Fujitsu Fujitsu SPARC64 V Floating-Point Exceptional Conditions and Results, UnfinishedFPop4

Models: Fujitsu SPARC64 V

1 255
Download 255 pages 53.5 Kb
Page 77
Image 77

summarizes the behavior of SPARC64 V floating-point hardware depending on FSR.NS.

Note – The result and behavior of SPARC64 V of the shaded column in the tables Table B-5 and Table B-6 conform to IEEE754-1985 standard.

Note – Throughout Table B-5 and Table B-6, lowercase exception conditions such as nx, uf, of, dv and nv are nontrapping IEEE 754 exceptions. Uppercase exception conditions such as NX, UF, OF, DZ and NV are trapping IEEE 754 exceptions.

TABLE B-5

Floating-Point Exceptional Conditions and Results

 

 

 

 

 

 

 

 

 

 

 

 

FSR.N

Denorm :

Result

Pessimistic

Pessimistic

 

 

 

 

 

S

Norm1

 

Denorm2

Zero

Overflow

UFM

OFM

NXM

Result

 

 

 

 

 

 

1

UF

 

 

 

 

Yes

 

0

1

NX

 

No

 

Yes

 

 

0

uf + nx, a signed zero, or a signed

 

 

 

 

 

 

 

 

 

 

Dmin3

 

 

 

 

No

 

1

UF

 

 

 

 

 

 

0

unfinished_FPop4

0

 

 

No

Conforms to IEEE754-1985

 

 

 

 

 

 

 

 

 

1

 

 

UF

 

 

 

 

Yes

0

 

1

NX

 

 

 

 

 

 

 

 

 

0

uf + nx, a signed zero, or a signed

 

Yes

 

n/a

 

 

 

 

 

 

Dmin

 

 

 

 

 

 

 

1

 

OF

 

 

 

 

No

Yes

0

 

1

NX

 

 

 

 

 

 

 

 

 

0

of + nx, a signed infinity, or a

 

 

 

 

 

 

 

 

 

 

signed Nmax5

 

 

 

 

 

No

 

unfinished_FPop

 

 

 

 

 

 

1

 

UF

 

No

 

Yes

 

 

0

 

1

NX

1

 

 

 

 

 

 

 

 

 

 

 

 

 

 

0

uf + nx, a signed zero

 

 

 

 

 

 

 

 

 

 

 

 

 

 

No

 

 

Conforms to IEEE754-1985

 

Yes

 

 

 

 

 

 

 

TABLE B-6

 

 

 

 

 

 

 

 

 

 

 

1.One of the operands is a denormalized number, and the other operand is a normal or a denormalized number (non- zero, non-NaN, and non-infinity).

2.The result before rounding turns out to be a denormalized number.

3.Dmin = denormalized minimum.

4.If the FPop is either FADD{s,d}, or FSUB{s,d} and the operation is 0 ± denormalized number, SPARC64 V does not generate an unfinished_FPop and generates a result according to IEEE754-1985 standard.

5.Nmax = normalized maximum.

66 SPARC JPS1 Implementation Supplement: Fujitsu SPARC64 V • Release 1.0, 1 July 2002

Page 77
Image 77
Fujitsu Fujitsu SPARC64 V manual Floating-Point Exceptional Conditions and Results, UnfinishedFPop4