Skip to content

[flang][cuda] Handle variable with initialization in device global pass #124307

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
merged 1 commit into from
Jan 24, 2025
Merged

[flang][cuda] Handle variable with initialization in device global pass #124307

Show file tree
Hide file tree
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
12 changes: 5 additions & 7 deletions flang/lib/Optimizer/Transforms/CUFDeviceGlobal.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -36,13 +36,11 @@ static void processAddrOfOp(fir::AddrOfOp addrOfOp,
addrOfOp.getSymbol().getRootReference().getValue())) {
// TO DO: limit candidates to non-scalars. Scalars appear to have been
// folded in already.
if (globalOp.getConstant()) {
if (recurseInGlobal)
globalOp.walk([&](fir::AddrOfOp op) {
processAddrOfOp(op, symbolTable, candidates, recurseInGlobal);
});
candidates.insert(globalOp);
}
if (recurseInGlobal)
globalOp.walk([&](fir::AddrOfOp op) {
processAddrOfOp(op, symbolTable, candidates, recurseInGlobal);
});
candidates.insert(globalOp);
}
}

Expand Down
21 changes: 21 additions & 0 deletions flang/test/Fir/CUDA/cuda-implicit-device-global.f90
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -308,3 +308,24 @@ // Test that global used in device function are flagged with the correct
// CHECK-DAG: fir.global linkonce_odr @_QM__mod1E.c.__builtin_c_devptr
// CHECK-DAG: fir.global linkonce_odr @_QM__mod1E.dt.__builtin_c_devptr
// CHECK-DAG: fir.global linkonce_odr @_QM__mod1E.n.__builtin_c_devptr

// -----

// Variables with initialization are promoted to non constant global.
//
// attributes(global) subroutine kernel4()
// integer :: a = 4
// end subroutine

func.func @_QPkernel4() attributes {cuf.proc_attr = #cuf.cuda_proc<global>} {
%0 = fir.address_of(@_QFkernel4Ea) : !fir.ref<i32>
return
}
fir.global internal @_QFkernel4Ea : i32 {
%c4_i32 = arith.constant 4 : i32
fir.has_value %c4_i32 : i32
}

// CHECK-LABEL: fir.global internal @_QFkernel4Ea : i32
// CHECK-LABEL: gpu.module @cuda_device_mod
// CHECK: fir.global internal @_QFkernel4Ea : i32
Loading