mirror of
https://github.com/ROCm/composable_kernel.git
synced 2026-05-03 21:21:22 +00:00
[CK-Tile] move out memory operation from cshuffle epilogue class (#3359)
* initial poc * factor out common parts in operator() * cv4 * rest of the universal gemm pipelines * fix test * remove boilerplate from tile engine * fix example * fix example * format * fix tests build for gemm * remove base pipeline codegen from gemm instance builder * unify v3 logic with the rest of universal gemm pipelines * fix build for multi abd test * fix test gemm multi d * fix build for weight preshuffle * fix grouped gemm test * fix grouped gemm multi d test * fix grouped gemm preshuffle * fix grouped gemm example except for quant * fix gemm preshuffle * fix splitk 2 stage example * fix batched gemm example * fix multid example * fix multiabd example * fix batched gemm test * fixup * fix examples build * fix grouped gemm test build * fix smoke builder * hacky poc * fix tile engine * kill the lambda * maybe fix test build * more fixes * clang-format * save temp * clang-format * mostly fix examples * clang-format * remove dead code * more cleanup * fix fmha bwd build (default epilogue set/add appears to be broken) * fix default epilogue tests but not correctness * clang-format * fix bquant * clang-format * cleanup dead code * rearrange make windows for readability * restore changes to IsSupportedArgument * fix smoke-builder * clang-format * fixup rename class * build fixes * clang-format * fix builder * fixup * remove set from builder tests * fix test * clang-format * re-refactor the kernels * clang-format * fix header license * remove memory operation from conv bwd test * clang-format * clang-format example,include * clang-format test * build fixes * clang-format * solve compilation error * fix the CI * solve compilation error * clang format * solve merge conflict * solve merge conflict * solve the gfx11 error * solve test error * moar build fixes * remove AtomicAddRequiresKBatchGreaterThanOne test since the property is removed from the kernel scope --------- Co-authored-by: Thomas Ning <Thomas.Ning@amd.com>
This commit is contained in:
Max Podkorytov
@@ -69,107 +69,88 @@ struct BasicInvoker
|
||||
|
||||
using CodegenGemmPipeline = ck_tile::GemmPipelineAGmemBGmemCRegV1<CodegenPipelineProblem>;
|
||||
|
||||
const auto Run = [&](const auto memory_operation_) {
|
||||
constexpr auto memory_operation = memory_operation_.value;
|
||||
using GemmEpilogue = ck_tile::CShuffleEpilogue<
|
||||
ck_tile::CShuffleEpilogueProblem<ADataType,
|
||||
BDataType,
|
||||
ck_tile::tuple<>,
|
||||
AccDataType,
|
||||
CDataType,
|
||||
ck_tile::tuple<>,
|
||||
CLayout,
|
||||
ck_tile::element_wise::PassThrough,
|
||||
TilePartitioner::MPerBlock,
|
||||
TilePartitioner::NPerBlock,
|
||||
M_Warp,
|
||||
N_Warp,
|
||||
M_Warp_Tile,
|
||||
N_Warp_Tile,
|
||||
K_Warp_Tile,
|
||||
CodegenPipelineProblem::TransposeC>>;
|
||||
|
||||
using GemmEpilogue = ck_tile::CShuffleEpilogue<
|
||||
ck_tile::CShuffleEpilogueProblem<ADataType,
|
||||
BDataType,
|
||||
ck_tile::tuple<>,
|
||||
AccDataType,
|
||||
CDataType,
|
||||
ck_tile::tuple<>,
|
||||
CLayout,
|
||||
ck_tile::element_wise::PassThrough,
|
||||
TilePartitioner::MPerBlock,
|
||||
TilePartitioner::NPerBlock,
|
||||
M_Warp,
|
||||
N_Warp,
|
||||
M_Warp_Tile,
|
||||
N_Warp_Tile,
|
||||
K_Warp_Tile,
|
||||
CodegenPipelineProblem::TransposeC,
|
||||
memory_operation>>;
|
||||
// ToDo: Will add the codegen part to test different pipeline policies in GEMM.
|
||||
// Now we only use the BlockGemmASmemBSmemCRegV1DefaultPolicy.
|
||||
using Kernel = ck_tile::GemmKernel<TilePartitioner, CodegenGemmPipeline, GemmEpilogue>;
|
||||
auto kargs = Kernel::MakeKernelArgs(args);
|
||||
|
||||
// ToDo: Will add the codegen part to test different pipeline policies in GEMM.
|
||||
// Now we only use the BlockGemmASmemBSmemCRegV1DefaultPolicy.
|
||||
using Kernel = ck_tile::GemmKernel<TilePartitioner, CodegenGemmPipeline, GemmEpilogue>;
|
||||
auto kargs = Kernel::MakeKernelArgs(args);
|
||||
const dim3 grids = Kernel::GridSize(args.M, args.N, args.k_batch);
|
||||
const dim3 blocks = Kernel::BlockSize();
|
||||
|
||||
const dim3 grids = Kernel::GridSize(args.M, args.N, args.k_batch);
|
||||
const dim3 blocks = Kernel::BlockSize();
|
||||
if(!Kernel::IsSupportedArgument(kargs))
|
||||
{
|
||||
throw std::runtime_error("Wrong! Arguments not supported! Skipping gemm!\n");
|
||||
}
|
||||
|
||||
if(!Kernel::IsSupportedArgument(kargs))
|
||||
{
|
||||
throw std::runtime_error("Wrong! Arguments not supported! Skipping gemm!\n");
|
||||
}
|
||||
if(s.log_level_ > 0)
|
||||
{
|
||||
std::cout << "Launching kernel with args: " << Kernel::GetName() << '\n'
|
||||
<< "shape: " << CodegenGemmShape::GetName() << '\n'
|
||||
<< "problem: " << CodegenPipelineProblem::GetName() << '\n'
|
||||
<< "pipeline: " << CodegenGemmPipeline::GetName() << '\n'
|
||||
<< "grid: {" << grids.x << ", " << grids.y << ", " << grids.z << "}"
|
||||
<< ", blocks: {" << blocks.x << ", " << blocks.y << ", " << blocks.z << "}"
|
||||
<< std::endl;
|
||||
}
|
||||
|
||||
if(s.log_level_ > 0)
|
||||
{
|
||||
std::cout << "Launching kernel with args: " << Kernel::GetName() << '\n'
|
||||
<< "shape: " << CodegenGemmShape::GetName() << '\n'
|
||||
<< "problem: " << CodegenPipelineProblem::GetName() << '\n'
|
||||
<< "pipeline: " << CodegenGemmPipeline::GetName() << '\n'
|
||||
<< "grid: {" << grids.x << ", " << grids.y << ", " << grids.z << "}"
|
||||
<< ", blocks: {" << blocks.x << ", " << blocks.y << ", " << blocks.z
|
||||
<< "}" << std::endl;
|
||||
}
|
||||
// Declare rotating_mem_ptr here so it stays in scope until it is needed
|
||||
std::unique_ptr<ck_tile::RotatingMemWrapper<ADataType, BDataType>> rotating_mem_ptr;
|
||||
std::function<void()> preprocess;
|
||||
|
||||
// Declare rotating_mem_ptr here so it stays in scope until it is needed
|
||||
std::unique_ptr<ck_tile::RotatingMemWrapper<ADataType, BDataType>> rotating_mem_ptr;
|
||||
std::function<void()> preprocess;
|
||||
|
||||
auto clear_gemm_output = [&]() {
|
||||
if(args.k_batch > 1)
|
||||
hipGetErrorString(hipMemsetAsync(
|
||||
args.e_ptr, 0, args.M * args.N * sizeof(CDataType), s.stream_id_));
|
||||
};
|
||||
|
||||
if(s.flush_cache_)
|
||||
{
|
||||
std::cout << "Flushing cache..." << std::endl;
|
||||
|
||||
ck_tile::HostTensor<ADataType> a_m(ck_tile::host_tensor_descriptor(
|
||||
args.M, args.K, args.stride_A, is_row_major(ALayout{})));
|
||||
ck_tile::HostTensor<BDataType> b_n(ck_tile::host_tensor_descriptor(
|
||||
args.K, args.N, args.stride_B, is_row_major(BLayout{})));
|
||||
|
||||
auto size_a_buffer = a_m.get_element_space_size_in_bytes();
|
||||
auto size_b_buffer = b_n.get_element_space_size_in_bytes();
|
||||
|
||||
rotating_mem_ptr =
|
||||
std::make_unique<ck_tile::RotatingMemWrapper<ADataType, BDataType>>(
|
||||
kargs.as_ptr[0],
|
||||
kargs.bs_ptr[0],
|
||||
s.rotating_count_,
|
||||
size_a_buffer,
|
||||
size_b_buffer);
|
||||
rotating_mem_ptr->Print();
|
||||
|
||||
preprocess = [&]() {
|
||||
ck_tile::flush_icache();
|
||||
rotating_mem_ptr->Next();
|
||||
clear_gemm_output();
|
||||
};
|
||||
}
|
||||
else
|
||||
{
|
||||
preprocess = clear_gemm_output;
|
||||
}
|
||||
|
||||
return ck_tile::launch_kernel_time_mask(
|
||||
s,
|
||||
preprocess,
|
||||
ck_tile::make_kernel<GemmConfig::kBlockPerCu>(Kernel{}, grids, blocks, 0, kargs));
|
||||
auto clear_gemm_output = [&]() {
|
||||
if(args.k_batch > 1)
|
||||
hipGetErrorString(hipMemsetAsync(
|
||||
args.e_ptr, 0, args.M * args.N * sizeof(CDataType), s.stream_id_));
|
||||
};
|
||||
|
||||
if(args.k_batch == 1)
|
||||
if(s.flush_cache_)
|
||||
{
|
||||
return Run(MemoryOpSet{});
|
||||
std::cout << "Flushing cache..." << std::endl;
|
||||
|
||||
ck_tile::HostTensor<ADataType> a_m(ck_tile::host_tensor_descriptor(
|
||||
args.M, args.K, args.stride_A, is_row_major(ALayout{})));
|
||||
ck_tile::HostTensor<BDataType> b_n(ck_tile::host_tensor_descriptor(
|
||||
args.K, args.N, args.stride_B, is_row_major(BLayout{})));
|
||||
|
||||
auto size_a_buffer = a_m.get_element_space_size_in_bytes();
|
||||
auto size_b_buffer = b_n.get_element_space_size_in_bytes();
|
||||
|
||||
rotating_mem_ptr = std::make_unique<ck_tile::RotatingMemWrapper<ADataType, BDataType>>(
|
||||
kargs.as_ptr[0], kargs.bs_ptr[0], s.rotating_count_, size_a_buffer, size_b_buffer);
|
||||
rotating_mem_ptr->Print();
|
||||
|
||||
preprocess = [&]() {
|
||||
ck_tile::flush_icache();
|
||||
rotating_mem_ptr->Next();
|
||||
clear_gemm_output();
|
||||
};
|
||||
}
|
||||
else
|
||||
{
|
||||
return Run(MemoryOpAtomicAdd{});
|
||||
preprocess = clear_gemm_output;
|
||||
}
|
||||
|
||||
return ck_tile::launch_kernel_time_mask(
|
||||
s,
|
||||
preprocess,
|
||||
ck_tile::make_kernel<GemmConfig::kBlockPerCu>(Kernel{}, grids, blocks, 0, kargs));
|
||||
}
|
||||
};
|
||||
|
||||
Reference in New Issue
Block a user