Energy Efficient Register Renaming 1
نویسندگان
چکیده
Modern microprocessor designs implement register renaming using register alias tables (RATs), which maintain the mapping between architectural and physical registers. Because of the non–trivial power that is dissipated in a disproportionately small area, the power density in the RAT is significantly higher than in some other datapath components. In this paper, we propose mechanisms to reduce the RAT power and the power density by exploiting the fundamental observation that most of the generated register values are used by the instructions in close proximity to the instruction producing a value. Our first technique disables the RAT lookup for a source register if that register is a destination of an earlier instruction dispatched in the same cycle. The second technique eliminates some of the remaining RAT read accesses even if the source register value is produced by an instruction dispatched in an earlier cycle. This is done by buffering a small number of recent register address translations in a set of external latches and satisfying some RAT lookup requests from these latches. The net result of applying both techniques is a 30% reduction in the RAT energy with no performance penalty, little additional complexity and no cycle time degradation.
منابع مشابه
Energy Efficient Application Specific Banked Register Files
Register files account for a significant fraction of the power dissipation in modern RISC processors. Register file banking is an effective alternative to monolithic register files in embedded systems. We propose a profile-based technique to arrive at a customized energy-efficient bank configuration for a given application on a dual bank register file. The technique consists of a register renam...
متن کاملAsymmetrically banked value-aware register files for low-energy and high-performance
Designing high-performance low-energy register files is of critical importance to the continuation of current performance advances in wide-issue and deeply pipelined superscalar microprocessors. In this paper, we propose a new microarchitecture, the asymmetrically banked value-aware register file (AB-VARF), to exploit the prevailing narrow-width register values for low-latency and energy-effici...
متن کاملReducing Register Pressure Through LAER Algorithm
When modern processors keep increasing the instruction window size and the issue width to exploit more instruction-level parallelism (ILP), the demand of larger physical register file is also on the increase. As a result, register file access time represents one of the critical delays and can easily become a bottleneck. In this paper, we first discuss the possibilities of reducing register pres...
متن کاملThe Design Space of Register Renaming Techniques
Register renaming is a technique to remove false data dependencies—write after read (WAR) and write after write (WAW)— that occur in straight line code between register operands of subsequent instructions. By eliminating related precedence requirements in the execution sequence of the instructions, renaming increases the average number of instructions that are available for parallel execution p...
متن کاملThe Design of a Register Renaming Unit
Register renaming is often used to improve performance in many high-ILP processors. However, there is a lack of publications regarding register renaming hardware design. This paper presents a detailed look at one possible implementation of a register renaming unit, as well as some possible optimizations.
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2003