Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

Multisequencing a Single Instruction Stream - Local Reshuffle of the Folded Dependency Depth Instruction Schedule

IP.com Disclosure Number: IPCOM000104691D
Original Publication Date: 1993-May-01
Included in the Prior Art Database: 2005-Mar-19
Document File: 4 page(s) / 196K

Publishing Venue

IBM

Related People

Ekanadham, K: AUTHOR [+5]

Abstract

Multisequencing a Single Instruction Stream (MSIS) is a uniprocessor organization in which a set of processing elements (PE) working in concert execute Segments of the instruction stream. The Segments are either P-Segments, normal uniprocessor instruction stream portions, that are processed in the E-MODE of MSIS and produce Z-Segments, or the Z-Segments that are processed in Z-MODE by MSIS. The main difference between E-MODE and Z-MODE is that during E-MODE each PE sees all instructions in the Segment and executes the ones that are assigned to it, but during Z-MODE, a PE only sees the instructions assigned to it.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 28% of the total text.

Multisequencing a Single Instruction Stream - Local Reshuffle of the Folded Dependency Depth Instruction Schedule

      Multisequencing a Single Instruction Stream (MSIS) is a
uniprocessor organization in which a set of processing elements (PE)
working in concert execute Segments of the instruction stream.  The
Segments are either P-Segments, normal uniprocessor instruction
stream portions, that are processed in the E-MODE of MSIS and produce
Z-Segments, or the Z-Segments that are processed in Z-MODE by MSIS.
The main difference between E-MODE and Z-MODE is that during E-MODE
each PE sees all instructions in the Segment and executes the ones
that are assigned to it, but during Z-MODE, a PE only sees the
instructions assigned to it.

      As all PEs see all instructions in E-MODE, each PE can create
the Z-CODE it will require to re-execute the Segment as a Z-Segment,
the Z-CODE being stored in the Z-CACHE, and associated with
instructions in the Z-CODE are S-LISTS and D-LISTS as appropriate.
An S-LIST instructs the PE, in the Z-MODE, that one or more of the
source registers in an instruction assigned to it is set by another
instruction that is executed on another PE, an  S-LIST is a receiving
obligation.  The D-LIST instructs the PE in the Z-MODE as to the
names of PEs that require the values of the register(s) that are
being set by an instruction that is assigned to it.  A D-LIST entry
is a sending obligation.

      The set of instructions assigned to a single PE can be further
delineated as THREADS.  A THREAD is a sequence of instructions in the
original conceptual order and a Thread is associated with a register
file which is either real or virtual.  There are no sending or
receiving obligations between instructions within a THREAD and the
THREAD is the smallest unit of aggregation of instructions from a
SEGMENT.

      The sequence of instructions that comprise a THREAD are in
conceptual sequence but THREADS can be interdigitated to form DECODER
STREAMS - the sequence of instructions that are issued by a single
DECODER.  Such a Z-CODE is called Out-Of-Sequence (OOS) Z-CODE.

      The interrelations between the instructions that set values of
registers and the instructions that use those values set can be
formulated in terms of a Directed Acyclic Graph, a DAG.  The nodes of
the DAG are instructions and the arcs represent interrelationships of
set/use.

      DETERMINING THE DDI - The DAG can be used to determine the
dependency depth (DDI) of each instruction.  The DDI corresponds to
the cycle of decode of an instruction when there are more PEs than
instructions and corresponds to the row index of the matrix that
represents the schedule in such a circumstance.  As instructions are
examined in conceptual sequence, in the order generated by the
sequence number, the assignment of the instruction based on DDI
represents the preferred way of breaking up the set of instructions
in the " &lambda.  sized chunks...