Skip to content
STIMSMITH

AV-Swap

Concept

AV-Swap is an operation in generalized tandem simulation in which the values of instruction-level architectural variables (ILAVs) are transferred to the corresponding RTL architectural variables (RTAVs) of the RTL-based execution model (RTEM). It is performed by the AV-Converter component in Scenario 3 of the tandem simulation flow and is the mechanism that enables jump-starting the RTEM from the state produced by the ILEM, reducing the warm-up overhead of cross-level simulation.

First seen 6/6/2026
Last seen 6/6/2026
Evidence 3 chunks
Wiki v1

WIKI

AV-Swap

Definition

In the generalized tandem simulation methodology proposed by Xing, Gupta, and Malik (ASPDAC 2022), AV-Swap denotes the operation of copying the values of the instruction-level architectural variables (ILAVs) maintained by the instruction-level execution model (ILEM) into the corresponding RTL architectural variables (RTAVs) of the RTL-based execution model (RTEM). The correspondence between ILAVs and RTAVs is established by the AV map, an augmentation of the ILA refinement map.

READ FULL ARTICLE →

NEIGHBORHOOD

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

explore full graph →

RELATIONSHIPS

3 connections
Jump-Starting (Warm-Up Simulation) ← uses 97% 2e
Jump-starting uses AV-Swap to transfer architectural variable values
Tandem Simulation ← uses 97% 1e
Tandem simulation uses AV-Swap for jump-starting
AV-Converter ← implements 97% 1e
AV-Converter implements the AV-Swap functionality

CITATIONS

3 sources
3 citations — click to collapse
[1] AV-Swap is invoked in Scenario 3 of the tandem simulation flow to swap AVs and perform jump-start, and is implemented by the AV-Converter based on the AV map. Generalizing Tandem Simulation: Connecting High-level and RTL Simulation Models
[2] Tandem simulation allows swapping in values from ILAVs to RTAVs (AV-Swap), which can be leveraged to jump-start the RTEM in the middle of an ILEM simulation, significantly reducing simulation time by leaving the warm-up part of the test to only the ILEM. Generalizing Tandem Simulation: Connecting High-level and RTL Simulation Models
[3] AV-Swap is automated via a cold-start map: a cold start sets RTL micro-architectural variables to reset values, the AV map then sets RTAVs from ILAVs, and the pre-swap cycle/sequence and swap-cycle fields of the cold-start map handle reset and the holding time for swapped RTAVs. Generalizing Tandem Simulation: Connecting High-level and RTL Simulation Models