Boosting Cross-Architectural Emulation Performance by Foregoing the Intermediate Representation Model

TL;DR

This paper demonstrates that removing intermediate representations in cross-architecture emulation can significantly boost performance, proposing a new tiered translation system to optimize speed and development effort.

Contribution

It introduces a novel two-tier engine system for QEMU, replacing the intermediate representation with direct binary translation for common architecture pairs.

Findings

Emulator without intermediate representation is up to 35x faster than QEMU with TCG.

Proposed tiered system offers a flexible trade-off between performance and development effort.

Proof of concept shows substantial performance improvements in cross-architecture emulation.

Abstract

As more applications utilize virtualization and emulation to run mission-critical tasks, the performance requirements of emulated and virtualized platforms continue to rise. Hardware virtualization is not universally available for all systems, and is incapable of emulating CPU architectures, requiring software emulation to be used. QEMU, the premier cross-architecture emulator for Linux and some BSD systems, currently uses dynamic binary translation (DBT) through intermediate representations using its Tiny Code Generator (TCG) model. While using intermediate representations of translated code allows QEMU to quickly add new host and guest architectures, it creates additional steps in the emulation pipeline which decrease performance. We construct a proof of concept emulator to demonstrate the slowdown caused by the usage of intermediate representations in TCG; this emulator performed up to 35x faster than QEMU with TCG, indicating substantial room for improvement in QEMU's design. We propose an expansion of QEMU's two-tier engine system (Linux KVM versus TCG) to include a middle tier using direct binary translation for commonly paired architectures such as RISC-V, x86, and ARM. This approach provides a slidable trade-off between development effort and performance depending on the needs of end users.