Intel 64 and IA-32 Architectures Software Developers Manual Volume 1, Basic Architecture
Vol. 1 E-19
GUIDELINES FOR WRITING SIMD FLOATING-POINT EXCEPTION HANDLERS
Table E-17. #U - Numeric Underflow
Instruction Condition Masked Response
Unmasked Response
and Exception Code
ADDPS
ADDSUBPS
HADDPS
SUBPS
HSUBPS
MULPS
DIVPS
ADDSS
SUBSS
MULSS
DIVSS
CVTPD2PS
CVTSD2SS
Result calculated with
unbounded exponent and
rounded to the
destination precision <
smallest single precision
finite normal value.
res = ±0, denormal, or
normal
#UE = 1 and #PE = 1,
but only if the result is
inexact
res = (result calculated with
unbounded exponent and
rounded to the destination
precision) * 2
192
•#UE = 1
• #PE = 1 if the result is
inexact
ADDPD
ADDSUBPD
HADDPD
SUBPD
HSUBPD
MULPD
DIVPD
ADDSD
SUBSD
MULSD
DIVSD
Result calculated with
unbounded exponent and
rounded to the
destination precision <
smallest double precision
finite normal value.
res = ±0, denormal or
normal
#UE = 1 and #PE = 1,
but only if the result is
inexact
res = (result calculated with
unbounded exponent and
rounded to the destination
precision) * 2
1536
•#UE = 1
• #PE = 1 if the result is
inexact