Search all intrinsics
SIMD ISAReturn TypeNameArgumentsInstruction Group
Neonint32_tvqrshls_s32(int32_t a, int32_t b)Shift / Left / Vector saturating rounding shift left
Description
Signed saturating Rounding Shift Left (register). This instruction takes each vector element in the first source SIMD&FP register, shifts it by a value from the least significant byte of the corresponding vector element of the second source SIMD&FP register, places the results into a vector, and writes the vector to the destination SIMD&FP register.
Results
Sd result
This intrinsic compiles to the following instructions:

SQRSHL Sd,Sn,Sm

Argument Preparation
a register: Snb register: Sm
Architectures
A64

Operation

CheckFPAdvSIMDEnabled64();
bits(datasize) operand1 = V[n];
bits(datasize) operand2 = V[m];
bits(datasize) result;

integer round_const = 0;
integer shift;
integer element;
boolean sat;

for e = 0 to elements-1
    shift = SInt(Elem[operand2, e, esize]<7:0>);
    if rounding then
        round_const = 1 << (-shift - 1); // 0 for left shift, 2^(n-1) for right shift 
    element = (Int(Elem[operand1, e, esize], unsigned) + round_const) << shift;
    if saturating then
        (Elem[result, e, esize], sat) = SatQ(element, esize, unsigned);
        if sat then FPSR.QC = '1';
    else
        Elem[result, e, esize] = element<esize-1:0>;

V[d] = result;