Skip to content

[flang][cuda] Carry over the cuf.proc_attr attribute to gpu.launch_func #124325

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] Carry over the cuf.proc_attr attribute to gpu.launch_func #124325

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
5 changes: 5 additions & 0 deletions flang/lib/Optimizer/Transforms/CUFOpConversion.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -806,6 +806,7 @@ struct CUFLaunchOpConversion
rewriter.getContext(),
op.getCallee().getLeafReference().getValue())});
mlir::Value clusterDimX, clusterDimY, clusterDimZ;
cuf::ProcAttributeAttr procAttr;
if (auto funcOp = symTab.lookup<mlir::func::FuncOp>(
op.getCallee().getLeafReference())) {
if (auto clusterDimsAttr = funcOp->getAttrOfType<cuf::ClusterDimsAttr>(
Expand All @@ -817,6 +818,8 @@ struct CUFLaunchOpConversion
clusterDimZ = rewriter.create<mlir::arith::ConstantIndexOp>(
loc, clusterDimsAttr.getZ().getInt());
}
procAttr =
funcOp->getAttrOfType<cuf::ProcAttributeAttr>(cuf::getProcAttrName());
}
llvm::SmallVector<mlir::Value> args;
for (mlir::Value arg : op.getArgs()) {
Expand Down Expand Up @@ -851,6 +854,8 @@ struct CUFLaunchOpConversion
gpuLaunchOp.getClusterSizeYMutable().assign(clusterDimY);
gpuLaunchOp.getClusterSizeZMutable().assign(clusterDimZ);
}
if (procAttr)
gpuLaunchOp->setAttr(cuf::getProcAttrName(), procAttr);
rewriter.replaceOp(op, gpuLaunchOp);
return mlir::success();
}
Expand Down
21 changes: 21 additions & 0 deletions flang/test/Fir/CUDA/cuda-launch.fir
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -104,3 +104,24 @@ module attributes {gpu.container_module, dlti.dl_spec = #dlti.dl_spec<#dlti.dl_e
// CHECK: %[[DEVADDR:.*]] = fir.call @_FortranACUFGetDeviceAddress(%[[CONV_ADDR]], %{{.*}}, %{{.*}}) : (!fir.llvm_ptr<i8>, !fir.ref<i8>, i32) -> !fir.llvm_ptr<i8>
// CHECK: %[[CONV_DEVADDR:.*]] = fir.convert %[[DEVADDR]] : (!fir.llvm_ptr<i8>) -> !fir.ref<!fir.box<!fir.ptr<!fir.array<?xf32>>>>
// CHECK: gpu.launch_func @cuda_device_mod::@_QMdevptrPtest blocks in (%{{.*}}, %{{.*}}, %{{.*}}) threads in (%{{.*}}, %{{.*}}, %{{.*}}) dynamic_shared_memory_size %{{.*}} args(%[[CONV_DEVADDR]] : !fir.ref<!fir.box<!fir.ptr<!fir.array<?xf32>>>>)

// -----

module attributes {gpu.container_module, dlti.dl_spec = #dlti.dl_spec<#dlti.dl_entry<f80, dense<128> : vector<2xi64>>, #dlti.dl_entry<i128, dense<128> : vector<2xi64>>, #dlti.dl_entry<i64, dense<64> : vector<2xi64>>, #dlti.dl_entry<!llvm.ptr<272>, dense<64> : vector<4xi64>>, #dlti.dl_entry<!llvm.ptr<271>, dense<32> : vector<4xi64>>, #dlti.dl_entry<!llvm.ptr<270>, dense<32> : vector<4xi64>>, #dlti.dl_entry<f128, dense<128> : vector<2xi64>>, #dlti.dl_entry<f64, dense<64> : vector<2xi64>>, #dlti.dl_entry<f16, dense<16> : vector<2xi64>>, #dlti.dl_entry<i32, dense<32> : vector<2xi64>>, #dlti.dl_entry<i16, dense<16> : vector<2xi64>>, #dlti.dl_entry<i8, dense<8> : vector<2xi64>>, #dlti.dl_entry<i1, dense<8> : vector<2xi64>>, #dlti.dl_entry<!llvm.ptr, dense<64> : vector<4xi64>>, #dlti.dl_entry<"dlti.endianness", "little">, #dlti.dl_entry<"dlti.stack_alignment", 128 : i64>>} {
gpu.module @cuda_device_mod {
gpu.func @_QMdevptrPtest() kernel {
gpu.return
}
}
func.func @_QMdevptrPtest() attributes {cuf.proc_attr = #cuf.cuda_proc<grid_global>} {
return
}
func.func @_QQmain() {
%c1_i32 = arith.constant 1 : i32
cuf.kernel_launch @_QMdevptrPtest<<<%c1_i32, %c1_i32, %c1_i32, %c1_i32, %c1_i32, %c1_i32>>>()
return
}
}

// CHECK-LABEL: func.func @_QQmain()
// CHECK: gpu.launch_func @cuda_device_mod::@_QMdevptrPtest blocks in (%{{.*}}, %{{.*}}, %{{.*}}) threads in (%{{.*}}, %{{.*}}, %{{.*}}) dynamic_shared_memory_size %{{.*}} {cuf.proc_attr = #cuf.cuda_proc<grid_global>}
Loading