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#
Go
Haskell
HLSL
Hook
Hylo
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
Swift
LLVM TableGen
Toit
TypeScript Native
V
Vala
Visual Basic
Zig
Javascript
GIMPLE
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 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 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 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 float @dot3_float3(float* dereferenceable(16) %a0, float* dereferenceable(16) %a1) { %bcx012 = bitcast float* %a0 to <3 x float>* %bcy012 = bitcast float* %a1 to <3 x float>* %x012 = load <3 x float>, <3 x float>* %bcx012, align 4 %y012 = load <3 x float>, <3 x float>* %bcy012, align 4 %mul012 = fmul <3 x float> %x012, %y012 %mul0 = extractelement <3 x float> %mul012, i32 0 %mul1 = extractelement <3 x float> %mul012, i32 1 %mul2 = extractelement <3 x float> %mul012, i32 2 %dot01 = fadd float %mul0, %mul1 %dot012 = fadd float %dot01, %mul2 ret float %dot012 }
llvm source #4
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 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 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 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 float @dot3_float3(float* dereferenceable(16) %a0, float* dereferenceable(16) %a1) { %bcx012 = bitcast float* %a0 to <4 x float>* %bcy012 = bitcast float* %a1 to <4 x float>* %x012.wide = load <4 x float>, <4 x float>* %bcx012, align 4 %y012.wide = load <4 x float>, <4 x float>* %bcy012, align 4 %x012 = shufflevector <4 x float> %x012.wide, <4 x float> undef, <3 x i32> <i32 0, i32 1, i32 2> %y012 = shufflevector <4 x float> %y012.wide, <4 x float> undef, <3 x i32> <i32 0, i32 1, i32 2> %mul012 = fmul <3 x float> %x012, %y012 %mul0 = extractelement <3 x float> %mul012, i32 0 %mul1 = extractelement <3 x float> %mul012, i32 1 %mul2 = extractelement <3 x float> %mul012, i32 2 %dot01 = fadd float %mul0, %mul1 %dot012 = fadd float %dot01, %mul2 ret float %dot012 }
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 (0.5.2)
llvm-mca (assertions trunk)
llvm-mca (trunk)
Options
Source code
vmovsd xmm0, qword ptr [rdi] # xmm0 = mem[0],zero vmovsd xmm1, qword ptr [rsi] # xmm1 = mem[0],zero vinsertps xmm0, xmm0, dword ptr [rdi + 8], 32 # xmm0 = xmm0[0,1],mem[0],xmm0[3] vinsertps xmm1, xmm1, dword ptr [rsi + 8], 32 # xmm1 = xmm1[0,1],mem[0],xmm1[3] vmulps xmm0, xmm0, xmm1 vmovshdup xmm1, xmm0 # xmm1 = xmm0[1,1,3,3] vpermilpd xmm2, xmm0, 1 # xmm2 = xmm0[1,0] vaddss xmm0, xmm0, xmm1 vaddss xmm0, xmm0, xmm2
analysis source #3
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 (0.5.2)
llvm-mca (assertions trunk)
llvm-mca (trunk)
Options
Source code
dot3_float3: # @dot3_float3 movups xmm0, xmmword ptr [rdi] movups xmm1, xmmword ptr [rsi] mulps xmm1, xmm0 movaps xmm0, xmm1 shufps xmm0, xmm1, 85 # xmm0 = xmm0[1,1],xmm1[1,1] addss xmm0, xmm1 movhlps xmm1, xmm1 # xmm1 = xmm1[1,1] addss xmm0, xmm1 ret
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