SearcharxivSearch

arXiv · 2511.19258

IOMMU Support for Virtual-Address Remote DMA in an ARMv8 environment

Abstract

In complex systems with many compute nodes containing multiple CPUs that are coherent within each node, a key challenge is maintaining efficient and correct coherence between nodes. The Unimem system addresses this by proposing a virtualized global address space that enables such coherence, relying on the I/O Memory Management Unit (IOMMU) in each node. The goal of this thesis is to support this approach by successfully testing and using the IOMMU of a single node. For this purpose, we used ARM's IOMMU, known as the System Memory Management Unit (SMMU), which translates virtual addresses to physical addresses. Because Linux documentation for the SMMU is limited and unclear, we implemented custom kernel modules to test and use its functionality. First, we tested the SMMU in the Processing System (PS) of the Xilinx Zynq UltraScale+ MPSoC by developing a module that inserted virtual-to-physical address mappings into the SMMU. We then triggered a DMA transfer to a virtual address and observed that the request passed through the SMMU for address translation. We repeated this experiment by initiating DMA transactions from the Programmable Logic (PL) and similarly confirmed that the transactions were translated by the SMMU. Finally, we developed a module that enables transactions from the PL without requiring explicit pre-mapping of virtual and physical address pairs. This was achieved by configuring the SMMU with the page table pointer of a user process, allowing it to translate all relevant virtual addresses dynamically. Overall, we successfully demonstrated the correct operation of the SMMU across all tested scenarios. Due to time constraints, further exploration of advanced SMMU features is left for future work.

Explore related subjects

Keep this discovery

BibTeXRIS

Antonis Psistakis. 2025-11-24. IOMMU Support for Virtual-Address Remote DMA in an ARMv8 environment. https://arxiv.org/abs/2511.19258

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

Online Treasure Hunt in Vertex-Permuted Dynamic Rings

We study the problem of treasure hunt by a group of $k \geq 1$ agents in vertex-permuted dynamic rings (VP). In this model, the $n$ vertices remain on a ring but are permuted at each time step. We first show that treasure hunt is impossible for any $k \leq n-3$ agents, if there are no restrictions on the sequence of permutations used in the dynamic ring. We then study the $VP(\delta)$ setting, in which for every pair $i, j$ of vertices, the edge $(i, j)$ is guaranteed to appear within $\delta$ steps. We show that the class $VP(\delta)$ is feasible only for $\delta \geq \left\lceil \frac{n-1}{2}\right\rceil$. For the one-agent case, we show a tight bound of $\Theta(\delta n)$ on the worst-case search time as well as competitive ratio of any online algorithm for treasure hunt, provided $\delta \geq 2n$. We then give an optimal algorithm for $k$ agents, thereby showing that $k$ agents can obtain a speedup of $k$ on the worst-case search time. Finally, in the R-VP setting, in which in every step, the vertices are arranged as a ring according to a random permutation, we show that treasure hunt takes expected $\Theta(n)$ steps against an oblivious adversary and $\Theta(n \log n)$ steps against an adaptive adversary.

cs.DC

The Computing Channel: How Modulation Programs the Airwaves

Distributed computing and distributed artificial intelligence require frequent exchanges of intermediate results, although many applications need only an aggregate rather than messages from individual devices. Conventional systems recover each message before computing the aggregate, whereas over-the-air computation (OAC) exploits simultaneous transmission to obtain it directly. However, dominant OAC implementations rely on analog signaling, creating a mismatch with finite-precision data and digital communication procedures. This article presents digital function-oriented communication, in which finite-alphabet symbol representations and receiver decisions are jointly designed so that multiple-access superposition encodes the desired function without recovering individual inputs. We introduce its computational-constellation principle, main design approaches, extensions, and implementation challenges. Federated edge learning illustrates how the framework can reduce user-dependent data-bearing resources while operating directly on quantized model updates.

cs.DC

Can AI Remediate Backend Failures Safely? GuardedAct with Blast-Radius-Aware Sandboxing

Large Language Models (LLMs) have shown promising capabilities in generating remediation actions for microservice failures. However, directly executing AI-generated repair actions in production risks cascading collateral damage. We propose GuardedAct, a sandbox-first remediation framework that interposes a blast-radius-aware verification layer between the LLM action generator and the production environment. GuardedAct operates in four phases: (1) ingesting a diagnosis report together with the live system topology and recent telemetry, (2) prompting an LLM to produce a ranked list of candidate remediation actions, (3) simulating each action in a lightweight digital-twin sandbox that estimates the blast radius and assigns a risk label, and (4) enforcing a rollback-confidence gate that auto-executes only low-risk actions while escalating high-risk ones for human review. We evaluate GuardedAct on five fault scenarios injected into the DeathStarBench social-network application. Experimental results show that GuardedAct achieves an overall recovery rate of 87.4% while reducing collateral damage by 79.7% relative to direct LLM execution (from 25.6% to 5.2%), at the cost of a modest sandbox-induced increase in mean time to recovery (approximately 8 s). Ablation studies confirm that each component contributes meaningfully to the safety-speed trade-off.

cs.DC