Saved in:
Bibliographic Details
Main Author: Pedroni, Felipe T.
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2605.24026
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910249983672320
author Pedroni, Felipe T.
author_facet Pedroni, Felipe T.
contents We present a formal bounding analysis for maximum credible interference in multicore processors under strict architectural invariants: direct-mapped L2 cache (1-way associativity), disabled Miss Status Handling Registers (MSHRs), single-bank main memory, deterministic pinned tasks with fixed physical memory mapping, and a pessimistic L2/memory arbitration policy. We prove that, under these invariants, the per-critical-access stall imposed on a target task T is bounded above by (N-1)Lmem, and that this bound is attained by a synchronized adversarial workload of N-1 congruent-different-tag memory requests issued in phase with T's critical accesses. The argument is per-access and direct, requiring no informal multiplicative interference function. The derivation is purely analytical and discussed in the context of DO-178C/CAST-32A certification objectives for airborne software. Limitations and conditions for applicability are explicitly stated. This work provides a traceable method for separating multicore interference from Worst-Case Execution Time (WCET) budgets under fixed architectural constraints.
format Preprint
id arxiv_https___arxiv_org_abs_2605_24026
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle A Per-Access Upper Bound for Shared-Resource Interference in Direct-Mapped Multicore Architectures
Pedroni, Felipe T.
Hardware Architecture
Operating Systems
Performance
C.1.4; B.3; C.3
We present a formal bounding analysis for maximum credible interference in multicore processors under strict architectural invariants: direct-mapped L2 cache (1-way associativity), disabled Miss Status Handling Registers (MSHRs), single-bank main memory, deterministic pinned tasks with fixed physical memory mapping, and a pessimistic L2/memory arbitration policy. We prove that, under these invariants, the per-critical-access stall imposed on a target task T is bounded above by (N-1)Lmem, and that this bound is attained by a synchronized adversarial workload of N-1 congruent-different-tag memory requests issued in phase with T's critical accesses. The argument is per-access and direct, requiring no informal multiplicative interference function. The derivation is purely analytical and discussed in the context of DO-178C/CAST-32A certification objectives for airborne software. Limitations and conditions for applicability are explicitly stated. This work provides a traceable method for separating multicore interference from Worst-Case Execution Time (WCET) budgets under fixed architectural constraints.
title A Per-Access Upper Bound for Shared-Resource Interference in Direct-Mapped Multicore Architectures
topic Hardware Architecture
Operating Systems
Performance
C.1.4; B.3; C.3
url https://arxiv.org/abs/2605.24026