User Guide

180 128-Bit Media and Scientific Programming
AMD64 Technology 24592—Rev. 3.15—November 2009
Result
infinite
A result of infinite precision, which is representable when the width of the exponent and the width
of the significand are both infinite.
Result
round
A r esult, after rounding, whose unbiased exponent is infinitely wide and whose s ignificand is the
width specified for the destination format. (Rounding is described in “Floating-Point Rounding”
on page 132.)
Result
round, denormal
A result, after rounding and denormalization. (Denormalization is described in “Denormalized
(Tiny) Numbers” on page 128.)
Masked and unmasked responses to the exceptions are described in “SIMD Floating-Point Exception
Masking” on page 184. The priority of the exceptions is described in “SIMD Floating-Point Exception
Priority” on page 182.
Invalid-Operation Exception (IE). The IE exception occurs due to one of the attempted invalid
operations shown in Table 4-11.
Table 4-11. Invalid-Operation Exception (IE) Causes
Operation Condition
Any Arithmetic Operation, and
CVTPS2PD, CVTPD2PS, CVTSS2SD, CVTSD2SS
A source operand is an SNaN
MAXPS, MAXPD, MAXSS, MAXSD
MINPS, MINPD, MINSS, MINSD
CMPPS, CMPPD, CMPSS, CMPSD
COMISS, COMISD
A source operand is a NaN (QNaN or SNaN)
ADDPS, ADDPD, ADDSS, ADDSD, ADDSUBPS.
ADDSUBPD, HADDPS, HADDPD
Source operands are infinities with opposite signs
SUBPS, SUBPD, SUBSS, SUBSD, ADDSUBPS,
ADDSUBPD, HSUBPS, HSUBPD
Source operands are infinities with same sign
MULPS, MULPD, MULSS, MULSD Source operands are zero and infinity
DIVPS, DIVPD, DIVSS, DIVSD Source operands are both infinity or both zero
SQRTPS, SQRTPD, SQRTSS, SQRTSD
Source operand is less than zero (except ±0, which
returns ±0)
Data conversion from floating-point to integer
(CVTPS2PI, CVTPD2PI, CVTSS2SI, CVTSD2SI,
CVTPS2DQ, CVTPD2DQ, CVTTPS2PI, CVTTPD2PI,
CVTTPD2DQ, CVTTPS2DQ, CVTTSS2SI,
CVTTSD2SI)
Source operand is a NaN, infinite, or not
representable in destination data type