Skip to content

Fix bfloat16 detection on aarch64 #55417

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

Closed
wants to merge 1 commit into from
Closed

Fix bfloat16 detection on aarch64 #55417

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
19 changes: 15 additions & 4 deletions src/llvm-demote-float16.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -73,12 +73,23 @@ static bool have_fp16(Function &caller, const Triple &TT) {
}

static bool have_bf16(Function &caller, const Triple &TT) {
if (caller.hasFnAttribute("julia.hasbf16")) {
return true;
Attribute FSAttr = caller.getFnAttribute("target-features");
StringRef FS = "";
if (FSAttr.isValid()) {
FS = FSAttr.getValueAsString();
} else if (jl_ExecutionEngine) {
FS = jl_ExecutionEngine->getTargetFeatureString();
}
if (TT.isAArch64()) {
if (FS.find("+bf16") != llvm::StringRef::npos) {
return true;
}
} else if (TT.getArch() == Triple::x86_64) {
if (caller.hasFnAttribute("julia.hasbf16")) {
return true;
}
}

// there's no targets that fully support bfloat yet;,
// AVX512BF16 only provides conversion and dot product instructions.
return false;
}

Expand Down