test/CodeGen/AArch64/sve2p1-intrinsics-stores.ll

; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc -mtriple=aarch64-linux-gnu -mattr=+sve2p1,+bf16 < %s | FileCheck %s
; RUN: llc -mtriple=aarch64-linux-gnu -mattr=+sme2,+bf16 < %s | FileCheck %s

; == Normal Multi-Vector Consecutive Stores ==

define void @st1_x2_i8(<vscale x 16 x i8> %unused, <vscale x 16 x i8> %zn0, <vscale x 16 x i8> %zn1, target("aarch64.svcount") %pn, ptr %ptr) nounwind {
; CHECK-LABEL: st1_x2_i8:
; CHECK:       // %bb.0:
; CHECK-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
; CHECK-NEXT:    addvl sp, sp, #-1
; CHECK-NEXT:    mov z3.d, z2.d
; CHECK-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
; CHECK-NEXT:    mov p8.b, p0.b
; CHECK-NEXT:    mov z2.d, z1.d
; CHECK-NEXT:    st1b { z2.b, z3.b }, pn8, [x0]
; CHECK-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
; CHECK-NEXT:    addvl sp, sp, #1
; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT:    ret
  call void @llvm.aarch64.sve.st1.pn.x2.nxv16i8(<vscale x 16 x i8> %zn0, <vscale x 16 x i8> %zn1, target("aarch64.svcount") %pn, ptr %ptr);
  ret void
}

define void @st1_x2_i16(<vscale x 16 x i8> %unused, <vscale x 8 x i16> %zn0, <vscale x 8 x i16> %zn1, target("aarch64.svcount") %pn, ptr %ptr) nounwind {
; CHECK-LABEL: st1_x2_i16:
; CHECK:       // %bb.0:
; CHECK-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
; CHECK-NEXT:    addvl sp, sp, #-1
; CHECK-NEXT:    mov z3.d, z2.d
; CHECK-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
; CHECK-NEXT:    mov p8.b, p0.b
; CHECK-NEXT:    mov z2.d, z1.d
; CHECK-NEXT:    st1h { z2.h, z3.h }, pn8, [x0]
; CHECK-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
; CHECK-NEXT:    addvl sp, sp, #1
; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT:    ret
  call void @llvm.aarch64.sve.st1.pn.x2.nxv8i16(<vscale x 8 x i16> %zn0, <vscale x 8 x i16> %zn1, target("aarch64.svcount") %pn, ptr %ptr);
  ret void
}

define void @st1_x2_i32(<vscale x 16 x i8> %unused, <vscale x 4 x i32> %zn0, <vscale x 4 x i32> %zn1, target("aarch64.svcount") %pn, ptr %ptr) nounwind {
; CHECK-LABEL: st1_x2_i32:
; CHECK:       // %bb.0:
; CHECK-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
; CHECK-NEXT:    addvl sp, sp, #-1
; CHECK-NEXT:    mov z3.d, z2.d
; CHECK-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
; CHECK-NEXT:    mov p8.b, p0.b
; CHECK-NEXT:    mov z2.d, z1.d
; CHECK-NEXT:    st1w { z2.s, z3.s }, pn8, [x0]
; CHECK-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
; CHECK-NEXT:    addvl sp, sp, #1
; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT:    ret
  call void @llvm.aarch64.sve.st1.pn.x2.nxv4i32(<vscale x 4 x i32> %zn0, <vscale x 4 x i32> %zn1, target("aarch64.svcount") %pn, ptr %ptr);
  ret void
}

define void @st1_x2_i64(<vscale x 16 x i8> %unused, <vscale x 2 x i64> %zn0, <vscale x 2 x i64> %zn1, target("aarch64.svcount") %pn, ptr %ptr) nounwind {
; CHECK-LABEL: st1_x2_i64:
; CHECK:       // %bb.0:
; CHECK-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
; CHECK-NEXT:    addvl sp, sp, #-1
; CHECK-NEXT:    mov z3.d, z2.d
; CHECK-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
; CHECK-NEXT:    mov p8.b, p0.b
; CHECK-NEXT:    mov z2.d, z1.d
; CHECK-NEXT:    st1d { z2.d, z3.d }, pn8, [x0]
; CHECK-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
; CHECK-NEXT:    addvl sp, sp, #1
; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT:    ret
  call void @llvm.aarch64.sve.st1.pn.x2.nxv2i64(<vscale x 2 x i64> %zn0, <vscale x 2 x i64> %zn1, target("aarch64.svcount") %pn, ptr %ptr);
  ret void
}

define void @st1_x2_f16(<vscale x 16 x i8> %unused, <vscale x 8 x half> %zn0, <vscale x 8 x half> %zn1, target("aarch64.svcount") %pn, ptr %ptr) nounwind {
; CHECK-LABEL: st1_x2_f16:
; CHECK:       // %bb.0:
; CHECK-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
; CHECK-NEXT:    addvl sp, sp, #-1
; CHECK-NEXT:    mov z3.d, z2.d
; CHECK-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
; CHECK-NEXT:    mov p8.b, p0.b
; CHECK-NEXT:    mov z2.d, z1.d
; CHECK-NEXT:    st1h { z2.h, z3.h }, pn8, [x0]
; CHECK-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
; CHECK-NEXT:    addvl sp, sp, #1
; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT:    ret
  call void @llvm.aarch64.sve.st1.pn.x2.nxv8f16(<vscale x 8 x half> %zn0, <vscale x 8 x half> %zn1, target("aarch64.svcount") %pn, ptr %ptr);
  ret void
}

define void @st1_x2_bf16(<vscale x 16 x i8> %unused, <vscale x 8 x bfloat> %zn0, <vscale x 8 x bfloat> %zn1, target("aarch64.svcount") %pn, ptr %ptr) nounwind {
; CHECK-LABEL: st1_x2_bf16:
; CHECK:       // %bb.0:
; CHECK-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
; CHECK-NEXT:    addvl sp, sp, #-1
; CHECK-NEXT:    mov z3.d, z2.d
; CHECK-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
; CHECK-NEXT:    mov p8.b, p0.b
; CHECK-NEXT:    mov z2.d, z1.d
; CHECK-NEXT:    st1h { z2.h, z3.h }, pn8, [x0]
; CHECK-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
; CHECK-NEXT:    addvl sp, sp, #1
; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT:    ret
  call void @llvm.aarch64.sve.st1.pn.x2.nxv8bf16(<vscale x 8 x bfloat> %zn0, <vscale x 8 x bfloat> %zn1, target("aarch64.svcount") %pn, ptr %ptr);
  ret void
}

define void @st1_x2_f32(<vscale x 16 x i8> %unused, <vscale x 4 x float> %zn0, <vscale x 4 x float> %zn1, target("aarch64.svcount") %pn, ptr %ptr) nounwind {
; CHECK-LABEL: st1_x2_f32:
; CHECK:       // %bb.0:
; CHECK-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
; CHECK-NEXT:    addvl sp, sp, #-1
; CHECK-NEXT:    mov z3.d, z2.d
; CHECK-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
; CHECK-NEXT:    mov p8.b, p0.b
; CHECK-NEXT:    mov z2.d, z1.d
; CHECK-NEXT:    st1w { z2.s, z3.s }, pn8, [x0]
; CHECK-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
; CHECK-NEXT:    addvl sp, sp, #1
; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT:    ret
  call void @llvm.aarch64.sve.st1.pn.x2.nxv4f32(<vscale x 4 x float> %zn0, <vscale x 4 x float> %zn1, target("aarch64.svcount") %pn, ptr %ptr);
  ret void
}

define void @st1_x2_f64(<vscale x 16 x i8> %unused, <vscale x 2 x double> %zn0, <vscale x 2 x double> %zn1, target("aarch64.svcount") %pn, ptr %ptr) nounwind {
; CHECK-LABEL: st1_x2_f64:
; CHECK:       // %bb.0:
; CHECK-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
; CHECK-NEXT:    addvl sp, sp, #-1
; CHECK-NEXT:    mov z3.d, z2.d
; CHECK-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
; CHECK-NEXT:    mov p8.b, p0.b
; CHECK-NEXT:    mov z2.d, z1.d
; CHECK-NEXT:    st1d { z2.d, z3.d }, pn8, [x0]
; CHECK-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
; CHECK-NEXT:    addvl sp, sp, #1
; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT:    ret
  call void @llvm.aarch64.sve.st1.pn.x2.nxv2f64(<vscale x 2 x double> %zn0, <vscale x 2 x double> %zn1, target("aarch64.svcount") %pn, ptr %ptr);
  ret void
}

define void @st1_x4_i8(<vscale x 16 x i8> %unused, <vscale x 16 x i8> %zn0, <vscale x 16 x i8> %zn1, <vscale x 16 x i8> %zn2, <vscale x 16 x i8> %zn3, target("aarch64.svcount") %pn, ptr %ptr) nounwind {
; CHECK-LABEL: st1_x4_i8:
; CHECK:       // %bb.0:
; CHECK-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
; CHECK-NEXT:    addvl sp, sp, #-1
; CHECK-NEXT:    mov z7.d, z4.d
; CHECK-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
; CHECK-NEXT:    mov z6.d, z3.d
; CHECK-NEXT:    mov p8.b, p0.b
; CHECK-NEXT:    mov z5.d, z2.d
; CHECK-NEXT:    mov z4.d, z1.d
; CHECK-NEXT:    st1b { z4.b - z7.b }, pn8, [x0]
; CHECK-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
; CHECK-NEXT:    addvl sp, sp, #1
; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT:    ret
  call void @llvm.aarch64.sve.st1.pn.x4.nxv16i8(<vscale x 16 x i8> %zn0, <vscale x 16 x i8> %zn1, <vscale x 16 x i8> %zn2, <vscale x 16 x i8> %zn3, target("aarch64.svcount") %pn, ptr %ptr);
  ret void
}

define void @st1_x4_i16(<vscale x 16 x i8> %unused, <vscale x 8 x i16> %zn0, <vscale x 8 x i16> %zn1, <vscale x 8 x i16> %zn2, <vscale x 8 x i16> %zn3, target("aarch64.svcount") %pn, ptr %ptr) nounwind {
; CHECK-LABEL: st1_x4_i16:
; CHECK:       // %bb.0:
; CHECK-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
; CHECK-NEXT:    addvl sp, sp, #-1
; CHECK-NEXT:    mov z7.d, z4.d
; CHECK-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
; CHECK-NEXT:    mov z6.d, z3.d
; CHECK-NEXT:    mov p8.b, p0.b
; CHECK-NEXT:    mov z5.d, z2.d
; CHECK-NEXT:    mov z4.d, z1.d
; CHECK-NEXT:    st1h { z4.h - z7.h }, pn8, [x0]
; CHECK-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
; CHECK-NEXT:    addvl sp, sp, #1
; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT:    ret
  call void @llvm.aarch64.sve.st1.pn.x4.nxv8i16(<vscale x 8 x i16> %zn0, <vscale x 8 x i16> %zn1, <vscale x 8 x i16> %zn2, <vscale x 8 x i16> %zn3, target("aarch64.svcount") %pn, ptr %ptr);
  ret void
}

define void @st1_x4_i32(<vscale x 16 x i8> %unused, <vscale x 4 x i32> %zn0, <vscale x 4 x i32> %zn1, <vscale x 4 x i32> %zn2, <vscale x 4 x i32> %zn3, target("aarch64.svcount") %pn, ptr %ptr) nounwind {
; CHECK-LABEL: st1_x4_i32:
; CHECK:       // %bb.0:
; CHECK-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
; CHECK-NEXT:    addvl sp, sp, #-1
; CHECK-NEXT:    mov z7.d, z4.d
; CHECK-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
; CHECK-NEXT:    mov z6.d, z3.d
; CHECK-NEXT:    mov p8.b, p0.b
; CHECK-NEXT:    mov z5.d, z2.d
; CHECK-NEXT:    mov z4.d, z1.d
; CHECK-NEXT:    st1w { z4.s - z7.s }, pn8, [x0]
; CHECK-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
; CHECK-NEXT:    addvl sp, sp, #1
; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT:    ret
  call void @llvm.aarch64.sve.st1.pn.x4.nxv4i32(<vscale x 4 x i32> %zn0, <vscale x 4 x i32> %zn1, <vscale x 4 x i32> %zn2, <vscale x 4 x i32> %zn3, target("aarch64.svcount") %pn, ptr %ptr);
  ret void
}

define void @st1_x4_i64(<vscale x 16 x i8> %unused, <vscale x 2 x i64> %zn0, <vscale x 2 x i64> %zn1, <vscale x 2 x i64> %zn2, <vscale x 2 x i64> %zn3, target("aarch64.svcount") %pn, ptr %ptr) nounwind {
; CHECK-LABEL: st1_x4_i64:
; CHECK:       // %bb.0:
; CHECK-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
; CHECK-NEXT:    addvl sp, sp, #-1
; CHECK-NEXT:    mov z7.d, z4.d
; CHECK-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
; CHECK-NEXT:    mov z6.d, z3.d
; CHECK-NEXT:    mov p8.b, p0.b
; CHECK-NEXT:    mov z5.d, z2.d
; CHECK-NEXT:    mov z4.d, z1.d
; CHECK-NEXT:    st1d { z4.d - z7.d }, pn8, [x0]
; CHECK-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
; CHECK-NEXT:    addvl sp, sp, #1
; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT:    ret
  call void @llvm.aarch64.sve.st1.pn.x4.nxv2i64(<vscale x 2 x i64> %zn0, <vscale x 2 x i64> %zn1, <vscale x 2 x i64> %zn2, <vscale x 2 x i64> %zn3, target("aarch64.svcount") %pn, ptr %ptr);
  ret void
}

define void @st1_x4_f16(<vscale x 16 x i8> %unused, <vscale x 8 x half> %zn0, <vscale x 8 x half> %zn1, <vscale x 8 x half> %zn2, <vscale x 8 x half> %zn3, target("aarch64.svcount") %pn, ptr %ptr) nounwind {
; CHECK-LABEL: st1_x4_f16:
; CHECK:       // %bb.0:
; CHECK-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
; CHECK-NEXT:    addvl sp, sp, #-1
; CHECK-NEXT:    mov z7.d, z4.d
; CHECK-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
; CHECK-NEXT:    mov z6.d, z3.d
; CHECK-NEXT:    mov p8.b, p0.b
; CHECK-NEXT:    mov z5.d, z2.d
; CHECK-NEXT:    mov z4.d, z1.d
; CHECK-NEXT:    st1h { z4.h - z7.h }, pn8, [x0]
; CHECK-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
; CHECK-NEXT:    addvl sp, sp, #1
; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT:    ret
  call void @llvm.aarch64.sve.st1.pn.x4.nxv8f16(<vscale x 8 x half> %zn0, <vscale x 8 x half> %zn1, <vscale x 8 x half> %zn2, <vscale x 8 x half> %zn3, target("aarch64.svcount") %pn, ptr %ptr);
  ret void
}

define void @st1_x4_bf16(<vscale x 16 x i8> %unused, <vscale x 8 x bfloat> %zn0, <vscale x 8 x bfloat> %zn1, <vscale x 8 x bfloat> %zn2, <vscale x 8 x bfloat> %zn3, target("aarch64.svcount") %pn, ptr %ptr) nounwind {
; CHECK-LABEL: st1_x4_bf16:
; CHECK:       // %bb.0:
; CHECK-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
; CHECK-NEXT:    addvl sp, sp, #-1
; CHECK-NEXT:    mov z7.d, z4.d
; CHECK-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
; CHECK-NEXT:    mov z6.d, z3.d
; CHECK-NEXT:    mov p8.b, p0.b
; CHECK-NEXT:    mov z5.d, z2.d
; CHECK-NEXT:    mov z4.d, z1.d
; CHECK-NEXT:    st1h { z4.h - z7.h }, pn8, [x0]
; CHECK-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
; CHECK-NEXT:    addvl sp, sp, #1
; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT:    ret
  call void @llvm.aarch64.sve.st1.pn.x4.nxv8bf16(<vscale x 8 x bfloat> %zn0, <vscale x 8 x bfloat> %zn1, <vscale x 8 x bfloat> %zn2, <vscale x 8 x bfloat> %zn3, target("aarch64.svcount") %pn, ptr %ptr);
  ret void
}

define void @st1_x4_f32(<vscale x 16 x i8> %unused, <vscale x 4 x float> %zn0, <vscale x 4 x float> %zn1, <vscale x 4 x float> %zn2, <vscale x 4 x float> %zn3, target("aarch64.svcount") %pn, ptr %ptr) nounwind {
; CHECK-LABEL: st1_x4_f32:
; CHECK:       // %bb.0:
; CHECK-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
; CHECK-NEXT:    addvl sp, sp, #-1
; CHECK-NEXT:    mov z7.d, z4.d
; CHECK-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
; CHECK-NEXT:    mov z6.d, z3.d
; CHECK-NEXT:    mov p8.b, p0.b
; CHECK-NEXT:    mov z5.d, z2.d
; CHECK-NEXT:    mov z4.d, z1.d
; CHECK-NEXT:    st1w { z4.s - z7.s }, pn8, [x0]
; CHECK-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
; CHECK-NEXT:    addvl sp, sp, #1
; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT:    ret
  call void @llvm.aarch64.sve.st1.pn.x4.nxv4f32(<vscale x 4 x float> %zn0, <vscale x 4 x float> %zn1, <vscale x 4 x float> %zn2, <vscale x 4 x float> %zn3, target("aarch64.svcount") %pn, ptr %ptr);
  ret void
}

define void @st1_x4_f64(<vscale x 16 x i8> %unused, <vscale x 2 x double> %zn0, <vscale x 2 x double> %zn1, <vscale x 2 x double> %zn2, <vscale x 2 x double> %zn3, target("aarch64.svcount") %pn, ptr %ptr) nounwind {
; CHECK-LABEL: st1_x4_f64:
; CHECK:       // %bb.0:
; CHECK-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
; CHECK-NEXT:    addvl sp, sp, #-1
; CHECK-NEXT:    mov z7.d, z4.d
; CHECK-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
; CHECK-NEXT:    mov z6.d, z3.d
; CHECK-NEXT:    mov p8.b, p0.b
; CHECK-NEXT:    mov z5.d, z2.d
; CHECK-NEXT:    mov z4.d, z1.d
; CHECK-NEXT:    st1d { z4.d - z7.d }, pn8, [x0]
; CHECK-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
; CHECK-NEXT:    addvl sp, sp, #1
; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT:    ret
  call void @llvm.aarch64.sve.st1.pn.x4.nxv2f64(<vscale x 2 x double> %zn0, <vscale x 2 x double> %zn1, <vscale x 2 x double> %zn2, <vscale x 2 x double> %zn3, target("aarch64.svcount") %pn, ptr %ptr);
  ret void
}

; == Non-temporal Multi-Vector Consecutive Stores ==

define void @stnt1_x2_i8(<vscale x 16 x i8> %unused, <vscale x 16 x i8> %zn0, <vscale x 16 x i8> %zn1, target("aarch64.svcount") %pn, ptr %ptr) nounwind {
; CHECK-LABEL: stnt1_x2_i8:
; CHECK:       // %bb.0:
; CHECK-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
; CHECK-NEXT:    addvl sp, sp, #-1
; CHECK-NEXT:    mov z3.d, z2.d
; CHECK-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
; CHECK-NEXT:    mov p8.b, p0.b
; CHECK-NEXT:    mov z2.d, z1.d
; CHECK-NEXT:    stnt1b { z2.b, z3.b }, pn8, [x0]
; CHECK-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
; CHECK-NEXT:    addvl sp, sp, #1
; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT:    ret
  call void @llvm.aarch64.sve.stnt1.pn.x2.nxv16i8(<vscale x 16 x i8> %zn0, <vscale x 16 x i8> %zn1, target("aarch64.svcount") %pn, ptr %ptr);
  ret void
}

define void @stnt1_x2_i16(<vscale x 16 x i8> %unused, <vscale x 8 x i16> %zn0, <vscale x 8 x i16> %zn1, target("aarch64.svcount") %pn, ptr %ptr) nounwind {
; CHECK-LABEL: stnt1_x2_i16:
; CHECK:       // %bb.0:
; CHECK-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
; CHECK-NEXT:    addvl sp, sp, #-1
; CHECK-NEXT:    mov z3.d, z2.d
; CHECK-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
; CHECK-NEXT:    mov p8.b, p0.b
; CHECK-NEXT:    mov z2.d, z1.d
; CHECK-NEXT:    stnt1h { z2.h, z3.h }, pn8, [x0]
; CHECK-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
; CHECK-NEXT:    addvl sp, sp, #1
; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT:    ret
  call void @llvm.aarch64.sve.stnt1.pn.x2.nxv8i16(<vscale x 8 x i16> %zn0, <vscale x 8 x i16> %zn1, target("aarch64.svcount") %pn, ptr %ptr);
  ret void
}

define void @stnt1_x2_i32(<vscale x 16 x i8> %unused, <vscale x 4 x i32> %zn0, <vscale x 4 x i32> %zn1, target("aarch64.svcount") %pn, ptr %ptr) nounwind {
; CHECK-LABEL: stnt1_x2_i32:
; CHECK:       // %bb.0:
; CHECK-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
; CHECK-NEXT:    addvl sp, sp, #-1
; CHECK-NEXT:    mov z3.d, z2.d
; CHECK-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
; CHECK-NEXT:    mov p8.b, p0.b
; CHECK-NEXT:    mov z2.d, z1.d
; CHECK-NEXT:    stnt1w { z2.s, z3.s }, pn8, [x0]
; CHECK-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
; CHECK-NEXT:    addvl sp, sp, #1
; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT:    ret
  call void @llvm.aarch64.sve.stnt1.pn.x2.nxv4i32(<vscale x 4 x i32> %zn0, <vscale x 4 x i32> %zn1, target("aarch64.svcount") %pn, ptr %ptr);
  ret void
}

define void @stnt1_x2_i64(<vscale x 16 x i8> %unused, <vscale x 2 x i64> %zn0, <vscale x 2 x i64> %zn1, target("aarch64.svcount") %pn, ptr %ptr) nounwind {
; CHECK-LABEL: stnt1_x2_i64:
; CHECK:       // %bb.0:
; CHECK-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
; CHECK-NEXT:    addvl sp, sp, #-1
; CHECK-NEXT:    mov z3.d, z2.d
; CHECK-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
; CHECK-NEXT:    mov p8.b, p0.b
; CHECK-NEXT:    mov z2.d, z1.d
; CHECK-NEXT:    stnt1d { z2.d, z3.d }, pn8, [x0]
; CHECK-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
; CHECK-NEXT:    addvl sp, sp, #1
; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT:    ret
  call void @llvm.aarch64.sve.stnt1.pn.x2.nxv2i64(<vscale x 2 x i64> %zn0, <vscale x 2 x i64> %zn1, target("aarch64.svcount") %pn, ptr %ptr);
  ret void
}

define void @stnt1_x2_f16(<vscale x 16 x i8> %unused, <vscale x 8 x half> %zn0, <vscale x 8 x half> %zn1, target("aarch64.svcount") %pn, ptr %ptr) nounwind {
; CHECK-LABEL: stnt1_x2_f16:
; CHECK:       // %bb.0:
; CHECK-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
; CHECK-NEXT:    addvl sp, sp, #-1
; CHECK-NEXT:    mov z3.d, z2.d
; CHECK-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
; CHECK-NEXT:    mov p8.b, p0.b
; CHECK-NEXT:    mov z2.d, z1.d
; CHECK-NEXT:    stnt1h { z2.h, z3.h }, pn8, [x0]
; CHECK-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
; CHECK-NEXT:    addvl sp, sp, #1
; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT:    ret
  call void @llvm.aarch64.sve.stnt1.pn.x2.nxv8f16(<vscale x 8 x half> %zn0, <vscale x 8 x half> %zn1, target("aarch64.svcount") %pn, ptr %ptr);
  ret void
}

define void @stnt1_x2_bf16(<vscale x 16 x i8> %unused, <vscale x 8 x bfloat> %zn0, <vscale x 8 x bfloat> %zn1, target("aarch64.svcount") %pn, ptr %ptr) nounwind {
; CHECK-LABEL: stnt1_x2_bf16:
; CHECK:       // %bb.0:
; CHECK-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
; CHECK-NEXT:    addvl sp, sp, #-1
; CHECK-NEXT:    mov z3.d, z2.d
; CHECK-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
; CHECK-NEXT:    mov p8.b, p0.b
; CHECK-NEXT:    mov z2.d, z1.d
; CHECK-NEXT:    stnt1h { z2.h, z3.h }, pn8, [x0]
; CHECK-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
; CHECK-NEXT:    addvl sp, sp, #1
; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT:    ret
  call void @llvm.aarch64.sve.stnt1.pn.x2.nxv8bf16(<vscale x 8 x bfloat> %zn0, <vscale x 8 x bfloat> %zn1, target("aarch64.svcount") %pn, ptr %ptr);
  ret void
}

define void @stnt1_x2_f32(<vscale x 16 x i8> %unused, <vscale x 4 x float> %zn0, <vscale x 4 x float> %zn1, target("aarch64.svcount") %pn, ptr %ptr) nounwind {
; CHECK-LABEL: stnt1_x2_f32:
; CHECK:       // %bb.0:
; CHECK-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
; CHECK-NEXT:    addvl sp, sp, #-1
; CHECK-NEXT:    mov z3.d, z2.d
; CHECK-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
; CHECK-NEXT:    mov p8.b, p0.b
; CHECK-NEXT:    mov z2.d, z1.d
; CHECK-NEXT:    stnt1w { z2.s, z3.s }, pn8, [x0]
; CHECK-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
; CHECK-NEXT:    addvl sp, sp, #1
; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT:    ret
  call void @llvm.aarch64.sve.stnt1.pn.x2.nxv4f32(<vscale x 4 x float> %zn0, <vscale x 4 x float> %zn1, target("aarch64.svcount") %pn, ptr %ptr);
  ret void
}

define void @stnt1_x2_f64(<vscale x 16 x i8> %unused, <vscale x 2 x double> %zn0, <vscale x 2 x double> %zn1, target("aarch64.svcount") %pn, ptr %ptr) nounwind {
; CHECK-LABEL: stnt1_x2_f64:
; CHECK:       // %bb.0:
; CHECK-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
; CHECK-NEXT:    addvl sp, sp, #-1
; CHECK-NEXT:    mov z3.d, z2.d
; CHECK-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
; CHECK-NEXT:    mov p8.b, p0.b
; CHECK-NEXT:    mov z2.d, z1.d
; CHECK-NEXT:    stnt1d { z2.d, z3.d }, pn8, [x0]
; CHECK-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
; CHECK-NEXT:    addvl sp, sp, #1
; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT:    ret
  call void @llvm.aarch64.sve.stnt1.pn.x2.nxv2f64(<vscale x 2 x double> %zn0, <vscale x 2 x double> %zn1, target("aarch64.svcount") %pn, ptr %ptr);
  ret void
}

define void @stnt1_x4_i8(<vscale x 16 x i8> %unused, <vscale x 16 x i8> %zn0, <vscale x 16 x i8> %zn1, <vscale x 16 x i8> %zn2, <vscale x 16 x i8> %zn3, target("aarch64.svcount") %pn, ptr %ptr) nounwind {
; CHECK-LABEL: stnt1_x4_i8:
; CHECK:       // %bb.0:
; CHECK-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
; CHECK-NEXT:    addvl sp, sp, #-1
; CHECK-NEXT:    mov z7.d, z4.d
; CHECK-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
; CHECK-NEXT:    mov z6.d, z3.d
; CHECK-NEXT:    mov p8.b, p0.b
; CHECK-NEXT:    mov z5.d, z2.d
; CHECK-NEXT:    mov z4.d, z1.d
; CHECK-NEXT:    stnt1b { z4.b - z7.b }, pn8, [x0]
; CHECK-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
; CHECK-NEXT:    addvl sp, sp, #1
; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT:    ret
  call void @llvm.aarch64.sve.stnt1.pn.x4.nxv16i8(<vscale x 16 x i8> %zn0, <vscale x 16 x i8> %zn1, <vscale x 16 x i8> %zn2, <vscale x 16 x i8> %zn3, target("aarch64.svcount") %pn, ptr %ptr);
  ret void
}

define void @stnt1_x4_i16(<vscale x 16 x i8> %unused, <vscale x 8 x i16> %zn0, <vscale x 8 x i16> %zn1, <vscale x 8 x i16> %zn2, <vscale x 8 x i16> %zn3, target("aarch64.svcount") %pn, ptr %ptr) nounwind {
; CHECK-LABEL: stnt1_x4_i16:
; CHECK:       // %bb.0:
; CHECK-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
; CHECK-NEXT:    addvl sp, sp, #-1
; CHECK-NEXT:    mov z7.d, z4.d
; CHECK-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
; CHECK-NEXT:    mov z6.d, z3.d
; CHECK-NEXT:    mov p8.b, p0.b
; CHECK-NEXT:    mov z5.d, z2.d
; CHECK-NEXT:    mov z4.d, z1.d
; CHECK-NEXT:    stnt1h { z4.h - z7.h }, pn8, [x0]
; CHECK-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
; CHECK-NEXT:    addvl sp, sp, #1
; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT:    ret
  call void @llvm.aarch64.sve.stnt1.pn.x4.nxv8i16(<vscale x 8 x i16> %zn0, <vscale x 8 x i16> %zn1, <vscale x 8 x i16> %zn2, <vscale x 8 x i16> %zn3, target("aarch64.svcount") %pn, ptr %ptr);
  ret void
}

define void @stnt1_x4_i32(<vscale x 16 x i8> %unused, <vscale x 4 x i32> %zn0, <vscale x 4 x i32> %zn1, <vscale x 4 x i32> %zn2, <vscale x 4 x i32> %zn3, target("aarch64.svcount") %pn, ptr %ptr) nounwind {
; CHECK-LABEL: stnt1_x4_i32:
; CHECK:       // %bb.0:
; CHECK-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
; CHECK-NEXT:    addvl sp, sp, #-1
; CHECK-NEXT:    mov z7.d, z4.d
; CHECK-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
; CHECK-NEXT:    mov z6.d, z3.d
; CHECK-NEXT:    mov p8.b, p0.b
; CHECK-NEXT:    mov z5.d, z2.d
; CHECK-NEXT:    mov z4.d, z1.d
; CHECK-NEXT:    stnt1w { z4.s - z7.s }, pn8, [x0]
; CHECK-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
; CHECK-NEXT:    addvl sp, sp, #1
; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT:    ret
  call void @llvm.aarch64.sve.stnt1.pn.x4.nxv4i32(<vscale x 4 x i32> %zn0, <vscale x 4 x i32> %zn1, <vscale x 4 x i32> %zn2, <vscale x 4 x i32> %zn3, target("aarch64.svcount") %pn, ptr %ptr);
  ret void
}

define void @stnt1_x4_i64(<vscale x 16 x i8> %unused, <vscale x 2 x i64> %zn0, <vscale x 2 x i64> %zn1, <vscale x 2 x i64> %zn2, <vscale x 2 x i64> %zn3, target("aarch64.svcount") %pn, ptr %ptr) nounwind {
; CHECK-LABEL: stnt1_x4_i64:
; CHECK:       // %bb.0:
; CHECK-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
; CHECK-NEXT:    addvl sp, sp, #-1
; CHECK-NEXT:    mov z7.d, z4.d
; CHECK-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
; CHECK-NEXT:    mov z6.d, z3.d
; CHECK-NEXT:    mov p8.b, p0.b
; CHECK-NEXT:    mov z5.d, z2.d
; CHECK-NEXT:    mov z4.d, z1.d
; CHECK-NEXT:    stnt1d { z4.d - z7.d }, pn8, [x0]
; CHECK-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
; CHECK-NEXT:    addvl sp, sp, #1
; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT:    ret
  call void @llvm.aarch64.sve.stnt1.pn.x4.nxv2i64(<vscale x 2 x i64> %zn0, <vscale x 2 x i64> %zn1, <vscale x 2 x i64> %zn2, <vscale x 2 x i64> %zn3, target("aarch64.svcount") %pn, ptr %ptr);
  ret void
}

define void @stnt1_x4_f16(<vscale x 16 x i8> %unused, <vscale x 8 x half> %zn0, <vscale x 8 x half> %zn1, <vscale x 8 x half> %zn2, <vscale x 8 x half> %zn3, target("aarch64.svcount") %pn, ptr %ptr) nounwind {
; CHECK-LABEL: stnt1_x4_f16:
; CHECK:       // %bb.0:
; CHECK-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
; CHECK-NEXT:    addvl sp, sp, #-1
; CHECK-NEXT:    mov z7.d, z4.d
; CHECK-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
; CHECK-NEXT:    mov z6.d, z3.d
; CHECK-NEXT:    mov p8.b, p0.b
; CHECK-NEXT:    mov z5.d, z2.d
; CHECK-NEXT:    mov z4.d, z1.d
; CHECK-NEXT:    stnt1h { z4.h - z7.h }, pn8, [x0]
; CHECK-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
; CHECK-NEXT:    addvl sp, sp, #1
; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT:    ret
  call void @llvm.aarch64.sve.stnt1.pn.x4.nxv8f16(<vscale x 8 x half> %zn0, <vscale x 8 x half> %zn1, <vscale x 8 x half> %zn2, <vscale x 8 x half> %zn3, target("aarch64.svcount") %pn, ptr %ptr);
  ret void
}

define void @stnt1_x4_bf16(<vscale x 16 x i8> %unused, <vscale x 8 x bfloat> %zn0, <vscale x 8 x bfloat> %zn1, <vscale x 8 x bfloat> %zn2, <vscale x 8 x bfloat> %zn3, target("aarch64.svcount") %pn, ptr %ptr) nounwind {
; CHECK-LABEL: stnt1_x4_bf16:
; CHECK:       // %bb.0:
; CHECK-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
; CHECK-NEXT:    addvl sp, sp, #-1
; CHECK-NEXT:    mov z7.d, z4.d
; CHECK-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
; CHECK-NEXT:    mov z6.d, z3.d
; CHECK-NEXT:    mov p8.b, p0.b
; CHECK-NEXT:    mov z5.d, z2.d
; CHECK-NEXT:    mov z4.d, z1.d
; CHECK-NEXT:    stnt1h { z4.h - z7.h }, pn8, [x0]
; CHECK-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
; CHECK-NEXT:    addvl sp, sp, #1
; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT:    ret
  call void @llvm.aarch64.sve.stnt1.pn.x4.nxv8bf16(<vscale x 8 x bfloat> %zn0, <vscale x 8 x bfloat> %zn1, <vscale x 8 x bfloat> %zn2, <vscale x 8 x bfloat> %zn3, target("aarch64.svcount") %pn, ptr %ptr);
  ret void
}

define void @stnt1_x4_f32(<vscale x 16 x i8> %unused, <vscale x 4 x float> %zn0, <vscale x 4 x float> %zn1, <vscale x 4 x float> %zn2, <vscale x 4 x float> %zn3, target("aarch64.svcount") %pn, ptr %ptr) nounwind {
; CHECK-LABEL: stnt1_x4_f32:
; CHECK:       // %bb.0:
; CHECK-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
; CHECK-NEXT:    addvl sp, sp, #-1
; CHECK-NEXT:    mov z7.d, z4.d
; CHECK-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
; CHECK-NEXT:    mov z6.d, z3.d
; CHECK-NEXT:    mov p8.b, p0.b
; CHECK-NEXT:    mov z5.d, z2.d
; CHECK-NEXT:    mov z4.d, z1.d
; CHECK-NEXT:    stnt1w { z4.s - z7.s }, pn8, [x0]
; CHECK-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
; CHECK-NEXT:    addvl sp, sp, #1
; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT:    ret
  call void @llvm.aarch64.sve.stnt1.pn.x4.nxv4f32(<vscale x 4 x float> %zn0, <vscale x 4 x float> %zn1, <vscale x 4 x float> %zn2, <vscale x 4 x float> %zn3, target("aarch64.svcount") %pn, ptr %ptr);
  ret void
}

define void @stnt1_x4_f64(<vscale x 16 x i8> %unused, <vscale x 2 x double> %zn0, <vscale x 2 x double> %zn1, <vscale x 2 x double> %zn2, <vscale x 2 x double> %zn3, target("aarch64.svcount") %pn, ptr %ptr) nounwind {
; CHECK-LABEL: stnt1_x4_f64:
; CHECK:       // %bb.0:
; CHECK-NEXT:    str x29, [sp, #-16]! // 8-byte Folded Spill
; CHECK-NEXT:    addvl sp, sp, #-1
; CHECK-NEXT:    mov z7.d, z4.d
; CHECK-NEXT:    str p8, [sp, #7, mul vl] // 2-byte Folded Spill
; CHECK-NEXT:    mov z6.d, z3.d
; CHECK-NEXT:    mov p8.b, p0.b
; CHECK-NEXT:    mov z5.d, z2.d
; CHECK-NEXT:    mov z4.d, z1.d
; CHECK-NEXT:    stnt1d { z4.d - z7.d }, pn8, [x0]
; CHECK-NEXT:    ldr p8, [sp, #7, mul vl] // 2-byte Folded Reload
; CHECK-NEXT:    addvl sp, sp, #1
; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT:    ret
  call void @llvm.aarch64.sve.stnt1.pn.x4.nxv2f64(<vscale x 2 x double> %zn0, <vscale x 2 x double> %zn1, <vscale x 2 x double> %zn2, <vscale x 2 x double> %zn3, target("aarch64.svcount") %pn, ptr %ptr);
  ret void
}

declare void @llvm.aarch64.sve.st1.pn.x2.nxv16i8(<vscale x 16 x i8>, <vscale x 16 x i8>, target("aarch64.svcount"), ptr)
declare void @llvm.aarch64.sve.st1.pn.x2.nxv8i16(<vscale x 8 x i16>, <vscale x 8 x i16>, target("aarch64.svcount"), ptr)
declare void @llvm.aarch64.sve.st1.pn.x2.nxv4i32(<vscale x 4 x i32>, <vscale x 4 x i32>, target("aarch64.svcount"), ptr)
declare void @llvm.aarch64.sve.st1.pn.x2.nxv2i64(<vscale x 2 x i64>, <vscale x 2 x i64>, target("aarch64.svcount"), ptr)
declare void @llvm.aarch64.sve.st1.pn.x2.nxv8f16(<vscale x 8 x half>, <vscale x 8 x half>, target("aarch64.svcount"), ptr)
declare void @llvm.aarch64.sve.st1.pn.x2.nxv8bf16(<vscale x 8 x bfloat>, <vscale x 8 x bfloat>, target("aarch64.svcount"), ptr)
declare void @llvm.aarch64.sve.st1.pn.x2.nxv4f32(<vscale x 4 x float>, <vscale x 4 x float>, target("aarch64.svcount"), ptr)
declare void @llvm.aarch64.sve.st1.pn.x2.nxv2f64(<vscale x 2 x double>, <vscale x 2 x double>, target("aarch64.svcount"), ptr)
declare void @llvm.aarch64.sve.stnt1.pn.x2.nxv16i8(<vscale x 16 x i8>, <vscale x 16 x i8>, target("aarch64.svcount"), ptr)
declare void @llvm.aarch64.sve.stnt1.pn.x2.nxv8i16(<vscale x 8 x i16>, <vscale x 8 x i16>, target("aarch64.svcount"), ptr)
declare void @llvm.aarch64.sve.stnt1.pn.x2.nxv4i32(<vscale x 4 x i32>, <vscale x 4 x i32>, target("aarch64.svcount"), ptr)
declare void @llvm.aarch64.sve.stnt1.pn.x2.nxv2i64(<vscale x 2 x i64>, <vscale x 2 x i64>, target("aarch64.svcount"), ptr)
declare void @llvm.aarch64.sve.stnt1.pn.x2.nxv8f16(<vscale x 8 x half>, <vscale x 8 x half>, target("aarch64.svcount"), ptr)
declare void @llvm.aarch64.sve.stnt1.pn.x2.nxv8bf16(<vscale x 8 x bfloat>, <vscale x 8 x bfloat>, target("aarch64.svcount"), ptr)
declare void @llvm.aarch64.sve.stnt1.pn.x2.nxv4f32(<vscale x 4 x float>, <vscale x 4 x float>, target("aarch64.svcount"), ptr)
declare void @llvm.aarch64.sve.stnt1.pn.x2.nxv2f64(<vscale x 2 x double>, <vscale x 2 x double>, target("aarch64.svcount"), ptr)


declare void @llvm.aarch64.sve.st1.pn.x4.nxv16i8(<vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, target("aarch64.svcount"), ptr)
declare void @llvm.aarch64.sve.st1.pn.x4.nxv8i16(<vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, target("aarch64.svcount"), ptr)
declare void @llvm.aarch64.sve.st1.pn.x4.nxv4i32(<vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, target("aarch64.svcount"), ptr)
declare void @llvm.aarch64.sve.st1.pn.x4.nxv2i64(<vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, target("aarch64.svcount"), ptr)
declare void @llvm.aarch64.sve.st1.pn.x4.nxv8f16(<vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half>, target("aarch64.svcount"), ptr)
declare void @llvm.aarch64.sve.st1.pn.x4.nxv8bf16(<vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, target("aarch64.svcount"), ptr)
declare void @llvm.aarch64.sve.st1.pn.x4.nxv4f32(<vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float>, target("aarch64.svcount"), ptr)
declare void @llvm.aarch64.sve.st1.pn.x4.nxv2f64(<vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double>, target("aarch64.svcount"), ptr)
declare void @llvm.aarch64.sve.stnt1.pn.x4.nxv16i8(<vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, target("aarch64.svcount"), ptr)
declare void @llvm.aarch64.sve.stnt1.pn.x4.nxv8i16(<vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, target("aarch64.svcount"), ptr)
declare void @llvm.aarch64.sve.stnt1.pn.x4.nxv4i32(<vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, target("aarch64.svcount"), ptr)
declare void @llvm.aarch64.sve.stnt1.pn.x4.nxv2i64(<vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, target("aarch64.svcount"), ptr)
declare void @llvm.aarch64.sve.stnt1.pn.x4.nxv8f16(<vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half>, target("aarch64.svcount"), ptr)
declare void @llvm.aarch64.sve.stnt1.pn.x4.nxv8bf16(<vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, target("aarch64.svcount"), ptr)
declare void @llvm.aarch64.sve.stnt1.pn.x4.nxv4f32(<vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float>, target("aarch64.svcount"), ptr)
declare void @llvm.aarch64.sve.stnt1.pn.x4.nxv2f64(<vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double>, target("aarch64.svcount"), ptr)