Thanks for using Compiler Explorer
Sponsors
Jakt
C++
Ada
Algol68
Analysis
Android Java
Android Kotlin
Assembly
C
C3
Carbon
C with Coccinelle
C++ with Coccinelle
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
Mojo
Nim
Numba
Nix
Objective-C
Objective-C++
OCaml
Odin
OpenCL C
Pascal
Pony
PTX
Python
Racket
Raku
Ruby
Rust
Sail
Snowball
Scala
Slang
Solidity
Spice
SPIR-V
Swift
LLVM TableGen
Toit
Triton
TypeScript Native
V
Vala
Visual Basic
Vyper
WASM
Zig
Javascript
GIMPLE
Ygen
sway
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 20.1.0
clang 21.1.0
clang 4.0.1
clang 5.0.0
clang 6.0.0
clang 7.0.0
clang 8.0.0
clang 9.0.0
hexagon-clang 16.0.5
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 20.1.0
llc 21.1.0
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 20.1.0
opt 21.1.0
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 @julia_mycdot_10439([2 x double]* noalias nocapture sret %0, { {}*, i8, i8 }* nocapture nonnull readonly align 8 dereferenceable(16) %1, { {}*, i8, i8 }* nocapture nonnull readonly align 8 dereferenceable(16) %2) { top: %3 = bitcast { {}*, i8, i8 }* %1 to {}*** %4 = load atomic {}**, {}*** %3 unordered, align 8 %5 = getelementptr inbounds {}*, {}** %4, i64 3 %6 = bitcast {}** %5 to i64* %7 = load i64, i64* %6, align 8 %8 = bitcast {}** %4 to i64* %9 = load i64, i64* %8, align 8 %10 = bitcast { {}*, i8, i8 }* %2 to i64** %11 = load atomic i64*, i64** %10 unordered, align 8 %12 = load i64, i64* %11, align 8 %13 = icmp ne i64 %7, 0 call void @llvm.assume(i1 %13) %ptr.0.i171 = inttoptr i64 %9 to double* %res.i174 = shl nuw i64 %7, 1 %indname.i = add i64 %res.i174, -32 %ptr.1.i172 = getelementptr inbounds double, double* %ptr.0.i171, i64 %indname.i %ptr.1.i169 = getelementptr inbounds double, double* %ptr.0.i171, i64 %res.i174 %ptr.2.i170 = ptrtoint double* %ptr.1.i169 to i64 %ptr.1.i166 = getelementptr inbounds double, double* %ptr.1.i172, i64 16 %ptr.2.i167 = ptrtoint double* %ptr.1.i166 to i64 %14 = icmp ult i64 %7, 16 br i1 %14, label %L129, label %L51 L51: ; preds = %L51, %top %value_phi = phi i64 [ %ptr.2.i132, %L51 ], [ %9, %top ] %value_phi8 = phi i64 [ %ptr.2.i129, %L51 ], [ %12, %top ] %value_phi14 = phi <8 x double> [ %res.i140, %L51 ], [ zeroinitializer, %top ] %value_phi15 = phi <8 x double> [ %res.i139, %L51 ], [ zeroinitializer, %top ] %value_phi18 = phi <8 x double> [ %res.i134, %L51 ], [ zeroinitializer, %top ] %value_phi19 = phi <8 x double> [ %res.i133, %L51 ], [ zeroinitializer, %top ] %ptr.1.i163 = inttoptr i64 %value_phi to <16 x double>* %res.i164 = load <16 x double>, <16 x double>* %ptr.1.i163, align 8 %res.i162 = shufflevector <16 x double> %res.i164, <16 x double> undef, <8 x i32> <i32 0, i32 2, i32 4, i32 6, i32 8, i32 10, i32 12, i32 14> %res.i161 = shufflevector <16 x double> %res.i164, <16 x double> undef, <8 x i32> <i32 1, i32 3, i32 5, i32 7, i32 9, i32 11, i32 13, i32 15> %ptr.0.i158 = inttoptr i64 %value_phi to double* %ptr.1.i159 = getelementptr inbounds double, double* %ptr.0.i158, i64 16 %ptr.1.i156 = bitcast double* %ptr.1.i159 to <16 x double>* %res.i157 = load <16 x double>, <16 x double>* %ptr.1.i156, align 8 %res.i155 = shufflevector <16 x double> %res.i157, <16 x double> undef, <8 x i32> <i32 0, i32 2, i32 4, i32 6, i32 8, i32 10, i32 12, i32 14> %res.i154 = shufflevector <16 x double> %res.i157, <16 x double> undef, <8 x i32> <i32 1, i32 3, i32 5, i32 7, i32 9, i32 11, i32 13, i32 15> %ptr.1.i152 = inttoptr i64 %value_phi8 to <16 x double>* %res.i153 = load <16 x double>, <16 x double>* %ptr.1.i152, align 8 %res.i151 = shufflevector <16 x double> %res.i153, <16 x double> undef, <8 x i32> <i32 0, i32 2, i32 4, i32 6, i32 8, i32 10, i32 12, i32 14> %res.i150 = shufflevector <16 x double> %res.i153, <16 x double> undef, <8 x i32> <i32 1, i32 3, i32 5, i32 7, i32 9, i32 11, i32 13, i32 15> %ptr.0.i147 = inttoptr i64 %value_phi8 to double* %ptr.1.i148 = getelementptr inbounds double, double* %ptr.0.i147, i64 16 %ptr.1.i145 = bitcast double* %ptr.1.i148 to <16 x double>* %res.i146 = load <16 x double>, <16 x double>* %ptr.1.i145, align 8 %res.i144 = shufflevector <16 x double> %res.i146, <16 x double> undef, <8 x i32> <i32 0, i32 2, i32 4, i32 6, i32 8, i32 10, i32 12, i32 14> %res.i143 = shufflevector <16 x double> %res.i146, <16 x double> undef, <8 x i32> <i32 1, i32 3, i32 5, i32 7, i32 9, i32 11, i32 13, i32 15> %res.i142 = call reassoc nsz arcp contract afn <8 x double> @llvm.fma.v8f64(<8 x double> %res.i161, <8 x double> %res.i150, <8 x double> %value_phi14) %res.i141 = call reassoc nsz arcp contract afn <8 x double> @llvm.fma.v8f64(<8 x double> %res.i154, <8 x double> %res.i143, <8 x double> %value_phi15) %res.i140 = call reassoc nsz arcp contract afn <8 x double> @llvm.fma.v8f64(<8 x double> %res.i162, <8 x double> %res.i151, <8 x double> %res.i142) %res.i139 = call reassoc nsz arcp contract afn <8 x double> @llvm.fma.v8f64(<8 x double> %res.i155, <8 x double> %res.i144, <8 x double> %res.i141) %res.i138 = fneg reassoc nsz arcp contract afn <8 x double> %res.i161 %res.i137 = call reassoc nsz arcp contract afn <8 x double> @llvm.fma.v8f64(<8 x double> %res.i138, <8 x double> %res.i151, <8 x double> %value_phi18) %res.i136 = fneg reassoc nsz arcp contract afn <8 x double> %res.i154 %res.i135 = call reassoc nsz arcp contract afn <8 x double> @llvm.fma.v8f64(<8 x double> %res.i136, <8 x double> %res.i144, <8 x double> %value_phi19) %res.i134 = call reassoc nsz arcp contract afn <8 x double> @llvm.fma.v8f64(<8 x double> %res.i162, <8 x double> %res.i150, <8 x double> %res.i137) %res.i133 = call reassoc nsz arcp contract afn <8 x double> @llvm.fma.v8f64(<8 x double> %res.i155, <8 x double> %res.i143, <8 x double> %res.i135) %ptr.1.i131 = getelementptr inbounds double, double* %ptr.0.i158, i64 32 %ptr.2.i132 = ptrtoint double* %ptr.1.i131 to i64 %ptr.1.i128 = getelementptr inbounds double, double* %ptr.0.i147, i64 32 %ptr.2.i129 = ptrtoint double* %ptr.1.i128 to i64 %.not = icmp ugt double* %ptr.1.i131, %ptr.1.i172 br i1 %.not, label %L110, label %L51 L110: ; preds = %L51 %res.i126 = fadd nsz contract <8 x double> %res.i140, %res.i139 %res.i125 = fadd nsz contract <8 x double> %res.i134, %res.i133 br label %L129 L129: ; preds = %L110, %top %value_phi36 = phi i64 [ %ptr.2.i132, %L110 ], [ %9, %top ] %value_phi42 = phi i64 [ %ptr.2.i129, %L110 ], [ %12, %top ] %value_phi46 = phi <8 x double> [ %res.i126, %L110 ], [ zeroinitializer, %top ] %value_phi47 = phi <8 x double> [ %res.i125, %L110 ], [ zeroinitializer, %top ] %value_phi42.ptr = inttoptr i64 %value_phi42 to double* %.not175177 = icmp ugt i64 %value_phi36, %ptr.2.i167 br i1 %.not175177, label %L187, label %L173 L173: ; preds = %L173, %L129 %value_phi68181 = phi i64 [ %ptr.2.i108, %L173 ], [ %value_phi42, %L129 ] %value_phi62180 = phi i64 [ %ptr.2.i111, %L173 ], [ %value_phi36, %L129 ] %value_phi58179 = phi <8 x double> [ %res.i112, %L173 ], [ %value_phi47, %L129 ] %value_phi54178 = phi <8 x double> [ %res.i115, %L173 ], [ %value_phi46, %L129 ] %ptr.1.i123 = inttoptr i64 %value_phi62180 to <16 x double>* %res.i124 = load <16 x double>, <16 x double>* %ptr.1.i123, align 8 %res.i122 = shufflevector <16 x double> %res.i124, <16 x double> undef, <8 x i32> <i32 0, i32 2, i32 4, i32 6, i32 8, i32 10, i32 12, i32 14> %res.i121 = shufflevector <16 x double> %res.i124, <16 x double> undef, <8 x i32> <i32 1, i32 3, i32 5, i32 7, i32 9, i32 11, i32 13, i32 15> %ptr.1.i119 = inttoptr i64 %value_phi68181 to <16 x double>* %res.i120 = load <16 x double>, <16 x double>* %ptr.1.i119, align 8 %res.i118 = shufflevector <16 x double> %res.i120, <16 x double> undef, <8 x i32> <i32 0, i32 2, i32 4, i32 6, i32 8, i32 10, i32 12, i32 14> %res.i117 = shufflevector <16 x double> %res.i120, <16 x double> undef, <8 x i32> <i32 1, i32 3, i32 5, i32 7, i32 9, i32 11, i32 13, i32 15> %res.i116 = call reassoc nsz arcp contract afn <8 x double> @llvm.fma.v8f64(<8 x double> %res.i121, <8 x double> %res.i117, <8 x double> %value_phi54178) %res.i115 = call reassoc nsz arcp contract afn <8 x double> @llvm.fma.v8f64(<8 x double> %res.i122, <8 x double> %res.i118, <8 x double> %res.i116) %res.i114 = fneg reassoc nsz arcp contract afn <8 x double> %res.i121 %res.i113 = call reassoc nsz arcp contract afn <8 x double> @llvm.fma.v8f64(<8 x double> %res.i114, <8 x double> %res.i118, <8 x double> %value_phi58179) %res.i112 = call reassoc nsz arcp contract afn <8 x double> @llvm.fma.v8f64(<8 x double> %res.i122, <8 x double> %res.i117, <8 x double> %res.i113) %ptr.0.i109 = inttoptr i64 %value_phi62180 to double* %ptr.1.i110 = getelementptr inbounds double, double* %ptr.0.i109, i64 16 %ptr.2.i111 = ptrtoint double* %ptr.1.i110 to i64 %ptr.0.i106 = inttoptr i64 %value_phi68181 to double* %ptr.1.i107 = getelementptr inbounds double, double* %ptr.0.i106, i64 16 %ptr.2.i108 = ptrtoint double* %ptr.1.i107 to i64 %.not175 = icmp ult double* %ptr.1.i172, %ptr.0.i109 br i1 %.not175, label %L187, label %L173 L187: ; preds = %L173, %L129 %value_phi54.lcssa = phi <8 x double> [ %value_phi46, %L129 ], [ %res.i115, %L173 ] %value_phi58.lcssa = phi <8 x double> [ %value_phi47, %L129 ], [ %res.i112, %L173 ] %value_phi62.lcssa = phi i64 [ %value_phi36, %L129 ], [ %ptr.2.i111, %L173 ] %value_phi68.lcssa.ptr = phi double* [ %value_phi42.ptr, %L129 ], [ %ptr.1.i107, %L173 ] %.not176 = icmp ult i64 %value_phi62.lcssa, %ptr.2.i170 br i1 %.not176, label %L191, label %L219 L191: ; preds = %L187 %15 = trunc i64 %7 to i8 %16 = add i8 %15, 7 %17 = and i8 %16, 7 %18 = xor i8 %17, 7 %19 = lshr i8 -1, %18 %ptr.0.i101 = inttoptr i64 %value_phi62.lcssa to double* %ptr.1.i102 = getelementptr inbounds double, double* %ptr.0.i101, <8 x i64> <i64 0, i64 2, i64 4, i64 6, i64 8, i64 10, i64 12, i64 14> %mask.0.i103 = bitcast i8 %19 to <8 x i1> %res.i104 = call <8 x double> @llvm.masked.gather.v8f64.v8p0f64(<8 x double*> %ptr.1.i102, i32 8, <8 x i1> %mask.0.i103, <8 x double> zeroinitializer) %ptr.1.i97 = getelementptr inbounds double, double* %ptr.0.i101, i64 1 %ptr.2.i98 = getelementptr inbounds double, double* %ptr.1.i97, <8 x i64> <i64 0, i64 2, i64 4, i64 6, i64 8, i64 10, i64 12, i64 14> %res.i100 = call <8 x double> @llvm.masked.gather.v8f64.v8p0f64(<8 x double*> %ptr.2.i98, i32 8, <8 x i1> %mask.0.i103, <8 x double> zeroinitializer) %ptr.1.i93 = getelementptr inbounds double, double* %value_phi68.lcssa.ptr, <8 x i64> <i64 0, i64 2, i64 4, i64 6, i64 8, i64 10, i64 12, i64 14> %res.i95 = call <8 x double> @llvm.masked.gather.v8f64.v8p0f64(<8 x double*> %ptr.1.i93, i32 8, <8 x i1> %mask.0.i103, <8 x double> zeroinitializer) %ptr.1.i88 = getelementptr inbounds double, double* %value_phi68.lcssa.ptr, i64 1 %ptr.2.i89 = getelementptr inbounds double, double* %ptr.1.i88, <8 x i64> <i64 0, i64 2, i64 4, i64 6, i64 8, i64 10, i64 12, i64 14> %res.i91 = call <8 x double> @llvm.masked.gather.v8f64.v8p0f64(<8 x double*> %ptr.2.i89, i32 8, <8 x i1> %mask.0.i103, <8 x double> zeroinitializer) %res.i86 = call reassoc nsz arcp contract afn <8 x double> @llvm.fma.v8f64(<8 x double> %res.i100, <8 x double> %res.i91, <8 x double> %value_phi54.lcssa) %res.i85 = call reassoc nsz arcp contract afn <8 x double> @llvm.fma.v8f64(<8 x double> %res.i104, <8 x double> %res.i95, <8 x double> %res.i86) %res.i84 = select reassoc nsz arcp contract <8 x i1> %mask.0.i103, <8 x double> %res.i85, <8 x double> %value_phi54.lcssa %res.i82 = fneg reassoc nsz arcp contract afn <8 x double> %res.i100 %res.i81 = call reassoc nsz arcp contract afn <8 x double> @llvm.fma.v8f64(<8 x double> %res.i82, <8 x double> %res.i95, <8 x double> %value_phi58.lcssa) %res.i80 = call reassoc nsz arcp contract afn <8 x double> @llvm.fma.v8f64(<8 x double> %res.i104, <8 x double> %res.i91, <8 x double> %res.i81) %res.i79 = select reassoc nsz arcp contract <8 x i1> %mask.0.i103, <8 x double> %res.i80, <8 x double> %value_phi58.lcssa br label %L219 L219: ; preds = %L191, %L187 %value_phi72 = phi <8 x double> [ %res.i84, %L191 ], [ %value_phi54.lcssa, %L187 ] %value_phi73 = phi <8 x double> [ %res.i79, %L191 ], [ %value_phi58.lcssa, %L187 ] %res.i75 = call reassoc nsz arcp contract afn double @llvm.experimental.vector.reduce.v2.fadd.f64.v8f64(double 0.000000e+00, <8 x double> %value_phi72) %res.i = call reassoc nsz arcp contract afn double @llvm.experimental.vector.reduce.v2.fadd.f64.v8f64(double 0.000000e+00, <8 x double> %value_phi73) %.sroa.0.0..sroa_idx = getelementptr inbounds [2 x double], [2 x double]* %0, i64 0, i64 0 store double %res.i75, double* %.sroa.0.0..sroa_idx, align 8 %.sroa.2.0..sroa_idx74 = getelementptr inbounds [2 x double], [2 x double]* %0, i64 0, i64 1 store double %res.i, double* %.sroa.2.0..sroa_idx74, align 8 ret void } declare void @llvm.assume(i1) #4 ; Function Attrs: nounwind readnone speculatable willreturn declare <8 x double> @llvm.fma.v8f64(<8 x double>, <8 x double>, <8 x double>) #5 ; Function Attrs: nounwind readonly willreturn declare <8 x double> @llvm.masked.gather.v8f64.v8p0f64(<8 x double*>, i32 immarg, <8 x i1>, <8 x double>) #6 ; Function Attrs: nounwind readnone willreturn declare double @llvm.experimental.vector.reduce.v2.fadd.f64.v8f64(double, <8 x double>) #7
Become a Patron
Sponsor on GitHub
Donate via PayPal
Compiler Explorer Shop
Source on GitHub
Mailing list
Installed libraries
Wiki
Report an issue
How it works
Contact the author
CE on Mastodon
CE on Bluesky
Statistics
Changelog
Version tree