The NEORV32 Processor is a customizable microcontroller-like system on chip (SoC) that is based on the RISC-V NEORV32 CPU. The project is intended as auxiliary processor in larger SoC designs or as ready-to-go stand-alone custom / customizable microcontroller.
Special focus is paid on execution safety to provide defined and predictable behavior at any time. Therefore, the CPU ensures that all memory access are acknowledged and no invalid/malformed instructions are executed. Whenever an unexpected situation occurs, the application code is informed via hardware exceptions.
🤔 Want to know more? Check out the project's rationale.
📚 For detailed information take a look at the NEORV32 documentation (online at GitHub-pages). The doxygen-based documentation of the software framework is also available online at GitHub-pages.
🏷️ The project's change log is available in CHANGELOG.md.
To see the changes between official releases visit the project's release page.
📦 The setups folder provides exemplary setups targeting
various FPGA boards and toolchains to get you started. Several example programs (including a FreeRTOS port) to be run on your setup
can be found in sw/example.
🪁 Supported by upstream Zephyr OS.
💡 Feel free to open a new issue or start a new discussion if you have questions, comments, ideas or if something is not working as expected. Or have a chat on our gitter channel.
🚀 Check out the quick links below or directly jump to the User Guide to get started setting up your NEORV32 setup!
- all-in-one: CPU plus SoC plus Software Framework & Tooling
- completely described in behavioral, platform-independent VHDL - no primitives, macros, etc.
- fully synchronous design, no latches, no gated clocks
- be as small as possible while being as RISC-V-compliant as possible – but with a reasonable size-performance trade-off: the processor (CPU including privileged architecture) fits into a Lattice iCE40 UltraPlus 5k low-power FPGA running at 24 MHz
- from zero to
printf("hello world!");- completely open source and documented - easy to use even for FPGA/RISC-V starters – intended to work out of the box
The NEORV32 Processor (top entity: rtl/core/neorv32_top.vhd)
provides a full-featured SoC build around the NEORV32 CPU. It is highly configurable via generics
to allow a flexible customization according to your needs. Note that all modules listed below are optional.
In-depth detailed information regarding the processor/SoC can be found in the 📚
Data Sheet: NEORV32 Processors (SoC).
Memory
- processor-internal data and instruction memories (DMEM / IMEM) & cache (iCACHE)
- bootloader (BOOTLDROM) with serial user interface
- allows booting application code via UART or from external SPI flash
Timers
- machine system timer, 64-bit (MTIME), RISC-V spec. compatible
- general purpose 32-bit timer (GPTMR)
- watchdog timer (WDT)
IO
- standard serial interfaces (UART, SPI, TWI / I²C)
- general purpose GPIO and PWM
- smart LED interface (NEOLED) to directly drive NeoPixel(TM) LEDs
SoC Connectivity and Integration
- 32-bit external bus interface, Wishbone b4 compatible (WISHBONE)
- 32-bit stream link interface with up to 8 independent RX and TX links
(SLINK)
- AXI4-Stream compatible
- external interrupt controller with up to 32 channels (XIRQ)
- custom functions subsystem (CFS) for tightly-coupled custom co-processor extensions
Advanced
- true random number generator (TRNG)
- on-chip debugger (OCD) accessible via JTAG interface - implementing the Minimal RISC-V Debug Specification Version 0.13.2 and compatible with OpenOCD and gdb
- bus keeper to monitor processor-internal bus transactions (BUSKEEPER)
ℹ️ It is recommended to use the processor setup even if you want to use the CPU in stand-alone mode. Just disable all optional processor-internal modules via the according generics and you will get a CPU wrapper that provides a minimal CPU environment and an external memory interface (like AXI4). This minimal setup allows to further use the default bootloader and software framework. From this base you can start building your own processor system.
The hardware resources used by a specific processor setup is defined by the implemented CPU extensions (see below), the configuration of the peripheral modules and some "glue logic". Section FPGA Implementation Results - Processor Modules of the online datasheet shows the resource utilization of each optional processor module to allow an estimation of the actual setup's hardware requirements.
ℹ️ The setups folder provides exemplary FPGA
setups targeting various FPGA boards and toolchains. These setups also provide resource utilization reports for different
SoC configurations
📚 In-depth detailed information regarding the CPU can be found in the Data Sheet: NEORV32 Central Processing Unit.
The CPU (top entity: rtl/core/neorv32_cpu.vhd)
implements the RISC-V 32-bit rv32 ISA with optional extensions (see below). It is compatible to subsets of the
Unprivileged ISA Specification (Version 2.2)
and the Privileged Architecture Specification (Version 1.12-draft).
Compatibility is checked by passing the official RISC-V architecture tests
(see sim/README).
The core is a little-endian Von-Neumann machine implemented as multi-cycle architecture.
However, the CPU's front end (instruction fetch) and back end (instruction execution) can work independently to increase performance.
Currently, three privilege levels (machine and optional user and debug_mode) are supported. The CPU implements all three standard RISC-V machine
interrupts (MTI, MEI, MSI) plus 16 fast interrupt requests as custom extensions.
It also supports all standard RISC-V exceptions (instruction/load/store misaligned address & bus access fault, illegal
instruction, breakpoint, environment calls).
Currently, the following optional RISC-V-compatible ISA extensions are implemented (linked to the according
documentation section). Note that the X extension is always enabled.
RV32
[I/
E]
[A]
[B]
[C]
[M]
[U]
[X]
[Zfinx]
[Zicsr]
[Zicntr]
[Zihpm]
[Zifencei]
[Zmmul]
[PMP]
[DEBUG]
B, Zfinx and Zmmul RISC-V extensions are frozen but not officially ratified yet. Hence, there is no
upstream gcc support. To circumvent this, the NEORV32 software framework provides intrinsic libraries for these extensions.
Implementation results for exemplary CPU configuration generated for an Intel Cyclone IV EP4CE22F17C6N FPGA using Intel Quartus Prime Lite 20.1 ("balanced implementation"). The timing information is derived from the Timing Analyzer / Slow 1200mV 0C Model. No constraints were used at all.
Results generated for hardware version 1.5.7.10.
| CPU Configuration | LEs | FFs | Memory bits | DSPs (9-bit) | f_max |
|---|---|---|---|---|---|
rv32i |
806 | 359 | 1024 | 0 | 125 MHz |
rv32i_Zicsr_Zicntr |
1729 | 813 | 1024 | 0 | 124 MHz |
rv32imac_Zicsr_Zicntr |
2511 | 1074 | 1024 | 0 | 124 MHz |
ℹ️ An incremental list of CPU extension's hardware utilization can found in the Data Sheet: FPGA Implementation Results - CPU.
ℹ️ The CPU (and also the SoC) provides advanced options to optimize for performance, area or energy. See User Guide: Application-Specific Processor Configuration for more information.
The NEORV32 CPU is based on a two-stages pipelined architecture. Since both stage use a multi-cycle processing scheme, each instruction requires several clock cycles to execute (2 cycles for ALU operations, up to 40 cycles for divisions). The average CPI (cycles per instruction) depends on the instruction mix of a specific applications and also on the available CPU extensions.
The following table shows the performance results (relative CoreMark score and average cycles per instruction) for
exemplary CPU configuration running 2000 iterations of the CoreMark CPU benchmark.
The source files are available in sw/example/coremark.
A simple(!) port of the Dhrystone benchmark is also available in
sw/example/dhrystone.
**CoreMark Setup**
Hardware: 32kB IMEM, 8kB DMEM, no caches, 100MHz clock
CoreMark: 2000 iterations, MEM_METHOD is MEM_STACK
Compiler: RISCV32-GCC 10.1.0 (rv32i toolchain)
Compiler flags: default, see makefile; optimization -O3
Results generated for hardware version 1.5.7.10.
| CPU Configuration | CoreMark Score | CoreMarks/MHz | Average CPI |
|---|---|---|---|
small (rv32i_Zicsr) |
33.89 | 0.3389 | 4.04 |
medium (rv32imc_Zicsr) |
62.50 | 0.6250 | 5.34 |
performance (rv32imc_Zicsr + perf. options) |
95.23 | 0.9523 | 3.54 |
ℹ️ More information regarding the CPU performance can be found in the Data Sheet: CPU Performance.
📚 In-depth detailed information regarding the software framework can be found in the Data Sheet: Software Framework.
- core libraries for high-level usage of the provided functions and peripherals
- application compilation based on GNU makefiles
- gcc-based toolchain (pre-compiled toolchains available)
- bootloader with UART interface console
- runtime environment for handling traps
- several example programs to get started including CoreMark, FreeRTOS and Conway's Game of Life
doxygen-based documentation, available on 📚 GitHub pages- supports implementation using open source tooling (GHDL, Yosys and nextpnr; in the future Verilog-to-Routing); both, software and hardware can be developed and debugged with open source tooling
- continuous integration
is available for:
- allowing users to see the expected execution/output of the tools
- ensuring specification compliance
- catching regressions
- providing ready-to-use and up-to-date bitstreams and documentation
This overview provides some quick links to the most important sections of the online Data Sheet and the online User Guide.
-
Rationale - NEORV32: why, how come, what for
-
NEORV32 Processor - the SoC
- Top Entity - Signals - how to connect to the processor
- Top Entity - Generics - configuration options
- Address Space - memory layout and boot configuration
- SoC Modules - available IO/peripheral modules and memories
- On-Chip Debugger - online & in-system debugging of the processor via JTAG
-
NEORV32 CPU - the CPU
- RISC-V compatibility - what is compatible to the specs. and what is not
- Full Virtualization - hardware execution safety
- ISA and Extensions - available RISC-V ISA extensions
- CSRs - control and status registers
- Traps - interrupts and exceptions
- Core Libraries - high-level functions for accessing the processor's peripherals
- Software Framework Documentation -
doxygen-based documentation
- Software Framework Documentation -
- Application Makefiles - turning your application into an executable
- Bootloader - the build-in NEORV32 bootloader
🚀 User Guides (see full User Guide)
- Toolchain Setup - install and setup RISC-V gcc
- General Hardware Setup - setup a new NEORV32 EDA project
- General Software Setup - configure the software framework
- Application Compilation - compile an application using
make - Upload via Bootloader - upload and execute executables
- Application-Specific Processor Configuration - tailor the processor to your needs
- Adding Custom Hardware Modules - add your custom hardware
- Debugging via the On-Chip Debugger - step through code online and in-system
- Simulation - simulate the whole SoC
- Hello World! - run a quick "hello world" simulation
- Overview - license, disclaimer, proprietary notice, ...
- Citing - citing information (DOI)
- Impressum - imprint (:de:)
A big shoutout to all contributors, who helped improving this project! ❤️
RISC-V - Instruction Sets Want To Be Free!
Continous integration provided by GitHub Actions and powered by GHDL.
Made with ☕ in Hanover, Germany 🇪🇺