| 1 |
// RUN: fir-opt --split-input-file --target-rewrite="target=x86_64-unknown-linux-gnu" %s | FileCheck %s |
| 2 |
|
| 3 |
module attributes {fir.defaultkind = "a1c4d8i4l4r4", fir.kindmap = "", llvm.data_layout = "e-m:e-p270:32:32-p271:32:32-p272:64:64-i64:64-i128:128-f80:128-n8:16:32:64-S128", llvm.target_triple = "x86_64-unknown-linux-gnu"} { |
| 4 |
|
| 5 |
gpu.module @testmod { |
| 6 |
gpu.func @_QPvcpowdk(%arg0: !fir.ref<complex<f64>> {cuf.data_attr = #cuf.cuda<device>, fir.bindc_name = "a"}) attributes {cuf.proc_attr = #cuf.cuda_proc<global>} { |
| 7 |
%0 = fir.alloca i64 |
| 8 |
%1 = fir.load %0 : !fir.ref<i64> |
| 9 |
%2 = fir.load %arg0 : !fir.ref<complex<f64>> |
| 10 |
%3 = fir.call @_FortranAzpowk(%2, %1) fastmath<contract> : (complex<f64>, i64) -> complex<f64> |
| 11 |
gpu.return |
| 12 |
} |
| 13 |
func.func private @_FortranAzpowk(complex<f64>, i64) -> complex<f64> attributes {fir.bindc_name = "_FortranAzpowk", fir.runtime} |
| 14 |
} |
| 15 |
|
| 16 |
// CHECK-LABEL: gpu.func @_QPvcpowdk |
| 17 |
// CHECK: %{{.*}} = fir.call @_FortranAzpowk(%{{.*}}, %{{.*}}, %{{.*}}) fastmath<contract> : (f64, f64, i64) -> tuple<f64, f64> |
| 18 |
// CHECK: func.func private @_FortranAzpowk(f64, f64, i64) -> tuple<f64, f64> attributes {fir.bindc_name = "_FortranAzpowk", fir.runtime} |
| 19 |
} |
| 20 |
|
| 21 |
// ----- |
| 22 |
|
| 23 |
module attributes {fir.defaultkind = "a1c4d8i4l4r4", fir.kindmap = "", llvm.data_layout = "e-m:e-p270:32:32-p271:32:32-p272:64:64-i64:64-i128:128-f80:128-n8:16:32:64-S128", llvm.target_triple = "x86_64-unknown-linux-gnu"} { |
| 24 |
|
| 25 |
gpu.module @testmod { |
| 26 |
gpu.func @_QPtest(%arg0: complex<f64>) -> (complex<f64>) { |
| 27 |
gpu.return %arg0 : complex<f64> |
| 28 |
} |
| 29 |
} |
| 30 |
} |
| 31 |
|
| 32 |
// CHECK-LABEL: gpu.func @_QPtest |
| 33 |
// CHECK-SAME: (%arg0: f64, %arg1: f64) -> tuple<f64, f64> { |
| 34 |
// CHECK: gpu.return %{{.*}} : tuple<f64, f64> |
| 35 |
|
| 36 |
// ----- |
| 37 |
module attributes {gpu.container_module, fir.defaultkind = "a1c4d8i4l4r4", fir.kindmap = "", llvm.data_layout = "e-m:e-p270:32:32-p271:32:32-p272:64:64-i64:64-i128:128-f80:128-n8:16:32:64-S128", llvm.target_triple = "x86_64-unknown-linux-gnu"} { |
| 38 |
|
| 39 |
gpu.module @testmod { |
| 40 |
gpu.func @_QPtest(%arg0: complex<f64>) -> () kernel { |
| 41 |
gpu.return |
| 42 |
} |
| 43 |
} |
| 44 |
|
| 45 |
func.func @main(%arg0: complex<f64>) { |
| 46 |
%0 = llvm.mlir.constant(0 : i64) : i64 |
| 47 |
%1 = llvm.mlir.constant(0 : i32) : i32 |
| 48 |
gpu.launch_func @testmod::@_QPtest blocks in (%0, %0, %0) threads in (%0, %0, %0) : i64 dynamic_shared_memory_size %1 args(%arg0 : complex<f64>) {cuf.proc_attr = #cuf.cuda_proc<global>} |
| 49 |
return |
| 50 |
} |
| 51 |
|
| 52 |
} |
| 53 |
|
| 54 |
// CHECK-LABEL: gpu.func @_QPtest |
| 55 |
// CHECK-SAME: (%arg0: f64, %arg1: f64) kernel { |
| 56 |
// CHECK: gpu.return |
| 57 |
// CHECK: gpu.launch_func @testmod::@_QPtest blocks in (%{{.*}}, %{{.*}}, %{{.*}}) threads in (%{{.*}}, %{{.*}}, %{{.*}}) : i64 dynamic_shared_memory_size %{{.*}} args(%{{.*}} : f64, %{{.*}} : f64) {cuf.proc_attr = #cuf.cuda_proc<global>} |