1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
|
//===- TestConvertGPUKernelToCubin.cpp - Test gpu kernel cubin lowering ---===//
//
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
// See https://llvm.org/LICENSE.txt for license information.
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
//
//===----------------------------------------------------------------------===//
#include "mlir/Dialect/GPU/Transforms/Passes.h"
#include "mlir/Pass/Pass.h"
#include "mlir/Target/LLVMIR/Dialect/NVVM/NVVMToLLVMIRTranslation.h"
#include "mlir/Target/LLVMIR/Export.h"
#include "llvm/Support/TargetSelect.h"
using namespace mlir;
#if MLIR_CUDA_CONVERSIONS_ENABLED
namespace {
class TestSerializeToCubinPass
: public PassWrapper<TestSerializeToCubinPass, gpu::SerializeToBlobPass> {
public:
MLIR_DEFINE_EXPLICIT_INTERNAL_INLINE_TYPE_ID(TestSerializeToCubinPass)
StringRef getArgument() const final { return "test-gpu-to-cubin"; }
StringRef getDescription() const final {
return "Lower GPU kernel function to CUBIN binary annotations";
}
TestSerializeToCubinPass();
private:
void getDependentDialects(DialectRegistry ®istry) const override;
// Serializes PTX to CUBIN.
std::unique_ptr<std::vector<char>>
serializeISA(const std::string &isa) override;
};
} // namespace
TestSerializeToCubinPass::TestSerializeToCubinPass() {
this->triple = "nvptx64-nvidia-cuda";
this->chip = "sm_35";
this->features = "+ptx60";
}
void TestSerializeToCubinPass::getDependentDialects(
DialectRegistry ®istry) const {
registerNVVMDialectTranslation(registry);
gpu::SerializeToBlobPass::getDependentDialects(registry);
}
std::unique_ptr<std::vector<char>>
TestSerializeToCubinPass::serializeISA(const std::string &) {
std::string data = "CUBIN";
return std::make_unique<std::vector<char>>(data.begin(), data.end());
}
namespace mlir {
namespace test {
// Register test pass to serialize GPU module to a CUBIN binary annotation.
void registerTestGpuSerializeToCubinPass() {
PassRegistration<TestSerializeToCubinPass>([] {
// Initialize LLVM NVPTX backend.
LLVMInitializeNVPTXTarget();
LLVMInitializeNVPTXTargetInfo();
LLVMInitializeNVPTXTargetMC();
LLVMInitializeNVPTXAsmPrinter();
return std::make_unique<TestSerializeToCubinPass>();
});
}
} // namespace test
} // namespace mlir
#endif // MLIR_CUDA_CONVERSIONS_ENABLED
|
