ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-05-04 05:31:24 +00:00
Code Issues Packages Projects Releases Wiki Activity

Add bias scalar vectorload = 1 for gemm bias gemm (#791)

Browse Source 
* first change bias load

* add bias dim and scalervector parameter

* make CDE0BlockTransferSrcVectorDim not work

* changse toinstance

* add limit for CDE0BlockTransferSrcScalarPerVector
This commit is contained in:
ltqin
2023-07-25 09:08:15 +08:00
committed by GitHub
parent 844b215d92
commit 50643dd555
5 changed files with 91 additions and 93 deletions
Download Patch File Download Diff File Expand all files Collapse all files

114
include/ck/tensor_operation/gpu/grid/gridwise_batched_gemm_multiple_d_gemm_multiple_d_xdl_cshuffle_v1.hpp
View File

@@ -67,6 +67,8 @@ template <typename A0B0B1DataType, // FIXME: don't assume A0/B0/B1 have same dat
index_t B0BlockTransferDstScalarPerVector_BK1,
bool B0ThreadTransferSrcResetCoordinateAfterRun, // ignored
index_t B0BlockLdsExtraN,
index_t CDE0BlockTransferSrcVectorDim,
index_t CDE0BlockTransferSrcScalarPerVector,
typename B1BlockTransferThreadClusterLengths_BK0_N_BK1,
typename B1BlockTransferThreadClusterArrangeOrder,
typename B1BlockTransferSrcAccessOrder,
@@ -710,13 +712,13 @@ struct GridwiseBatchedGemmMultipleDGemmMultipleD_Xdl_CShuffle
constexpr auto d0_thread_desc_m0_n0_m1_n1_m2_n2_m3_n3_n4_n5 =
make_naive_tensor_descriptor_packed(make_tuple(I1, // MBlockId
I1, // NBlockID
I1, // MRepeat
I1, // NRepeat
I1, // MWaveId
I1, // NWaveId
I1, // MPerXdl
I1, // NGroupNum
I1, // NInputNum
m0, // MRepeat
n0, // NRepeat
m1, // MWaveId
n1, // NWaveId
m2, // MPerXdl
n2, // NGroupNum
n3, // NInputNum
n4)); // registerNum
auto d0s_thread_buf = generate_tuple(
@@ -732,8 +734,9 @@ struct GridwiseBatchedGemmMultipleDGemmMultipleD_Xdl_CShuffle
const auto wave_id = GetGemm0WaveIdx();
const auto wave_m_n_id = GetGemm0WaveMNIdx(wave_id[I2]); // I2: 0~63
constexpr auto acc0_thread_desc = make_naive_tensor_descriptor_packed(
make_tuple(Number<Gemm0MXdlPerWave>{}, Number<Gemm0NXdlPerWave>{}, n2, n4));
static_assert(CDE0BlockTransferSrcScalarPerVector <= n4,
"vector load must be not greater than n4");
static_assert(n4 % CDE0BlockTransferSrcScalarPerVector == 0);
auto d0s_threadwise_copy = generate_tuple(
[&](auto i) {
@@ -742,10 +745,19 @@ struct GridwiseBatchedGemmMultipleDGemmMultipleD_Xdl_CShuffle
A0B0B1DataType,
decltype(d0s_griddesc_m0_n0_m1_n1_m2_n2_m3_n3_n4_n5[i]),
decltype(d0_thread_desc_m0_n0_m1_n1_m2_n2_m3_n3_n4_n5),
Sequence<I1, I1, I1, I1, I1, I1, I1, I1, I1, n4>,
Sequence<I1, // MBlockId
I1, // NBlockID
m0, // MRepeat
n0, // NRepeat
m1, // MWaveId
n1, // NWaveId
m2, // MPerXdl
n2, // NGroupNum
n3, // NInputNum
n4>,
Sequence<0, 1, 2, 3, 4, 5, 6, 7, 8, 9>,
9,
n4,
9, // CDE0BlockTransferSrcVectorDim
CDE0BlockTransferSrcScalarPerVector,
1,
false>(d0s_griddesc_m0_n0_m1_n1_m2_n2_m3_n3_n4_n5[i],
make_multi_index(block_work_idx[I0], // MBlockId
@@ -898,66 +910,42 @@ struct GridwiseBatchedGemmMultipleDGemmMultipleD_Xdl_CShuffle
blockwise_gemm0,
acc0_thread_buf,
num_k_block_main_loop);
// bias+gelu
// multiple d
if constexpr(NumD0Tensor)
{
static_for<0, Gemm0MXdlPerWave, 1>{}([&](auto mr) {
static_for<0, Gemm0NXdlPerWave, 1>{}([&](auto nr) {
static_for<0, n2, 1>{}([&](auto groupid) {
static_for<0, NumD0Tensor, 1>{}([&](auto i) {
d0s_threadwise_copy(i).Run(
d0s_griddesc_m0_n0_m1_n1_m2_n2_m3_n3_n4_n5[i],
d0s_grid_buf[i],
d0_thread_desc_m0_n0_m1_n1_m2_n2_m3_n3_n4_n5,
make_tuple(I0, I0, I0, I0, I0, I0, I0, I0, I0, I0),
d0s_thread_buf(i));
});
static_for<0, NumD0Tensor, 1>{}([&](auto i) {
d0s_threadwise_copy(i).Run(d0s_griddesc_m0_n0_m1_n1_m2_n2_m3_n3_n4_n5[i],
d0s_grid_buf[i],
d0_thread_desc_m0_n0_m1_n1_m2_n2_m3_n3_n4_n5,
make_tuple(I0, I0, I0, I0, I0, I0, I0, I0, I0, I0),
d0s_thread_buf(i));
});
static_for<0, m0 * n0 * n2 * n4, 1>{}([&](auto i) {
// get reference to src data
const auto src_data_refs = generate_tie(
// return type should be lvalue
[&](auto iSrc) -> const auto& { return d0s_thread_buf[iSrc][i]; },
Number<NumD0Tensor>{});
static_for<0, n4, 1>{}([&](auto i) {
constexpr index_t c_offset = acc0_thread_desc.CalculateOffset(
make_tuple(mr, nr, groupid, i));
// get reference to dst data
auto dst_data_refs = generate_tie(
// return type should be lvalue
[&](auto) -> auto& { return acc0_thread_buf(i); },
Number<2>{});
// get reference to src data
const auto src_data_refs = generate_tie(
// return type should be lvalue
[&](auto iSrc) -> const auto& {
return d0s_thread_buf[iSrc][i];
},
Number<NumD0Tensor>{});
// get reference to dst data
auto dst_data_refs = generate_tie(
// return type should be lvalue
[&](auto) -> auto& {
return acc0_thread_buf(Number<c_offset>{});
},
Number<2>{});
unpack2(cde0_element_op, dst_data_refs, src_data_refs);
});
static_for<0, NumD0Tensor, 1>{}([&](auto i) {
d0s_threadwise_copy(i).MoveSrcSliceWindow(
d0s_griddesc_m0_n0_m1_n1_m2_n2_m3_n3_n4_n5[i],
make_multi_index(0, 0, 0, 0, 0, 0, 0, 1, 0, 0));
});
});
static_for<0, NumD0Tensor, 1>{}([&](auto i) {
d0s_threadwise_copy(i).MoveSrcSliceWindow(
d0s_griddesc_m0_n0_m1_n1_m2_n2_m3_n3_n4_n5[i],
make_multi_index(0, 0, 0, 1, 0, 0, 0, -n2.value, 0, 0));
});
});
static_for<0, NumD0Tensor, 1>{}([&](auto i) {
d0s_threadwise_copy(i).MoveSrcSliceWindow(
d0s_griddesc_m0_n0_m1_n1_m2_n2_m3_n3_n4_n5[i],
make_multi_index(0, 0, 1, -Gemm0NXdlPerWave, 0, 0, 0, 0, 0, 0));
});
unpack2(cde0_element_op, dst_data_refs, src_data_refs);
});
static_for<0, NumD0Tensor, 1>{}([&](auto i) {
d0s_threadwise_copy(i).MoveSrcSliceWindow(
d0s_griddesc_m0_n0_m1_n1_m2_n2_m3_n3_n4_n5[i],
make_multi_index(0, 1, -Gemm0MXdlPerWave, 0, 0, 0, 0, 0, 0, 0));
make_multi_index(0, 1, 0, 0, 0, 0, 0, 0, 0, 0));
});
}
else
{
static_for<0, acc0_thread_buf.Size(), 1>{}(
[&](auto i) { cde0_element_op(acc_thread_buf(i), acc0_thread_buf[i]); });
}
// gemm1
{
// TODO: explore using dynamic buffer for a1 thread buffer