mlir.cpp

#include <migraphx/gpu/mlir.hpp>
#include <migraphx/gpu/target.hpp>
#include <migraphx/ref/target.hpp>
#include <migraphx/module.hpp>
#include <migraphx/program.hpp>
#include <migraphx/make_op.hpp>
#include <migraphx/ranges.hpp>
#include <migraphx/stringutils.hpp>
#include <migraphx/generate.hpp>
#include <migraphx/verify_args.hpp>
#include <test.hpp>

using migraphx::trim;

std::string encode(std::string s)
{
    std::stringstream ss;
    bool prespace = false;
    for(auto c : s)
    {
        if(std::isspace(c))
        {
            if(not prespace)
                ss << "  ";
            prespace = true;
        }
        else if(std::isprint(c))
        {
            ss << c;
            prespace = false;
        }
    }
    return migraphx::trim(ss.str());
}

migraphx::program create_program_from_mlir(const migraphx::module& mmlir)
{
    migraphx::program p;
    auto* mm   = p.get_main_module();
    auto names = mmlir.get_parameter_names();
    std::vector<migraphx::instruction_ref> inputs;
    std::transform(names.begin(), names.end(), std::back_inserter(inputs), [&](const auto& name) {
        return mm->add_parameter(name, mmlir.get_parameter_shape(name));
    });
    inputs.push_back(mm->add_parameter("output", mmlir.get_output_shapes().front()));
    migraphx::gpu::insert_mlir(*mm, mm->end(), mmlir, inputs);
    return p;
}

migraphx::parameter_map generate_params(const migraphx::program& p)
{
    migraphx::parameter_map m;
    std::size_t i = 0;
    for(auto&& x : p.get_parameter_shapes())
    {
        m[x.first] = migraphx::generate_argument(x.second, i++);
    }
    return m;
}

migraphx::argument run_gpu(migraphx::program p, const migraphx::parameter_map& inputs)
{
    migraphx::gpu::target t;
    p.compile(t);
    migraphx::parameter_map m;
    for(auto&& input : inputs)
    {
        m[input.first] = t.copy_to(input.second);
    }
    for(auto&& x : p.get_parameter_shapes())
    {
        if(m.count(x.first) == 0)
        {
            m[x.first] = t.allocate(x.second);
        }
    }
    return t.copy_from(p.eval(m).front());
}

migraphx::argument run_ref(migraphx::program p, const migraphx::parameter_map& inputs)
{
    p.compile(migraphx::ref::target{});
    return p.eval(inputs).front();
}

bool verify_mlir(const migraphx::module& mmlir)
{
    migraphx::program ref;
    ref.get_main_module()->insert_module_instructions(ref.get_main_module()->end(), &mmlir);

    auto inputs = generate_params(ref);

    auto mlir = create_program_from_mlir(mmlir);
    return migraphx::verify_args("mlir", run_ref(ref, inputs), run_gpu(mlir, inputs));
}

TEST_CASE(conv)
{
    const std::string mlir_output = R"__migraphx__(
module  {
  func @main(%arg0: tensor<1x8x4x4xf32>, %arg1: tensor<2x8x3x3xf32>) -> tensor<1x2x2x2xf32> {
    %0 = migraphx.convolution(%arg0, %arg1) {dilation = [1, 1], group = 1 : i64, padding = [0, 0, 0, 0], padding_mode = 0 : i64, stride = [1, 1]} : (tensor<1x8x4x4xf32>, tensor<2x8x3x3xf32>) -> tensor<1x2x2x2xf32>
    return %0 : tensor<1x2x2x2xf32>
  }
}
)__migraphx__";
    migraphx::module m;
    auto x    = m.add_parameter("x", {migraphx::shape::float_type, {1, 8, 4, 4}});
    auto w    = m.add_parameter("w", {migraphx::shape::float_type, {2, 8, 3, 3}});
    auto conv = m.add_instruction(migraphx::make_op("convolution"), x, w);
    m.add_return({conv});
    auto s = migraphx::gpu::dump_mlir(m);
    // Skip test if MLIR is not enabled
    if(s.empty())
        return;
    std::cout << s << std::endl;
    EXPECT(encode(s) == encode(mlir_output));
    EXPECT(verify_mlir(m));
}

TEST_CASE(conv_add_relu)
{
    const std::string mlir_output = R"__migraphx__(
module  {
  func @main(%arg0: tensor<1x8x4x4xf32>, %arg1: tensor<2x8x3x3xf32>, %arg2: tensor<1x2x2x2xf32>) -> tensor<1x2x2x2xf32> {
    %0 = migraphx.convolution(%arg0, %arg1) {dilation = [1, 1], group = 1 : i64, padding = [0, 0, 0, 0], padding_mode = 0 : i64, stride = [1, 1]} : (tensor<1x8x4x4xf32>, tensor<2x8x3x3xf32>) -> tensor<1x2x2x2xf32>
    %1 = migraphx.add(%0, %arg2) : (tensor<1x2x2x2xf32>, tensor<1x2x2x2xf32>) -> tensor<1x2x2x2xf32>
    %2 = migraphx.relu(%1) : (tensor<1x2x2x2xf32>) -> tensor<1x2x2x2xf32>
    return %2 : tensor<1x2x2x2xf32>
  }
}
)__migraphx__";
    migraphx::module m;
    auto x    = m.add_parameter("x", {migraphx::shape::float_type, {1, 8, 4, 4}});
    auto w    = m.add_parameter("w", {migraphx::shape::float_type, {2, 8, 3, 3}});
    auto b    = m.add_parameter("b", {migraphx::shape::float_type, {1, 2, 2, 2}});
    auto conv = m.add_instruction(migraphx::make_op("convolution"), x, w);
    auto add  = m.add_instruction(migraphx::make_op("add"), conv, b);
    auto relu = m.add_instruction(migraphx::make_op("relu"), add);
    m.add_return({relu});
    auto s = migraphx::gpu::dump_mlir(m);
    // Skip test if MLIR is not enabled
    if(s.empty())
        return;
    std::cout << s << std::endl;
    EXPECT(encode(s) == encode(mlir_output));
    EXPECT(verify_mlir(m));
}

int main(int argc, const char* argv[]) { test::run(argc, argv); }