Raspberry Pi hardware - Raspberry Pi Documentation
Raspberry Pi hardware - Raspberry Pi Documentation
We use optional cookies, as detailed in our cookie policy, to remember your settings and understand how you use our website.
Introduction
mechanical drawings
Edit this on GitHub
Computers
Product compliance and
Frequency managemen
Raspberry Pi OS The Flagship series, often referred to by the shorthand "Raspberry Pi", offers high-
and thermal control
performance hardware, a full Linux operating system, and a variety of common ports in
Configuration
a form factor roughly the size of a credit card. Raspberry Pi boot EEPR
The Zero series offers a full Linux operating system and essential ports at an affordable USB mass storage boot
The Linux kernel
price point in a minimal form factor with low power consumption. Network booting
Remote access GPIO boot mode
The Compute Module series, often referred to by the shorthand "CM", offers high-
NVMe SSD boot
Camera software performance hardware and a full Linux operating system in a minimal form factor
suitable for industrial and embedded applications. Compute Module models feature HTTP boot
AI Kit and AI HAT+
hardware equivalent to the corresponding flagship models, but with fewer ports and no Boot sequence
software
on-board GPIO pins. Instead, users should connect Compute Modules to a separate EEPROM boot flow
Raspberry Pi hardware baseboard that provides the ports and pins required for a given application. Raspberry Pi bootloade
Introduction Additionally, Raspberry Pi makes the Pico series of tiny, versatile microcontroller boards.
configuration
Pico models do not run Linux or allow for removable storage, but instead allow Display Parallel Interfac
Flagship series
programming by flashing a binary onto on-board flash storage. (DPI)
Keyboard series GPIO and the 40-pin hea
For more information about the ports on the Raspberry Pi flagship series, see the
Schematics and mechanical drawings.
Keyboard series
Keyboard series devices use model identifiers of the form <X00>, where X indicates the
corresponding Flagship series device. For instance, "Raspberry Pi 500" is the keyboard
version of the Raspberry Pi 5.
Zero series
Models with the H suffix have header pins pre-soldered to the GPIO header. Models that
lack the H suffix do not come with header pins attached to the GPIO header; the user must
solder pins manually or attach a third-party pin kit.
2× micro USB ports (one for input power, one for external devices)
Since version 1.3 of the original Zero, all Zero models also include:
a mini 22-pin, 0.5mm (fine) pitch, 11.5mm width, CSI (camera) port
Raspberry Pi Zero
Raspberry Pi Zero W
Model SoC Memory GPIO Wireless Connectivity
Raspberry Pi Zero WH
Raspberry Pi Zero 2 W
Raspberry Pi Zero 2 WH
NOTE
Compute Modules that use the physical DDR2 SO-DIMM form factor are not
compatible with DDR2 SO-DIMM electrical specifications.
For more information about Raspberry Pi Compute Modules, see the Compute Module
documentation.
Pico microcontrollers
Models with the H suffix have header pins pre-soldered to the GPIO header. Models that
lack the H suffix do not come with header pins attached to the GPIO header; the user must
solder pins manually or attach a third-party pin kit.
Raspberry Pi Pico
Model SoC Memory Storage GPIO Wireless
Connectivity
Raspberry Pi Pico H
Raspberry Pi Pico W
Raspberry Pi Pico WH
Raspberry Pi Pico 2
Raspberry Pi Pico 2 W
For more information about Raspberry Pi Pico models, see the Pico documentation.
Raspberry Pi 5
Mechanical drawings, PDF
Raspberry Pi 4 Model B
Schematics, revision 4.0
Raspberry Pi 3 Model B+
Schematics, revision 1.0
Raspberry Pi 3 Model A+
Schematics, revision 1.0
Raspberry Pi 3 Model B
Schematics, revision 1.2
Raspberry Pi 2 Model B
Schematics, revision 1.2
Raspberry Pi 1 Model B+
Schematics, revision 1.2
Raspberry Pi 1 Model A+
Schematics, revision 1.1
NOTE
Mechanical drawings for the Raspberry Pi 3 Model A+ are also applicable to the
Raspberry Pi 1 Model A+.
Raspberry Pi Zero 2 W
Schematics
The Raspberry Pi Zero 2 W has a number of test pad locations used during production of
the board.
Label Function X (mm from origin) Y (mm from origin)
Raspberry Pi Zero W
Schematics, revision 1.1
Raspberry Pi Zero
Schematics, revision 1.3
All Raspberry Pi products have undergone extensive compliance testing. For more
information see the Product Information Portal.
Flammability rating
The PCBs used in Raspberry Pi devices adhere to UL94-V0.
NOTE
Powered by Raspberry Pi
The Powered by Raspberry Pi program provides a process for companies wanting to use a
form of the Raspberry Pi logo, and covers products with Raspberry Pi computers or silicon
inside, and services provided by a Raspberry Pi. If you wish to start the process to apply
you can do so online.
When the core temperature is between 80°C and 85°C, the Arm cores will be progressively
throttled back. If the temperature reaches 85°C, both the Arm cores and the GPU will be
throttled back.
For Raspberry Pi 3 Model B+, the PCB technology has been changed to provide better heat
dissipation and increased thermal mass. In addition, a soft temperature limit has been
introduced, with the goal of maximising the time for which a device can "sprint" before
reaching the hard limit at 85°C. When the soft limit is reached, the clock speed is reduced
from 1.4GHz to 1.2GHz, and the operating voltage is reduced slightly. This reduces the rate
of temperature increase: we trade a short period at 1.4GHz for a longer period at 1.2GHz.
By default, the soft limit is 60°C, and this can be changed via the temp_soft_limit setting
in config.txt.
The Raspberry Pi 4 Model B continues with the same PCB technology as the Raspberry Pi
3 Model B+, to help dissipate excess heat. There is currently no soft limit defined.
Use DVFS
NOTE
Discussion of DVFS applies to 4-series devices only (Raspberry Pi 4, Compute Module
4, and Pi 400).
Raspberry Pi 4 devices implement dynamic voltage and frequency scaling (DVFS). This
technique allows 4-series devices to run at lower temperatures whilst still providing the
same performance.
Various clocks (e.g. Arm, Core, V3D, ISP, H264, HEVC) inside the SoC are monitored by the
firmware, and whenever they are not running at full speed, the voltage supplied to the
particular part of the chip driven by the clock is reduced relative to the reduction from full
speed. In effect, only enough voltage is supplied to keep the block running correctly at the
specific speed at which it is running. This can result in significant reductions in power used
by the SoC, and therefore in the overall heat being produced.
Due to possible system stability problems involved with running an undervoltage, especially
when using undervoltaged fixed clock peripherals (eg. PCIe), three DVFS modes are
available and can be configured in /boot/firmware/config.txt with the below properties.
Most systems should use dvfs=3, headless systems may benefit from a small power
reduction with dvfs=1 at the risk of PCIe stability issues.
property=value Description
NOTE
This setting has been removed on 5-series devices and is effectively always mode 3.
In addition, a more stepped CPU governor is also used to produce finer-grained control of
ARM core frequencies, which means the DVFS is more effective. The steps are now
1500MHz, 1000MHz, 750MHz, and 600MHz. These steps can also help when the SoC is
being throttled, and mean that throttling all the way back to 600MHz is much less likely,
giving an overall increase in fully loaded performance.
The default CPU governor is ondemand. The governor can be manually changed with the
cpufreq-set command (from the cpufrequtils package) to reduce idle power
consumption:
Measure temperatures
Due to the architecture of the SoCs used on Raspberry Pi devices, and the use of the
upstream temperature monitoring code in the Raspberry Pi OS distribution, Linux-based
temperature measurements can be inaccurate. However, the vcgencmd command provides
an accurate and instantaneous reading of the current SoC temperature, as it
communicates with the GPU directly:
$ vcgencmd measure_temp
Fan cases
To ensure the best performance for your Raspberry Pi, use an active cooling solution such
as a fan. Raspberry Pi firmware manages fan speeds for all official fans.
Raspberry Pi 4 fan
For Raspberry Pi 4, add the Raspberry Pi 4 Case Fan to the lid of the Raspberry Pi 4 case.
Raspberry Pi 5 fans
For Raspberry Pi 5, use one of the official fan options:
Active Cooler
Both of the Raspberry Pi 5 fan options plug into the four-pin JST-SH PWM fan connector
located in the upper right of the board between the 40-pin GPIO header and the USB 2
ports. The fan connector pulls from the same current limit as USB peripherals. We
recommend the Active Cooler case for overclockers, since it provides better cooling
performance.
As the temperature of the Raspberry Pi 5 increases, the fan reacts in the following way:
below 50°C, the fan does not spin at all (0% speed)
Temperature decreases use the same mapping with a 5°C hysteresis; fan speed decreases
when the temperature drops to 5°C below each of the above thresholds.
At boot the fan is turned on, and the tachometer input is checked to see if the fan is
spinning. If it is, then the cooling_fan device tree overlay is enabled. This overlay is in
bcm2712-rpi-5-b.dtb by default, but with status=disabled.
The Raspberry Pi 5 fan connector is a 1mm pitch JST-SH socket containing the following
four pins:
1 +5V Red
2 PWM Blue
3 GND Black
4 Tach Yellow
All other models of Raspberry Pi computer use the bootcode.bin file located in the boot
filesystem.
NOTE
You can find the scripts and pre-compiled binaries used to create rpi-eeprom in the
rpi-eeprom GitHub repository.
Diagnostics
If an error occurs during boot, then an error code will be displayed via the green LED. Newer
versions of the bootloader will display a diagnostic message on all HDMI displays.
Flagship models since Raspberry Pi 4B; Compute Modules since CM5; Keyboard
models since Pi 400
Raspberry Pi OS automatically updates the bootloader for important bug fixes. To manually
update the bootloader or change the boot order, use raspi-config.
NOTE
Raspberry Pi Imager provides a GUI for updating the bootloader and selecting the boot
mode.
4. Select Choose OS
6. Select Bootloader for your version of Raspberry Pi (Pi 400 is part of the 4 family)
10. Boot the Raspberry Pi with the new image and wait for at least ten seconds
11. When the green activity LED blinks with a steady pattern and the HDMI display shows a
green screen, you have successfully written the bootloader
To change the boot-mode or bootloader version from within Raspberry Pi OS, run raspi-
config.
5. Select Default for factory default settings or Latest for the latest bootloader release.
Advanced users can switch to the latest bootloader to get the latest functionality.
First, ensure that your Raspberry Pi runs the latest software. Run the following command
to update:
$ sudo raspi-config
Navigate to Advanced Options > Bootloader Version. Select Latest, then choose Yes
to confirm. Select Finish and confirm that you want to reboot.
If you run sudo rpi-eeprom-update, you should see that a more recent version of the
bootloader is available and it’s the latest release.
$ sudo rpi-eeprom-update -a
$ sudo reboot
Reboot, then run sudo rpi-eeprom-update. You should now see that the CURRENT date
has updated to the latest version of the bootloader:
BOOTLOADER: up to date
CURRENT: Mon 22 Jan 10:41:21 UTC 2024 (1705920081)
LATEST: Mon 22 Jan 10:41:21 UTC 2024 (1705920081)
RELEASE: latest (/lib/firmware/raspberrypi/bootloader-2711/latest)
Use raspi-config to change the release.
To view the configuration used by the current running bootloader, run the following
command:
$ rpi-eeprom-config
$ rpi-eeprom-config pieeprom.bin
The following command loads the current bootloader configuration into a text editor. When
the editor is closed, rpi-eeprom-config applies the updated configuration to latest
available bootloader release and uses rpi-eeprom-update to schedule an update when
the system is rebooted:
The following command applies boot.conf to the latest available bootloader image and
uses rpi-eeprom-update to schedule an update when the system is rebooted.
Automatic updates
The rpi-eeprom-update systemd service runs at startup and applies an update if a new
image is available, automatically migrating the current bootloader configuration.
NOTE
If the FREEZE_VERSION bootloader config is set then the update service will skip any
automatic updates. This removes the need to individually disable the update service if
there are multiple operating systems installed, or when swapping SD cards.
rpi-eeprom-update
$ vcgencmd bootloader_version
$ sudo rpi-eeprom-update
$ sudo rpi-eeprom-update -a
$ sudo reboot
The -d flag instructs rpi-eeprom-update to use the configuration in the specified image
file instead of automatically migrating the current configuration.
$ rpi-eeprom-update -h
default - Updated for new hardware support, critical bug fixes and periodic update for
new features that have been tested via the latest release
Since the release status string is just a subdirectory name, it is possible to create your own
release streams e.g. a pinned release or custom network boot configuration.
NOTE
You can change which release stream is to be used during an update by editing the
/etc/default/rpi-eeprom-update file and changing the FIRMWARE_RELEASE_STATUS
entry to the appropriate stream.
recovery.bin
At power on, the ROM found on BCM2711 and BCM2712 looks for a file called
recovery.bin in the root directory of the boot partition on the SD card. If a valid
recovery.bin is found then the ROM executes this instead of the contents of the
EEPROM. This mechanism ensures that the bootloader flash image can always be reset to
a valid image with factory default settings.
Filename Purpose
The .sig files contain the hexadecimal sha256 checksum of the corresponding image
file; additional fields may be added in the future.
The ROM found on BCM2711 and BCM2712 does not support loading recovery.bin
from USB mass storage or TFTP. Instead, newer versions of the bootloader support a
self-update mechanism where the bootloader is able to reflash the SPI flash itself. See
ENABLE_SELF_UPDATE on the bootloader configuration page.
The temporary EEPROM update files are automatically deleted by the rpi-eeprom-
update service at startup.
For more information about the rpi-eeprom-update configuration file see rpi-eeprom-
update -h.
Both the bootloader and VLI EEPROMs support hardware write protection. See the
eeprom_write_protect option for more information about how to enable this when flashing
the EEPROMs.
Boot diagnostics
Edit this on GitHub
This diagnostics page will also appear if the bootloader is unable to boot from any boot
media or network boot. This can happen if there is no bootable image on the boot media, if
the boot media is defective, or if network boot parameters are incorrect.
To reboot while displaying the diagnostics page, power cycle the device. You can
disconnect, then reconnect the power supply, or press and hold the power button, if your
device has one.
The top line describes the model of Raspberry Pi and its memory capacity. The QR code is
a link to the downloads page.
Line Information
bootloader Bootloader git version - RO (if EEPROM is write protected) - software build
date
boot mode (current boot mode name and number) order (the BOOT ORDER
configuration) retry (retry count in the current boot mode) restart (number
of cycles through the list of boot modes)
net Network boot: link status (up/down), client IP address (ip), subnet (sn),
default gateway (gw)
display Indicates whether hotplug was detected (HPD=1) and if so whether the
EDID was read successfully (EDID=ok) for each HDMI output
To disable this diagnostics display, use the DISABLE_HDMI option in the bootloader
configuration.
The Raspberry Pi has a number of different stages of booting. This document explains how
the boot modes work, and which ones are supported for Linux booting.
NOTE
Since Raspberry Pi 4, flagship devices do not use the bootcode.bin file. Instead, these
devices use a bootloader located in an on-board EEPROM chip. For more information,
see the documentation on EEPROM bootflow and SPI boot EEPROM.
Format an SD card as FAT32 and copy over the latest bootcode.bin. The SD card must be
present in the Raspberry Pi for it to boot. Once bootcode.bin is loaded from the SD card,
the Raspberry Pi continues booting using USB host mode.
This is useful for the Raspberry Pi 1, 2, and Zero models, which are based on the BCM2835
and BCM2836 chips, and in situations where a Raspberry Pi 3 fails to boot (the latest
bootcode.bin includes additional bugfixes for the Raspberry Pi 3B, compared to the boot
code burned into the BCM2837A0).
If you have a problem with a mass storage device still not working, even with this
bootcode.bin, then add a new file called "timeout" to the SD card. This will extend to six
seconds the time for which it waits for the mass storage device to initialise.
For information on enabling UART with the EEPROM bootloader, see the bootloader
configuration documentation.
It is possible to enable an early stage UART to debug booting issues (useful with the above
bootcode.bin only boot mode). To do this, make sure you’ve got a recent version of the
firmware (including bootcode.bin). To check if UART is supported in your current
firmware:
$ strings bootcode.bin | grep BOOT_UART
Next, connect a suitable USB serial cable to your host computer (a Raspberry Pi will work,
although you may find that the easiest path is to use a USB serial cable, since it’ll work out
the box without any pesky config.txt settings). Use the standard pins 6, 8 and 10 (GND,
GPIO14, GPIO15) on a Raspberry Pi or Compute Module.
Then use screen on Linux or macOS or putty on Windows to connect to the serial.
Set up your serial to receive at 115200-8-N-1, and then boot your Raspberry Pi. You should
get an immediate serial output from the device as bootcode.bin runs.
The firmware chooses between the two modes at boot time based on the OTP bits. Two
bits control USB boot. The first enables USB device boot and is enabled by default; the
second enables USB host boot.
If the USB host boot mode bit is set, the processor reads the OTGID pin to decide whether
to boot as a host (driven to zero as on any Raspberry Pi Model B/B+) or as a device (left
floating). The Raspberry Pi Zero has access to the OTGID pin through the USB connector;
the Compute Module has access to the OTGID pin on the edge connector.
Some other OTP bits allow certain GPIO pins to select the boot modes.
USB device boot is available on the Compute Module series, Zero series, and Model A
variants of the flagship series.
When this boot mode is activated (usually after a failure to boot from the SD card), the
Raspberry Pi puts its USB port into device mode and awaits a USB reset from the host.
Example code showing how the host needs to talk to the Raspberry Pi can be found on
Github.
The host first sends a structure to the device down control endpoint 0. This contains the
size and signature for the boot (security is not enabled, so no signature is required).
Secondly, code is transmitted down endpoint 1 (bootcode.bin). Finally, the device will
reply with one of the following codes:
0 - Success
0x80 - Failure
Host boot is available on the Compute Module series since Compute Module 3, Zero
series since Zero 2 W, Raspberry Pi 2B (version 1.2), Raspberry Pi 3B, and all flagship
series devices since Raspberry Pi 3B+. Raspberry Pi 3A+ supports mass storage boot,
but not network boot.
1. Enable the USB port and wait for D+ line to be pulled high indicating a USB 2.0 device
(we only support USB2.0)
i. Release from reset and wait for D+ to be driven high to indicate that a device is
connected
WARNING
Any change you make to the OTP is permanent and cannot be undone.
On Raspberry Pi 3A+, setting the OTP bit to enable USB host boot mode will permanently
prevent that Raspberry Pi from booting in USB device mode.
NOTE
Available on the Compute Module series since Compute Module 3, Zero series since
Zero 2 W, and all flagship series devices since Raspberry Pi 2B (version 1.2).
USB mass storage boot enables you to boot your Raspberry Pi from a USB mass storage
device such as a flash drive or USB disk. When attaching USB devices, particularly hard
disks and SSDs, be mindful of their power requirements. Attaching more than one disk
typically requires additional external power from either a powered disk enclosure or a
powered USB hub.
NOTE
Models prior to Raspberry Pi 4B have known issues which prevent booting with some
USB devices.
NOTE
Early editions of Raspberry Pi 4 may require a bootloader update to boot from USB.
NOTE
Early editions of Compute Module 4 may require a bootloader update to boot from USB.
Raspberry Pi 3B+
The Raspberry Pi 3B+ supports USB mass storage boot out of the box.
NOTE
To enable USB host boot mode on these devices, set the USB host bit in OTP (one-time
programmable) memory. To set the bit, boot from an SD card where
/boot/firmware/config.txt contains the line program_usb_boot_mode=1. Once you set
the bit, you can boot from USB without the SD card.
WARNING
Any change you make to OTP (one-time programmable) memory is permanent and cannot
be undone.
On Raspberry Pi 3A+, setting the OTP bit to enable USB host boot mode will permanently
prevent that Raspberry Pi from booting in USB device mode.
Use any SD card flashed with Raspberry Pi OS to program the OTP bit.
To enable USB host boot mode, add the following line to config.txt:
program_usb_boot_mode=1
Then, use sudo reboot to reboot your Raspberry Pi. To check that the OTP has been
programmed correctly, run the following command:
If the output reads 0x3020000a, the OTP has been successfully programmed. If you see
different output, try the programming procedure again. Make sure there is no blank line at
the end of config.txt.
You can now boot from a USB mass storage device in the same way as booting from an
SD card. See the following section for further information.
After preparing the storage device, connect the drive and power up the Raspberry Pi, being
aware of the extra USB power requirements of the external drive.
After five to ten seconds, the Raspberry Pi should begin booting and show the rainbow
splash screen on an attached display. Make sure that you do not have an SD card inserted
in the Raspberry Pi, since if you do, it will boot from that first.
See the boot modes documentation for the boot sequence and alternative boot modes
(network, USB device, GPIO or SD boot).
Known issues
The default timeout for checking bootable USB devices is two seconds. Some flash
drives and hard disks power up too slowly. It is possible to extend this timeout to five
seconds (add a new file timeout to the SD card), but note that some devices take even
longer to respond.
Some flash drives have a very specific protocol requirement that is not handled by the
bootcode and may thus be incompatible.
Hardware compatibility
Before booting from a USB mass storage device, verify that the device works correctly
under Linux. Boot using an SD card and plug in the USB mass storage device. This should
appear as a removable drive. This is especially important with USB SATA adapters, which
may be supported by the bootloader in mass storage mode, but fail if Linux selects USB
Attached SCSI-UAS mode.
Hard disk drives (HDDs) typically require a powered USB hub. Even if everything appears to
work, you may encounter intermittent failures without a powered USB hub.
NOTE
Use config.txt file conditional filters to select alternate firmware in complex device
configurations.
Network booting
Edit this on GitHub
This section describes how network booting works on Raspberry Pi 3B, 3B+ and 2B v1.2.
Network booting works only for the wired adapter built into the above models of Raspberry
Pi. Booting over wireless LAN is not supported, nor is booting from any other wired network
device.
Send DHCP request (with Vendor Class identifier DHCP option 60 set to
PXEClient:Arch:00000:UNDI:002001)
File not found: Server replies with TFTP error response with textual error message
File exists: Server will reply with the first block (512 bytes) of data for the file with a
block number in the header
Raspberry Pi replies with TFTP ACK packet containing the block number, and
repeats until the last block which is not 512 bytes
This will normally result in an error file not found. This is to be expected, and
TFTP boot servers should be able to handle it.
From this point the bootcode.bin code continues to load the system. The first file it will try
to access is <serial_number>/start.elf. If this does not result in an error then any
other files to be read will be prepended with the serial_number. This is useful because it
enables you to create separate directories with separate start.elf / kernels for your
Raspberry Pis.
To get the serial number for the device you can either try this boot mode and see what file
is accessed using tcpdump / wireshark, or you can run a standard Raspberry Pi OS SD card
and cat /proc/cpuinfo.
If you put all your files into the root of your TFTP directory then all following files will be
accessed from there.
If row 17 contains 3020000a then the OTP is correctly programmed. You should now be
able to remove the SD card, plug in Ethernet, and then the Ethernet LEDs should light up
around 5 seconds after the Raspberry Pi powers up.
To capture the Ethernet packets on the server, use tcpdump on the tftpboot server (or
DHCP server if they are different). You will need to capture the packets there otherwise you
will not be able to see packets that get sent directly because network switches are not
hubs!
This will write everything from eth0 to a file named dump.pcap. You can then post-process
or upload the packets to cloudshark for communication.
As a minimum you should see a DHCP request and reply which looks like the following:
Vendor-Option Option 43 contains the important part of the reply. This must contain the
string "Raspberry Pi Boot". Due to a bug in the boot ROM, you may need to add three
spaces to the end of the string.
Known problems
There are a number of known problems with the Ethernet boot mode. Since the
implementation of the boot modes is in the chip itself, there are no workarounds other than
to use an SD card with just the bootcode.bin file.
The Raspberry Pi will attempt a DHCP request five times with five seconds in between, for
a total period of 25 seconds. If the server is not available to respond in this time, then the
Raspberry Pi will drop into a low-power state. There is no workaround for this other than
bootcode.bin on an SD card.
The DHCP check also checked if the hops value was 1, which it wouldn’t be with DHCP
relay.
The "Raspberry Pi Boot " string in the DHCP reply requires the extra three spaces due to an
error calculating the string length.
Fixed in Raspberry Pi 3 Model B+; the value is set to the 32-bit serial number.
The Raspberry Pi will only respond to ARP requests when it is in the initialisation phase;
once it has begun transferring data, it’ll fail to continue responding.
At boot time, Raspberry Pi broadcasts a DHCPDISCOVER packet. The DHCP server replies
with a DHCPOFFER packet. The Raspberry Pi then continues booting without doing a
DHCPREQUEST or waiting for DHCPACK. This may result in two separate devices being
offered the same IP address and using it without it being properly assigned to the client.
Different DHCP servers have different behaviours in this situation. dnsmasq (depending
upon settings) will hash the MAC address to determine the IP address, and ping the IP
address to make sure it isn’t already in use. This reduces the chances of this happening
because it requires a collision in the hash.
GPIO boot mode
Edit this on GitHub
NOTE
GPIO boot mode is only available on the Raspberry Pi 3A+, 3B, 3B+, Compute Module 3
and 3+.
Earlier Raspberry Pis can be configured to allow the boot mode to be selected at power-on
using hardware attached to the GPIO connector. This is done by setting bits in the OTP
memory of the SoC. Once the bits are set, they permanently allocate five GPIOs to allow
this selection to be made. Once the OTP bits are set, they cannot be unset. You should
think carefully about enabling this, since those five GPIO lines will always control booting.
Although you can use the GPIOs for some other function once the Raspberry Pi has
booted, you must set them up so that they enable the desired boot modes when the
Raspberry Pi boots.
To enable GPIO boot mode, add the following line to the config.txt file:
program_gpio_bootmode=n
Where n is the bank of GPIOs which you wish to use. Then reboot the Raspberry Pi once to
program the OTP with this setting. Bank 1 is GPIOs 22-26, Bank 2 is GPIOs 39-43. Unless
you have a Compute Module, you must use bank 1: the GPIOs in Bank 2 are only available
on the Compute Module. Because of the way the OTP bits are arranged, if you first program
GPIO boot mode for Bank 1, you then have the option of selecting Bank 2 later. The reverse
is not true: once Bank 2 has been selected for GPIO boot mode, you cannot select Bank 1.
Once GPIO boot mode is enabled, the Raspberry Pi will no longer boot. You must pull up at
least one boot-mode GPIO pin in order for the Raspberry Pi to boot.
Pin assignments
Raspberry Pi 3B and Compute Module 3
22 39 SD0
23 40 SD1
26 43 USB
USB in the table above selects both USB device boot mode and USB host boot mode. In
order to use a USB boot mode, it must be enabled in the OTP memory. For more
information, see USB device boot and USB host boot.
Later Raspberry Pi 3B (BCM2837B0 with the metal lid), Raspberry Pi 3A+, 3B+ and
Compute Module 3+
20 37 SD0
21 38 SD1
24 41 USB device
NOTE
The various boot modes are attempted in the numerical order of the GPIO lines, i.e.
SD0, then SD1, then NAND and so on.
Boot flow
SD0 is the Broadcom SD card/MMC interface. When the boot ROM within the SoC runs, it
always connects SD0 to the built-in microSD card slot. On Compute Modules with an
eMMC device, SD0 is connected to that; on the Compute Module Lite SD0 is available on
the edge connector and connects to the microSD card slot in the CMIO carrier board. SD1
is the Arasan SD card/MMC interface which is also capable of SDIO. All Raspberry Pi
models with built-in wireless LAN use SD1 to connect to the wireless chip via SDIO.
The default pull resistance on the GPIO lines is 50KΩ, as documented on page 102 of the
BCM2835 ARM peripherals datasheet. A pull resistance of 5KΩ is recommended to pull a
GPIO line up: this will allow the GPIO to function but not consume too much power.
NVMe (Non-Volatile Memory express) is a standard for external storage access over a PCIe
bus. You can connect NVMe drives via the PCIe slot on Compute Module 4 IO Board, the
M.2 slot on Compute Module 5 IO Board, and Raspberry Pi 5 using an M.2 HAT+. With
some additional configuration, you can boot from an NVMe drive.
Prerequisites
Hardware
For Raspberry Pi 5, we recommend the M.2 HAT+, which converts from the
Raspberry Pi’s PCIe FFC slot to an M Key interface.
For the CM4, search for a "PCI-E 3.0 ×1 lane to M.2 NGFF M-Key SSD NVMe PCI
Express adapter card"
To check that your NVMe drive is connected correctly, boot your Raspberry Pi from another
storage device (such as an SD card) and run ls -l /dev/nvme*. Example output is shown
below.
Software
First, ensure that your Raspberry Pi runs the latest software. Run the following command
to update:
$ sudo raspi-config
Under Advanced Options > Bootloader Order, specify that the bootloader should
attempt to boot from NVMe first:
$ sudo rpi-eeprom-update -a
Then, reboot with sudo reboot. Your Raspberry Pi should boot from NVMe.
For CM4, use rpiboot to update the bootloader. You can find instructions for building
rpiboot and configuring the IO board to switch the ROM to usbboot mode in the USB boot
GitHub repository.
For versions of CM4 with an eMMC, make sure you have set NVMe first in the boot order.
Remember to add the NVMe boot mode 6 to BOOT_ORDER in recovery/boot.conf.
CM4 Lite automatically boots from NVMe when the SD card slot is empty.
NVMe BOOT_ORDER
The BOOT_ORDER setting in EEPROM configuration controls boot behaviour. For NVMe boot,
use boot mode 6. For more information, see Raspberry Pi bootloader configuration.
Example
Below is an example of UART output when the bootloader detects the NVMe drive:
In Linux the SSD appears as /dev/nvme0 and the "namespace" as /dev/nvme0n1. There
will be two partitions /dev/nvme0n1p1 (FAT) and /dev/nvme0n1p2 (EXT4). Use lsblk to
check the partition assignments:
Troubleshooting
If the boot process fails, please file an issue on the rpi-eeprom GitHub repository, being
sure to attach a copy of the console and anything displayed on the screen during boot.
HTTP boot
Edit this on GitHub
The network install feature uses HTTP over Ethernet to boot the Raspberry Pi into
embedded Raspberry Pi Imager.
In addition to network install, you can explicitly boot your device with files downloaded via
HTTP with boot-mode 7. You can still use this even if network install on boot is disabled.
You could, for example, add this to your BOOT_ORDER as a fall-back boot method, or put it
behind a GPIO conditional to initiate HTTP boot from your own server when a GPIO pin is
pulled low.
For example, if you added the following to your EEPROM config and GPIO 8 (which has a
default state of 1 or HIGH) were to be pulled low, the files
http://downloads.raspberrypi.org:80/net_install/boot.img and
http://downloads.raspberrypi.org:80/net_install/boot.sig would be
downloaded. If network install on boot were enabled, it would use the same URL. If GPIO 8
were not pulled low the behaviour would be unchanged.
[gpio8=0]
BOOT_ORDER=0xf7
HTTP_HOST=downloads.raspberrypi.org
NET_INSTALL_ENABLED=0
boot.img and the boot.sig signature file is a ram disk containing a boot file system. For
more details, see boot_ramdisk.
HTTP in the BOOT_ORDER will be ignored if secure boot is enabled and HTTP_HOST is not
set.
Requirements
To use HTTP boot, update to a bootloader released 10th March 2022 or later. HTTP boot
requires a wired Ethernet connection.
To use custom CA certificates, update to a bootloader released 5th April 2024 or later. Only
devices running the BCM2712 CPU support custom CA certificates.
Keys
All HTTP downloads must be signed. The bootloader includes a public key for the files on
the default host fw-download-alias1.raspberrypi.com. This key will be used to verify
the network install image, unless you set HTTP_HOST and include a public key in the
EEPROM. This allows you to host the Raspberry Pi network install images on your own
server.
WARNING
Using your own network install image will require you to sign the image and add your
public key to the EEPROM. If you then apply a public EEPROM update, your key will be
lost and will need to be re-added.
Use the following command to add your public key to the EEPROM. boot.conf contains
your modifications:
Then, use the following command to sign the network install image with your private key:
Finally, put boot.img and boot.sig on your web server to use your own signed network
install image.
Certificates
For security, Network Install uses HTTPS to download OS images from the Raspberry Pi
website. This feature uses our own CA root included in the bootloader to verify the host.
You can add your own custom CA certificate to your device EEPROM to securely download
images from your own website. Use the --cacertder option of the rpi-eeprom-config
tool to add the DER-encoded certificate. You must place a hash of the certificate in the
EEPROM config settings to ensure that the certificate is not modified.
Then, run the following command to generate a SHA-256 hash of the certificate:
$ sha256sum cert.der
701bd97f67b0f5483a9734e6e5cf72f9a123407b346088638f597878563193fc cert.d
er
<path_to_files> with the path to your OS image hosted on your website, e.g.
path/to/files
[gpio8=0]
BOOT_ORDER=0xf7
NET_INSTALL_ENABLED=0
HTTP_HOST=<your_website>
HTTP_PATH=<path_to_files>
HTTP_CACERT_HASH=<hash>
When you specify a HTTP_CACERT_HASH, Network Install downloads the image using
HTTPS over port 443. Without a hash, Network install downloads the image using HTTP
over port 80.
During network boot, your Raspberry Pi should use HTTPS instead of HTTP. To see the full
HTTPS URL resolved by Network Install for the download, check the boot output:
Secure boot
If secure boot is enabled, then the Raspberry Pi can only run code signed by the customer’s
private key. So if you want to use network install or HTTP boot mode with secure boot, you
must sign boot.img and generate boot.sig with your own key and host these files
somewhere for download. The public key in the EEPROM will be used to verify the image.
If secure boot is enabled and HTTP_HOST is not set, then network install and HTTP boot
will be disabled.
Boot sequence
Edit this on GitHub
IMPORTANT
The following boot sequence applies to the BCM2837 and BCM2837B0 based models
of Raspberry Pi only. On models prior to this, the Raspberry Pi will try SD card boot,
followed by USB device mode boot. For the Raspberry Pi 4 and Raspberry Pi 5 boot
sequence please see the EEPROM bootflow section.
USB boot defaults on Raspberry Pi 3 will depend on which version is being used. See this
page for information on enabling USB boot modes when not enabled by default.
When the BCM2837 boots, it uses two different sources to determine which boot modes to
enable. Firstly, the one-time-programmable (OTP) memory block is checked to see which
boot modes are enabled. If the GPIO boot mode setting is enabled, then the relevant GPIO
lines are tested to select which of the OTP-enabled boot modes should be attempted. Note
that GPIO boot mode can only be used to select boot modes that are already enabled in the
OTP. See GPIO boot mode for details on configuring GPIO boot mode. GPIO boot mode is
disabled by default.
Next, the boot ROM checks each of the boot sources for a file called bootcode.bin; if it is
successful it will load the code into the local 128K cache and jump to it. The overall boot
mode process is as follows:
BCM2837 boots
If GPIO boot mode enabled, use GPIO boot mode to refine list of enabled boot modes
Success - boot
Success - boot
If OTG pin == 0
Enable USB, wait for valid USB 2.0 devices (two seconds)
Device found:
If there is no SD card inserted, the SD boot mode takes five seconds to fail. To reduce
this and fall back to USB more quickly, you can either insert an SD card with nothing on it
or use the GPIO bootmode OTP setting described above to only enable USB.
The default pull for the GPIOs is defined on page 102 of the ARM Peripherals datasheet.
If the value at boot time does not equal the default pull, then that boot mode is enabled.
USB enumeration is a means of enabling power to the downstream devices on a hub,
then waiting for the device to pull the D+ and D- lines to indicate if it is either USB 1 or
USB 2. This can take time: on some devices it can take up to three seconds for a hard
disk drive to spin up and start the enumeration process. Because this is the only way of
detecting that the hardware is attached, we have to wait for a minimum amount of time
(two seconds). If the device fails to respond after this maximum timeout, it is possible to
increase the timeout to five seconds using program_usb_boot_timeout=1 in
config.txt.
MSD boot takes precedence over Ethernet boot.
It is no longer necessary for the first partition to be the FAT partition, as the MSD boot
will continue to search for a FAT partition beyond the first one.
The boot ROM also now supports GUID partitioning and has been tested with hard drives
partitioned using Mac, Windows, and Linux.
The LAN951x is detected using the Vendor ID 0x0424 and Product ID 0xec00: this is
different to the standalone LAN9500 device, which has a product ID of 0x9500 or
0x9e00. To use the standalone LAN9500, an I2C EEPROM would need to be added to
change these IDs to match the LAN951x.
The primary SD card boot mode is, as standard, set to be GPIOs 49-53. It is possible to
boot from the secondary SD card on a second set of pins, i.e. to add a secondary SD card
to the GPIO pins. However, we have not yet enabled this ability.
NAND boot and SPI boot modes do work, although they do not yet have full GPU support.
The USB device boot mode is enabled by default at the time of manufacture, but the USB
host boot mode is only enabled with program_usb_boot_mode=1. Once enabled, the
processor will use the value of the OTGID pin on the processor to decide between the two
modes. On any Raspberry Pi Model B/B+, the OTGID pin is driven to 0 and therefore will
only boot via host mode once enabled (it is not possible to boot through device mode
because the LAN951x device is in the way).
The USB will boot as a USB device on the Raspberry Pi Zero or Compute Module if the
OTGID pin is left floating (when plugged into a PC for example), so you can push the
bootcode.bin into the device. The usbboot code for doing this is available on GitHub.
Since Raspberry Pi 4, Raspberry Pi flagship devices use an EEPROM bootloader. The main
difference between these and previous products is that the second-stage bootloader is
loaded from SPI flash EEPROM instead of the bootcode.bin file used on previous
products.
SoC powers up
Fail - continue
Fail - continue
While True
Success - run recovery.bin and update the SPI EEPROM or switch to USB mass
storage device mode
NOTE
recovery.bin is a minimal second stage program used to reflash the bootloader SPI
EEPROM image.
Please see the bootloader configuration page for more information about each boot mode,
and the boot folder page for a description of the GPU firmware files loaded by this stage.
Else if WAKE_ON_GPIO is 1
Sleep
While True
Read the next boot-mode from the BOOT_ORDER parameter in the EEPROM config
file.
If boot-mode == RESTART
Failure - continue
If the firmware is not found or a timeout or network error occurs then continue
Failure - continue
Attempt to load firmware using USB device mode from the USB OTG port - see
USB boot. There is no timeout for RPIBOOT mode.
Differences on Raspberry Pi 5
The power button is used to wake up from PMIC STANDBY or HALT instead of GPIO 3.
Instead of loading start.elf, the firmware loads the Linux kernel. Effectively, the
bootloader has an embedded version of start.elf.
Bootloader updates
The bootloader may also be updated before the firmware is started if a pieeprom.upd file
is found. See the bootloader EEPROM page for more information about bootloader
updates.
To set the tryboot flag, add tryboot after the partition number in the reboot command.
Normally, the partition number defaults to zero but it must be specified if extra arguments
are added. Always use quotes when passing arguments to reboot: it accepts only a single
argument:
All Raspberry Pi models support tryboot, however, on Raspberry Pi 4 Model B revision 1.0
and 1.1 the EEPROM must not be write protected. This is because older Raspberry Pi 4B
devices have to reset the power supply (losing the tryboot state), so this is stored inside the
EEPROM instead.
tryboot_a_b mode
If the tryboot_a_b property in autoboot.txt is set to 1 then config.txt is loaded instead
of tryboot.txt. This is because the tryboot switch has already been made at a higher
level (the partition), so it’s unnecessary to have a tryboot.txt file within alternate partition
itself.
The tryboot_a_b property is implicitly set to 1 when loading files from within a boot.img
ramdisk.
$ rpi-eeprom-config
To edit the current EEPROM configuration and apply the updates to latest EEPROM release,
run the following command:
For more information about the EEPROM update process, see boot EEPROM.
Configuration properties
This section describes all the configuration items available in the bootloader. The syntax is
the same as config.txt but the properties are specific to the bootloader. Conditional filters
are also supported except for EDID.
BOOT_UART
If 1 then enable UART debug output on GPIO 14 and 15. Configure the receiving debug
terminal at 115200bps, 8 bits, no parity bits, 1 stop bit.
Default: 0
UART_BAUD
Raspberry Pi 5 only.
Supported values: 9600, 19200, 38400, 57600, 115200, 230400, 460800, 921600
Default: 115200
WAKE_ON_GPIO
If 1 then sudo halt will run in a lower power mode until either GPIO3 or GLOBAL_EN are
shorted to ground.
This setting is not relevant on Flagship models since Raspberry Pi 5, Compute Modules
since CM5, and Keyboard models since Pi 500 because the dedicated power button may
always be used to wake from HALT or STANDBY.
Default: 1
POWER_OFF_ON_HALT
If 1 and WAKE_ON_GPIO=0 then sudo halt will switch off all PMIC outputs. This is lowest
possible power state for halt but may cause problems with some HATs because 5V will still
be on. GLOBAL_EN must be shorted to ground to boot.
The dedicated power button on Pi 400 operates even if the processor is switched off. This
behaviour is enabled by default, however, WAKE_ON_GPIO=2 may be set to use an external
GPIO power button instead of the dedicated power button.
On Flagship models since Raspberry Pi 5 and Keyboard models since Pi 500, this places
the PMIC in STANDBY mode where all outputs are switched off. There is no need to set
WAKE_ON_GPIO and pressing the dedicated power button boots the device.
BOOT_ORDER
The BOOT_ORDER setting allows flexible configuration for the priority of different boot
modes. It is represented as a 32-bit unsigned integer where each nibble represents a boot-
mode. The boot modes are attempted in lowest significant nibble to highest significant
nibble order.
BOOT_ORDER fields
The BOOT_ORDER property defines the sequence for the different boot modes. It is read
right to left, and up to eight digits may be defined.
0x5 BCM-USB-MSD USB 2.0 boot from USB Type C socket (CM4:
USB type A socket on CM4IO board). Not
available on Raspberry Pi 5.
0x7 HTTP HTTP boot over ethernet. See HTTP boot for
more details.
RPIBOOT is intended for use with Compute Module 4 to load a custom debug image (e.g. a
Linux RAM-disk) instead of the normal boot. This should be the last boot option because it
does not currently support timeouts or retries.
BOOT_ORDER examples
BOOT_ORDER Description
MAX_RESTARTS
If the RESTART (0xf) boot-mode is encountered more than MAX_RESTARTS times then a
watchdog reset is triggered. This isn’t recommended for general use but may be useful for
test or remote systems where a full reset is needed to resolve issues with hardware or
network interfaces.
Default: -1 (infinite)
SD_BOOT_MAX_RETRIES
The number of times that SD boot will be retried after failure before moving to the next
boot-mode defined by BOOT_ORDER.
Default: 0
NET_BOOT_MAX_RETRIES
The number of times that network boot will be retried after failure before moving to the
next boot-mode defined by BOOT_ORDER.
Default: 0
DHCP_TIMEOUT
The timeout in milliseconds for the entire DHCP sequence before failing the current
iteration.
Minimum: 5000
Default: 45000
DHCP_REQ_TIMEOUT
Minimum: 500
Default: 4000
TFTP_FILE_TIMEOUT
Minimum: 5000
Default: 30000
TFTP_IP
Optional dotted decimal ip address (e.g. 192.168.1.99) for the TFTP server which
overrides the server-ip from the DHCP request.
This may be useful on home networks because tftpd-hpa can be used instead of dnsmasq
where broadband router is the DHCP server.
Default: ""
TFTP_PREFIX
In order to support unique TFTP boot directories for each Raspberry Pi, the bootloader
prefixes the filenames with a device-specific directory. If neither start4.elf nor start.elf are
found in the prefixed directory then the prefix is cleared.
On earlier models the serial number is used as the prefix, however on Raspberry Pi 4 and 5
the MAC address is no longer generated from the serial number, making it difficult to
automatically create tftpboot directories on the server by inspecting DHCPDISCOVER
packets. To support this the TFTP_PREFIX may be customized to either be the MAC
address, a fixed value or the serial number (default).
Value Description
Default: 0
TFTP_PREFIX_STR
Specify the custom directory prefix string used when TFTP_PREFIX is set to 1. For
example:- TFTP_PREFIX_STR=tftp_test/
Default: ""
PXE_OPTION43
Overrides the PXE Option43 match string with a different string. It’s normally better to apply
customisations to the DHCP server than change the client behaviour, but this option is
provided in case that’s not possible.
DHCP_OPTION97
In earlier releases the client GUID (Option97) was just the serial number repeated four
times. By default, the new GUID format is the concatenation of the four-character code
(FourCC) (RPi4 0x34695052 for Raspberry Pi 4 or RPi5 0x35695052 for Raspberry Pi 5), the
board revision (e.g. 0x00c03111 or 0x00d04170) (4-bytes), the least significant 4 bytes of
the mac address and the 4-byte serial number. This is intended to be unique but also
provides structured information to the DHCP server, allowing Raspberry Pi 4 and 5
computers to be identified without relying upon the Ethernet MAC OUID.
Default: 0x34695052
MAC_ADDRESS
Overrides the Raspberry Pi Ethernet MAC address with the given value. e.g.
dc:a6:32:01:36:c2
Default: ""
MAC_ADDRESS_OTP
Overrides the Raspberry Pi Ethernet MAC address with a value stored in the Customer OTP
registers.
For example, to use a MAC address stored in rows 0 and 1 of the Customer OTP.
MAC_ADDRESS_OTP=0,1
The first value (row 0 in the example) contains the OUI and the most significant 8 bits of
the MAC address. The second value (row 1 in the example) stores the remaining 16-bits of
the MAC address. This is the same format as used for the Raspberry Pi MAC address
programmed at manufacture.
Any two customer rows may be selected and combined in either order.
The Customer OTP rows are OTP registers 36 to 43 in the vcgencmd otp_dump output so if
the first two rows are programmed as follows then MAC_ADDRESS_OTP=0,1 would give a
MAC address of e4:5f:01:20:24:7e.
36:247e0000
37:e45f0120
Default: ""
If TFTP_IP and the following options are set then DHCP is skipped and the static IP
configuration is applied. If the TFTP server is on the same subnet as the client then
GATEWAY may be omitted.
CLIENT_IP
Default: ""
SUBNET
Default: ""
GATEWAY
The gateway address to use if the TFTP server is on a different subnet e.g. 192.168.0.1
Default: ""
DISABLE_HDMI
The HDMI boot diagnostics display is disabled if DISABLE_HDMI=1. Other non-zero values
are reserved for future use.
Default: 0
HDMI_DELAY
Skip rendering of the HDMI diagnostics display for up to N seconds (default 5) unless a
fatal error occurs. The default behaviour is designed to avoid the bootloader diagnostics
screen from briefly appearing during a normal SD/USB boot.
Default: 5
ENABLE_SELF_UPDATE
Enables the bootloader to update itself from a TFTP or USB mass storage device (MSD)
boot filesystem.
If self-update is enabled then the bootloader will look for the update files (.sig/.upd) in the
boot file system. If the update image differs from the current image then the update is
applied and system is reset. Otherwise, if the EEPROM images are byte-for-byte identical
then boot continues as normal.
Notes:
Prior to 2022, self-update was not enabled in SD boot. On a Raspberry Pi 4, the ROM can
already load recovery.bin from the SD card. On a CM4, neither self-update nor
recovery.bin have any effect and USB boot is required (see the Compute Module
EEPROM bootloader docs).
For network boot make sure that the TFTP boot directory can be mounted via NFS and
that rpi-eeprom-update can write to it.
Default: 1
FREEZE_VERSION
Previously this property was only checked by the rpi-eeprom-update script. However,
now that self-update is enabled the bootloader will also check this property. If set to 1, this
overrides ENABLE_SELF_UPDATE to stop automatic updates. To disable FREEZE_VERSION
you will have to use SD card boot with recovery.bin.
Custom EEPROM update scripts must also check this flag.
Default: 0
HTTP_HOST
If network install or HTTP boot is initiated, boot.img and boot.sig are downloaded from
this server.
Invalid host names will be ignored. They should only contain lower case alphanumeric
characters and - or .. If HTTP_HOST is set then HTTPS is disabled and plain HTTP used
instead. You can specify an IP address to avoid the need for a DNS lookup. Don`t include
the HTTP scheme or any forward slashes in the hostname.
Default: fw-download-alias1.raspberrypi.com
HTTP_PORT
You can use this property to change the port used for network install and HTTP boot.
HTTPS is enabled when using the default host fw-download-alias1.raspberrypi.com.
If HTTP_HOST is changed then HTTPS is disabled and plain HTTP will be used instead.
When HTTPS is disabled, plain HTTP will still be used even if HTTP_PORT is changed to 443.
HTTP_PATH
Case-sensitive. Use forward (Linux) slashes for the path separator. Leading and trailing
forward slashes are not required.
If HTTP_HOST is not set, HTTP_PATH is ignored and the URL will be https://fw-download-
alias1.raspberrypi.com:443/net_install/boot.img. If HTTP_HOST is set the URL will
be http://<HTTP_HOST>:<HTTP_PORT>/<HTTP_PATH>/boot.img
Default: net_install
IMAGER_REPO_URL
The embedded Raspberry Pi Imager application is configured with a JSON file downloaded
at startup.
You can change the URL of the JSON file used by the embedded Raspberry Pi Imager
application to get it to offer your own images. You can test this with the standard
Raspberry Pi Imager application by passing the URL via the --repo argument.
Default: http://downloads.raspberrypi.org/os_list_imagingutility_v3.json
NET_INSTALL_ENABLED
When network install is enabled, the bootloader displays the network install screen on boot
if it detects a keyboard.
In order to detect the keyboard, network install must initialise the USB controller and
enumerate devices. This increases boot time by approximately 1 second so it may be
advantageous to disable network install in some embedded applications.
NET_INSTALL_AT_POWER_ON
When set to 1, displays the network install UI briefly after a cold boot to make this feature
more obvious to new users. Overrides NET_INSTALL_ENABLED if the settings conflict.
The default bootloader images set this value to 1 in the bootloader config. To disable the
brief network install UI display, use the Advanced Options menu in raspi-config or
manually delete this line in rpi-eeprom-config:
NET_INSTALL_KEYBOARD_WAIT
If network install is enabled, the bootloader attempts to detect a keyboard and the SHIFT
key to initiate network install. You can change the length of this wait in milliseconds with
this property.
Setting this to 0 disables the keyboard wait, although network install can still be initiated if
no boot files are found and USB boot-mode 4 is in BOOT_ORDER.
NOTE
Default: 900
NETCONSOLE duplicates debug messages to the network interface. The IP addresses and
ports are defined by the NETCONSOLE string.
NOTE
NETCONSOLE blocks until the Ethernet link is established or a timeout occurs. The
timeout value is DHCP_TIMEOUT although DHCP is not attempted unless network boot is
requested.
Format
src_port@src_ip/dev_name,dst_port@dst_ip/dst_mac
E.g. [email protected]/,6666@/
In order to simplify parsing, the bootloader requires every field separator to be present. You
must specify the source IP address, but you can leave the following fields blank to use their
default values:
src_port - 6665
dst_port - 6666
dst_ip - 255.255.255.255
dst_mac - 00:00:00:00:00
One way to view the data is to connect the test Raspberry Pi 4 to another Raspberry Pi
running WireShark and select udp.srcport == 6665 as a filter and select Analyze →
Follow → UDP stream to view as an ASCII log.
NETCONSOLE should not be enabled by default because it may cause network problems. It
can be enabled on demand via a GPIO filter:
PARTITION
The PARTITION option may be used to specify the boot partition number, if it has not
explicitly been set by the reboot command (e.g. sudo reboot N) or by
boot_partition=N in autoboot.txt. This could be used to boot from a rescue partition if
the user presses a button.
Default: 0
PSU_MAX_CURRENT
Raspberry Pi 5 only.
If set, this property instructions the firmware to skip USB power-delivery negotiation and
assume that it is connected to a power supply with the given current rating. Typically, this
would either be set to 3000 or 5000 i.e. low or high-current capable power supply.
Default: ""
USB_MSD_EXCLUDE_VID_PID
A list of up to four VID/PID pairs specifying devices which the bootloader should ignore. If
this matches a HUB then the HUB won’t be enumerated, causing all downstream devices to
be excluded. This is intended to allow problematic (e.g. very slow to enumerate) devices to
be ignored during boot enumeration. This is specific to the bootloader and is not passed to
the OS.
The format is a comma-separated list of hexadecimal values with the VID as most
significant nibble. Spaces are not allowed. E.g. 034700a0,a4231234
Default: ""
USB_MSD_DISCOVER_TIMEOUT
If no USB mass storage devices are found within this timeout then USB-MSD is stopped
and the next boot-mode is selected.
USB_MSD_LUN_TIMEOUT
How long to wait in milliseconds before advancing to the next LUN e.g. a multi-slot SD-
CARD reader. This is still being tweaked but may help speed up boot if old/slow devices are
connected as well as a fast USB-MSD device containing the OS.
Minimum: 100
USB_MSD_PWR_OFF_TIME
Raspberry Pi 4 only.
When the Pi is rebooted power USB power is switched off by the hardware. A short power
off time can cause problems with some USB devices so this parameter may be used to
force a longer power off as though the cable was physically removed.
On RaspberryPi 4 version 1.3 and older, the configurable/long power off requires the XHCI
controller to be enabled so there is actually a short power off followed by a longer
configurable power off. The longer configurable power off may be skipped by setting this
parameter to zero.
On newer revisions the hardware ensures that USB power is off from reboot and the
bootloader only enables power after this timeout has elapsed. This is happens after
memory is initialised ensuring that USB power is off for at least two seconds. Therefore,
this parameter generally has no effect on newer hardware revisions.
Minimum: 0
Maximum: 5000
USB_MSD_STARTUP_DELAY
If defined, delays USB enumeration for the given timeout after the USB host controller has
initialised. If a USB hard disk drive takes a long time to initialise and triggers USB timeouts
then this delay can be used to give the driver additional time to initialise. It may also be
necessary to increase the overall USB timeout (USB_MSD_DISCOVER_TIMEOUT).
Minimum: 0
Default: 0
VL805
Compute Module 4 only.
If the VL805 property is set to 1 then the bootloader will search for a VL805 PCIe XHCI
controller and attempt to initialise it with VL805 firmware embedded in the bootloader
EEPROM. This enables industrial designs to use VL805 XHCI controllers without providing
a dedicated SPI EEPROM for the VL805 firmware.
On Compute Module 4 the bootloader never writes to the dedicated VL805 SPI
EEPROM. This option just configures the controller to load the firmware from SDRAM.
Do not use this option if the VL805 XHCI controller has a dedicated EEPROM. It will fail
to initialise because the VL805 ROM will attempt to use a dedicated SPI EEPROM if
fitted.
The embedded VL805 firmware assumes the same USB configuration as Raspberry Pi
4B (two USB 3.0 ports and four USB 2.0 ports). There is no support for loading alternate
VL805 firmware images, a dedicated VL805 SPI EEPROM should be used instead for
such configurations.
Default: 0
XHCI_DEBUG
This property is a bit-field which controls the verbosity of USB debug messages for mass
storage boot-mode. Enabling all of these messages generates a huge amount of log data
which will slow down booting and may even cause boot to fail. For verbose logs it’s best to
use NETCONSOLE.
Value Log
[config.txt] section
After reading config.txt the GPU firmware start4.elf reads the bootloader EEPROM
config and checks for a section called [config.txt]. If the [config.txt] section exists
then the contents from the start of this section to the end of the file is appended in
memory, to the contents of the config.txt file read from the boot partition. This can be
used to automatically apply settings to every operating system, for example, dtoverlays.
WARNING
If you configure the bootloader with an invalid configuration that fails to boot, you must
re-flash the bootloader EEPROM with a valid configuration to boot.
TIP
Some configuration properties live in config.txt. For more information about those
properties, see configuration properties.
WHITE PAPER
An up-to-24-bit parallel RGB interface is available on all Raspberry Pi boards with the 40
way header and the Compute Modules. This interface allows parallel RGB displays to be
attached to the Raspberry Pi GPIO either in RGB24 (8 bits for red, green and blue) or
RGB666 (6 bits per colour) or RGB565 (5 bits red, 6 green, and 5 blue).
This interface is controlled by the GPU firmware and can be programmed by a user via
special config.txt parameters and by enabling the correct Linux Device Tree overlay.
GPIO pins
One of the alternate functions selectable on Bank 0 of the Raspberry Pi GPIO is DPI
(Display Parallel Interface) which is a simple clocked parallel interface (up to 8 bits of R, G
and B; clock, enable, hsync, and vsync). This interface is available as alternate function 2
(ALT2) on GPIO Bank 0:
GPIO ALT2
GPIO0 PCLK
GPIO1 DE
GPIO2 LCD_VSYNC
GPIO3 LCD_HSYNC
GPIO4 DPI_D0
GPIO5 DPI_D1
GPIO6 DPI_D2
GPIO7 DPI_D3
GPIO8 DPI_D4
GPIO9 DPI_D5
GPIO10 DPI_D6
GPIO11 DPI_D7
GPIO12 DPI_D8
GPIO13 DPI_D9
GPIO14 DPI_D10
GPIO15 DPI_D11
GPIO16 DPI_D12
GPIO17 DPI_D13
GPIO18 DPI_D14
GPIO19 DPI_D15
GPIO20 DPI_D16
GPIO21 DPI_D17
GPIO22 DPI_D18
GPIO23 DPI_D19
GPIO24 DPI_D20
GPIO25 DPI_D21
GPIO26 DPI_D22
GPIO27 DPI_D23
NOTE
There are various ways that the colour values can be presented on the DPI output pins
in either 565, 666, or 24-bit modes (see the following table and the output_format part
of the dpi_output_format parameter below):
M R GPIO
o G
d B
e b 2 2 2 2 2 2 2 2 1 1 1 1 1 1 1 1 1 1 9 8 7 6 5 4
it 7 6 5 4 3 2 1 0 9 8 7 6 5 4 3 2 1 0
s
1 - - - - - - - - - - - - - - - - - - - - - - - - -
2 5 - - - - - - - - 7 6 5 4 3 7 6 5 4 3 2 7 6 5 4 3
6
5
3 5 - - - 7 6 5 4 3 - - 7 6 5 4 3 2 - - - 7 6 5 4 3
6
5
4 5 - - 7 6 5 4 3 - - - 7 6 5 4 3 2 - - 7 6 5 4 3 -
6
5
5 6 - - - - - - 7 6 5 4 3 2 7 6 5 4 3 2 7 6 5 4 3 2
6
6
6 6 - - 7 6 5 4 3 2 - - 7 6 5 4 3 2 - - 7 6 5 4 3 2
6
6
7 8 7 6 5 4 3 2 1 0 7 6 5 4 3 2 1 0 7 6 5 4 3 2 1 0
8
8
dtparam=i2c_arm=off
dtparam=spi=off
Configure a display
The Kernel Mode Setting (KMS) generic display interface enables output to arbitrary
displays, as long as you have an appropriate driver.
Auto detect
Auto detect allows your Raspberry Pi to connect with a display without a manually
configured device tree overlay. Auto detection is enabled by default. You can enable display
auto detect by adding the following line to config.txt:
display_auto_detect=1
Replace the 1 with a 0 to disable auto detect. When you connect the official Raspberry Pi
display with auto detect enabled, KMS determines the display model automatically and
configures the appropriate display settings.
NOTE
To use any display other than the official Raspberry Pi display, you must specify a
dtoverlay entry in config.txt. The panel manufacturer should configure timings for your
display in Linux kernel code and provide an overlay to enable those settings. See the
Adafruit Kippah display entry for an example. The following example demonstrates how to
set a dtoverlay entry for the Kippah display in your /boot/firmware/config.txt file:
dtoverlay=vc4-kms-kippah-7inch-overlay
Display timings are usually defined in the kernel, but you can also define them in the
provided panel-dpi driver. If your panel lacks a defined overlay in kernel code, you can use
the panel-dpi driver to define display timings as parameters. This enables you to
manually configure a device tree entry for any display.
The following example demonstrates how you can define timings using device tree
parameters:
dtoverlay=vc4-kms-v3d
dtoverlay=vc4-kms-dpi-generic,hactive=480,hfp=26,hsync=16,hbp=10
dtparam=vactive=640,vfp=25,vsync=10,vbp=16
dtparam=clock-frequency=32000000,rgb666-padhi
NOTE
Device tree line length must not exceed 80 characters. When a setting requires a line
longer than 80 characters, split the assignment of that parameter across multiple lines.
Option Description
You can find a 40-pin GPIO (general-purpose input/output) header on all current Raspberry
Pi boards. The GPIO headers on all boards have a 0.1in (2.54mm) pin pitch.
NOTE
The header is unpopulated (has no headers) on Zero and Pico devices that lack the "H"
suffix.
General Purpose I/O (GPIO) pins can be configured as either general-purpose input,
general-purpose output, or as one of up to six special alternate settings, the functions of
which are pin-dependent.
NOTE
The GPIO pin numbering scheme is not in numerical order. GPIO pins 0 and 1 are
present on the board (physical pins 27 and 28), but are reserved for advanced use.
Outputs
A GPIO pin designated as an output pin can be set to high (3.3V) or low (0V).
Inputs
A GPIO pin designated as an input pin can be read as high (3.3V) or low (0V). This is made
easier with the use of internal pull-up or pull-down resistors. Pins GPIO2 and GPIO3 have
fixed pull-up resistors, but for other pins this can be configured in software.
WARNING
While connecting simple components to GPIO pins is safe, be careful how you wire
things up. LEDs should have resistors to limit the current passing through them. Do not
use 5V for 3.3V components. Do not connect motors directly to the GPIO pins, instead
use an H-bridge circuit or a motor controller board.
Permissions
In order to use the GPIO ports, your user must be a member of the gpio group. The default
user account is a member by default, but you must add other users manually using the
following command:
The pads are configurable CMOS push-pull output drivers/input buffers. Register-based
control settings are available for:
Power-on states
All GPIO pins revert to general-purpose inputs on power-on reset. The default pull states
are also applied, which are detailed in the alternate function table in the Arm peripherals
datasheet. Most GPIOs have a default pull applied.
Interrupts
Each GPIO pin, when configured as a general-purpose input, can be configured as an
interrupt source to the Arm. Several interrupt generation sources are configurable:
Level-sensitive (high/low)
Rising/falling edge
Level interrupts maintain the interrupt status until the level has been cleared by system
software (e.g. by servicing the attached peripheral generating the interrupt).
The normal rising/falling edge detection has a small amount of synchronisation built into
the detection. At the system clock frequency, the pin is sampled with the criteria for
generation of an interrupt being a stable transition within a three-cycle window, i.e. a record
of 1 0 0 or 0 1 1. Asynchronous detection bypasses this synchronisation to enable the
detection of very narrow events.
Alternative functions
Almost all of the GPIO pins have alternative functions. Peripheral blocks internal to the SoC
can be selected to appear on one or more of a set of GPIO pins, for example the I2C buses
can be configured to at least three separate locations. Pad control, such as drive strength
or Schmitt filtering, still applies when the pin is configured as an alternate function.
SPI
SPI0: MOSI (GPIO10); MISO (GPIO9); SCLK (GPIO11); CE0 (GPIO8), CE1 (GPIO7)
SPI1: MOSI (GPIO20); MISO (GPIO19); SCLK (GPIO21); CE0 (GPIO18); CE1 (GPIO17);
CE2 (GPIO16)
I2C
Serial
TX (GPIO14); RX (GPIO15)
Voltage specifications
Two 5V pins and two 3.3V pins are present on the board, as well as a number of ground
pins (GND), which can not be reconfigured. The remaining pins are all general-purpose 3.3V
pins, meaning outputs are set to 3.3V and inputs are 3.3V-tolerant.
The table below gives the various voltage specifications for the GPIO pins for BCM2835,
BCM2836, BCM2837 and RP3A0-based products (e.g. Raspberry Pi Zero or Raspberry Pi
3+). For information about Compute Modules you should see the relevant datasheets.
voltagea
IIL Input TA = - - 5 µA
leakage +85 C◦
current
CIN Input - - 5 - pF
capacitanc
e
voltageb
voltageb
currentc
RPU Pullup - 50 - 65 kΩ
resistor
RPD Pulldown - 50 - 65 kΩ
resistor
a
Hysteresis enabled
b
Default drive strength (8mA)
c
Maximum drive strength (16mA)
The table below gives the voltage specifications for the GPIO pins on BCM2711-based
products (4-series devices). For information about Compute Modules you should see the
relevant datasheets.
IIL Input TA = - - 10 µA
leakage +85 C◦
current
voltageb
currentc
currentc
RPU Pullup - 33 - 73 kΩ
resistor
Symbol Parameter Conditions Min Typical Max Unit
RPD Pulldown - 33 - 73 kΩ
resistor
a
Hysteresis enabled
b
Default drive strength (4mA)
c
Maximum drive strength (8mA)
GPIO drive strengths do not indicate a maximum current, but a maximum current under
which the pad will still meet the specification. You should set the GPIO drive strengths to
match the device being attached in order for the device to work correctly.
WARNING
On 4-series devices, the current level is half the value shown in the diagram.
Current value
The current value specifies the maximum current under which the pad will still meet the
specification.
Current value is not the current that the pad will deliver, and is not a current limit.
If set high, the pad will try to drive the output to the rail voltage (3.3V)
If set low, the pad will try to drive the output to ground (0V)
The pad will try to drive the output high or low. Success will depend on the requirements of
what is connected. If the pad is shorted to ground, it will not be able to drive high. It will try
to deliver as much current as it can, and the current is only limited by the internal
resistance.
If the pad is driven high and it is shorted to ground, in due time it will fail. The same holds
true if you connect it to 3.3V and drive it low.
Meeting the specification is determined by the guaranteed voltage levels. Because the pads
are digital, there are two voltage levels, high and low. The I/O ports have two parameters
which deal with the output level:
VOL=0.14V means that if the output is Low, it will be <= 0.14V. VOH=3.0V means that if the
As an example, a drive strength of 16mA means that if you set the pad high, you can draw
up to 16mA, and the output voltage is guaranteed to be >=VOH. This also means that if you
set a drive strength of 2mA and you draw 16mA, the voltage will not be VOH but lower. In
NOTE
On the Compute Module devices, it is possible to change the VDD IO from the standard
3.3V. In this case, VOL and VOH will change according to the table in the GPIO section.
The Raspberry Pi 3.3V supply was designed with a maximum current of ~3mA per GPIO
pin. If you load each pin with 16mA, the total current is 272mA. The 3.3V supply will
collapse under that level of load. Big current spikes will happen, especially if you have a
capacitive load. Spikes will bounce around all the other pins near them. This is likely to
cause interference with the SD card, or even the SDRAM behaviour.
Safe current
All the electronics of the pads are designed for 16mA. This is a safe value under which you
will not damage the device. Even if you set the drive strength to 2mA and then load it so
16mA comes out, this will not damage the device. Other than that, there is no guaranteed
maximum safe current.
GPIO addresses
0x 7e10 002c PADS (GPIO 0-27)
1 = 4mA
2 = 6mA
3 = 8mA
4 = 10mA
5 = 12mA
6 = 14mA
7 = 16mA
Raspberry Pi is often used as part of another product. This documentation describes some
extra facilities available to use other capabilities of your Raspberry Pi.
There are a number of OTP values that can be used. To see a list of all the OTP values, run
the following command:
$ vcgencmd otp_dump
28 - Serial number
Also, from 36 to 43 (inclusive), there are eight rows of 32 bits available for the customer.
NOTE
On BCM2712 devices these numbers are different. Row 31 is the serial number and row
32 is the board revision number. The customer rows are 77 to 84 inclusive.
Some of these rows can be programmed with vcmailbox. This is a Linux driver interface to
the firmware which will handle the programming of the rows. To do this, please refer to the
documentation, and the vcmailbox example application.
The vcmailbox application can be used directly from the command line on Raspberry Pi
OS. An example usage would be:
$ vcmailbox 0x00010004 8 8 0 0
The above uses the mailbox property interface GET_BOARD_SERIAL with a request size of 8
bytes and response size of 8 bytes (sending two integers for the request 0, 0). The
response to this will be two integers (0x00000020 and 0x80000000) followed by the tag
code, the request length, the response length (with the 31st bit set to indicate that it is a
response) then the 64-bit serial number (where the MS 32 bits are always 0).
The OTP values are one-time programmable. Once a bit has been changed from 0 to 1,
it can’t be changed back.
To set the customer OTP values you will need to use the SET_CUSTOMER_OTP (0x38021) tag
as follows:
$ vcmailbox 0x00030021 20 20 4 3 0 0 0
If you’d like to integrate this functionality into your own code, you should be able to achieve
this by using the vcmailbox.c code as an example.
This can be done using a special argument with the OTP write mailbox:
Once locked, the customer OTP values can no longer be altered. Note that this locking
operation is irreversible.
$ vcmailbox 0x00030086 4 4 0
OTP is only locked until the device is reset, so OTP locks need to be reapplied on every
boot.
This operation is unlikely to be useful for the vast majority of users, and is irreversible.
$ vcmailbox 0x00030083 6 6 0 0
0x00000020 0x80000000 0x00030083 0x00000006 0x80000006 0xddccbbaa 0x0000
ffee 0x00000000
In order to set a customer MAC address, it has to be sent as two 32 words with the bytes in
the right order. You can run a command to check it’s formatted properly:
Check the log to see if the MAC address matches your expectations:
$ sudo vclog -m
1057826.701: read mac address 11:22:33:44:55:66
A multicast address is not considered valid. The least significant bit in the most significant
octet of a MAC address is the multicast bit, so make sure this is not set.
You can then set the customer MAC address with the command vcmailbox
0x00038082/3/4 6 6 <row1> <row0>:
These rows can be programmed and read using similar vcmailbox commands to those
used for managing customer OTP rows. If secure-boot / file-system encryption is not
required, then the device private key rows can be used to store general-purpose
information.
The device private key rows can only be read via the vcmailbox command which
requires access to /dev/vcio which is restricted to the video group on Raspberry Pi
OS.
NOTE
The rpi-otp-private-key script only works on devices that use the Broadcom
BCM2711 or BCM2712 processors.
The rpi-otp-private-key script wraps the device private key vcmailbox APIs to make it
easier to read and write a key in the OpenSSL format.
NOTE
The usbboot repository contains all the tools you need, including rpi-eeprom as a Git
submodule.
Example output:
f8dbc7b0a4fcfb1d706e298ac9d0485c2226ce8df7f7596ac77337bd09fbe160
WARNING
NOTE
To specify the number of OTP rows to use, pass -l <word count>. To specify a start
location in the key store, pass -o <word offset>.
$ vcmailbox 0x00030081 40 40 0 8 0 0 0 0 0 0 0 0
Example output:
Write all of the row (replace the trailing eight zeros with the key data):
$ vcmailbox 0x00038081 40 40 0 8 0 0 0 0 0 0 0 0
All SoCs used by the Raspberry Pi range have a inbuilt one-time programmable (OTP)
memory block. A few locations have factory-programmed data.
$ vcgencmd otp_dump
16
OTP control register - BCM2711
17
bootmode register
Bit 15: disables ROM RSA key 0 - (secure boot enabled if set) (BCM2711)
Bit 29: enables USB host booting (ethernet and mass storage)
NOTE
18
copy of bootmode register
28
serial number
29
~(serial number)
30
revision code 1
33
board revision extended - the meaning depends on the board model. This is available
via device-tree in /proc/device-tree/chosen/rpi-boardrev-ext and for testing
purposes this OTP value can be temporarily overridden by setting board_rev_ext in
config.txt.
Compute Module 4
Bit 30: Whether the Compute Module has a Wi-Fi module fitted
0 - Wi-Fi
1 - No Wi-Fi
Bit 31: Whether the Compute Module has an EMMC module fitted
0 - EMMC
1 - No EMMC (Lite)
Raspberry Pi 400
36-43
customer OTP values
45
MPG2 decode key
46
WVC1 decode key
47-54
SHA256 of RSA public key for secure-boot
55
secure-boot flags (reserved for use by the bootloader)
56-63
256-bit device-specific private key
64-65
MAC address; if set, system will use this in preference to the automatically generated
address based on the serial number
66
advanced boot register (not BCM2711)
Bits 0-6: GPIO for ETH_CLK output pin
0 - 25MHz
1 - 24MHz
1
Also contains bits to disable overvoltage, OTP programming, and OTP reading.
22
bootmode register
23
copy of bootmode register
29
advanced boot mode
31
lower 32 bits of serial number
32
board revision
33
board attributes - the meaning depends on the board model. This is available via
device-tree in /proc/device-tree/chosen/rpi-boardrev-ext
35
upper 32 bits of serial number The full 64 bit serial number is available in
/proc/device-tree/serial-number
50-51
Ethernet MAC address This is passed to the operating system in the Device Tree, e.g.
/proc/device-tree/axi/pcie@120000/rp1/ethernet@100000/local-mac-
address
52-53
Wi-Fi MAC address This is passed to the operating system in the Device Tree, e.g.
/proc/device-tree/axi/mmc@1100000/wifi@1/local-mac-address
54-55
Bluetooth MAC address This is passed to the operating system in the Device Tree,
e.g. /proc/device-tree/soc/serial@7d50c000/bluetooth/local-bd-address
77-84
customer OTP values
86
board country - The default keyboard country code used by piwiz If set, this is
available via Device Tree in /proc/device-tree/chosen/rpi-country-code
87-88
customer Ethernet MAC address Overrides OTP rows 50-51 if set
89-90
customer Wi-Fi MAC address Overrides OTP rows 52-53 if set
89-90
customer Bluetooth MAC address Overrides OTP rows 54-55 if set
109-114
Factory device UUID Currently a 16-digit numerical id which should match the bar
code on the device. Padded with zero characters and c40 encoded.
This is available via device-tree in /proc/device-tree/chosen/rpi-duid.
Raspberry Pi 5 has an FPC connector on the right-hand side of the board. This connector
breaks out a PCIe Gen 2.0 ×1 interface for fast peripherals.
To connect a PCIe HAT+ device, connect it to your Raspberry Pi. Your Raspberry Pi should
automatically detect the device. To connect a non-HAT+ device, connect it to your
Raspberry Pi, then manually enable PCIe.
For more information about the PCIe FPC connector pinout and other details needed to
create third-party devices, accessories, and HATs, see the Raspberry Pi Connector for PCIe
standards document. It should be read alongside the Raspberry Pi HAT+ Specification.
NOTE
Enable PCIe
By default, the PCIe connector is not enabled unless connected to a HAT+ device. To
enable the connector, add the following line to /boot/firmware/config.txt:
dtparam=pciex1
Reboot with sudo reboot for the configuration changes to take effect.
NOTE
BOOT_ORDER=0xf416
To boot from a non-HAT+ device, also add the following line:
PCIE_PROBE=1
After saving your changes, reboot your Raspberry Pi with sudo reboot to update the
EEPROM.
The Raspberry Pi 5 is not certified for Gen 3.0 speeds. PCIe Gen 3.0 connections may
be unstable.
By default, Raspberry Pi 5 uses Gen 2.0 speeds (5 GT/s). Use one of the following
approaches to force Gen 3.0 (8 GT/s) speeds:
config.txt raspi-config
dtparam=pciex1_gen=3
Reboot your Raspberry Pi with sudo reboot for these settings to take effect.
Power button
Edit this on GitHub
NOTE
This section only applies to Raspberry Pi models with a power button, such as the
Raspberry Pi 5.
When you plug your Raspberry Pi into power for the first time, it will automatically turn on
and boot into the operating system without having to push the button.
If you run Raspberry Pi Desktop, you can initiate a clean shutdown by briefly pressing the
power button. A window will appear asking whether you want to shutdown, reboot, or
logout.
Select an option or press the power button again to initiate a clean shutdown.
NOTE
If you run Raspberry Pi Desktop, you can press the power button twice in quick
succession to shut down. If you run Raspberry Pi OS Lite without a desktop, press the
power button a single time to initiate a shutdown.
Restart
If the Raspberry Pi board is turned off, but still connected to power, pressing the power
button restarts the board.
NOTE
Resetting the Power Management Integrated Circuit (PMIC) can also restart the board.
Connecting a HAT can reset the PMIC. Always disconnect your device from the power
supply before connecting a HAT.
Hard shutdown
To force a hard shutdown, press and hold the power button.
The J2 jumper is located between the RTC battery connector and the board edge. This
breakout allows you to add your own power button to Raspberry Pi 5 by adding a Normally
Open (NO) momentary switch bridging the two pads. Briefly closing this switch will perform
the same actions as the onboard power button.
Power supply
Edit this on GitHub
The power supply requirements differ by Raspberry Pi model. All models require a 5.1V
supply, but the current required generally increases according to model. All models up to
the Raspberry Pi 3 require a micro USB power connector, while Raspberry Pi 4, Raspberry
Pi 400, and Raspberry Pi 5 use a USB-C connector.
NOTE
NOTE
If you use a third-party USB-PD multi-port power supply, plugging an additional device
into the supply when your Raspberry Pi is connected causes a renegotiation between
the supply and the Raspberry Pi. If the Raspberry Pi is powered, this happens
seamlessly. If the Raspberry Pi is powered down, this renegotiation may cause the
Raspberry Pi to boot.
The Ethernet jack on Raspberry Pi 5 is PoE+ capable, supporting the IEEE 802.3at-2009
PoE standard.
The Ethernet jack on Raspberry Pi 4B and Pi 3B+ is PoE capable, supporting the IEEE
802.3af-2003 PoE standard.
All Raspberry Pi models with a PoE-capable Ethernet jack require a HAT to draw power
through the Ethernet port. For models that support PoE, we recommend the PoE HAT. For
models that support PoE+, we recommend the PoE+ HAT.
NOTE
Most Raspberry Pis provide enough current to USB peripherals to power most USB devices,
including keyboards, mice, and adapters. However, some devices require additional current,
including modems, external disks, and high-powered antenna. To connect a USB device
with power requirements that exceed the values specified in the table above, connect it
using an externally-powered USB hub.
The power requirements of the Raspberry Pi increase as you make use of the various
interfaces on the Raspberry Pi. Combined, the GPIO pins can draw 50mA safely; each pin
can individually draw up to 16mA. The HDMI port uses 50mA. The Camera Module requires
250mA. USB keyboards and mice can take as little as 100mA or as much as 1000mA.
Check the power rating of the devices you plan to connect to the Raspberry Pi and
purchase a power supply accordingly. If you’re not sure, use an externally-powered USB
hub.
Run the following command to check the status of power output to the USB ports:
The following table describes the amount of power (in amps) drawn by different Raspberry
Pi models during various workloads:
NOTE
WHITE PAPER
By default, the Raspberry Pi 5 consumes around 1W to 1.4W of power when turned off.
This can be decreased by manually editing the EEPROM configuration with sudo rpi-
eeprom-config -e. Change the settings to the following:
BOOT_UART=1
POWER_OFF_ON_HALT=1
BOOT_ORDER=0xf416
This should drop the power consumption when powered down to around 0.01W.
If you see warnings, switch to a higher quality power supply and cable. Low quality power
supplies can corrupt storage or cause unpredictable behaviour within the Raspberry Pi.
Voltages can drop for a variety of reasons. You may have plugged in too many high-
demand USB devices. The power supply could be inadequate. Or the power supply cable
could use wires that are too thin.
WHITE PAPER
max_current
The max current in mA
uspd_power_data_objects
A dump of the PDOs - debug for advanced users
usb_max_current_enable
Whether the current limiter was set to high or low
usb_over_current_detected
Whether any USB over current occurred during boot before transferring control to the
OS
reset_event
The PMIC reset reason e.g. watchdog, over- or under-voltage, over-temperature
The PMIC has built-in ADCs that, among other things, can measure the supply voltage
EXT5V_V. Use the following command to view ADC measurements:
$ vcgencmd pmic_read_adc
You can’t see USB current or anything else connected directly to 5V, because this bypasses
the PMIC. You should not expect this to add up to the wattage of the source power supply.
However, it can be useful to monitor things like the core voltage.
Back-powering
The USB specification requires that USB devices must not supply current to upstream
devices. If a USB device does supply current to an upstream device, then this is called
back-powering. Often this happens when a badly-made powered USB hub is connected,
and will result in the powered USB hub supplying power to the host Raspberry Pi. This is
not recommended since the power being supplied to the Raspberry Pi via the hub will
bypass the protection circuitry built into the Raspberry Pi, leaving it vulnerable to damage in
the event of a power surge.
The Raspberry Pi 5 includes an RTC module. This can be battery powered via the J5 (BAT)
connector on the board located to the right of the USB-C power connector.
The J5 battery connector
You can set a wake alarm which will switch the board to a very low-power state
(approximately 3mA). When the alarm time is reached, the board will power back on. This
can be useful for periodic jobs like time-lapse imagery.
To support the low-power mode for wake alarms, edit the bootloader configuration:
POWER_OFF_ON_HALT=1
WAKE_ON_GPIO=0
That will halt the board into a very low-power state, then wake and restart after 10 minutes.
The RTC also provides the time on boot e.g. in dmesg, for use cases that lack access to
NTP:
NOTE
The RTC is still usable even when there is no backup battery attached to the J5
connector.
The official battery part is a rechargeable lithium manganese coin cell, with a pre-fitted two-
pin JST-SH plug and an adhesive mounting pad. This is suitable for powering the RTC when
the main power supply for the board is disconnected. Since the current draw when
powered down measures in single-digit µA, the retention time measures in months.
NOTE
We do not recommend using a primary (non-rechargeable) lithium cell for the RTC. The
RTC backup current consumption is higher than most dedicated RTC modules and will
result in a short service life.
WARNING
Charging of the battery is disabled by default. There are sysfs files that show the charging
voltage and limits:
/sys/devices/platform/soc/soc:rpi_rtc/rtc/rtc0/charging_voltage:0
/sys/devices/platform/soc/soc:rpi_rtc/rtc/rtc0/charging_voltage_max:4400
000
/sys/devices/platform/soc/soc:rpi_rtc/rtc/rtc0/charging_voltage_min:1300
000
dtparam=rtc_bbat_vchg=3000000
Reboot with sudo reboot to use the new voltage setting. Check the sysfs files to ensure
that the charging voltage was correctly set.
Raspberry Pi computers are equipped with a number of SPI buses. SPI can be used to
connect a wide variety of peripherals - displays, network controllers (Ethernet, CAN bus),
UARTs, etc. These devices are best supported by kernel device drivers, but the spidev API
allows userspace drivers to be written in a wide array of languages.
SPI hardware
Raspberry Pi Zero, 1, 2 and 3 have three SPI controllers:
SPI0, with two hardware chip selects, is available on the header of all Raspberry Pis;
there is also an alternate mapping that is only available on Compute Modules.
SPI1, with three hardware chip selects, is available on all Raspberry Pi models except
the original Raspberry Pi 1 Model A and Model B.
SPI2, also with three hardware chip selects, is only available on Compute Module 1, 3
and 3+.
On the Raspberry Pi 4, 400 and Compute Module 4 there are four additional SPI buses:
SPI3 to SPI6, each with two hardware chip selects. These extra SPI buses are available via
alternate function assignments on certain GPIO pins. For more information, see the
BCM2711 Arm peripherals datasheet.
Chapter 10 in the BCM2835 Arm peripherals datasheet describes the main controller.
Chapter 2.3 describes the auxiliary controller.
Pin/GPIO mappings
SPI0
SPI1
Master modes
SCLK
serial clock
CE
chip enable (often called chip select)
MOSI
master out slave in
MISO
master in slave out
MOMI
master out master in
Standard mode
In Standard SPI mode the peripheral implements the standard three-wire serial protocol
(SCLK, MOSI and MISO).
Bidirectional mode
In bidirectional SPI mode the same SPI standard is implemented, except that a single wire
is used for data (MOMI) instead of the two used in standard mode (MISO and MOSI). In this
mode, the MOSI pin serves as MOMI pin.
The LoSSI standard allows issuing of commands to peripherals (LCD) and to transfer data
to and from them. LoSSI commands and parameters are 8 bits long, but an extra bit is
used to indicate whether the byte is a command or parameter/data. This extra bit is set
high for data and low for a command. The resulting 9-bit value is serialised to the output.
LoSSI is commonly used with MIPI DBI type C compatible LCD controllers.
NOTE
Some commands trigger an automatic read by the SPI controller, so this mode cannot
be used as a multipurpose 9-bit SPI.
Transfer modes
Polled
Interrupt
DMA
Speed
The clock divider (CDIV) field of the CLK register sets the SPI clock speed:
SCLK
Core Clock / CDIV
If CDIV is set to 0, the divisor is 65536. The divisor must be a multiple of 2, with odd
numbers rounded down. Note that not all possible clock rates are usable because of
analogue electrical issues (rise times, drive strengths, etc).
Chip selects
Setup and hold times related to the automatic assertion and de-assertion of the CS lines
when operating in DMA mode are as follows:
The CS line will be asserted at least three core clock cycles before the msb of the first
byte of the transfer.
The CS line will be de-asserted no earlier than one core clock cycle after the trailing edge
of the final clock pulse.
SPI software
Linux driver
SPI0 is disabled by default. To enable it, use raspi-config, or ensure the line
dtparam=spi=on is not commented out in /boot/firmware/config.txt. By default it uses
two chip select lines, but this can be reduced to one using dtoverlay=spi0-1cs. There is
also dtoverlay=spi0-2cs; without any parameters it is equivalent to dtparam=spi=on.
To enable SPI1, you can use 1, 2 or 3 chip select lines. Add the appropriate lines to
/boot/firmware/config.txt:"
#1 chip select
dtoverlay=spi1-1cs
#2 chip select
dtoverlay=spi1-2cs
#3 chip select
dtoverlay=spi1-3cs
Similar overlays exist for SPI2, SPI3, SPI4, SPI5 and SPI6.
The driver does not make use of the hardware chip select lines because of some
limitations. Instead, it can use an arbitrary number of GPIOs as software/GPIO chip selects.
This means you are free to choose any spare GPIO as a CS line, and all of these SPI
overlays include that control - see /boot/firmware/overlays/README for details, or run
(for example) dtoverlay -h spi0-2cs (dtoverlay -a | grep spi might be helpful to
list them all).
Speed
The driver supports all speeds which are even integer divisors of the core clock, although
as said above not all of these speeds will support data transfer due to limits in the GPIOs
and in the devices attached. As a rule of thumb, anything over 50MHz is unlikely to work,
but your mileage may vary.
SPI_CPOL
clock polarity
SPI_CPHA
clock phase
SPI_CS_HIGH
chip select active high
SPI_NO_CS
1 device per bus, no Chip select
SPI_3WIRE
bidirectional mode, data in and out pin shared
Bidirectional mode, also called 3-wire mode, is supported by the spi-bcm2835 kernel
module. Please note that in this mode, either the tx or rx field of the spi_transfer struct
must be a NULL pointer, since only half-duplex communication is possible. Otherwise, the
transfer will fail. The spidev_test.c source code does not consider this correctly, and
therefore does not work at all in 3-wire mode.
8 - normal
Transfer modes
Interrupt mode is supported on all SPI buses. SPI0, and SPI3-6 also support DMA transfers.
spidev
There is a loopback test program in the Linux documentation that can be used as a
starting point. See the Troubleshooting section.
There are several Python libraries that provide access to spidev, including spidev (pip
install spidev - see https://pypi.org/project/spidev/) and SPI-Py
(https://github.com/lthiery/SPI-Py).
There are other user space libraries that provide SPI control by directly manipulating the
hardware: this is not recommended.
Troubleshooting
Loopback test
This can be used to test SPI send and receive. Put a wire between MOSI and MISO. It does
not test CE0 and CE1.
$ wget https://raw.githubusercontent.com/raspberrypi/linux/rpi-6.1.y/too
ls/spi/spidev_test.c
$ gcc -o spidev_test spidev_test.c
$ ./spidev_test -D /dev/spidev0.0
spi mode: 0
bits per word: 8
max speed: 500000 Hz (500 KHz)
FF FF FF FF FF FF
40 00 00 00 00 95
FF FF FF FF FF FF
FF FF FF FF FF FF
FF FF FF FF FF FF
DE AD BE EF BA AD
F0 0D
Some of the content above has been copied from the elinux SPI page, which also borrows
from here. Both are covered by the CC-SA licence.
1. For the original Raspberry Pi 1 Model B the limit is 100mA per port.
Raspberry Pi 5
The Raspberry Pi 5 requires a good quality USB-C power supply capable of delivering 3A at
+5V (15W) in order to boot. However, using such a supply will restrict current draw to
peripherals. If you are using a power supply that cannot provide 5A at +5V on first boot you
will be warned by the operating system that the current draw to peripherals will be
restricted to 600mA.
For users who wish to drive high-power peripherals like hard drives and SSDs, while
retaining margin for peak workloads, a USB-PD enabled power supply capable of supplying
a 5A at +5V (25W) should be used. If the Raspberry Pi 5 firmware detects such a supply, it
increases the USB current limit for peripherals to 1.6A, providing 5W of extra power for
downstream USB devices, and 5W of extra onboard power budget.
NOTE
The power budget is shared between the USB ports and the fan header.
Raspberry Pi 4
Raspberry Pi 4 offers two USB 3.0 ports and two USB 2.0 ports which are connected to a
VL805 USB controller. The USB 2.0 lines on all four ports are connected to a single USB 2.0
hub within the VL805. This limits the total available bandwidth for USB 1.1 and USB 2.0
devices to that of a single USB 2.0 port.
On Raspberry Pi 4, the USB controller used on previous models is located on the USB type
C port and is disabled by default.
The USB controller on models prior to Raspberry Pi 4 has only a basic level of support for
certain devices, which presents a higher software processing overhead. It also supports
only one root USB port: all traffic from connected devices is funnelled down this single bus,
which operates at a maximum speed of 480Mbps.
The USB 2.0 specification defines three device speeds - low, full and high. Most mice and
keyboards are low speed, most USB sound devices are full speed, and most video devices
(webcams or video capture) are high speed.
Generally, there are no issues with connecting multiple high speed USB devices to a
Raspberry Pi.
The software overhead incurred when talking to low- and full-speed devices means that
there are limitations on the number of simultaneously active low- and full-speed devices.
Small numbers of these types of devices connected to a Raspberry Pi will cause no issues.
USB 2.0 high speed devices, including USB 2.0 hubs, operate correctly when connected via
a USB 3.0 hub.
Avoid connecting low or full speed devices into a USB 3.0 hub. As a workaround, plug a
USB 2.0 hub into the downstream port of the USB 3.0 hub and connect the low-speed
device, or use a USB 2.0 hub between the Raspberry Pi and the USB 3.0 hub, then plug low-
speed devices into the USB 2.0 hub.
Old webcams may be full-speed devices. Because these devices transfer a lot of data and
incur additional software overhead, reliable operation is not guaranteed. As a workaround,
try to use the camera at a lower resolution.
Expensive audiophile sound cards typically use large amounts of USB bandwidth. Reliable
operation with 96kHz/192kHz DACs is not guaranteed. As a workaround, forcing the output
stream to be CD quality (44.1kHz/48kHz 16-bit) will reduce the stream bandwidth to
reliable levels.
USB 2.0 and 3.0 hubs have a mechanism for talking to full- or low-speed devices connected
to their downstream ports called a transaction translator (TT). This device buffers high
speed requests from the host and transmits them at full or low speed to the downstream
device. Two configurations of hub are allowed by the USB specification: Single TT (one TT
for all ports) and Multi TT (one TT per port). Because of a hardware limitation, if too many
full- or low-speed devices are plugged into a single TT hub, the devices may behave
unreliably. It is recommended to use a Multi TT hub to interface with multiple full and low
speed devices. As a workaround, spread full- and low-speed devices out between the
Raspberry Pi’s own USB port and the single TT hub.
Each distinct Raspberry Pi model revision has a unique revision code. You can look up a
Raspberry Pi’s revision code by running:
$ cat /proc/cpuinfo
The last three lines show the hardware type, the revision code, and the Raspberry Pi’s
unique serial number. For example:
Hardware : BCM2835
Revision : a02082
Serial : 00000000765fc593
NOTE
All Raspberry Pi computers report BCM2835, even those with BCM2836, BCM2837,
BCM2711, and BCM2712 processors. You should not use this string to detect the
processor. Decode the revision code using the information below, or cat
/sys/firmware/devicetree/base/model.
NOQuuuWuFMMMCCCCPPPPTTTTTTTTRRRR
1: Overvoltage disallowed
1: OTP programming
disallowed
0: old-style revision
1: 512MB
2: 1GB
3: 2GB
4: 4GB
5: 8GB
6: 16GB
1: Egoman
2: Embest
3: Sony Japan
4: Embest
5: Stadium
Part Represents Options
1: BCM2836
2: BCM2837
3: BCM2711
4: BCM2712
0x01: B
0x02: A+
0x03: B+
0x04: 2B
0x06: CM1
0x08: 3B
0x09: Zero
0x0a: CM3
0x0c: Zero W
0x0d: 3B+
0x0e: 3A+
0x10: CM3+
0x11: 4B
0x12: Zero 2 W
0x13: 400
0x14: CM4
0x15: CM4S
0x17: 5
0x18: CM5
0x19: 500
1
Information on programming the OTP bits.
2
The warranty bit is never set on Raspberry Pi 4.
This list is not exhaustive - there may be codes in use that are not in this table. Please
see the next section for best practices on using revision codes to identify boards.
In this example above, we have a hexadecimal revision code of c03111. Converting this to
binary, we get 0 0 0 000 0 0 1 100 0000 0011 00010001 0001. Spaces have been
inserted to show the borders between each section of the revision code, according to the
above table.
Starting from the lowest order bits, the bottom four (0-3) are the board revision number, so
this board has a revision of 1. The next eight bits (4-11) are the board type, in this case
binary 00010001, hex 11, so this is a Raspberry Pi 4B. Using the same process, we can
determine that the processor is a BCM2711, the board was manufactured by Sony UK, and
it has 4GB of RAM.
Obviously there are so many programming languages out there it’s not possible to give
examples for all of them, but here are two quick examples for C and Python. Both these
examples use a system call to run a bash command that gets the cpuinfo and pipes the
result to awk to recover the required revision code. They then use bit operations to extract
the New, Model, and Memory fields from the code.
#include <stdio.h>
#include <stdlib.h>
if (new && model == 0x11 && mem >= 3) // Note, 3 in the mem field is
2GB
printf("We are a 4B with at least 2GB of RAM!\n" );
return 0;
}
import subprocess
if new and model == 0x11 and mem >= 3 : # Note, 3 in the mem field is 2G
B
print("We are a 4B with at least 2GB RAM!")
A naive implementation uses a list of supported revision codes, comparing the detected
code with the list to decide if the device is supported. This breaks when a new board
revision comes out or if the production location changes: each creates a new revision code
not in the supported revision code list. This would cause rejections of new revisions of the
same board type, despite the fact that they are always backwards-compatible. Every time a
new revision appears, you would have to release a new supported revision code list
containing the new revision code - an onerous support burden.
Instead, use one of the following approaches:
Filter by the board-type field (3A, 4B, etc.), which indicates the model, but not the
revision.
Filter by the amount-of-memory field, since RAM vaguely corresponds to the computing
power of a board.
For instance, you could limit support to Raspberry Pi 4B models with 2GB of RAM or more.
The examples in the previous section use this recommended approach.
NOTE
Always check bit 23, the 'New' flag, to ensure that the revision code is the new version
before checking any other fields.
Support and formatting for /proc/cpuinfo varies across Linux distributions. To check the
model or CPU of a Raspberry Pi device on any Linux distribution (including Raspberry Pi
OS), check the device tree:
This outputs two null-separated string values, each containing a comma-separated make
and model. For instance, the Raspberry Pi 5 outputs the board and CPU strings above.
These correspond to the following values:
You can view and edit the Raspberry Pi documentation source on Github. Please read our usage and contributions policy before you make a Pull
Request.
Raspberry Pi documentation is copyright © 2012-2025 Raspberry Pi Ltd and is licensed under a Creative Commons Attribution-ShareAlike 4.0
International (CC BY-SA) licence.
Some content originates from the eLinux wiki, and is licensed under a Creative Commons Attribution-ShareAlike 3.0 Unported licence.
The terms HDMI, HDMI High-Definition Multimedia Interface, HDMI trade dress and the HDMI Logos are trademarks or registered trademarks of HDMI
Licensing Administrator, Inc