Thanks for using Compiler Explorer
Sponsors
Jakt
C++
Ada
Analysis
Android Java
Android Kotlin
Assembly
C
C3
Carbon
C++ (Circle)
CIRCT
Clean
CMake
CMakeScript
COBOL
C++ for OpenCL
MLIR
Cppx
Cppx-Blue
Cppx-Gold
Cpp2-cppfront
Crystal
C#
CUDA C++
D
Dart
Elixir
Erlang
Fortran
F#
GLSL
Go
Haskell
HLSL
Hook
Hylo
IL
ispc
Java
Julia
Kotlin
LLVM IR
LLVM MIR
Modula-2
Nim
Objective-C
Objective-C++
OCaml
OpenCL C
Pascal
Pony
Python
Racket
Ruby
Rust
Snowball
Scala
Solidity
Spice
SPIR-V
Swift
LLVM TableGen
Toit
TypeScript Native
V
Vala
Visual Basic
WASM
Zig
Javascript
GIMPLE
Ygen
llvm source #1
Output
Compile to binary object
Link to binary
Execute the code
Intel asm syntax
Demangle identifiers
Verbose demangling
Filters
Unused labels
Library functions
Directives
Comments
Horizontal whitespace
Debug intrinsics
Compiler
clang (assertions trunk)
clang (trunk)
clang 10.0.0
clang 10.0.1
clang 11.0.0
clang 11.0.1
clang 12.0.0
clang 12.0.1
clang 13.0.0
clang 14.0.0
clang 15.0.0
clang 16.0.0
clang 17.0.1
clang 18.1.0
clang 19.1.0
clang 4.0.1
clang 5.0.0
clang 6.0.0
clang 7.0.0
clang 8.0.0
clang 9.0.0
llc (assertions trunk)
llc (trunk)
llc 10.0.0
llc 10.0.1
llc 11.0.0
llc 11.0.1
llc 12.0.0
llc 12.0.1
llc 13.0.0
llc 14.0.0
llc 15.0.0
llc 16.0.0
llc 17.0.1
llc 18.1.0
llc 19.1.0
llc 3.2
llc 3.3
llc 3.9.1
llc 4.0.0
llc 4.0.1
llc 5.0.0
llc 6.0.0
llc 7.0.0
llc 8.0.0
llc 9.0.0
opt (assertions trunk)
opt (trunk)
opt 10.0.0
opt 10.0.1
opt 11.0.0
opt 11.0.1
opt 12.0.0
opt 12.0.1
opt 13.0.0
opt 14.0.0
opt 15.0.0
opt 16.0.0
opt 17.0.1
opt 18.1.0
opt 19.1.0
opt 3.2
opt 3.3
opt 3.9.1
opt 4.0.0
opt 4.0.1
opt 5.0.0
opt 6.0.0
opt 7.0.0
opt 8.0.0
opt 9.0.0
Options
Source code
define void @store_i32_stride6_vf8(<8 x i32>* %in.vecptr0, <8 x i32>* %in.vecptr1, <8 x i32>* %in.vecptr2, <8 x i32>* %in.vecptr3, <8 x i32>* %in.vecptr4, <8 x i32>* %in.vecptr5, <48 x i32>* %out.vec) nounwind { %in.vec0 = load <8 x i32>, <8 x i32>* %in.vecptr0, align 32 %in.vec1 = load <8 x i32>, <8 x i32>* %in.vecptr1, align 32 %in.vec2 = load <8 x i32>, <8 x i32>* %in.vecptr2, align 32 %in.vec3 = load <8 x i32>, <8 x i32>* %in.vecptr3, align 32 %in.vec4 = load <8 x i32>, <8 x i32>* %in.vecptr4, align 32 %in.vec5 = load <8 x i32>, <8 x i32>* %in.vecptr5, align 32 tail call void asm sideeffect "# LLVM-MCA-BEGIN", "~{dirflag},~{fpsr},~{flags}"() %concat01 = shufflevector <8 x i32> %in.vec0, <8 x i32> %in.vec1, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15> %concat23 = shufflevector <8 x i32> %in.vec2, <8 x i32> %in.vec3, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15> %concat45 = shufflevector <8 x i32> %in.vec4, <8 x i32> %in.vec5, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15> %concat0123 = shufflevector <16 x i32> %concat01, <16 x i32> %concat23, <32 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 16, i32 17, i32 18, i32 19, i32 20, i32 21, i32 22, i32 23, i32 24, i32 25, i32 26, i32 27, i32 28, i32 29, i32 30, i32 31> %concat45uu = shufflevector <16 x i32> %concat45, <16 x i32> poison, <32 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef> %concat012345 = shufflevector <32 x i32> %concat0123, <32 x i32> %concat45uu, <48 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 16, i32 17, i32 18, i32 19, i32 20, i32 21, i32 22, i32 23, i32 24, i32 25, i32 26, i32 27, i32 28, i32 29, i32 30, i32 31, i32 32, i32 33, i32 34, i32 35, i32 36, i32 37, i32 38, i32 39, i32 40, i32 41, i32 42, i32 43, i32 44, i32 45, i32 46, i32 47> %interleaved.vec = shufflevector <48 x i32> %concat012345, <48 x i32> poison, <48 x i32> <i32 0, i32 8, i32 16, i32 24, i32 32, i32 40, i32 1, i32 9, i32 17, i32 25, i32 33, i32 41, i32 2, i32 10, i32 18, i32 26, i32 34, i32 42, i32 3, i32 11, i32 19, i32 27, i32 35, i32 43, i32 4, i32 12, i32 20, i32 28, i32 36, i32 44, i32 5, i32 13, i32 21, i32 29, i32 37, i32 45, i32 6, i32 14, i32 22, i32 30, i32 38, i32 46, i32 7, i32 15, i32 23, i32 31, i32 39, i32 47> tail call void asm sideeffect "# LLVM-MCA-END", "~{dirflag},~{fpsr},~{flags}"() store <48 x i32> %interleaved.vec, <48 x i32>* %out.vec, align 32 ret void }
analysis source #2
Output
Compile to binary object
Link to binary
Execute the code
Intel asm syntax
Demangle identifiers
Verbose demangling
Filters
Unused labels
Library functions
Directives
Comments
Horizontal whitespace
Debug intrinsics
Compiler
OSACA (0.6.1)
llvm-mca (assertions trunk)
llvm-mca (trunk)
Options
Source code
vbroadcastss xmm11, xmm15 vunpcklps xmm10, xmm0, xmm1 # xmm10 = xmm0[0],xmm1[0],xmm0[1],xmm1[1] vpermilps xmm3, xmm7, 164 # xmm3 = xmm7[0,1,2,2] vpermilps xmm2, xmm6, 229 # xmm2 = xmm6[1,1,2,3] vblendps xmm2, xmm2, xmm3, 10 # xmm2 = xmm2[0],xmm3[1],xmm2[2],xmm3[3] vpermpd ymm2, ymm2, 100 # ymm2 = ymm2[0,1,2,1] vblendps ymm2, ymm2, ymm10, 12 # ymm2 = ymm2[0,1],ymm10[2,3],ymm2[4,5,6,7] vunpckhps xmm12, xmm5, xmm4 # xmm12 = xmm5[2],xmm4[2],xmm5[3],xmm4[3] vinsertf128 ymm3, ymm0, xmm12, 1 vblendps ymm8, ymm2, ymm3, 48 # ymm8 = ymm2[0,1,2,3],ymm3[4,5],ymm2[6,7] vbroadcastss xmm2, xmm2 vunpcklps xmm2, xmm2, xmm11 # xmm2 = xmm2[0],xmm11[0],xmm2[1],xmm11[1] vunpcklps xmm4, xmm5, xmm4 # xmm4 = xmm5[0],xmm4[0],xmm5[1],xmm4[1] vpermpd ymm4, ymm4, 100 # ymm4 = ymm4[0,1,2,1] vblendps ymm2, ymm4, ymm2, 12 # ymm2 = ymm4[0,1],ymm2[2,3],ymm4[4,5,6,7] vinsertf128 ymm10, ymm0, xmm10, 1 vblendps ymm11, ymm2, ymm10, 48 # ymm11 = ymm2[0,1,2,3],ymm10[4,5],ymm2[6,7] vunpckhps xmm0, xmm0, xmm1 # xmm0 = xmm0[2],xmm1[2],xmm0[3],xmm1[3] vpermpd ymm0, ymm0, 100 # ymm0 = ymm0[0,1,2,1] vblendps ymm0, ymm0, ymm12, 12 # ymm0 = ymm0[0,1],ymm12[2,3],ymm0[4,5,6,7] vpmovzxdq xmm2, xmm2 # xmm2 = mem[0],zero,mem[1],zero vunpckhps xmm6, xmm6, xmm7 # xmm6 = xmm6[2],xmm7[2],xmm6[3],xmm7[3] vpbroadcastd ymm7, xmm14 vpermilps xmm6, xmm6, 238 # xmm6 = xmm6[2,3,2,3] vinsertf128 ymm6, ymm0, xmm6, 1 vblendps ymm0, ymm0, ymm6, 48 # ymm0 = ymm0[0,1,2,3],ymm6[4,5],ymm0[6,7] vpblendd xmm2, xmm2, xmm7, 8 # xmm2 = xmm2[0,1,2],xmm7[3] vpshufd ymm6, ymm4, 164 # ymm6 = ymm4[0,1,2,2,4,5,6,6] vpshufd ymm7, ymm5, 229 # ymm7 = ymm5[1,1,2,3,5,5,6,7] vpblendd ymm6, ymm7, ymm6, 170 # ymm6 = ymm7[0],ymm6[1],ymm7[2],ymm6[3],ymm7[4],ymm6[5],ymm7[6],ymm6[7] vpermq ymm6, ymm6, 230 # ymm6 = ymm6[2,1,2,3] vpblendd ymm2, ymm6, ymm2, 12 # ymm2 = ymm6[0,1],ymm2[2,3],ymm6[4,5,6,7] vpunpckhdq ymm6, ymm9, ymm3 # ymm6 = ymm9[2],ymm3[2],ymm9[3],ymm3[3],ymm9[6],ymm3[6],ymm9[7],ymm3[7] vpblendd ymm2, ymm2, ymm6, 48 # ymm2 = ymm2[0,1,2,3],ymm6[4,5],ymm2[6,7] vextracti128 xmm6, ymm6, 1 vpunpckhdq ymm7, ymm10, ymm1 # ymm7 = ymm10[2],ymm1[2],ymm10[3],ymm1[3],ymm10[6],ymm1[6],ymm10[7],ymm1[7] vpermq ymm7, ymm7, 230 # ymm7 = ymm7[2,1,2,3] vpblendd ymm6, ymm7, ymm6, 12 # ymm6 = ymm7[0,1],ymm6[2,3],ymm7[4,5,6,7] vpunpckhdq ymm7, ymm5, ymm4 # ymm7 = ymm5[2],ymm4[2],ymm5[3],ymm4[3],ymm5[6],ymm4[6],ymm5[7],ymm4[7] vpshufd ymm7, ymm7, 238 # ymm7 = ymm7[2,3,2,3,6,7,6,7] vpblendd ymm6, ymm6, ymm7, 48 # ymm6 = ymm6[0,1,2,3],ymm7[4,5],ymm6[6,7] vpunpckldq ymm4, ymm5, ymm4 # ymm4 = ymm5[0],ymm4[0],ymm5[1],ymm4[1],ymm5[4],ymm4[4],ymm5[5],ymm4[5] vpunpckldq ymm3, ymm9, ymm3 # ymm3 = ymm9[0],ymm3[0],ymm9[1],ymm3[1],ymm9[4],ymm3[4],ymm9[5],ymm3[5] vpermq ymm4, ymm4, 170 # ymm4 = ymm4[2,2,2,2] vpermq ymm3, ymm3, 230 # ymm3 = ymm3[2,1,2,3] vpblendd ymm3, ymm3, ymm4, 12 # ymm3 = ymm3[0,1],ymm4[2,3],ymm3[4,5,6,7] vpunpckldq ymm1, ymm10, ymm1 # ymm1 = ymm10[0],ymm1[0],ymm10[1],ymm1[1],ymm10[4],ymm1[4],ymm10[5],ymm1[5] vpblendd ymm1, ymm3, ymm1, 48 # ymm1 = ymm3[0,1,2,3],ymm1[4,5],ymm3[6,7]
Become a Patron
Sponsor on GitHub
Donate via PayPal
Source on GitHub
Mailing list
Installed libraries
Wiki
Report an issue
How it works
Contact the author
CE on Mastodon
About the author
Statistics
Changelog
Version tree