Compiler Explorer

Source code

target datalayout = "e-m:e-p270:32:32-p271:32:32-p272:64:64-i64:64-i128:128-f80:128-n8:16:32:64-S128"
target triple = "x86_64-unknown-linux-gnu"

define void @test_interleaved___un_3C_Cunf_3E_un_3C_Cunf_3E_un_3C_Cunf_3E_un_3C_uni_3E_(ptr noalias nocapture readonly %xin, ptr noalias nocapture readonly %yin, ptr noalias nocapture readonly %zin, ptr noalias nocapture %data, <4 x i32> %__mask) local_unnamed_addr #0 {
allocas:
  %invariant.gep = getelementptr i8, ptr %data, i64 8
  %invariant.gep320 = getelementptr i8, ptr %data, i64 16
  %invariant.gep322 = getelementptr i8, ptr %data, i64 24
  %invariant.gep324 = getelementptr i8, ptr %data, i64 4
  %invariant.gep326 = getelementptr i8, ptr %data, i64 12
  %invariant.gep328 = getelementptr i8, ptr %data, i64 20
  %invariant.gep330 = getelementptr i8, ptr %data, i64 28
  br label %foreach_full_body

foreach_full_body:
  %indvars.iv = phi i64 [ 0, %allocas ], [ %indvars.iv.next, %foreach_full_body ]
  %0 = shl nuw nsw i64 %indvars.iv, 2
  %ptr = getelementptr i8, ptr %xin, i64 %0
  %ptr_masked_load240 = load <4 x float>, ptr %ptr, align 4
  %mul__xin_load_offset_load = fmul <4 x float> %ptr_masked_load240, splat (float 2.000000e+00)
  %ptr248 = getelementptr i8, ptr %yin, i64 %0
  %ptr248_masked_load249 = load <4 x float>, ptr %ptr248, align 4
  %mul__yin_load_offset_load = fmul <4 x float> %ptr248_masked_load249, splat (float 3.000000e+00)
  %add_mul__xin_load_offset_load_mul__yin_load_offset_load = fadd <4 x float> %mul__xin_load_offset_load, %mul__yin_load_offset_load
  %ptr257 = getelementptr i8, ptr %zin, i64 %0
  %ptr257_masked_load258 = load <4 x float>, ptr %ptr257, align 4
  %mul__zin_load_offset_load = fmul <4 x float> %ptr257_masked_load258, splat (float 2.000000e+00)
  %add_add_mul__xin_load_offset_load_mul__yin_load_offset_load_mul__zin_load_offset_load = fadd <4 x float> %add_mul__xin_load_offset_load_mul__yin_load_offset_load, %mul__zin_load_offset_load
  %div_add_add_mul__xin_load_offset_load_mul__yin_load_offset_load_mul__zin_load_offset_load_ = fdiv <4 x float> %add_add_mul__xin_load_offset_load_mul__yin_load_offset_load_mul__zin_load_offset_load, splat (float 7.000000e+00)
  %div_add_add_mul__xin_load_offset_load_mul__yin_load_offset_load_mul__zin_load_offset_load__to_int32 = fptosi <4 x float> %div_add_add_mul__xin_load_offset_load_mul__yin_load_offset_load_mul__zin_load_offset_load_ to <4 x i32>
  %add_xin_load44_offset_load_yin_load51_offset_load = fadd <4 x float> %ptr_masked_load240, %ptr248_masked_load249
  %add_add_xin_load44_offset_load_yin_load51_offset_load_zin_load58_offset_load = fadd <4 x float> %add_xin_load44_offset_load_yin_load51_offset_load, %ptr257_masked_load258
  %div_add_add_xin_load44_offset_load_yin_load51_offset_load_zin_load58_offset_load_ = fdiv <4 x float> %add_add_xin_load44_offset_load_yin_load51_offset_load_zin_load58_offset_load, splat (float 3.000000e+00)
  %div_add_add_xin_load44_offset_load_yin_load51_offset_load_zin_load58_offset_load__to_int32 = fptosi <4 x float> %div_add_add_xin_load44_offset_load_yin_load51_offset_load_zin_load58_offset_load_ to <4 x i32>
  %offset32.i.i = shl nuw nsw i64 %indvars.iv, 3
  %offset64.i.i = and i64 %offset32.i.i, 4294967264
  %ptroffset.i.i = getelementptr i8, ptr %data, i64 %offset64.i.i
  %storeval.i.i = extractelement <4 x i32> %div_add_add_mul__xin_load_offset_load_mul__yin_load_offset_load_mul__zin_load_offset_load__to_int32, i64 0
  store i32 %storeval.i.i, ptr %ptroffset.i.i, align 4
  %gep = getelementptr i8, ptr %invariant.gep, i64 %offset64.i.i
  %storeval.i9.i = extractelement <4 x i32> %div_add_add_mul__xin_load_offset_load_mul__yin_load_offset_load_mul__zin_load_offset_load__to_int32, i64 1
  store i32 %storeval.i9.i, ptr %gep, align 4
  %gep321 = getelementptr i8, ptr %invariant.gep320, i64 %offset64.i.i
  %storeval.i18.i = extractelement <4 x i32> %div_add_add_mul__xin_load_offset_load_mul__yin_load_offset_load_mul__zin_load_offset_load__to_int32, i64 2
  store i32 %storeval.i18.i, ptr %gep321, align 4
  %gep323 = getelementptr i8, ptr %invariant.gep322, i64 %offset64.i.i
  %storeval.i27.i = extractelement <4 x i32> %div_add_add_mul__xin_load_offset_load_mul__yin_load_offset_load_mul__zin_load_offset_load__to_int32, i64 3
  store i32 %storeval.i27.i, ptr %gep323, align 4
  %gep325 = getelementptr i8, ptr %invariant.gep324, i64 %offset64.i.i
  %storeval.i.i302 = extractelement <4 x i32> %div_add_add_xin_load44_offset_load_yin_load51_offset_load_zin_load58_offset_load__to_int32, i64 0
  store i32 %storeval.i.i302, ptr %gep325, align 4
  %gep327 = getelementptr i8, ptr %invariant.gep326, i64 %offset64.i.i
  %storeval.i9.i307 = extractelement <4 x i32> %div_add_add_xin_load44_offset_load_yin_load51_offset_load_zin_load58_offset_load__to_int32, i64 1
  store i32 %storeval.i9.i307, ptr %gep327, align 4
  %gep329 = getelementptr i8, ptr %invariant.gep328, i64 %offset64.i.i
  %storeval.i18.i312 = extractelement <4 x i32> %div_add_add_xin_load44_offset_load_yin_load51_offset_load_zin_load58_offset_load__to_int32, i64 2
  store i32 %storeval.i18.i312, ptr %gep329, align 4
  %gep331 = getelementptr i8, ptr %invariant.gep330, i64 %offset64.i.i
  %storeval.i27.i317 = extractelement <4 x i32> %div_add_add_xin_load44_offset_load_yin_load51_offset_load_zin_load58_offset_load__to_int32, i64 3
  store i32 %storeval.i27.i317, ptr %gep331, align 4
  %indvars.iv.next = add nuw nsw i64 %indvars.iv, 4
  %before_aligned_end18 = icmp samesign ult i64 %indvars.iv, 252
  br i1 %before_aligned_end18, label %foreach_full_body, label %partial_inner_all_outer

partial_inner_all_outer:
  ret void
}

attributes #0 = { nounwind }
attributes #1 = { nocallback nofree nosync nounwind speculatable willreturn memory(none) }