== "D" Extension for Double-Precision Floating-Point, Version 2.2

This chapter describes the standard double-precision floating-point
instruction-set extension, which is named "D" and adds
double-precision floating-point computational instructions compliant
with the IEEE 754-2008 arithmetic standard. The D extension depends on
the base single-precision instruction subset F.
(((double-precision, floating point)))
(((floating point, double precision)))

=== D Register State

The D extension widens the 32 floating-point registers, `f0-f31`, to
64 bits (FLEN=64 in <<fprs>>. The `f` registers can
now hold either 32-bit or 64-bit floating-point values as described
below in <<nanboxing>>.

[NOTE]
====
FLEN can be 32, 64, or 128 depending on which of the F, D, and Q
extensions are supported. There can be up to four different
floating-point precisions supported, including H, F, D, and Q.
====
(((floating-point, supported precisions)))

[[nanboxing]]
=== NaN Boxing of Narrower Values

When multiple floating-point precisions are supported, then valid values
of narrower _n_-bit types, _n_<FLEN, are represented in the lower _n_ bits of an FLEN-bit NaN value, in a process termed NaN-boxing. The upper bits of a valid NaN-boxed value must be all 1s. Valid NaN-boxed _n_-bit values
therefore appear as negative quiet NaNs (qNaNs) when viewed as any wider
_m_-bit value, _n_ < _m_ &#8804; FLEN. Any operation that writes a narrower result to an 'f' register must write all 1s to the uppermost FLEN-_n_ bits to yield a legal NaN-boxedvalue.
(((floating-point, requirements)))

[NOTE]
====
Software might not know the current type of data stored in a
floating-point register but has to be able to save and restore the
register values, hence the result of using wider operations to transfer
narrower values has to be defined. A common case is for callee-saved
registers, but a standard convention is also desirable for features
including varargs, user-level threading libraries, virtual machine
migration, and debugging.
====

Floating-point _n_-bit transfer operations move external
values held in IEEE standard formats into and out of the `f` registers,
and comprise floating-point loads and stores (FL__n__/FS__n__) and floating-point move instructions (FMV._n_.X/FMV.X._n_). A narrower _n_-bit transfer, _n_<FLEN, into the `f` registers will create a valid NaN-boxed value. A narrower
_n_-bit transfer out of the floating-point registers will
transfer the lower _n_ bits of the register ignoring the
upper FLEN-_n_ bits.

Apart from transfer operations described in the previous paragraph, all
other floating-point operations on narrower __n__-bit
operations, _n_<FLEN, check if the input operands are
correctly NaN-boxed, i.e., all upper FLEN-_n_ bits are 1. If
so, the _n_ least-significant bits of the input are used as
the input value, otherwise the input value is treated as an
_n_-bit canonical NaN.

[TIP]
====
Earlier versions of this document did not define the behavior of feeding
the results of narrower or wider operands into an operation, except to
require that wider saves and restores would preserve the value of a
narrower operand. The new definition removes this
implementation-specific behavior, while still accommodating both
non-recoded and recoded implementations of the floating-point unit. The
new definition also helps catch software errors by propagating NaNs if
values are used incorrectly.

Non-recoded implementations unpack and pack the operands to IEEE
standard format on the input and output of every floating-point
operation. The NaN-boxing cost to a non-recoded implementation is
primarily in checking if the upper bits of a narrower operation
represent a legal NaN-boxed value, and in writing all 1s to the upper
bits of a result.

Recoded implementations use a more convenient internal format to
represent floating-point values, with an added exponent bit to allow all
values to be held normalized. The cost to the recoded implementation is
primarily the extra tagging needed to track the internal types and sign
bits, but this can be done without adding new state bits by recoding
NaNs internally in the exponent field. Small modifications are needed to
the pipelines used to transfer values in and out of the recoded format,
but the datapath and latency costs are minimal. The recoding process has
to handle shifting of input subnormal values for wide operands in any
case, and extracting the NaN-boxed value is a similar process to
normalization except for skipping over leading-1 bits instead of
skipping over leading-0 bits, allowing the datapath muxing to be shared.
====

[[fld_fsd]]
=== Double-Precision Load and Store Instructions

The FLD instruction loads a double-precision floating-point value from
memory into floating-point register _rd_. FSD stores a double-precision
value from the floating-point registers to memory.
(((floating point, load and store)))

[NOTE]
====
The double-precision value may be a NaN-boxed single-precision value.
====

include::images/wavedrom/double-ls.adoc[]
[[double-ls]]
//.Double-precision load and store

FLD and FSD are only guaranteed to execute atomically if the effective
address is naturally aligned and XLEN&#8805;64.

FLD and FSD do not modify the bits being transferred; in particular, the
payloads of non-canonical NaNs are preserved.

=== Double-Precision Floating-Point Computational Instructions

The double-precision floating-point computational instructions are
defined analogously to their single-precision counterparts, but operate
on double-precision operands and produce double-precision results.

include::images/wavedrom/double-fl-compute.adoc[]
[[fl-compute]]
//.Double-precision float computational

=== Double-Precision Floating-Point Conversion and Move Instructions

Floating-point-to-integer and integer-to-floating-point conversion
instructions are encoded in the OP-FP major opcode space. FCVT.W.D or
FCVT.L.D converts a double-precision floating-point number in
floating-point register _rs1_ to a signed 32-bit or 64-bit integer,
respectively, in integer register _rd_. FCVT.D.W or FCVT.D.L converts a
32-bit or 64-bit signed integer, respectively, in integer register _rs1_
into a double-precision floating-point number in floating-point register
_rd_. FCVT.WU.D, FCVT.LU.D, FCVT.D.WU, and FCVT.D.LU variants convert to
or from unsigned integer values. For RV64, FCVT.W[U].D sign-extends the
32-bit result. FCVT.L[U].D and FCVT.D.L[U] are RV64-only instructions.
The range of valid inputs for FCVT._int_.D and the behavior for invalid
inputs are the same as for FCVT._int_.S.
(((floating point, convert and move)))

All floating-point to integer and integer to floating-point conversion
instructions round according to the _rm_ field. Note FCVT.D.W[U] always
produces an exact result and is unaffected by rounding mode.

include::images/wavedrom/double-fl-convert-mv.adoc[]
[[fl-convert-mv]]
//.Double-precision float convert and move

The double-precision to single-precision and single-precision to
double-precision conversion instructions, FCVT.S.D and FCVT.D.S, are
encoded in the OP-FP major opcode space and both the source and
destination are floating-point registers. The _rs2_ field encodes the
datatype of the source, and the _fmt_ field encodes the datatype of the
destination. FCVT.S.D rounds according to the RM field; FCVT.D.S will
never round.
(((double-precision, to single-precision)))
(((single-precision, to double-precision )))

include::images/wavedrom/fcvt-sd-ds.adoc[]
[[fcvt-sd-ds]]
//.Double-precision FCVT.S.D and FCVT.D.S

Floating-point to floating-point sign-injection instructions, FSGNJ.D,
FSGNJN.D, and FSGNJX.D are defined analogously to the single-precision
sign-injection instruction.
//FSGNJ.D, FSGNJN.D, and FSGNJX.D

include::images/wavedrom/fsjgnjnx-d.adoc[]
//.Double-precision sign-injection

For XLEN&#8805;64 only, instructions are provided to move bit
patterns between the floating-point and integer registers. FMV.X.D moves
the double-precision value in floating-point register _rs1_ to a
representation in IEEE 754-2008 standard encoding in integer register
_rd_. FMV.D.X moves the double-precision value encoded in IEEE 754-2008
standard encoding from the integer register _rs1_ to the floating-point
register _rd_.

FMV.X.D and FMV.D.X do not modify the bits being transferred; in
particular, the payloads of non-canonical NaNs are preserved.

include::images/wavedrom/d-xwwx.adoc[]
[[fmvxddx]]
//.Double-precision float move to _rd_

[TIP]
====
Early versions of the RISC-V ISA had additional instructions to allow
RV32 systems to transfer between the upper and lower portions of a
64-bit floating-point register and an integer register. However, these
would be the only instructions with partial register writes and would
add complexity in implementations with recoded floating-point or
register renaming, requiring a pipeline read-modify-write sequence.
Scaling up to handling quad-precision for RV32 and RV64 would also
require additional instructions if they were to follow this pattern. The
ISA was defined to reduce the number of explicit int-float register
moves, by having conversions and comparisons write results to the
appropriate register file, so we expect the benefit of these
instructions to be lower than for other ISAs.

We note that for systems that implement a 64-bit floating-point unit
including fused multiply-add support and 64-bit floating-point loads and
stores, the marginal hardware cost of moving from a 32-bit to a 64-bit
integer datapath is low, and a software ABI supporting 32-bit wide
address-space and pointers can be used to avoid growth of static data
and dynamic memory traffic.
====

=== Double-Precision Floating-Point Compare Instructions

The double-precision floating-point compare instructions are defined
analogously to their single-precision counterparts, but operate on
double-precision operands.
(((floating-point, compare)))

include::images/wavedrom/double-fl-compare.adoc[]
[[fl-compare]]
//.Double-precision float compare

=== Double-Precision Floating-Point Classify Instruction

The double-precision floating-point classify instruction, FCLASS.D, is
defined analogously to its single-precision counterpart, but operates on
double-precision operands.
(((floating-point, classify)))

include::images/wavedrom/double-fl-class.adoc[]
[[fl-class]]
//.Double-precision float classify