Intel ITANIUM ARCHITECTURE - SOFTWARE DEVELOPERS MANUAL VOLUME 1 REV 2.3 Manual page 1056
Hide thumbs
Also See for ITANIUM ARCHITECTURE - SOFTWARE DEVELOPERS MANUAL VOLUME 1 REV 2.3:
Table of Contents
Advertisement
ldf — Floating-point Load
(
) ldf
.
Format:
qp
fsz
fldtype
(
) ldf
.
qp
fsz
fldtype
(
) ldf
.
qp
fsz
fldtype
(
) ldf8.
qp
fldtype
(
) ldf8.
qp
fldtype
(
) ldf8.
qp
fldtype
(qp) ldf.fill.
(
) ldf.fill.
qp
ldhint f
(
) ldf.fill.
qp
ldhint f
A value consisting of fsz bytes is read from memory starting at the address specified by
Description:
the value in GR
and placed in FR
on conversion to floating-point register format. The values of the fsz completer are
given in
Table
described in
For the integer_form, an 8-byte value is loaded and placed in the significand field of FR
without conversion. The exponent field of FR
f
1
(0x1003E) and the sign field of FR
For the fill_form, a 16-byte value is loaded, and the appropriate fields are placed in FR
without conversion. This instruction is used for reloading a spilled register. See
f
1
Section 4.4.4, "Control Speculation" on page 1:60
In the base update forms, the value in GR
value (
imm
register update is done after the load, and does not affect the load address. In the
reg_base_update_form, if the NaT bit corresponding to GR
corresponding to GR
Table 2-35.
fsz Completer
Table 2-36.
fldtype
Completer
none
s
a
Volume 3: Instruction Reference
.
= [
]
ldhint f
r
1
3
.
= [
],
ldhint f
r
r
1
3
2
.
= [
],
ldhint f
r
imm
1
3
9
.
= [
]
ldhint f
r
1
3
.
= [
],
ldhint f
r
r
1
3
2
.
= [
],
ldhint f
r
imm
1
3
9
= [
]
ldhint f
r
1
3
= [
],
r
r
1
3
2
= [
],
r
imm
1
3
9
. The value is then converted into the floating-point register format
r
3
. See
Section 5.1, "Data Types and Formats" on page 1:85
f
1
2-35. The fldtype completer specifies special load operations, which are
Table
2-36.
) or a value from GR
r
9
2
is set and no fault is raised.
r
3
fsz Completers
s
d
e
FP Load Types
Interpretation
Normal load
Speculative load
Certain exceptions may be deferred rather than generating a fault.
Deferral causes NaTVal to be placed in the target register. The NaTVal
value is later used to detect deferral.
Advanced load
An entry is added to the ALAT. This allows later instructions to check for
colliding stores. If the referenced data page has a non-speculative
attribute, no ALAT entry is added to the ALAT and the target register is
set as follows: for the integer_form, the exponent is set to 0x1003E and
the sign and significand are set to zero; for all other forms, the sign,
exponent and significand are set to zero. The absence of an ALAT entry
is later used to detect deferral or collision.
integer_form, no_base_update_form
integer_form, reg_base_update_form
integer_form, imm_base_update_form
fill_form, no_base_update_form
fill_form, reg_base_update_form
fill_form, imm_base_update_form
is set to the biased exponent for 2.0
f
1
is set to positive (0).
f
1
for details.
is added to either a signed immediate
r
3
, and the result is placed back in GR
Bytes Accessed
4 bytes
8 bytes
10 bytes
Special Load Operation
no_base_update_form
reg_base_update_form
imm_base_update_form
for details
. This base
r
3
is set, then the NaT bit
r
2
Memory Format
Single precision
Double precision
Extended precision
ldf
M9
M7
M8
M9
M7
M8
M9
M7
M8
63
3:157
Advertisement
Table of Contents
