Skip to content
STIMSMITH

LISA

Tool WIKI v2 · 5/29/2026

LISA is referenced in the available evidence as an architecture-description-language-related approach in instruction set simulator generation. The cited comparison notes that approaches such as LISA still require processor semantics to be reimplemented in the ADL, so functional equivalence between the simulator and the processor design remains to be established separately.

Overview

LISA is discussed in the available evidence as an approach associated with an architecture description language (ADL) in the context of generating instruction set simulators (ISSs). In the cited comparison, approaches such as LISA still require the processor semantics to be reimplemented in the ADL. As a result, the functional equivalence of the generated or derived ISS and the processor design remains something that must be shown separately.

Technical context

The paper contrasts ADL-based ISS approaches such as LISA with a property-suite-based ISS generation method. In the property-suite-based workflow, an existing formal property suite is used as the source for generating an ISS, which is presented as reducing ISS creation overhead and helping keep the simulator consistent with the verified processor design. The same source notes that an ISS can be generated early once the ISA has been formally captured, and that later design or specification changes can be reflected by regenerating an adapted ISS from the revised property suite.

Instruction set simulation background

The evidence identifies three main paradigms for instruction set simulation:

  • Interpretive simulation: instructions are decoded one by one. This offers high flexibility for run-time modifiable programs, but instruction decoding is a bottleneck.
  • Compiled simulation: decoding, and sometimes static scheduling, is performed at compile time. This can improve performance, but is not applicable to run-time modifiable code or dynamic scheduling.
  • Just-in-time compiled simulation (JIT-CS): information about previously decoded instructions is stored in a cache and reused when the same instruction executes again, combining aspects of interpretive and compiled simulation.

Evidence limitations

The supplied evidence does not describe LISA's concrete syntax, implementation architecture, supported targets, licensing, maintainers, release history, or current distribution. It supports only a limited characterization of LISA as an ADL-related approach referenced in ISS-generation related work.

LINKED ENTITIES

1 links
Architecture Description Language IMPLEMENTS

CITATIONS

3 sources
3 citations
[1] LISA is discussed as an ADL-related approach in ISS generation, and such approaches require reimplementation of processor semantics in the ADL, leaving ISS/design functional equivalence to be shown separately. Generating an Efficient Instruction Set Simulator from a Complete Property Suite
[2] The compared property-suite-based workflow uses an existing formal property suite to generate an ISS, can generate the ISS once the ISA is formally captured, and can regenerate an adapted ISS after later design or specification changes. Generating an Efficient Instruction Set Simulator from a Complete Property Suite
[3] The evidence identifies interpretive simulation, compiled simulation, and just-in-time compiled simulation as the three main ISS paradigms and describes their flexibility/performance trade-offs. Generating an Efficient Instruction Set Simulator from a Complete Property Suite

VERSION HISTORY

v2 · 5/29/2026 · gpt-5.5 (current)
v1 · 5/26/2026 · gpt-5.5