Skip to content
STIMSMITH

UVM Based Design Verification of a RISC-V CPU Core

Paper

“UVM Based Design Verification of a RISC-V CPU Core” is a technical work on CPU-core verification methods, centered on UVM/SystemVerilog verification practice for RISC-V designs. The supplied evidence shows that it motivates CPU verification as a difficult state-space problem, explains verification planning and UVM testbench concepts, and surveys open-source RISC-V verification environments such as Ibex and Core-V-Verif, including co-simulation with Spike, random instruction generation, coverage planning, scoreboards, BSP-compatible test programs, and RVFI-based checking.

First seen 5/27/2026
Last seen 5/28/2026
Evidence 12 chunks
Wiki v1

WIKI

Overview

“UVM Based Design Verification of a RISC-V CPU Core” is a technical work available as a POLITesi PDF that discusses verification of RISC-V CPU cores using UVM-oriented design-verification practices. The work frames CPU verification as a demanding problem because processors are complex state machines with many states and corner cases; verification must check instruction correctness, exception handling, memory accesses, timing behavior, and expected functional outcomes. It also notes that the practical number of possible test cases can be far beyond exhaustive exploration, making systematic verification strategy essential. [C1]

Verification planning

READ FULL ARTICLE →

NEIGHBORHOOD

No graph connections found for this entity yet. It may appear in future ingestion runs.

explore full graph →

RELATIONSHIPS

27 connections
Scoreboard uses → 95% 2e
The paper uses a scoreboard for checking functional correctness in the testbench.
Constrained Random Verification uses → 90% 2e
The paper uses constrained random verification as part of its verification methodology.
formal verification uses → 75% 2e
The paper discusses formal verification as a technique used in CPU verification.
Testbench uses → 100% 2e
The paper presents a UVM-based testbench for RISC-V verification.
Instruction Set Compliance Testing uses → 95% 2e
The paper uses the RISC-V compliance test suite for instruction set compliance testing.
Ibex Core mentions → 95% 2e
The paper discusses the Ibex Core as a state-of-the-art RISC-V CPU verification example.
Core-V-Verif mentions → 95% 2e
The paper discusses Core-V-Verif as a state-of-the-art verification project.
Co-simulation uses → 90% 2e
The paper employs co-simulation methodology using Spike as a reference model.
Direct Tests uses → 100% 1e
The paper uses specifically designed direct tests for RISC-V core verification.
Benchmarks uses → 100% 1e
The paper uses benchmarks to evaluate the functional performance of the RISC-V core.
RISC-V Toolchain uses → 100% 1e
The paper integrates the RISC-V toolchain as part of its verification infrastructure.
Fuzzing uses → 90% 1e
The paper includes fuzzing as part of its experimental evaluation.
AHB Verification IP uses → 90% 1e
The paper includes AHB verification IP in its testbench architecture.
Dhrystone evaluates → 95% 1e
The paper evaluates the RISC-V core using the Dhrystone benchmark.
Coremark evaluates → 95% 1e
The paper evaluates the RISC-V core using the Coremark benchmark.
Functional Coverage uses → 90% 1e
The paper implements a functional coverage model in the verification infrastructure.
Verification Test Plan uses → 90% 1e
The paper discusses and employs a verification test plan as part of the verification infrastructure.
RISC-V Formal Interface uses → 85% 1e
The paper uses the RISC-V Formal Interface to provide information about retired instructions for co-simulation checking.
Renato Occhineri authored by → 100% 1e
The thesis paper is authored by Renato Occhineri.
CV32E40P mentions → 90% 1e
The paper mentions CV32E40P as the primary target core of the Core-V-Verif environment.
Franco Zappa authored by → 95% 1e
Franco Zappa is the academic advisor of the thesis.
UVM uses → 100% 1e
The paper presents a UVM-based verification infrastructure for a RISC-V core.
SystemVerilog uses → 100% 1e
The paper uses SystemVerilog for the verification infrastructure.
RISC-V evaluates → 100% 1e
The paper targets verification of a RISC-V CPU core.
spike uses → 100% 1e
The paper incorporates Spike as a RISC-V ISS for co-simulation validation.
Random Instruction Generator uses → 100% 1e
The paper includes a random instruction generator in the verification infrastructure.
Coverage-Driven Verification uses → 100% 1e
The paper employs coverage-driven verification enabled by UVM and SystemVerilog.

CITATIONS

8 sources
8 citations — click to expand
[1] C1: The work motivates CPU verification as difficult because CPUs are complex state machines requiring checks of instruction correctness, exceptions, memory accesses, timing, and functional outcomes, while exhaustive testing is impractical. [PDF] UVM based design veri cation of a RISC-V CPU core - POLITesi
[2] C2: A verification test plan derives from the design specification, records features/configurations and combinations to verify, and commonly uses constrained-random or coverage-driven simulation, with formal verification for selected areas. [PDF] UVM based design veri cation of a RISC-V CPU core - POLITesi
[3] C3: UVM is presented as a SystemVerilog class framework with drivers, monitors, stimulus generators, scoreboards, sequence items, components, factory substitution, and phased execution. [PDF] UVM based design veri cation of a RISC-V CPU core - POLITesi
[4] C4: The work states that proprietary CPU verification practices are often opaque, while open-source RISC-V enables examination of production-ready verification processes. [PDF] UVM based design veri cation of a RISC-V CPU core - POLITesi
[5] C5: Ibex is described as a SystemVerilog 32-bit RISC-V core verified with a UVM-based co-simulation testbench using Spike, RISC-DV-generated binaries, randomized stimuli, and test/coverage planning. [PDF] UVM based design veri cation of a RISC-V CPU core - POLITesi
[6] C6: Core-V-Verif is described as an OpenHW functional-verification project for CORE-V RISC-V cores, initially focused on CV32E40P, using UVM/SystemVerilog class libraries in a vendor-independent simulation environment. [PDF] UVM based design veri cation of a RISC-V CPU core - POLITesi
[7] C7: Core-V-Verif test programs must be BSP-compatible, may be pre-existing or generated and self-checking or non-self-checking, and checker-monitors can fail simulations by issuing uvm_error. [PDF] UVM based design veri cation of a RISC-V CPU core - POLITesi
[8] C8: The evidence states that Spike is required for the co-simulation system and that RVFI supplies information about retired instructions and synchronous traps for checking. [PDF] UVM based design veri cation of a RISC-V CPU core - POLITesi