Jonathan Bachrach

Chisel: constructing hardware in a Scala embedded language

In this paper we introduce Chisel, a new hardware construction language that supports advanced hardware design using highly parameterized generators and layered domain-specific hardware languages. By embedding Chisel in the Scala programming language, we raise the level of hardware design abstraction by providing concepts including object orientation, functional programming, parameterized types, and type inference. Chisel can generate a high-speed C++-based cycle-accurate software simulator, or low-level Verilog designed to map to either FPGAs or to a standard ASIC flow for synthesis. This paper presents Chisel, its embedding in Scala, hardware examples, and results for C++ simulation, Verilog emulation and ASIC synthesis.

An Agile Approach to Building RISC-V Microprocessors

The final phase of CMOS technology scaling provides continued increases in already vast transistor counts, but only minimal improvements in energy efficiency, thus requiring innovation in circuits and architectures. However, even huge teams are struggling to complete large, complex designs on schedule using traditional rigid development flows. This article presents an agile hardware development methodology, which the authors adopted for 11 RISC-V microprocessor tape-outs on modern 28-nm and 45-nm CMOS processes in the past five years. The authors discuss how this approach enabled small teams to build energy-efficient, cost-effective, and industry-competitive high-performance microprocessors in a matter of months. Their agile methodology relies on rapid iterative improvement of fabricatable prototypes using hardware generators written in Chisel, a new hardware description language embedded in a modern programming language. The parameterized generators construct highly customized systems based on the free, open, and extensible RISC-V platform. The authors present a case study of one such prototype featuring a RISC-V vector microprocessor integrated with a switched-capacitor DC-DC converter alongside an adaptive clock generator in a 28-nm, fully depleted silicon-on-insulator process.

Towards a minimal architecture for a printable, modular, and robust sensing skin

This work presents a low-complexity modular sensor grid architecture to provide a smart skin to non-convex shapes, such as a robot body and legs. To configure a sensing skin shaped by arbitrary cuts and rapid changes in designs, we use a wavefront planning approach to generate a minimum-depth spanning tree of an arbitrary topology of contiguous, regularly arranged modular sensing units on a flexible substrate wired network. A Finite State Machine protocol for extracting this topology and sensor information is shown that is robust to destructive sensor loss, device failure, and transmission noise. The architecture is designed to require as little state complexity at each node as possible to minimize the area and cost of such a network implemented in printable semiconductor technology. Simulation data show recovery from network failures and extension of the architecture to larger networks with arbitrary geometry, and a sample synthesis of the verified architecture logic is shown to have a very low state and combinational logic complexity. A proof-of-concept implementation of the architecture using microcontrollers and optical proximity sensors on a flexible substrate show integration with a Scaled Composite Manufacturing process used for Biomimetic Millirobots.

Strober: fast and accurate sample-based energy simulation for arbitrary RTL

This paper presents a sample-based energy simulation methodology that enables fast and accurate estimations of performance and average power for arbitrary RTL designs. Our approach uses an FPGA to simultaneously simulate the performance of an RTL design and to collect samples containing exact RTL state snapshots. Each snapshot is then replayed in gate-level simulation, resulting in a workload-specific average power estimate with confidence intervals. For arbitrary RTL and workloads, our methodology guarantees a minimum of four-orders-of-magnitude speedup over commercial CAD gate-level simulation tools and gives average energy estimates guaranteed to be within 5% of the true average energy with 99% confidence. We believe our open-source sample-based energy simulation tool Strober can not only rapidly provide ground truth for more abstract power models, but can enable productive design-space exploration early in the RTL design process.

A generator of memory-based, runtime-reconfigurable 2N3M5K FFT engines

Runtime-reconfigurable, mixed-radix FFT/IFFT engines are essential for modern wireless communication systems. To comply with varying standards requirements, these engines are customized for each modem. The Chisel hardware construction language has been used in this work to create a generator of runtime-reconfigurable 2n3m5k FFT engines targeting software-defined radios (SDR) for modern communications, but with flexibility to support a wide range of applications. The generator uses a conflict-free, in-place, multi-bank SRAM design, and exploits the duality of decimation-in-frequency (DIF) and decimation-in-time (DIT) FFTs to support continuous data flow with only 2N memory blocks. DFT decomposition using the prime-factor algorithm (PFA) followed by the Cooley-Tukey algorithm (CTA) reduces twiddle ROM sizes. A programmable Winograds Fourier Transform (WFTA) butterfly supporting radix-2/3/4/5/7 operations reuses radix-7 hardware to support reconfigurability with minimal area penalty. The generated FFTs use 50% less memory than iterative FFTs from Spiral. The twiddle ROM size of the generated LTE/WiFi FFT engine is 16% smaller than that of a 2048-pt Spiral design.

A Hardware Accelerator for Computing an Exact Dot Product

We study the implementation of a hardware accelerator that computes a dot product of IEEE-754 floating-point numbers exactly. The accelerator uses a wide (640 or 4288 bits for single or double-precision respectively) fixed-point representation into which intermediate floating-point products are accumulated. We designed the accelerator as a generator in Chisel, which can synthesize various configurations of the accelerator that make different area-performance trade-offs.We integrated eight different configurations into an SoC comprised of RISC-V in-order scalar core, split L1 instruction and data caches, and unified L2 cache. In a TSMC 45 nm technology, the accelerator area ranges from 0.05 mm2 to 0.32 mm2, and all configurations could be clocked at frequencies in excess of 900MHz. The accelerator successfully saturates the SoCs memory system, achieving the same per-element efficiency (1 cycle-per-element) as Intel MKL running on an x86 machine with a similar cache configuration.

Firesim: FPGA-accelerated cycle-exact scale-out system simulation in the public cloud

We present FireSim, an open-source simulation platform that enables cycle-exact microarchitectural simulation of large scale-out clusters by combining FPGA-accelerated simulation of silicon-proven RTL designs with a scalable, distributed network simulation. Unlike prior FPGA-accelerated simulation tools, FireSim runs on Amazon EC2 F1, a public cloud FPGA platform, which greatly improves usability, provides elasticity, and lowers the cost of large-scale FPGA-based experiments. We describe the design and implementation of FireSim and show how it can provide sufficient performance to run modern applications at scale, to enable true hardware-software co-design. As an example, we demonstrate automatically generating and deploying a target cluster of 1,024 3.2 GHz quad-core server nodes, each with 16 GB of DRAM, interconnected by a 200 Gbit/s network with 2 microsecond latency, which simulates at a 3.4 MHz processor clock rate (less than 1,000x slowdown over real-time). In aggregate, this FireSim instantiation simulates 4,096 cores and 16 TB of memory, runs ~14 billion instructions per second, and harnesses 12.8 million dollars worth of FPGAs—at a total cost of only ~

Efficient sampling of SAT solutions for testing

100 per simulation hour to the user. We present several examples to show how FireSim can be used to explore various research directions in warehouse-scale machine design, including modeling networks with high-bandwidth and low-latency, integrating arbitrary RTL designs for a variety of commodity and specialized datacenter nodes, and modeling a variety of datacenter organizations, as well as reusing the scale-out FireSim infrastructure to enable fast, massively parallel cycle-exact single-node microarchitectural experimentation.

ACED: a hardware library for generating DSP systems

In software and hardware testing, generating multiple inputs which satisfy a given set of constraints is an important problem with applications in fuzz testing and stimulus generation. However, it is a challenge to perform the sampling efficiently, while generating a diverse set of inputs which satisfy the constraints. We developed a new algorithm QuickSampler which requires a small number of solver calls to produce millions of samples which satisfy the constraints with high probability. We evaluate QuickSampler on large real-world benchmarks and show that it can produce unique valid solutions orders of magnitude faster than other state-of-the-art sampling tools, with a distribution which is reasonably close to uniform in practice.

Specialization for energy efficiency using agile development

Designers translate DSP algorithms into application-specific hardware via primitives composed in various ways for different architectural realizations. Despite sharing underlying algorithms and hardware constructs, designs are often difficult to reuse, leading to redeveloping/reverifying conceptually similar instances. Hardware generators are attractive solutions for effectively balancing fine-grained control of implementation details with simple, retargetable hardware descriptions. This work presents ACED, a hardware library for generating DSP systems. It extends the Chisel hardware construction language and FIRRTL compiler and operates on three principles: zero-cost abstraction, unobtrusive downstream optimization/specialization promoting generator reusability, and unified, portable systems modeling and verification.