SVE Instruction List by Dougall Johnson
ST1W (scalar plus scalar, consecutive registers): Contiguous store of words from multiple consecutive vectors (scalar index)
ST1W { Zt1.S, Zt2.S }, PNg, [Xn, Xm, LSL #2] (SVE2.1 (SME2+S
128-bit SVE
Store 32-bit values from the two consecutive registers (2) and (3) to the memory operand (4). After decoding the predicate (1) from its predicate-as-counter representation to a double-length predicate, if the predicate bit corresponding to an element is zero, that store is skipped, and cannot cause a fault, and the corresponding value in memory is unchanged. The first register number (2) must be divisible by two.
256-bit SVE
Store 32-bit values from the two consecutive registers (2) and (3) to the memory operand (4). After decoding the predicate (1) from its predicate-as-counter representation to a double-length predicate, if the predicate bit corresponding to an element is zero, that store is skipped, and cannot cause a fault, and the corresponding value in memory is unchanged. The first register number (2) must be divisible by two.
512-bit SVE
Store 32-bit values from the two consecutive registers (2) and (3) to the memory operand (4). After decoding the predicate (1) from its predicate-as-counter representation to a double-length predicate, if the predicate bit corresponding to an element is zero, that store is skipped, and cannot cause a fault, and the corresponding value in memory is unchanged. The first register number (2) must be divisible by two.
Larger sizes
1024-bit SVE
Store 32-bit values from the two consecutive registers (2) and (3) to the memory operand (4). After decoding the predicate (1) from its predicate-as-counter representation to a double-length predicate, if the predicate bit corresponding to an element is zero, that store is skipped, and cannot cause a fault, and the corresponding value in memory is unchanged. The first register number (2) must be divisible by two.
2048-bit SVE
Store 32-bit values from the two consecutive registers (2) and (3) to the memory operand (4). After decoding the predicate (1) from its predicate-as-counter representation to a double-length predicate, if the predicate bit corresponding to an element is zero, that store is skipped, and cannot cause a fault, and the corresponding value in memory is unchanged. The first register number (2) must be divisible by two.
Report mistakes or give feedback
Inspired by and based on the x86/x64 SIMD Instruction List by Daytime.