VeeR EH2 RISC-V Core

This repository contains the VeeR EH2 RISC-V Core design RTL.

Overview

VeeR EH2 is a machine-mode (M-mode) only, 32-bit CPU core which supports RISC-V’s integer (I), compressed instruction (C), multiplication and division (M), atomic (A), and instruction-fetch fence, CSR, and subset of bit manipulation instructions (Zb*) extensions. The core is a 9-stage, dual-threaded, dual-issue, superscalar, mostly in-order pipeline with some out-of-order execution capability.

License

By contributing to this project, you agree that your contribution is governed by Apache-2.0.
Files under the tools directory may be available under a different license. Please review individual files for details.

Directory Structure

├── configs                 # Configurations Dir
├── design                  # Design root dir
│   ├── dbg                 #   Debugger
│   ├── dec                 #   Decode, Registers and Exceptions
│   ├── dmi                 #   DMI block
│   ├── exu                 #   EXU (ALU/MUL/DIV)
│   ├── ifu                 #   Fetch & Branch Prediction
│   ├── include             
│   ├── lib
│   └── lsu                 #   Load/Store
├── docs
├── tools                   # Scripts/Makefiles
└── testbench               # (Very) simple testbench
    ├── asm                 #   Example assembly files
    └── hex                 #   Canned demo hex files

Dependencies

• Verilator (4.102 or later) must be installed on the system if running with Verilator
• If adding/removing instructions, espresso must be installed (used by tools/coredecode)
• A RISC-V tool chain (based on gcc version 7.3 or higher) must be installed so that it can be used to prepare RISC-V binaries to run.

Quickstart guide

1. Clone the repository
2. Setup RV_ROOT to point to the path in your local filesystem
3. Determine your configuration {optional}
4. Run make with tools/Makefile

Release Notes for this version

Please see release notes for changes and bug fixes in this version of VeeR.

Configurations

VeeR can be configured by running the $RV_ROOT/configs/veer.config script:

% $RV_ROOT/configs/veer.config -h for detailed help options

For example to build with a DCCM of size 64 Kb:

% $RV_ROOT/configs/veer.config -dccm_size=64

This will update the default snapshot in $PWD/snapshots/default/ with parameters for a 64K DCCM.

Add -snapshot=dccm64, for example, if you wish to name your build snapshot dccm64 and refer to it during the build.

There are 4 predefined target configurations: default, default_mt, typical_pd and high_perf that can be selected via the -target=name option to veer.config. See configs/README.md for a description of these targets.

Building an FPGA speed optimized model: Use -fpga_optimize=1 option to veer.config to build a model that removes clock gating logic from flop model so that the FPGA builds can run at higher speeds. This is now the default option for targets other than typical_pd.

Building a Power optimized model (ASIC flows): Use -fpga_optimize=0 option to veer.config to build a model that enables clock gating logic into the flop model so that the ASIC flows get a better power footprint. This is now the default option for target typical_pd.

This script derives the following consistent set of include files:

$PWD/snapshots/<snapshot_name>
├── common_defines.vh                       # `defines for testbench or design
├── defines.h                               # #defines for C/assembly headers
├── eh2_param.vh                            # Design parameters
├── eh2_pdef.vh                             # Parameter structure
├── pd_defines.vh                           # `defines for physical design
├── perl_configs.pl                         # Perl %configs hash for scripting
├── pic_map_auto.h                          # PIC memory map based on configure size
└── whisper.json                            # JSON file for veer-iss
└── link.ld                                 # default linker control file

Building a model

While in a work directory:

1. Set the RV_ROOT environment variable to the root of the VeeR directory structure.

   Example for bash shell: export RV_ROOT=/path/to/veer Example for csh or its derivatives: setenv RV_ROOT /path/to/veer

2. Create your specific configuration

   (Skip if default is sufficient)
   (Name your snapshot to distinguish it from the default. Without an explicit name, it will update/override the default snapshot)

   For example if mybuild is the name for the snapshot, set the BUILD_PATH environment variable:

   setenv BUILD_PATH snapshots/mybuild

   And then:

   $RV_ROOT/configs/veer.config [configuration options..] -snapshot=mybuild

   Snapshots are placed in the $BUILD_PATH directory

3. Running a simple Hello World program (verilator)

   make -f $RV_ROOT/tools/Makefile target=default_mt

This command will build a verilator model of VeeR EH2 with AXI bus, and execute a short sequence of instructions that writes out "HELLO WORLD" to the bus.

The simulation produces output on the screen like:

VerilatorTB: Start of sim

-------------------------------
Hello World from VeeR EH2 hart0
-------------------------------
-------------------------------
Hello World from VeeR EH2 hart1
-------------------------------
TEST_PASSED

Finished hart0 : minstret = 1158, mcycle = 2895
Finished hart1 : minstret = 1733, mcycle = 2822
See "exec.log" for execution trace with register updates..

The simulation generates the following files:

• console.log contains what the cpu writes to the console address of 0xd0580000.
• exec.log shows instruction trace with GPR updates.
• trace_port.csv contains a log of the trace port.

When debug=1 is provided, a vcd file sim.vcd is created and can be browsed by gtkwave or similar waveform viewers.

You can re-execute simulation using: ./obj_dir/Vtb_top or make -f $RV_ROOT/tools/Makefile verilator

The simulation run/build command has the following generic form:

make -f $RV_ROOT/tools/Makefile [<simulator>] [debug=1] [snapshot=<snap>] [target=<target>] [TEST=<test>] [TEST_DIR=<path_to_test_dir>][CONF_PARAMS=<conf_switches>]

where:

<simulator> - can be 'verilator' (by default) 'irun' - Cadence xrun, 'vcs' - Synopsys VCS, 'vlog' - Mentor Questa
              if not provided, 'make' cleans work directory, builds verilator executable and runs a test.
debug=1     - allows VCD generation for verilator and VCS and SHM waves for irun option.
<target>    - predefined CPU configurations 'default' ( by default), 'default_mt', 'typical_pd', 'high_perf' 
TEST        - allows to run a C (<test>.c) or assembly (<test>.s) test, hello_world is run by default 
TEST_DIR    - alternative to test source directory testbench/asm
<snap>      - run and build executable model of custom CPU configuration, remember to provide 'snapshot' argument 
              for runs on custom configurations.
CONF_PARAMS - allows to provide veer.config command line arguments like -set=dccm_size=32 or -unset=iccm_enable..

Example:

make -f $RV_ROOT/tools/Makefile verilator TEST=cmark

will simulate the testbench/asm/cmark.c program with Verilator on the default target

If you want to compile a test only, you can run:

make -f $RV_ROOT/tools/Makefile program.hex TEST=<test> [TEST_DIR=/path/to/dir]

The Makefile uses $RV_ROOT/testbench/linker.ld file by default to build the test executable.
The user can provide a test-specific linker file in <test_name>.ld to build the test executable, in the same directory with the test source.

The user also can create a test-specific Makefile in <test_name>.makefile, contaning building instructions how to create program.hex files used by simulation. The private Makefile should be in the same directory as the test source. (the program.hex file is loaded to instruction and data bus memory slaves and optionally to DCCM/ICCM at the beginning of simulation).

Note: You may need to delete the program.hex file from the work directory, before running a new test.

The $RV_ROOT/testbench/asm directory contains the following tests ready to simulate:

hello_world       - default tes to run, prints Hello World message to screen and console.log
hello_world_dccm  - the same as above, but takes the string from preloaded DCCM.
hello_world_iccm  - the same as hello_world, but loads ICCM via LSU-DMA bridge and then executes from ICCM
cmark             - coremark benchmark running with code and data in external memories
cmark_dccm        - the same as above, running data and stack from DCCM (faster)
cmark_iccm        - the same as above, but preloading and running from ICCM 
cmark_mt          - coremark benchmark running with code and data in external memories for MT configs
cmark_dccm_mt     - the same as above, running data and stack from DCCM (faster) for MT configs
cmark_iccm_mt     - the same as above, but preloading and running from ICCM for MT configs
dhry              - dhrystone benchmark as example of multisouce program from testbench/tests/dhry directory
dhry_mt           - similar to above, but running two harts ( need to be run on MT configs )

The $RV_ROOT/testbench/hex directory contains precompiled hex files of the tests, ready for simulation in case RISC-V SW tools are not installed.

Note: The testbench has a simple synthesizable bridge that allows you to load the ICCM via load/store instructions. This is only supported for AXI4 builds.