shrn
Shift Right Narrow
SHRN <Vd>.<Tb>, <Vn>.<Ta>, #<shift>
Shifts wide vector right, narrowing to destination (Upper/Lower).
Details
Shifts each element of a wide vector right by an immediate amount, narrows the result to half the width, and stores in the destination register. The Q bit selects whether the operation produces the lower (Q=0) or upper (Q=1) half of the result. No flags are modified by this instruction. This is an AArch64 NEON instruction only.
Pseudocode Operation
for i = 0 to elements-1 do
temp ← Vn[i] >> shift
Vd[i] ← temp[element_width_dest-1:0]
endfor
Example
SHRN v0.4s.Tb, v1.4s.Ta, #LSL
Encoding
Binary Layout
0
Q
0
011110
immh
immb
1000
0
1
Rn
Rd
Operands
-
Vd
Dest (Narrow) -
Vn
Src (Wide) -
shift
Imm
Reference (Arm A64 ISA)
Instruction Forms
| Encoding | Instruction | ISA | Bit pattern | ||
|---|---|---|---|---|---|
| 0x0F008400 | SHRN{2} <Vd>.<Tb>, <Vn>.<Ta>, #<shift> | A64 | 0 | Q | 0 | 011110 | immh | immb | 1000 | 0 | 1 | Rn | Rd |
Description
Shift Right Narrow (immediate). This instruction reads each unsigned integer value from the source SIMD&FP register, right shifts each result by an immediate value, puts the final result into a vector, and writes the vector to the lower or upper half of the destination SIMD&FP register. The destination vector elements are half as long as the source vector elements. The results are truncated. For rounded results, see RSHRN.
The RSHRN instruction writes the vector to the lower half of the destination register and clears the upper half, while the RSHRN2 instruction writes the vector to the upper half of the destination register without affecting the other bits of the register.
Depending on the settings in the CPACR_EL1, CPTR_EL2, and CPTR_EL3 registers, and the current Security state and Exception level, an attempt to execute the instruction might be trapped.
Operation
CheckFPAdvSIMDEnabled64();
bits(datasize*2) operand = V[n, datasize*2];
bits(datasize) result;
integer element;
for e = 0 to elements-1
element = RShr(UInt(Elem[operand, e, 2*esize]), shift, round);
Elem[result, e, esize] = element<esize-1:0>;
Vpart[d, part, datasize] = result;