; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 5
; RUN: llc -mtriple=aarch64-gnu-linux -mattr=+neon,+sve -vector-library=sleefgnuabi < %s | FileCheck %s -check-prefix=SLEEF
; RUN: llc -mtriple=aarch64-gnu-linux -mattr=+neon,+sve -vector-library=ArmPL < %s | FileCheck %s -check-prefix=ARMPL

define <4 x float> @test_modf_v4f32(<4 x float> %x, ptr %out_integral) {
; SLEEF-LABEL: test_modf_v4f32:
; SLEEF:       // %bb.0:
; SLEEF-NEXT:    str x30, [sp, #-16]! // 8-byte Folded Spill
; SLEEF-NEXT:    .cfi_def_cfa_offset 16
; SLEEF-NEXT:    .cfi_offset w30, -16
; SLEEF-NEXT:    bl _ZGVnN4vl4_modff
; SLEEF-NEXT:    ldr x30, [sp], #16 // 8-byte Folded Reload
; SLEEF-NEXT:    ret
;
; ARMPL-LABEL: test_modf_v4f32:
; ARMPL:       // %bb.0:
; ARMPL-NEXT:    str x30, [sp, #-16]! // 8-byte Folded Spill
; ARMPL-NEXT:    .cfi_def_cfa_offset 16
; ARMPL-NEXT:    .cfi_offset w30, -16
; ARMPL-NEXT:    bl armpl_vmodfq_f32
; ARMPL-NEXT:    ldr x30, [sp], #16 // 8-byte Folded Reload
; ARMPL-NEXT:    ret
  %result = call { <4 x float>, <4 x float> } @llvm.modf.v4f32(<4 x float> %x)
  %result.0 = extractvalue { <4 x float>, <4 x float> } %result, 0
  %result.1 = extractvalue { <4 x float>, <4 x float> } %result, 1
  store <4 x float> %result.1, ptr %out_integral, align 4
  ret <4 x float> %result.0
}

define <2 x double> @test_modf_v2f64(<2 x double> %x, ptr %out_integral) {
; SLEEF-LABEL: test_modf_v2f64:
; SLEEF:       // %bb.0:
; SLEEF-NEXT:    str x30, [sp, #-16]! // 8-byte Folded Spill
; SLEEF-NEXT:    .cfi_def_cfa_offset 16
; SLEEF-NEXT:    .cfi_offset w30, -16
; SLEEF-NEXT:    bl _ZGVnN2vl8_modf
; SLEEF-NEXT:    ldr x30, [sp], #16 // 8-byte Folded Reload
; SLEEF-NEXT:    ret
;
; ARMPL-LABEL: test_modf_v2f64:
; ARMPL:       // %bb.0:
; ARMPL-NEXT:    str x30, [sp, #-16]! // 8-byte Folded Spill
; ARMPL-NEXT:    .cfi_def_cfa_offset 16
; ARMPL-NEXT:    .cfi_offset w30, -16
; ARMPL-NEXT:    bl armpl_vmodfq_f64
; ARMPL-NEXT:    ldr x30, [sp], #16 // 8-byte Folded Reload
; ARMPL-NEXT:    ret
  %result = call { <2 x double>, <2 x double> } @llvm.modf.v2f64(<2 x double> %x)
  %result.0 = extractvalue { <2 x double>, <2 x double> } %result, 0
  %result.1 = extractvalue { <2 x double>, <2 x double> } %result, 1
  store <2 x double> %result.1, ptr %out_integral, align 8
  ret <2 x double> %result.0
}

define <vscale x 4 x float> @test_modf_nxv4f32(<vscale x 4 x float> %x, ptr %out_integral) {
; SLEEF-LABEL: test_modf_nxv4f32:
; SLEEF:       // %bb.0:
; SLEEF-NEXT:    str x30, [sp, #-16]! // 8-byte Folded Spill
; SLEEF-NEXT:    .cfi_def_cfa_offset 16
; SLEEF-NEXT:    .cfi_offset w30, -16
; SLEEF-NEXT:    bl _ZGVsNxvl4_modff
; SLEEF-NEXT:    ldr x30, [sp], #16 // 8-byte Folded Reload
; SLEEF-NEXT:    ret
;
; ARMPL-LABEL: test_modf_nxv4f32:
; ARMPL:       // %bb.0:
; ARMPL-NEXT:    str x30, [sp, #-16]! // 8-byte Folded Spill
; ARMPL-NEXT:    .cfi_def_cfa_offset 16
; ARMPL-NEXT:    .cfi_offset w30, -16
; ARMPL-NEXT:    ptrue p0.s
; ARMPL-NEXT:    bl armpl_svmodf_f32_x
; ARMPL-NEXT:    ldr x30, [sp], #16 // 8-byte Folded Reload
; ARMPL-NEXT:    ret
  %result = call { <vscale x 4 x float>, <vscale x 4 x float> } @llvm.modf.nxv4f32(<vscale x 4 x float> %x)
  %result.0 = extractvalue { <vscale x 4 x float>, <vscale x 4 x float> } %result, 0
  %result.1 = extractvalue { <vscale x 4 x float>, <vscale x 4 x float> } %result, 1
  store <vscale x 4 x float> %result.1, ptr %out_integral, align 4
  ret <vscale x 4 x float> %result.0
}

define <vscale x 2 x double> @test_modf_nxv2f64(<vscale x 2 x double> %x, ptr %out_integral) {
; SLEEF-LABEL: test_modf_nxv2f64:
; SLEEF:       // %bb.0:
; SLEEF-NEXT:    str x30, [sp, #-16]! // 8-byte Folded Spill
; SLEEF-NEXT:    .cfi_def_cfa_offset 16
; SLEEF-NEXT:    .cfi_offset w30, -16
; SLEEF-NEXT:    bl _ZGVsNxvl8_modf
; SLEEF-NEXT:    ldr x30, [sp], #16 // 8-byte Folded Reload
; SLEEF-NEXT:    ret
;
; ARMPL-LABEL: test_modf_nxv2f64:
; ARMPL:       // %bb.0:
; ARMPL-NEXT:    str x30, [sp, #-16]! // 8-byte Folded Spill
; ARMPL-NEXT:    .cfi_def_cfa_offset 16
; ARMPL-NEXT:    .cfi_offset w30, -16
; ARMPL-NEXT:    ptrue p0.d
; ARMPL-NEXT:    bl armpl_svmodf_f64_x
; ARMPL-NEXT:    ldr x30, [sp], #16 // 8-byte Folded Reload
; ARMPL-NEXT:    ret
  %result = call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.modf.nxv2f64(<vscale x 2 x double> %x)
  %result.0 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %result, 0
  %result.1 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %result, 1
  store <vscale x 2 x double> %result.1, ptr %out_integral, align 8
  ret <vscale x 2 x double> %result.0
}

define <4 x float> @modf_store_merging_load_before_store(<4 x float> %x, ptr %out_integral) {
; SLEEF-LABEL: modf_store_merging_load_before_store:
; SLEEF:       // %bb.0:
; SLEEF-NEXT:    sub sp, sp, #32
; SLEEF-NEXT:    str x30, [sp, #16] // 8-byte Spill
; SLEEF-NEXT:    .cfi_def_cfa_offset 32
; SLEEF-NEXT:    .cfi_offset w30, -16
; SLEEF-NEXT:    ldr q1, [x0]
; SLEEF-NEXT:    str q1, [sp] // 16-byte Spill
; SLEEF-NEXT:    bl _ZGVnN4vl4_modff
; SLEEF-NEXT:    ldr q1, [sp] // 16-byte Reload
; SLEEF-NEXT:    ldr x30, [sp, #16] // 8-byte Reload
; SLEEF-NEXT:    fadd v0.4s, v1.4s, v0.4s
; SLEEF-NEXT:    add sp, sp, #32
; SLEEF-NEXT:    ret
;
; ARMPL-LABEL: modf_store_merging_load_before_store:
; ARMPL:       // %bb.0:
; ARMPL-NEXT:    sub sp, sp, #32
; ARMPL-NEXT:    str x30, [sp, #16] // 8-byte Spill
; ARMPL-NEXT:    .cfi_def_cfa_offset 32
; ARMPL-NEXT:    .cfi_offset w30, -16
; ARMPL-NEXT:    ldr q1, [x0]
; ARMPL-NEXT:    str q1, [sp] // 16-byte Spill
; ARMPL-NEXT:    bl armpl_vmodfq_f32
; ARMPL-NEXT:    ldr q1, [sp] // 16-byte Reload
; ARMPL-NEXT:    ldr x30, [sp, #16] // 8-byte Reload
; ARMPL-NEXT:    fadd v0.4s, v1.4s, v0.4s
; ARMPL-NEXT:    add sp, sp, #32
; ARMPL-NEXT:    ret
  %result = call { <4 x float>, <4 x float> } @llvm.modf.v4f32(<4 x float> %x)
  %result.0 = extractvalue { <4 x float>, <4 x float> } %result, 0
  %result.1 = extractvalue { <4 x float>, <4 x float> } %result, 1
  %original_intergral = load <4 x float>, ptr %out_integral, align 4
  store <4 x float> %result.1, ptr %out_integral, align 4
  %return = fadd <4 x float> %original_intergral, %result.0
  ret <4 x float> %return
}