Related papers: Analyzing Latency Hiding and Parallelism in an MLI…

Computation intensive kernels, such as convolutions, matrix multiplication and Fourier transform, are fundamental to edge-computing AI, signal processing and cryptographic applications. Interleaved-Multi-Threading (IMT) processor cores are…

In the era of diminishing returns from Moores Law, heterogeneous computing systems have emerged as a vital approach to enhance computational efficiency. This paper introduces a novel MLIR-based dialect, named hyper, designed to optimize…

Current AI code generation systems suffer from significant latency bottlenecks due to CPU-GPU data transfers during compilation, execution, and testing phases. We establish theoretical foundations for three complementary approaches to…

The rapidly evolving landscape of AI and machine learning workloads has widened the gap between high-level domain operations and efficient hardware utilization. Achieving near-peak performance still demands deep hardware expertise-experts…

Existing iterative compilation and machine-learning-based optimization techniques have been proven very successful in achieving better optimizations than the standard optimization levels of a compiler. However, they were not engineered to…

Instruction-Level Redundancy (ILR) is a well-known approach to tolerate transient CPU faults. It replicates instructions in a program and inserts periodic checks to detect and correct CPU faults using majority voting, which essentially…

Hardware architectures and machine learning (ML) libraries evolve rapidly. Traditional compilers often fail to generate high-performance code across the spectrum of new hardware offerings. To mitigate, engineers develop hand-tuned kernels…

Modern out-of-order processors have increased capacity to exploit instruction level parallelism (ILP) and memory level parallelism (MLP), e.g., by using wide superscalar pipelines and vector execution units, as well as deep buffers for…

This work proposes a compilation flow using open-source compiler passes to build a framework to achieve ninja performance from a generic linear algebra high-level abstraction. We demonstrate this flow with a proof-of-concept MLIR project…

Creating high performance implementations of deep learning primitives on CPUs is a challenging task. Multiple considerations including multi-level cache hierarchy, and wide SIMD units of CPU platforms influence the choice of program…

Edge computing's growing prominence, due to its ability to reduce communication latency and enable real-time processing, is promoting the rise of high-performance, heterogeneous System-on-Chip solutions. While current approaches often…

In this paper, we present Hexagon-MLIR,an open-source compilation stack that targets Qualcomm Hexagon Neural Processing Unit (NPU) and provides unified support for lowering Triton kernels and PyTorch models . Built using the MLIR framework,…

Reinforcement learning (RL) workloads take a notoriously long time to train due to the large number of samples collected at run-time from simulators. Unfortunately, cluster scale-up approaches remain expensive, and commonly used CPU…

Modern GPU workloads, especially large language model (LLM) inference, suffer from kernel launch overheads and coarse synchronization that limit inter-kernel parallelism. Recent megakernel techniques fuse multiple operators into a single…

Multi-level intermediate representations (MLIR) show great promise for reducing the cost of building domain-specific compilers by providing a reusable and extensible compiler infrastructure. This work presents TPU-MLIR, an end-to-end…

This report presents some early results on code generation targeting tensor cores on NVIDIA GPUs using the MLIR compiler infrastructure. The state-of-the-art in high-performance deep learning today is primarily driven by manually optimized…

Neural networks are increasingly used in real-time systems, such as automated driving applications. This requires high-performance hardware with predictable timing behavior. State-of-the-art real-time hardware is limited in memory and…

Optimizing deep learning models is generally performed in two steps: (i) high-level graph optimizations such as kernel fusion and (ii) low level kernel optimizations such as those found in vendor libraries. This approach often leaves…

Driven by the increasing demand for low-latency and real-time processing, machine learning applications are steadily migrating toward edge computing platforms, where Field-Programmable Gate Arrays (FPGAs) are widely adopted for their energy…

Computation in-memory is a promising non-von Neumann approach aiming at completely diminishing the data transfer to and from the memory subsystem. Although a lot of architectures have been proposed, compiler support for such architectures…