Dynamic coalescing for 16-bit instructions

The ARM processor is a dual width ISA processor that provides a 16-bit Thumb instruction set in addition to the 32-bit ARM instruction set. The compromises made in designing the Thumb instruction set leads to significantly increased instruction counts. This increase is in part due to the fact that only half of the register file is visible to most instructions in Thumb code. In this paper we address this inefficiency by providing a new instruction, the SetMask instruction, using which the compiler can change the visible subset of registers at any program point. Thus, through the use of this instruction the compiler can make use of all registers in all instructions. We present compiler techniques for allocating invisible registers and introducing SetMask instructions in a manner that the number of introduced instructions is minimized so that the increase in code size is insignificant. We implement this new instruction using the dynamic instruction coalescing framework which enables the SetMask instruction to have zero execution time cost. Our techniques eliminated 11.7% of MOV instructions from Thumb code while causing negligible code size increase.

The ARM processor core is a leading processor design for the embedded domain. In the embedded domain, both memory and energy are important concerns. For this reason the 32 bit ARM processor also supports the 16 bit Thumb instruction set. For a given program, typically the Thumb code is smaller than the ARM code. Therefore by using Thumb code the I-cache activity, and hence the energy consumed by the I-cache, can be reduced. However, the limitations of the Thumb instruction set, in comparison to the ARM instruction set, can often lead to generation of poorer quality code. Thus, while Thumb code may be smaller than ARM code, it may perform poorly and thus may not lead to overall energy savings.

The advancement of modern computing machines today is seen in embedded systems by making the system do more functions faster and in real time. One technique is to use multiple processors executing in parallel. Also, because embedded systems have limited memory size, adding more functions in the system will limit the data that can be stored in the memory. This project is an implementation of a compiler for an ARM7TDMI-S microprocessor with dual-core capabilities. Aside from the 32-bit ARM7 instructions, the compiler also includes 16-bit thumb instructions and atomic instructions as its target output. Using thumb instructions compresses the size of the program in the memory to increase the number of data that can be stored in it. Code optimizations and memory optimizations were applied to further take advantage of the use of dual-core microprocessors. Result shows that the compiler gives a good distribution of instructions to the two cores.

In this paper we concentrate on the effect of instruction set architecture on the performance potential of a computer system. These issues are key in considerations of what instruction set is most appropriate for the support of high level languages on general purpose machines. Two possible approaches are so called complex instruction sets, such as those of the VAX and IAPX 432, and the “reduced” instruction set of the RISC I [2] microcomputer, which is expected to have performance similar to that of a VAX 11-780. We propose a method of instruction set interpretation that takes advantage of the architectural features of complex instruction sets. These methods have been simulated executing real programs and in the case of the VAX instruction set have resulted in a typical improvement of a factor of two, assuming the same cycle time as the VAX 11-780. The techniques presented exploit the context available in a complex instruction and retain this information for use in subsequent execution of that instruction. Since this context is available only within a single instruction, low level instruction sets cannot benefit from this technique. In this sense the reduced instruction set machines as implemented in RISC are at their architectural limits, while complex instruction sets such as the VAX are far from theirs. For these reasons, complex instruction sets can have a significantly greater performance potential than a RISC instruction set for a given technology.

This paper summarizes a project on the implementation of the ARM9TDM, a 32-bit RISC processor based on the ARM9TDMI. This core is the successor to the ARM7TDMI-S which is used for embedded applications requiring low power, small chip area, and high processing speed. The main features of the ARM9TDM are its use of a 5-stage pipelined datapath and a Harvard architecture that has separate data and instruction interfaces. It supports the ARMv4T instruction set architecture (ISA) that uses both the 32-bit ARM instructions and 16-bit Thumb instructions. It includes a high-speed multiplier and debug capabilities using JTAG boundary scan test interface. It does not include an EmbeddedICE-RT module. The project was coded using the Verilog Hardware Description Language and was simulated using Synopsys VCS. The verified code was synthesized in 0.25-micrometer standard cells using Synopsys Design Vision. The layout generated by Synopsys Astro was characterized as having a maximum operating frequency of 34.13 MHz, an average power consumption of 16 mW and a chip size of 1.5335 sq. mm.

The use of code compression in embedded systems based on standard RISC instruction set architectures (ISA) has been shown in the past to be of benefit in reducing overall system cost. The 16-bit THUMB ISA from ARM Ltd has a significantly higher density than the original 32-bits ARM ISA. Our proposed memory compression architecture has showed a further size reduction of 15% to 20% on the THUMB code. In this paper we propose to use a high-speed data lossless hardware decompressor to improve the timing performance of the architecture. We simulated the architecture on the SimpleScalar platform and show that for some applications, the time overheads are limited within 5% of the original application.

Today, as deep learning (DL) is applied more often in daily life, dedicated processors such as CPUs and GPUs have become very important for accelerating model executions. With the growth of technology, people are becoming accustomed to using edge devices, such as mobile phones, smart watches, and VR devices in their daily lives. A variety of technologies using DL are gradually being applied to these edge devices. However, there is a large number of computations in DL. It faces a challenging problem how to provide solutions in the edge devices. In this article, the proposed method enables a flow with the RISC-V Packed extension (P extension) in TVM. TVM, an open deep learning compiler for neural network models, is growing as a key infrastructure for DL computing. RISC-V is an open instruction set architecture (ISA) with customized and flexible features. The Packed-SIMD extension is a RISC-V extension that enables subword single-instruction multiple-data (SIMD) computations in RISC-V architectures to support fallback engines in AI computing. In the proposed flow, a fixed-point type that is supported by an integer of 16-bit type and saturation instructions is added to replace the original 32-bit float type. In addition, an auto-tuning method is proposed to use a uniform selector mechanism (USM) to find the binary point position for fixed-point type use. The tensorization feature of TVM can be used to optimize specific hardware such as subword SIMD instructions with RISC-V P extension. With our experiment on the Spike simulator, the proposed method with the USM can improve performance by approximately 2.54 to 6.15× in terms of instruction counts with little accuracy loss.

Recent advances in different instruction set architectures (ISAs) and their implementations have revived the argument on the role of ISAs in the overall performance and energy efficiency of a processor. Many computer architects believe that with current compiler and microarchitecture developments, the choice of ISA is not a decisive matter anymore. On the other hand, some believe that ISAs can still play an important role in the overall performance and energy efficiency of a computer system. Our objective is to compare and contrast ISAs by finding the differences in performance and energy consumption across ISAs and the reasons behind those differences. Our work shows that ISAs affect the performance and energy efficiency of applications differently based on their inherent characteristics.

This paper examines the causes and extent of code size overhead caused by the ARM calling convention in Thumb-2 binaries. We show that binaries generated from C++ source files generally have higher amounts of calling convention overhead, and present a binary file optimizer to eliminate some of that overhead. Calling convention overhead can negatively impact power consumption, flash memory costs, and chip size in embedded or otherwise resource-constrained domains. This is particularly true on platforms using "compressed" instruction sets, such as the 16-bit ARM Thumb and Thumb-2 instruction sets, used in virtually all smartphones and in many other smaller-scale embedded devices. In this paper, we examine the extent of calling convention overhead in practical software, and compare the results of C and C++ programs, and find that C++ programs generally have a higher percentage of calling-convention overhead. Finally, we demonstrate a tool capable of eliminating some of this overhead, particularly in the case of C++ programs, by modifying the calling conventions on a per-procedure basis.

A stack-based instruction set which includes two-operand, one-operand and zero-operand instructions is proposed, and is shown to produce a more compact code and fewer instructions for evaluating assignment statements than one-operand and zero-operand stack instruction sets. If this instruction set is then extended to include two-operand memory-to-memory instructions, the combined instruction set also produces more compact code than a simple memory-to-memory instruction set, for a set of typical assignment statements weighed by the frequency of their occurrence in real programs. Finally various alternatives for optimizing the proposed instruction set are discussed.

For more than fifty years, computer engineers have sought to construct minimal computers using only a single instruction computer. While it might appear to be a simple academic exercise, remarkably, a rich computation paradigm can be developed using this approach, with important applications and implications in reconfigurable, chemical, optical and biological computing. More recently, the widespread use of the Field Programmable Gate Array (FPGA) has made such an approach not only desirable, but also practical. In this paper the history and motivation behind single instruction or one instruction computing (OISC) is reviewed. It is then shown how the paradigm can be used to implement a variety of imaging operations efficiently. Finally, a practical application and future work in languages and tools are presented.

To the present day, the performance of microprocessors has progressed dramatically. Recently, almost all computer systems use reduced instruction set computer (RISC) architectures. However, about 30 years ago, complex instruction set computer (CISC) architectures were widely used for almost all computer systems. The advantages and successes of RISC architectures are attributable to their simplified structures. Conventional complex instruction set computer (CISC) architectures invariably included various and numerous instruction sets. Each instruction was able to execute a complicated multi-step operation. For that reason, the CISC architectures were useful in assembler-based programming environments and in systems with small amounts of memory. However, such complicated architectures prevent increases in clock frequency or a processor’s processing power. Therefore, RISC architectures—which use simple architectures based on single-step instruction sets—have been developed. The RISC architectures present advantages in terms of higher clock frequency, smaller implementation area, and lower power consumption than conventional complex instruction set computer (CISC) architectures. Observation of many examples reveals that, in circuit implementations, a simple structure is best to increase the overall performance. That principle is also applicable to programmable devices. If clock-by-clock reconfigurable devices are used, even a single instruction set computer (SISC) can be implemented onto them. A single instruction set computer is one in which a processor has only a single instruction. During production, various single instruction set computers are prepared: a single instruction set computer with an AND logic function, a single instruction set computer with an adder function, and so on. These processor units are implemented at necessary times and at necessary places of a programmable device. In CISC and RISC architectures, the hardware is fixed. Its operations are switched using software commands, as portrayed in Fig. 1(a). In contrast, in a single instruction set computer, the operation changes are executed by hardware reconfigurations, as shown in Figs. 1(b) and 1(c). Therefore, in a single instruction set computer, a processor with a certain function itself can be reconfigured to another processor with another function. The implementation of such single instruction set computers provides the following advantages under programmable device implementations. A single instruction set computer with the simplest architecture can operate at the highest clock frequency among all processor architectures. In RISC architectures, many selectors to change functions are implemented; such selectors have a certain delay. However, single instruction set computers

The ever-increasing demand for faster execution time, smaller resource usage and lower energy consumption has compelled architects of embedded processors to adopt more specialized hardware features with irregular data paths and heterogeneous registers that are customized to the needs of their target applications. These processors consequently provide a rich set of specialized instructions in order to enable programmers to access these features. Such an instruction is typically a multi-output instruction (MOI), which outputs multiple results parallely in order to exploit inherent underlying hardware parallelism. Earlier study has exhibited that MOIs help to enhance performance in aspect of instruction counts and code size. However, as MOIs require more operands, they tend to increase not only the size of the instruction set but also the size of individual instructions. This can be a serious setback for embedded processors, which are mostly subject to strong resource limitations (particularly in this case, limited instruction encoding space). For this reason, these processors are often allowed to include only a very small subset of the total desired MOIs in their instruction sets, despite there can be sufficient silicon real estate to accommodate these specialized MOIs. To attack this problem, we introduce a novel instruction encoding scheme based on the dynamic implied addressing mode (DIAM). In this paper, we will discuss how we have overcome the encoding space problem for our target embedded processor whose instruction set has been augmented with a variety of MOIs. Our DIAM-based encoding scheme employs a small on-chip buffer to supplement extra encoding information for MOIs at run time. The empirical results are promising: the scheme allows us to encode many more MOIs for our processor; thereby helping us to achieve considerable reduction of code size as well as running time after the DIAM is additively implemented in the original architecture.

This paper analyses the performance of the state-of-the-art media ISA (instruction set architecture) extensions in a general-purpose processor, when executing a video encoder based on an affine motion model. In addition to SIMD (single instruction multiple data) fixed-point instructions, these ISA extensions include SIMD floating-point instructions, special-purpose SIMD fixed-point instructions, and cacheability control instructions. In this study, eight time-consuming kernels of the video encoder were hand-optimized, using instructions in all four instruction categories of these media ISA extensions (the FLP version). These kernels were also hand-optimized using only SIMD fixed-point ISA extensions, without special-purpose instructions (the FXP version). For the FLP version, this study resulted in an average kernel-level speedup of 1.37X and an application-level speedup of 1.11X, compared to the FXP version, and an application-level speedup of 3.41X, compared to the C version.

An oil/water, 3-D fully implicit Reservoir Simulator has been implemented on the ICL Distributed Array Processor (DAP), by a joint BNOC/ICL project team. The DAP is a single instruction, multiple data stream machine with 4096 processors in parallel. The major computational areas of code have been implemented on the DAP, leaving I/O and well calculations as serial code on the host 2976 machine. Matrix assembly is inherently parallel and can be implemented easily and efficiently on the DAP with compact code. Table look-up, though not obviously parallel, has been coded using a very efficient parallel algorithm. The Linear Solver, of Line Gauss-Seidel type, has been chosen for its efficiency on the DAP, with odd and even reduction used for solution of the resulting tri-diagonal equations. Further research will be undertaken to increase the robustness of this solver. The initial implementation has a 1-1 mapping of active grid blocks to processors, thus allowing up to 4096 active grid blocks. It is planned to extend to much larger models, and to 3 phase.