An Asynchronous Divider Implementation

We present an asynchronous implementation of a novel division algorithm previously patented. Our implementation exploits the average-case behavior of the algorithm and uses the versatility of GasP circuits to implement the data-dependent latencies in the algorithm. On average, the delay per quotient bit for our implementation is 6.3 FO4 gate delays compared to 9.5 FO4 gate delays for a similar SRT divider implementation.

[1]  James E. Robertson,et al.  A New Class of Digital Division Methods , 1958, IRE Trans. Electron. Comput..

[2]  Marc Renaudin,et al.  A new asynchronous pipeline scheme: application to the design of a self-timed ring divider , 1996 .

[3]  Mark Horowitz,et al.  SRT division architectures and implementations , 1997, Proceedings 13th IEEE Sympsoium on Computer Arithmetic.

[4]  Robert E Goldschmidt,et al.  Applications of division by convergence , 1964 .

[5]  Mark Horowitz,et al.  A zero-overhead self-timed 160-ns 54-b CMOS divider , 1991 .

[6]  Milos D. Ercegovac,et al.  Digit-recurrence algorithms for division and square root with limited precision primitives , 2003, The Thrity-Seventh Asilomar Conference on Signals, Systems & Computers, 2003.

[7]  Ivan E. Sutherland,et al.  GasP: a minimal FIFO control , 2001, Proceedings Seventh International Symposium on Asynchronous Circuits and Systems. ASYNC 2001.

[8]  Michael J. Flynn,et al.  Design Issues in Division and Other Floating-Point Operations , 1997, IEEE Trans. Computers.

[9]  Peter A. Beerel,et al.  A Designer's Guide to Asynchronous VLSI , 2010 .

[10]  Michael J. Flynn,et al.  Division Algorithms and Implementations , 1997, IEEE Trans. Computers.

[11]  Wei Liu,et al.  Power dissipation in division , 2008, 2008 42nd Asilomar Conference on Signals, Systems and Computers.

[12]  J. Ebergen,et al.  New division algorithms by digit recurrence , 2004, Conference Record of the Thirty-Eighth Asilomar Conference on Signals, Systems and Computers, 2004..