Floating point processing unit with forced arithmetic results

Details Floating point processing unit with forced arithmetic results Floating point processing unit with forced arithmetic results

1995-09-26

1998-05-05

Malzahn, David H.

Boots, shoes, and leggings

3647365, G06F 700

Patent

active

057485160

ABSTRACT:
Logic for selectively forcing arithmetic results allows a floating point unit to bypass the normal flow through arithmetic units and pipelines depending on the particular floating point operation and operand conditions. Certain forced results (e.g., forced zeros, infinities, and those corresponding to certain invalid operand conditions) may bypass arithmetic units or pipelines and rounding circuitry entirely. On the other hand, other operand dependent results (e.g., the result of X+0 and results of operations involving a NaN operand or operands) may only partially bypass the normal flow. By providing logic for selectively forcing results, arithmetic pipelines may be freed for subsequent instructions in the instruction stream. Logic for selectively forcing arithmetic results may be particularly attractive in a superscalar processor. In a superscalar processor which includes a floating point unit with forced arithmetic results, microcode to handle special cases, pipeline bypass, and early result generation can be avoided because architectural approaches for handling out-of-order results allow dependencies to be resolved irrespective of result reordering. Therefore, the early and out-of-order generation of forced results may be handled by a reorder buffer.

REFERENCES:
patent: 4161784 (1979-07-01), Cushing et al.
patent: 4631696 (1986-12-01), Sakamoto
patent: 4789956 (1988-12-01), Hildebrandt
patent: 4823260 (1989-04-01), Imel et al.
patent: 4928223 (1990-05-01), Dao et al.
patent: 4941120 (1990-07-01), Brown et al.
patent: 4961161 (1990-10-01), Kojima
patent: 5038313 (1991-08-01), Kojima
patent: 5053631 (1991-10-01), Perlman et al.
patent: 5058048 (1991-10-01), Gupta et al.
patent: 5095458 (1992-03-01), Lynch et al.
patent: 5128888 (1992-07-01), Tamura et al.
patent: 5128891 (1992-07-01), Lynch et al.
patent: 5155816 (1992-10-01), Kohn
patent: 5155820 (1902-10-01), Gibson
patent: 5206828 (1993-04-01), Shah et al.
patent: 5226126 (1993-07-01), McFarland et al.
patent: 5257214 (1993-10-01), Mason et al.
patent: 5262973 (1993-11-01), Richardson
patent: 5339266 (1994-08-01), Hinds et al.
patent: 5481489 (1996-01-01), Yanagida et al.
patent: 5481686 (1996-01-01), Dockser
patent: 5499352 (1996-03-01), Clift et al.
patent: 5574928 (1996-11-01), White et al.
patent: 5632023 (1997-05-01), White et al.
patent: 5651125 (1997-07-01), Witt et al.
IBM Technical Disclosure, "Special Case" Arithmetic Handling for IEEE Floating Point Hardware, May 1989, pp. 469-471, vol. 31 No. 12.
IBM Technical Disclosure, FPU Special Case Logic Without Madd Array Dataflow, Nov. 1991, pp. 236-237, vol. 34 No. 6.
IBM Technical Disclosure, Floating Point Bypass Dataflow, Sep. 1994, pp. 545-550, vol. 37 No. 09.
Vojin G. Oklobdzija, "An Algorithmic and Novel Design of a Leading Zero Detector Circuit: Comparison with Logic Synthesis", IEEE, .COPYRGT.1994, pp. 124-128.
R.M. Tomsaulo, "An Efficient Algorithm For Exploiting Multiple Arithmetic Units", Jan. 1967, pp. 293-305.
Kai Hwang et al, "Computer Architecture and Parallel Processing", Copyright 1984, pp. 170-174.
Intel, "Microprocessor Family Programmer's Reference Manual", 1992, pp. 15-1 thru 16-28.
IEEE, "IEEE Standard For Radix-Independent Floating-Point Arithmetic", Copyright 1987, pp. 7-16.
IEEE, "IEEE Standard For Binary Floating-Point Arithmetic", Copyright 1985, pp. 7-17.

Affiliated with

Also associated with

Rating

Floating point processing unit with forced arithmetic results does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Floating point processing unit with forced arithmetic results, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Floating point processing unit with forced arithmetic results will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-62354