Patent US4757444A successfully refreshed from Google Patents.
← Back to Patent Explorer
US4757444A Expired / Lapsed US ⏱️ 0.0 years remaining

Vector processor capable of performing iterative processing

Assignee: Hitachi Ltd
•
Filing: 1984-12-21
•
Granted: 1988-07-12
•
Expires: 2004-12-21
📄 Original PDF 🌐 Google Patents 🔄 Sync Now

Abstract

Abstract There is provided a vector processor based on a pipeline control method in which a cyclic operation is divided into a plurality of stages and processed. This processor comprises a vector register controller for dividing an operating process into a plurality of fundamental process units and controlling these units, and a phase generator for allowing the vector register controller to time-sharingly make the vector processor operative. This vector processor reads out data from vector registers in which vector elements are stored, operates this data and writes the result of operation into the vector register. With this vector processor, a cyclic operation can be processed in parallel at a high speed without causing a remarkable increase in hardware.

Inventors

  • 👤 Tomoo Aoyama
  • 👤 Yuuji Aoki
  • 👤 Hiroshi Murayama

Bibliographic Details

Application Number —
Jurisdiction US
Data Source google_patents
Last Synchronized 2026-10-26 20:50
CPC Classifications
G; G06; G06F; G06F15/00; G06F15/76; G06F15/80; G06F15/8053; G06F15/8076

Claims Analysis

Total Claims: 3
Claim 1
Independent Claim 321 words
A vector processor comprising: a plurality of vector registers for storing vector elements of respective vectors; an arithmetic logic unit for carrying out an operation between vector elements successively read out from said vector registers and for writing results of said operation between vector elements successively into one of said vector register; and control means for pipeline-controlling said arithmetic logic unit so that said operation performed thereby is divided into a plurality of stages and said vector elements for said plurality of stages are processed in parallel thereby performing an iterative operation between first vector data elements (x i ) and second vector data elements (a i ), where the first vector data elements (x i ) form a last operation result of said operation and the second vector data elements (a i ) are mutually independent from element to element; said control means including: (1) first latch means for holding the first vector data elements read from a vector register or supplied from said arithmetic logic unit; (2) second latch means for holding at least two succeeding vector data elements (a i-1 ) and (a i ) of the second vector data elements read from said vector register; (3) third latch means for holding the output of said arithmetic logic unit based on the operation on said second data elements (a i-1 ) and (a i ) held in said second latch means; (4) selection means for selecting vector data held in said first, second and third latch means and for supplying the selected data to said arithmetic logic unit; and (5) phase generating means for generating a phase signal which controls said selection means to select (a) the vector data elements (a i-1 ) and (a i ) of said second latch means or (b) the vector data elements of said first and third latch means on a time-sharing basis and to supply them to said arithmetic logic unit.
Claim 2
Dependent Claim 321 words
A vector processor comprising: a plurality of vector registers for storing vector elements of respective vectors; an arithmetic logic unit for carrying out an operation between vector elements successively read out from said vector registers and for writing results of said operation between vector elements successively into one of said vector register; and control means for pipeline-controlling said arithmetic logic unit so that said operation performed thereby is divided into a plurality of stages and said vector elements for said plurality of stages are processed in parallel thereby performing an iterative operation between first vector data elements (x i ) and second vector data elements (a i ), where the first vector data elements (x i ) form a last operation result of said operation and the second vector data elements (a i ) are mutually independent from element to element; said control means including: (1) first latch means for holding the first vector data elements read from a vector register or supplied from said arithmetic logic unit; (2) second latch means for holding at least two succeeding vector data elements (a i-1 ) and (a i ) of the second vector data elements read from said vector register; (3) third latch means for holding the output of said arithmetic logic unit based on the operation on said second data elements (a i-1 ) and (a i ) held in said second latch means; (4) selection means for selecting vector data held in said first, second and third latch means and for supplying the selected data to said arithmetic logic unit; and (5) phase generating means for generating a phase signal which controls said selection means to select (a) the vector data elements (a i-1 ) and (a i ) of said second latch means or (b) the vector data elements of said first and third latch means on a time-sharing basis and to supply them to said arithmetic logic unit.
Claim 3
Dependent Claim 63 words
A vector processor according to claim 1, further comprising: delay means for delaying said phase signal by a timing equal to the number of stages of said arithmetic logic unit; and selection means for controlling the output of said arithmetic logic unit to be supplied to either one of said first and third latch means based on the output of said delay means.

References Cited (Backward Citations)

US4128880A Cited
US4270181A Cited
US4302818A Cited
US4400768A Cited
US4435765A Cited
US4525796A Cited
US4507728A Cited

USPTO Annuity & Maintenance Windows

Annuity Stage Window Opens Payment Deadline Surcharge Period Until Estimated Fee Status
3.5 Years (Year 4) 1991-07-12 1992-01-11 1992-07-12 $2000.00 Window Closed / Lapsed
7.5 Years (Year 8) 1995-07-12 1996-01-11 1996-07-12 $3760.00 Window Closed / Lapsed
11.5 Years (Year 12) 1999-07-12 2000-01-11 2000-07-12 $7700.00 Window Closed / Lapsed

Assigned Categories & Tags

No categories assigned to this patent yet.

Assign Category Tag