As the volume of data processed by applications has increased, considerable attention has been paid to data address translation overheads, leading to the widespread use of larger page sizes (“superpages”) and multi-level translation lookaside buffers (TLBs). However, far less attention has been paid to instruction address translation and its relation to TLB and pipeline structure. In prior work, we quantified the impact of using code superpages on a variety of widely used applications, ranging from compilers to web user-interface frameworks, and the impact of sharing page table pages for executables and shared libraries. Within this article, we augment those results by first uncovering the effects that microarchitectural differences between Intel Skylake and AMD Zen+, particularly their different TLB organizations, have on instruction address translation overhead. This analysis provides some key insights into the microarchitectural design decisions that impact the cost of instruction address translation. First, a lower-level (level 2) TLB that has both instruction and data mappings competing for space within the same structure allows better overall performance and utilization when using code superpages. Code superpages not only reduce instruction address translation overhead but also indirectly reduce data address translation overhead. In fact, for a few applications, the use of just a few code superpages has a larger impact on overall performance than the use of a much larger number of data superpages. Second, a level 1 (L1) TLB with separate structures for different page sizes may require careful tuning of the superpage promotion policy for code, and a correspondingly suboptimal utilization of the level 2 TLB. In particular, increasing the number of superpages when the size of the L1 superpage structure is small may result in more L1 TLB misses for some applications. Moreover, on some microarchitectures, the cost of these misses can be highly variable, because replacement is delayed until all of the in-flight instructions mapped by the victim entry are retired. Hence, more superpage promotions can result in a performance regression. Finally, our findings also make a case for first-class OS support for superpages on ordinary files containing executables and shared libraries, as well as a more aggressive superpage policy for code.
more »
« less
PDL: a high-level hardware design language for pipelined processors
Processors are typically designed in Register Transfer Level (RTL) languages, which give designers low-level control over circuit structure and timing. To achieve good performance, processors are pipelined, with multiple instructions executing concurrently in different parts of the circuit. Thus even though processors implement a fundamentally sequential specification (the instruction set architecture), the implementation is highly concurrent. The interactions of multiple instructions---potentially speculative---can cause incorrect behavior.
We present PDL, a novel hardware description language targeted at the construction of pipelined processors. PDL provides one instruction at a time semantics: the first language to enforce that the generated pipelined circuit has the same behavior as a sequential specification. This enforcement facilitates design-space exploration. Adding or removing pipeline stages, moving operations across stages, or otherwise changing pipeline structure normally requires careful analysis of bypass paths and stall logic; with PDL, this analysis is handled by the PDL compiler. At the same time, PDL still offers designers fine-grained control over performance-critical microarchitectural choices such as timing of operations, data forwarding, and speculation. We demonstrate PDL's expressive power and ease of design exploration by implementing several RISC-V cores with differing microarchitectures. Our results show that PDL does not impose significant performance or area overhead compared to a standard HDL.
more »
« less
- Award ID(s):
- 1717554
- NSF-PAR ID:
- 10389500
- Date Published:
- Journal Name:
- ACM SIGPLAN Conf. on Programming Language Design and Implementation (PLDI)
- Page Range / eLocation ID:
- 719 to 732
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Bertoni G.M., Regazzoni F. (Ed.)The design of software countermeasures against active and passive adversaries is a challenging problem that has been addressed by many authors in recent years. The proposed solutions adopt a theoretical foundation (such as a leakage model) but often do not offer concrete reference implementations to validate the foundation. Contributing to the experimental dimension of this body of work, we propose a customized processor called SKIVA that supports experiments with the design of countermeasures against a broad range of implementation attacks. Based on bitslice programming and recent advances in the literature, SKIVA offers a flexible and modular combination of countermeasures against power-based and timing-based side-channel leakage and fault injection. Multiple configurations of side-channel protection and fault protection enable the programmer to select the desired number of shares and the desired redundancy level for each slice. Recurring and security-sensitive operations are supported in hardware through custom instruction-set extensions. The new instructions support bitslicing, secret-share generation, redundant logic computation, and fault detection. We demonstrate and analyze multiple versions of AES from a side-channel analysis and a fault-injection perspective, in addition to providing a detailed performance evaluation of the protected designs. To our knowledge, this is the first validated end-to-end implementation of a modular bitslice-oriented countermeasure.more » « less
-
The tidal waves of modern electronic/electrical devices have led to increasing demands for ubiquitous application-specific power converters. A conventional manual design procedure of such power converters is computation- and labor-intensive, which involves selecting and connecting component devices, tuning component-wise parameters and control schemes, and iteratively evaluating and optimizing the design. To automate and speed up this design process, we propose an automatic framework that designs custom power converters from design specifications using Monte Carlo Tree Search. Specifically, the framework embraces the upper-confidence-bound-tree (UCT), a variant of Monte Carlo Tree Search, to automate topology space exploration with circuit design specification-encoded reward signals. Moreover, our UCT-based approach can exploit small offline data via the specially designed default policy and can run in parallel to accelerate topology space exploration. Further, it utilizes a hybrid circuit evaluation strategy to substantially reduce design evaluation costs. Empirically, we demonstrated that our framework could generate energy-efficient circuit topologies for various target voltage conversion ratios. Compared to existing automatic topology optimization strategies, the proposed method is much more computationally efficient—the sequential version can generate topologies with the same quality while being up to 67% faster. The parallelization schemes can further achieve high speedups compared to the sequential version.more » « less
-
Moghaddam, Mohsen ; Marion, Tucker ; Holtta-Otto, Katja ; Fu, Kate ; Olechowski, Alison ; McComb, Christopher (Ed.)The early-stage product design and development (PDD) process fundamentally involves the processing, synthesis, and communication of a large amount of information to make a series of key decisions on design exploration and specification, concept generation and evaluation, and prototyping. Although most current PDD practices depend heavily on human intuition, advances in computing, communication, and human–computer interaction technologies can transform PDD processes by combining the creativity and ingenuity of human designers with the speed and precision of computers. Emerging technologies like artificial intelligence (AI), cloud computing, and extended reality (XR) stand to substantially change the way designers process information and make decisions in the early stages of PDD by enabling new methods such as natural language processing, generative modeling, cloud-based virtual collaboration, and immersive design and prototyping. These new technologies are unlikely to render the human designer obsolete, but rather do change the role that the human designer plays. Thus, it is essential to understand the designer's role as an individual, a team, and a group that forms an organization. The purpose of this special issue is to synthesize the state-of-the-art research on technologies and methods that augment the performance of designers in the front-end of PDD—from understanding user needs to conceptual design, prototyping, and development of systems architecture while also emphasizing the critical need to understand the designer and their role as well.more » « less
-
Dense matrix multiply (MM) serves as one of the most heavily used kernels in deep learning applications. To cope with the high computation demands of these applications, heterogeneous architectures featuring both FPGA and dedicated ASIC accelerators have emerged as promising platforms. For example, the AMD/Xilinx Versal ACAP architecture combines general-purpose CPU cores and programmable logic (PL) with AI Engine processors (AIE) optimized for AI/ML. An array of 400 AI Engine processors executing at 1 GHz can theoretically provide up to 6.4 TFLOPs performance for 32-bit floating-point (fp32) data. However, machine learning models often contain both large and small MM operations. While large MM operations can be parallelized efficiently across many cores, small MM operations typically cannot. In our investigation, we observe that executing some small MM layers from the BERT natural language processing model on a large, monolithic MM accelerator in Versal ACAP achieved less than 5% of the theoretical peak performance. Therefore, one key question arises: How can we design accelerators to fully use the abundant computation resources under limited communication bandwidth for end-to-end applications with multiple MM layers of diverse sizes? We identify the biggest system throughput bottleneck resulting from the mismatch of massive computation resources of one monolithic accelerator and the various MM layers of small sizes in the application. To resolve this problem, we propose the CHARM framework to compose multiple diverse MM accelerator architectures working concurrently towards different layers within one application. CHARM includes analytical models which guide design space exploration to determine accelerator partitions and layer scheduling. To facilitate the system designs, CHARM automatically generates code, enabling thorough onboard design verification. We deploy the CHARM framework for four different deep learning applications, including BERT, ViT, NCF, MLP, on the AMD/Xilinx Versal ACAP VCK190 evaluation board. Our experiments show that we achieve 1.46 TFLOPs, 1.61 TFLOPs, 1.74 TFLOPs, and 2.94 TFLOPs inference throughput for BERT, ViT, NCF, MLP, respectively, which obtain 5.40x, 32.51x, 1.00x and 1.00x throughput gains compared to one monolithic accelerator.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.1145/3519939.3523455
