1
0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-11-23 11:13:28 +01:00
llvm-mirror/test/CodeGen/X86/horizontal-reduce-fadd.ll
Simon Pilgrim 21418c85a1 [X86][SSE] getV4X86ShuffleImm8 - canonicalize broadcast masks
If the mask input to getV4X86ShuffleImm8 only refers to a single source element (+ undefs) then canonicalize to a full broadcast.

getV4X86ShuffleImm8 defaults to inline values for undefs, which can be useful for shuffle widening/narrowing but does leave SimplifyDemanded* calls thinking the shuffle depends on unnecessary elements.

I'm still investigating what we should do more generally to avoid these undemanded elements, but broadcast cases was a simpler win.
2020-07-29 11:32:44 +01:00

353 lines
14 KiB
LLVM

; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc < %s -mtriple=x86_64-unknown -mattr=+sse2 | FileCheck %s --check-prefixes=SSE,SSE2
; RUN: llc < %s -mtriple=x86_64-unknown -mattr=+ssse3 | FileCheck %s --check-prefixes=SSE,SSSE3,SSSE3-SLOW
; RUN: llc < %s -mtriple=x86_64-unknown -mattr=+ssse3,fast-hops | FileCheck %s --check-prefixes=SSE,SSSE3,SSSE3-FAST
; RUN: llc < %s -mtriple=x86_64-unknown -mattr=+avx | FileCheck %s --check-prefixes=AVX,AVX1,AVX1-SLOW
; RUN: llc < %s -mtriple=x86_64-unknown -mattr=+avx,fast-hops | FileCheck %s --check-prefixes=AVX,AVX1,AVX1-FAST
; RUN: llc < %s -mtriple=x86_64-unknown -mattr=+avx2 | FileCheck %s --check-prefixes=AVX,AVX2
; PR37890 - subvector reduction followed by shuffle reduction
define float @PR37890_v4f32(<4 x float> %a) {
; SSE2-LABEL: PR37890_v4f32:
; SSE2: # %bb.0:
; SSE2-NEXT: movaps %xmm0, %xmm1
; SSE2-NEXT: unpckhpd {{.*#+}} xmm1 = xmm1[1],xmm0[1]
; SSE2-NEXT: addps %xmm0, %xmm1
; SSE2-NEXT: movaps %xmm1, %xmm0
; SSE2-NEXT: shufps {{.*#+}} xmm0 = xmm0[1,1],xmm1[1,1]
; SSE2-NEXT: addss %xmm1, %xmm0
; SSE2-NEXT: retq
;
; SSSE3-SLOW-LABEL: PR37890_v4f32:
; SSSE3-SLOW: # %bb.0:
; SSSE3-SLOW-NEXT: movaps %xmm0, %xmm1
; SSSE3-SLOW-NEXT: unpckhpd {{.*#+}} xmm1 = xmm1[1],xmm0[1]
; SSSE3-SLOW-NEXT: addps %xmm0, %xmm1
; SSSE3-SLOW-NEXT: movshdup {{.*#+}} xmm0 = xmm1[1,1,3,3]
; SSSE3-SLOW-NEXT: addss %xmm0, %xmm1
; SSSE3-SLOW-NEXT: movaps %xmm1, %xmm0
; SSSE3-SLOW-NEXT: retq
;
; SSSE3-FAST-LABEL: PR37890_v4f32:
; SSSE3-FAST: # %bb.0:
; SSSE3-FAST-NEXT: haddps %xmm0, %xmm0
; SSSE3-FAST-NEXT: haddps %xmm0, %xmm0
; SSSE3-FAST-NEXT: retq
;
; AVX1-SLOW-LABEL: PR37890_v4f32:
; AVX1-SLOW: # %bb.0:
; AVX1-SLOW-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
; AVX1-SLOW-NEXT: vaddps %xmm1, %xmm0, %xmm0
; AVX1-SLOW-NEXT: vmovshdup {{.*#+}} xmm1 = xmm0[1,1,3,3]
; AVX1-SLOW-NEXT: vaddss %xmm1, %xmm0, %xmm0
; AVX1-SLOW-NEXT: retq
;
; AVX1-FAST-LABEL: PR37890_v4f32:
; AVX1-FAST: # %bb.0:
; AVX1-FAST-NEXT: vhaddps %xmm0, %xmm0, %xmm0
; AVX1-FAST-NEXT: vhaddps %xmm0, %xmm0, %xmm0
; AVX1-FAST-NEXT: retq
;
; AVX2-LABEL: PR37890_v4f32:
; AVX2: # %bb.0:
; AVX2-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
; AVX2-NEXT: vaddps %xmm1, %xmm0, %xmm0
; AVX2-NEXT: vmovshdup {{.*#+}} xmm1 = xmm0[1,1,3,3]
; AVX2-NEXT: vaddss %xmm1, %xmm0, %xmm0
; AVX2-NEXT: retq
%hi0 = shufflevector <4 x float> %a, <4 x float> undef, <2 x i32> <i32 2, i32 3>
%lo0 = shufflevector <4 x float> %a, <4 x float> undef, <2 x i32> <i32 0, i32 1>
%sum0 = fadd fast <2 x float> %lo0, %hi0
%hi1 = shufflevector <2 x float> %sum0, <2 x float> undef, <2 x i32> <i32 1, i32 undef>
%sum1 = fadd fast <2 x float> %sum0, %hi1
%e = extractelement <2 x float> %sum1, i32 0
ret float %e
}
define double @PR37890_v4f64(<4 x double> %a) {
; SSE2-LABEL: PR37890_v4f64:
; SSE2: # %bb.0:
; SSE2-NEXT: addpd %xmm1, %xmm0
; SSE2-NEXT: movapd %xmm0, %xmm1
; SSE2-NEXT: unpckhpd {{.*#+}} xmm1 = xmm1[1],xmm0[1]
; SSE2-NEXT: addsd %xmm0, %xmm1
; SSE2-NEXT: movapd %xmm1, %xmm0
; SSE2-NEXT: retq
;
; SSSE3-SLOW-LABEL: PR37890_v4f64:
; SSSE3-SLOW: # %bb.0:
; SSSE3-SLOW-NEXT: addpd %xmm1, %xmm0
; SSSE3-SLOW-NEXT: movapd %xmm0, %xmm1
; SSSE3-SLOW-NEXT: unpckhpd {{.*#+}} xmm1 = xmm1[1],xmm0[1]
; SSSE3-SLOW-NEXT: addsd %xmm0, %xmm1
; SSSE3-SLOW-NEXT: movapd %xmm1, %xmm0
; SSSE3-SLOW-NEXT: retq
;
; SSSE3-FAST-LABEL: PR37890_v4f64:
; SSSE3-FAST: # %bb.0:
; SSSE3-FAST-NEXT: addpd %xmm1, %xmm0
; SSSE3-FAST-NEXT: haddpd %xmm0, %xmm0
; SSSE3-FAST-NEXT: retq
;
; AVX1-SLOW-LABEL: PR37890_v4f64:
; AVX1-SLOW: # %bb.0:
; AVX1-SLOW-NEXT: vextractf128 $1, %ymm0, %xmm1
; AVX1-SLOW-NEXT: vaddpd %xmm1, %xmm0, %xmm0
; AVX1-SLOW-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
; AVX1-SLOW-NEXT: vaddsd %xmm1, %xmm0, %xmm0
; AVX1-SLOW-NEXT: vzeroupper
; AVX1-SLOW-NEXT: retq
;
; AVX1-FAST-LABEL: PR37890_v4f64:
; AVX1-FAST: # %bb.0:
; AVX1-FAST-NEXT: vextractf128 $1, %ymm0, %xmm1
; AVX1-FAST-NEXT: vhaddpd %xmm0, %xmm1, %xmm0
; AVX1-FAST-NEXT: vhaddpd %xmm0, %xmm0, %xmm0
; AVX1-FAST-NEXT: vzeroupper
; AVX1-FAST-NEXT: retq
;
; AVX2-LABEL: PR37890_v4f64:
; AVX2: # %bb.0:
; AVX2-NEXT: vextractf128 $1, %ymm0, %xmm1
; AVX2-NEXT: vaddpd %xmm1, %xmm0, %xmm0
; AVX2-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
; AVX2-NEXT: vaddsd %xmm1, %xmm0, %xmm0
; AVX2-NEXT: vzeroupper
; AVX2-NEXT: retq
%hi0 = shufflevector <4 x double> %a, <4 x double> undef, <2 x i32> <i32 2, i32 3>
%lo0 = shufflevector <4 x double> %a, <4 x double> undef, <2 x i32> <i32 0, i32 1>
%sum0 = fadd fast <2 x double> %lo0, %hi0
%hi1 = shufflevector <2 x double> %sum0, <2 x double> undef, <2 x i32> <i32 1, i32 undef>
%sum1 = fadd fast <2 x double> %sum0, %hi1
%e = extractelement <2 x double> %sum1, i32 0
ret double %e
}
define float @PR37890_v8f32(<8 x float> %a) {
; SSE2-LABEL: PR37890_v8f32:
; SSE2: # %bb.0:
; SSE2-NEXT: addps %xmm1, %xmm0
; SSE2-NEXT: movaps %xmm0, %xmm1
; SSE2-NEXT: unpckhpd {{.*#+}} xmm1 = xmm1[1],xmm0[1]
; SSE2-NEXT: addps %xmm0, %xmm1
; SSE2-NEXT: movaps %xmm1, %xmm0
; SSE2-NEXT: shufps {{.*#+}} xmm0 = xmm0[1,1],xmm1[1,1]
; SSE2-NEXT: addss %xmm1, %xmm0
; SSE2-NEXT: retq
;
; SSSE3-SLOW-LABEL: PR37890_v8f32:
; SSSE3-SLOW: # %bb.0:
; SSSE3-SLOW-NEXT: addps %xmm1, %xmm0
; SSSE3-SLOW-NEXT: movaps %xmm0, %xmm1
; SSSE3-SLOW-NEXT: unpckhpd {{.*#+}} xmm1 = xmm1[1],xmm0[1]
; SSSE3-SLOW-NEXT: addps %xmm0, %xmm1
; SSSE3-SLOW-NEXT: movshdup {{.*#+}} xmm0 = xmm1[1,1,3,3]
; SSSE3-SLOW-NEXT: addss %xmm0, %xmm1
; SSSE3-SLOW-NEXT: movaps %xmm1, %xmm0
; SSSE3-SLOW-NEXT: retq
;
; SSSE3-FAST-LABEL: PR37890_v8f32:
; SSSE3-FAST: # %bb.0:
; SSSE3-FAST-NEXT: addps %xmm1, %xmm0
; SSSE3-FAST-NEXT: haddps %xmm0, %xmm0
; SSSE3-FAST-NEXT: haddps %xmm0, %xmm0
; SSSE3-FAST-NEXT: retq
;
; AVX1-SLOW-LABEL: PR37890_v8f32:
; AVX1-SLOW: # %bb.0:
; AVX1-SLOW-NEXT: vextractf128 $1, %ymm0, %xmm1
; AVX1-SLOW-NEXT: vaddps %xmm1, %xmm0, %xmm0
; AVX1-SLOW-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
; AVX1-SLOW-NEXT: vaddps %xmm1, %xmm0, %xmm0
; AVX1-SLOW-NEXT: vmovshdup {{.*#+}} xmm1 = xmm0[1,1,3,3]
; AVX1-SLOW-NEXT: vaddss %xmm1, %xmm0, %xmm0
; AVX1-SLOW-NEXT: vzeroupper
; AVX1-SLOW-NEXT: retq
;
; AVX1-FAST-LABEL: PR37890_v8f32:
; AVX1-FAST: # %bb.0:
; AVX1-FAST-NEXT: vextractf128 $1, %ymm0, %xmm1
; AVX1-FAST-NEXT: vhaddps %xmm0, %xmm1, %xmm0
; AVX1-FAST-NEXT: vhaddps %xmm0, %xmm0, %xmm0
; AVX1-FAST-NEXT: vhaddps %xmm0, %xmm0, %xmm0
; AVX1-FAST-NEXT: vzeroupper
; AVX1-FAST-NEXT: retq
;
; AVX2-LABEL: PR37890_v8f32:
; AVX2: # %bb.0:
; AVX2-NEXT: vextractf128 $1, %ymm0, %xmm1
; AVX2-NEXT: vaddps %xmm1, %xmm0, %xmm0
; AVX2-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
; AVX2-NEXT: vaddps %xmm1, %xmm0, %xmm0
; AVX2-NEXT: vmovshdup {{.*#+}} xmm1 = xmm0[1,1,3,3]
; AVX2-NEXT: vaddss %xmm1, %xmm0, %xmm0
; AVX2-NEXT: vzeroupper
; AVX2-NEXT: retq
%hi0 = shufflevector <8 x float> %a, <8 x float> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
%lo0 = shufflevector <8 x float> %a, <8 x float> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
%sum0 = fadd fast <4 x float> %lo0, %hi0
%hi1 = shufflevector <4 x float> %sum0, <4 x float> undef, <2 x i32> <i32 2, i32 3>
%lo1 = shufflevector <4 x float> %sum0, <4 x float> undef, <2 x i32> <i32 0, i32 1>
%sum1 = fadd fast <2 x float> %lo1, %hi1
%hi2 = shufflevector <2 x float> %sum1, <2 x float> undef, <2 x i32> <i32 1, i32 undef>
%sum2 = fadd fast <2 x float> %sum1, %hi2
%e = extractelement <2 x float> %sum2, i32 0
ret float %e
}
define double @PR37890_v8f64(<8 x double> %a) {
; SSE2-LABEL: PR37890_v8f64:
; SSE2: # %bb.0:
; SSE2-NEXT: addpd %xmm3, %xmm1
; SSE2-NEXT: addpd %xmm2, %xmm1
; SSE2-NEXT: addpd %xmm0, %xmm1
; SSE2-NEXT: movapd %xmm1, %xmm0
; SSE2-NEXT: unpckhpd {{.*#+}} xmm0 = xmm0[1],xmm1[1]
; SSE2-NEXT: addsd %xmm1, %xmm0
; SSE2-NEXT: retq
;
; SSSE3-SLOW-LABEL: PR37890_v8f64:
; SSSE3-SLOW: # %bb.0:
; SSSE3-SLOW-NEXT: addpd %xmm3, %xmm1
; SSSE3-SLOW-NEXT: addpd %xmm2, %xmm1
; SSSE3-SLOW-NEXT: addpd %xmm0, %xmm1
; SSSE3-SLOW-NEXT: movapd %xmm1, %xmm0
; SSSE3-SLOW-NEXT: unpckhpd {{.*#+}} xmm0 = xmm0[1],xmm1[1]
; SSSE3-SLOW-NEXT: addsd %xmm1, %xmm0
; SSSE3-SLOW-NEXT: retq
;
; SSSE3-FAST-LABEL: PR37890_v8f64:
; SSSE3-FAST: # %bb.0:
; SSSE3-FAST-NEXT: addpd %xmm3, %xmm1
; SSSE3-FAST-NEXT: addpd %xmm2, %xmm1
; SSSE3-FAST-NEXT: addpd %xmm0, %xmm1
; SSSE3-FAST-NEXT: haddpd %xmm1, %xmm1
; SSSE3-FAST-NEXT: movapd %xmm1, %xmm0
; SSSE3-FAST-NEXT: retq
;
; AVX1-SLOW-LABEL: PR37890_v8f64:
; AVX1-SLOW: # %bb.0:
; AVX1-SLOW-NEXT: vaddpd %ymm1, %ymm0, %ymm0
; AVX1-SLOW-NEXT: vextractf128 $1, %ymm0, %xmm1
; AVX1-SLOW-NEXT: vaddpd %xmm1, %xmm0, %xmm0
; AVX1-SLOW-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
; AVX1-SLOW-NEXT: vaddsd %xmm1, %xmm0, %xmm0
; AVX1-SLOW-NEXT: vzeroupper
; AVX1-SLOW-NEXT: retq
;
; AVX1-FAST-LABEL: PR37890_v8f64:
; AVX1-FAST: # %bb.0:
; AVX1-FAST-NEXT: vaddpd %ymm1, %ymm0, %ymm0
; AVX1-FAST-NEXT: vextractf128 $1, %ymm0, %xmm1
; AVX1-FAST-NEXT: vhaddpd %xmm0, %xmm1, %xmm0
; AVX1-FAST-NEXT: vhaddpd %xmm0, %xmm0, %xmm0
; AVX1-FAST-NEXT: vzeroupper
; AVX1-FAST-NEXT: retq
;
; AVX2-LABEL: PR37890_v8f64:
; AVX2: # %bb.0:
; AVX2-NEXT: vaddpd %ymm1, %ymm0, %ymm0
; AVX2-NEXT: vextractf128 $1, %ymm0, %xmm1
; AVX2-NEXT: vaddpd %xmm1, %xmm0, %xmm0
; AVX2-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
; AVX2-NEXT: vaddsd %xmm1, %xmm0, %xmm0
; AVX2-NEXT: vzeroupper
; AVX2-NEXT: retq
%hi0 = shufflevector <8 x double> %a, <8 x double> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
%lo0 = shufflevector <8 x double> %a, <8 x double> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
%sum0 = fadd fast <4 x double> %lo0, %hi0
%hi1 = shufflevector <4 x double> %sum0, <4 x double> undef, <2 x i32> <i32 2, i32 3>
%lo1 = shufflevector <4 x double> %sum0, <4 x double> undef, <2 x i32> <i32 0, i32 1>
%sum1 = fadd fast <2 x double> %lo1, %hi1
%hi2 = shufflevector <2 x double> %sum1, <2 x double> undef, <2 x i32> <i32 1, i32 undef>
%sum2 = fadd fast <2 x double> %sum1, %hi2
%e = extractelement <2 x double> %sum2, i32 0
ret double %e
}
define float @PR37890_v16f32(<16 x float> %a) {
; SSE2-LABEL: PR37890_v16f32:
; SSE2: # %bb.0:
; SSE2-NEXT: addps %xmm3, %xmm1
; SSE2-NEXT: addps %xmm2, %xmm1
; SSE2-NEXT: addps %xmm0, %xmm1
; SSE2-NEXT: movaps %xmm1, %xmm2
; SSE2-NEXT: unpckhpd {{.*#+}} xmm2 = xmm2[1],xmm1[1]
; SSE2-NEXT: addps %xmm1, %xmm2
; SSE2-NEXT: movaps %xmm2, %xmm0
; SSE2-NEXT: shufps {{.*#+}} xmm0 = xmm0[1,1],xmm2[1,1]
; SSE2-NEXT: addss %xmm2, %xmm0
; SSE2-NEXT: retq
;
; SSSE3-SLOW-LABEL: PR37890_v16f32:
; SSSE3-SLOW: # %bb.0:
; SSSE3-SLOW-NEXT: addps %xmm3, %xmm1
; SSSE3-SLOW-NEXT: addps %xmm2, %xmm1
; SSSE3-SLOW-NEXT: addps %xmm0, %xmm1
; SSSE3-SLOW-NEXT: movaps %xmm1, %xmm2
; SSSE3-SLOW-NEXT: unpckhpd {{.*#+}} xmm2 = xmm2[1],xmm1[1]
; SSSE3-SLOW-NEXT: addps %xmm1, %xmm2
; SSSE3-SLOW-NEXT: movshdup {{.*#+}} xmm0 = xmm2[1,1,3,3]
; SSSE3-SLOW-NEXT: addss %xmm2, %xmm0
; SSSE3-SLOW-NEXT: retq
;
; SSSE3-FAST-LABEL: PR37890_v16f32:
; SSSE3-FAST: # %bb.0:
; SSSE3-FAST-NEXT: addps %xmm3, %xmm1
; SSSE3-FAST-NEXT: addps %xmm2, %xmm1
; SSSE3-FAST-NEXT: addps %xmm0, %xmm1
; SSSE3-FAST-NEXT: movaps %xmm1, %xmm0
; SSSE3-FAST-NEXT: unpckhpd {{.*#+}} xmm0 = xmm0[1],xmm1[1]
; SSSE3-FAST-NEXT: addps %xmm1, %xmm0
; SSSE3-FAST-NEXT: haddps %xmm0, %xmm0
; SSSE3-FAST-NEXT: retq
;
; AVX1-SLOW-LABEL: PR37890_v16f32:
; AVX1-SLOW: # %bb.0:
; AVX1-SLOW-NEXT: vaddps %ymm1, %ymm0, %ymm0
; AVX1-SLOW-NEXT: vextractf128 $1, %ymm0, %xmm1
; AVX1-SLOW-NEXT: vaddps %xmm1, %xmm0, %xmm0
; AVX1-SLOW-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
; AVX1-SLOW-NEXT: vaddps %xmm1, %xmm0, %xmm0
; AVX1-SLOW-NEXT: vmovshdup {{.*#+}} xmm1 = xmm0[1,1,3,3]
; AVX1-SLOW-NEXT: vaddss %xmm1, %xmm0, %xmm0
; AVX1-SLOW-NEXT: vzeroupper
; AVX1-SLOW-NEXT: retq
;
; AVX1-FAST-LABEL: PR37890_v16f32:
; AVX1-FAST: # %bb.0:
; AVX1-FAST-NEXT: vaddps %ymm1, %ymm0, %ymm0
; AVX1-FAST-NEXT: vextractf128 $1, %ymm0, %xmm1
; AVX1-FAST-NEXT: vhaddps %xmm0, %xmm1, %xmm0
; AVX1-FAST-NEXT: vhaddps %xmm0, %xmm0, %xmm0
; AVX1-FAST-NEXT: vhaddps %xmm0, %xmm0, %xmm0
; AVX1-FAST-NEXT: vzeroupper
; AVX1-FAST-NEXT: retq
;
; AVX2-LABEL: PR37890_v16f32:
; AVX2: # %bb.0:
; AVX2-NEXT: vaddps %ymm1, %ymm0, %ymm0
; AVX2-NEXT: vextractf128 $1, %ymm0, %xmm1
; AVX2-NEXT: vaddps %xmm1, %xmm0, %xmm0
; AVX2-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
; AVX2-NEXT: vaddps %xmm1, %xmm0, %xmm0
; AVX2-NEXT: vmovshdup {{.*#+}} xmm1 = xmm0[1,1,3,3]
; AVX2-NEXT: vaddss %xmm1, %xmm0, %xmm0
; AVX2-NEXT: vzeroupper
; AVX2-NEXT: retq
%hi0 = shufflevector <16 x float> %a, <16 x float> undef, <8 x i32> <i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
%lo0 = shufflevector <16 x float> %a, <16 x float> undef, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
%sum0 = fadd fast <8 x float> %lo0, %hi0
%hi1 = shufflevector <8 x float> %sum0, <8 x float> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
%lo1 = shufflevector <8 x float> %sum0, <8 x float> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
%sum1 = fadd fast <4 x float> %lo1, %hi1
%hi2 = shufflevector <4 x float> %sum1, <4 x float> undef, <2 x i32> <i32 2, i32 3>
%lo2 = shufflevector <4 x float> %sum1, <4 x float> undef, <2 x i32> <i32 0, i32 1>
%sum2 = fadd fast <2 x float> %lo2, %hi2
%hi3 = shufflevector <2 x float> %sum2, <2 x float> undef, <2 x i32> <i32 1, i32 undef>
%sum3 = fadd fast <2 x float> %sum2, %hi3
%e = extractelement <2 x float> %sum3, i32 0
ret float %e
}