Browse Prior Art Database

RISC Superscalar Workload Analysis Tool

IP.com Disclosure Number: IPCOM000112092D
Original Publication Date: 1994-Apr-01
Included in the Prior Art Database: 2005-Mar-26
Document File: 2 page(s) / 70K

Publishing Venue

IBM

Related People

Gariazzo, RE: AUTHOR [+2]

Abstract

The RISC Superscalar Workload Analysis Tool was developed to better understand the working environment of the Graphics Floating Point Engine.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 52% of the total text.

RISC Superscalar Workload Analysis Tool

      The RISC Superscalar Workload Analysis Tool was developed to
better understand the working environment of the Graphics Floating
<SR> Point Engine.

      The Graphics Floating Point Engine (or GFPE), is a superscalar
microprocessor based on the RISC System/6000* POWER PC architecture,
used in Graphics and Multimedia applications.  It is a single chip
implementation that contains an Instruction cache, a Data cache, a
Branch Processor, a Superscalar Unit, two Fixed Point Units, two Load
Store Units, and three Floating Point Units.

      The performance of any CPU is highly dependent on the
environment in which it operates.  This environment not only includes
the hardware functions it performs, but also to the instruction
streams that it processes.

      A RISC Superscalar implementation is very sensitive to the
instruction stream.  The combination of instructions, or the sequence
they have been scheduled at, will have a dramatic impact on the
efficiency of the CPU.  This performance dependency on the structure
of the instruction stream has created the necessity to meticulously
study the characteristics of the CPU's workload.  Knowledge obtained
is then used in performance studies and compiler design to speed
program execution.

      The solution presented in this disclosure is a software tool to
dissect the instruction stream that will be executed by the GFPE RISC
Superscalar Pipeline.  The goal is to learn the most about the most
common combination of instructions, their requirements on the
pipeline, their interaction between consecutive instructions, the
utilization of hardware resources, and their pipeline external
dependencies.

      Advantages - This tool gives an internal view of the operation
of...