LLVM AVX2 transpose 8x8

gistfile1.txt

; Further compiled with:
; clang -x ir -emit-llvm -S -mcpu=haswell -O3 - -o - | llc -O3 -mcpu=haswell - -o -

; ModuleID = 'LLVMDialectModule'
source_filename = "LLVMDialectModule"

declare i8* @malloc(i64)

declare void @free(i8*)

; Function Attrs: noinline
define void @transpose_2d_on_tensors(float* %0, float* %1, i64 %2, i64 %3, i64 %4, i64 %5, i64 %6, float* %7, float* %8, i64 %9, i64 %10, i64 %11, i64 %12, i64 %13) #0  {
  %15 = getelementptr float, float* %1, i64 0
  %16 = bitcast float* %15 to <8 x float>*
  %17 = load <8 x float>, <8 x float>* %16, align 4
  %18 = getelementptr float, float* %1, i64 8
  %19 = bitcast float* %18 to <8 x float>*
  %20 = load <8 x float>, <8 x float>* %19, align 4
  %21 = getelementptr float, float* %1, i64 16
  %22 = bitcast float* %21 to <8 x float>*
  %23 = load <8 x float>, <8 x float>* %22, align 4
  %24 = getelementptr float, float* %1, i64 24
  %25 = bitcast float* %24 to <8 x float>*
  %26 = load <8 x float>, <8 x float>* %25, align 4
  %27 = getelementptr float, float* %1, i64 32
  %28 = bitcast float* %27 to <8 x float>*
  %29 = load <8 x float>, <8 x float>* %28, align 4
  %30 = getelementptr float, float* %1, i64 40
  %31 = bitcast float* %30 to <8 x float>*
  %32 = load <8 x float>, <8 x float>* %31, align 4
  %33 = getelementptr float, float* %1, i64 48
  %34 = bitcast float* %33 to <8 x float>*
  %35 = load <8 x float>, <8 x float>* %34, align 4
  %36 = getelementptr float, float* %1, i64 56
  %37 = bitcast float* %36 to <8 x float>*
  %38 = load <8 x float>, <8 x float>* %37, align 4
  %39 = shufflevector <8 x float> %17, <8 x float> %20, <8 x i32> <i32 0, i32 8, i32 1, i32 9, i32 4, i32 12, i32 5, i32 13>
  %40 = shufflevector <8 x float> %17, <8 x float> %20, <8 x i32> <i32 2, i32 10, i32 3, i32 11, i32 6, i32 14, i32 7, i32 15>
  %41 = shufflevector <8 x float> %23, <8 x float> %26, <8 x i32> <i32 0, i32 8, i32 1, i32 9, i32 4, i32 12, i32 5, i32 13>
  %42 = shufflevector <8 x float> %23, <8 x float> %26, <8 x i32> <i32 2, i32 10, i32 3, i32 11, i32 6, i32 14, i32 7, i32 15>
  %43 = shufflevector <8 x float> %29, <8 x float> %32, <8 x i32> <i32 0, i32 8, i32 1, i32 9, i32 4, i32 12, i32 5, i32 13>
  %44 = shufflevector <8 x float> %29, <8 x float> %32, <8 x i32> <i32 2, i32 10, i32 3, i32 11, i32 6, i32 14, i32 7, i32 15>
  %45 = shufflevector <8 x float> %35, <8 x float> %38, <8 x i32> <i32 0, i32 8, i32 1, i32 9, i32 4, i32 12, i32 5, i32 13>
  %46 = shufflevector <8 x float> %35, <8 x float> %38, <8 x i32> <i32 2, i32 10, i32 3, i32 11, i32 6, i32 14, i32 7, i32 15>
  %47 = shufflevector <8 x float> %39, <8 x float> %41, <8 x i32> <i32 2, i32 3, i32 8, i32 9, i32 6, i32 7, i32 12, i32 13>
  %48 = shufflevector <8 x float> %39, <8 x float> %47, <8 x i32> <i32 0, i32 1, i32 10, i32 11, i32 4, i32 5, i32 14, i32 15>
  %49 = shufflevector <8 x float> %41, <8 x float> %47, <8 x i32> <i32 8, i32 9, i32 2, i32 3, i32 12, i32 13, i32 6, i32 7>
  %50 = shufflevector <8 x float> %40, <8 x float> %42, <8 x i32> <i32 2, i32 3, i32 8, i32 9, i32 6, i32 7, i32 12, i32 13>
  %51 = shufflevector <8 x float> %40, <8 x float> %50, <8 x i32> <i32 0, i32 1, i32 10, i32 11, i32 4, i32 5, i32 14, i32 15>
  %52 = shufflevector <8 x float> %42, <8 x float> %50, <8 x i32> <i32 8, i32 9, i32 2, i32 3, i32 12, i32 13, i32 6, i32 7>
  %53 = shufflevector <8 x float> %43, <8 x float> %45, <8 x i32> <i32 2, i32 3, i32 8, i32 9, i32 6, i32 7, i32 12, i32 13>
  %54 = shufflevector <8 x float> %43, <8 x float> %53, <8 x i32> <i32 0, i32 1, i32 10, i32 11, i32 4, i32 5, i32 14, i32 15>
  %55 = shufflevector <8 x float> %45, <8 x float> %53, <8 x i32> <i32 8, i32 9, i32 2, i32 3, i32 12, i32 13, i32 6, i32 7>
  %56 = shufflevector <8 x float> %44, <8 x float> %46, <8 x i32> <i32 2, i32 3, i32 8, i32 9, i32 6, i32 7, i32 12, i32 13>
  %57 = shufflevector <8 x float> %44, <8 x float> %56, <8 x i32> <i32 0, i32 1, i32 10, i32 11, i32 4, i32 5, i32 14, i32 15>
  %58 = shufflevector <8 x float> %46, <8 x float> %56, <8 x i32> <i32 8, i32 9, i32 2, i32 3, i32 12, i32 13, i32 6, i32 7>
  %59 = shufflevector <8 x float> %48, <8 x float> %54, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 8, i32 9, i32 10, i32 11>
  %60 = shufflevector <8 x float> %49, <8 x float> %55, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 8, i32 9, i32 10, i32 11>
  %61 = shufflevector <8 x float> %51, <8 x float> %57, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 8, i32 9, i32 10, i32 11>
  %62 = shufflevector <8 x float> %52, <8 x float> %58, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 8, i32 9, i32 10, i32 11>
  %63 = shufflevector <8 x float> %48, <8 x float> %54, <8 x i32> <i32 4, i32 5, i32 6, i32 7, i32 12, i32 13, i32 14, i32 15>
  %64 = shufflevector <8 x float> %49, <8 x float> %55, <8 x i32> <i32 4, i32 5, i32 6, i32 7, i32 12, i32 13, i32 14, i32 15>
  %65 = shufflevector <8 x float> %51, <8 x float> %57, <8 x i32> <i32 4, i32 5, i32 6, i32 7, i32 12, i32 13, i32 14, i32 15>
  %66 = shufflevector <8 x float> %52, <8 x float> %58, <8 x i32> <i32 4, i32 5, i32 6, i32 7, i32 12, i32 13, i32 14, i32 15>
  %67 = getelementptr float, float* %8, i64 0
  %68 = bitcast float* %67 to <8 x float>*
  store <8 x float> %59, <8 x float>* %68, align 4
  %69 = getelementptr float, float* %8, i64 8
  %70 = bitcast float* %69 to <8 x float>*
  store <8 x float> %60, <8 x float>* %70, align 4
  %71 = getelementptr float, float* %8, i64 16
  %72 = bitcast float* %71 to <8 x float>*
  store <8 x float> %61, <8 x float>* %72, align 4
  %73 = getelementptr float, float* %8, i64 24
  %74 = bitcast float* %73 to <8 x float>*
  store <8 x float> %62, <8 x float>* %74, align 4
  %75 = getelementptr float, float* %8, i64 32
  %76 = bitcast float* %75 to <8 x float>*
  store <8 x float> %63, <8 x float>* %76, align 4
  %77 = getelementptr float, float* %8, i64 40
  %78 = bitcast float* %77 to <8 x float>*
  store <8 x float> %64, <8 x float>* %78, align 4
  %79 = getelementptr float, float* %8, i64 48
  %80 = bitcast float* %79 to <8 x float>*
  store <8 x float> %65, <8 x float>* %80, align 4
  %81 = getelementptr float, float* %8, i64 56
  %82 = bitcast float* %81 to <8 x float>*
  store <8 x float> %66, <8 x float>* %82, align 4
  ret void
}