neorv32/docs/userguide/simulating_the_processor.adoc

248 lines
14 KiB
Plaintext

<<<
:sectnums:
== Simulating the Processor
The NEORV32 project includes a core CPU, built-in peripherals in the Processor Subsystem, and additional peripherals in
the templates and examples.
Therefore, there is a wide range of possible testing and verification strategies.
On the one hand, a simple smoke testbench allows ensuring that functionality is correct from a software point of view.
That is used for running the RISC-V architecture tests, in order to guarantee compliance with the ISA specification(s).
On the other hand, http://vunit.github.io/[VUnit] and http://vunit.github.io/verification_components/user_guide.html[Verification Components]
are used for verifying the functionality of the various peripherals from a hardware point of view.
.Xilinx Vivado / ISIM
[IMPORTANT]
When using Xilinx Vivado (ISIM for simulation) make sure to **turn of** "incremental compilation" (_Project Setting_
-> _Simulation_ -> _Advanced_ -> _Enable incremental compilation). This will slow down simulation relaunch but will
ensure that all application images (`*_image.vhd`) are reanalyzed when recompiling the NEORV32 application or bootloader
[TIP]
The processor can check if it is being _simulated_ by checking the SYSINFO _SYSINFO_SOC_IS_SIM_ flag
(see https://stnolting.github.io/neorv32/#_system_configuration_information_memory_sysinfo).
Note that this flag is not guaranteed to be set correctly (depending on the HDL toolchain's pragma support).
:sectnums:
=== Testbench
A plain-VHDL (no third-party libraries) testbench (`sim/simple/neorv32_tb.simple.vhd`) can be used for simulating and
testing the processor.
This testbench features a 100MHz clock and enables all optional peripheral and CPU extensions except for the `E`.
[IMPORTANT]
In the simple testbench several optional extensions are disabled, such as C or E.
If software is compiled using instructions corresponding to disabled extensions, the whole processor will hang in an eternal exception loop and, therefore, the simulation will timeout.
The `MARCH` must be a subset of the extensions enabled in the testbench.
.True Random Number Generator
[NOTE]
The NEORV32 TRNG will be set to "simulation mode" when enabled for simulation (replacing the ring-oscillators
by pseudo-random LFSRs). See the neoTRNG documentation for more information.
The simulation setup is configured via the "User Configuration" section located right at the beginning of
the testbench's architecture. Each configuration constant provides comments to explain the functionality.
Besides the actual NEORV32 Processor, the testbench also simulates "external" components that are connected
to the processor's external bus/memory interface. These components are:
* an external instruction memory (that also allows booting from it)
* an external data memory
* an external memory to simulate "external IO devices"
* a memory-mapped registers to trigger the processor's interrupt signals
The following table shows the base addresses of these four components and their default configuration and
properties:
[NOTE]
====
Attributes:
* `r` = read
* `w` = write
* `e` = execute
* `8` = byte-accessible
* `16` = half-word-accessible
* `32` = word-accessible
====
.Testbench: processor-external memories
[cols="^4,>3,^5,<11"]
[options="header",grid="rows"]
|=======================
| Base address | Size | Attributes | Description
| `0x00000000` | `imem_size_c` | `r/w/e 8/16/32` | external IMEM (initialized with application image)
| `0x80000000` | `dmem_size_c` | `r/w/e 8/16/32` | external DMEM
| `0xf0000000` | 64 bytes | `r/w/e 8/16/32` | external "IO" memory
| `0xff000000` | 4 bytes | `-/w/- -/-/32` | memory-mapped register to trigger "machine external", "machine software" and "SoC Fast Interrupt" interrupts
|=======================
[IMPORTANT]
The simulated NEORV32 does not use the bootloader and _directly boots_ the current application image (from
the `rtl/core/neorv32_application_image.vhd` image file).
.UART output during simulation
[IMPORTANT]
Data written to the NEORV32 UART0 / UART1 transmitter is send to a virtual UART receiver implemented
as part of the testbench. Received chars are send to the simulator console and are also stored to a log file
(`neorv32.testbench_uart0.out` for UART0, `neorv32.testbench_uart1.out` for UART1) inside the simulation's home folder.
**Please note that printing via the native UART receiver takes a lot of time.** For faster simulation console output
see section <<_faster_simulation_console_output>>.
:sectnums:
=== Faster Simulation Console Output
When printing data via the physical UART the communication speed will always be based on the configured BAUD
rate. For a simulation this might take some time. To have faster output you can enable the **simulation mode**
for UART0/UART1 (see section https://stnolting.github.io/neorv32/#_primary_universal_asynchronous_receiver_and_transmitter_uart0[Documentation: Primary Universal Asynchronous Receiver and Transmitter (UART0)]).
ASCII data sent to UART0|UART1 will be immediately printed to the simulator console and logged to files in the simulator
execution directory:
* `neorv32.uart?.sim_mode.text.out`: ASCII data.
You can "automatically" enable the simulation mode of UART0/UART1 when compiling an application.
In this case, the "real" UART0/UART1 transmitter unit is permanently disabled.
To enable the simulation mode just compile and install your application and add _UART?_SIM_MODE_ to the compiler's
_USER_FLAGS_ variable (do not forget the `-D` suffix flag):
[source, bash]
----
sw/example/demo_blink_led$ make USER_FLAGS+=-DUART0_SIM_MODE clean_all all
----
The provided define will change the default UART0/UART1 setup function in order to set the simulation
mode flag in the according UART's control register.
[NOTE]
The UART simulation output (to file and to screen) outputs "complete lines" at once. A line is
completed with a line feed (newline, ASCII `\n` = 10).
:sectnums:
=== Simulation using a shell script (with GHDL)
To simulate the processor using _GHDL_ navigate to the `sim/simple/` folder and run the provided shell script.
Any arguments that are provided while executing this script are passed to GHDL.
For example the simulation time can be set to 20ms using `--stop-time=20ms` as argument.
[source, bash]
----
neorv32/sim/simple$ sh ghdl.sh --stop-time=20ms
----
:sectnums:
=== Simulation using Application Makefiles (In-Console with GHDL)
To directly compile and run a program in the console (using the default testbench and GHDL
as simulator) you can use the `sim` makefile target. Make sure to use the UART simulation mode
(`USER_FLAGS+=-DUART0_SIM_MODE` and/or `USER_FLAGS+=-DUART1_SIM_MODE`) to get
faster / direct-to-console UART output.
[source, bash]
----
sw/example/demo_blink_led$ make USER_FLAGS+=-DUART0_SIM_MODE clean_all sim
[...]
Blinking LED demo program
----
:sectnums:
==== Hello World!
To do a quick test of the NEORV32 make sure to have https://github.com/ghdl/ghdl[GHDL] and a
https://github.com/stnolting/riscv-gcc-prebuilt[RISC-V gcc toolchain] installed.
Navigate to the project's `sw/example/hello_world` folder and run `make USER_FLAGS+=-DUART0_SIM_MODE clean_all sim`:
[TIP]
The simulator will output some _sanity check_ notes (and warnings or even errors if something is ill-configured)
right at the beginning of the simulation to give a brief overview of the actual NEORV32 SoC and CPU configurations.
[source, bash]
----
neorv32/sw/example/hello_world$ make USER_FLAGS+=-DUART0_SIM_MODE clean_all sim
../../../sw/lib/source/neorv32_uart.c: In function 'neorv32_uart_setup':
../../../sw/lib/source/neorv32_uart.c:116:2: warning: #warning UART0_SIM_MODE (primary UART) enabled! Sending all UART0.TX data to text.io simulation output instead of real UART0 transmitter. Use this for simulations only! [-Wcpp]
116 | #warning UART0_SIM_MODE (primary UART) enabled! Sending all UART0.TX data to text.io simulation output instead of real UART0 transmitter. Use this for simulations only! <1>
| ^~~~~~~
Memory utilization:
text data bss dec hex filename
4664 0 116 4780 12ac main.elf <2>
Compiling ../../../sw/image_gen/image_gen
Installing application image to ../../../rtl/core/neorv32_application_image.vhd <3>
Simulating neorv32_application_image.vhd...
Tip: Compile application with USER_FLAGS+=-DUART[0/1]_SIM_MODE to auto-enable UART[0/1]'s simulation mode (redirect UART output to simulator console). <4>
Using simulation run arguments: --stop-time=10ms <5>
../../rtl/core/neorv32_top.vhd:355:5:@0ms:(assertion note): [NEORV32] The NEORV32 RISC-V Processor (version 0x01090504), github.com/stnolting/neorv32 <6>
../../rtl/core/neorv32_top.vhd:361:5:@0ms:(assertion note): [NEORV32] Processor Configuration: IMEM DMEM I-CACHE D-CACHE WISHBONE GPIO MTIME UART0 UART1 SPI SDI TWI PWM WDT TRNG CFS NEOLED XIRQ GPTMR XIP ONEWIRE DMA SLINK CRC SYSINFO OCD
../../rtl/core/neorv32_clockgate.vhd:60:3:@0ms:(assertion warning): [NEORV32] Clock gating enabled (using generic clock switch).
../../rtl/core/neorv32_cpu.vhd:142:3:@0ms:(assertion note): [NEORV32] CPU ISA: rv32imabu_zicsr_zicntr_zicond_zifencei_zfinx_zihpm_zxcfu_sdext_sdtrig_smpmp
../../rtl/core/neorv32_cpu.vhd:163:3:@0ms:(assertion note): [NEORV32] CPU tuning options: fast_mul fast_shift
../../rtl/core/neorv32_cpu.vhd:170:3:@0ms:(assertion warning): [NEORV32] Assuming this is a simulation.
../../rtl/core/neorv32_cpu_cp_bitmanip.vhd:172:3:@0ms:(assertion note): [NEORV32] Implementing bit-manipulation (B) sub-extensions Zba Zbb Zbc Zbs
../../rtl/core/neorv32_cpu_cp_fpu.vhd:292:3:@0ms:(assertion warning): [NEORV32] The floating-point unit (Zfinx) is still in experimental state.
../../rtl/core/mem/neorv32_imem.legacy.vhd:72:3:@0ms:(assertion note): [NEORV32] Implementing LEGACY processor-internal IMEM as pre-initialized ROM.
../../rtl/core/neorv32_wishbone.vhd:117:3:@0ms:(assertion note): [NEORV32] Ext. Bus Interface (WISHBONE) - PIPELINED Wishbone protocol, auto-timeout, LITTLE-endian byte order, registered RX, registered TX
../../rtl/core/neorv32_trng.vhd:343:3:@0ms:(assertion note): [neoTRNG NOTE] << neoTRNG V3 - A Tiny and Platform-Independent True Random Number Generator >>
../../rtl/core/neorv32_trng.vhd:545:5:@0ms:(assertion warning): [neoTRNG WARNING] Implementing non-physical pseudo-RNG!
../../rtl/core/neorv32_trng.vhd:545:5:@0ms:(assertion warning): [neoTRNG WARNING] Implementing non-physical pseudo-RNG!
../../rtl/core/neorv32_trng.vhd:545:5:@0ms:(assertion warning): [neoTRNG WARNING] Implementing non-physical pseudo-RNG!
../../rtl/core/neorv32_debug_dm.vhd:227:3:@0ms:(assertion note): [NEORV32] OCD DM compatible to debug spec. version 1.0
<7>
## ## ## ##
## ## ######### ######## ######## ## ## ######## ######## ## ################
#### ## ## ## ## ## ## ## ## ## ## ## ## ## #### ####
## ## ## ## ## ## ## ## ## ## ## ## ## ## ###### ##
## ## ## ######### ## ## ######### ## ## ##### ## ## #### ###### ####
## ## ## ## ## ## ## ## ## ## ## ## ## ## ###### ##
## #### ## ## ## ## ## ## ## ## ## ## ## #### ####
## ## ######### ######## ## ## ## ######## ########## ## ################
## ## ## ##
Hello world! :)
----
<1> Notifier that "simulation mode" of UART0 is enabled (by the `USER_FLAGS+=-DUART0_SIM_MODE` makefile flag). All UART0 output is send to the simulator console.
<2> Final executable size (`text`) and _static_ data memory requirements (`data`, `bss`).
<3> The application code is _installed_ as pre-initialized IMEM. This is the default approach for simulation.
<4> A note regarding UART "simulation mode", but we have already enabled that.
<5> List of (default) arguments that were send to the simulator. Here: maximum simulation time (10ms).
<6> "Sanity checks" from the core's VHDL files. These reports give some brief information about the SoC/CPU configuration (-> generics). If there are problems with the current configuration, an ERROR will appear.
<7> Execution of the actual program starts.
:sectnums:
=== Advanced Simulation using VUnit
https://vunit.github.io/[VUnit] is an open source unit testing framework for VHDL/SystemVerilog.
It allows continuous and automated testing of HDL code by complementing traditional testing methodologies.
The motto of VUnit is _"testing early and often"_ through automation.
VUnit is composed by a http://vunit.github.io/py/ui.html[Python interface] and multiple optional
http://vunit.github.io/vhdl_libraries.html[VHDL libraries].
The Python interface allows declaring sources and simulation options, and it handles the compilation, execution and
gathering of the results regardless of the simulator used.
That allows having a single `run.py` script to be used with GHDL, ModelSim/QuestaSim, Riviera PRO, etc.
On the other hand, the VUnit's VHDL libraries provide utilities for assertions, logging, having virtual queues, handling CSV files, etc.
The http://vunit.github.io/verification_components/user_guide.html[Verification Component Library] uses those features
for abstracting away bit-toggling when verifying standard interfaces such as Wishbone, AXI, Avalon, UARTs, etc.
Testbench sources in `sim` (such as `sim/neorv32_tb.vhd` and `sim/uart_rx*.vhd`) use VUnit's VHDL libraries for testing
NEORV32 and peripherals.
The entry-point for executing the tests is `sim/run.py`.
[source, bash]
----
# ./sim/run.py -l
neorv32.neorv32_tb.all
Listed 1 tests
# ./sim/run.py -v
Compiling into neorv32: rtl/core/neorv32_uart.vhd passed
Compiling into neorv32: rtl/core/neorv32_twi.vhd passed
Compiling into neorv32: rtl/core/neorv32_trng.vhd passed
...
----
See http://vunit.github.io/user_guide.html[VUnit: User Guide] and http://vunit.github.io/cli.html[VUnit: Command Line Interface] for further info about VUnit's features.