KR970016935A - Sticky bit detection method in floating point adder - Google Patents

Sticky bit detection method in floating point adder Download PDF

Info

Publication number
KR970016935A
KR970016935A KR1019950032938A KR19950032938A KR970016935A KR 970016935 A KR970016935 A KR 970016935A KR 1019950032938 A KR1019950032938 A KR 1019950032938A KR 19950032938 A KR19950032938 A KR 19950032938A KR 970016935 A KR970016935 A KR 970016935A
Authority
KR
South Korea
Prior art keywords
exponent
operators
sticky bit
floating point
sticky
Prior art date
Application number
KR1019950032938A
Other languages
Korean (ko)
Inventor
강정엽
Original Assignee
김광호
삼성전자 주식회사
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 김광호, 삼성전자 주식회사 filed Critical 김광호
Priority to KR1019950032938A priority Critical patent/KR970016935A/en
Publication of KR970016935A publication Critical patent/KR970016935A/en

Links

Landscapes

  • Complex Calculations (AREA)

Abstract

본 발명은 부동소숫점 덧셈기에서 스티키비트(sticky bit)을 구하는 방법에 관한 것으로, 두 연산자(A,B)에 대한 부동소숫점 덧셈기에서 스티키비트를 검출하기 위한 방법에 있어서, 두 연산자(A,B) 중 어느 쪽의 지수가 큰가를 판단하는 과정; 두 연산자(A,B)의 지수차(exp-diff)를 구하는 과정; 두 연산자(A,B)에서 각각 LSB로부터 출현한 0의 개수를 MSB쪽으로 세어 Acount 및 Bcount를 구하는 과정; 및 만일 지수차(exp-diff)가 Acount보다 크고 B의 지수가 A의 지수보다 크거나, 지수차(exp-diff)가 Bcount보다 크고 A의 지수가 B의 지수보다 크면, 스티키비트의 값은 “1”, 그렇지 않으면 스티키비트의 값은 “0”이 되는 과정을 포함함을 특징으로 한다.The present invention relates to a method for obtaining a sticky bit in a floating point adder, and in a method for detecting a sticky bit in a floating point adder for two operators (A, B), two operators (A, B). Determining which of the indices is large; Obtaining an exp-diff of two operators A and B; Calculating Acount and Bcount by counting the number of zeros appearing from the LSB to the MSB in each of the two operators A and B; And if the exp-diff is greater than Acount and the exponent of B is greater than the exponent of A, or if the exp-diff is greater than Bcount and the exponent of A is greater than the exponent of B, the value of the sticky bit is "1", otherwise, the sticky bit value includes "0".

따라서 본 발명에 의하면, 덧셈 과정과는 별개로 스티키비트를 구할 수 있으므로, 덧셈연산에서 필요한 A와 B의 순서변환이나 배열과정을 거치지 않고 빠르게 스티키비트를 구할 수 있다.Therefore, according to the present invention, since the sticky bits can be obtained separately from the addition process, the sticky bits can be quickly obtained without undergoing the A and B order conversion or arrangement process required for the addition operation.

Description

부동소숫점 덧셈기에서 스티키비트 검출방법Sticky bit detection method in floating point adder

본 내용은 요부공개 건이므로 전문내용을 수록하지 않았음Since this is an open matter, no full text was included.

제1도는 본 발명에 의한 부동소숫점 덧셈방법을 설명하기 위한 도면.1 is a view for explaining a floating point addition method according to the present invention.

Claims (1)

두 연산자(A,B)에 대한 부동소숫점 덧셈기에서 스티키비트를 검출하기 위한 방법에 있어서, 두 연산자(A,B) 중 어느 쪽의 지수가 큰가를 판단하는 과정; 두 연산자(A,B)의 지수차(exp-diff)를 구하는 과정; 두 연산자(A,B)에서 각각 LSB로부터 출현한 0의 개수를 MSB쪽으로 세어 Acount 및 Bcount를 구하는 과정; 및 만일 지수차(exp-diff)가 Acount보다 크고 B의 지수가 A의 지수보다 크거나, 지수차(exp-diff)가 Bcount보다 크고 A의 지수가 B의 지수보다 크면, 스티키비트의 값은 “1”, 그렇지 않으면 스티키비트의 값은 “0”이 되는 과정을 포함함을 특징으로 하는 부동소숫점 덧셈기에서 스티키비트 검출방법.A method for detecting sticky bits in a floating point adder for two operators A and B, the method comprising: determining which exponent of two operators A and B is large; Obtaining an exp-diff of two operators A and B; Calculating Acount and Bcount by counting the number of zeros appearing from the LSB to the MSB in each of the two operators A and B; And if the exp-diff is greater than Acount and the exponent of B is greater than the exponent of A, or if the exp-diff is greater than Bcount and the exponent of A is greater than the exponent of B, the value of the sticky bit is Sticky bit detection method in a floating point adder, characterized in that "1", otherwise the value of the sticky bit includes the process of becoming "0". ※ 참고사항 : 최초출원 내용에 의하여 공개하는 것임.※ Note: The disclosure is based on the initial application.
KR1019950032938A 1995-09-29 1995-09-29 Sticky bit detection method in floating point adder KR970016935A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
KR1019950032938A KR970016935A (en) 1995-09-29 1995-09-29 Sticky bit detection method in floating point adder

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
KR1019950032938A KR970016935A (en) 1995-09-29 1995-09-29 Sticky bit detection method in floating point adder

Publications (1)

Publication Number Publication Date
KR970016935A true KR970016935A (en) 1997-04-28

Family

ID=66616476

Family Applications (1)

Application Number Title Priority Date Filing Date
KR1019950032938A KR970016935A (en) 1995-09-29 1995-09-29 Sticky bit detection method in floating point adder

Country Status (1)

Country Link
KR (1) KR970016935A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR101528340B1 (en) * 2009-03-16 2015-06-11 어드밴스드 마이크로 디바이시즈, 인코포레이티드 Mechanism for fast detection of overshift in a floating point unit

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR101528340B1 (en) * 2009-03-16 2015-06-11 어드밴스드 마이크로 디바이시즈, 인코포레이티드 Mechanism for fast detection of overshift in a floating point unit

Similar Documents

Publication Publication Date Title
KR920001332A (en) Method and device for predicting branch operation of high performance processor
KR910020641A (en) Noise Prediction Device and Signal Processing Device Using It
KR910015962A (en) Voice signal processing device
KR920700496A (en) Vector Quantizer Codebook Processing Circuit
KR970016935A (en) Sticky bit detection method in floating point adder
KR970077995A (en) Flash Analog-to-Digital Converter
KR920019198A (en) Pattern Feature Extraction Method
KR950035383A (en) Motion vector detection device
US6128636A (en) Method for interfacing floating point and integer processes in a computer system
KR920007434A (en) Operating amount conversion system for digital image processing
KR920003238A (en) Sound source data generation system
KR970049447A (en) Rounding method for floating point addition / subtraction
KR970012116A (en) How to get sticky beat
JPH03100722A (en) Processing system for accuracy conversion instruction
KR940004981A (en) Error Interpolation Circuit in D2-Multiplex Analog Component (D2-MAC) Voice Decoder
JPH0784755A (en) Digital signal processor
KR970066845A (en) Hardware that implements μ-law among speech signal coding techniques
Lozier et al. Error-bounding in level-index computer arithmetic
SU896786A1 (en) Discrete information registering device
KR970050158A (en) Digital V. C. audio data sync number extraction method in 625/50 system
JPH0119159B2 (en)
JPH03100723A (en) Processing system for accuracy conversion instruction
JPH02184962A (en) Performance evaluation processing system for japanese sentence error detector
KR970050160A (en) Digital V. C. audio data sync number extraction method in 525/60 system
KR950007571A (en) Representative Vector Error Detection Method of Classified Vector Quantized Image

Legal Events

Date Code Title Description
WITN Withdrawal due to no request for examination