Table B–2: IEEE Floating-Point Trap Handling (Continued)

 

 

 

OS

 

User

 

 

 

Hardware1

PAL-

Completion

Signal

 

 

Alpha Instructions

Code

Handler

Handler

 

MULx OUTPUT Exceptions:

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Exponent overflow

Trap

Trap

Supply

 

[Overflow3]

 

 

 

 

+/–Inf

 

Scale

by

bias

 

 

 

+/–MAX

 

adjust

 

 

Exponent underflow and disabled

Supply +0

 

 

 

Exponent underflow and enabled

Supply +0

Trap

Supply

 

[Underflow3]

 

and Trap

 

+/–MIN

 

Scale

by

bias

 

 

 

denorm

 

adjust

 

 

 

 

 

+/–0

 

 

 

 

Inexact and disabled

 

 

 

Inexact and enabled

Supply prod.

Trap

 

[Inexact]

 

 

and trap

 

 

 

 

 

 

 

 

 

 

 

 

 

 

DIVx INPUT Exceptions:

 

 

 

 

 

 

 

 

 

 

 

 

Denormal operand

Trap

Trap

Supply quot.

[Denormal Op2]

+/-Inf operand

Trap

Trap

Supply quot.

 

 

QNaN operand

Trap

Trap

Supply QNaN

 

 

SNaN operand

Trap

Trap

Supply QNaN

[Invalid Op]

 

0/0 or Inf/Inf

Trap

Trap

Supply QNaN

[Invalid Op]

 

A/0

Trap

Trap

Supply

 

[Div. Zero]

 

 

 

 

+/– Inf

 

 

 

 

 

 

 

 

 

 

 

 

DIVx OUTPUT Exceptions:

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Exponent overflow

Trap

Trap

Supply

 

[Overflow3]

 

 

 

 

+/–Inf

 

Scale

by

bias

 

 

 

+/– MAX

 

adjust

 

 

Exponent underflow and disabled

Supply +0

 

 

 

Exponent underflow and enabled

Supply +0

Trap

Supply

 

[Underflow3]

 

and trap

 

+/– MIN

 

Scale

by

bias

 

 

 

denorm

 

adjust

 

 

 

 

 

+/–0

 

 

 

 

Inexact and disabled

 

 

 

Inexact and enabled

Supply quot.

Trap

 

[Inexact]

 

 

and trap

 

 

 

 

 

 

 

 

 

 

 

 

 

CMPTEQ CMPTUN INPUT Exceptions:

 

 

 

 

 

 

 

 

 

 

 

Denormal operand

Trap

Trap

Supply (=)

[Denormal Op2]

QNaN operand

Trap

Trap

Supply

False

 

 

 

 

 

for EQ,

True

 

 

 

 

 

 

for UN

 

 

 

 

SNaN operand

Trap

Trap

Supply

 

[Invalid Op]

 

 

 

 

False/ True

 

 

 

 

 

 

 

 

 

 

 

IEEE Floating-Point Conformance B–9

Page 299
Image 299
Compaq ECQD2KCTE manual MULx Output Exceptions