12ba74e181
Runtime function call to a void function are producing a ssa value because the FunctionType result is set to NoneType with is later translated to a empty struct. This is not an issue when going to LLVM IR but it breaks when lowering a gpu module to PTX. This patch update the RTModel to correctly set the FunctionType result type to nothing. This is one runtime call before this patch at the LLVM IR dialect step. ``` %45 = llvm.call @_FortranAAssign(%arg0, %1, %44, %4) : (!llvm.ptr, !llvm.ptr, !llvm.ptr, i32) -> !llvm.struct<()> ``` After the patch the call would be correctly formed ``` llvm.call @_FortranAAssign(%arg0, %1, %44, %4) : (!llvm.ptr, !llvm.ptr, !llvm.ptr, i32) -> () ``` Without the patch it would lead to error like: ``` ptxas /tmp/mlir-cuda_device_mod-nvptx64-nvidia-cuda-sm_60-e804b6.ptx, line 10; error : Output parameter cannot be an incomplete array. ptxas /tmp/mlir-cuda_device_mod-nvptx64-nvidia-cuda-sm_60-e804b6.ptx, line 125; error : Call has wrong number of parameters ``` The change is pretty much mechanical.
33 lines
2.8 KiB
Plaintext
33 lines
2.8 KiB
Plaintext
// RUN: fir-opt %s --abstract-result | FileCheck %s
|
|
|
|
// Check that the attributes are shifted along with their corresponding arguments
|
|
|
|
// CHECK-LABEL: func.func @_QMi8Pintrinsic_pack0
|
|
// CHECK-SAME:(%arg0: !fir.ref<!fir.box<!fir.heap<!fir.array<?xi32>>>>, %arg1: !fir.box<!fir.array<?x?x?xi32>> {fir.bindc_name = "array"}, %arg2: !fir.ref<!fir.logical<4>> {fir.bindc_name = "mask"}, %arg3: !fir.box<!fir.array<?xi32>> {fir.bindc_name = "vector", fir.optional})
|
|
|
|
func.func @_QMi8Pintrinsic_pack0(%arg0: !fir.box<!fir.array<?x?x?xi32>> {fir.bindc_name = "array"}, %arg1: !fir.ref<!fir.logical<4>> {fir.bindc_name = "mask"}, %arg2: !fir.box<!fir.array<?xi32>> {fir.bindc_name = "vector", fir.optional}) -> !fir.box<!fir.heap<!fir.array<?xi32>>> {
|
|
%0 = fir.alloca !fir.box<!fir.heap<!fir.array<?xi32>>>
|
|
%1 = fir.alloca !fir.box<!fir.heap<!fir.array<?xi32>>> {bindc_name = "intrinsic_pack0", uniq_name = "_QMi8Fintrinsic_pack0Eintrinsic_pack0"}
|
|
%2 = fir.zero_bits !fir.heap<!fir.array<?xi32>>
|
|
%c0 = arith.constant 0 : index
|
|
%3 = fir.shape %c0 : (index) -> !fir.shape<1>
|
|
%4 = fir.embox %2(%3) : (!fir.heap<!fir.array<?xi32>>, !fir.shape<1>) -> !fir.box<!fir.heap<!fir.array<?xi32>>>
|
|
fir.store %4 to %1 : !fir.ref<!fir.box<!fir.heap<!fir.array<?xi32>>>>
|
|
%5 = fir.embox %arg1 : (!fir.ref<!fir.logical<4>>) -> !fir.box<!fir.logical<4>>
|
|
%6 = fir.zero_bits !fir.heap<!fir.array<?xi32>>
|
|
%7 = fir.embox %6(%3) : (!fir.heap<!fir.array<?xi32>>, !fir.shape<1>) -> !fir.box<!fir.heap<!fir.array<?xi32>>>
|
|
fir.store %7 to %0 : !fir.ref<!fir.box<!fir.heap<!fir.array<?xi32>>>>
|
|
%8 = fir.convert %0 : (!fir.ref<!fir.box<!fir.heap<!fir.array<?xi32>>>>) -> !fir.ref<!fir.box<none>>
|
|
%9 = fir.convert %arg0 : (!fir.box<!fir.array<?x?x?xi32>>) -> !fir.box<none>
|
|
%10 = fir.convert %5 : (!fir.box<!fir.logical<4>>) -> !fir.box<none>
|
|
%11 = fir.convert %arg2 : (!fir.box<!fir.array<?xi32>>) -> !fir.box<none>
|
|
fir.call @_FortranAPack(%8, %9, %10, %11) : (!fir.ref<!fir.box<none>>, !fir.box<none>, !fir.box<none>, !fir.box<none>) -> ()
|
|
%13 = fir.load %1 : !fir.ref<!fir.box<!fir.heap<!fir.array<?xi32>>>>
|
|
return %13 : !fir.box<!fir.heap<!fir.array<?xi32>>>
|
|
}
|
|
func.func private @_FortranAPack(!fir.ref<!fir.box<none>>, !fir.box<none>, !fir.box<none>, !fir.box<none>) -> () attributes {fir.runtime}
|
|
|
|
// CHECK-LABEL: func.func private @empty
|
|
// CHECK-SAME:(!fir.ref<!fir.box<!fir.heap<!fir.array<?xi32>>>>, !fir.box<!fir.array<?x?x?xi32>> {fir.bindc_name = "array"}, !fir.ref<!fir.logical<4>> {fir.bindc_name = "mask"}, !fir.box<!fir.array<?xi32>> {fir.bindc_name = "vector", fir.optional})
|
|
func.func private @empty(%arg0: !fir.box<!fir.array<?x?x?xi32>> {fir.bindc_name = "array"}, %arg1: !fir.ref<!fir.logical<4>> {fir.bindc_name = "mask"}, %arg2: !fir.box<!fir.array<?xi32>> {fir.bindc_name = "vector", fir.optional}) -> !fir.box<!fir.heap<!fir.array<?xi32>>>
|