Patents by Inventor Eric W. Mahurin
Eric W. Mahurin has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).
-
Patent number: 10656947Abstract: Receiving an instruction indicating a source operand and a destination operand. Storing a result in the destination operand in response to the instruction. The result operand may have: (1) first range of bits having a first end explicitly specified by the instruction in which each bit is identical in value to a bit of the source operand in a corresponding position; and (2) second range of bits that all have a same value regardless of values of bits of the source operand in corresponding positions. Execution of instruction may complete without moving the first range of the result relative to the bits of identical value in the corresponding positions of the source operand, regardless of the location of the first range of bits in the result. Execution units to execute such instructions, computer systems having processors to execute such instructions, and machine-readable medium storing such an instruction are also disclosed.Type: GrantFiled: March 14, 2018Date of Patent: May 19, 2020Assignee: Intel CorporationInventors: Maxim Loktyukhin, Eric W. Mahurin, Bret L. Toll, Martin G. Dixon, Sean P. Mirkes, David L. Kreitzer, Elmoustapha Ould-Ahmed-Vall, Vinodh Gopal
-
Patent number: 10579380Abstract: Receiving an instruction indicating a source operand and a destination operand. Storing a result in the destination operand in response to the instruction. The result operand may have: (1) first range of bits having a first end explicitly specified by the instruction in which each bit is identical in value to a bit of the source operand in a corresponding position; and (2) second range of bits that all have a same value regardless of values of bits of the source operand in corresponding positions. Execution of instruction may complete without moving the first range of the result relative to the bits of identical value in the corresponding positions of the source operand, regardless of the location of the first range of bits in the result. Execution units to execute such instructions, computer systems having processors to execute such instructions, and machine-readable medium storing such an instruction are also disclosed.Type: GrantFiled: December 12, 2014Date of Patent: March 3, 2020Assignee: INTEL CORPORATIONInventors: Maxim Loktyukhin, Eric W Mahurin, Bret L Toll, Martin G Dixon, Sean P Mirkes, David L Kreitzer, Elmoustapha Ould-Ahmed-Vall, Vinodh Gopal
-
Patent number: 10579379Abstract: Receiving an instruction indicating a source operand and a destination operand. Storing a result in the destination operand in response to the instruction. The result operand may have: (1) first range of bits having a first end explicitly specified by the instruction in which each bit is identical in value to a bit of the source operand in a corresponding position; and (2) second range of bits that all have a same value regardless of values of bits of the source operand in corresponding positions. Execution of instruction may complete without moving the first range of the result relative to the bits of identical value in the corresponding positions of the source operand, regardless of the location of the first range of bits in the result. Execution units to execute such instructions, computer systems having processors to execute such instructions, and machine-readable medium storing such an instruction are also disclosed.Type: GrantFiled: December 12, 2014Date of Patent: March 3, 2020Assignee: INTEL CORPORATIONInventors: Maxim Loktyukhin, Eric W Mahurin, Bret L Toll, Martin G Dixon, Sean P Mirkes, David L Kreitzer, Elmoustapha Ould-Ahmed-Vall, Vinodh Gopal
-
Patent number: 10372455Abstract: Receiving an instruction indicating a source operand and a destination operand. Storing a result in the destination operand in response to the instruction. The result operand may have: (1) first range of bits having a first end explicitly specified by the instruction in which each bit is identical in value to a bit of the source operand in a corresponding position; and (2) second range of bits that all have a same value regardless of values of bits of the source operand in corresponding positions. Execution of instruction may complete without moving the first range of the result relative to the bits of identical value in the corresponding positions of the source operand, regardless of the location of the first range of bits in the result. Execution units to execute such instructions, computer systems having processors to execute such instructions, and machine-readable medium storing such an instruction are also disclosed.Type: GrantFiled: December 12, 2014Date of Patent: August 6, 2019Assignee: Intel CorporationInventors: Maxim Loktyukhin, Eric W Mahurin, Bret L Toll, Martin G Dixon, Sean P Mirkes, David L Kreitzer, Elmoustapha Ould-Ahmed-Vall, Vinodh Gopal
-
Publication number: 20180203698Abstract: Receiving an instruction indicating a source operand and a destination operand. Storing a result in the destination operand in response to the instruction. The result operand may have: (1) first range of bits having a first end explicitly specified by the instruction in which each bit is identical in value to a bit of the source operand in a corresponding position; and (2) second range of bits that all have a same value regardless of values of bits of the source operand in corresponding positions. Execution of instruction may complete without moving the first range of the result relative to the bits of identical value in the corresponding positions of the source operand, regardless of the location of the first range of bits in the result. Execution units to execute such instructions, computer systems having processors to execute such instructions, and machine-readable medium storing such an instruction are also disclosed.Type: ApplicationFiled: March 14, 2018Publication date: July 19, 2018Applicant: Intel CorporationInventors: Maxim Loktyukhin, Eric W. Mahurin, Bret L. Toll, Martin G. Dixon, Sean P. Mirkes, David L. Kreitzer, Elmoustapha Ould-Ahmed-Vall, Vinodh Gopal
-
Patent number: 9092213Abstract: A method of performing vector operations on a semiconductor chip is described. The method includes performing a first vector instruction with a vector functional unit implemented on the semiconductor chip and performing a second vector instruction with the vector functional unit. The first vector instruction is a vector multiply add instruction. The second vector instruction is a vector leading zeros count instruction.Type: GrantFiled: September 24, 2010Date of Patent: July 28, 2015Assignee: Intel CorporationInventors: Jeff Wiedemeier, Sridhar Samudrala, Roger Golliver, Eric W. Mahurin
-
Publication number: 20150143084Abstract: Receiving an instruction indicating a source operand and a destination operand. Storing a result in the destination operand in response to the instruction. The result operand may have: (1) first range of bits having a first end explicitly specified by the instruction in which each bit is identical in value to a bit of the source operand in a corresponding position; and (2) second range of bits that all have a same value regardless of values of bits of the source operand in corresponding positions. Execution of instruction may complete without moving the first range of the result relative to the bits of identical value in the corresponding positions of the source operand, regardless of the location of the first range of bits in the result. Execution units to execute such instructions, computer systems having processors to execute such instructions, and machine-readable medium storing such an instruction are also disclosed.Type: ApplicationFiled: December 12, 2014Publication date: May 21, 2015Applicant: INTEL CORPORATIONInventors: Maxim Loktyukhin, Eric W. Mahurin, Bret L. Toll, Martin G. Dixon, Sean P. Mirkes, David L. Kreitzer, ELMOUSTAPHA OULD-AHMED-VALL, Vinodh Gopal
-
Patent number: 9009449Abstract: A system that executes program instructions on a processor is described. During a normal-execution mode, the system issues instructions for execution in program order. Upon encountering an unresolved data dependency during execution of an instruction, the system speculatively executes subsequent instructions in a lookahead mode to prefetch future loads. While executing in the lookahead mode, if the processor determines that the lookahead mode is unlikely to uncover any additional outer-level cache misses, the system terminates the lookahead mode. Then, after the unresolved data dependency is resolved, the system recommences execution in the normal-execution mode from the instruction that triggered the lookahead mode.Type: GrantFiled: November 10, 2011Date of Patent: April 14, 2015Assignee: Oracle International CorporationInventors: Yuan C. Chou, Eric W. Mahurin
-
Publication number: 20150100761Abstract: Receiving an instruction indicating a source operand and a destination operand. Storing a result in the destination operand in response to the instruction. The result operand may have: (1) first range of bits having a first end explicitly specified by the instruction in which each bit is identical in value to a bit of the source operand in a corresponding position; and (2) second range of bits that all have a same value regardless of values of bits of the source operand in corresponding positions. Execution of instruction may complete without moving the first range of the result relative to the bits of identical value in the corresponding positions of the source operand, regardless of the location of the first range of bits in the result. Execution units to execute such instructions, computer systems having processors to execute such instructions, and machine-readable medium storing such an instruction are also disclosed.Type: ApplicationFiled: December 12, 2014Publication date: April 9, 2015Applicant: INTEL CORPORATIONInventors: Maxim Loktyukhin, Eric W Mahurin, Bret L Toll, Martin G Dixon, Sean P Mirkes, David L Kreitzer, ELMOUSTAPHA OULD-AHMED-VALL, Vinodh Gopal
-
Publication number: 20150100760Abstract: Receiving an instruction indicating a source operand and a destination operand. Storing a result in the destination operand in response to the instruction. The result operand may have: (1) first range of bits having a first end explicitly specified by the instruction in which each bit is identical in value to a bit of the source operand in a corresponding position; and (2) second range of bits that all have a same value regardless of values of bits of the source operand in corresponding positions. Execution of instruction may complete without moving the first range of the result relative to the bits of identical value in the corresponding positions of the source operand, regardless of the location of the first range of bits in the result. Execution units to execute such instructions, computer systems having processors to execute such instructions, and machine-readable medium storing such an instruction are also disclosed.Type: ApplicationFiled: December 12, 2014Publication date: April 9, 2015Applicant: INTEL CORPORATIONInventors: Maxim Loktyukhin, Eric W Mahurin, Bret L Toll, Martin G Dixon, Sean P Mirkes, David L Kreitzer, ELMOUSTAPHA OULD-AHMED-VALL, Vinodh Gopal
-
Patent number: 9003170Abstract: Receiving an instruction indicating a source operand and a destination operand. Storing a result in the destination operand in response to the instruction. The result operand may have: (1) first range of bits having a first end explicitly specified by the instruction in which each bit is identical in value to a bit of the source operand in a corresponding position; and (2) second range of bits that all have a same value regardless of values of bits of the source operand in corresponding positions. Execution of instruction may complete without moving the first range of the result relative to the bits of identical value in the corresponding positions of the source operand, regardless of the location of the first range of bits in the result. Execution units to execute such instructions, computer systems having processors to execute such instructions, and machine-readable medium storing such an instruction are also disclosed.Type: GrantFiled: December 22, 2009Date of Patent: April 7, 2015Assignee: Intel CorporationInventors: Maxim Loktyukhin, Eric W. Mahurin, Bret L. Toll, Martin G. Dixon, Sean P. Mirkes, David L. Kreitzer, El Moustapha Ould-Ahmed-Vall, Vinodh Gopal
-
Patent number: 8918626Abstract: The disclosed embodiments relate to a system that executes program instructions on a processor. During a normal-execution mode, the system issues instructions for execution in program order. Upon encountering an unresolved data dependency during execution of an instruction, the system speculatively executes subsequent instructions in a lookahead mode to prefetch future loads. When an instruction retires during the lookahead mode, a working register which serves as a destination register for the instruction is not copied to a corresponding architectural register. Instead the architectural register is marked as invalid. Note that by not updating architectural registers during lookahead mode, the system eliminates the need to checkpoint the architectural registers prior to entering lookahead mode.Type: GrantFiled: November 10, 2011Date of Patent: December 23, 2014Assignee: Oracle International CorporationInventors: Yuan C. Chou, Eric W. Mahurin
-
Publication number: 20130124829Abstract: The disclosed embodiments relate to a system that executes program instructions on a processor. During a normal-execution mode, the system issues instructions for execution in program order. Upon encountering an unresolved data dependency during execution of an instruction, the system speculatively executes subsequent instructions in a lookahead mode to prefetch future loads. While executing in the lookahead mode, if the processor determines that the lookahead mode is unlikely to uncover any additional outer-level cache misses, the system terminates the lookahead mode. Then, after the unresolved data dependency is resolved, the system recommences execution in the normal-execution mode from the instruction that triggered the lookahead mode.Type: ApplicationFiled: November 10, 2011Publication date: May 16, 2013Applicant: ORACLE INTERNATIONAL CORPORATIONInventors: Yuan C. Chou, Eric W. Mahurin
-
Publication number: 20130124828Abstract: The disclosed embodiments relate to a system that executes program instructions on a processor. During a normal-execution mode, the system issues instructions for execution in program order. Upon encountering an unresolved data dependency during execution of an instruction, the system speculatively executes subsequent instructions in a lookahead mode to prefetch future loads. When an instruction retires during the lookahead mode, a working register which serves as a destination register for the instruction is not copied to a corresponding architectural register. Instead the architectural register is marked as invalid. Note that by not updating architectural registers during lookahead mode, the system eliminates the need to checkpoint the architectural registers prior to entering lookahead mode.Type: ApplicationFiled: November 10, 2011Publication date: May 16, 2013Applicant: ORACLE INTERNATIONAL CORPORATIONInventors: Yuan C. Chou, Eric W. Mahurin
-
Publication number: 20120079253Abstract: A method of performing vector operations on a semiconductor chip is described. The method includes performing a first vector instruction with a vector functional unit implemented on the semiconductor chip and performing a second vector instruction with the vector functional unit. The first vector instruction is a vector multiply add instruction. The second vector instruction is a vector leading zeros count instruction.Type: ApplicationFiled: September 24, 2010Publication date: March 29, 2012Inventors: Jeff Wiedemeier, Sridhar Samudrala, Roger Golliver, Eric W. Mahurin
-
Publication number: 20110153997Abstract: Receiving an instruction indicating a source operand and a destination operand. Storing a result in the destination operand in response to the instruction. The result operand may have: (1) first range of bits having a first end explicitly specified by the instruction in which each bit is identical in value to a bit of the source operand in a corresponding position; and (2) second range of bits that all have a same value regardless of values of bits of the source operand in corresponding positions. Execution of instruction may complete without moving the first range of the result relative to the bits of identical value in the corresponding positions of the source operand, regardless of the location of the first range of bits in the result. Execution units to execute such instructions, computer systems having processors to execute such instructions, and machine-readable medium storing such an instruction are also disclosed.Type: ApplicationFiled: December 22, 2009Publication date: June 23, 2011Inventors: Maxim Loktyukhin, Eric W. Mahurin, Bret L. Toll, Martin G. Dixon, Sean P. Mirkes, David L. Kreitzer, El Moustapha Ould-Ahmed-Vall, Vinodh Gopal
-
Patent number: 7401310Abstract: A method for designing integrated circuits may include custom designing, at the transistor level, individual cells to be incorporated into cell-based macros. A macro-level function of an integrated circuit's design specification requiring custom, transistor-level design may be identified and custom cells may be designed at the transistor-level to meet the design specification. Custom designed cells may be included in cell-based macros, thus allowing cell-based simulation and verification methodologies and tools to be used on the integrated circuit design. Static timing analysis, circuit extraction and other characteristics may be defined for each custom cell and the timing analysis and circuit extraction for cell-based macros may be defined based on the timing and extraction information for the custom cells included in the macro.Type: GrantFiled: April 4, 2006Date of Patent: July 15, 2008Assignee: Advanced Micro Devices, Inc.Inventors: Robert C. Thaden, Steven C. Hesley, Christopher B. Ang, Jeffery E. Short, Eric W. Mahurin
-
Patent number: 7242219Abstract: A circuit for a parity tree is disclosed. In one embodiment, a circuit for a parity tree includes a pull-up circuit, a pull-down circuit, and a cross-couple circuit. The circuit, an XOR/XNOR circuit, includes both an output node and an inverted output node. For a given set of input signals, a pull-up path exists through the cross-couple circuit for one of the output node and the inverted output node, and wherein a pull-down path exists through the cross-couple circuit for the other one of the output node and the inverted output node.Type: GrantFiled: September 8, 2005Date of Patent: July 10, 2007Assignee: Advanced Micro Devices, Inc.Inventors: Eric W. Mahurin, Dimitry Patent
-
Patent number: 6539470Abstract: An instruction decode unit is described including circuitry coupled to receive an instruction. The instruction identifies multiple operands, one of which is a destination operand. The circuitry responds to the instruction by producing: (i) operand codes specifying the operands, wherein the operand codes are produced in the order in which the operands are identified within the instruction, and (ii) a destination operand signal identifying the destination operand. In one embodiment, the decode unit responds to the instruction by producing the operand codes, operand address information, control signals, and the destination operand signal. A processor including the instruction decode unit is also described, as is a computer system including the processor. The instruction may include operand information which identifies the operands. The instruction may also include destination operand information which indicates which of the operands is the destination operand.Type: GrantFiled: November 16, 1999Date of Patent: March 25, 2003Assignee: Advanced Micro Devices, Inc.Inventors: Eric W. Mahurin, Brian D. McMinn
-
Patent number: 6535972Abstract: A system and method for shared dependency checking of status flags. In certain instruction set architectures, the reading of some status flags by the instruction set is exclusive, or nearly exclusive, with respect to the reading of other status flags. Hardware for exclusively read flags may be shared in dependency checking circuitry, allowing the reading of either one flag or the other for during dependency checking of a given instruction. This may allow circuit area to be saved. For an instruction that may need access to both flags, system firmware (e.g. microcode) can be used to break the instruction into two separate instructions or operations, thereby allowing the flags to maintain their exclusivity with respect to each other.Type: GrantFiled: November 16, 1999Date of Patent: March 18, 2003Assignee: Advanced Micro Devices, Inc.Inventors: Eric W. Mahurin, Michael T. Clark