Search Images Maps Play YouTube News Gmail Drive More »
Sign in
Screen reader users: click this link for accessible mode. Accessible mode has the same essential features but works better with your reader.

Patents

  1. Advanced Patent Search
Publication numberUS3711692 A
Publication typeGrant
Publication dateJan 16, 1973
Filing dateMar 15, 1971
Priority dateMar 15, 1971
Publication numberUS 3711692 A, US 3711692A, US-A-3711692, US3711692 A, US3711692A
InventorsBatcher K
Original AssigneeGoodyear Aerospace Corp
Export CitationBiBTeX, EndNote, RefMan
External Links: USPTO, USPTO Assignment, Espacenet
Determination of number of ones in a data field by addition
US 3711692 A
Abstract
An arrangement for counting the number of a given data such as ones in a data field. The system employs full adders which count the number of ones in basic three element subsets of the data field. Additional full adders total the results of the first series of full adders and also count any additional ones of the data field.
Images(2)
Previous page
Next page
Claims  available in
Description  (OCR text may contain errors)

United States Patent Batcher [451 Jan. 16, 1973 [54] DETERMINATION OF NUMBER OF 3.466.433 9/969 Dudact al .235/l75 3,603,776 9/!97] Wcinherger ..235/l75 ONES IN A DATA FIELD BY ADDITION [75] Inventor: Kenneth E. Batcher, Stow, Ohio [73] Assignee: Goodyear Aerospace Corporation,

Akron, Ohio [22] Filed: March 15, 1971 [2]] Appl. No.: l24,08 9

[52] US. Cl ..235/l75 [51] Int. Cl ..G06f 7/50 [58] Field of Search ..235/l75, 92, 92 CP, 92 SA [56] References Cited UNITED STATES PATENTS 3,535,502 l0/l970 Clapper ..235/l75 X Primary Examiner-Eugene G. Botz Assistant ExaminerDavid H. Malzahn Att0rney.l. G. Pere and L. A. Germain [57] ABSTRACT An arrangement for counting the number of a given data such as ones in a data field. The system employs full adders which count the number of ones in basic three element subsets of the data field. Additional full adders total the results of the first series of full adders and also count any additional ones of the data field.

5 Claims, 3 Drawing Figures OUTPUT PATENTEDJAN 1 6 I973 SHEET 2 0F 2 mmE INVENTOR KENNETH E. BATCHER WWW,

ATTORNEYS DETERMINATION OF NUMBER OF ONES IN A DATA FIELD BY ADDITION It is the primary object of the present invention to provide means for counting the number of occurrences ofa given value in a data field.

It is also an object of the invention to provide means for counting the number of occurrences of a given value in a data field which provides for the fast counting of the selected value occurrences.

The above and other objects of the invention which will become apparent in the following detailed description are achieved by providing a counting arrangement in which a first series of full adders count the number of occurrences of a selected value simultaneously in three element subsets of the data field, and includes an additional series of full adders for combining the totals produced by each of the adders of the first series.

For a more complete understanding of the invention and the objects and advantages thereof reference should be had to the following detailed descriptionand the accompanying drawings wherein there is shown a preferred embodiment of the invention.

In the drawing:

FIG. 1 is a schematic showing of the counting arrangement of the present invention for a six element field;

FIG. 2 is a schematic showing of the counting arrangement for a seven element field; and

FIG. 3 is a schematic showing of a counting arrangement for a fifteen element field.

The data fields referred to in the following description may be stored in any suitable data storage means, such as the response store of an associative memory or associative processor. Regardless of the particular storage device involved, the data is stored as individual bits in the usual digital manner each having a O or 1 value.

FIG. 1 shows a six element data field 10 and illustrates the basic concept of the present invention. Each of the elements b of the data field 10 may have either a or. I value. While the elements are shown grouped into two sets of three elements each, this is for illustration only and the data field is not necessarily so arranged physically. In order to count the total number of bits b which are of a particular value, for example which have the value 1, there is provided an arrangement of full adders 12-18. A typical full adder would be a MCl0l9 integrated circuit manufactured by Motorola Semiconductor Products, Inc.. Each of the full adders 12-18 is capable of receiving three binary inputs and adding these inputs to produce the two digit binary sum thereof, with the least significant digit of the sum being produced at the output s and the more significant digit being produced at the output 0. The full adder 12 receives as inputs the values of three of the bits of the data field 10 and the full adder 14 receives as inputs the values of the remaining three data bits. If the data field is expressed in a form other than a binary code, converting means may be provided between the elements of the field and the full adders. Such converting means produces a 0 output if the element is not of the selected value and a 1 output if the element is of the selected value. Thus, the full adders 12 and 14 will simultaneously produce the total count of the 1's in their respective subsets of the data field 10. A third full adder 16 receives as inputs the least significant digit outputs of the first full adders l2 and 14. The output of this full adder 16 represents the sum of the least significant digits of the outputs of the adders I2 and 14. The least significant digit of the output of the adder 16 is thus the least significant digit of the sum of the 1's in the data field 10. This digit is supplied to the output device 20. A fourth full adder 18 receives as inputs the most significant digit output of the full adder l6 and the most significant digit outputs of the full adders I2 and 14. The least significant digit of the addition performed by the full adder 18 is the second significant digit of the total count while the most significant digit of the output is the most significant digit of the full count. Thus, the output of the least significant digit of the full adder l6 and the full output of the adder 18 comprise the full count of the number of ones in the data field 10.

FIG. 2 illustrates the arrangement of full adders employed with a data field 22 having seven elements. A first set of full adders 24 and 26 each count the values of three positions of the data field 22. An additional set of full adders 28 and 30 count, respectively, the single previously uncounted bit value of the data field 22, the least significant digits of the outputs of the counters 24 and 26, and the most significant digit of the output of the counter 28 and the most significant digits of the outputs of the counters 24 and 26. Again, the least significant digit output of the counter 28 is the least significant digit of the total count while the outputs of the full adder 30 provide the second and third significant digits of the total count. These three outputs are supplied to the output device 32 and represent the total count of the number of l s in the data field 22.

The basic counting system described in the above paragraphs may be expanded to count the number of ls in a field of any length. Essentially, this is accomplished by dividing the data field into subsets which may be counted by one of the counting arrangements described above and by adding the totals achieved for each subset by means of additional full adders. Thus, as is shown in FIG. 3, a fifteen element data field 48 may be divided into three subsets: S, which consists ofa single bit of the data field 48, S which consists of the next seven data bits of the field 48; and S which consists of the remaining seven data bits. It should be noted that the order in which the data field 48 is divided into the subsets S S and S is purely arbitrary. The arrangement shown in which the first element is assigned to the subset the next seven elements to the subset S ,-and the remaining seven elements to subset S is chosen for convenience of illustration. However, any other arrangement may be employed so long as the subsets S and 8;, are of equal length and the subset S, contains not more than one element.

The total count of 1's in the subsets S and 8;, may be determined by full counters 50 and 52 which are each equivalent to the full adder arrangement of FIG. 2. The counts produced by the seven position counters 50 and 52 and the count of the least significant bit subset S are added by means of additional full adders 54-58. These full adders 54-58 add, respectively, the least significant digit values, the next most significant digit values and the carry value of the previous adder, and the most significant digit values and the carry value from the previous adder. The outputs of the full adders 54-58 provide the successively higher ranked significant digits of the total count and these outputs are furnished to the output device 60. It should be noted that the maximum time required to achieve the total count of the number of I s in the data field is equal to the delay imposed by a full adder multiplied by the total number of full adders involved in producing one digit of the output or final count. Referring again to the arrangement of FIG. 3 it will be seen that the maximum delay is five times the delay of one full adder. Since the adders 62 and 64 operate simultaneiously these two full adders impose only one time delay in the count. The adder 68, however, receives inputs both from the adder 62 and from the full adder 66. As a result, two additional delays are imposed. Likewise, the full adder 56 receives inputs both from the adder 68 and the adder 54 so that two additional time delays are again imposed. In general, if a field having 11 bit positions can be counted with a maximum delay of P full adders, a field having 2n 1 bit positions can be counted with a maximum delay off 2 full adders. For a field having 2" 1 bit positions, where m is greater than or equal to 2, the maximum delay will be 2m 3 multiplied by the delay of one full adder.

The counting arrangement of the present invention is applicable to date fields of any length. The data field is first divided into three primary subsets S S and S with the subsets S and 8;, being of equal length and the subset S containing at most a single element. The subsets S,, S and 8;, are mutually exclusive and exhaust the elements of the data field. lf the primary subsets S and 8;, each contain more than three elements, they are further divided into secondary subsets S S and S and S S and S respectively. The secondary subsets are also mutually exclusive and exhaustive of the elements of their respective primary subsets. The division of the primary subsets is accomplished in the same manner as the original division of the data field. Thus, the subsets S and S are of equal length and the subset S is either void or contains a single element, depending on whether the primary subset S contains an even or odd number of elements. Division of the multiple element subsets is continued in the same manner until a series of ultimate subsets, none of which contains more than three elements of the data field, is achieved.

A full adder is provided for each of the ultimate subsets which contains more than a single element. The full adders produce the counts of the number of 1's in the multiple element ultimate subsets. The counts of the penultimate subsets are obtained by means of additional full adders, two additional full adders being provided for each penultimate subset with one full adder combining the count of the single element subset, if present, and the least significant digit outputs of the ultimate subset adders and the second full adder combining the most significant digit of the output of the first full adder and the most significant digits of the ultimate subset full adders. By means of additional full adders arranged in the same manner, the counts of the successive subsets of the hierarchical rank of the subsets are obtained. The final series of adders combines the counts of the subsets 5,, S and S to produce the total count for the entire data field.

It will be understood that while only the best known embodiment of the invention has been described in detail, in accordance with the Patent Statutes, the invention is not so limited. Reference should therefore be had to the appended claims in determining the true scope of the invention.

What is claimed is:

1. The method of determining the number of occurrences of a selected value in a data field having at least fifteen elements, which comprises the steps of l. dividing the data field into mutually exclusive, ex-

haustive subsets S S and S where S, and S, are of equal length and 8, includes at most one element of the data field;

2. counting the occurrences of the selected value in each of the subsets S, and S 3. adding, by means ofa full adder, the number of occurrences of the selected value in subset S and the least significant digit of the count of each of the subsets S and S 4. adding, by means of an additional full adder, the most significant digit of the output of the previous full adder and the next significant digit of the count of each of the subsets S and S 5. repeating step (4) until all the digits of the count of each of the subsets S, and 5;, have been exhausted, an additional full adder being employed for each addition; and

6. transmitting to an output device the least significant digit of the output of each full adder and both digits of the output of the last full adder, the least significant digit of the output of the first full adder being the least significant digit of the total count, the least significant digits of the outputs of successive full adders being the successive digits of the full count, and the most significant digit of the output of the final full adder being the most significant digit of the full count.

2. The method according to claim 1 wherein the count of each multiple element subset of the data field is determined by the steps of:

a. dividing .the subset into three mutually exclusive, exhaustive subsets, the first-of which has at most one element, a second and third of which are of equal length;

b. determining if any subset produced in the immediately preceeding division contains more than three elements;

c. dividing all elements produced in the immediately preceeding division which contain more than three elements into three subsets satisfying the conditions of step (a);

. repeating steps (b) and (0) until an ultimate series of subsets no one of which contains more than three elements is obtained;

e. summing, by means of full adders, the number of occurrences of the selected value of each ultimate subset having more than one element; and

f. by means of additional full adders, progressively combining andadding to the sums the number of occurrences of the selected value in those subsets having one element in the reverse order of their creation in steps (a) through (d) until the total count of each multiple element subset has been achieved.

3. Apparatus for counting the number of occurrences of a predetermined value in a data field of at least fifteen elements, comprising:

first and second counting means for counting the number of occurrences in first and second portions, respectively, of the data field, the first and second portions being non-overlapping, of equal length, and including at least all but one element of the data field;

a first full adder receiving as inputs the value of the data field element excluded from the first and second portions and the least significant digits of the output of the first and second counting means;

additional full adders receiving as inputs, respectively, the more significant digit output of the preceeding full adder and the succeedingly more significant digits of the outputs of the first and second counting means; and

an output device receiving the least significant digit outputs of each full adder and the most significant digit output of the final full adder, the output of the first full adder being the least significant digit of the total and the outputs of successive full adders being the successively more significant digits of the total.

4. Apparatus according to claim 3 wherein the first and second counting means each comprise a plurality of full adders.

5. Apparatus according to claim 4 wherein the full adders comprising each counting means are arranged in a plurality of series, the full adders of the first series counting the occurrences in two to three element subsets of the portion, the full adders of successive series combining the totals of the previous series of full adders.

Patent Citations
Cited PatentFiling datePublication dateApplicantTitle
US3466433 *Dec 14, 1965Sep 9, 1969IbmOptical parallel adder
US3535502 *Nov 15, 1967Oct 20, 1970IbmMultiple input binary adder
US3603776 *Jan 15, 1969Sep 7, 1971IbmBinary batch adder utilizing threshold counters
Referenced by
Citing PatentFiling datePublication dateApplicantTitle
US4336600 *Apr 10, 1980Jun 22, 1982Thomson-CsfBinary word processing method using a high-speed sequential adder
US4399517 *Mar 19, 1981Aug 16, 1983Texas Instruments IncorporatedMultiple-input binary adder
US4488253 *Apr 30, 1982Dec 11, 1984Itt Industries, Inc.Parallel counter and application to binary adders
US4713786 *Feb 15, 1985Dec 15, 1987Harris CorporationDigital hardware selection filter
US5148388 *May 17, 1991Sep 15, 1992Advanced Micro Devices, Inc.7 to 3 counter circuit
US5539683 *Jun 24, 1994Jul 23, 1996Fujitsu LimitedMethod and device for processing, and detecting a state of, binary data
US5541865 *Jul 6, 1995Jul 30, 1996Intel CorporationMethod and apparatus for performing a population count operation
US5619437 *Sep 27, 1995Apr 8, 1997Ando Electric Co., Ltd.Parallel data counter circuit
US5642306 *May 15, 1996Jun 24, 1997Intel CorporationMethod and apparatus for a single instruction multiple data early-out zero-skip multiplier
US5666298 *Aug 22, 1996Sep 9, 1997Intel CorporationMethod for performing shift operations on packed data
US5675526 *Nov 26, 1996Oct 7, 1997Intel CorporationProcessor performing packed data multiplication
US5677862 *Apr 2, 1996Oct 14, 1997Intel CorporationMethod for multiplying packed data
US5701508 *Dec 19, 1995Dec 23, 1997Intel CorporationExecuting different instructions that cause different data type operations to be performed on single logical register file
US5721892 *Nov 6, 1995Feb 24, 1998Intel CorporationProcessor
US5740392 *Dec 27, 1995Apr 14, 1998Intel CorporationMethod and apparatus for fast decoding of 00H and OFH mapped instructions
US5742529 *Dec 21, 1995Apr 21, 1998Intel CorporationMethod and an apparatus for providing the absolute difference of unsigned values
US5752001 *Jun 1, 1995May 12, 1998Intel CorporationMethod and apparatus employing Viterbi scoring using SIMD instructions for data recognition
US5757432 *Dec 18, 1995May 26, 1998Intel CorporationManipulating video and audio signals using a processor which supports SIMD instructions
US5764943 *Dec 28, 1995Jun 9, 1998Intel CorporationData path circuitry for processor having multiple instruction pipelines
US5787026 *Dec 20, 1995Jul 28, 1998Intel CorporationMethod and apparatus for providing memory access in a processor pipeline
US5793661 *Dec 26, 1995Aug 11, 1998Intel CorporationMethod and apparatus for performing multiply and accumulate operations on packed data
US5802336 *Jan 27, 1997Sep 1, 1998Intel CorporationMicroprocessor capable of unpacking packed data
US5815421 *Dec 18, 1995Sep 29, 1998Intel CorporationMethod for transposing a two-dimensional array
US5818739 *Apr 17, 1997Oct 6, 1998Intel CorporationProcessor for performing shift operations on packed data
US5819101 *Jul 21, 1997Oct 6, 1998Intel CorporationMethod for packing a plurality of packed data elements in response to a pack instruction
US5822232 *Mar 1, 1996Oct 13, 1998Intel CorporationMethod for performing box filter
US5822459 *Sep 28, 1995Oct 13, 1998Intel CorporationMethod for processing wavelet bands
US5831885 *Mar 4, 1996Nov 3, 1998Intel CorporationFor generating a quotient
US5835392 *Dec 28, 1995Nov 10, 1998Intel CorporationIn a computer system
US5835748 *Dec 19, 1995Nov 10, 1998Intel CorporationMethod for executing different sets of instructions that cause a processor to perform different data type operations on different physical registers files that logically appear to software as a single aliased register file
US5835782 *Mar 4, 1996Nov 10, 1998Intel CorporationPacked/add and packed subtract operations
US5852726 *Dec 19, 1995Dec 22, 1998Intel CorporationMethod and apparatus for executing two types of instructions that specify registers of a shared logical register file in a stack and a non-stack referenced manner
US5857096 *Dec 19, 1995Jan 5, 1999Intel CorporationMicroarchitecture for implementing an instruction to clear the tags of a stack reference register file
US5859997 *Aug 20, 1996Jan 12, 1999Intel CorporationComputer-implemented method
US5862067 *Dec 29, 1995Jan 19, 1999Intel CorporationMethod and apparatus for providing high numerical accuracy with packed multiply-add or multiply-subtract operations
US5880979 *Dec 21, 1995Mar 9, 1999Intel CorporationSystem for providing the absolute difference of unsigned values
US5881279 *Nov 25, 1996Mar 9, 1999Intel CorporationMethod and apparatus for handling invalid opcode faults via execution of an event-signaling micro-operation
US5898601 *Dec 17, 1996Apr 27, 1999Intel CorporationComputer implemented method for compressing 24 bit pixels to 16 bit pixels
US5907842 *Dec 20, 1995May 25, 1999Intel CorporationIn a processor for processing instructions
US5935240 *Dec 15, 1995Aug 10, 1999Intel CorporationComputer implemented method for transferring packed data between register files and memory
US5936872 *Dec 20, 1995Aug 10, 1999Intel CorporationIn a data processing system
US5940859 *Dec 19, 1995Aug 17, 1999Intel CorporationEmptying packed data state during execution of packed data instructions
US5959636 *Feb 23, 1996Sep 28, 1999Intel CorporationMethod and apparatus for performing saturation instructions using saturation limit values
US5983253 *Dec 20, 1995Nov 9, 1999Intel CorporationComputer system for performing complex digital filters
US5983256 *Oct 29, 1997Nov 9, 1999Intel CorporationApparatus for performing multiply-add operations on packed data
US5983257 *Dec 26, 1995Nov 9, 1999Intel CorporationSystem for signal processing using multiply-add operations
US5984515 *Aug 21, 1997Nov 16, 1999Intel CorporationComputer implemented method for providing a two dimensional rotation of packed data
US6009191 *Feb 15, 1996Dec 28, 1999Intel CorporationComputer implemented method for compressing 48-bit pixels to 16-bit pixels
US6014684 *Mar 24, 1997Jan 11, 2000Intel CorporationMethod and apparatus for performing N bit by 2*N-1 bit signed multiplication
US6018351 *Oct 30, 1997Jan 25, 2000Intel CorporationComputer system performing a two-dimensional rotation of packed data representing multimedia information
US6035316 *Feb 23, 1996Mar 7, 2000Intel CorporationApparatus for performing multiply-add operations on packed data
US6036350 *May 20, 1997Mar 14, 2000Intel CorporationMethod of sorting signed numbers and solving absolute differences using packed instructions
US6049864 *Aug 20, 1996Apr 11, 2000Intel CorporationMethod for scheduling a flag generating instruction and a subsequent instruction by executing the flag generating instruction in a microprocessor
US6058408 *Dec 20, 1995May 2, 2000Intel CorporationMethod and apparatus for multiplying and accumulating complex numbers in a digital filter
US6067034 *Apr 7, 1998May 23, 2000Vocal Technologies Ltd.Maximal bit packing method
US6070237 *Mar 4, 1996May 30, 2000Intel CorporationMethod for performing population counts on packed data types
US6081824 *Mar 5, 1998Jun 27, 2000Intel CorporationMethod and apparatus for fast unsigned integral division
US6092184 *Dec 28, 1995Jul 18, 2000Intel CorporationParallel processing of pipelined instructions having register dependencies
US6128614 *Feb 8, 1999Oct 3, 2000Intel CorporationMethod of sorting numbers to obtain maxima/minima values with ordering
US6170997Jul 22, 1997Jan 9, 2001Intel CorporationMethod for executing instructions that operate on different data types stored in the same single logical register file
US6237016Jul 31, 1997May 22, 2001Intel CorporationMethod and apparatus for multiplying and accumulating data samples and complex coefficients
US6266686Mar 4, 1999Jul 24, 2001Intel CorporationEmptying packed data state during execution of packed data instructions
US6275834Mar 4, 1996Aug 14, 2001Intel CorporationApparatus for performing packed shift operations
US6370559Jul 13, 1999Apr 9, 2002Intel CorportionMethod and apparatus for performing N bit by 2*N−1 bit signed multiplications
US6385634Aug 31, 1995May 7, 2002Intel CorporationMethod for performing multiply-add operations on packed data
US6418529Mar 31, 1998Jul 9, 2002Intel CorporationApparatus and method for performing intra-add operation
US6430251 *Oct 24, 2000Aug 6, 2002Sun Microsystems, Inc.4-Bit population count circuit
US6470370Jan 16, 2001Oct 22, 2002Intel CorporationMethod and apparatus for multiplying and accumulating complex numbers in a digital filter
US6516406Sep 8, 2000Feb 4, 2003Intel CorporationProcessor executing unpack instruction to interleave data elements from two packed data
US6631389Dec 22, 2000Oct 7, 2003Intel CorporationApparatus for performing packed shift operations
US6738793Jan 14, 2001May 18, 2004Intel CorporationProcessor capable of executing packed shift operations
US6751725Feb 16, 2001Jun 15, 2004Intel CorporationMethods and apparatuses to clear state for operation of a stack
US6792523Jul 27, 1999Sep 14, 2004Intel CorporationProcessor with instructions that operate on different data types stored in the same single logical register file
US6823353Aug 2, 2002Nov 23, 2004Intel CorporationMethod and apparatus for multiplying and accumulating complex numbers in a digital filter
US6901420Jul 18, 2003May 31, 2005Intel CorporationMethod and apparatus for performing packed shift operations
US6904114 *Apr 25, 2003Jun 7, 2005J. Barry ShacklefordOnes counter employing two dimensional cellular array
US6909767Jan 14, 2004Jun 21, 2005Arithmatica LimitedLogic circuit
US6961845Jul 9, 2002Nov 1, 2005Intel CorporationSystem to perform horizontal additions
US7047383Jul 11, 2002May 16, 2006Intel CorporationByte swap operation for a 64 bit operand
US7117232May 27, 2005Oct 3, 2006Intel CorporationMethod and apparatus for providing packed shift operations in a processor
US7149882May 11, 2004Dec 12, 2006Intel CorporationProcessor with instructions that operate on different data types stored in the same single logical register file
US7155601Feb 14, 2001Dec 26, 2006Intel CorporationMulti-element operand sub-portion shuffle instruction execution
US7260595Nov 14, 2003Aug 21, 2007Arithmatica LimitedLogic circuit and method for carry and sum generation and method of designing such a logic circuit
US7373490Mar 19, 2004May 13, 2008Intel CorporationEmptying packed data state during execution of packed data instructions
US7392275Jun 30, 2003Jun 24, 2008Intel CorporationMethod and apparatus for performing efficient transformations with horizontal addition and subtraction
US7395298Jun 30, 2003Jul 1, 2008Intel CorporationMethod and apparatus for performing multiply-add operations on packed data
US7395302Jun 30, 2003Jul 1, 2008Intel CorporationMethod and apparatus for performing horizontal addition and subtraction
US7424505Nov 19, 2001Sep 9, 2008Intel CorporationMethod and apparatus for performing multiply-add operations on packed data
US7430578Jun 30, 2003Sep 30, 2008Intel CorporationMethod and apparatus for performing multiply-add operations on packed byte data
US7451169Jun 15, 2006Nov 11, 2008Intel CorporationMethod and apparatus for providing packed shift operations in a processor
US7461109Jun 6, 2007Dec 2, 2008Intel CorporationMethod and apparatus for providing packed shift operations in a processor
US7480686May 14, 2004Jan 20, 2009Intel CorporationMethod and apparatus for executing packed shift operations
US7509367Jun 4, 2004Mar 24, 2009Intel CorporationMethod and apparatus for performing multiply-add operations on packed data
US7624138Dec 30, 2003Nov 24, 2009Intel CorporationMethod and apparatus for efficient integer transform
US7631025Jun 30, 2003Dec 8, 2009Intel CorporationMethod and apparatus for rearranging data between multiple registers
US7685212Oct 25, 2002Mar 23, 2010Intel CorporationFast full search motion estimation with SIMD merge instruction
US7725521Oct 10, 2003May 25, 2010Intel CorporationMethod and apparatus for computing matrix transformations
US7739319Jul 1, 2003Jun 15, 2010Intel CorporationMethod and apparatus for parallel table lookup using SIMD instructions
US7818356Jul 1, 2003Oct 19, 2010Intel CorporationBitstream buffer manipulation with a SIMD merge instruction
US7966482Jun 12, 2006Jun 21, 2011Intel CorporationInterleaving saturated lower half of data elements from two source registers of packed data
US8078836Dec 30, 2007Dec 13, 2011Intel CorporationVector shuffle instructions operating on multiple lanes each having a plurality of data elements using a common set of per-lane control bits
US8185571Mar 23, 2009May 22, 2012Intel CorporationProcessor for performing multiply-add operations on packed data
US8190867May 16, 2011May 29, 2012Intel CorporationPacking two packed signed data in registers with saturation
US8214626Mar 31, 2009Jul 3, 2012Intel CorporationMethod and apparatus for shuffling data
US8225075Oct 8, 2010Jul 17, 2012Intel CorporationMethod and apparatus for shuffling data
US8346838Sep 15, 2009Jan 1, 2013Intel CorporationMethod and apparatus for efficient integer transform
US8396915Sep 4, 2012Mar 12, 2013Intel CorporationProcessor for performing multiply-add operations on packed data
US8495123Oct 1, 2012Jul 23, 2013Intel CorporationProcessor for performing multiply-add operations on packed data
US8495346Apr 11, 2012Jul 23, 2013Intel CorporationProcessor executing pack and unpack instructions
US8510355Oct 19, 2010Aug 13, 2013Intel CorporationBitstream buffer manipulation with a SIMD merge instruction
US8521994Dec 22, 2010Aug 27, 2013Intel CorporationInterleaving corresponding data elements from part of two source registers to destination register in processor operable to perform saturation
US8560586 *Mar 29, 2010Oct 15, 2013Meltin BellLinear bit counting implementations
US8601246Jun 27, 2002Dec 3, 2013Intel CorporationExecution of instruction with element size control bit to interleavingly store half packed data elements of source registers in same size destination register
US8626814Jul 1, 2011Jan 7, 2014Intel CorporationMethod and apparatus for performing multiply-add operations on packed data
US8639914Dec 29, 2012Jan 28, 2014Intel CorporationPacking signed word elements from two source registers to saturated signed byte elements in destination register
US8688959Sep 10, 2012Apr 1, 2014Intel CorporationMethod and apparatus for shuffling data
US8725787Apr 26, 2012May 13, 2014Intel CorporationProcessor for performing multiply-add operations on packed data
US8745119Mar 13, 2013Jun 3, 2014Intel CorporationProcessor for performing multiply-add operations on packed data
US8745358Sep 4, 2012Jun 3, 2014Intel CorporationProcessor to execute shift right merge instructions
US8782377May 22, 2012Jul 15, 2014Intel CorporationProcessor to execute shift right merge instructions
US8793299Mar 13, 2013Jul 29, 2014Intel CorporationProcessor for performing multiply-add operations on packed data
US8793475Dec 29, 2012Jul 29, 2014Intel CorporationMethod and apparatus for unpacking and moving packed data
US8838946Dec 29, 2012Sep 16, 2014Intel CorporationPacking lower half bits of signed data elements in two source registers in a destination register with saturation
US20110238717 *Mar 29, 2010Sep 29, 2011Meltin BellLinear Bit Counting Implementations
EP0195284A2 *Feb 27, 1986Sep 24, 1986Siemens AktiengesellschaftDevice for counting the number of 1/0 bits contained in an n-bits binary word
EP0388506A2 *Jun 19, 1989Sep 26, 1990Digital Equipment CorporationNormalizer
WO1996017289A1 *Dec 1, 1995Jun 6, 1996Intel CorpA novel processor having shift operations
WO2004064254A2 *Jan 14, 2004Jul 29, 2004Arithmatica LtdA logic circuit
Classifications
U.S. Classification708/210
International ClassificationG11C15/00, G11C15/04, G06F7/60
Cooperative ClassificationG06F7/607, G11C15/04
European ClassificationG06F7/60P, G11C15/04
Legal Events
DateCodeEventDescription
Feb 22, 1988ASAssignment
Owner name: LORAL CORPORATION, 600 THIRD AVENUE, NEW YORK, NEW
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST.;ASSIGNOR:GOODYEAR AEROSPACE CORPORATION;REEL/FRAME:004869/0167
Effective date: 19871218
Owner name: LORAL CORPORATION,NEW YORK
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:GOODYEAR AEROSPACE CORPORATION;REEL/FRAME:004869/0167