Skip to content
STIMSMITH

DifuzzRTL: Differential Fuzz Testing to Find CPU Bugs

Paper
First seen 6/28/2026
Last seen 6/28/2026
Evidence 1 chunks

NEIGHBORHOOD

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

explore full graph →

RELATIONSHIPS

13 connections
DiFuzzRTL introduces → 99% 1e
The paper introduces the DIFUZZRTL tool.
Register-Coverage Guided Fuzzing introduces → 98% 1e
The paper presents a new register-coverage guided fuzzing technique.
Unified CPU input format with TileLink protocols introduces → 97% 1e
The paper introduces a unified CPU input format leveraging TileLink protocols.
Cycle-sensitive register coverage guiding introduces → 97% 1e
Cycle-sensitive register coverage guiding is introduced as part of the new techniques.
Asynchronous interrupt events handling introduces → 96% 1e
Asynchronous interrupt events handling is introduced as a new technique in the paper.
Drop-in-replacement designs introduces → 95% 1e
Drop-in-replacement designs are introduced to support various CPU RTLs.
Differential Fuzz Testing uses → 90% 1e
The study applies differential fuzz testing to find CPU bugs.
OpenRISC Mor1kx Cappuccino evaluates → 98% 1e
The evaluation includes the OpenRISC Mor1kx Cappuccino RTL core.
RISC-V Rocket Core evaluates → 98% 1e
The evaluation includes the RISC-V Rocket Core.
RISC-V BOOM Core evaluates → 98% 1e
The evaluation includes the RISC-V BOOM Core.
DiFuzzRTL evaluates → 78% 1e
The paper evaluates the proposed DIFUZZRTL by applying it to real-world CPU RTLs.
DiFuzzRTL uses → 85% 1e
The authors use their DIFUZZRTL tool in the study.
TileLink mentions → 92% 1e
TileLink protocols are mentioned as part of the unified input format.