Intel Stratix 10 Configuration User Guide

RSS

UG-S10CONFIG | 2021.10.04

Version Information

Updated for:
Intel® Quartus® Prime Design Suite 21.2

1. Intel Stratix 10 Configuration User Guide

1.1. 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 such as the on-board quad SPI flash device that is storing the FPGA bitstreams.

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

Table 1. Intel^® Stratix^® 10 Configuration Scheme, Data Width, and MSEL
Configuration Scheme		Data Width (bits)	MSEL[2:0]
Passive	Avalon^® -ST	32¹	000
		16¹	101
		8	110
	JTAG	1	111
	Configuration via Protocol (CvP)	x1, x2, x4, x8, x16 lanes	001 ²
Active	AS - fast mode	4¹	001
Active	AS - normal mode	4¹	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 pins 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. Intel^® Stratix^® 10 devices support two CvP modes, CvP initialization and CvP update.

Note: Typically, the data rate of the device's internal configuration data path, not the PCIe* link width, limits the configuration data rate. The maximum data rate depends on the PCIe* generation and number of lanes.

CvP initialization process includes the following two steps:

During the board power up, the CvP uses quad SPI memory in AS x4 mode to configure the FPGA with the periphery image to enable CvP interface that includes the PCIe* IP. The 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 initialization 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 to change core algorithms logic blocks.
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 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 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 that this mode does not delay for 10 ms before beginning configuration. Use this mode to meet the 100 ms of link up requirement for PCIe* or for other systems with strict timing requirements.

In AS fast mode, the power-on sequence must ensure that the quad SPI flash memory is out of reset before the SDM because the Intel^® Stratix^® 10 device 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 report that 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.

¹ This configuration scheme is not supported for Intel^® Stratix^® 10 GX 10M devices.

² 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.

1.1.1. 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.

1.1.2. 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.

1.2. 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 integrity checks in user mode
Other sub-systems:
- Hard processor system (HPS)
- High Bandwidth Memory (HBM2)
- Transceiver tiles

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

1.2.1. 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: During the configuration state, the SDM authenticates the Intel-generated configuration firmware and configuration bitstream, ensuring that configuration bitstream is from a trusted source. All Intel^® Stratix^® 10 support authentication.
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.

The following security features are available in Intel^® Stratix^® 10 devices that support advanced security. Devices with advanced security enabled can only load firmware shipped with Intel^® Quartus^® Prime Pro Edition software. Intel recommends Intel^® Stratix^® 10 devices with a -BK ordering part number (OPN) suffix for use with the black key provisioning feature.

Table 3. Supported Security Features in Intel^® Stratix^® 10 Devices
Intel^® Stratix^® 10	Authentication	Advanced Security
GX	Yes	-AS suffix devices
GX 10M	Yes	No
SX	Yes	-AS suffix devices
MX	Yes	-AS suffix devices
TX	Yes	-AS suffix devices
DX	Yes	Yes

Figure 3. SDM Block Diagram

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

SDM uses temperature sensor for SmartVID feature to communicate to the 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 a special hardware storage mechanism.
The AS enables active configuration schemes via dedicated SDM pins.
The Avalon^® -ST x8 configuration scheme uses SDM I/O pins. The Avalon^® -ST x16 and x32 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. All Intel^® Stratix^® 10 devices compress the configuration bitstream. This feature is always enabled. When specifying an encrypted configuration bitstream, the Intel^® Quartus^® Prime Pro Edition software compresses the configuration bitstream before encryption.
A specific PCIe* block included in the Intel^® Stratix^® 10 device supports CvP.

1.2.1.1. Restricting Security Features

Attention: The -V device does not support device attestation and anti-tamper security features. If you are using these security features, you must select a non-V device.

For a non-V device, when you use attestation and/or anti-tamper security features, you cannot use the conditional public key entry or the owner root key virtual fusing security features. Similarly, if you need to use conditional public key entry and/or the virtual fusing, you cannot turn on device attestation and/or anti-tamper security features in your design.

Table 4. Available Security Features for a Non-V Device
Security Features	Attestation / Anti-Tamper
Security Features	On	Off
Conditional public key entry	Not supported	Supported
Owner root key virtual fusing	Not supported	Supported

If you need to turn on or off the anti-tamper or attestation features in your design, you do not need to recompile your design. You can regenerate the .sof file by selecting the Intel^® Quartus^® Prime Pro Edition Processing > Start > Start Assembler menu.

1.2.1.2. Updating the SDM Firmware

When you generate a configuration bitstream using the File > Programming File Generator menu item, the bitstream assembler adds all firmware (including the SDM firmware) that matches the Intel^® Quartus^® Prime Pro Edition Release to the bitstream generated from the .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

For more information about the output file types, refer to the Intel^® Quartus^® Prime Pro Edition User Guide: Programmer.

Newer versions of the Intel^® Quartus^® Prime software typically include new or updated SDM features implemented in firmware. When regenerating your configuration bitstream, Intel recommends using the latest version of the Intel^® Quartus^® Prime Pro Edition Software which includes the latest firmware. You do not need to recompile your .sof to use the firmware from a newer version of the Intel^® Quartus^® Prime Pro Edition Software. You can simply regenerate your configuration bitstream with the new version of the Programming File Generator.

1.2.1.3. 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. In this mode the fabric begins functioning just before the HPS exits reset. Note that FPGA First option does not allow FPGA reconfiguration by using HPS. This user guide defines a state when the FPGA is functional. Configuration and initialization are complete.

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.

2. Intel Stratix 10 Configuration Details

2.1. Intel Stratix 10 Configuration Timing Diagram

Figure 4. Power-On, Configuration, and Reconfiguration Timing Diagram

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

Power-On 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 state is unknown.

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 Device Datasheet.

Initial Configuration Timing

The first section of the figure shows the expected timing for initial configuration after a normal power-on reset . Initially, the application logic drives the nCONFIG signal low (POR). Under normal conditions nSTATUS follows nCONFIG because nSTATUS reflects the current configuration state. nCONFIG must only change when it has the same value as nSTATUS.

When an error occurs, nSTATUS pulses low and asserts high when the device is ready to accept reconfiguration.

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. Holding nCONFIG signal low during power up is optional.
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. Throughout the configuration, 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 to indicate the FPGA has entered 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 U-Boot or the operating system (OS) initiates the configuration, the FPGA configures and enters user mode.
The entire device does not enter user mode simultaneously. Intel requires you to include reset release as described in the Including the Reset Release Intel FPGA IP in Your Design. Use the nINIT_DONE output of the Reset Release Intel^® FPGA 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 in 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. nCONFIG signal must be held low until the device drives the nSTATUS 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 and starts to reconfigure.
Note: If you do not monitor the nSTATUS signal, pulse the nCONFIG signal low for at least 1000 ms to initiate a reconfiguration request.

Recoverable Configuration Error

Figure 5. Recoverable Error during Reconfiguration Timing Diagram

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

The SDM drives nSTATUS signal low for a period of time specified in the Intel^® Stratix^® 10 Device Datasheet to indicate a recoverable configuration error. The Intel^® Stratix^® 10 device may not assert CONF_DONE indicating that device did not receive the complete configuration bitstream. The device does not assert INIT_DONE indicating that the configuration did not complete successfully. nCONFIG should continue to be driven high until after nSTATUS has returned back to high state.
If an error occurs during JTAG configuration, the SDM does not change the state of the nSTATUS signal. You can monitor the error messages that the Intel^® Quartus^® Prime Pro Edition Programmer generates for error reporting.
The SDM enters the error state.
The SDM enters the idle state if the nCONFIG signal drives to low. 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.

Unrecoverable Configuration Error

Figure 6. Unrecoverable Error during Reconfiguration Timing Diagram

In rare instances, a configuration error or a security event may be unrecoverable. In these cases, the SDM drives nSTATUS low and it stays low. You must perform a power cycle to restart the reconfiguration process. To ensure error recovery under all reconfiguration circumstances, Intel recommends that you design your system to continuously monitor nSTATUS and enable device power cycling if needed.

Note that in the case of an anti-tamper event or a double bit ECC error in the SDM RAM, nSTATUS is also asserted low and stays low.

2.2. Configuration Flow Diagram

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

Figure 7. Intel^® Stratix^® 10 FPGA Configuration Flow

Note: You can perform JTAG configuration anytime from any state as long as the device is powered up and the power is intact. 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.

Power-On

The Intel^® Stratix^® 10 power supplies follow 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.
During power-on stage, internal circuitry pulls the SDM_IO0, SDM_IO8, and SDM_IO16 low internally. Internal circuitry pulls the remaining SDM_IO pins to a weak high.
After POR, internal circuitry also pulls all GPIO pins to a weak high until the device enters user mode.

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.

FPGA Configuration

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.
The power management activity is ongoing during the device configuration. For more information, refer to the Intel^® Stratix^® 10 Power Management User Guide.
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.

Failed FPGA Configuration

A low pulse on the nSTATUS pin indicates a configuration error.
An internal device wipe occurs followed by errors requiring 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. After CONF_DONE asserts and before INIT_DONE asserts, parts of the device start to enter user mode. The assertion of INIT_DONE indicates that the entire device entered user mode. Intel requires you to include the Reset Release in your design. Use the nINIT_DONE output of the Reset Release Intel^® FPGA 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.
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.

2.3. Device Response to Configuration and Reset Events

The following table summarizes the device response to various external configuration and reset events.

Note: HPS_COLD_nRESET is a SDM input pin that manages the HPS reset.

Table 5. Device Response Due To Configuration and Reset EventsEvents marked by a tick (√) require reset initiated by provided reset type.
Action	Reset Type
Action	Power Cycle	`nCONFIG`	`HPS_COLD_nRESET`
Wipe the FPGA	√	√	—
Sample `MSEL` pins	√	—	—
Read fuses	√	—	—
Run the SDM boot ROM code	√	—	—
Reset the SDM	√	—	—
Reset the HPS	√	√	√

Note: When using QSPI, you can use Remote System Update (RSU) to load a specific image with the same device responses as nCONFIG.

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

The Intel^® Stratix^® 10 device has specific clock requirements for PCIe* , HPS EMIF, eSRAM, and the High Bandwidth Memory (HBM2) IP. These clock requirements must be met before the FPGA configuration begins.

FPGA Configuration

To avoid configuration failures, the Intel^® Stratix^® 10 device requires clocks for the PCIe* , HPS EMIF, eSRAM, the HBM2 IP, and all E-tile transceiver reference clocks. You must provide a free-running, stable reference clock to these blocks before configuration begins and throughout the entire user mode. The clock frequencies must match the frequency settings specified in the Intel^® Quartus^® Prime software during configuration. Stopping the reference clock during the user mode may result in a functional failure. 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: HPS_OSC_CLK, when HPS enabled ³
HPS EMIF: pll_ref_clk
L- and H-tile PCIe* channels: REFCLK_GXB
E-tile: REFCLK_GXE
In the Intel^® Stratix^® 10 TX/MX devices, when using PRESERVE_UNUSED_XCVR_CHANNEL_QSF assignment to protect unused channels by enabling transceiver circuits, you must provide a free-running and stable reference clock to the transceiver circuit.

Note: The transceiver power supplies must be at 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^® Quartus^® Prime Pro Edition software allows you to configure the HPS prior the FPGA configuration. To enable this option, select HPS First in the Assignments > Device > Device and Pin Options > Configuration > HPS/FPGA Configuration order dialog box.

HPS First Configuration

Intel^® Stratix^® 10 devices have the option of booting the HPS before configuring the FPGA core logic. This method is known as the HPS first configuration. When you choose this option in the Intel^® Quartus^® Prime Pro Edition software, the following clocks must be operational prior to the FPGA I/O, HPS I/O, and HPS boot, also called a phase 1 configuration:

HPS reference clock: HPS_OSC_CLK
HPS EMIF (when in use): pll_ref_clk
E-tile transceivers: REFCLK_GXE

The remaining clocks specified in the FPGA Configuration must be fully operational prior the FPGA core logic configuration, also called phase 2 configuration. For more information on HPS boot first mode, refer to the Intel^® Stratix^® 10 SoC FPGA Boot User Guide.

³ If you use the FPGA to HPS free clock as the HPS PLL reference clock, the HPS_OSC_CLK clock may not be required.

2.5. Intel Stratix 10 Configuration Pins

The Intel^® Stratix^® 10 device uses SDM_IO pins for device configuration. Control of SDM I/O pins passes from internal FPGA circuitry, to the Boot ROM, and finally to the value your application logic specifies.

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 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.

2.5.1. 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 Assignments > Device > Device and Pin Options dialog box in the Intel^® Quartus^® Prime software.

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 6. 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_nCSOO`
`SDM_IO6`	—	`AVSTx8_DATA4`	`AS_DATA3`
`SDM_IO7`	`MSEL1`	—	`AS_nCSO2`
`SDM_IO8`	—	`AVST_READY`	`AS_nCSO3`
`SDM_IO9`	`MSEL2`	—	`AS_nCSO1`
`SDM_IO10`	—	`AVSTx8_DATA7`	—
`SDM_IO11`	—	`AVSTx8_VALID`	—
`SDM_IO12`	—	—	—
`SDM_IO13`	—	`AVSTx8_DATA5`	—
`SDM_IO14`	—	`AVSTx8_CLK`	—
`SDM_IO15`	—	`AVSTx8_DATA6`	—
`SDM_IO16`	—	—	—

2.5.2. 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 7. 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) ⁴	001
AS (Normal mode) ⁵	011
JTAG only⁶	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

⁴ If you use AS Fast mode, you must ramp all power supplies to the recommended operating condition within 10 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.

⁵ If you use AS Normal mode, you must fully ramp the V_{CCIO_SDM} supply to the recommended operating condition within 10 ms.

⁶ JTAG configuration works with any valid MSEL settings, unless disabled for security.

2.5.3. 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 as an indicator to ensure that configuration is successful. 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 8. 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 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.

2.5.3.1. 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.

2.5.3.1.1. nCONFIG

The nCONFIG pin is a dedicated, input pin of 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 drive nCONFIG low to initiate device cleaning. Then the host should drive nCONFIG high to initiate configuration. If the host drives nCONFIG low during a configuration cycle, that configuration cycle stops. The SDM expects a new configuration cycle to begin.

2.5.3.1.2. 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 or failures using the JTAG configuration scheme.

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 0.5 ms to 10 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. Intel^® Stratix^® 10 always drives nSTATUS. Consequently, you cannot wire OR an Intel^® Stratix^® 10 nSTATUS signal with the nSTATUS signal from earlier device families.

nSTATUS must be pulled high externally and the SDM must sample nSTATUS high when V_{CCIO_SDM} ramps up to the recommended operating voltage.

2.5.3.1.3. 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. The CONF_DONE signal indicates that the configuration bitstream is received successfully. The INIT_DONE pin indicates that the device operates within the design.

In the current implementation, you cannot wire 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 Including the Reset Release Intel FPGA IP in Your Design 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.

2.5.3.1.4. 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 other SDM_IO pins weakly high during power-on.

Figure 10. Configuration Pin Selection in the Intel^® Quartus^® Prime Pro Edition Software

Figure 11. Fitter Report and SDM_IO Pin Reporting

2.5.3.2. 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.

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

Table 9. 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_lv
`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_io_d8s1_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_io_d8s1_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_io_d8s1_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.

2.5.3.4. 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 ED8401, EM21XX, EM22XX, ISL82XX, or LTM4677 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. Select a regulator you use on your board. If you are using a different PMBus regulator, change the default setting from LTM4677 to Other.

Figure 12. Specifying the Slave Device Type for Power Management and VID

Intel^® Quartus^® Prime Pro Edition software allows you to control the payload value of the Page command. Some PMBus devices contain more than one page per bank of registers, the Page command allows you to select the target Page per bank registers. The Page payload available range is between 0x00 and 0xFF.

Note: Prior the Intel^® Quartus^® Prime Pro Edition software release, the Page command selected all banks by default.

Figure 13. Specifying the Page Command Setting

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

2.5.3.5. 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^® FPGA IP.

2.6. Configuration Clocks

2.6.1. 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:

To specify OSC_CLK_1 as the clock source, 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.

2.6.2. OSC_CLK_1 Clock Input

OSC_CLK_1 Requirements

When you drive the 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. If you use transceivers, you must provide an external clock to this pin.

When you specify OSC_CLK_1 for configuration, the OSC_CLK_1 clock must be a stable and free-running clock.

When you specify AS configuration scheme and nCONFIG is pull high, the SDM starts the configuration once the device exits the POR state. Ensure the OSC_CLK_1 clock is available before SDM starts to load the bitstream from the quad SPI flash or you need to supply a stable free-running clock before/at the same time V_{CCIO_SDM} ramps up to the typical voltage level.

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 or the OSC_CLK_1 is not stable and free running due to an interruption or a frequency change.
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.

Configuration Clock Requirements for Reconfiguration Without Power Cycling the Device

When you specify OSC_CLK_1 for configuration and reconfigure without powering down the Intel^® Stratix^® 10 device, the device can only reconfigure with OSC_CLK_1. In this scenario, OSC_CLK_1 must be a free-running clock.

Configuration Clock Requirements for Configuration After Powering Cycling the Device

After a power-down, when you specify OSC_CLK_1 for configuration, the Intel^® Stratix^® 10 device uses the internal oscillator to load the first section of the bitstream and OSC_CLK_1 for the remainder.

2.7. Maximum Configuration Time Estimation

Hyper Initialization is an option that can be enabled or disabled through the setting in the Intel^® Quartus^® Prime software to initialize or reset the Intel^® Hyperflex™ registers to a known state at device configuration. By default, this option is off. For information about register initialization, refer to Device Initialization. For more information about Intel^® Hyperflex™ registers, refer to the Intel^® Hyperflex™ Architecture High-Performance Design Handbook.

The configuration time is the time estimated for the entire configuration state. For more information, refer to Figure 4.

The maximum configuration time is estimated when the device starts configuration until CONF_DONE is asserted to high.

Table 10. Maximum Configuration Time Estimation for Intel^® Stratix^® 10 Devices ( Avalon^® -ST)
Note: Provided configuration times are based on the AVST_CLK = 125 MHz.
Variant	Product Line	Maximum Configuration Time (ms) [Hyper Initialization Off/Hyper Initialization On]
		AVST ×8 ⁷		AVST ×16 ⁷		AVST ×32 ⁷
		Configuration Clock Source: Internal Oscillator	Configuration Clock Source: OSC_CLK_1 (25/100/125 MHz)	Configuration Clock Source: Internal Oscillator	Configuration Clock Source: OSC_CLK_1 (25/100/125 MHz)	Configuration Clock Source: Internal Oscillator	Configuration Clock Source: OSC_CLK_1 (25/100/125 MHz)
Intel^® Stratix^® 10 GX, SX, TX, MX, and DX	GX 400, SX 400, TX 400	183/222	122/148	108/147	72/98	90/129	60/86
	GX 650, SX 650	274/334	182/222	154/216	102/144	120/184	80/122
	GX 850, GX 1100, SX 850, SX 1100, TX 850, TX 1100, DX 1100	456/1,200	304/378	246/358	164/238	190/300	126/200
	GX 1660, GX 2110, TX 1650, TX 2100, MX 1650, MX 2100, DX 2100	754/852	502/568	394/496	262/330	214/316	142/210
	GX 1650, GX 2100, GX 2500, GX 2800, SX 1650, SX 2100, SX 2500, SX 2800, TX 2500, TX 2800, DX 2800	1,102/1,240	734/826	568/708	378/472	300/442	200/294

Table 11. Maximum Configuration Time Estimation for Intel^® Stratix^® 10 Devices (AS x4)
Note: Provided configuration times are based on the AS_CLK = 115 MHz for Internal Oscillator and AS_CLK = 125 MHz for OSC_CLK_1 (25/100/125 MHz).
Variant	Product Line	Maximum Configuration Time (ms) [Hyper Initialization Off/Hyper Initialization On]
		AS ×4
		Configuration Clock Source: Internal Oscillator	Configuration Clock Source: OSC_CLK_1 (25/100/125 MHz)
Intel^® Stratix^® 10 GX, SX, TX, MX, and DX	GX 400, SX 400, TX 400	408/447	272/298
	GX 650, SX 650	568/630	378/420
	GX 850, GX 1100, SX 850, SX 1100, TX 850, TX 1100, DX 1100	900/1,012	600/674
	GX 1660, GX 2110, TX 1650, TX 2100, MX 1650, MX 2100, DX 2100	1,432/1,534	954/1,022
	GX 1650, GX 2100, GX 2500, GX 2800, SX 1650, SX 2100, SX 2500, SX 2800, TX 2500, TX 2800, DX 2800	2,058/2,200	1,372/1,466

⁷ The maximum configuration time does not include the time incurred from external storage and control logic, and transceiver calibration time.

3. Intel Stratix 10 Configuration Schemes

3.1. 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 12. Avalon^® -ST Configuration Data Width, Clock Rates, and Data RatesMbps 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 Mbps	000
	16	125 MHz	2000 Mbps	101
	8	125 MHz	1000 Mbps	110

Table 13. 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 > Device and Pin Options > Configuration 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 > 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	Input	V_{CCIO_SDM}
`CONF_DONE` ⁸	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 INIT_DONE configuration signal is not required for configuration, Intel recommends that you use this signal. The SDM drives the INIT_DONE signal high to indicate the device is fully in user mode. This signal is 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.

⁸ CONF_DONE is required if you are using the Intel FPGA Parallel Flash Loader II IP as the configuration host.

3.1.1. 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 14. Components and Design Flow for .pof Programming

Table 14. 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.

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.

3.1.2. 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.

3.1.3. 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.

If you use the Parallel Flash Loader II Intel^® FPGA IP core as the configuration host, the AVST_READY synchronizer logic is included.

Figure 15. 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.

3.1.4. 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 15. 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 16. 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

3.1.5. 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 16. Connections for Avalon-ST x8 Single-Device Configuration

Figure 17. Connections for Avalon-ST x16 Single-Device Configuration

Figure 18. 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 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.

3.1.6. 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 the host device does not drive configuration pins before the Intel^® Stratix^® 10 device powers up.
Ensure nCONFIG remains high during the Avalon^® -ST configuration scheme.
Verify that CONF_DONE and INIT_DONE pins correlate to the Intel^® Quartus^® Prime SDM I/O pins assignment and the board-level connections.
Ensure that during the power up, no external component drives the nSTATUS signal low.

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

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

Note: The current implementation does not support programming two QSPI devices with two separate PFL images in a single programming cycle. To program multiple QSPI devices, you must program each QSPI flash device with a single PFL image separately.

Note: The Parallel Flash Loader II Intel FPGA IP does not support HPS cold reset.

3.1.7.1.1. 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 MT28EW CFI flash memory devices to the host in parallel using the same data bus, clock, and control signals. Intel does not support connecting two of non-MT28W CFI flash memory devices to PFL II IP core in parallel. During FPGA configuration, the AVST_CLK frequency is four times faster than the flash_clk frequency.

Figure 20. PFL II IP core with Dual MT28EW CFI Flash Memory DevicesThe flash memory devices must have the same memory density from the same device family and manufacturer.

3.1.7.1.2. 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.

3.1.7.1.3. 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 21. 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 22. 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 17. 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` ⁹	.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.

⁹ The .pof version occupies only one byte in the option bits sector.

3.1.7.1.4. 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 18. 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 19. 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.

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

The following figure shows a 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 23. 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 24. 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 20. 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`

3.1.7.2. Designing with the PFL II IP Core for Avalon-ST Single Device Configuration

This section describes the procedures on how to use the PFL II IP core.

To target a MAX^® II, MAX^® V, or Intel^® MAX^® 10 device requires the use of Intel^® Quartus^® Prime Standard Edition whereas targeting a Intel^® Stratix^® 10 requires Intel^® Quartus^® Prime Pro Edition.

The process of creating the Avalon-ST single device configuration design targeting a MAX10/MAX V/MAX II device involves three steps.

Generate the AVST design for the MAX device with the default option address.
Create the Intel^® Stratix^® 10 .pof file in setting the appropriate option bits.
Regenerate the Parallel Flash Loader II Intel FPGA IP (PFL II) with the option bits used to generate the Intel^® Stratix^® 10 .pof file and recompile the Intel^® MAX^® 10 design.

You can find a MAX^® V system design example that implements the PFL II IP for AVST x16 configuration mode in the installer package of the Intel^® Stratix^® 10 GX FPGA Development Kit.

Figure 25. Process for Using the PFL IP CoreFigure shows the process for using the PFL IP core, using MAX^® II as an example.

3.1.7.2.1. PFL II Parameters

Table 21. 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 22. 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 23. 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 24. 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 (MT28EW) 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 (MT28EW)—applicable tor MT28EW 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.

3.1.7.2.2. PFL II Signals

Table 25. 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. You can use `pfl_nreconfigure` to drive the `fpga_nconfig` output signal initiating reconfiguration. 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. ¹⁰
`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. ¹⁰
`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.

¹⁰ 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.

3.1.7.3. Generating the PFL II IP Core

3.1.7.3.1. 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 26. 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

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

Figure 28. 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 29. 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 30. 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.

3.1.7.3.3. Creating a Single PFL II IP for Programming and Configuration

Follow these steps to create a single PFL II IP instantiation 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 and FPGA Configuration.
In the same tab, for What is the targeted flash?, select CFI Parallel Flash.
Specify the parameters on the Flash Interface Settings tab:
1. Select 1 for a single flash device.
2. Select CFI 1 Gbit as the largest used flash device.
3. Select 16 bits for the flash interface data width.
Specify the parameters on the FPGA Configuration to match your design.
Compile and generate a .pof for the flash memory device. Ensure that you tri-state all unused I/O pins.

3.1.7.3.4. 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 CPLD 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.

3.1.7.4. Constraining the PFL II IP Core

You can specify many design constraints in Intel^® Quartus^® Prime project settings by editing the .sdc files.

3.1.7.4.1. PFL II IP Recommended Design Constraints to FPGA Avalon-ST Pins

Create a `pfl_clk` clock and a generated `AVST_CLK` clock

Example below creates a pfl_clk clock running at 50 MHz, supplied by the clk_50m_sysmax input clock.

set pfl_clk_period 20.000
create_clock -name {clk_50m_sysmax} -period $pfl_clk_period [get_ports {clk_50m_sysmax}]
create_generated_clock -name AVST_CLK -source [get_ports {clk_50m_sysmax}] [get_ports {avst_clk}]

Set output delay for PFL II IP output pins

Example below sets the output delay for the AvST_DATA and AvST_VALID pins.

set avst_data_tracemax 0.250
set avst_data_tracemin 0.000
set avst_clk_tracemax 0.250
set avst_clk_tracemin 0.000
set fpga_Tsu 5.500
set fpga_Th 0.000
set fpga_out_max_dly [expr $avst_data_tracemax + $fpga_Tsu - $avst_clk_tracemin]
set fpga_out_min_dly [expr $avst_data_tracemin - $fpga_Th - $avst_clk_tracemax]

set_output_delay -add_delay -max -clock [get_clocks {AVST_CLK}] $fpga_out_max_dly [get_ports {avst_d[*] avst_valid}]
set_output_delay -add_delay -min -clock [get_clocks {AVST_CLK}] $fpga_out_min_dly [get_ports {avst_d[*] avst_valid}]

Setting a false path

You can set the AVST_READY input pin to a false path since this pin is not synchronous to the AVST_CLK clock. The host must synchronize the AVST_READY signal to the AVST_CLK signal using a 2-stage register synchronizer.

set_false_path -from [get_ports {avst_ready}] -to *

3.1.7.4.2. PFL II IP Recommended Design Constraints for Using QSPI Flash

Create a `FLASH_CLK` clock

Example below creates assigns QSPI flash clock pin (flash_dc1_ic0) to the flash clock.

create_generated_clock -name FLASH_CLK -source [get_ports {clk_50m_sysmax}] [get_ports {flash_dc1_io0}]

Set output delay for PFL II IP output pins

Example below sets the output delay for the QSPI flash data and chip select pins.

#flash_dc1_io1/3/4/5 = QSPI flash data pins, 
#flash_dc1_io2 = QSPI flash chip select pins
set flash_data_tracemax 0.250
set flash_data_tracemin 0.000
set flash_clk_tracemax 0.250
set flash_clk_tracemin 0.000
set flash_Tsu 2.700
set flash_Th 2.000
set flash_out_max_dly [expr $flash_data_tracemax + $flash_Tsu - $flash_clk_tracemin]
set flash_out_min_dly [expr $flash_data_tracemin - $flash_Th - $flash_clk_tracemax]

set_output_delay -add_delay -max -clock [get_clocks {FLASH_CLK}] $flash_out_max_dly [get_ports {flash_dc1_io1 flash_dc1_io3 flash_dc1_io4 flash_dc1_io5 flash_dc1_io2}]
set_output_delay -add_delay -min -clock [get_clocks {FLASH_CLK}] $flash_out_min_dly [get_ports { flash_dc1_io1 flash_dc1_io3 flash_dc1_io4 flash_dc1_io5 flash_dc1_io2}]

Set input delay for input pins

Example below sets the input delay for the QSPI flash data.

set flash_tco_max 7.000
set flash_tco_min 1.000
set in_max_dly [expr $flash_data_tracemax + $flash_tco_max + $flash_clk_tracemax]
set in_min_dly [expr $flash_data_tracemin + $flash_tco_min + $flash_clk_tracemin]

set_input_delay -clock { FLASH_CLK } -max $in_max_dly [get_ports {flash_dc1_io1 flash_dc1_io3 flash_dc1_io4 flash_dc1_io5}]
set_input_delay -clock { FLASH_CLK } -min $in_min_dly [get_ports {flash_dc1_io1 flash_dc1_io3 flash_dc1_io4 flash_dc1_io5}]

3.1.7.4.3. PFL II IP Recommended Design Constraints for using CFI Flash

Create a `FLASH_CLK` clock

Example below assigns CFI flash clock pin (flash_clk) to the flash clock. You constrain the flash_clk pin only when using burst mode.

create_generated_clock -name FLASH_CLK -source [get_ports {clk_50m_max5}] [get_ports {flash_clk}]

Set output delay for output pins

Example below sets the output delay for CFI flash output pins.

set flash_data_tracemax 0.250
set flash_data_tracemin 0.000
set flash_clk_tracemax 0.250
set flash_clk_tracemin 0.000
set flash_Tsu 3.500
set flash_Th 2.000
set flash_out_max_dly [expr $flash_data_tracemax + $flash_Tsu - $flash_clk_tracemin]
set flash_out_min_dly [expr $flash_data_tracemin - $flash_Th - $flash_clk_tracemax]

#Note: For normal mode, the clock is referred to input pfl_clk clock(clk_50m_max5) of PFL II IP. If burst mode is used, the clock is referred to flash clock of PFL II IP.   
set_output_delay -add_delay -max -clock [get_clocks {clk_50m_max5}] $flash_out_max_dly [get_ports {flash_nce[0] flash_nce[1] flash_noe flash_nwe flash_addr[*] flash_data[*]}]
set_output_delay -add_delay -min -clock [get_clocks {clk_50m_max5}] $flash_out_min_dly [get_ports { flash_nce[0] flash_nce[1] flash_noe flash_nwe flash_addr[*] flash_data[*]}]
#Only need to constraint flash_advn pin when using burst mode.
set_output_delay -add_delay -max -clock [get_clocks { FLASH_CLK }] $flash_out_max_dly [get_ports {flash_nadv}]
set_output_delay -add_delay -min -clock [get_clocks { FLASH_CLK }] $flash_out_min_dly [get_ports {flash_nadv}]

Set input delay for input pins

Example below sets the input delay for CFI flash data.

# For Normal Mode
set flash_noe_tracemax 0.250
set flash_noe_tracemin 0.000
set flash_tco_max 7.000
set flash_tco_min 0.000
set normal_in_max_dly [expr $flash_data_tracemax + $flash_tco_max + $ flash_noe_tracemax]
set normal_in_min_dly [expr $flash_data_tracemin + $flash_tco_min + $ flash_noe_tracemin]

set_input_delay -clock { clk_50m_max5 } -max $normal_in_max_dly [get_ports {flash_data[*]}]
set_input_delay -clock { clk_50m_max5 } -min $normal_in_min_dly [get_ports {flash_data[*]}]

# For Burst mode
set flash_tco_max 5.500
set flash_tco_min 2.000
set burst_in_max_dly [expr $flash_data_tracemax + $flash_tco_max + $flash_clk_tracemax]
set burst_in_min_dly [expr $flash_data_tracemin + $flash_tco_min + $flash_clk_tracemin]

set_input_delay -clock { FLASH_CLK } -max $burst_in_max_dly [get_ports {flash_data[*]}]
set_input_delay -clock { FLASH_CLK } -min $burst_in_min_dly [get_ports {flash_data[*]}]

3.1.7.4.4. PFL II IP Recommended Constraints for Other Input Pins

Set a false path for PFL II IP input pins

You can set the pfl_nreset input reset pin to a false path since this pin is asynchronous.

set_false_path -from [get_ports {pfl_nreset}] -to *

Set input delay to PFL II IP input pins

Example below sets the input delay for the pfl_flash_access_granted pin.

You don't have to constraint the path when you use the device arbiter logic to control the pin,
You don't have to constraint the path when not using the device arbiter logic or the external processor to control the pin and you loopback the pfl_flash_access_request signal to the pfl_flash_access_granted pin.
You can constraint the path when a processor or an external device controls the pfl_flash_access_granted pin.

set_input_delay -clock {clk_50m_sysmax} -max [<pfl_flash_access_granted_tco_max> + <pfl_flash_access_granted_tracemax>] [get_ports {pfl_flash_access_granted}]
set_input_delay -clock {clk_50m_sysmax} -min [<pfl_flash_access_granted_tco_min> + <pfl_flash_access_granted_tracemin>] [get_ports {pfl_flash_access_granted}]

Set a false path for `fpga_pgm[]` input pin

You can set a false path to quasi-static signals such as reset and configuration signals (fpga_conf_done, fpga_nstatus) that are stable for a long periods of time.

set_false_path -from [get_ports {fpga_pgm[]}] -to *

Set input delay to `pfl_nreconfigure` input pin

If you use the external component to drive this pin, you must set the input delay path to drive the pfl_nreconfigure pin.

set_input_delay -clock {clk_50m_sysmax} -max [<pfl_nreconfigure_tco_max> + <pfl_nreconfigure_tracemax>] [get_ports {pfl_nreconfigure}]
set_input_delay -clock {clk_50m_sysmax} -min [<pfl_nreconfigure_tco_min> + <pfl_nreconfigure_tracemin>] [get_ports {pfl_nreconfigure}]

Set input delay to `pfl_reset_watchdog` pin

If you use the external component to drive this pin, you must set the input delay path to drive the pfl_reset_watchdog pin.

set_input_delay -clock {clk_50m_sysmax} -max [$pfl_reset_watchdog_tco_max + $pfl_reset_watchdog_tracemax] [get_ports {pfl_nreconfigure}]
set_input_delay -clock {clk_50m_sysmax} -min [$pfl_reset_watchdog_tco_min + $pfl_reset_watchdog_tracemin] [get_ports {pfl_nreconfigure}]

3.1.7.4.5. PFL II IP Recommended Constraints for Other Output Pins

Set output delay to PFL II IP output pins

Example below sets the output delay for the pfl_flash_access_request output pin.

You don't have to constraint the path when this signal feeds arbiter logic or device tristate logic,
You don't have to constraint the path when this signal feeds the pfl_flash_access_granted input pin while not using the device arbiter logic or the external processor.
You can constraint the path when this signal feeds the processor or an external device controls.

set_output_delay -add_delay -clock [get_clocks { clk_50m_sysmax }] -max $flash_access_request_tracemax [get_ports {pfl_flash_access_request}]
set_output_delay -add_delay -clock [get_clocks { clk_50m_sysmax }] -min $flash_access_request_tracemin [get_ports {pfl_flash_access_request}]

Set output delay to `flash_nreset` output pin

The flash_nreset output pin is available in Burst mode only.

set_output_delay -add_delay -max -clock [get_clocks { FLASH_CLK }] $flash_out_max_dly [get_ports {flash_nreset}]
set_output_delay -add_delay -min -clock [get_clocks { FLASH_CLK }] $flash_out_min_dly [get_ports {flash_nreset}]

Set a false path for `fpga_nconfig` output pin

You can set the fpga_nconfig output pin to a false path since the nCONFIG is an asynchronous input pin..

set_false_path -from [get_ports {fpga_nconfig}] -to *

Set output delay to `pfl_watchdog_error` output pin

You don't have to constraint the path when the signal feeds the internal logic.
You can constraint the path when signal feeds an external host.

set_output_delay -add_delay -clock [get_clocks { clk_50m_sysmax }] -max $pfl_watchdog_error_tracemax [get_ports {pfl_watchdog_error}]
set_output_delay -add_delay -clock [get_clocks { clk_50m_sysmax }] -min $pfl_watchdog_error_tracemin [get_ports {pfl_watchdog_error}]

3.1.7.5. Using the PFL II IP Core

3.1.7.5.1. 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.

View the video guide and complete the following steps to convert .sof file to a .pof:

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 31. Programming File Generator Output Files Tab
To specify a .sof that contains the configuration bitstream, on the Input Files tab, click Add Bitstream.

Figure 32. Input Files Tab
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 33. Edit Partition: OPTIONS for Flash Device
With the flash device selected, click Add Partition to specify a partition in flash memory.

Figure 34. 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.

3.1.7.5.2. 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.

3.1.7.5.3. 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.

3.1.7.5.4. 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 26. 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 35. Using the Programmer Edit Menu to Define a New Flash Device

Table 27. 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.

3.2. 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. For more information about the power sequence, refer to the Intel^® Stratix^® 10 Power Management User Guide.

Important: Do not reset the quad SPI flash when used as the configuration device and data storage device with FPGA. Resetting the quad SPI flash during the FPGA configuration and reconfiguration, or in the QSPI's READ/WRITE/ERASE operations, causes undefined behavior for quad SPI flash and the FPGA. To recover from the unresponsive behavior, you must power cycle your device. To reset the quad SPI flash via the external host, you must first complete the FPGA configuration and reconfiguration, or a quad SPI operation, and only then toggle the reset. The quad SPI operation is complete when the exclusive access to the quad SPI flash is closed by issuing the QSPI_CLOSE command via the Mailbox Client Intel^® FPGA IP or CLOSE command via the Serial Flash Mailbox Client Intel^® FPGA IP.

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

Table 28. 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	125 MHz	500 Mbps	Fast mode - 001 Normal mode - 011

Table 29. Required Configuration Signals for the AS Configuration Scheme
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	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.

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_nCSOO asserts low when the SDM starts to communicate with the QSPI flash. The remaining chip select pins, AS_nCSO1 - AS_nCSO3 deassert high.

Note: MSEL[0] and AS_nCSOO pins share the same SDM I/O. Refer to Table 6 for more details.

3.2.1. 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.

For PCIe designs including the Configuration via Protocol (CvP), Intel recommends that you use Micron QSPI flash memory. When using the Micron QSPI flash memory, the boot ROM uses the AS x4 mode to load the initial configuration firmware faster to meet the PCIe wake-up time for host enumeration.

When using other flash memory types for PCIe designs, the boot ROM reads the firmware using AS x1 mode. Intel recommends asserting the PERST# signal low for a minimum of 200 ms counting from the device exiting the power-on reset (POR) state. This ensures the PCIe endpoint enters the link training state prior to the PERST# signal deasserts.

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

Figure 36. 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 30. 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.

3.2.2. 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 37. Connections for AS x4 Single-Device Configuration

3.2.3. 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 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 38. 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 31. 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	125	115
30	100	77
37	71.5	77
80	50	58
140	25	25

3.2.4. AS Configuration Timing Parameters

Figure 39. AS Configuration Serial Output Timing Diagram

Figure 40. AS Configuration Serial Input Timing Diagram

Table 32. T_{ext_delay} as a Function of `AS_CLK` Frequency
Symbol	Configuration Clock Source	Frequency	Max (ns)
T_{ext_delay}	Internal Oscillator	115 MHz	20
		77 MHz	20
		58 MHz	20
		25 MHz	24
	`OSC_CLK_1`	125 Mhz	18
		100 MHz	24
		71.5 MHz	35
		50 MHz	24
		25 MHz	24

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

3.2.5. 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 33. 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	125 MHz	—	—	4.00
		115 MHz	—	—	4.20
		100 MHz	—	—	5.0
		<100 MHz	—	—	5.0

3.2.6. 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

3.2.6.1. 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_nCSO0-AS_nCSO3. The Intel^® Quartus^® Prime Programmer programs the flash memory devices via the AS header. If you are using the Generic Serial Flash Interface Intel^® FPGA IP to write the flash memory the flash device must be connected to GPIO to access the flash device.

Attention: When you power up the Intel^® Stratix^® 10 device with an empty serial flash device and use the AS interface to program the .rpd file into this serial flash device, you must power cycle the Intel^® Stratix^® 10 device to configure the device from the flash successfully.

Figure 41. AS Programming Using Intel^® Quartus^® Prime or Third-Party Programmer

3.2.6.2. 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 42. Programming Your Serial Configuration Device Using JTAG and SDM Emulation of AS

Figure 43. 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.

3.2.7. 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 44. 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

3.2.7.1. 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.

3.2.8. 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 125 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 34. Supported Configuration Clock Source and `AS_CLK` Frequencies in Intel^® Stratix^® 10 DevicesThe table displays valid `AS_CLK` settings for the respective configuration clock source.
Configuration Clock Source	`AS_CLK` Frequency (MHz)
Internal oscillator	25 58 77 115
`OSC_CLK_1` (25/100/125 MHz)	25 50 71.5 100 ¹¹ 125 ¹²

Note: The configuration fails if the firmware receives bitstream with an invalid AS_CLK setting. For firmware in Intel^® Quartus^® Prime software version before 21.1, if AS_CLK is set incorrectly, the firmware defaults the AS_CLK frequency to 50 MHz to complete the AS x4 configuration.

¹¹ The .sof files, generated with selected 108 MHz AS_CLK frequency in the earlier versions of Intel^® Quartus^® Prime software, default to this frequency when you generate a programming file (.jic/.pof/.rbf/.rpd) using Intel^® Quartus^® Prime Programming File Generator version 21.1 or newer.

¹² The .sof files, generated with selected 133 MHz AS_CLK frequency in the earlier versions of Intel^® Quartus^® Prime software, default to this frequency when you generate a programming file (.jic/.pof/.rbf/.rpd) using Intel^® Quartus^® Prime Programming File Generator version 21.1 or newer.

3.2.9. 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.

3.2.10. Intel Quartus Prime Programming Steps

3.2.10.1. 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.

View the video guide and complete the following steps to generate the programming file(s) 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 45. 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 46. Programing File Generator Input Files
On the Configuration Device tab, click Add Device. You can select your flash device 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 47. 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 .pof 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(s).
1. 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 use this file to regenerate the programming file by running the following command:
```
quartus_pfg -c <configuration_file>.pfg
```

3.2.10.2. 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.

3.2.10.3. 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.

3.2.11. 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_nCSO0-AS_nCSO3, 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, you must ramp all power supplies to the recommended operating condition within 10 ms. This ramp-up requirement ensures that the AS x4 device is within its operating voltage range when the Intel^® Stratix^® 10 device begins accessing 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, all the power supplies 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, all power supplies must ramp to the recommended operating condition within 10 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 does not resolve the problem, you have two possible solutions depending on the components you are using for configuration:
- If you are using a third-party programmer to configure the flash directly from an AS or JTAG header as shown in Figure 41 change the MSEL setting to JTAG. Setting MSEL to JTAG prevents the corrupt image from loading automatically at power-on. Then, update the image in quad serial flash through the AS or JTAG header.
- If you are programming the flash device using the JTAG header as shown in Figure 42, force the nCONFIG signal to low. When nCONFIG is low, the image cannot load from the quad SPI flash device. Then, update the image in quad serial flash through the JTAG header.
If you are using AS x4 flash memories with AS Fast mode, you must ramp up all power supplies to the recommended operating condition within 10 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.
If you are using the OSC_CLK_1 clock source for configuration, ensure OSC_CLK_1 is free running and stable before SDM starts to load a bitstream from the Quad SPI device. The SDM starts configuration after the device exits the POR state if nCONFIG is held high.
Try a lower AS clock frequency setting.
Reconfiguration in AS mode may fail if a system asserts the external reset of the serial flash device used for configuration, before reconfiguration is triggered. Ensure you are not resetting the serial flash device prior to a reconfiguration.
When you power up the Intel^® Stratix^® 10 with an empty serial flash device and use the AS interface to program the .rpd file into this serial flash device, you must power cycle the Intel^® Stratix^® 10 device to configure the device from the flash successfully
If you drive nCONFIG signal using the external host, ensure it remains high during the AS x4 configuration scheme.
If you pulse nCONFIG low for reconfiguration, ensure that the nSTATUS acknowledges nCONFIG. If nSTATUS does not follow the nCONFIG signal, the FPGA may not exit power on reset state. You may need to power cycle the PCB.
Ensure no external component drives the nSTATUS signal low during the power up.

3.2.12. QSF Assignments for AS

You can specify many Intel^® Quartus^® Prime project settings using Intel^® Quartus^® Prime Software GUI. The following assignments in the .qsf show typical settings for a Intel^® Stratix^® 10 device using AS 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 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
# ======================

# The following setting also supports 
               Intel^® 
               Stratix^® 10 devices
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 48. Set SDM_IO Configuration Pins Using the Intel^® Quartus^® Prime Software

3.3. 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/.rbf file. Configuration using the JTAG device chain allows faster development because it does not require you to program external flash memory. You can also use JTAG to reprogram if the image stored in quad SPI memory. You can also use the JTAG configuration scheme to reprogram the quad SPI memory if the quad SPI content is corrupted or invalid.

The Intel^® Quartus^® Prime software generates a .sof/.rbf file containing the FPGA design information. You can use the .sof/.rbf file 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 35. Intel^® Stratix^® 10 Configuration Data Width, Clock Rates, and Data RatesMbps is an abbreviation for Megabits per second.
Mode		Data Width (bits)	Max Clock Rate	Max Data Rate	MSEL[2:0]
Passive	JTAG	1	30 MHz	30 Mbps	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 36. 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. Intel^® Quartus^® Prime software enables the HPS secure debug by default.
Configuration Function	Pin Type	Direction	Powered by
`TCK`	Fixed	Input	V_{CCIO_SDM}
`TDI` ¹³	Fixed	Input	V_{CCIO_SDM}
`TMS` ¹³	Fixed	Input	V_{CCIO_SDM}
`TDO` ¹³	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	Input	V_{CCIO_SDM}

¹³ The JTAG pins can access the HPS JTAG chain in Intel^® Stratix^® 10 SoC devices.

3.3.1. 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 49. JTAG Configuration Scheme

Figure 50. Components and Design Flow for JTAG Programming

3.3.2. 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.

3.3.2.1. 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 51. Connection Setup for JTAG Single-Device Configuration using Download Cable

3.3.2.2. 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 52. Connection Setup for JTAG Single-Device Configuration using a Microprocessor

3.3.3. 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.

3.3.3.1. JTAG Multi-Device Configuration using Download Cable

Figure 53. Connection Setup for JTAG Multi Device Configuration using Download Cable

3.3.4. 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, nSTATUS 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 does not change state 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 the MSEL setting on the PCB is not JTAG, if 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 you 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.
Clearing the RSU_STATUS command after the JTAG reconfiguration depends on your Intel^® Quartus^® Prime software version. In earlier Intel^® Quartus^® Prime software versions, the RSU_STATUS is not cleared after the JTAG reconfiguration. Starting with the Intel^® Quartus^® Prime 20.3 version, the system clears the RSU_STATUS after the JTAG reconfiguration.
Ensure that during the power up, no external component drives the nSTATUS signal low.

4. Including the Reset Release Intel FPGA IP in Your Design

Intel requires that you either use the Reset Release Intel^® FPGA IP to hold your design in reset until configuration is complete.

The Reset Release Intel^® FPGA 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 hold your design in reset while the nINIT_DONE signal is high or while the INIT_DONE pin is low. 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 54. Reset Release Intel^® FPGA IP nINIT_DONE Internal Connection

4.1. Understanding the Reset Release IP Requirement

Intel^® Stratix^® 10 devices use a parallel, sector-based architecture that distributes the core fabric logic across multiple sectors. Device configuration proceeds in parallel with each Local Sector Manager (LSM) configuring its own sector. Consequently, FPGA registers and core logic do not exit reset at exactly the same time, as has always been the case in previous families.

The continual increases in clock frequency, device size, and design complexity now necessitate a reset strategy that considers the possible effects of slight differences in the release from reset. The Reset Release Intel FPGA IP holds a control circuit in reset until the device has fully entered user mode. The Reset Release FPGA IP generates an inverted version of the internal INIT_DONE signal, nINIT_DONE for use in your design.

After nINIT_DONE asserts (low), all logic is in user mode and operates normally. You can use the nINIT_DONE signal in one of the following ways:

To gate an external or internal reset.
To gate the reset input to the transceiver and I/O PLLs.
To gate the write enable of design blocks such as embedded memory blocks, state machine, and shift registers.
To synchronously drive register reset input ports in your design.

Attention: When you instantiate Reset Release Intel FPGA IP in your design, the Intel^® Quartus^® Prime Fitter selects one Local Sector Manager (LSM) to output the nINIT_DONE signal. Multiple instances results in some skew between the nINIT_DONE signals.

4.2. Instantiating the Reset Release IP In Your Design

The Reset Release IP is available in the IP Catalog in the Basic Functions > Configuration and Programming category. This IP has no parameters.

Complete the following steps to instantiate the Reset Release IP in your design.

In the IP Catalog, type reset release in the search window to find the Reset Release Intel FPGA IP.

Figure 55. Locate Reset ReleaseIntel FPGA IP in IP Catalog
Double click the Reset Release Intel FPGA IP to add the Reset Release IP to your design.
In the New IP Variant dialog box, browse to your IP directory and specify a file name for the Reset Release IP. Then click Create. The Reset Release IP is now included in your project.

4.3. Gating the PLL Reset Signal

In older FPGA device families, designs frequently used the PLL lock signal to hold the custom FPGA logic in reset until the PLL locked. In newer Intel device families the lock time of PLLs can be less than the initialization time. In some cases the PLL may lock before the device completes initialization. Consequently, if you use the locked output of the PLL to control resets in the Intel^® Stratix^® 10 device, you should gate the PLL reset input with nINIT_DONE as shown the figure.

Figure 56. Using nINIT_DONE to Gate the PLL_Reset Signal

Another alternative if you are using PLL_Lock in your reset sequence is to gate the PLL_Lock output with the nINIT_DONE signal, (PLL_Lock && !nINIT_DONE).

4.4. Guidance When Using Partial Reconfiguration (PR)

The PR Region Controller IP provides reset logic that ensures that the static region of the device and the PR personas do not interact during PR.

The Reset Release IP is only necessary to manage reset for full FPGA core configuration and subsequent full FPGA core reconfigurations. The Reset Release IP is not necessary to prevent interaction between the static and PR personas during the PR process. For more information about PR refer to the Intel^® Quartus^® Prime Pro Edition User Guide: Partial Reconfiguration.

4.5. Detailed Description of Device Configuration

Each Local Sector Manager (LSM) configures its own sector. A sector comprises multiple logic array block (LAB) rows. A logical function can span multiple rows and multiple sectors.

During configuration global configuration control signals hold the core fabric in a frozen state to prevent electrical contention. The LSMs work in parallel to asynchronously unfreeze the sectors. Within a sector the LSM unfreezes LAB rows and registers in the LABs sequentially. The LSMs work to unfreeze the fabric in parallel across all sectors without synchronization. Consequently, logic in different sectors or in the same sector but in different rows could begin to operate while other logic is still frozen. The INIT_DONE signal asserts when all the LSMs have entered user mode.

Figure 57. Releasing LAB Rows and Registers in the LABs Sequentially and Asynchronously Across Sectors

The following topics provide more detail about device configuration and initialization, and possible consequences if you do not use the Reset Release IP to hold the Intel^® Stratix^® 10 device in reset until entire fabric enters user mode.

4.5.1. Device Initialization

The following steps summarize device initialization:

An external host drives a configuration request to the Secure Device Manager (SDM) by driving nCONFIG high. The SDM exits the IDLE state and signals the beginning of configuration by driving nSTATUS high and driving configuration data.
The SDM asserts CONF_DONE indicating that the Intel FPGA has successfully received all the configuration data.
The SDM uses the configuration logic to start non-gated clocks in the fabric. Intel^® Hyperflex™ registers begin shifting data. Consequently, the initial conditions of Intel^® Hyperflex™ registers can be random. Use the Disable Register Power-up Initialization setting in the Intel^® Quartus^® Prime Configuration dialog box to disable Intel^® Hyperflex™ register initialization during power-on as explained below.
The SDM uses the configuration logic to enable and initialize user registers in the LABs, DSP, and embedded memory blocks.
The SDM drives INIT_DONE to indicate that the device has fully entered user mode. The Reset Release IP asserts nINIT_DONE. Intel recommends that you use nINIT_DONE to gate your reset logic.
The FPGA is now in user mode and ready for operation.

4.5.2. Preventing Register Initialization During Power-On

If not held in reset, both ALM and Intel^® Hyperflex™ registers may lose their initial state if they initialize before their respective source.

You can prevent registers from initializing during power-on by enabling an option in the Intel^® Quartus^® Prime software. Complete the following steps to turn on this option:

On the Assignments menu select Device > Device and Pin Options > Configuration.
In the Configuration dialog box, turn on Disable Register Power-up Initialization.

Figure 58. Disabling Register Initialization During Power-On

Note:

Coming out of reset, you cannot rely on the value of registers with initial conditions unless you gate your system reset using one of the following options:

Designs that include the Reset Release IP must route nINIT_DONE to the system reset.
Designs that do not include the Reset Release IP must route INIT_DONE to an external pin and feed INIT_DONE back into the FPGA as an input to system reset.

4.5.3. Embedded Memory Block Initial Conditions

Initialized content of embedded memory blocks is stable during configuration. However, designs that contain logic to modify embedded memory can result in spurious writes. Spurious writes can occur if you fail to gate the write enable with an appropriate reset.

4.5.4. Protecting State Machine Logic

To guarantee correct operation of state machines, your reset logic must hold the FPGA fabric in reset until the entire fabric enters user mode.

The following example shows how an inadequate reset strategy might result in an illegal state in a one-hot state machine. In this example, the design does not reset any of the state machine registers. The state machine design depends on registers entering an initial state. Without an adequate reset, this state machine begins operating when part of the device is active. Nearby logic included in the state machine remains frozen, before INIT_DONE asserts.

Figure 59. Partially Initialized Design - INIT_DONE = 0

Register B in the active section is operational and takes on the value of Register A in the next clock cycle. Register A is still in the freeze register state and does not respond to the clock edge. Register A remains in the current state.

Figure 60. Advance One Clock Cycle, Device Completely In User Mode - INIT_DONE = 1

The entire fabric is now in user mode. The state machine enters an illegal or unknown state with two ones in a one-hot state machine. To prevent this illegal state, use the Reset Release IP to hold the circuit in reset until INIT_DONE asserts indicating that the entire fabric has entered user mode.

5. 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^® FPGA IP you can write configuration bitstreams to the AS x4 flash device. Then you can use the Mailbox Client Intel^® FPGA IP 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^® FPGA IP 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 61. Typical Remote System Update Process

5.1. Remote System Update Functional Description

5.1.1. RSU Glossary

Table 37. 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 backup application image that the RSU loads when all attempts to load an application image fail. The factory image should provide enough functionality for the device to recover when all application images are corrupt. Once the factory image loads, you can program new application images to replace the failing images. The SDM loads the factory image in the following circumstances: You assign the `Direct to Factory Image` function to an SDM I/O pin and assert the pin after a power-on reset or `nCONFIG` deassertion. All SDM attempts to load the application images fail. You make a request to the SDM to load the factory image. The configuration system treats the factory image in the same way as it does an application image.
Initial RSU flash image	Contains the factory image, the application images, the decision firmware, and the associated RSU data structures.
Factory update image	An image that updates the following RSU-related items in flash: The factory image The decision firmware The decision firmware data Intel recommends that your factory update image include the minimum amount of logic necessary to debug successfully your design if your application image or images fail to load.

5.1.2. 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 Mailbox Client Intel^® FPGA IP 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.
Figure 62. 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

If you use the Mailbox IP in designs compiled in Intel^® Quartus^® Prime Pro Edition software version 19.2 or later, you must only use SDM firmware starting from version 19.2 or later to configure the FPGA.

5.1.3. 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 seven 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.

5.1.4. Remote System Update Configuration Sequence

Figure 63. 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, 256 K, 512 K, or 768 K to initialize the SDM. The same configuration firmware is present in each of these locations. This firmware is part of the initial RSU flash image. ( Refer to Step 2 of Guidelines for Performing Remote System Update Functions for Non-HPS for step-by-step details for programming the initial RSU flash image 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.

Note: You must keep nCONFIG high until the device enters the user mode.

Keep the nCONFIG signal high after the device powers up and throughout the entire device configuration to load an application or factory image.
Keep the nCONFIG signal high during the remote update to other application image or factory image by using the RSU_IMAGE_UPDATE command.

You drive nCONFIG low only when the device is in user mode to trigger the reconfiguration.

5.1.5. RSU Recovery from Corrupted Images

When an RSU fails, the Mailbox Client Intel^® 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 64. Multiple Corrupt Images

5.1.6. 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.
Trigger reconfiguration to load the update image from the starting address using one of the following methods:
- Trigger reconfiguration using the RSU_IMAGE_UPDATE command. You don't need to update CPB0 and CPB1.
- Trigger reconfiguration by pulsing the nCONFIG signal and power cycling the device. Write the factory update image start address in both CPB0 and CPB1. After factory image updates, the SDM automatically removes the address from CPB0 and CPB1.
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 CPB 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 and the CPB.

5.2. Guidelines for Performing Remote System Update Functions for Non-HPS

Figure 65. 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^® 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^® 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

5.3. Commands and Responses

The host controller communicates with the SDM using command and response packets via the Mailbox Client Intel^® FPGA IP.

The first word of the command and response packets is a header that provides basic information about the command or response.

Block Diagram

The following figure illustrates the role of the Mailbox Client Intel^® FPGA IP 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 66. Command and Response Header Format

Note: The LENGTH field in the command header must match the command length of corresponding command. Your client must read all the response words, even if your client does not interpret all the response words.

The following table describes the fields of the header command.

Table 38. 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. Refer to Operation Commands for command descriptions.
`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. If the command succeeds, the `Error Code` is 0. If the command fails, refer to the error codes defined in the Error Code Responses.

5.3.1. Operation Commands

Resetting Quad SPI Flash

Important: For Intel^® Stratix^® 10 devices, do not reset the quad SPI flash when used as the configuration device and data storage device with FPGA. Resetting the quad SPI flash during the FPGA configuration and reconfiguration, or in the QSPI's READ/WRITE/ERASE operations, causes undefined behavior for quad SPI flash and the FPGA. To recover from the unresponsive behavior, you must power cycle your device. To reset the quad SPI flash via the external host, you must first complete the FPGA configuration and reconfiguration, or a quad SPI operation, and only then toggle the reset. The quad SPI operation is complete when the exclusive access to the quad SPI flash is closed by issuing the QSPI_CLOSE command via the Mailbox Client Intel^® FPGA IP or CLOSE command via the Serial Flash Mailbox Client Intel^® FPGA IP.

Table 39. Command List and Description
Command	Code (Hex)	Command Length ¹⁴	Response Length ¹⁴	Description
`RSU_IMAGE_UPDATE`	5C	2	0	Triggers reconfiguration from the data source that can be either the factory or an application image. This command takes an optional 64-bit argument that specifies the reconfiguration data address in the flash. When sending the argument to the IP, you first send bits [31:0] followed by bits [63:32]. If you do not provide this argument its value is assumed to be 0. Bit [31:0]: The start address of an application image. Bit [63:32]: Reserved (write as 0). Once the device processes this command, it returns the response header to response FIFO before it proceeds to reconfigure the device. Ensure the host PC or host controller stops servicing other interrupts and focuses on reading the response header data to indicate the command completed successfully. Otherwise, the host PC or host controller may not be able to receive the response once the reconfiguration process started. Once the device proceeds with reconfiguration, the link between the external host and FPGA is lost. If you use PCIe in your design, you need to re-enumerate the PCIe link. Important: When resetting quad SPI, you must follow instructions specified in Resetting Quad SPI Flash.
`RSU_GET_SPT`	5A	0	4	`RSU_GET_SPT` retrieves the quad SPI flash location for the two sub-partition tables that the RSU uses: SPT0 and SPT1. The 4-word response contains the following information:
				Word	Name	Description
				0	SPT0[63:32]	SPT0 address in quad SPI flash.
				1	SPT0[31:0]	SPT0 address in quad SPI flash.
				2	SPT1[63:32]	SPT1 address in quad SPI flash.
				3	SPT1[31:0]	SPT1 address in quad SPI flash.
`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 information:
				Word	Summary	Description
				0	State	Describes the most recent configuration related error. Returns 0 when there are no configuration errors. The error field has 2 fields: Upper 16 bits: Major error code. Lower 16 bits: Minor error code. Refer to Appendix: CONFIG_STATUS and RSU_STATUS Error Code Descriptions in the Mailbox Client Intel^® FPGA IP User Guide for more information.
				1	Version	The version of the RSU data structure.
				2	Pin status	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	Contains the value of each of the soft functions, even if you have not assigned the function to an SDM pin. Bit [31:6]: Reserved Bit [5]: `HPS_WARMRESET` Bit [4]: `HPS_COLDRESET` 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 if there are no errors.
				5	Error details	Contains the error details. Returns 0 if there are no errors.
`RSU_STATUS`	5B	0	9	Reports the current remote system upgrade status. You can use this command to check the configuration status during configuration and after it has completed. This command returns the following responses:
				Word	Summary	Description
				0-1	Current image	Flash offset of the currently running application image.
				2-3	Failing image	Flash offset of the highest priority failing application image. If multiple images are available in flash memory, stores the value of the first image that failed. A value of all 0s indicates no failing images. If there are no failing images, the remainder of the remaining words of the status information do not store valid information. Note: A rising edge on `nCONFIG` to reconfigure from ASx4, does not clear this field. Information about failing image only updates when the Mailbox Client receives a new `RSU_IMAGE_UPDATE` command and successfully configures from the update image.
				4	State	Failure code of the failing image. The error field has two parts: Bit [31:16]: Major error code Bit [15:0]: Minor error code Returns 0 for no failures. Refer to Appendix: CONFIG_STATUS and RSU_STATUS Error Code Descriptions in the Mailbox Client Intel^® FPGA IP User Guide for more information.
				5	Version	The version of the RSU software. For information about version fields, refer to Hard Processor System Remote System Update User Guide.
				6	Error location	Stores the error location of the failing image. Returns 0 for no errors.
				7	Error details	Stores the error details for the failing image. Returns 0 if there are no errors.
				8	Current image retry counter	Count of the number of retries that have been attempted for the current image. The counter is 0 initially. The counter is set to 1 after the first retry, then 2 after a second retry. Specify the maximum number of retries in your Intel^® Quartus^® Prime Settings File (.qsf). The command is: `set_global_assignment -name RSU_MAX_RETRY_COUNT 3`. Valid values for the `MAX_RETRY` counter are 1-3. The actual number of available retries is `MAX_RETRY -1` This field was added in version 19.3 of the Intel^® Quartus^® Prime Pro Edition software.
`RSU_NOTIFY`	5D	1	0	Clears all error information in the `RSU_STATUS` response and resets the retry counter. The one-word argument has the following fields: 0x00050000: Clear current reset retry counter. Resetting the current retry counter sets the counter back to zero, as if the current image was successfully loaded for the first time. 0x00060000: Clear error status information. All other values are reserved. This command is not available before version 19.3 of the Intel^® Quartus^® Prime Pro Edition software.
`QSPI_OPEN`	32	0	0	Requests exclusive access to the quad SPI. You issue this request before any other QSPI requests. The SDM accepts the request if the quad SPI is not in use and the SDM is not configuring the device. Returns OK if the SDM grants access. The SDM grants exclusive access to the client using this mailbox. Other clients cannot access the quad SPI until the active client relinquishes access using the `QSPI_CLOSE` command. Access to the quad SPI flash memory devices via any mailbox client IP is not available by default in designs that include the HPS, unless you disable the QSPI in HPS software configuration. Important: When resetting quad SPI, you must follow instructions specified in Resetting Quad SPI Flash.
`QSPI_CLOSE`	33	0	0	Closes the exclusive access to the quad SPI interface. Important: When resetting quad SPI, you must follow instructions specified in Resetting Quad SPI Flash.
`QSPI_SET_CS`	34	1	0	Specifies one of the attached quad SPI devices via the chip select lines. Takes a one-word argument as described below: Bits[31:28]: Flash device to select. The value `4'b0000` selects the flash that corresponds to `nCSO[0]`. `nCSO[0]` is the only signal that the FPGA can use to access the quad SPI flash device. Bits[27:0]: Reserved (write as 0). Note: The HPS can use `nCSO[3:1]` to access 3 additional quad SPI devices. This command is optional for the AS x4 configuration scheme, the chip select line follows the last executed `QSPI_SET_CS` command or defaults to `nCSO[0]` after the AS x4 configuration. The JTAG configuration scheme requires executing this command to access the QSPI flash that connects the SDM_IO pins. Access to the QSPI flash memory devices using SDM_IO pins is only available for the AS x4 configuration scheme, JTAG configuration, and a design compiled for AS x4 configuration. For the Avalon^® streaming interface ( Avalon^® ST) configuration scheme, you must connect QSPI flash memories to GPIO pins. Important: When resetting quad SPI, you must follow instructions specified in Resetting Quad SPI Flash.
`QSPI_READ`	3A	2	N	Reads the attached quad SPI device. The maximum transfer size is 4 kilobytes (KB) or 1024 words. Takes two arguments: The quad SPI flash address (one word). The address must be word aligned. The device returns the `0x1` error code for non-aligned addresses. Number of words to read (one word). When successful, returns OK followed by the read data from the quad SPI device. A failure response returns an error code. For a partially successful read, `QSPI_READ` may erroneously return the `OK` status. Note: You cannot run the `QSPI_READ` command while device configuration is in progress. Important: When resetting quad SPI, you must follow instructions specified in Resetting Quad SPI Flash.
`QSPI_WRITE`	39	2+N	0	Writes data to the quad SPI device. The maximum transfer size is 4 kilobytes (KB) or 1024 words. Takes three arguments: The flash address offset (one word). The write address must be word aligned. The number of words to write (one word). The data to be written (one or more words). A successful write returns the OK response code. To prepare memory for writes, use the `QSPI_ERASE` command before issuing this command. Note: You cannot run the `QSPI_WRITE` command while device configuration is in progress. Important: When resetting quad SPI, you must follow instructions specified in Resetting Quad SPI Flash.
`QSPI_ERASE`	38	2	0	Erases a 4/32/64 KB sector of the quad SPI device. Takes two arguments: The flash address offset to start the erase (one word). Depending on the number of words to erase, the start address must be: 4 KB aligned if number words to erase is 0x400 32 KB aligned if number words to erase is 0x2000 64 KB aligned if number words to erase is 0x4000 Returns an error for non-4/32/64 KB aligned addresses. The number of words to erase is specified in multiples of: 0x400 to erase 4 KB (100 words) of data. This option is the minimum erase size. 0x2000 to erase 32 KB (500 words) of data 0x4000 to erase 64 KB (1000 words) of data A successful erase returns the OK response code. Important: When resetting quad SPI, you must follow instructions specified in Resetting Quad SPI Flash.
`QSPI_READ_DEVICE_REG`	35	2	N	Reads registers from the quad SPI device. The maximum read is 8 bytes. Takes two arguments: The opcode for the read command. The number of bytes to read. A successful read returns the OK response code followed by the data read from the device. The read data return is in multiple of 4 bytes. If the bytes to read is not an exact multiple of 4 bytes, it is padded with multiple of 4 bytes until the next word boundary and the padded bit value is zero. Important: When resetting quad SPI, you must follow instructions specified in Resetting Quad SPI Flash.
`QSPI_WRITE_DEVICE_REG`	36	2+N	0	Writes to registers of the quad SPI. The maximum write is 8 bytes. Takes three arguments: The opcode for the write command. The number of bytes to write. The data to write. To perform a sector erase or sub-sector erase, you must specify the serial flash address in most significant byte (MSB) to least significant byte (LSB) order as the following example illustrates. To erase a sector of a Micron 2 gigabit (Gb) flash at address 0x04FF0000 using the `QSPI_WRITE_DEVICE_REG` command, write the flash address in MSB to LSB order as shown here: Header: 0x00003036 Opcode: 0x000000DC Number of bytes to write: 0x00000004 Flash address: 0x0000FF04 A successful write returns the OK response code. This command pads data that is not a multiple of 4 bytes to the next word boundary. The command pads the data with zero. Important: When resetting quad SPI, you must follow instructions specified in Resetting Quad SPI Flash.
`QSPI_SEND_DEVICE_OP`	37	1	0	Sends a command opcode to the quad SPI. Takes one argument: The opcode to send the quad SPI device. A successful command returns the OK response code. Important: When resetting quad SPI, you must follow instructions specified in Resetting Quad SPI Flash.

For CONFIG_STATUS and RSU_STATUS major and minor error code descriptions, refer to Appendix: CONFIG_STATUS and RSU_STATUS Error Code Descriptions in the Mailbox Client Intel^® FPGA IP User Guide .

¹⁴ This number does not include the command or response header.

5.3.2. Error Code Responses

Table 40. Error Codes
Value (Hex)	Error Code Response	Description
0	`OK`	Indicates that the command completed successfully. A command may erroneously return the `OK` status if a command, such as `QSPI_READ` is partially successful.
1	`INVALID_COMMAND`	Indicates that the currently loaded boot ROM cannot decode or recognize the command code.
3	`UNKNOWN_COMMAND`	Indicates that the currently loaded firmware cannot decode the command code.
4	`INVALID_COMMAND_PARAMETERS`	Indicates that the command is incorrectly formatted. For example, the length field setting in header is not valid.
6	`COMMAND_INVALID_ON_SOURCE`	Indicates that the command is from a source for which it is not enabled.
8	`CLIENT_ID_NO_MATCH`	Indicates that the Client ID cannot complete the request to close the exclusive access to quad SPI. The Client ID does not match the existing client with the current exclusive access to quad SPI.
9	`INVALID_ADDRESS`	The address is invalid. This error indicates one of the following conditions: An unaligned address An address range problem A read permission problem An invalid chip select value, displaying value of more than 3 An invalid address in RSU case
A	`AUTHENTICATION_FAIL`	Indicates the configuration bitstream signature authentication failure.
B	`TIMEOUT`	This error indicates time out due to the following conditions: Command Waiting for `QSPI_READ` operation to complete Waiting for the requested temperature reading from one of the temperature sensors. May indicate a potential hardware error in the temperature sensor.
C	`HW_NOT_READY`	Indicates one of the following conditions: The hardware is not ready. Can indicate either an initialization or configuration problem. The hardware may refer to quad SPI. RSU image is not used to configure the FPGA.
D	`HW_ERROR`	Indicates that the command completed unsuccessfully due to unrecoverable hardware error.
80 - 8F	`COMMAND_SPECIFIC_ERROR`	Indicates a command specific error due to an SDM command you used.
		SDM Command	Error Name	Error code	Description
		`GET_CHIPID`	`EFUSE_SYSTEM_FAILURE`	0x82	Indicates that the eFuse cache pointer is invalid.
		`QSPI_OPEN`/ `QSPI_CLOSE`/ `QSPI_SET_CS`/ `QSPI_READ_DEVICE_REG`/ `QSPI_WRITE_DEVICE_REG`/ `QSPI_SEND_DEVICE_OP`/ `QSPI_READ`	`QSPI_HW_ERROR`	0x80	Indicates QSPI flash memory error. This error indicates one of the following conditions: A QSPI flash chip select setting problem A QSPI flash initialization problem A QSPI flash resetting problem A QSPI flash settings update problem
			`QSPI_ALREADY_OPEN`	0x81	Indicates that the client's exclusive access to QSPI flash via `QSPI_OPEN` command is already open.
100	`NOT_CONFIGURED`	Indicates that the device is not configured.
1FF	`ALT_SDM_MBOX_RESP_DEVICE_ BUSY`	Indicates that the device is busy due to following use cases: RSU: Firmware is unable to transition to different version due to an internal error. HPS: HPS is busy when in HPS reconfiguration process or HPS cold reset.
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.

5.3.3. Error Code Recovery

The table below describes possible steps to recover from an error code. Error recovery depends on specific use case.

Table 41. Error Code Recovery for known Error Codes
Value	Error Code Response	Error Code Recovery
4	`INVALID_COMMAND_PARAMETERS`	Resend the command header or header with arguments with corrected parameters. For example, ensures that the length field setting in header is sent with the correct value.
6	`COMMAND_INVALID_ON_SOURCE`	Resend the command from valid source such as JTAG, HPS, or core fabric.
8	`CLIENT_ID_NO_MATCH`	Wait for the client who opened the access to quad SPI to complete its access and then closes the exclusive access to quad SPI.
9	`INVALID_ADDRESS`	Possible error recovery steps: For QSPI operation: Send command with a valid chip select. Send command with a valid QSPI flash address. For RSU: Send command with a valid start address of the factory image or application.
B	`TIMEOUT`	Possible recovery steps: For QSPI operation: Check signal integrity of QSPI interfaces and attempt command again. For HPS restart operation: Retry to send the command again.
C	`HW_NOT_READY`	Possible recovery steps: For QSPI operation: Reconfigure the device via source. Ensure that IP used to build your design allows access to the QSPI flash. For RSU: Configure the device with RSU image.
80	`QSPI_HW_ERROR`	Check the QSPI interface signal integrity and to ensure the QSPI device is not damaged.
81	`QSPI_ALREADY_OPEN`	Client already opened QSPI. Continue with the next operation.
82	`EFUSE_SYSTEM_FAILURE`	Attempt reconfiguration or power cycle. If error persists after reconfiguration or power cycle, the device may be damaged and unrecoverable.
100	`NOT_CONFIGURED`	Send a bitstream that configures the HPS.
1FF	`ALT_SDM_MBOX_RESP_DEVICE_ BUSY`	Possible error recovery steps: For QSPI operation: Wait for ongoing configuration or other client to complete operation. For RSU: Reconfigure device to recover from internal error. For HPS restart operation: Wait for reconfiguration via HPS or HPS Cold Reset to complete.

5.4. Quad SPI Flash Layout

5.4.1. High Level Flash Layout

5.4.1.1. Standard (non-RSU) Image Layout in Flash

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 application image in flash.

Figure 67. Image Layout in Flash: Non-RSUIn this figure, each of the four firmware copies points to the Application Image.

Note: When the PUF feature is used, two more 32 K blocks are required for storing PUF data. For more information about PUF and the detailed flash layout, refer to the Intel^® Stratix^® 10 Device Security User Guide.

5.4.1.2. RSU Image Layout in Flash – 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 firmware data
One factory image
Two configuration pointer blocks (CPBs)

Figure 68. RSU Image Layout in Flash : SDM PerspectiveIn this figure:

Each of the Decision Firmware copies point to the Factory Image and both Pointer Block 0 and Pointer Block 1.
Both Pointer Block 0 and Pointer Block 1 point to Application Images.

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
Note: You can set this pin on the following menu in the factory image project: Assignments > Device > Device and Pin Options > Configuration > Configuration Pin Options
The max_retry parameter value.
The max_retry value, stored as an 8-bit value at address 0x10018C in flash, is the only accessible field from the decision firmware data structure. The value in flash is max_retry - 1. That is, a value of zero means each image is tried just once. Modification of the max_retry value directly in the bitstream stored in flash is typically used by RSU via HPS. For more information, refer to Intel^® Stratix^® 10 Hard Processor System Remote System Update User Guide.

The pointer blocks contain a list of application images to try until one of them is successful. If none is 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, that 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.

5.4.1.2.1. Firmware Version Information

The Intel^® Quartus^® Prime firmware version is stored as a 32bit value at offset 0x420 in the following:

Decision firmware (each of the four copies),
Application images,
Factory update images,
Decision firmware update images,
Combined application images.

The format for the version field is described below:

Table 42. Firmware Version Fields
Bit Field	Bits	Description
`version_major`	`31:24`	Major release number for Intel^® Quartus^® Prime. Example: set to 20 for release 20.1.
`version_minor`	`23:16`	Minor release number for Intel^® Quartus^® Prime. Example: set to 1 for release 20.1.
`version_update`	`15:8`	Update release number for Intel^® Quartus^® Prime. Example: set to 2 for release 20.1.2
`reserved`	`7:0`	Set to zero

Both U-Boot and LibRSU enable the decision firmware versions to be queried. For examples on querying the decision firmware version for both U-Boot and Linux, refer to the Remote System Update Example section. The decision firmware version must first be queried in U-Boot for the value to be available in Linux.

5.4.1.3. RSU Image Layout – Your Perspective

The sub-partition table (SPT) is used for managing the allocation of the quad SPI flash.

The Intel^® Quartus^® Prime Programming File Generator creates the SPT when creating the initial manufacturing image. To ensure reliable operation, the Programming File Generator creates two copies of the sub-partition table and the configuration pointer block, SPT0 and SPT1 and CPB0 and CPB1

The initial RSU image stored in flash typically contains the following partitions:

Table 43. Typical Sub-Partitions of the RSU Image
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
P1	Application image 1
P2	Application image 2

Figure 69. RSU Image Layout - Your PerspectiveIn this figure:

SPT0 and SPT1 point to everything:
- BOOT_INFO
- Factory Image
- Pointer Block 0 and Pointer Block 1
- All Application Images
Pointer Block 0 and Pointer Block 1 point to all Application Images

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.
You have access to the sub-partition tables. The sub-partition tables provide access to the data structures required for remote system update.

5.4.2. Detailed Quad SPI Flash Layout

5.4.2.1. RSU Image Sub-Partitions Layout

The RSU Image Sub-Partitions Layout table shows the layout of RSU image stored in QSPI flash.

If you anticipate changes to the factory or application images, you may consider reserving additional memory space. By default, the Intel^® Quartus^® Prime Programming File Generator reserves additional 256 kB of memory space for a factory image. To increase the partition size, update the End Address value in the Edit Partition dialog box window as described in the Generating the Initial RSU Image.

Table 44. RSU Image Sub-Partitions Layout
Flash Offset	Size (in bytes)	Contents	Sub-Partition Name
0 K	256 K	Decision firmware	BOOT_INFO
256 K	256 K	Decision firmware
512 K	256 K	Decision firmware
768 K	256 K	Decision firmware
1 M	8 K + 24 K pad	Decision firmware data
1 M+32 K	32 K	Reserved for SDM
1 M+64 K	Varies	Factory image	FACTORY_IMAGE
Next	4 K + 28 K pad	Sub-partition table (copy 0)	SPT0
Next	4 K + 28 K pad	Sub-partition table (copy 1)	SPT1
Next	4 K + 28 K pad	Pointer block (copy 0)	CPB0
Next	4 K + 28 K pad	Pointer block (copy 1)	CPB1
Varies	Varies	Application image 1	You assign
Varies	Varies	Application image 2	You assign

The Intel^® Quartus^® Prime Programming File Generator allows you to create many user partitions. 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 seven partitions for application images. There are no limitations on creating empty partitions.

5.4.2.2. Sub-Partition Table Layout

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

Note: The firmware never reads or updates the SPT, it only reports the SPT locations in flash.

Table 45. Sub-partition Table Layout
Offset	Size (in bytes)	Description
`0x000`	4	Magic number 0x57713427
`0x004`	4	Version number: 0 - before Intel^® Quartus^® Prime Pro Edition software version 20.4 1 - starting with Intel^® Quartus^® Prime Pro Edition software version 20.4
`0x008`	4	Number of entries
`0x00C`	4	Checksum: 0 - before Intel^® Quartus^® Prime Pro Edition software version 20.4 CRC32 checksum - starting with Intel^® Quartus^® Prime Pro Edition software version 20.4
`0x010`	16	Reserved
`0x020`	32	Sub-partition Descriptor 1
`0x040`	32	Sub-partition Descriptor 2
`0xFE0`	32	Sub-partition Descriptor 126

Starting with Intel^® Quartus^® Prime Pro Edition software version 20.4, the SPT header contains a CRC32 checksum that is computed over the whole SPT. The value of the CRC32 checksum filed itself is assumed as zero when the checksum is computed. Refer to Application Image Layout for the algorithm used to compute the CRC32 checksum. The checksum is provided as a convenience so that SPT corruptions can better be detected by HPS software. By default the feature is turned off.

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

Table 46. 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

Two flags are currently defined:

System set to 1: Reserved for RSU system. For partition offset value, refer to Table 44.
Read only set to 1: The system protects partition against direct writes.

The Intel Quartus Programming File Generator sets these flags as follows at image creation time, then they are not changed afterward:

Table 47. Flags Specifying Contents and Access
Partition	System	Read Only
`BOOT_INFO`	1	1
`FACTORY_IMAGE`	1	1
`SPT0`	1	0
`SPT1`	1	0
`CPB0`	1	0
`CPB1`	1	0
`P1`	0	0
`P2`	0	0

5.4.2.3. Configuration Pointer Block Layout

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

Table 48. 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	Reserved
`0x10`	4	Offset to image pointers (IPTAB)
`0x14`	4	Number of image pointer slots (`NSLOTS`)
`0x18`	8	Reserved
`0x20` (IPTAB) ¹⁵	8	First (lowest priority) image pointer slot (IPTAB)
	8	Second (2nd lowest priority) image pointer slot
	8	…
	8	Last (highest priority) image pointer

The configuration pointer block can contain up to 508 application image pointers. The actual number is listed as NSLOTS. A typical configuration 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 updates before the pointer block must be erased. The erase procedure is called pointer block compression. This procedure is power failure safe, as there are two copies of the pointer block. The copies are in different flash erase sectors. While one copy is being updated the other copy is still valid.

¹⁵ The offset may vary in future firmware updates.

5.4.2.4. Modifying the List of Application Images

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

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

When the configuration pointer block is erased, all entries are marked as unused. To add an application image to the list, the client finds the first unused location and writes the application image address to this location. To remove an application image from the list, the client finds the application image address in the pointer block list and writes all 0's to this address.

If the configuration pointer block runs out of space for new application images, the client compresses the pointer block by completing the following actions:

Read all the valid entries from the configuration
Erase the pointer block
Add all previously valid entries
Add the new image

When using HPS to manage RSU, both the U-Boot and LIBRSU clients implement the block compression. For designs that drive RSU from FPGA logic, 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 configuration pointer blocks: a primary (CPB0) and a backup (CPB1). Two blocks enable the list of application images to be protected if a power failure occurs just after erasing one of them. When a CPB is erased and re-created, the header is written last. The CPB header is checked prior to use to prevent accidental use if a power failure occurred. For more information, refer to the Configuration Pointer Block Layout topic. When compressing, the client compresses (erases and rewrites) 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 is the last word written in the CPB. (After this number is written only image pointer slot values can be changed.)

When the client writes the application image to flash, it ensures that the pointers within the main image pointer of its first signature block are updated to point to the correct locations in flash. When using HPS to manage RSU, both the U-Boot and LIBRSU clients implement the required pointer updates. For more information, refer to the Application Image Layout section.

5.4.2.5. Application Image Layout

The application image comprises SDM firmware and the configuration data. The configuration data includes up to four sections. The SDM 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.

Table 49. 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. This option may not be practical as different systems may have a different set of updates applied, which may result in different slots being suitable to store the new application image.
You can generate the application image as if it is located at address zero, then update the pointers to match the actual location.
- In Intel^® Quartus^® Prime software version 21.1 or later, a Use relative address option is the default option to generate a single application bitstream. You do not have to specify the start address since you can program the image to any available location in the QSPI flash memory. To load the image, you must correctly point to the starting address of the stored image in the flash memory.
  If you choose to disable this option, unselect the Use relative address checkbox and provide the Start address for the image in flash memory. For more information, refer to Generating an Application Image.

When using the HPS to manage RSU, both U-Boot and LIBRSU clients implement the below procedure to relocate application images targeting address zero in the actual destination slot address.

The procedure to update the pointers from an application image created for INITIAL_ADDRESS to NEW_ADDRESS is:

Create the application image, targeting the INITIAL_ADDRESS.
Read the 32-bit value from offset 0x1F00 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 0x1F00 + (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, which requires a different procedure for the pointer updates. When using the HPS to manage RSU, both U-Boot and LIBRSU clients implement this procedure for relocating the factory update image.

Note: The combined application image has a different format, with no pointers that need to be relocated. The image can be placed unmodified in flash at any address.

5.5. 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.

5.5.1. Generating the Initial RSU Image

Follow below steps to generate the initial RSU Image.

For generic applications, you can generate the initial image using the .sof file.

For security application when you need to generate a signed or encrypted .rbf file, you need to first generate .rbf file from a .sof file, and only then generate the initial RSU image from the .rbf file. Refer to Intel^® Stratix^® 10 Device Security User Guide for more information about the generation of the signed or encrypted .rbf file.

Important: You must update the decision firmware (CMF) in the Intel^® Quartus^® Prime software version 21.1 or later if you want to use an RSU image generated with Intel^® Quartus^® Prime software version 20.4 or prior along with any factory or application image(s) generated in Intel^® Quartus^® Prime software version 21.1 or later. For more information, refer to Updating Decision Firmware.

5.5.1.1. Generating the Initial RSU Image Using .sof Files

View the video guide and complete the following 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 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:
- 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 image .sof file and click Open. Repeat this step for the application image .sof.
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 seven partitions for seven application images. The page 1 application image is the highest priority, and the page 7 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. Verification fails because the verify operation compares the updated image to the initial flash data. If you want to verify the updated flash image, you read back the updated image from flash and compare it to the expected .rpd file.

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 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.

Note: If you are using HPS to drive the remote system update, follow the information in the Intel^® Stratix^® 10 Hard Processor System Remote System Update User Guide.

5.5.1.2. Generating the Initial RSU Image Using .rbf Files

Follow these steps to generate the initial RSU image using .rbf file:

Run the following command to generate a .rbf file from a factory or application image file (.sof).
```
quartus_pfg -c factory.sof factory.rbf
quartus_pfg -c app1.sof app1.rbf
quartus_pfg -c app2.sof app2.rbf
```
Run the following command to generate a boot .rbf file from a factory image file (.sof)
```
quartus_pfg -c factory.sof boot.rbf -o rsu_boot=ON
```
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 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)
On the Input Files tab, click Add Bitstream, select the factory image .rbf file and click Open. Repeat this step for the application image .rbf.
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.
1. Assign boot.rbf to the BOOT_INFO partition.
2. Assign factory.rbf to the FACTORY_IMAGE partition.
3. Assign app1.rbf and app2.rbf to the P1 and P2 respectively.
  
  Note: P1 and P2 are user-defined partition names.
If you generate the .jic file, select the Flash Loader. On the Configuration Device tab, select Flash Loader, click Select. Select device family and device name. 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.

5.5.1.3. Updating Decision Firmware

This section describes scenario when you must update the decision firmware in flash memory.

The RSU image consists of decision firmware, factory image, application image 1, and application image 2. Starting with Intel^® Quartus^® Prime software version 21.1, if you use an RSU image generated prior to Intel^® Quartus^® Prime software version 21.1, you must update the decision firmware with the Intel^® Quartus^® Prime software version 21.1 or later if you plan to deploy factory or application images from Intel^® Quartus^® Prime software version 21.1 or later.

You can choose to only update decision firmware or update decision firmware along with the factory image.

To update decision firmware and the factory image, run the following command to generate the RSU image:
```
quartus_pfg –c <input.sof> <output.rpd> -o mode=ASx4 -o rsu_upgrade=ON
```
To update decision firmware only, run the following command to generate the RSU image:
```
quartus_pfg –c <input.sof> <output.rpd> -o mode=ASx4 -o rsu_upgrade=ON -o firmware_only=ON
```

You can also use a combined application image to update decision firmware or decision firmware data in flash. For more information, refer to the Combined Application Images section in the Intel^® Stratix^® 10 Hard Processors System Remote System Update User Guide.

Typically, you update the decision firmware after the .rpd file is created as described in the Updates with the Factory Update Image.

5.5.2. Generating an Application Image

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

Note: If you selected Use relative address option, specifying start_address is optional.

Note: The rsu1.tcl script that Intel provides performs the bit swap operation. Consequently, if you are using this script, set bitswap=OFF in the command above.

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)
In Intel^® Quartus^® Prime software version 20.4 or earlier, 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. If the address is incorrect, the image does not load successfully when you trigger an update with this image.

In Intel^® Quartus^® Prime software version 21.1 or later, a Use relative address option is the default option to generate a single application bitstream. You do not have to specify the start address since you may program the image to any available location in the QSPI flash memory. To load the image, you must correctly point to the starting address of the stored image in flash memory.

If you choose to disable this option, unselect the Use relative address checkbox. You must provide the Start address for the image in flash memory.

Figure 70. 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.

Note: The rsu1.tcl script that Intel provides performs the bit swap operation. Consequently, if you are using this script, set Bit swap to Off.
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 71. 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.

5.5.3. Generating a Factory Update Image

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

Note: The rsu1.tcl script that Intel provides performs the bit swap operation. Consequently, if you are using this script, set bitswap=OFF in the command above.

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. 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 factory update image in flash memory. You do not have to create a separate partition for the factory update image. The factory update image typically includes the minimum amount of logic necessary to successfully debug your design if your application image fails to load. Consequently the Start address can be the sector boundary of unused space in flash memory. This Start address must match the starting address of the target sector in flash memory. If the address is incorrect, the image does not load when the host tries to use this image as a backup image.

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 72. 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.

Note: The rsu1.tcl script that Intel provides performs the bit swap operation. Consequently, if you are using this script, set Bit swap to Off.
On the Input Files tab, click Add Bitstream. If necessary, change the Files of type to SRAM Object File (*.sof). Then, select factory image .sof file and click Open.

Figure 73. Specify the .sof File
Select the .sof and then click Properties. Turn On Generate RSU factory update image. Specify the Bootloader file.

Note: You only have to specify the Bootloader file for Intel^® Stratix^® 10 SX devices.

Figure 74. 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.

5.5.4. 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 39 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 JTAG 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.

5.6. 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.

5.6.1. 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^® 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.
In addition, your design must include the Reset Release Intel^® FPGA IP. This component holds the design in reset until the entire FPGA fabric has entered user mode.
The ninit_done_reset and reset_bridge_1 components create a two-stage reset synchronizer to release the Mailbox Client Intel^® FPGA IP and JTAG to Avalon Master Bridge Intel^® FPGA IP from reset when the device configuration is complete and the device is in user mode.
The ninit_done output signal from Reset Release IP gates this reset by connecting to the ninit_done_reset in_reset pin.
The reset_in Reset Bridge Intel^® FPGA IP provides a user mode reset. In this design, the exported resetpin connects to application logic.
Figure 75. Required Communication and Host Components for the Remote System Update Design Example

5.6.2. 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). Generating the .rpd file is optional.
Figure 76. Creating Initial Flash Image
On the Input Files tab, click Add Bitstream, select the factory image .sof file and click Open. Repeat this step for the application image .sof.
1. Bitstream_1 is the bitstream for factory image.
2. Bitstream_2 is the bitstream for application image.
Figure 77. Input Files Tab: Specifying the .sof
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 the Page 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 1SX280LU2 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
Figure 78. Configuration Tab: Add Device, Partition, Flash Loader and Generate

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

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.

5.6.3. Programming Flash Memory with the Initial Remote System Update Image

You can program the initial remote system update image from the command line. In the following command, substitute your .jic for output_file.jic if necessary.

quartus_pgm -c 1 -m jtag -o "pvi;./output_file.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 (output_file.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.

Figure 79. Programming the Flash Memory with the Initial RSU Image

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 the start address for the application image printed in the .map file.
Figure 80. Running Tcl Commands Available in rsu1.tcl

5.6.4. 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
    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
    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.
Figure 81. Verify Current Image Using rsu_status Command
```
$ source rsu1.tcl
/channels/local/top/master_1
$ rsu_status
current image address 0x00110000
first failing image address 0x00000000
failing code 0x00000000
error location 0x00000000
0x00000000
```

5.6.5. 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.
Figure 82. Program New Application Image
$ source rsu1.tcl /channels/local/top/master_1 $ program_flash new_application_image.rpd 0x03ff0000 1024 total number of words is 458752 total number of page is 448 total number of sector is 28 reading rpd is completed start erasing flash erasing flash is completed start writing flash writing flash is completed
Write the new application image start address to a new image pointer entry in the configuration firmware pointer block (CPB) using the QSPI_WRITE command. Ensure that the new image pointer entry value is 0xFFFFFFFF before initiating the write.

Note: When using HPS to manage RSU, you must update both copies of the Configuration Pointer Block (CPB0 and CPB1) and the sub-partition table (SPT). In a non-HPS case, while updates to both copies of the pointer blocks are mandatory, the updates to the sub-partition table are not required. For more details about the SPT and CPB, refer to Table 1 for the sub-partition table layout and Table 1 for the pointer block layout.

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 entry value must be 0xFFFFFFFF before you write the start address of the new application image to the next image pointer entry.

Table 50. Configuration Firmware Pointer Block Contents
CPB Start Address + `0x20`	Content	Value
CPB0 + `0x20` = `0x002E4020`	Current application image pointer entry (highest priority)	`0x002F4000`
CPB0 + `0x28` = `0x002E4028`	Next image pointer entry	`0xFFFFFFFF`
CPB1 + `0x20` = `0x002EC020`	Current application image pointer entry (highest priority)	`0x002F4000`
CPB1 + `0x28` = `0x002EC028`	Next image pointer entry	`0xFFFFFFFF`

Table 51. Configuration Firmware Pointer Block Contents
CPB Start Address + `0x20`	Content	Value
CPB0 + `0x20` = `0x004A0020`	Current application image pointer entry (highest priority)	`0x004B0000`
CPB0 + `0x28` = `0x004A0028`	Next image pointer entry	`0xFFFFFFFF`
CPB1 + `0x20` = `0x004A8020`	Current application image pointer entry (highest priority)	`0x004B0000`
CPB1 + `0x28` = `0x002EC028`	Next image pointer entry	`0xFFFFFFFF`

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

Start address
Number of words to read

Figure 83. Verifying that the New Image Pointer entry Value is 0xFFFFFFFF

$ qspi_read 0x002e4020 1
0x002f4000
$ qspi_read 0x002e4028 1
0xffffffff
% qspi_read 0x002ec020 1
ISR is empty
0x002f40000
% qspi_read 0x002ec028 1
0xffffffff

You can now proceed to write the new application image address to next image entry 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 84. Writing an Address Pointer to the New Image Pointer Entry

% qspi_write_one_word 0x002e4028 0x03ff0000
% qspi_write_one_word 0x002ec028 0x03ff

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

Verifying the Update to the New Image Pointer

% qspi_read 0x002e4028 1
0x03ff0000
% qspi_read 0x002ec028 1
0x03ff0000

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 entries in reverse order. The new image has the highest priority when you restart the device.

5.6.6. 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 entry 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 entry 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 entries for CPB0 and CPB1 for offsets 0x20 and 0x28 :

Table 52. Configuration Firmware Pointer Block Contents
CPB Start Address + `0x20`	Content	Value
CPB0 + `0x20` = `0x002E4020`	Old application image pointer entry (lower priority)	`0x002F4000`
CPB0 + `0x28` = `0x002E4028`	Current/new application image pointer entry (highest priority)	`0x03FF0000`
CPB1 + `0x20` = `0x002EC020`	Old application image pointer entry (lower priority)	`0x002F4000`
CPB1 + `0x28` = `0x002EC028`	Current/New application image pointer entry (highest priority)	`0x03FF0000`

Figure 85. Read Current CPB Values

% qspi_read 0x002e4020 1
0x002f400
% qspi_read 0x002e4028 1
0x03ff0000
% qspi_read 0x002ec020 1
0x002f4000
% qspi_read 0x002ec028 1
ISR is empty
0x03ff0000

You can now remove the current or new application image address image pointer entry 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.

Figure 86. Remove Application Image

% qspi_write_one_word 0x002e4028 0x00000000
% qspi_write_one_word 0x002ec028 0x00000000

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

Figure 87. Verify the Writes

% qspi_read 0x002e4028 1
% qspi_read 0x002ec028 1

Figure 88. Verify the Writes

% qspi_read 0x004A0028 1
% qspi_read 0x004A8028 1

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.

6. Intel Stratix 10 Configuration Features

6.1. 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

6.2. 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 89. 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 received 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.

6.3. 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.

7. Intel Stratix 10 Debugging Guide

This section describes best practices in debugging configuration issues on SDM devices. View the video guide below for a quick walk-through.

7.1. Configuration Debugging Checklist

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

Table 53. 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`, `PWRMGT_SDA`, `PWRMGT_SCL`).	☐
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	Verify that selected clocks match the frequency setting specified in the Intel^® Quartus^® Prime software during configuration.	☐
13	Does your design include the Reset Release IP?	☐
14	To avoid configuration failures, disconnect the `PMBus` regulator’s JTAG download cable before configuring Intel^® Stratix^® 10 -V devices.	☐
15	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.	☐
16	Is the configuration clock source chosen appropriately? You can use an internal oscillator or the `OSC_CLK_1` pin.	☐
17	For designs driving the `OSC_CLK_1` pin is the frequency 25, 100, or 125 MHz?	☐
18	For Intel^® Stratix^® 10 SX parts ensure that the HPS and EMIF IOPLL reference clocks are stable and free running before configuration begins. The actual frequency should match the setting specified in Platform Designer.	☐
19	Are proper slave addresses set for the PMBus voltage regulator modules using the Intel^® Quartus^® Prime Software?	☐
20	For designs that use 3 V I/0, verify that the transceiver tiles are powered up before configuration begins.	☐

7.2. 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.

7.3. Understanding Configuration Status Using quartus_pgm command

You can read the device configuration status from the command line directly. In the following command, specify the cable index and the JTAG chain device index to obtain the configuration status.

quartus_pgm -c <cable index> -m jtag -device=<device order in jtag chain> -status

The following example output shows the device status with cable index set to 1 and the second device selected as the targeted device in the JTAG chain after a successful configuration.

quartus_pgm -c 1 -m jtag -device=2 -status

Info (21597): Response of CONFIG_STATUS    
   Device is running in user mode
   00006000  RESPONSE_CODE=OK, LENGTH=6
   00000000  STATE=IDLE
   00000000  Version
   C000000F  MSEL=JTAG, nSTATUS=1, nCONFIG=1
   00000003  CONF_DONE=1, INIT_DONE=1, CVP_DONE=0, SEU_ERROR=0
   00000000  Error location
   00000000  Error details

7.4. 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 90. 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.
Figure 91. Temperature Measurements

HPS Reset Control

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

Figure 92. HPS Reset Options

QSPI Flash

Provides options to detect, read, erase, and program QSPI flash memory device.

Auto-detect QSPI Flash option provides the flash memory device information.
Read Registers option provides the current value of Read/Write memory operations.
Erase Sectors provides option to erase specific memory sectors.
Load RPD File provides option to read flash image and save data in a .rpd file.
Program RPD provides option to program a flash image.

Figure 93. QSPI Flash Options

7.4.1. 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. For information about decoding the State output value, refer to the CONFIG_STATUS command.

Figure 94. Read Configuration Status Command

7.5. 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 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 - 115 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 125 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]`	`SDM_IO8` (`AS_nCSO3`) `SDM_IO7` (`AS_nCSO2`)`SDM_IO9` (`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_nCSO0`-`AS_nCSO3`, `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`

7.6. 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 95. 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. The version of firmware that the Programmer adds depends on the version of the Programmer you are using.

7.7. 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.

7.8. Reading the Unique 64-Bit CHIP ID

The Chip ID Intel^® FPGA IP in each Intel^® Stratix^® 10 device stores a unique 64-bit chip ID. After the Chip ID Intel^® FPGA IP receives a valid clock input and readid signal 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 User Guide.

7.9. E-Tile Transceivers May Fail To Configure

Making the PRESERVE_UNUSED_XCVR_CHANNEL assignment to completely unused E-tile transceivers may cause configuration failures in Intel^® Stratix^® 10 TX or MX devices.

The Intel^® Quartus^® Prime Programmer detects an internal error and fails to configure your device under the following conditions:

You have made the PRESERVE_UNUSED_XCVR_CHANNEL assignment to an entire unused E-tile.
Your design does not provide a reference clock to this unused E-tile.

The reference clock is necessary to generate a pseudo-random data signal to prevent the transceiver from degrading over time. You must instantiate at least one dummy channel in the E-tile using the Native PHY IP GUI. Provide this channel at least one reference clock. All preserved channels in a single E-tile can use the same reference clock.

When your design uses some channels in an E-tile, you can use the per-pin PRESERVE_UNUSED_XCVR_CHANNEL QSF assignment to preserve only the channels in the E-tile that you intend to use. If you never intend to use a channel, you should not add the per-pin PRESERVE_UNUSED_XCVR_CHANNEL QSF assignment.

Here are some examples of PRESERVE_UNUSED_XCVR_CHANNEL QSF assignments.

#Global QSF assignment
set_global_assignment -name PRESERVE_UNUSED_XCVR_CHANNEL ON 

#Per-pin QSF assignment 
set_instance_assignment -name PRESERVE_UNUSED_XCVR_CHANNEL ON -to AA75

7.10. 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.

8. Intel Stratix 10 Configuration User Guide Archives

If an Intel^® Quartus^® Prime version is not listed, the user guide for the previous Intel^® Quartus^® Prime version applies.

Intel^® Quartus^® Prime Version	User Guide
21.1	Intel^® Stratix^® 10 Configuration User Guide
20.4	Intel^® Stratix^® 10 Configuration User Guide
20.3	Intel^® Stratix^® 10 Configuration User Guide
20.2	Intel^® Stratix^® 10 Configuration User Guide
20.1	Intel^® Stratix^® 10 Configuration User Guide
19.4	Intel^® Stratix^® 10 Configuration User Guide
19.3	Intel^® Stratix^® 10 Configuration User Guide
19.2	Intel^® Stratix^® 10 Configuration 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

9. Document Revision History for the Intel Stratix 10 Configuration User Guide

Document Version	Intel^® Quartus^® Prime Version	Changes
2021.06.21	21.2	Made the following changes: Added a CvP-related note in the Intel^® Stratix^® 10 Configuration Overview. Revised block diagram descriptions in the Intel^® Stratix^® 10 Configuration Architecture. Revised Secure Device Manager. Added recommendation to use -BK ordering part number suffix for Intel^® Stratix^® 10 devices with the black key provisioning feature. Added new section: Restricting Security Features Revised Intel^® Stratix^® 10 Configuration Timing Diagram. Added a note in the Reconfiguration Timing section. Re-ordered sections for clarity. Revised Intel^® Stratix^® 10 Configuration Flow Diagram section. Renamed Power Up section to Power-On to align the description with the figure. Merged Configuration Start and Configuration Pass sections into the FPGA Configuration section. Renamed Configuration Error section to Failed FPGA Configuration. Minor Re-ordered sections for clarity. Removed JTAG Configuration section. Reposition the existing JTAG configuration note. Moved device response content to a new section: Device Response to Configuration and Reset Events. Revised text and figure in SDM I/O Pins for Power Management and SmartVID Revised `OSC_CLK_1` requirements in OSC_CLK_1 Clock Input Added new PFL II IP-related topics: PFL II IP Recommended Constraints for Other Input Pins PFL II IP Recommended Constraints for Other Output Pins Revised AS_CLK topic: Updated `OS_CLK_1` description in the Supported Configuration Clock Source and `AS_CLK` Frequencies in Intel^® Stratix^® 10 Devices table. Added table's description. Added note to clarify configuration behavior for invalid `AS_CLK` setting. Reworded attention note in the Remote System Update Using AS Configuration section. Revised the steps for image update in the Updates with the Factory Update Image section. Revised Command List and Description table. Updated description for: `RSU_STATUS` `QSPI_OPEN` `QSPI_SET_CS` `QSPI_ERASE` Revised RSU Image Layout in Flash - SDM Perspective. Updated `max_retry` parameter value description. Revised step 2 in the Command Sequence To Perform Quad SPI Operations. The `QSPI_SET_CS*` command is optional for the AS x4 configuration and mandatory for the JTAG configuration scheme. Added new topic: Firmware Version Information Clarified usage of Use relative address option in the Application Image Layout and Generating an Application Image sections. Revised Updating Decision Firmware. Added statement about updating decision firmware using a combined application image. Added a new video guide about debugging SDM-related configuration issues in Intel^® Stratix^® 10 Debugging Guide. Updated the following figures and diagrams: Intel^® Stratix^® 10 Configuration Interfaces Power-On, Configuration, and Reconfiguration Timing Diagram Recoverable Error during Reconfiguration Timing Diagram Intel^® Stratix^® 10 FPGA Configuration Flow SDM I/O pins selection in the Specifying Optional Configuration Pins section Configuration Pin Selection in the Intel^® Quartus^® Prime Pro Edition Software Dual-purpose pins selection in the Enabling Dual-Purpose Pins section Specifying the Slave Device Type for Power Management and VID Specifying the Page Command Setting Configuration clock source selection in the Setting Configuration Clock Source section AS configuration scheme setting in the Active Serial Configuration Software Settings section Corrected minor errors and spelling mistakes.
2021.03.29	21.1	Made the following changes: Globally removed support for 133 MHz and 108 MHz `AS_CLK` frequencies. Updated the maximum `AS_CLK` clock rate from 133 MHz to 125 MHz. Updated the maximum `AS_CLK` data rate from 532 MHz to 500 MHz. Updated MSEL Settings topic. Updated footnote for AS Fast mode. To support this mode, all power supplies must ramp-up to the recommended operating condition within 10 ms. Added footnote for AS Normal mode. To support the mode, the `V_{CCIO_SDM}` supply must ramp-up to the recommended operation condition within 10 ms. Revised the Maximum Configuration Time Estimation section. Added hyper initialization description. Restructured PFL II IP content in the Avalon^® -ST Configuration chapter. Added statement in The AVST_READY Signal. The PFL II IP core includes the `AVST_READY` synchronizer logic if you use PFL II IP core as the configuration host. Added note in the PFL II IP Functional Description. The PFL II IP does not support HPS cold reset. Added new topics:

Avalon-ST

JTAG

CvP

AS Normal Mode

AS Fast Mode

Power-On Status

Initial Configuration Timing

Reconfiguration Timing

Recoverable Configuration Error

Unrecoverable Configuration Error

Power-On

SDM Startup

Idle

FPGA Configuration

Failed FPGA Configuration

User Mode

Device Clean

FPGA Configuration

HPS First 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

Configuration Clock Requirements for Reconfiguration Without Power Cycling the Device

Configuration Clock Requirements for Configuration After Powering Cycling the Device

Notes for Figure:

Debugging Suggestions

Create a pfl_clk clock and a generated AVST_CLK clock

Set output delay for PFL II IP output pins

Setting a false path

Create a FLASH_CLK clock

Set output delay for PFL II IP output pins

Set input delay for input pins

Create a FLASH_CLK clock

Set output delay for output pins

Set input delay for input pins

Set a false path for PFL II IP input pins

Set input delay to PFL II IP input pins

Set a false path for fpga_pgm[] input pin

Set input delay to pfl_nreconfigure input pin

Set input delay to pfl_reset_watchdog pin

Set output delay to PFL II IP output pins

Set output delay to flash_nreset output pin

Set a false path for fpga_nconfig output pin

Set output delay to pfl_watchdog_error output pin

MSEL Pin Function for the AS x4 Configuration Scheme

Debugging Suggestions

Debugging Suggestions

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

Create a `pfl_clk` clock and a generated `AVST_CLK` clock

Create a `FLASH_CLK` clock

Create a `FLASH_CLK` clock

Set a false path for `fpga_pgm[]` input pin

Set input delay to `pfl_nreconfigure` input pin

Set input delay to `pfl_reset_watchdog` pin

Set output delay to `flash_nreset` output pin

Set a false path for `fpga_nconfig` output pin

Set output delay to `pfl_watchdog_error` output pin