llvm/test/CodeGen/RISCV/rvv/vector-interleave-store.ll

; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc < %s -mtriple=riscv32 -mattr=+v,+zvfh,+zvfbfmin | FileCheck --check-prefixes=CHECK,RV32 %s
; RUN: llc < %s -mtriple=riscv64 -mattr=+v,+zvfh,+zvfbfmin | FileCheck --check-prefixes=CHECK,RV64 %s
; RUN: llc < %s -mtriple=riscv32 -mattr=+v,+zvfhmin,+zvfbfmin | FileCheck --check-prefixes=CHECK,RV32 %s
; RUN: llc < %s -mtriple=riscv64 -mattr=+v,+zvfhmin,+zvfbfmin | FileCheck --check-prefixes=CHECK,RV64 %s

; Integers

define void @vector_interleave_store_nxv32i1_nxv16i1(<vscale x 16 x i1> %a, <vscale x 16 x i1> %b, ptr %p) {
; CHECK-LABEL: vector_interleave_store_nxv32i1_nxv16i1:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli a1, zero, e8, m2, ta, ma
; CHECK-NEXT:    vmv1r.v v9, v0
; CHECK-NEXT:    vmv1r.v v0, v8
; CHECK-NEXT:    vmv.v.i v10, 0
; CHECK-NEXT:    li a1, -1
; CHECK-NEXT:    csrr a2, vlenb
; CHECK-NEXT:    vmerge.vim v12, v10, 1, v0
; CHECK-NEXT:    vmv1r.v v0, v9
; CHECK-NEXT:    vmerge.vim v8, v10, 1, v0
; CHECK-NEXT:    srli a2, a2, 2
; CHECK-NEXT:    vwaddu.vv v16, v8, v12
; CHECK-NEXT:    vwmaccu.vx v16, a1, v12
; CHECK-NEXT:    vmsne.vi v8, v18, 0
; CHECK-NEXT:    vmsne.vi v9, v16, 0
; CHECK-NEXT:    add a1, a2, a2
; CHECK-NEXT:    vsetvli zero, a1, e8, mf2, ta, ma
; CHECK-NEXT:    vslideup.vx v9, v8, a2
; CHECK-NEXT:    vsetvli a1, zero, e8, m4, ta, ma
; CHECK-NEXT:    vsm.v v9, (a0)
; CHECK-NEXT:    ret
  %res = call <vscale x 32 x i1> @llvm.vector.interleave2.nxv32i1(<vscale x 16 x i1> %a, <vscale x 16 x i1> %b)
  store <vscale x 32 x i1> %res, ptr %p
  ret void
}

; Shouldn't be lowered to vsseg because it's unaligned
define void @vector_interleave_store_nxv16i16_nxv8i16_align1(<vscale x 8 x i16> %a, <vscale x 8 x i16> %b, ptr %p) {
; CHECK-LABEL: vector_interleave_store_nxv16i16_nxv8i16_align1:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli a1, zero, e16, m2, ta, ma
; CHECK-NEXT:    vwaddu.vv v12, v8, v10
; CHECK-NEXT:    li a1, -1
; CHECK-NEXT:    vwmaccu.vx v12, a1, v10
; CHECK-NEXT:    vs4r.v v12, (a0)
; CHECK-NEXT:    ret
  %res = call <vscale x 16 x i16> @llvm.vector.interleave2.nxv16i16(<vscale x 8 x i16> %a, <vscale x 8 x i16> %b)
  store <vscale x 16 x i16> %res, ptr %p, align 1
  ret void
}

define void @vector_interleave_store_nxv16i16_nxv8i16(<vscale x 8 x i16> %a, <vscale x 8 x i16> %b, ptr %p) {
; CHECK-LABEL: vector_interleave_store_nxv16i16_nxv8i16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli a1, zero, e16, m2, ta, ma
; CHECK-NEXT:    vsseg2e16.v v8, (a0)
; CHECK-NEXT:    ret
  %res = call <vscale x 16 x i16> @llvm.vector.interleave2.nxv16i16(<vscale x 8 x i16> %a, <vscale x 8 x i16> %b)
  store <vscale x 16 x i16> %res, ptr %p
  ret void
}

define void @vector_interleave_store_nxv8i32_nxv4i32(<vscale x 4 x i32> %a, <vscale x 4 x i32> %b, ptr %p) {
; CHECK-LABEL: vector_interleave_store_nxv8i32_nxv4i32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli a1, zero, e32, m2, ta, ma
; CHECK-NEXT:    vsseg2e32.v v8, (a0)
; CHECK-NEXT:    ret
  %res = call <vscale x 8 x i32> @llvm.vector.interleave2.nxv8i32(<vscale x 4 x i32> %a, <vscale x 4 x i32> %b)
  store <vscale x 8 x i32> %res, ptr %p
  ret void
}

define void @vector_interleave_store_nxv4i64_nxv2i64(<vscale x 2 x i64> %a, <vscale x 2 x i64> %b, ptr %p) {
; CHECK-LABEL: vector_interleave_store_nxv4i64_nxv2i64:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli a1, zero, e64, m2, ta, ma
; CHECK-NEXT:    vsseg2e64.v v8, (a0)
; CHECK-NEXT:    ret
  %res = call <vscale x 4 x i64> @llvm.vector.interleave2.nxv4i64(<vscale x 2 x i64> %a, <vscale x 2 x i64> %b)
  store <vscale x 4 x i64> %res, ptr %p
  ret void
}

define void @vector_interleave_store_nxv8i64_nxv4i64(<vscale x 4 x i64> %a, <vscale x 4 x i64> %b, ptr %p) {
; CHECK-LABEL: vector_interleave_store_nxv8i64_nxv4i64:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli a1, zero, e64, m4, ta, ma
; CHECK-NEXT:    vsseg2e64.v v8, (a0)
; CHECK-NEXT:    ret
  %res = call <vscale x 8 x i64> @llvm.vector.interleave2.nxv8i64(<vscale x 4 x i64> %a, <vscale x 4 x i64> %b)
  store <vscale x 8 x i64> %res, ptr %p
  ret void
}

; This shouldn't be lowered to a vsseg because EMUL * NFIELDS >= 8
define void @vector_interleave_store_nxv16i64_nxv8i64(<vscale x 8 x i64> %a, <vscale x 8 x i64> %b, ptr %p) {
; CHECK-LABEL: vector_interleave_store_nxv16i64_nxv8i64:
; CHECK:       # %bb.0:
; CHECK-NEXT:    csrr a1, vlenb
; CHECK-NEXT:    vsetvli a2, zero, e16, m2, ta, mu
; CHECK-NEXT:    vid.v v6
; CHECK-NEXT:    vmv8r.v v24, v8
; CHECK-NEXT:    srli a2, a1, 1
; CHECK-NEXT:    vmv4r.v v28, v16
; CHECK-NEXT:    vmv4r.v v16, v12
; CHECK-NEXT:    vsrl.vi v8, v6, 1
; CHECK-NEXT:    vand.vi v10, v6, 1
; CHECK-NEXT:    slli a1, a1, 3
; CHECK-NEXT:    vmsne.vi v0, v10, 0
; CHECK-NEXT:    add a1, a0, a1
; CHECK-NEXT:    vadd.vx v8, v8, a2, v0.t
; CHECK-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
; CHECK-NEXT:    vrgatherei16.vv v0, v24, v8
; CHECK-NEXT:    vrgatherei16.vv v24, v16, v8
; CHECK-NEXT:    vs8r.v v24, (a1)
; CHECK-NEXT:    vs8r.v v0, (a0)
; CHECK-NEXT:    ret
  %res = call <vscale x 16 x i64> @llvm.vector.interleave2.nxv16i64(<vscale x 8 x i64> %a, <vscale x 8 x i64> %b)
  store <vscale x 16 x i64> %res, ptr %p
  ret void
}

; Floats

define void @vector_interleave_store_nxv4bf16_nxv2bf16(<vscale x 2 x bfloat> %a, <vscale x 2 x bfloat> %b, ptr %p) {
; CHECK-LABEL: vector_interleave_store_nxv4bf16_nxv2bf16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli a1, zero, e16, mf2, ta, ma
; CHECK-NEXT:    vsseg2e16.v v8, (a0)
; CHECK-NEXT:    ret
  %res = call <vscale x 4 x bfloat> @llvm.vector.interleave2.nxv4bf16(<vscale x 2 x bfloat> %a, <vscale x 2 x bfloat> %b)
  store <vscale x 4 x bfloat> %res, ptr %p
  ret void
}

define void @vector_interleave_store_nxv8bf16_nxv4bf16(<vscale x 4 x bfloat> %a, <vscale x 4 x bfloat> %b, ptr %p) {
; CHECK-LABEL: vector_interleave_store_nxv8bf16_nxv4bf16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli a1, zero, e16, m1, ta, ma
; CHECK-NEXT:    vsseg2e16.v v8, (a0)
; CHECK-NEXT:    ret
  %res = call <vscale x 8 x bfloat> @llvm.vector.interleave2.nxv8bf16(<vscale x 4 x bfloat> %a, <vscale x 4 x bfloat> %b)
  store <vscale x 8 x bfloat> %res, ptr %p
  ret void
}

define void @vector_interleave_store_nxv4f16_nxv2f16(<vscale x 2 x half> %a, <vscale x 2 x half> %b, ptr %p) {
; CHECK-LABEL: vector_interleave_store_nxv4f16_nxv2f16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli a1, zero, e16, mf2, ta, ma
; CHECK-NEXT:    vsseg2e16.v v8, (a0)
; CHECK-NEXT:    ret
  %res = call <vscale x 4 x half> @llvm.vector.interleave2.nxv4f16(<vscale x 2 x half> %a, <vscale x 2 x half> %b)
  store <vscale x 4 x half> %res, ptr %p
  ret void
}

define void @vector_interleave_store_nxv8f16_nxv4f16(<vscale x 4 x half> %a, <vscale x 4 x half> %b, ptr %p) {
; CHECK-LABEL: vector_interleave_store_nxv8f16_nxv4f16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli a1, zero, e16, m1, ta, ma
; CHECK-NEXT:    vsseg2e16.v v8, (a0)
; CHECK-NEXT:    ret
  %res = call <vscale x 8 x half> @llvm.vector.interleave2.nxv8f16(<vscale x 4 x half> %a, <vscale x 4 x half> %b)
  store <vscale x 8 x half> %res, ptr %p
  ret void
}

define void @vector_interleave_store_nxv4f32_nxv2f32(<vscale x 2 x float> %a, <vscale x 2 x float> %b, ptr %p) {
; CHECK-LABEL: vector_interleave_store_nxv4f32_nxv2f32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli a1, zero, e32, m1, ta, ma
; CHECK-NEXT:    vsseg2e32.v v8, (a0)
; CHECK-NEXT:    ret
  %res = call <vscale x 4 x float> @llvm.vector.interleave2.nxv4f32(<vscale x 2 x float> %a, <vscale x 2 x float> %b)
  store <vscale x 4 x float> %res, ptr %p
  ret void
}

define void @vector_interleave_store_nxv16bf16_nxv8bf16(<vscale x 8 x bfloat> %a, <vscale x 8 x bfloat> %b, ptr %p) {
; CHECK-LABEL: vector_interleave_store_nxv16bf16_nxv8bf16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli a1, zero, e16, m2, ta, ma
; CHECK-NEXT:    vsseg2e16.v v8, (a0)
; CHECK-NEXT:    ret
  %res = call <vscale x 16 x bfloat> @llvm.vector.interleave2.nxv16bf16(<vscale x 8 x bfloat> %a, <vscale x 8 x bfloat> %b)
  store <vscale x 16 x bfloat> %res, ptr %p
  ret void
}

define void @vector_interleave_store_nxv16f16_nxv8f16(<vscale x 8 x half> %a, <vscale x 8 x half> %b, ptr %p) {
; CHECK-LABEL: vector_interleave_store_nxv16f16_nxv8f16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli a1, zero, e16, m2, ta, ma
; CHECK-NEXT:    vsseg2e16.v v8, (a0)
; CHECK-NEXT:    ret
  %res = call <vscale x 16 x half> @llvm.vector.interleave2.nxv16f16(<vscale x 8 x half> %a, <vscale x 8 x half> %b)
  store <vscale x 16 x half> %res, ptr %p
  ret void
}

define void @vector_interleave_store_nxv8f32_nxv4f32(<vscale x 4 x float> %a, <vscale x 4 x float> %b, ptr %p) {
; CHECK-LABEL: vector_interleave_store_nxv8f32_nxv4f32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli a1, zero, e32, m2, ta, ma
; CHECK-NEXT:    vsseg2e32.v v8, (a0)
; CHECK-NEXT:    ret
  %res = call <vscale x 8 x float> @llvm.vector.interleave2.nxv8f32(<vscale x 4 x float> %a, <vscale x 4 x float> %b)
  store <vscale x 8 x float> %res, ptr %p
  ret void
}

define void @vector_interleave_store_nxv4f64_nxv2f64(<vscale x 2 x double> %a, <vscale x 2 x double> %b, ptr %p) {
; CHECK-LABEL: vector_interleave_store_nxv4f64_nxv2f64:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli a1, zero, e64, m2, ta, ma
; CHECK-NEXT:    vsseg2e64.v v8, (a0)
; CHECK-NEXT:    ret
  %res = call <vscale x 4 x double> @llvm.vector.interleave2.nxv4f64(<vscale x 2 x double> %a, <vscale x 2 x double> %b)
  store <vscale x 4 x double> %res, ptr %p
  ret void
}

define void @vector_interleave_store_nxv4p0_nxv2p0(<vscale x 2 x ptr> %a, <vscale x 2 x ptr> %b, ptr %p) {
; RV32-LABEL: vector_interleave_store_nxv4p0_nxv2p0:
; RV32:       # %bb.0:
; RV32-NEXT:    vsetvli a1, zero, e32, m1, ta, ma
; RV32-NEXT:    vsseg2e32.v v8, (a0)
; RV32-NEXT:    ret
;
; RV64-LABEL: vector_interleave_store_nxv4p0_nxv2p0:
; RV64:       # %bb.0:
; RV64-NEXT:    vsetvli a1, zero, e64, m2, ta, ma
; RV64-NEXT:    vsseg2e64.v v8, (a0)
; RV64-NEXT:    ret
  %res = call <vscale x 4 x ptr> @llvm.vector.interleave2.nxv4p0(<vscale x 2 x ptr> %a, <vscale x 2 x ptr> %b)
  store <vscale x 4 x ptr> %res, ptr %p
  ret void
}

define void @vector_interleave_store_factor4(<vscale x 4 x i32> %a, <vscale x 4 x i32> %b, <vscale x 4 x i32> %c, <vscale x 4 x i32> %d, ptr %p) {
; CHECK-LABEL: vector_interleave_store_factor4:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli a1, zero, e32, m2, ta, ma
; CHECK-NEXT:    vsseg4e32.v v8, (a0)
; CHECK-NEXT:    ret
  %v0 = call <vscale x 8 x i32> @llvm.vector.interleave2.nxv8i32(<vscale x 4 x i32> %a, <vscale x 4 x i32> %c)
  %v1 = call <vscale x 8 x i32> @llvm.vector.interleave2.nxv8i32(<vscale x 4 x i32> %b, <vscale x 4 x i32> %d)
  %v2 = call <vscale x 16 x i32> @llvm.vector.interleave2.nxv16i32(<vscale x 8 x i32> %v0, <vscale x 8 x i32> %v1)
  store <vscale x 16 x i32> %v2, ptr %p
  ret void
}

define void @vector_interleave_store_factor8(<vscale x 2 x i32> %a, <vscale x 2 x i32> %b, <vscale x 2 x i32> %c, <vscale x 2 x i32> %d, <vscale x 2 x i32> %e, <vscale x 2 x i32> %f, <vscale x 2 x i32> %g, <vscale x 2 x i32> %h, ptr %p) {
; CHECK-LABEL: vector_interleave_store_factor8:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli a1, zero, e32, m1, ta, ma
; CHECK-NEXT:    vsseg8e32.v v8, (a0)
; CHECK-NEXT:    ret
  %v0 = call <vscale x 4 x i32> @llvm.vector.interleave2.nxv4i32(<vscale x 2 x i32> %a, <vscale x 2 x i32> %e)
  %v1 = call <vscale x 4 x i32> @llvm.vector.interleave2.nxv4i32(<vscale x 2 x i32> %c, <vscale x 2 x i32> %g)
  %v2 = call <vscale x 8 x i32> @llvm.vector.interleave2.nxv8i32(<vscale x 4 x i32> %v0, <vscale x 4 x i32> %v1)

  %v3 = call <vscale x 4 x i32> @llvm.vector.interleave2.nxv4i32(<vscale x 2 x i32> %b, <vscale x 2 x i32> %f)
  %v4 = call <vscale x 4 x i32> @llvm.vector.interleave2.nxv4i32(<vscale x 2 x i32> %d, <vscale x 2 x i32> %h)
  %v5 = call <vscale x 8 x i32> @llvm.vector.interleave2.nxv8i32(<vscale x 4 x i32> %v3, <vscale x 4 x i32> %v4)

  %v6 = call <vscale x 16 x i32> @llvm.vector.interleave2.nxv16i32(<vscale x 8 x i32> %v2, <vscale x 8 x i32> %v5)
  store <vscale x 16 x i32> %v6, ptr %p
  ret void
}