You copied the Doc URL to your clipboard.

SCVTF

Signed integer convert to floating-point (predicated).

Convert to floating-point from the signed integer in each active element of the source vector, and place the results in the corresponding elements of the destination vector. Inactive elements in the destination vector register remain unmodified.

If the input and result types have a different size the smaller type is held unpacked in the least significant bits of elements of the larger size. When the input is the smaller type the upper bits of each source element are ignored. When the result is the smaller type the results are zero-extended to fill each destination element.

It has encodings from 7 classes: 16-bit to half-precision , 32-bit to half-precision , 32-bit to single-precision , 32-bit to double-precision , 64-bit to half-precision , 64-bit to single-precision and 64-bit to double-precision

16-bit to half-precision

313029282726252423222120191817161514131211109876543210
0110010101010010101PgZnZd
int_U

16-bit to half-precision

SCVTF <Zd>.H, <Pg>/M, <Zn>.H

if !HaveSVE() then UNDEFINED;
integer esize = 16;
integer g = UInt(Pg);
integer n = UInt(Zn);
integer d = UInt(Zd);
integer s_esize = 16;
integer d_esize = 16;
boolean unsigned = FALSE;
FPRounding rounding = FPRoundingMode(FPCR<31:0>);

32-bit to half-precision

313029282726252423222120191817161514131211109876543210
0110010101010100101PgZnZd
int_U

32-bit to half-precision

SCVTF <Zd>.H, <Pg>/M, <Zn>.S

if !HaveSVE() then UNDEFINED;
integer esize = 32;
integer g = UInt(Pg);
integer n = UInt(Zn);
integer d = UInt(Zd);
integer s_esize = 32;
integer d_esize = 16;
boolean unsigned = FALSE;
FPRounding rounding = FPRoundingMode(FPCR<31:0>);

32-bit to single-precision

313029282726252423222120191817161514131211109876543210
0110010110010100101PgZnZd
int_U

32-bit to single-precision

SCVTF <Zd>.S, <Pg>/M, <Zn>.S

if !HaveSVE() then UNDEFINED;
integer esize = 32;
integer g = UInt(Pg);
integer n = UInt(Zn);
integer d = UInt(Zd);
integer s_esize = 32;
integer d_esize = 32;
boolean unsigned = FALSE;
FPRounding rounding = FPRoundingMode(FPCR<31:0>);

32-bit to double-precision

313029282726252423222120191817161514131211109876543210
0110010111010000101PgZnZd
int_U

32-bit to double-precision

SCVTF <Zd>.D, <Pg>/M, <Zn>.S

if !HaveSVE() then UNDEFINED;
integer esize = 64;
integer g = UInt(Pg);
integer n = UInt(Zn);
integer d = UInt(Zd);
integer s_esize = 32;
integer d_esize = 64;
boolean unsigned = FALSE;
FPRounding rounding = FPRoundingMode(FPCR<31:0>);

64-bit to half-precision

313029282726252423222120191817161514131211109876543210
0110010101010110101PgZnZd
int_U

64-bit to half-precision

SCVTF <Zd>.H, <Pg>/M, <Zn>.D

if !HaveSVE() then UNDEFINED;
integer esize = 64;
integer g = UInt(Pg);
integer n = UInt(Zn);
integer d = UInt(Zd);
integer s_esize = 64;
integer d_esize = 16;
boolean unsigned = FALSE;
FPRounding rounding = FPRoundingMode(FPCR<31:0>);

64-bit to single-precision

313029282726252423222120191817161514131211109876543210
0110010111010100101PgZnZd
int_U

64-bit to single-precision

SCVTF <Zd>.S, <Pg>/M, <Zn>.D

if !HaveSVE() then UNDEFINED;
integer esize = 64;
integer g = UInt(Pg);
integer n = UInt(Zn);
integer d = UInt(Zd);
integer s_esize = 64;
integer d_esize = 32;
boolean unsigned = FALSE;
FPRounding rounding = FPRoundingMode(FPCR<31:0>);

64-bit to double-precision

313029282726252423222120191817161514131211109876543210
0110010111010110101PgZnZd
int_U

64-bit to double-precision

SCVTF <Zd>.D, <Pg>/M, <Zn>.D

if !HaveSVE() then UNDEFINED;
integer esize = 64;
integer g = UInt(Pg);
integer n = UInt(Zn);
integer d = UInt(Zd);
integer s_esize = 64;
integer d_esize = 64;
boolean unsigned = FALSE;
FPRounding rounding = FPRoundingMode(FPCR<31:0>);

Assembler Symbols

<Zd>

Is the name of the destination scalable vector register, encoded in the "Zd" field.

<Pg>

Is the name of the governing scalable predicate register P0-P7, encoded in the "Pg" field.

<Zn>

Is the name of the source scalable vector register, encoded in the "Zn" field.

Operation

CheckSVEEnabled();
integer elements = VL DIV esize;
bits(PL) mask = P[g];
bits(VL) operand  = Z[n];
bits(VL) result = Z[d];

for e = 0 to elements-1
    bits(esize) element = Elem[operand, e, esize];
    if ElemP[mask, e, esize] == '1' then
        bits(d_esize) fpval = FixedToFP(element<s_esize-1:0>, 0, unsigned, FPCR<31:0>, rounding);
        Elem[result, e, esize] = ZeroExtend(fpval);

Z[d] = result;

Operational information

This instruction might be immediately preceded in program order by a MOVPRFX instruction. The MOVPRFX instruction must conform to all of the following requirements, otherwise the behavior of the MOVPRFX and this instruction is unpredictable:

  • The MOVPRFX instruction must be unpredicated, or be predicated using the same governing predicate register and source element size as this instruction.
  • The MOVPRFX instruction must specify the same destination register as this instruction.
  • The destination register must not refer to architectural register state referenced by any other source operand register of this instruction.