Floating-Point Parallel Merge Negative Sign Operation; Floating-Point Parallel Merge Sign Operation - Intel ITANIUM ARCHITECTURE - SOFTWARE DEVELOPERS MANUAL VOLUME 3 REV 2.3 Manual

fpmerge — Floating-point Parallel Merge
( ) fpmerge.ns
Format: qp
( ) fpmerge.s
qp
( ) fpmerge.se
qp
For the neg_sign_form, the signs of the pair of single precision values in the significand
Description:
field of FR
the pair of single precision values in the significand field of FR
significand field of FR
floating-point numbers by using the same register for
For the sign_form, the signs of the pair of single precision values in the significand field
of FR are concatenated with the exponents and the significands of the pair of single
f
2
precision values in the significand field of FR
For the sign_exp_form, the signs and exponents of the pair of single precision values in
the significand field of FR
significands in the significand field of FR
For all forms, the exponent field of FR
(0x1003E) and the sign field of FR
For all forms, if either FR
computed result.
Figure 2-15.
FR
f
2
Figure 2-16.
FR
f
2
Volume 3: Instruction Reference
= ,
f f f
1 2 3
= ,
f f f
1 2 3
= ,
f f f
1 2 3
are negated and concatenated with the exponents and the significands of
f
2
. This form can be used to negate a pair of single precision
f
1
are concatenated with the pair of single precision
f
2
or FR
f
2

Floating-point Parallel Merge Negative Sign Operation

81 80 64 63 62 32 31 30
Negated
Sign Bits
81 80
FR 1003E
f
0
1

Floating-point Parallel Merge Sign Operation

81 80 64 63 62 32 31 30
81 80
FR 1003E
f 0
1
and
f
2
and stored in FR
f
3
and stored in the significand field of FR
f
3
is set to the biased exponent for 2.0
f
1
is set to positive (0).
f
1
is a NaTVal, FR is set to NaTVal instead of the
f f
3 1
0 81 80 64
FR
f
3
64 63 62 32 31 30
0 81 80 64
FR
f
3
64 63 62 32 31 30
fpmerge
neg_sign_form
sign_form
sign_exp_form
and stored in the
f
3
.
f
3
.
f
1
63
63 62 32 31 30 0
0
63 62 32 31 30 0
0
F9
F9
F9
.
f
1
3:111