Patents by Inventor Wolf C. Witt

Wolf C. Witt 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: 8793299
    Abstract: A method and apparatus for including in a processor instructions for performing multiply-add operations on packed data. In one embodiment, a processor is coupled to a memory. The memory has stored therein a first packed data and a second packed data. The processor performs operations on data elements in said first packed data and said second packed data to generate a third packed data in response to receiving an instruction. At least two of the data elements in this third packed data storing the result of performing multiply-add operations on data elements in the first and second packed data.
    Type: Grant
    Filed: March 13, 2013
    Date of Patent: July 29, 2014
    Assignee: Intel Corporation
    Inventors: Alexander Peleg, Millind Mittal, Larry M. Mennemeier, Benny Eitan, Carole Dulong, Eiichi Kowashi, Wolf C. Witt
  • Patent number: 8745119
    Abstract: A method and apparatus for including in a processor instructions for performing multiply-add operations on packed data. In one embodiment, a processor is coupled to a memory. The memory has stored therein a first packed data and a second packed data. The processor performs operations on data elements in said first packed data and said second packed data to generate a third packed data in response to receiving an instruction. At least two of the data elements in this third packed data storing the result of performing multiply-add operations on data elements in the first and second packed data.
    Type: Grant
    Filed: March 13, 2013
    Date of Patent: June 3, 2014
    Assignee: Intel Corporation
    Inventors: Alexander Peleg, Millind Mittal, Larry M. Mennemeier, Benny Eitan, Carole Dulong, Eiichi Kowashi, Wolf C. Witt
  • Patent number: 8626814
    Abstract: A method and apparatus for including in a processor instructions for performing multiply-add operations on packed data. In one embodiment, a processor is coupled to a memory. The memory has stored therein a first packed data and a second packed data. The processor performs operations on data elements in said first packed data and said second packed data to generate a third packed data in response to receiving an instruction. At least two of the data elements in this third packed data storing the result of performing multiply-add operations on data elements in the first and second packed data.
    Type: Grant
    Filed: July 1, 2011
    Date of Patent: January 7, 2014
    Assignee: Intel Corporation
    Inventors: Alexander Peleg, Milland Mittal, Larry M. Mennemeier, Benny Eitan, Carole Dulong, Eiichi Kowashi, Wolf C. Witt
  • Publication number: 20130262836
    Abstract: A method and apparatus for including in a processor instructions for performing multiply-subtract operations on packed data. In one embodiment, a processor is coupled to a memory. The memory has stored therein a first packed data and a second packed data. The processor performs operations on data elements in said first packed data and said second packed data to generate a third packed data in response to receiving an instruction. At least two of the data elements in this third packed data storing the result of performing multiply-subtract operations on data elements in the first and second packed data.
    Type: Application
    Filed: May 30, 2013
    Publication date: October 3, 2013
    Inventors: Alexander Peleg, Milland Mittal, Larry M. Mennemeier, Benny Eitan, Carole Dulong, Eiichi Kowashi, Wolf C. Witt
  • Publication number: 20130262547
    Abstract: A method and apparatus for including in a processor instructions for performing multiply-add operations on packed data. In one embodiment, a processor is coupled to a memory. The memory has stored therein a first packed data and a second packed data. The processor performs operations on data elements in said first packed data and said second packed data to generate a third packed data in response to receiving an instruction. At least two of the data elements in this third packed data storing the result of performing multiply-add operations on data elements in the first and second packed data.
    Type: Application
    Filed: May 30, 2013
    Publication date: October 3, 2013
    Inventors: Alexander Peleg, Milland Mittal, Larry M. Mennemeier, Benny Eitan, Carole Dulong, Eiichi Kowashi, Wolf C. Witt
  • Publication number: 20130219151
    Abstract: A method and apparatus for including in a processor instructions for performing multiply-add operations on packed data. In one embodiment, a processor is coupled to a memory. The memory has stored therein a first packed data and a second packed data. The processor performs operations on data elements in said first packed data and said second packed data to generate a third packed data in response to receiving an instruction. At least two of the data elements in this third packed data storing the result of performing multiply-add operations on data elements in the first and second packed data.
    Type: Application
    Filed: March 13, 2013
    Publication date: August 22, 2013
    Inventors: Alexander Peleg, Millind Mittal, Larry M. Mennemeier, Benny Eitan, Carole Dulong, Eiichi Kowashi, Wolf C. Witt
  • Publication number: 20130198254
    Abstract: A method and apparatus for including in a processor instructions for performing multiply-add operations on packed data. In one embodiment, a processor is coupled to a memory. The memory has stored therein a first packed data and a second packed data. The processor performs operations on data elements in said first packed data and said second packed data to generate a third packed data in response to receiving an instruction. At least two of the data elements in this third packed data storing the result of performing multiply-add operations on data elements in the first and second packed data.
    Type: Application
    Filed: March 13, 2013
    Publication date: August 1, 2013
    Inventors: Alexander Peleg, Millind Mittal, Larry M. Mennemeier, Benny Eitan, Carole Dulong, Eiichi Kowashi, Wolf C. Witt
  • Patent number: 8495123
    Abstract: A method and apparatus for including in a processor instructions for performing multiply-add operations on packed data. In one embodiment, a processor is coupled to a memory. The memory has stored therein a first packed data and a second packed data. The processor performs operations on data elements in said first packed data and said second packed data to generate a third packed data in response to receiving an instruction. At least two of the data elements in this third packed data storing the result of performing multiply-add operations on data elements in the first and second packed data.
    Type: Grant
    Filed: October 1, 2012
    Date of Patent: July 23, 2013
    Assignee: Intel Corporation
    Inventors: Alexander Peleg, Millind Mittal, Larry M. Mennemeier, Benny Eitan, Carole Dulong, Eiichi Kowashi, Wolf C. Witt
  • Publication number: 20130091190
    Abstract: A method and apparatus for including in a processor instructions for performing multiply-add operations on packed data. In one embodiment, a processor is coupled to a memory. The memory has stored therein a first packed data and a second packed data. The processor performs operations on data elements in said first packed data and said second packed data to generate a third packed data in response to receiving an instruction. At least two of the data elements in this third packed data storing the result of performing multiply-add operations on data elements in the first and second packed data.
    Type: Application
    Filed: October 1, 2012
    Publication date: April 11, 2013
    Applicant: INTEL CORPORATION
    Inventors: Alexander Peleg, Millind Mittal, Larry M. Mennemeier, Benny Eitan, Carole Dulong, Eiichi Kowashi, Wolf C. Witt
  • Patent number: 8396915
    Abstract: A method and apparatus for including in a processor instructions for performing multiply-add operations on packed data. In one embodiment, a processor is coupled to a memory. The memory has stored therein a first packed data and a second packed data. The processor performs operations on data elements in said first packed data and said second packed data to generate a third packed data in response to receiving an instruction. At least two of the data elements in this third packed data storing the result of performing multiply-add operations on data elements in the first and second packed data.
    Type: Grant
    Filed: September 4, 2012
    Date of Patent: March 12, 2013
    Assignee: Intel Corporation
    Inventors: Alexander Peleg, Millind Mittal, Larry M. Mennemeier, Benny Eitan, Carole Dulong, Eiichi Kowashi, Wolf C. Witt
  • Patent number: 6275840
    Abstract: A method for detecting overflow in an add operation on first and second decoded bit-vectors is provided, the method including generating a one-ahead vector using the first decoded bit-vector. The method also includes selecting an overflow bit from bits of the one-ahead vector using the second decoded bit-vector.
    Type: Grant
    Filed: December 10, 1998
    Date of Patent: August 14, 2001
    Assignee: Intel Corporation
    Inventors: Umair A. Khan, Wolf C. Witt
  • Patent number: 5642306
    Abstract: A method and apparatus for multiple parallel multiplications of multiple packed data using a single multiplier is provided. Given multiple packed data as multiplicand blocks and as multiplier blocks, an early-out zero-skip feature examines a multiplicand block to be multiplied to determine if the multiplicand block consists of all zeros. If the multiplicand block consists of all zeros, then the corresponding multiplication is skipped. The early-out zero skip multiplier also examines the most significant bits of a multiplier block to be multiplied to determine if the most significant bits consist of all zeros. If the most significant bits of the multiplier block to be multiplied consist of all zeros, then the multiplicand block is multiplied with only the least significant bits of the corresponding multiplier block. Otherwise, if the most significant bits of the multiplier block consist of both zeros and ones, then the corresponding multiplicand block is multiplied with the entire multiplier block.
    Type: Grant
    Filed: May 15, 1996
    Date of Patent: June 24, 1997
    Assignee: Intel Corporation
    Inventors: Larry M. Mennemeier, Wolf C. Witt