Intel ITANIUM ARCHITECTURE - SOFTWARE DEVELOPERS MANUAL VOLUME 1 REV 2.3 Manual page 1177
Hide thumbs
Also See for ITANIUM ARCHITECTURE - SOFTWARE DEVELOPERS MANUAL VOLUME 1 REV 2.3:
Table of Contents
Advertisement
xmpy
xmpy — Fixed-Point Multiply
(
) xmpy.l
Format:
qp
(
) xmpy.lu
qp
(
) xmpy.h
qp
(
) xmpy.hu
qp
Two source operands (FR
Description:
and multiplied. The upper or lower 64 bits of the resultant product are selected and
placed in FR
In the high_unsigned_form, the significand fields of FR
unsigned integers and multiplied to produce a full 128-bit unsigned result. The most
significant 64-bits of the resultant product are placed in the significand field of FR
In the high_form, the significand fields of FR
and multiplied to produce a full 128-bit signed result. The most significant 64-bits of
the resultant product are placed in the significand field of FR
In the other forms, the significand fields of FR
and multiplied to produce a full 128-bit signed result. The least significant 64-bits of the
resultant product are placed in the significand field of FR
In all forms, the exponent field of FR
(0x1003E) and the sign field of FR
an integer 1; it is just the register file format's 1.0 value.
Operation:
See "xma — Fixed-Point Multiply Add" on page 3:276.
3:278
=
,
f
f
f
1
3
4
=
,
f
f
f
1
3
4
=
,
f
f
f
1
3
4
=
,
f
f
f
1
3
4
and FR
f
3
.
f
1
pseudo-op of: (
pseudo-op of: (
pseudo-op of: (
pseudo-op of: (
) are treated as either signed or unsigned integers
f
4
and FR
f
3
and FR
are treated as signed integers
f
f
3
4
and FR
are treated as signed integers
f
f
3
4
f
1
is set to the biased exponent for 2.0
f
1
is set to positive (0). Note: f1 as an operand is not
f
1
) xma.l
=
,
, f0
qp
f
f
f
1
3
4
) xma.l
=
,
, f0
qp
f
f
f
1
3
4
) xma.h
=
,
f0
qp
f
f
f
1
3
4,
) xma.hu
=
,
, f0
qp
f
f
f
1
3
4
are treated as
f
4
.
f
1
.
63
Volume 3: Instruction Reference
.
f
1
Advertisement
Table of Contents
