CHERI Performance Enhancement for a Bytecode Interpreter

TL;DR

This paper identifies and addresses performance issues encountered when porting the MicroPython bytecode interpreter to the CHERI-based Arm Morello platform, achieving significant speedups by optimizing for larger capability pointers.

Contribution

It demonstrates how to mitigate software-induced runtime overhead caused by larger CHERI capabilities, improving interpreter performance on the Morello platform.

Findings

Reduced benchmark slowdown from 5x to 1.7x after optimization

Greatly improved worst-case slowdown from 100x to 2x

Showed that pointer size assumptions impact performance, not just correctness

Abstract

During our port of the MicroPython bytecode interpreter to the CHERI-based Arm Morello platform, we encountered a number of serious performance degradations. This paper explores several of these performance issues in detail, in each case we characterize the cause of the problem, the fix, and the corresponding interpreter performance improvement over a set of standard Python benchmarks. While we recognize that Morello is a prototypical physical instantiation of the CHERI concept, we show that it is possible to eliminate certain kinds of software-induced runtime overhead that occur due to the larger size of CHERI capabilities (128 bits) relative to native pointers (generally 64 bits). In our case, we reduce a geometric mean benchmark slowdown from 5x (before optimization) to 1.7x (after optimization) relative to AArch64, non-capability, execution. The worst-case slowdowns are greatly improved, from 100x (before optimization) to 2x (after optimization). The key insight is that implicit pointer size presuppositions pervade systems code; whereas previous CHERI porting projects highlighted compile-time and execution-time errors exposed by pointer size assumptions, we instead focus on the performance implications of such assumptions.