Device name. Click OK.
Click Generate to generate the remote system update programming files. The Programming File Generator generates the following files:
1. Initial_RSU_Image.jic
2. Initial_RSU_Image_jic.map

The following example output shows the generated .map file. The .map lists the start addresses of the factory image, CPB0, CPB1, and one application image. The remote system update requires these addresses.

BLOCK                         START ADDRESS   END ADDRESS
BOOT_INFO                     0x00000000      0x0010FFFF
FACTORY_IMAGE                 0x00110000      0x002D3FFF
SPT0                          0x002D4000      0x002DBFFF
SPT1                          0x002DC000      0x002E3FFF
CPB0                          0x002E4000      0x002EBFFF
CPB1                          0x002EC000      0x002F3FFF
Application Image             0x002F4000      0x004B7FFF


Configuration device: 1SX280LU3S2

Configuration mode: Active Serial x4
Quad-Serial configuration device dummy clock cycle: 15


Notes:

- Data checksum for this conversion is 0xBFFB90A5

- All the addresses in this file are byte addresses

After generating the programming file, you can program the flash memory.

Programming Flash Memory with the Initial Remote System Update Image

You can program the initial remote system update image from the command line, by running the following command:

quartus_pgm -c 1 -m jtag -o "pvi;./Initial_RSU_Image.jic

Alternatively, you can use the Intel^® Quartus^® Prime Programmer to program the initial RSU update image by completing the following procedure:

open the Programmer and click Add File. Select the generated .jic file (Initial_RSU_Image.jic) and click Open.
Turn on the Program/Configure for the attached .jic file.
To begin programming the flash memory with the initial remote system update image, click Start.
Configuration is complete when the progress bar reaches 100%. Power cycle the board to automatically configure the Intel^® Stratix^® 10 device with the application image using the AS x4 configuration scheme.

Note: This example does not assign the Direct to Factory Image pin. Consequently, the Programmer configures the device with the application image. The application image is the default image if the design does not use the Direct to Factory Image pin.
Use the RSU_STATUS command to determine which bitstream image the Programmer is using as shown in the following example:
1. In the Intel^® Quartus^® Prime software, select Tools > System Debugging Tools > System Console to launch the system console.
2. In the Tcl Console pane, type source rsu1.tcl to open the example of Tcl script to perform the remote system update commands. Refer to the Related Information for a link to rsu1.tcl.
3. Type the rsu_status command to report the current remote system update status. You can retrieve the current running image address from the remote system update status report. The current image address must match with the start address for the application image printed in the .map file.

Reconfiguring the Device with an Application or Factory Image

The following steps describe the process to reconfigure the device with a different application image or the factory image using operation commands after the device is in user mode.

The remote system update host sends the RSU_IMAGE_UPDATE command to perform the remote system update to the new application image or factory image.
1. For example, in the Tcl console of the System Console, type the following command to initiate a remote system update to the factory image.
  1. rsu_image_update 0x00110000
    Note: This command reconfigures the device with factory image. Address 0x00110000 is the start address of the factory image as shown in the .map file. The JTAG host automatically disconnects from the System Console once the device reconfiguration is successful. You must restart the System Console to re-establish the connection with the device to perform next command.
  2. rsu_image_update 0x002F4000
    Note: This command reconfigures the device with the application image. Address 0x002F4000 is the start address of the application image as shown in the .map file.
Optional: Retrieve the remote system update status by using the rsu_status command to ensure you have successfully reconfigured the device.
In the Tcl console of the System Console, type rsu_status to verify the current image. The following figure shows the device is being reconfigured with the factory image.

Adding an Application Image

Complete the following steps to add an application image to flash memory:

Set up exclusive access to the AS x4 interface and flash memory by running the QSPI_OPEN and QSPI_SET_CS commands in the Tcl Console window. You now have exclusive access to the AS x4 interface and flash until you relinquish access by running the QSPI_CLOSE command. Write the new application image to the flash memory using the QSPI_WRITE command.
Alternatively, the rsu1.tcl script includes the program_flash function that programs a new application image into flash memory. The following command accomplishes this task:
```
program_flash new_application_image.rpd 0x03FF0000 1024
```
The program_flash function takes three arguments:
1. The .rpd file to write to flash memory.
2. The start address.
3. Number of words to write for each QSPI_WRITE command. The QSPI_WRITE supports up to 1024 words per write instruction.
Write the new application image start address to a new image pointer slot in the configuration firmware pointer block (CPB) using the QSPI_WRITE command. Ensure that the new image pointer slot value is 0xFFFFFFFF before initiating the write.

Note: You must update both copies (CBP0 and CBP1) when editing the configuration firmware pointer block and sub-partition table. Refer to for more details about the configuration firmware pointer block.

Based on the example described above, the address offset 0x20 in the CPB0 and CPB1 must point to the start address of the application image. The next new image pointer slot value must be 0xFFFFFFFF before you write the start address of the new application image to the next image pointer slot.

Table 45. Configuration Firmware Point Block Contents
CPB Start Address + `0x20`	Content	Value
CPB0 + `0x20` = `0x002E4020`	Current application image pointer slot (highest priority)	`0x002F4000`
CPB0 + `0x28` = `0x002E4028`	Next image pointer slot	`0xFFFFFFFF`
CPB1 + `0x20` = `0x002EC020`	Current application image pointer slot (highest priority)	`0x002F4000`
CPB1 + `0x28` = `0x002EC028`	Next image pointer slot	`0xFFFFFFFF`

You can use the QSPI_read function verify that the new image pointer slot value is 0xFFFFFFFF . The QSPI_read function takes in two arguments:

Start address
Number of words to read

Figure 68. Verifying that the New Image Pointer Slot Value is 0xffffffff

You can now proceed to write the new application image address to next image slot by using the QSPI_write_one_word function. The QSPI_write_one_word function takes in two arguments:

Address
The value of the word

Figure 69. Writing an Address Pointer to the New Image Pointer Slot at 0x00234028

You can now do a QSPI_read function to the next image pointer slot to ensure that it is written with the start address of the desired new application image.

Verifying the Update to the New Image Pointer Slot at 0x00234028

Host software can now reconfigure the Intel^® Stratix^® 10 FPGA with the new application image by asserting the nCONFIG pin. Alternatively, you can power cycle the PCB. After reconfiguration, check the current image address. The expected address is 0x03ff0000. After adding a new image, your application image list includes the newly added application image and the old application image, which is now a secondary image. The newly added application image has the highest priority.

Note: When the remote system update host loads an application image, the decision firmware traverses the image pointer slots in reverse order. The new image has the highest priority when you restart the device.

Removing an Application Image

Set up exclusive access to the AS x4 interface and flash memory by running the QSPI_OPEN and QSPI_SET_CS commands in the Tcl Console window.. You now have exclusive access to the AS x4 interface and flash until you relinquish access by running the QSPI_CLOSE command. Write the new application image to the flash memory using the QSPI_WRITE command.
Write 0x00000000 to the application image start address stored in the image pointer slot of the configuration firmware pointer block (CPB0 and CPB1) using the QSPI_WRITE command.

Note: You must update both copies (copy0 and copy1) when editing the configuration firmware pointer block and sub-partition table.
Erase the application image content in the flash memory using the QSPI_ERASE command.

To remove a new application image, add another new application image in the next or subsequent image pointer slot or allow the device to fall back to the previous or secondary application image in your application image list. The following table shows correct entries for image pointer slots for CPB0 and CPB1 for offsets 0x20 and 0x28 :

CPB Start Address + `0x20`	Content	Value
CPB0 + `0x20` = `0x002E4020`	Old application image pointer slot (lower priority)	`0x002F4000`
CPB0 + `0x28` = `0x002E4028`	Current/new application image pointer slot (highest priority)	`0x03FF0000`
CPB1 + `0x20` = `0x002EC020`	Old application image pointer slot (lower priority)	`0x002F4000`
CPB1 + `0x28` = `0x002EC028`	Current/New application image pointer slot (highest priority)	`0x03FF0000`

You can now remove the current or new application image address image pointer slot by writing the value to 0x00000000 using the QSPI_write_one_word function as shown in the following example. The QSPI_write_one_word function takes address and data arguments. Be sure to erase the application content that you just removed from flash memory.

You can use a QSPI_read to the image pointer slot at offset 0x28 for CBP0 and CPB1 to verify completion of the QSPI_write_one_word commands .

You can now configure the device with the old application image. The old application image has the highest priority if you power cycle the device or the host asserts the nCONFIG pin. You can run the rsu_status report to check the status of the current image address, 0x002f4000.

Intel Stratix 10 Configuration Features

Device Security

Note: Contact your Intel^® sales representative for more information about the device security support in Intel^® Stratix^® 10 devices.

The Intel^® Stratix^® 10 device provides the following flexible and robust security features to protect sensitive data and intellectual property:

User image authentication and encryption
Public-Key based authentication
Advanced Encryption Standard (AES)-256 Encryption
JTAG Disable
JTAG Debug Disable/Enable
Side channel protection
Physical anti-tampering

Configuration via Protocol

The CvP configuration scheme creates separate images for the periphery and core logic. You can store the periphery image in a local configuration device and the core image in host memory, reducing system costs and increasing the security for the proprietary core image. CvP configures the FPGA fabric through the PCI Express* ( PCIe* ) link and is available for Endpoint variants only.

Figure 70. Intel^® Stratix^® 10 CvP Configuration Block Diagram

The CvP configuration scheme supports the following modes:

CvP Initialization Mode:
In this mode an external configuration device stores the periphery image and it loads into the FPGA through the Active Serial x4 (Fast mode) configuration scheme. The host memory stores the core image and it loads into the FPGA through the PCIe* link.

After the periphery image configuration completes, the CONF_DONE signal goes high and the FPGA starts PCIe* link training. When PCIe* link training completes, the PCIe* link transitions to the Link Training and Status State Machine (LTSSM) L0 state and then through PCIe* enumeration. The PCIe* host then configures the core through the PCIe* link. The PCIe* reference clock must be running for the link for link training.

After the core image configuration is complete, the CvP_CONFDONE pin (if enabled) goes high, indicating the FPGA has receiver the full configuration bitstream over the PCIe* link. INIT_DONE indicates that configuration is complete.
CvP Update Mode:
CvP update mode is a reconfiguration scheme that uses the PCIe* link to deliver an updated bitstream to a target device after the device enters user mode. The periphery images which includes the PCIe* link remains active, allowing CvP update to use this link to reconfigure the core fabric. In this mode, the FPGA device initializes by loading the full configuration image from the external local configuration device to the FPGA or after CvP initialization.

You can perform CvP update on a device that you originally configure using CvP initialization or any other configuration scheme.

Partial Reconfiguration

Partial reconfiguration (PR) allows you to reconfigure a portion of the FPGA dynamically, while the remaining FPGA design continues to function. You can define multiple personas for a region in your design, without impacting operation in areas outside this region. This methodology is effective in systems with multiple functions that time-share the same FPGA device resources. PR enables the implementation of more complex FPGA systems.

Intel Stratix 10 Debugging Guide

Configuration Debugging Checklist

Work through this checklist to identify issues that may result in operational failures.

Table 46. General Configuration Debugging Checklist
	Checklist Item	Complete?
1	Verify that the V_cc ,V_ccp ,V_{ccio_sdm} V_ccpt ,V_cceram, V_ccadc supplies are in the proper range by using SDM Debug Toolkit.	☐
2	Verify that all configuration resistors are correctly connected (`MSEL`, `nCONFIG`, `nSTATUS`, `CONF_DONE`, `INIT_DONE`).	☐
3	Verify that you are following the correct power-up and power-down sequences.	☐
4	Verify that the SDM I/Os assignments are correct by checking the Intel^® Quartus^® Prime Compilation QSF and Fitter reports.	☐
5	For SmartVID devices (-V), ensure that all `PMBus` pins are connected to Intel^® Stratix^® 10 device.	☐
6	Verify that SmartVID settings follow the recommendations in the Intel^® Stratix^® 10 Power Management User Guide	☐
7	Verify that the Intel^® Stratix^® 10 -V device has its own voltage regulator module for V_CC and V_CCP.	☐
8	After configuration are the `nCONFIG`, `nSTATUS` `CONF_DONE`, and `INIT_DONE` pins high? Use the SDM Debug Toolkit to determine these levels.	☐
9	Is the SDM operating Boot ROM code or configuration firmware? Use the SDM Debug Toolkit to answer this question.	☐
10	Are the `MSEL` pins correctly connected on board? Use the SDM Debug Toolkit to answer this question.	☐
11	For designs that use transceivers, HBM2, PCIe* , or EMIF, are the reference clocks stable and free running before configuration begins?	☐
12	Does your design include the Reset Release IP?	☐
13	To avoid configuration failures, disconnect the `PMBus` regulator’s JTAG download cable before configuring Intel^® Stratix^® 10 -V devices.	☐
14	If the SDM Debug Toolkit is not operational, verify that the Intel^® Stratix^® 10 device has exited POR by checking `nCONFIG`, `nSTATUS`, `CONF_DONE` and `INIT_DONE` pins using an oscilloscope.	☐
15	Is the configuration clock source chosen appropriately? You can use an internal oscillator or the `OSC_CLK_1` pin.	☐
16	For designs driving the `OSC_CLK_1` pin is the frequency 25, 100, or 125 MHz?	☐
17	For Intel^® Stratix^® 10 SX parts ensure that the HPS and EMIF IOPLL are stable and free running before configuration begins. The actual frequency should match the setting specified in Platform Designer.	☐
18	Are proper slave addresses set for the PMBus voltage regulator modules using the Intel^® Quartus^® Prime Software?	☐
19	For designs that use 3 V I/0, verify that the transceiver tiles are powered up before configuration begins.	☐

Intel Stratix 10 Configuration Architecture Overview

Intel^® Stratix^® 10 devices employ a new configuration architecture. The Secure Device Manager (SDM), a dedicated hard processor, controls and monitors all aspects of device configuration from device power-on reset. This configuration architecture differs from previous Intel FPGA device families where state machines control configuration.

There are important differences between Intel^® Stratix^® 10 and previous device families with respect to available configuration modes, configuration pin behavior, and connection guidelines. In addition, the bitstream format is different. Knowing about these differences and how these pins behave can help you understand and debug configuration issues.

SDM Debug Toolkit Overview

You launch the Intel^® Stratix^® 10 SDM Debug Toolkit from the System Console in the Intel^® Quartus^® Prime software, Tools > System Debugging Tools > System Console > Intel^® Stratix^® 10 Debug Toolkit.

The SDM Debug Toolkit provides access to current status of the Intel^® Stratix^® 10 device. To use these commands you must have a valid design loaded that includes the module that you intend to access The SDM Debug Toolkit includes the four tabs:

Configuration Status

The Read Configuration Status option provides the current value of MSEL, Configuration Pin Values, and Chip ID. The following information would be useful for debugging with the help of Intel: State, Error Location and Error Detail,

Voltage Sensor

The Read option in the External Channel window reads the voltage on available channels. The Read option in the Internal Power Supplies window provides the values of internal power supplies.
Figure 71. Read External Channel and Read Internal Power Supplies

For more information about using the voltage sensor refer to Intel^® Stratix^® 10 Voltage Sensor in the Intel^® Stratix^® 10 Analog to Digital Converter Uses Guide.

Temperature Sensor

Read option reads the temperature in Celsius of the Intel^® Stratix^® 10 device, the HSSI channels, and the UIB_TOP and UIB_BOTTOM. The universal interface bus (UIB) blocks are general-purpose SiP interfaces for HBM2.
HPS Reset Control: Provides the following two options to reset the HPS: Release HPS from Reset option and HPS Cold Reset.
Figure 72. HPS Reset Options

HPS Reset Control

Provides the following two options to reset the HPS: Release HPS from Reset option and HPS Cold Reset.

Figure 73. HPS Reset Options

Using the SDM Debug Toolkit

To use the Intel^® Stratix^® 10 SDM Debug Toolkit, bring up the System Console in the same version of the Intel^® Quartus^® Prime software that you used to configure the Intel^® Stratix^® 10 device.

Complete the following steps to become familiar with the Intel^® Stratix^® 10 SDM Debug Toolkit:

In a Nios^® II command shell, type the following command:
```
% system-console
```
Under Intel^® Stratix^® 10 SDM Debug Toolkit click Launch.
Verify that the Intel^® FPGA Download Cable II connects to the PC and the 10-pin JTAG header connects to the Intel^® Stratix^® 10 device.
In the Intel^® Stratix^® 10 SDM Debug Toolkit, click Refresh Connections. The Refresh Connections is above the tabs in the SDM Debug Toolkit GUI.
On the Configuration Status tab, click Read Configuration Status. The following figure shows the Configuration Status after a successful read.

Figure 74. Read Configuration Status Command

Configuration Pin Differences from Previous Device Families

Intel^® Stratix^® 10 configuration pin behavior is different from earlier device families. Knowing about these differences and how these pins behave can help you understand and debug configuration issues.

Configuration Pin Names (Pre- Intel^® Stratix^® 10)	Intel^® Stratix^® 10 Pin Names	Notes
`TRST`	Not Available	Use the `TMS` reset sequence. Hold `TMS` high for 5 `TCK` cycles.
`CLKUSR`	`OSC_CLK_1`	An external source you can supply to increase the configuration throughput to 250 MHz. Using an external clock source Transceivers, the HPS, PCIe* , and the High Bandwidth Memory (HBM2) require you this external clock. 25 100 125 Refer to Setting Configuration Clock Source for instructions on setting the clock source and frequency in the Intel^® Quartus^® Prime Pro Edition software.
CRC_ERROR	Any unused `SDM_IO` (`SEU_ERROR`)	No dedicated location. Now called `SEU_ERROR`. Ignore until after `CONF_DONE` asserts.
`CONF_DONE`	`SDM_IO5`, `SDM_IO16` (`CONF_DONE`)	No single dedicated pin location. No longer Open Drain. External pull-up Is not mandatory.
`DCLK` (PS - FPP)	`AVST_CLK`, `AVSTx8_CLK`	x8 mode has a dedicated clock input on `SDM_IO14` (`AVSTx8_CLK`). For other Avalon^® -ST modes, use AVST_CLK. AVST_CLK and AVSTx8_CLK must be continuous and cannot pause during configuration.
`DCLK` (AS)	`SDM_IO2 (AS_CLK)`	When using the internal oscillator in AS mode, the `AS_CLK` runs in the range of 57 - 133 based on `AS_CLK` selection. If you provide a 25 MHz, 100 MHz or 125 MHz clock to the `OSC_CLK_1` pin, the `AS_CLK` can run up to 133 MHz.
`DEV_OE`	Not Available
`DEV_CLRn`	Not Available
`INIT_DONE`	`SDM_IO0` `SDM_IO16` `INIT_DONE`	No longer Open Drain.
`MSEL[0]`	`SDM_IO5` (`MSEL[0]`)	After the SDM samples `MSEL` this pin functions as per the configuration mode selected. Do not connect directly to power. Use 4.7 KΩ pull-up or pull-downs, as appropriate.
`MSEL[1]`	`SDM_IO7` (`MSEL[1]`)	After the SDM samples `MSEL`, this pin functions as per the configuration mode selected. Do not connect directly to power. Use 4.7 KΩ pull-up or pull-downs, as appropriate.
`MSEL[2]`	`SDM_IO9` (`MSEL[2]`)	After the SDM samples `MSEL`, this pin functions as per the configuration mode selected. Do not connect directly to power. Use 4.7 KΩ pull-up or pull-downs, as appropriate.
`NSTATUS`	`nSTATUS`	No longer Open Drain. Intel recommends a 10 KΩ pull-up to V_{CCIO_SDM}.
`NCE`	Not Available	Multi-device configuration is not supported.
`NCEO`	Not Available	Multi-device configuration is not supported.
`DATA[31:0]` (PP32/PP16)	`AVST_DATA[31:0]`	Avalon^® -ST x8 uses SDM pins for data pins.
`DATA[7:0]` (PP8)	`SDM _IO` pins (`AVSTx8_DATAn`)
`nCSO[2:0]`	`SDMIO_8` (`AS_nCSO3`) `SDMI_O7` (`AS_nCSO2`)`SDMI_O9` (`AS_nCSO1`) `SDM_IO5` (`AS_nCSO0`)	Intel^® Stratix^® 10 supports up to 4 cascaded AS devices
`nIO_PULLUP`	Not Available	Use a JTAG instruction to invoke.
`AS_DATA0_ASDO`	`SDM_IO4` (`AS_DATA0`)
`AS_DATA[3:1]`	`SDM_IO6` (`AS_DATA3`) `SDM_IO3` (`AS_DATA2`) `SDM_IO1` (`AS_DATA1`)	Unlike earlier device families, the AS interface does not automatically tristate at power-on. When you set `MSEL` to JTAG, the SDM drives the `AS_CLK`, `AS_DATA0`-`AS_DATA3`, and `AS_nCS0`-`AS_nCS3`, `MSEL` pins until POR.
`PR_REQUEST`	`GPIO*`	No dedicated location.
`PR_READY`	`GPIO*`	No dedicated location.
`PR_ERROR`	`GPIO*`	No dedicated location.
`PR_DONE`	`GPIO*`	No dedicated location.
`CVP_CONFDONE`	Any unused `SDM_IO CVP_CONFDONE`

Configuration File Format Differences

Detailed information about the configuration file format is proprietary. This topic explains the general structure and differences from previous device families.

The configuration file format differs significantly from previous device families. The configuration bitstream begins with a SDM firmware section. The SDM loads the boot ROM firmware during power-on reset. Design sections for I/O configuration, HPS boot code (if applicable), and fabric configuration follow the firmware section. Configuration begins after the SDM boot ROM performs device consistency checks.

Figure 75. Example of an Intel^® Stratix^® 10 Configuration Bitstream Structure

The firmware section is not part of the .sof file. The Intel^® Quartus^® Prime Pro Edition Programmer adds the firmware to the .sof. The programmer adds the firmware when configuring an Intel^® Stratix^® 10 device or when it converts the .sof to another format.

Understanding SEUs

SEUs are rare, unintended changes in the state of an FPGA's internal memory elements caused by cosmic radiation effects. The change in state is a soft error and the FPGA incurs no permanent damage. Because of the unintended memory state, the FPGA may operate erroneously until background scrubbing fixes the upset.

The Intel^® Quartus^® Prime software offers several features to detect and correct the effects of SEU, or soft errors, and to characterize the effects of SEU on your designs. LSM firmware provides SEU single bit error correction and multi-bit error detection per LSM. Additionally, some Intel FPGAs contain dedicated circuitry to help detect and correct errors.

For more information about SEUs, refer to Intel^® Stratix^® 10 SEU Mitigation User Guide.

Reading the Unique 64-Bit CHIP ID

The Chip ID Intel IP core in each Intel^® Stratix^® 10 device stores a unique 64-bit chip ID. After the Chip ID Intel IP core receives a valid clock input, the chip ID is available on the chip_id[63:0] output port. You can read the chip ID using the JTAG interface. The chip ID may be useful for debugging. For more information about the chip ID refer to the Chip ID Intel FPGA IP Cores User Guide .

Understanding and Troubleshooting Configuration Pin Behavior

Configuration typically fails for one of the following reasons:

The host times outs
A configuration data error occurs
An external event interrupts configuration
An internal error occurs

Here are some very common causes of configuration failures:

Check OSC_CLK_1 frequency. It must match the frequency you specified in the Intel^® Quartus^® Prime Software and the clock source on your board.
Ensure a free running reference clock is present for designs using transceivers, PCIe* , or HBM2. These reference clocks must be available until the device enters user mode.
For designs using the HPS and the external memory interface (EMIF), ensure that the EMIF clock is present.
For designs using SmartVID (-V devices), ensure that this feature is set-up and operating correctly. Ensure that the voltage regulator supports SmartVID.

Here are some debugging suggestions that apply to any configuration mode:

To rule out issues with OSC_CLK_1 select the Internal Oscillator option in the Intel^® Quartus^® Prime.
Try configuring the Intel^® Stratix^® 10 device with a simple design that does not contain any IP. If configuration via a non-JTAG scheme fails with a simple design, try JTAG configuration with the MSEL pins set specifically to JTAG.

The following topics describe the expected behavior of configuration pins. In addition, these topics provide some suggestions to assist in debugging configuration failures. Refer to the separate sections on each configuration scheme for debugging suggestions that pertain to a specific configuration scheme.

nCONFIG

The nCONFIG pin is a dedicated, input pin in the SDM. nCONFIG has two functions:

Hold-off initial configuration
Initiate FPGA reconfiguration

The nCONFIG pin transition from low to high signals a configuration or reconfiguration request. The nSTATUS pin indicates device readiness to initiate FPGA configuration.

The configuration source can only change the state of the nCONFIG pin when it has the same value as nSTATUS. When the Intel^® Stratix^® 10 device is ready it drives nSTATUS to follow nCONFIG.

The host should assert nCONFIG to clear the device. Then the host should deassert nCONFIG initiate configuration. If nCONFIG asserts during a configuration cycle, that configuration cycle stops. The SDM expects a new configuration cycle to begin.

Debugging Suggestions

The host drives nCONFIG. Be sure that it is not floating or stuck low. nCONFIG should remain high during configuration.

nSTATUS

nSTATUS has the following two functions:

To behave as an acknowledge for nCONFIG.
To behave as an error status signal. It is important to monitor nSTATUS to identify configuration failures.

Note: nSTATUS does not go low for PR failures.

Generally, the Intel^® Stratix^® 10 device changes the value of nSTATUS to follow the value of nCONFIG, except after an error. For example, after POR, nSTATUS asserts after nCONFIG asserts. When the host drives nCONFIG high, the Intel^® Stratix^® 10 device drives nSTATUS high.

In previous device families the deassertion of nSTATUS indicates the device is ready for configuration. For Intel^® Stratix^® 10 devices, when using Avalon^® -ST configuration scheme, after the Intel^® Stratix^® 10 device drives nSTATUS high, you must also monitor the AVST_READY signal to determine when the device is ready to accept configuration data.

nSTATUS asserts if an error occurs during configuration. The pulse ranges from .5 ms to 1.5 ms.

nSTATUS assertion is asynchronous to data error detection. Intel^® Stratix^® 10 devices do not support the auto-restart configuration after error option.

Previous device families implement the nSTATUS as an open drain with a weak internal pull-up. Consequently, you cannot wire OR an Intel^® Stratix^® 10 nSTATUS signal with the nSTATUS signal from earlier device families.

Debugging Suggestions

Ensure nSTATUS acknowledges nCONFIG. If nSTATUS is not following nCONFIG, the FPGA may not have exited POR. You may need to power cycle the PCB.

CONF_DONE and INIT_DONE

For Intel^® Stratix^® 10 devices, both CONF_DONE and INIT_DONE share multiplexed SDM_IO pins. Previous device families implement the CONF_DONE and INIT_DONE pins as open drains with a weak internal pull-up. Consequently, you cannot wire OR an Intel^® Stratix^® 10 CONF_DONE or INIT_DONE signal with the nSTATUS signal from previous device families. Otherwise, CONF_DONE and INIT_DONE behave as these signals behaved in earlier device families. If you assign CONF_DONE and INIT_DONE to SDM_IO16 and SDM_IO0, weak internal pull-downs pull these pins low at power-on reset. Ensure you specify these pins in the Intel^® Quartus^® Prime Software or in the Intel^® Quartus^® Prime settings file, (.qsf). CONF_DONE and INIT_DONE are low prior to and during configuration. CONF_DONE asserts when the device finishes receiving configuration data. INIT_DONE asserts when the device enters user mode.

Note: The entire device does not enter user mode simultaneously.Intel recommends that you include the Intel Stratix 10 Reset Release IP to hold your application logic in the reset state until the entire FPGA fabric is in user mode.

CONF_DONE and INIT_DONE are optional signals. You can use these pins for other functions that the Intel^® Quartus^® Prime Pro Edition Device and Pin Options menu defines.

Debugging Suggestions

Place the CONF_DONE and INIT_DONE pins on the SDM_IO pins that correlate with the board-level connection. Refer to SDM Pin Mapping and Setting Additional Configuration Pins for more information.

SDM_IO Pins

Intel^® Stratix^® 10 devices include 17 SDM_IO pins that you can configure to implement specific functions such as CONF_DONE and INIT_DONE. The chosen function must follow the GX, MX, TX, and SX Device Family Pin Connections Guidelines . The configuration bitstream controls the pin locations for the SDM_IO pins.

Internal Intel^® Stratix^® 10 circuitry pulls SDM_IO0, SDM_IO8 and SDM_IO16 weakly low through a 25 kΩ resistor. Internal Intel^® Stratix^® 10 circuitry pulls all SDM_IO pins weakly high during power-on.

Debugging Suggestions

Check the Intel^® Quartus^® Prime Pro Edition settings and Fitter report to ensure that the SDM_IO configuration matches your PCB design. The following screen shots show where to configure these signals and how to confirm the SDM_IO pin settings in the Fitter report.

Figure 76. Configuration Pin Selection in the Intel^® Quartus^® Prime Pro Edition Software

Figure 77. Fitter Report and SDM_IO Pin Reporting

Starting with the Intel^® Quartus^® Prime Pro Edition Software, version 18.1, an SDM Debug Toolkit is available through the System Console, Tools > System Debugging Tools > System Console > Intel^® Stratix^® 10 SDM Debug Toolkit.

Intel Stratix 10 Configuration User Guide Archives

If an IP core version is not listed, the user guide for the previous IP core version applies.

IP Core Version	User Guide
19.1	Intel^® Stratix^® 10 Configuration User Guide
18.1	Intel^® Stratix^® 10 Configuration User Guide
18.0	Intel^® Stratix^® 10 Configuration User Guide
17.1	Intel^® Stratix^® 10 Configuration User Guide

Document Revision History for the Intel Stratix 10 Configuration User Guide

Document Version	Intel^® Quartus^® Prime Version	Changes
2019.07.19	19.2	Made the following changes: Corrected numbers on Configuration, Reconfiguration, and Error Timing Diagram timing diagram. The number 3 now labels the `nCONFIG` rising edge. Renumbered associated text under the Initial Configuration Timing heading. In the Additional Clock Requirements for HPS, PCIe* , eSRAM, and HBM2 topic, removed the following items from the list of components requiring a free-running clock before configuration begins: EMIF E-Tile transceiver Note: HPS EMIF retains this requirement.
2019.07.08	19.2	Made the following changes: Revised and reorganized all topics covering configuration pin assignments: Clarified the behavior of the `MSEL` pins in AS x4 mode. Added information about the SDM_IO pin states during power-on and after device cleaning to the Intel^® Stratix^® 10 Configuration Pins topic. Created separate topics covering partial configuration and SmartVID signals. Made the following changes to the RSU chapter: Added the following topics: RSU Glossary Standard (non-RSU) Flash Layout RSU Flash Layout – SDM Perspective RSU Flash Layout – Your Perspective Detailed Quad SPI Flash Layout Sub-partitions Layout Sub-Partition Table Layout Pointer Block Layout Modifying the List of Application Images Application Image Layout The static firmware has been replaced by decision firmware. The update image now includes the factory image, the decision firmware and the decision firmware data. The `QSPI_ERASE` command is now 4 KB aligned. The number of words to erase must be a multiple of 1024. Added definitions of major and minor error codes for `RSU_STATUS` and `CONFIG_STATUS`. Added footnote explaining that before you can use CvP you must configure either the periphery image or the full image via the AS configuration scheme. Then, you can configure the core image using CvP. Added recommendation to use the Analog Devices LTM4677 device to regulate the PMBus for SmartVID devices. You set this parameter here: Device > Device and Pin Options > Power Management & VID > Slave device type. Added two restrictions to the dual-purpose use of Avalon^® -ST pins. For more information, refer to the Enabling Dual-Purpose Pins topic. Corrected maximum speed and data rate in the Intel^® Stratix^® 10 Configuration Data Width, Clock Rates, and Data Rates table. The Max Clock Rate is 33 MHz. The Max Data Rate is 33 Mb. Updated the Intel^® Stratix^® 10 Reset Release IP to include reference to the new An 891: Using the Reset Release FPGA IP. Removed recommendation to gate Intel^® Hyperflex™ registers using the `nINIT_DONE` signal. Added the eSRAM clocks to the list of free-running clocks that must be stable before configuration begins. Added Attention warning that designs including the Intel^® Stratix^® 10 Mailbox Client FPGA IP using Intel^® Quartus^® Prime Programmer 19.2 or later whose .sof was generated in Intel^® Quartus^® Prime Programmer 19.1 or earlier must be regenerate the .sof. Added 10 kΩ pull-up resistor to `nCONFIG` in the following figures: Connections for Avalon^® -ST x8 Single-Device Configuration Connections for Avalon^® -ST x16 Single-Device Configuration Connections for Avalon^® -ST x32 Single-Device Configuration PFL II IP core with Dual CFI Flash Memory Devices Removed discrete synchronizers for the `AVST_READY` signal in the following figures: Connections for Avalon^® -ST x8 Single-Device Configuration Connections for Avalon^® -ST x16 Single-Device Configuration Connections for Avalon^® -ST x32 Single-Device Configuration If necessary, you can implement synchronizers in the host controller if the host is an FPGA or CPLD. Validation has shown that external synchronizers are not required. Made a global change recommending that you use the newer Intel^® Quartus^® Prime Programming File Generator instead of the legacy Intel^® Quartus^® Prime Convert Programming Files conversion program to generate programming files. Changed all file conversion topics to use the Programming File Generator. Revised all topics providing steps for file conversion to use Programming File Generator instead of the legacy Convert Programming Files dialog box. Clarified statement on quad SPI flash byte addressing: The SDM configures the Quad SPI flash device to operate using 4-byte addressing if the flash size in 256 MB or greater. Corrected flash memory sizes in Understanding Quad SPI Flash Byte-Addressing topic. All sizes in in megabits or gigabits, not megabytes or gigabytes. Generalized Figure 2. Intel^® Stratix^® 10 Configuration Architecture Block Diagram. This figure no longer lists specific variants of Intel^® Stratix^® 10 device. Corrected Step 3 in the Initial Configuration Timing description. The step should say, with `nConfig` low, the SDM enters Idle mode after booting. Corrected the Intel^® Stratix^® 10 FPGA Configuration Flow diagram. The transition between `FPGA Config*` and `User Mode` should say `INIT_DONE = HIGH`. Corrected the following statement in the Debugging Guidelines for the JTAG Configuration Scheme topic: An `nSTATUS` falling edge terminates any JTAG access and the device reverts to the `MSEL`-specified boot source. `nSTATUS` must be stable during JTAG configuration.. In both sentence, `nSTATUS` should be `nCONFIG`. Removed pin assignments for `CVP_CONFDONE` for the Avalon^® -ST in the Available SDM I/O Pin Assignments for Configuration Signals that Do Not Use Dedicated SDM I/O Pins table. CvP does not the support Avalon^® -ST x8 configuration scheme in Intel^® Stratix^® 10 devices.
2019.04.10	19.1	Updated the transceiver reference clocks.
2019.04.01	19.1	Made the following additions and enhancements: Added Intel^® Stratix^® 10 Reset Release IP. Use the `nINIT_DONE` output of this IP to hold your application logic in reset until the entire FPGA fabric enters user mode. Added Configuration Scheme Components and File Types topics illustrating the software flow and programming file outputs for the AS, Avalon^® -ST, and JTAG programming schemes. Added the following topics to the Stratix 10 Configuration Debugging Guide chapter: Debugging Checklist SDM Debug Toolkit Overview Using the SDM Debug Toolkit Reading the Unique 64-Bit CHIP ID Understanding SEUs Added Maximum Allowable External AS_DATA Pin Skew Delay Guidelines topic. Added Generating an Update Image for Static Firmware and Factory Image topic. Added topics covering SDM_IO pin assignments and QSF settings for the Avalon^® -ST x8, x16, x32, and AS configuration schemes. Added note in the Avalon^® -ST Configuration Timing topic. This note covers special requirements for driving configuration data Avalon^® -ST x16 and x32 configurations. Added the following signals to the Intel^® Stratix^® 10 Configuration Timing Diagram: `nINIT_DONE`, `Data<n>-1:0]`, `AVST_READY`, `AVST_VALID`, and `AS_CS0`. Added a 10K Ω pull-up resistor to `nCONFIG` and corrected file type for flash image in the following figures: Connections for AS x4 Single-Device Configuration Connections for AS Configuration with Multiple Serial Flash Devices Connections for Programming the Serial Flash Devices using the JTAG Interface Added the IBIS model name to the Intel^® Stratix^® 10 Configuration Pins I/O Standard and Drive Strength table. Renamed this table Configuration Pins I/O Standard, Drive Strength, and IBIS Model. Added Updating the SDM Firmware topic in the Intel^® Stratix^® 10 Configuration Overview chapter. Added the following guidance in Debugging Guidelines for the JTAG Configuration Scheme topic: When you use the JTAG interface for reconfiguration after an initial reconfiguration using AS or the Avalon^® -ST interface, the .sof must be in the file format you specified in the Intel^® Quartus^® Prime project. Added the following signals to the list of device configuration pins that do not have fixed assignments: `CONF_DONE` `INIT_DONE` `HPS_COLD_nRESET` Improved definitions of programming file output types. Edited Using the PFL II IP Core for clarity and style. Added many screenshots illustrating the step to complete a task. Updated the supported flash memory devices and supported SD* card types in the Intel^® Stratix^® 10 Configuration Overview topic. Updated the SDM Pin Mapping table to include the following: Avalon^® -ST x16, and x32 configuration scheme Pins for SmartVID Added definition of the `GETDESIGN_HASH` command to the Mailbox Client Intel^® Stratix^® 10 FPGA IP Command List and Description table. Renamed the topic title Commands and Error Codes to Commands and Responses. Updated the descriptions for `Length` and `Command Code/Error Code` in the Mailbox Client Intel^® Stratix^® 10 FPGA IP Command and Response Header Description table. Added PLL reference clock requirement to the Additional Clock Requirements for Transceivers, HPS, PCIe* , High Bandwidth Memory (HBM2) and SmartVID topic. Updated the Generating a Single RSU Image topic to clarify that the .rpd for Intel^® Stratix^® 10 devices now includes firmware pointer information for image addresses and is not compatible with earlier generation methods. Removed a note in Remote System Upgrade Configuration Images topic, saying that the application image is optional and can be added later. The initial RSU setup requires both a factory image and an application image. Added the following note to the Configuration Firmware Pointer Block (CPB) topic: Note: Application images must align to partition boundaries in the flash device. If an application image is less than a full partition, the rest of the sector cannot be used. Added a new topic RSU Recovery from Corrupted Images that explains how the SDM recovers from attempts to load corrupted images. Added 71.5 MHz as a supported frequency for the `OSC_CLK_1` for AS configuration. Added description of optional 16-byte .rpd Version ID to the Remote System Upgrade Flash Device Layout topic. Removed Supported Flash Devices appendix. This appendix has been replace by the following web page Supported Flash Devices for Intel^® Stratix^® 10 Devices which provides more information about flash devices for different purposes. Removed references to P30 and P33 flash memory devices. These CFI flash devices are no longer available. Made the following corrections: Corrected the following statement: Because Intel^® Stratix^® 10 devices operate at 1.8 volt and all SD MMC I/Os operate between 2.7 - 3.6 volts, an intermediate voltage level translator is necessary for SD* cards. This statement is only true for SD* cards. Corrected the value of `MSEL` for Avalon^® -ST x16 configuration in Table 1 and Table 9. The correct value is 101. Corrected PFL II IP core with Dual P30 or P33 CFI Flash Memory Devices figure. The `nCONFIG` signal should not have a pulldown resistor. Removed the statement that Remote system upgrade cannot use partial reconfiguration (PR) images for the application image from the Remote System Upgrade Using AS Configuration topic. Remote system upgrade does support PR. In the Mailbox Client Intel^® Stratix^® 10FPGA IP Command List and Description table, for `CONFIG_STATUS`, corrected the size of `MSEL`. `MSEL` is 3 bits. Corrected the end address in step 14a of Generating a Standard RSU Image. It should be `0x00523FF.` Corrected the definition of `QSPI_ERASE`. The number of words to erase must be a multiple of 4000 (hexadecimal) words. Changed Number of Commands and Number of Responses to Command Length and Response Length in the Mailbox Client Intel^® Stratix^® 10 FPGA IP Command List and Descriptions (RSU Functions for Non-HPS Variants) table. Corrected the fields of the `RSU_STATUS` command. The `Last failing image` field should be called the `First failing image`. This field reports the flash offset of the first failing application image. In the Remote System Upgrade Flash Memory Layout table, changed the amount of reserved flash memory image from 64k to 256k.
2018.11.02	18.1	Updated Figure 39: Intel^® Stratix^® 10 Modules and Interfaces to Implement RSU Using Images Stored in Flash Memory to exclude SD and MMC memory. These memory types are not supported in the current release.
2018.10.23	18.1	Added the following statement to the description of Avalon^® -ST Configuration Timing topic: The `AVST_READY` signal is only valid when the `nSTATUS` pin is high.
2018.10.10	18.1	Made the following changes: Changed the number of remote system upgrade images supported from more than 500 to 507 in Remote System Upgrade Configuration Images. Updated the last two entries in the Configuration Firmware Pointer Block Format table.
2018.10.04	18.1	Made the following changes: Corrected statement in the Remote System Upgrade topic. A command to the Mailbox Client Intel^® Stratix^® 10 FPGA Mailbox Client IP Core initiates reconfiguration. Corrected the Intel Stratix 10 Remote System Upgrade Components figure and Related Information link. The mailbox component is the Mailbox Client Intel^® Stratix^® 10 FPGA IP Core.
2018.09.21	18.1	Made the following changes: Added new chapter, Remote System Upgrade Added new chapter, Intel^® Stratix^® 10 Debugging Guide Added separate Debugging Guidelines topics in the Avalon^® -ST, AS, and JTAG configuration scheme sections. Significantly expanded Stratix 10 Configuration Overview Configuration Overview chapter. Added Additional Clock and SmartVID Requirements for Transceivers, HPS, PCIe* , High Bandwidth Memory (HBM2) and SmartVID topic. Expanded OSC_CLK_1 Clock Input topic to include additional usage requirements. Added AS Using Multiple Serial Flash Devices topic. Added numerous screenshots illustrating Intel^® Quartus^® Prime Pro Edition procedures. Improved many figures illustrating configuration schemes. Added the fact that you must have system administrator privileges to define a new flash device in the Defining New CFI Flash Memory Device topic. Added MT28EW to the list of PFL II flash devices supported. Moved almost all of the material describing the PFL II flash from an appendix to the Intel^® Stratix^® 10 Configuration Schemes chapter. Edited entire document for clarity and style. Corrected minor errors and typos.
2018.05.07	18.0	Removed Estimating the Active Serial Configuration Time section. Updated the `OSC_CLK_1` supported frequency. Added selecting flash loader step to Generating Programming Files using Convert Programming Files. Added a note to `TCK`, `TDI`, `TMS`, and `TDO` stating that they are available for HPS JTAG chaining in SoC devices. Removed instruction to drive nCONFIG low from POR in the following diagrams: Connections for AS x4 Single-Device Configuration Connection Setup for AS Configuration with Multiple EPCQ-L Devices Connection Setup for Programming the EPCQ-L Devices using the JTAG Interface Added a note in OSC_CLK_1 Clock Input stating that reference clocks to EMIF and PCIe IP cores must be stable and free running. Removed .ekp file from Overview of Intel^® Quartus^® Prime Supported Files and Tools for Configuration and Programming figure. Updated the Configuring Intel^® Stratix^® 10 Devices using AS Configuration section title to Generating and Programming AS Configuration Programming Files. Updated Configuration Schemes and Features Overview in Intel Stratix 10 Devices table: Added a note stating to contact sales representative for more information about support readiness. Added a note stating to contact sales representative for more information about flash support other than EPCQ-L devices. Removed NAND configuration support. Updated Configuration Sequence in Intel Stratix 10 Devices figure by adding a looped flow arrow during Idle state. Updated the MSEL note in Intel Stratix 10 Device Configuration Pins table. Added a note to recommend `OSC_CLK_1` for configuration clock source in OSC_CLK_1 Clock Input. Updated CvP data width and maximum data rate in Configuration Schemes and Features Overview in Intel^® Stratix^® 10 Devices table. Removed the multiple EPCQ-L configuration device support.

Date	Version	Changes
November 2017	2017.11.09	Removed link to the Configuration via Protocol (CvP) Implementation User Guide. Updated titles for Device Security, Partial Reconfiguration, and Configuration via Protocol.
November 2017	2017.11.06	Updated Option Bits Sector Format table. Updated a step in Setting Additional Configuration Pins. Added Converting .sof to .pof File and Programming CPLDs and Flash Memory Devices. Updated the .pof version value in Storing Option Bits. Added information about restoring start and end address for option bits in Restoring Option Bit Start and End Address. Added note about pull-down resistor is recommended for `CONF_DONE` and `INIT_DONE` pins in Additional Configuration Pin Functions. Added new subsection Multiple EPCQ-L Devices Support. Added Configuration Pins I/O Standard and Drive Strength table. Updated information about maximum additional data words when using 2-stage register synchronizer. Updated the equation for minimum AS configuration time estimation. Added RBF Configuration File Format section explaining the format of the .rbf file. Updated Configuration Sequence to state that a firmware which is part of the configuration data if loaded in the device initially. Updated description for Number of flash devices used parameter in the PFL II Flash Interface Setting Parameters table. Updated Configuration via Protocol overview and added link to the Configuration via Protocol (CvP) Implementation User Guide. Updated Partial Reconfiguration overview and added link to the Creating a Partial Reconfiguration Design chapter of the Handbook Volume 1: Design and Compilation. Updated Design Security Overview descriptions. Added note for Partial Reconfiguration feature and link to Partial Reconfiguration Solutions IP User Guide in Intel^® Stratix^® 10 Configuration Overview. Removed SDM pin notes in Intel^® Stratix^® 10 Configuration Overview. Updated internal oscillator's `AS_CLK` frequency in Supported configuration clock source and `AS_CLK` Frequencies in Intel^® Stratix^® 10 Devices table.
May 2017	2017.05.22	Updated Connection Setup for Programming the EPCQ-L Device using the AS Interface figure. Updated guideline to program the EPCQ-L device in Programming EPCQ-L Devices using the Active Serial Interface.
April 2017	2017.04.10	Updated note for AS Fast Mode in MSEL Settings for Each Configuration Scheme of Devices table. Added note to Configuration via Protocol recommending user to use AS x4 fast mode for CvP application. Updated instances of Spansion to Cypress. Updated note and description in Configuration Overview. Removed AS x1 support. Added Connection Setup for SD/MMC Single-Device Configuration figure. Updated Connections for AS x4 Single-Device Configuration, Connection Setup for AS Configuration with Multiple EPCQ-L Devices, Connection Setup for Programming the EPCQ-L Devices using the JTAG Interface, Connection Setup for NAND Flash Single-Device Configuration, and Connection Setup for SD/MMC Single-Device Configuration to include note about nCONFIG test point. Added note in Avalon-ST Configuration stating that `AVST_CLK` should be continuous.
February 2017	2017.02.13	Updated Configuring Stratix 10 Devices using AS Configuration section and subsections to include .jic for AS configuration scheme. Added Programming .jic files into EPCQ-L Device. Updated the SDM description. Updated SDM block diagram by adding Mailbox block and note for Avalon-ST x8 configuration scheme. Updated Configuration Sequence Diagram. Updated configuration sequence descriptions. Updated Avalon-ST Bus Timing Waveform figure. Added note to Avalon-ST in Stratix 10 Configuration Overview table. Updated ASx4 max data rate in Stratix 10 Configuration Overview table. Removed Configurable Node subsection.
December 2016	2016.12.09	Updated max data rate for ASx1. Updated the Configuration Sequence in Stratix 10 Devices figure. Updated configuration sequence description. Added JTAG configuration sequence description. Added Parallel Flash Loader II IP core.
October 2016	2016.10.31	Initial release

Avalon-ST

JTAG

CvP

AS Normal Mode

AS Fast Mode

SD/MMC

Initial Configuration Timing

Reconfiguration Timing

Configuration Error

Power Supply Status

Power Up

SDM Startup

Idle

Configuration Start

Configuration Pass

Configuration Error

User Mode

Device Clean

JTAG Configuration

Fixed SDM I/O Pin Assignments for Avalon® -ST x8 and AS x4

Device Configuration Pins without Fixed Assignments

Unused SDM Pins

OSC_CLK_1 Requirements

Notes for Figure:

Debugging Suggestions

MSEL Pin Function for the AS x4 Configuration Scheme

Debugging Suggestions

Debugging Suggestions

Multiple Corrupted Images

Block Diagram

Fixed SDM I/O Pin Assignments for Avalon^® -ST x8 and AS x4