Skip to content

add intrinsics for portable packed simd vector reductions #48983

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
bors merged 10 commits into from
Mar 17, 2018
Merged

add intrinsics for portable packed simd vector reductions #48983

Changes from 1 commit
Commits
Show all changes
10 commits
Select commit Hold shift + click to select a range
01cc5b3
add intrinsics for portable packed simd vector reductions
gnzlbg Mar 13, 2018
07ce659
expose ordered/unordered/nanless intirnsics
gnzlbg Mar 14, 2018
51832c3
fix style
gnzlbg Mar 14, 2018
c990fa0
add dummy symbols for LLVM<6
gnzlbg Mar 14, 2018
3125a30
error on vector reduction usage if LLVM version is < 5.0
gnzlbg Mar 15, 2018
8478fa2
add min-llvm version to reduction tests
gnzlbg Mar 15, 2018
4fe6acf
add compile fail tests
gnzlbg Mar 15, 2018
19b81f6
error via bug! instead of stderr+terminate
gnzlbg Mar 15, 2018
f173a4c
add missing min-llvm-version
gnzlbg Mar 15, 2018
06148cb
ignore emscripten
gnzlbg Mar 16, 2018
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
47 changes: 47 additions & 0 deletions src/rustllvm/RustWrapper.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1442,4 +1442,51 @@ extern "C" LLVMValueRef
LLVMRustBuildVectorReduceFMax(LLVMBuilderRef B, LLVMValueRef Src, bool NoNaN) {
return wrap(unwrap(B)->CreateFPMaxReduce(unwrap(Src), NoNaN));
}

#else

extern "C" LLVMValueRef
LLVMRustBuildVectorReduceFAdd(LLVMBuilderRef, LLVMValueRef, LLVMValueRef Src) {
return Src;
}
extern "C" LLVMValueRef
LLVMRustBuildVectorReduceFMul(LLVMBuilderRef, LLVMValueRef, LLVMValueRef Src) {
return Src;

@hanna-kruppe hanna-kruppe Mar 14, 2018

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This is going to lead to broken IR (you'll wind up with vectors where scalars are expected). To truly support earlier LLVM versions, you'll need a polyfill that does the reduction "by hand".

}
extern "C" LLVMValueRef
LLVMRustBuildVectorReduceAdd(LLVMBuilderRef, LLVMValueRef Src) {
return Src;
}
extern "C" LLVMValueRef
LLVMRustBuildVectorReduceMul(LLVMBuilderRef, LLVMValueRef Src) {
return Src;
}
extern "C" LLVMValueRef
LLVMRustBuildVectorReduceAnd(LLVMBuilderRef, LLVMValueRef Src) {
return Src;
}
extern "C" LLVMValueRef
LLVMRustBuildVectorReduceOr(LLVMBuilderRef, LLVMValueRef Src) {
return Src;
}
extern "C" LLVMValueRef
LLVMRustBuildVectorReduceXor(LLVMBuilderRef, LLVMValueRef Src) {
return Src;
}
extern "C" LLVMValueRef
LLVMRustBuildVectorReduceMin(LLVMBuilderRef, LLVMValueRef Src, bool) {
return Src;
}
extern "C" LLVMValueRef
LLVMRustBuildVectorReduceMax(LLVMBuilderRef, LLVMValueRef Src, bool) {
return Src;
}
extern "C" LLVMValueRef
LLVMRustBuildVectorReduceFMin(LLVMBuilderRef, LLVMValueRef Src, bool) {
return Src;
}
extern "C" LLVMValueRef
LLVMRustBuildVectorReduceFMax(LLVMBuilderRef, LLVMValueRef Src, bool) {
return Src;
}
#endif