llvm/test/CodeGen/AArch64/sve-fcopysign.ll

; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc -mattr=+sve,+bf16 < %s -o - | FileCheck %s

target triple = "aarch64-unknown-linux-gnu"

;
; llvm.copysign.nxv2f16
;

define <vscale x 2 x half> @copysign_nxv2f16_nxv2f16(<vscale x 2 x half> %a, <vscale x 2 x half> %b) {
; CHECK-LABEL: copysign_nxv2f16_nxv2f16:
; CHECK:       // %bb.0:
; CHECK-NEXT:    and z1.h, z1.h, #0x8000
; CHECK-NEXT:    and z0.h, z0.h, #0x7fff
; CHECK-NEXT:    orr z0.d, z0.d, z1.d
; CHECK-NEXT:    ret
  %r = call <vscale x 2 x half> @llvm.copysign.nxv2f16(<vscale x 2 x half> %a, <vscale x 2 x half> %b)
  ret <vscale x 2 x half> %r
}

define <vscale x 2 x half> @copysign_nxv2f16_nxv2f32(<vscale x 2 x half> %a, <vscale x 2 x float> %b) {
; CHECK-LABEL: copysign_nxv2f16_nxv2f32:
; CHECK:       // %bb.0:
; CHECK-NEXT:    ptrue p0.d
; CHECK-NEXT:    and z0.h, z0.h, #0x7fff
; CHECK-NEXT:    fcvt z1.h, p0/m, z1.s
; CHECK-NEXT:    and z1.h, z1.h, #0x8000
; CHECK-NEXT:    orr z0.d, z0.d, z1.d
; CHECK-NEXT:    ret
  %b.trunc = fptrunc <vscale x 2 x float> %b to <vscale x 2 x half>
  %r = call <vscale x 2 x half> @llvm.copysign.nxv2f16(<vscale x 2 x half> %a, <vscale x 2 x half> %b.trunc)
  ret <vscale x 2 x half> %r
}

define <vscale x 2 x half> @copysign_nxv2f16_nxv2f64(<vscale x 2 x half> %a, <vscale x 2 x double> %b) {
; CHECK-LABEL: copysign_nxv2f16_nxv2f64:
; CHECK:       // %bb.0:
; CHECK-NEXT:    ptrue p0.d
; CHECK-NEXT:    and z0.h, z0.h, #0x7fff
; CHECK-NEXT:    fcvt z1.h, p0/m, z1.d
; CHECK-NEXT:    and z1.h, z1.h, #0x8000
; CHECK-NEXT:    orr z0.d, z0.d, z1.d
; CHECK-NEXT:    ret
  %b.trunc = fptrunc <vscale x 2 x double> %b to <vscale x 2 x half>
  %r = call <vscale x 2 x half> @llvm.copysign.nxv2f16(<vscale x 2 x half> %a, <vscale x 2 x half> %b.trunc)
  ret <vscale x 2 x half> %r
}

;
; llvm.copysign.nxv4f16
;

define <vscale x 4 x half> @copysign_nxv4f16_nxv4f16(<vscale x 4 x half> %a, <vscale x 4 x half> %b) {
; CHECK-LABEL: copysign_nxv4f16_nxv4f16:
; CHECK:       // %bb.0:
; CHECK-NEXT:    and z1.h, z1.h, #0x8000
; CHECK-NEXT:    and z0.h, z0.h, #0x7fff
; CHECK-NEXT:    orr z0.d, z0.d, z1.d
; CHECK-NEXT:    ret
  %r = call <vscale x 4 x half> @llvm.copysign.nxv4f16(<vscale x 4 x half> %a, <vscale x 4 x half> %b)
  ret <vscale x 4 x half> %r
}

define <vscale x 4 x half> @copysign_nxv4f16_nxv4f32(<vscale x 4 x half> %a, <vscale x 4 x float> %b) {
; CHECK-LABEL: copysign_nxv4f16_nxv4f32:
; CHECK:       // %bb.0:
; CHECK-NEXT:    ptrue p0.s
; CHECK-NEXT:    and z0.h, z0.h, #0x7fff
; CHECK-NEXT:    fcvt z1.h, p0/m, z1.s
; CHECK-NEXT:    and z1.h, z1.h, #0x8000
; CHECK-NEXT:    orr z0.d, z0.d, z1.d
; CHECK-NEXT:    ret
  %b.trunc = fptrunc <vscale x 4 x float> %b to <vscale x 4 x half>
  %r = call <vscale x 4 x half> @llvm.copysign.nxv4f16(<vscale x 4 x half> %a, <vscale x 4 x half> %b.trunc)
  ret <vscale x 4 x half> %r
}

define <vscale x 4 x half> @copysign_nxv4f16_nxv4f64(<vscale x 4 x half> %a, <vscale x 4 x double> %b) {
; CHECK-LABEL: copysign_nxv4f16_nxv4f64:
; CHECK:       // %bb.0:
; CHECK-NEXT:    ptrue p0.d
; CHECK-NEXT:    and z0.h, z0.h, #0x7fff
; CHECK-NEXT:    fcvt z2.h, p0/m, z2.d
; CHECK-NEXT:    fcvt z1.h, p0/m, z1.d
; CHECK-NEXT:    uzp1 z1.s, z1.s, z2.s
; CHECK-NEXT:    and z1.h, z1.h, #0x8000
; CHECK-NEXT:    orr z0.d, z0.d, z1.d
; CHECK-NEXT:    ret
  %b.trunc = fptrunc <vscale x 4 x double> %b to <vscale x 4 x half>
  %r = call <vscale x 4 x half> @llvm.copysign.nxv4f16(<vscale x 4 x half> %a, <vscale x 4 x half> %b.trunc)
  ret <vscale x 4 x half> %r
}

;
; llvm.copysign.nxv8f16
;

define <vscale x 8 x half> @copysign_nxv8f16_nxv8f16(<vscale x 8 x half> %a, <vscale x 8 x half> %b) {
; CHECK-LABEL: copysign_nxv8f16_nxv8f16:
; CHECK:       // %bb.0:
; CHECK-NEXT:    and z1.h, z1.h, #0x8000
; CHECK-NEXT:    and z0.h, z0.h, #0x7fff
; CHECK-NEXT:    orr z0.d, z0.d, z1.d
; CHECK-NEXT:    ret
  %r = call <vscale x 8 x half> @llvm.copysign.nxv8f16(<vscale x 8 x half> %a, <vscale x 8 x half> %b)
  ret <vscale x 8 x half> %r
}

define <vscale x 8 x half> @copysign_nxv8f16_nxv8f32(<vscale x 8 x half> %a, <vscale x 8 x float> %b) {
; CHECK-LABEL: copysign_nxv8f16_nxv8f32:
; CHECK:       // %bb.0:
; CHECK-NEXT:    ptrue p0.s
; CHECK-NEXT:    and z0.h, z0.h, #0x7fff
; CHECK-NEXT:    fcvt z2.h, p0/m, z2.s
; CHECK-NEXT:    fcvt z1.h, p0/m, z1.s
; CHECK-NEXT:    uzp1 z1.h, z1.h, z2.h
; CHECK-NEXT:    and z1.h, z1.h, #0x8000
; CHECK-NEXT:    orr z0.d, z0.d, z1.d
; CHECK-NEXT:    ret
  %b.trunc = fptrunc <vscale x 8 x float> %b to <vscale x 8 x half>
  %r = call <vscale x 8 x half> @llvm.copysign.nxv8f16(<vscale x 8 x half> %a, <vscale x 8 x half> %b.trunc)
  ret <vscale x 8 x half> %r
}

define <vscale x 8 x half> @copysign_nxv8f16_nxv8f64(<vscale x 8 x half> %a, <vscale x 8 x double> %b) {
; CHECK-LABEL: copysign_nxv8f16_nxv8f64:
; CHECK:       // %bb.0:
; CHECK-NEXT:    ptrue p0.d
; CHECK-NEXT:    and z0.h, z0.h, #0x7fff
; CHECK-NEXT:    fcvt z4.h, p0/m, z4.d
; CHECK-NEXT:    fcvt z3.h, p0/m, z3.d
; CHECK-NEXT:    fcvt z2.h, p0/m, z2.d
; CHECK-NEXT:    fcvt z1.h, p0/m, z1.d
; CHECK-NEXT:    uzp1 z3.s, z3.s, z4.s
; CHECK-NEXT:    uzp1 z1.s, z1.s, z2.s
; CHECK-NEXT:    uzp1 z1.h, z1.h, z3.h
; CHECK-NEXT:    and z1.h, z1.h, #0x8000
; CHECK-NEXT:    orr z0.d, z0.d, z1.d
; CHECK-NEXT:    ret
  %b.trunc = fptrunc <vscale x 8 x double> %b to <vscale x 8 x half>
  %r = call <vscale x 8 x half> @llvm.copysign.nxv8f16(<vscale x 8 x half> %a, <vscale x 8 x half> %b.trunc)
  ret <vscale x 8 x half> %r
}

;
; llvm.copysign.nxv2f32
;

define <vscale x 2 x float> @copysign_nxv2f32_nxv2f16(<vscale x 2 x float> %a, <vscale x 2 x half> %b) {
; CHECK-LABEL: copysign_nxv2f32_nxv2f16:
; CHECK:       // %bb.0:
; CHECK-NEXT:    ptrue p0.d
; CHECK-NEXT:    and z0.s, z0.s, #0x7fffffff
; CHECK-NEXT:    fcvt z1.s, p0/m, z1.h
; CHECK-NEXT:    and z1.s, z1.s, #0x80000000
; CHECK-NEXT:    orr z0.d, z0.d, z1.d
; CHECK-NEXT:    ret
  %b.ext = fpext <vscale x 2 x half> %b to <vscale x 2 x float>
  %r = call <vscale x 2 x float> @llvm.copysign.nxv2f32(<vscale x 2 x float> %a, <vscale x 2 x float> %b.ext)
  ret <vscale x 2 x float> %r
}

define <vscale x 2 x float> @copysign_nxv2f32_nxv2f32(<vscale x 2 x float> %a, <vscale x 2 x float> %b) {
; CHECK-LABEL: copysign_nxv2f32_nxv2f32:
; CHECK:       // %bb.0:
; CHECK-NEXT:    and z1.s, z1.s, #0x80000000
; CHECK-NEXT:    and z0.s, z0.s, #0x7fffffff
; CHECK-NEXT:    orr z0.d, z0.d, z1.d
; CHECK-NEXT:    ret
  %r = call <vscale x 2 x float> @llvm.copysign.nxv2f32(<vscale x 2 x float> %a, <vscale x 2 x float> %b)
  ret <vscale x 2 x float> %r
}

define <vscale x 2 x float> @copysign_nxv2f32_nxv2f64(<vscale x 2 x float> %a, <vscale x 2 x double> %b) {
; CHECK-LABEL: copysign_nxv2f32_nxv2f64:
; CHECK:       // %bb.0:
; CHECK-NEXT:    ptrue p0.d
; CHECK-NEXT:    and z0.s, z0.s, #0x7fffffff
; CHECK-NEXT:    fcvt z1.s, p0/m, z1.d
; CHECK-NEXT:    and z1.s, z1.s, #0x80000000
; CHECK-NEXT:    orr z0.d, z0.d, z1.d
; CHECK-NEXT:    ret
  %b.trunc = fptrunc <vscale x 2 x double> %b to <vscale x 2 x float>
  %r = call <vscale x 2 x float> @llvm.copysign.nxv2f32(<vscale x 2 x float> %a, <vscale x 2 x float> %b.trunc)
  ret <vscale x 2 x float> %r
}

define <vscale x 2 x float> @copysign_nxv2f32_nxv2bf16(<vscale x 2 x float> %a, <vscale x 2 x bfloat> %b) {
; CHECK-LABEL: copysign_nxv2f32_nxv2bf16:
; CHECK:       // %bb.0:
; CHECK-NEXT:    lsl z1.s, z1.s, #16
; CHECK-NEXT:    and z0.s, z0.s, #0x7fffffff
; CHECK-NEXT:    and z1.s, z1.s, #0x80000000
; CHECK-NEXT:    orr z0.d, z0.d, z1.d
; CHECK-NEXT:    ret
  %tmp0 = fpext <vscale x 2 x bfloat> %b to <vscale x 2 x float>
  %r = call <vscale x 2 x float> @llvm.copysign.nxv2f32(<vscale x 2 x float> %a, <vscale x 2 x float> %tmp0)
  ret <vscale x 2 x float> %r
}

;
; llvm.copysign.nxv4f32
;

define <vscale x 4 x float> @copysign_nxv4f32_nxv4f16(<vscale x 4 x float> %a, <vscale x 4 x half> %b) {
; CHECK-LABEL: copysign_nxv4f32_nxv4f16:
; CHECK:       // %bb.0:
; CHECK-NEXT:    ptrue p0.s
; CHECK-NEXT:    and z0.s, z0.s, #0x7fffffff
; CHECK-NEXT:    fcvt z1.s, p0/m, z1.h
; CHECK-NEXT:    and z1.s, z1.s, #0x80000000
; CHECK-NEXT:    orr z0.d, z0.d, z1.d
; CHECK-NEXT:    ret
  %b.ext = fpext <vscale x 4 x half> %b to <vscale x 4 x float>
  %r = call <vscale x 4 x float> @llvm.copysign.nxv4f32(<vscale x 4 x float> %a, <vscale x 4 x float> %b.ext)
  ret <vscale x 4 x float> %r
}

define <vscale x 4 x float> @copysign_nxv4f32_nxv4f32(<vscale x 4 x float> %a, <vscale x 4 x float> %b) {
; CHECK-LABEL: copysign_nxv4f32_nxv4f32:
; CHECK:       // %bb.0:
; CHECK-NEXT:    and z1.s, z1.s, #0x80000000
; CHECK-NEXT:    and z0.s, z0.s, #0x7fffffff
; CHECK-NEXT:    orr z0.d, z0.d, z1.d
; CHECK-NEXT:    ret
  %r = call <vscale x 4 x float> @llvm.copysign.nxv4f32(<vscale x 4 x float> %a, <vscale x 4 x float> %b)
  ret <vscale x 4 x float> %r
}

define <vscale x 4 x float> @copysign_nxv4f32_nxv4f64(<vscale x 4 x float> %a, <vscale x 4 x double> %b) {
; CHECK-LABEL: copysign_nxv4f32_nxv4f64:
; CHECK:       // %bb.0:
; CHECK-NEXT:    ptrue p0.d
; CHECK-NEXT:    and z0.s, z0.s, #0x7fffffff
; CHECK-NEXT:    fcvt z2.s, p0/m, z2.d
; CHECK-NEXT:    fcvt z1.s, p0/m, z1.d
; CHECK-NEXT:    uzp1 z1.s, z1.s, z2.s
; CHECK-NEXT:    and z1.s, z1.s, #0x80000000
; CHECK-NEXT:    orr z0.d, z0.d, z1.d
; CHECK-NEXT:    ret
  %b.trunc = fptrunc <vscale x 4 x double> %b to <vscale x 4 x float>
  %r = call <vscale x 4 x float> @llvm.copysign.nxv4f32(<vscale x 4 x float> %a, <vscale x 4 x float> %b.trunc)
  ret <vscale x 4 x float> %r
}

define <vscale x 4 x float> @copysign_nxv4f32_nxv4bf16(<vscale x 4 x float> %a, <vscale x 4 x bfloat> %b) {
; CHECK-LABEL: copysign_nxv4f32_nxv4bf16:
; CHECK:       // %bb.0:
; CHECK-NEXT:    lsl z1.s, z1.s, #16
; CHECK-NEXT:    and z0.s, z0.s, #0x7fffffff
; CHECK-NEXT:    and z1.s, z1.s, #0x80000000
; CHECK-NEXT:    orr z0.d, z0.d, z1.d
; CHECK-NEXT:    ret
  %tmp0 = fpext <vscale x 4 x bfloat> %b to <vscale x 4 x float>
  %r = call <vscale x 4 x float> @llvm.copysign.nxv4f32(<vscale x 4 x float> %a, <vscale x 4 x float> %tmp0)
  ret <vscale x 4 x float> %r
}

;
; llvm.copysign.nxv2f64
;

define <vscale x 2 x double> @copysign_nxv2f64_nxv2f16(<vscale x 2 x double> %a, <vscale x 2 x half> %b) {
; CHECK-LABEL: copysign_nxv2f64_nxv2f16:
; CHECK:       // %bb.0:
; CHECK-NEXT:    ptrue p0.d
; CHECK-NEXT:    and z0.d, z0.d, #0x7fffffffffffffff
; CHECK-NEXT:    fcvt z1.d, p0/m, z1.h
; CHECK-NEXT:    and z1.d, z1.d, #0x8000000000000000
; CHECK-NEXT:    orr z0.d, z0.d, z1.d
; CHECK-NEXT:    ret
  %b.ext = fpext <vscale x 2 x half> %b to <vscale x 2 x double>
  %r = call <vscale x 2 x double> @llvm.copysign.nxv2f64(<vscale x 2 x double> %a, <vscale x 2 x double> %b.ext)
  ret <vscale x 2 x double> %r
}

define <vscale x 2 x double> @copysign_nxv2f64_nxv2f32(<vscale x 2 x double> %a, <vscale x 2 x float> %b) {
; CHECK-LABEL: copysign_nxv2f64_nxv2f32:
; CHECK:       // %bb.0:
; CHECK-NEXT:    ptrue p0.d
; CHECK-NEXT:    and z0.d, z0.d, #0x7fffffffffffffff
; CHECK-NEXT:    fcvt z1.d, p0/m, z1.s
; CHECK-NEXT:    and z1.d, z1.d, #0x8000000000000000
; CHECK-NEXT:    orr z0.d, z0.d, z1.d
; CHECK-NEXT:    ret
  %b.ext = fpext <vscale x 2 x float> %b to <vscale x 2 x double>
  %r = call <vscale x 2 x double> @llvm.copysign.nxv2f64(<vscale x 2 x double> %a, <vscale x 2 x double> %b.ext)
  ret <vscale x 2 x double> %r
}

define <vscale x 2 x double> @copysign_nxv2f64_nxv2f64(<vscale x 2 x double> %a, <vscale x 2 x double> %b) {
; CHECK-LABEL: copysign_nxv2f64_nxv2f64:
; CHECK:       // %bb.0:
; CHECK-NEXT:    and z1.d, z1.d, #0x8000000000000000
; CHECK-NEXT:    and z0.d, z0.d, #0x7fffffffffffffff
; CHECK-NEXT:    orr z0.d, z0.d, z1.d
; CHECK-NEXT:    ret
  %r = call <vscale x 2 x double> @llvm.copysign.nxv2f64(<vscale x 2 x double> %a, <vscale x 2 x double> %b)
  ret <vscale x 2 x double> %r
}

define <vscale x 2 x double> @copysign_nxv2f64_nxv2bf16(<vscale x 2 x double> %a, <vscale x 2 x bfloat> %b) {
; CHECK-LABEL: copysign_nxv2f64_nxv2bf16:
; CHECK:       // %bb.0:
; CHECK-NEXT:    lsl z1.s, z1.s, #16
; CHECK-NEXT:    ptrue p0.d
; CHECK-NEXT:    and z0.d, z0.d, #0x7fffffffffffffff
; CHECK-NEXT:    fcvt z1.d, p0/m, z1.s
; CHECK-NEXT:    and z1.d, z1.d, #0x8000000000000000
; CHECK-NEXT:    orr z0.d, z0.d, z1.d
; CHECK-NEXT:    ret
  %b.ext = fpext <vscale x 2 x bfloat> %b to <vscale x 2 x double>
  %r = call <vscale x 2 x double> @llvm.copysign.nxv2f64(<vscale x 2 x double> %a, <vscale x 2 x double> %b.ext)
  ret <vscale x 2 x double> %r
}

;
; llvm.copysign.nxv2bf16
;

define <vscale x 2 x bfloat> @copysign_nxv2bf16_nxv2bf16(<vscale x 2 x bfloat> %a, <vscale x 2 x bfloat> %b) {
; CHECK-LABEL: copysign_nxv2bf16_nxv2bf16:
; CHECK:       // %bb.0:
; CHECK-NEXT:    and z1.h, z1.h, #0x8000
; CHECK-NEXT:    and z0.h, z0.h, #0x7fff
; CHECK-NEXT:    orr z0.d, z0.d, z1.d
; CHECK-NEXT:    ret
  %r = call <vscale x 2 x bfloat> @llvm.copysign.nxv2bf16(<vscale x 2 x bfloat> %a, <vscale x 2 x bfloat> %b)
  ret <vscale x 2 x bfloat> %r
}

define <vscale x 2 x bfloat> @copysign_nxv2bf16_nxv2f32(<vscale x 2 x bfloat> %a, <vscale x 2 x float> %b) {
; CHECK-LABEL: copysign_nxv2bf16_nxv2f32:
; CHECK:       // %bb.0:
; CHECK-NEXT:    ptrue p0.d
; CHECK-NEXT:    and z0.h, z0.h, #0x7fff
; CHECK-NEXT:    bfcvt z1.h, p0/m, z1.s
; CHECK-NEXT:    and z1.h, z1.h, #0x8000
; CHECK-NEXT:    orr z0.d, z0.d, z1.d
; CHECK-NEXT:    ret
  %tmp0 = fptrunc <vscale x 2 x float> %b to <vscale x 2 x bfloat>
  %r = call <vscale x 2 x bfloat> @llvm.copysign.nxv2bf16(<vscale x 2 x bfloat> %a, <vscale x 2 x bfloat> %tmp0)
  ret <vscale x 2 x bfloat> %r
}

; TODO: Cannot downconvert from double to bfloat
;define <vscale x 2 x bfloat> @copysign_nxv2bf16_nxv2f64(<vscale x 2 x bfloat> %a, <vscale x 2 x double> %b) {
;  %tmp0 = fptrunc <vscale x 2 x double> %b to <vscale x 2 x bfloat>
;  %r = call <vscale x 2 x bfloat> @llvm.copysign.nxv2bf16(<vscale x 2 x bfloat> %a, <vscale x 2 x bfloat> %tmp0)
;  ret <vscale x 2 x bfloat> %r
;}

;
; llvm.copysign.nxv2bf16
;

define <vscale x 4 x bfloat> @copysign_nxv4bf16_nxv4bf16(<vscale x 4 x bfloat> %a, <vscale x 4 x bfloat> %b) {
; CHECK-LABEL: copysign_nxv4bf16_nxv4bf16:
; CHECK:       // %bb.0:
; CHECK-NEXT:    and z1.h, z1.h, #0x8000
; CHECK-NEXT:    and z0.h, z0.h, #0x7fff
; CHECK-NEXT:    orr z0.d, z0.d, z1.d
; CHECK-NEXT:    ret
  %r = call <vscale x 4 x bfloat> @llvm.copysign.nxv4bf16(<vscale x 4 x bfloat> %a, <vscale x 4 x bfloat> %b)
  ret <vscale x 4 x bfloat> %r
}

define <vscale x 4 x bfloat> @copysign_nxv4bf16_nxv4f32(<vscale x 4 x bfloat> %a, <vscale x 4 x float> %b) {
; CHECK-LABEL: copysign_nxv4bf16_nxv4f32:
; CHECK:       // %bb.0:
; CHECK-NEXT:    ptrue p0.s
; CHECK-NEXT:    and z0.h, z0.h, #0x7fff
; CHECK-NEXT:    bfcvt z1.h, p0/m, z1.s
; CHECK-NEXT:    and z1.h, z1.h, #0x8000
; CHECK-NEXT:    orr z0.d, z0.d, z1.d
; CHECK-NEXT:    ret
  %b.trunc = fptrunc <vscale x 4 x float> %b to <vscale x 4 x bfloat>
  %r = call <vscale x 4 x bfloat> @llvm.copysign.nxv4bf16(<vscale x 4 x bfloat> %a, <vscale x 4 x bfloat> %b.trunc)
  ret <vscale x 4 x bfloat> %r
}

; TODO: Cannot downconvert from double to bfloat
;define <vscale x 4 x bfloat> @copysign_nxv4bf16_nxv4f64(<vscale x 4 x bfloat> %a, <vscale x 4 x double> %b) {
;  %b.trunc = fptrunc <vscale x 4 x double> %b to <vscale x 4 x bfloat>
;  %r = call <vscale x 4 x bfloat> @llvm.copysign.nxv4bf16(<vscale x 4 x bfloat> %a, <vscale x 4 x bfloat> %b.trunc)
;  ret <vscale x 4 x bfloat> %r
;}

;
; llvm.copysign.nxv8bf16
;

define <vscale x 8 x bfloat> @copysign_nxv8bf16_nxv8bf16(<vscale x 8 x bfloat> %a, <vscale x 8 x bfloat> %b) {
; CHECK-LABEL: copysign_nxv8bf16_nxv8bf16:
; CHECK:       // %bb.0:
; CHECK-NEXT:    and z1.h, z1.h, #0x8000
; CHECK-NEXT:    and z0.h, z0.h, #0x7fff
; CHECK-NEXT:    orr z0.d, z0.d, z1.d
; CHECK-NEXT:    ret
  %r = call <vscale x 8 x bfloat> @llvm.copysign.nxv8bf16(<vscale x 8 x bfloat> %a, <vscale x 8 x bfloat> %b)
  ret <vscale x 8 x bfloat> %r
}

define <vscale x 8 x bfloat> @copysign_nxv8bf16_nxv8f32(<vscale x 8 x bfloat> %a, <vscale x 8 x float> %b) {
; CHECK-LABEL: copysign_nxv8bf16_nxv8f32:
; CHECK:       // %bb.0:
; CHECK-NEXT:    ptrue p0.s
; CHECK-NEXT:    and z0.h, z0.h, #0x7fff
; CHECK-NEXT:    bfcvt z2.h, p0/m, z2.s
; CHECK-NEXT:    bfcvt z1.h, p0/m, z1.s
; CHECK-NEXT:    uzp1 z1.h, z1.h, z2.h
; CHECK-NEXT:    and z1.h, z1.h, #0x8000
; CHECK-NEXT:    orr z0.d, z0.d, z1.d
; CHECK-NEXT:    ret
  %b.trunc = fptrunc <vscale x 8 x float> %b to <vscale x 8 x bfloat>
  %r = call <vscale x 8 x bfloat> @llvm.copysign.nxv8bf16(<vscale x 8 x bfloat> %a, <vscale x 8 x bfloat> %b.trunc)
  ret <vscale x 8 x bfloat> %r
}

; TODO: Cannot downconvert from double to bfloat
;define <vscale x 8 x bfloat> @copysign_nxv8bf16_nxv8f64(<vscale x 8 x bfloat> %a, <vscale x 8 x double> %b) {
;  %b.trunc = fptrunc <vscale x 8 x double> %b to <vscale x 8 x bfloat>
;  %r = call <vscale x 8 x bfloat> @llvm.copysign.nxv8bf16(<vscale x 8 x bfloat> %a, <vscale x 8 x bfloat> %b.trunc)
;  ret <vscale x 8 x bfloat> %r
;}

declare <vscale x 2 x half> @llvm.copysign.nxv2f16(<vscale x 2 x half> %a, <vscale x 2 x half> %b)
declare <vscale x 4 x half> @llvm.copysign.nxv4f16(<vscale x 4 x half> %a, <vscale x 4 x half> %b)
declare <vscale x 8 x half> @llvm.copysign.nxv8f16(<vscale x 8 x half> %a, <vscale x 8 x half> %b)
declare <vscale x 2 x float> @llvm.copysign.nxv2f32(<vscale x 2 x float> %a, <vscale x 2 x float> %b)
declare <vscale x 4 x float> @llvm.copysign.nxv4f32(<vscale x 4 x float> %a, <vscale x 4 x float> %b)
declare <vscale x 2 x double> @llvm.copysign.nxv2f64(<vscale x 2 x double> %a, <vscale x 2 x double> %b)
declare <vscale x 2 x bfloat> @llvm.copysign.nxv2bf16(<vscale x 2 x bfloat> %a, <vscale x 2 x bfloat> %b)
declare <vscale x 4 x bfloat> @llvm.copysign.nxv4bf16(<vscale x 4 x bfloat> %a, <vscale x 4 x bfloat> %b)
declare <vscale x 8 x bfloat> @llvm.copysign.nxv8bf16(<vscale x 8 x bfloat> %a, <vscale x 8 x bfloat> %b)