Kim E O'Donnell

US Patent:

54210003, May 30, 1995

Filed:

Jul 26, 1993

Appl. No.:

8/097931

Inventors:

Ronald N. Fortino - Raleigh NC
Harry I. Linzer - Raleigh NC
Kim E. O'Donnell - Raleigh NC

Assignee:

International Business Machines Corp. - Armonk NY

International Classification:

G06F 1200

US Classification:

395425

Abstract:

A computer memory subsystem is comprised of one or more Dynamic Random Access Memory (DRAM) arrays with on-chip sense latches for storing data outputted from the DRAM, an on-chip Static Random Access Memory (SRAM) functioning as a Distributed Cache and an on-chip multiplexor. A first data bus interconnects the sense latches, the SRAM and the multiplexor. A second data bus interconnects the multiplexor and the SRAM. A memory controller generates signals which cause information to be extracted from the DRAM while the contents of the SRAM is unchanged or vice versa.

US Patent:

53718720, Dec 6, 1994

Filed:

Oct 28, 1991

Appl. No.:

7/783551

Inventors:

Larry D. Larsen - Raleigh NC
David W. Nuechterlein - Durham NC
Kim E. O'Donnell - Raleigh NC
Lee S. Rogers - Raleigh NC
Thomas A. Sartorius - Raleigh NC
Kenneth D. Schultz - Cary NC
Harry I. Linzer - Raleigh NC

Assignee:

International Business Machines Corporation - Armonk NY

International Classification:

G06F 1208

US Classification:

395425

Abstract:

The use of a high speed cache memory may be selectively controlled when a data processing task is interrupted in response to an interrupt signal, in order to prevent the interrupt from chilling the cache when insufficient performance enhancement will be realized. Disturbing the cache memory during performance of an interrupting task is prevented, thereby increasing the hit ratio of the cache when the interrupted task is resumed. Cache control information may be incorporated into a program status vector or program status word which is loaded into a program status register on occurrence of an interrupt.

US Patent:

50420348, Aug 20, 1991

Filed:

Oct 27, 1989

Appl. No.:

7/427549

Inventors:

Anthony Correale - Raleigh NC
Richard M. Doney - Durham NC
Kim E. O'Donnell - Raleigh NC
Andrew Kegl - Raleigh NC
Erwin A. Tate - Raleigh NC
David M. Wu - Melbourne FL

Assignee:

International Business Machines Corporation - Armonk NY

International Classification:

G06F 1100
G01R 3128

US Classification:

371 223

Abstract:

The present invention implements self testable boundary logic by using a tristate pass gate and a tristate receiver in combination with a linear feedback shift register, a storage register, and level sensitive scan design (LSSD) techniques. The linear feedback shift register (LFSR) shifts data into a storage register which is connected to the data inputs of the boundary logic through the tristate pass gate. The outputs of the tristate input receiver are also connected to the inputs of the boundary logic so that the boundary logic can receive data from both the data input of the integrated circuit (data path) or from the storage register connected to the LFSR. The tristate pass gate and receiver are enabled through a self test signal such that when the pass gate is enabled the receiver is not enabled and vice versa. In this way the boundary logic can only get data from either the storage register or through the receiver but not both. In this configuration data from the storage register can be input into the boundary logic without going through a multiplexer in the data path and incurring the associated delay.

US Patent:

53676530, Nov 22, 1994

Filed:

Dec 26, 1991

Appl. No.:

7/813971

Inventors:

William E. Coyle - Cary NC
David W. Nuechterlein - Durham NC
Kim E. O'Donnell - Raleigh NC
Thomas A. Sartorius - Raleigh NC
Kenneth D. Schultz - Cary NC
Emmy M. Wolters - Raleigh NC

Assignee:

International Business Machines Corporation - Armonk NY

International Classification:

G11C 1300

US Classification:

395400

Abstract:

A reconfigurable set associative cache memory can be reconfigured from 2. sup. x way to 2. sup. y way set associative cache memory by effectively merging a predetermined number of least significant bits of the tag field of the main memory address with the line field of the main memory address. The effective merging is provided by logically merging least significant bits of the tag field with a reconfiguration designation. As a result, Y-X+1 different configurations of cache memory can be obtained using the Y-X least significant bits of the tag field merged with the cache memory address.