Skip to content

GitLab

Commit 104aeabc authored by danyao12's avatar danyao12
Browse files

split bwd mqa/gqa

parent 198558c5
Expand all Show whitespace changes
Inline Side-by-side
Showing with 482 additions and 114 deletions
example/32_batched_gemm_scale_softmax_gemm/batched_multihead_attention_backward_v3.cpp
View file @ 104aeabc
...@@ -277,7 +277,8 @@ int run(int argc, char* argv[]) ...@@ -277,7 +277,8 @@ int run(int argc, char* argv[])
ck::index_t K = DIM; ck::index_t K = DIM;
ck::index_t O = DIM; ck::index_t O = DIM;
ck::index_t G0 = 4; ck::index_t G0 = 4;
ck::index_t G1 = 6; ck::index_t G1 = 6; // h_q
ck::index_t G2 = 6; // h_kv
bool input_permute = false; bool input_permute = false;
bool output_permute = false; bool output_permute = false;
...@@ -296,7 +297,7 @@ int run(int argc, char* argv[]) ...@@ -296,7 +297,7 @@ int run(int argc, char* argv[])
init_method = std::stoi(argv[2]); init_method = std::stoi(argv[2]);
time_kernel = std::stoi(argv[3]); time_kernel = std::stoi(argv[3]);
} }
else if(argc == 13) else if(argc == 14)
{ {
do_verification = std::stoi(argv[1]); do_verification = std::stoi(argv[1]);
init_method = std::stoi(argv[2]); init_method = std::stoi(argv[2]);
...@@ -308,20 +309,21 @@ int run(int argc, char* argv[]) ...@@ -308,20 +309,21 @@ int run(int argc, char* argv[])
O = std::stoi(argv[7]); O = std::stoi(argv[7]);
G0 = std::stoi(argv[8]); G0 = std::stoi(argv[8]);
G1 = std::stoi(argv[9]); G1 = std::stoi(argv[9]);
G2 = std::stoi(argv[10]);
p_drop = std::stof(argv[10]); p_drop = std::stof(argv[11]);
input_permute = std::stoi(argv[11]); input_permute = std::stoi(argv[12]);
output_permute = std::stoi(argv[12]); output_permute = std::stoi(argv[13]);
} }
else else
{ {
printf("arg1: verification (0=no, 1=yes)\n"); printf("arg1: verification (0=no, 1=yes)\n");
printf("arg2: initialization (0=no init, 1=integer value, 2=decimal value)\n"); printf("arg2: initialization (0=no init, 1=integer value, 2=decimal value)\n");
printf("arg3: time kernel (0=no, 1=yes)\n"); printf("arg3: time kernel (0=no, 1=yes)\n");
printf("arg4 to 11: M, N, K, O, G0, G1\n"); printf("arg4 to 10: M, N, K, O, G0, G1, G2\n");
printf("arg10: scale (alpha)\n"); printf("arg11: p_drop\n");
printf("arg11 to 12: input / output permute\n"); printf("arg12 to 13: input / output permute\n");
exit(0); exit(0);
} }
...@@ -339,6 +341,7 @@ int run(int argc, char* argv[]) ...@@ -339,6 +341,7 @@ int run(int argc, char* argv[])
std::cout << "O: " << O << std::endl; std::cout << "O: " << O << std::endl;
std::cout << "G0: " << G0 << std::endl; std::cout << "G0: " << G0 << std::endl;
std::cout << "G1: " << G1 << std::endl; std::cout << "G1: " << G1 << std::endl;
std::cout << "G2: " << G2 << std::endl;
std::cout << "alpha: " << alpha << std::endl; std::cout << "alpha: " << alpha << std::endl;
std::cout << "input_permute: " << input_permute << std::endl; std::cout << "input_permute: " << input_permute << std::endl;
std::cout << "output_permute: " << output_permute << std::endl; std::cout << "output_permute: " << output_permute << std::endl;
...@@ -354,17 +357,17 @@ int run(int argc, char* argv[]) ...@@ -354,17 +357,17 @@ int run(int argc, char* argv[])
? std::vector<ck::index_t>{M * G1 * K, K, G1 * K, 1} // Q layout [G0, M, G1, K] ? std::vector<ck::index_t>{M * G1 * K, K, G1 * K, 1} // Q layout [G0, M, G1, K]
: std::vector<ck::index_t>{G1 * M * K, M * K, K, 1}; // Q layout [G0, G1, M, K] : std::vector<ck::index_t>{G1 * M * K, M * K, K, 1}; // Q layout [G0, G1, M, K]
std::vector<ck::index_t> k_gs_ns_ks_lengths{G0, G1, N, K}; std::vector<ck::index_t> k_gs_ns_ks_lengths{G0, G2, N, K};
std::vector<ck::index_t> k_gs_ns_ks_strides = std::vector<ck::index_t> k_gs_ns_ks_strides =
input_permute input_permute
? std::vector<ck::index_t>{N * G1 * K, K, G1 * K, 1} // K layout [G0, N, G1, K] ? std::vector<ck::index_t>{N * G2 * K, K, G2 * K, 1} // K layout [G0, N, G2, K]
: std::vector<ck::index_t>{G1 * N * K, N * K, K, 1}; // K layout [G0, G1, N, K] : std::vector<ck::index_t>{G2 * N * K, N * K, K, 1}; // K layout [G0, G2, N, K]
std::vector<ck::index_t> v_gs_os_ns_lengths{G0, G1, O, N}; std::vector<ck::index_t> v_gs_os_ns_lengths{G0, G2, O, N};
std::vector<ck::index_t> v_gs_os_ns_strides = std::vector<ck::index_t> v_gs_os_ns_strides =
input_permute input_permute
? std::vector<ck::index_t>{N * G1 * O, O, 1, G1 * O} // V layout [G0, N, G1, O] ? std::vector<ck::index_t>{N * G2 * O, O, 1, G2 * O} // V layout [G0, N, G2, O]
: std::vector<ck::index_t>{G1 * N * O, N * O, 1, O}; // V layout [G0, G1, N, O] : std::vector<ck::index_t>{G2 * N * O, N * O, 1, O}; // V layout [G0, G2, N, O]
std::vector<ck::index_t> y_gs_ms_os_lengths{G0, G1, M, O}; std::vector<ck::index_t> y_gs_ms_os_lengths{G0, G1, M, O};
std::vector<ck::index_t> y_gs_ms_os_strides = std::vector<ck::index_t> y_gs_ms_os_strides =
...@@ -377,6 +380,18 @@ int run(int argc, char* argv[]) ...@@ -377,6 +380,18 @@ int run(int argc, char* argv[])
input_permute input_permute
? std::vector<ck::index_t>{M * G1 * N, N, G1 * N, 1} // Z layout [G0, M, G1, N] ? std::vector<ck::index_t>{M * G1 * N, N, G1 * N, 1} // Z layout [G0, M, G1, N]
: std::vector<ck::index_t>{G1 * M * N, M * N, N, 1}; // Z layout [G0, G1, M, N] : std::vector<ck::index_t>{G1 * M * N, M * N, N, 1}; // Z layout [G0, G1, M, N]
std::vector<ck::index_t> kgrad_gs_ns_ks_lengths{G0, G1, N, K};
std::vector<ck::index_t> kgrad_gs_ns_ks_strides =
input_permute
? std::vector<ck::index_t>{N * G1 * K, K, G1 * K, 1} // KGrad layout [G0, N, G1, K]
: std::vector<ck::index_t>{G1 * N * K, N * K, K, 1}; // KGrad layout [G0, G1, N, K]
std::vector<ck::index_t> vgrad_gs_os_ns_lengths{G0, G1, O, N};
std::vector<ck::index_t> vgrad_gs_os_ns_strides =
input_permute
? std::vector<ck::index_t>{N * G1 * O, O, 1, G1 * O} // VGrad layout [G0, N, G1, O]
: std::vector<ck::index_t>{G1 * N * O, N * O, 1, O}; // VGrad layout [G0, G1, N, O]
// The softmax stat log-sum-exp (LSE) is used to speed up softmax calculation in backward pass // The softmax stat log-sum-exp (LSE) is used to speed up softmax calculation in backward pass
// Pi = exp(Si) / sum(exp(S0) + exp(S1) + ...) // Pi = exp(Si) / sum(exp(S0) + exp(S1) + ...)
// = exp(Si) / exp(log(sum(exp() + ...))) // = exp(Si) / exp(log(sum(exp() + ...)))
...@@ -394,6 +409,8 @@ int run(int argc, char* argv[]) ...@@ -394,6 +409,8 @@ int run(int argc, char* argv[])
Tensor<InputDataType> ygrad_gs_ms_os(y_gs_ms_os_lengths, y_gs_ms_os_strides); Tensor<InputDataType> ygrad_gs_ms_os(y_gs_ms_os_lengths, y_gs_ms_os_strides);
Tensor<LSEDataType> lse_gs_ms(lse_gs_ms_lengths, lse_gs_ms_strides); Tensor<LSEDataType> lse_gs_ms(lse_gs_ms_lengths, lse_gs_ms_strides);
Tensor<DDataType> d_gs_ms(lse_gs_ms_lengths, lse_gs_ms_strides); Tensor<DDataType> d_gs_ms(lse_gs_ms_lengths, lse_gs_ms_strides);
Tensor<OutputDataType> kgrad_gs_ns_ks(kgrad_gs_ns_ks_lengths, kgrad_gs_ns_ks_strides);
Tensor<OutputDataType> vgrad_gs_os_ns(vgrad_gs_os_ns_lengths, vgrad_gs_os_ns_strides);
std::cout << "q_gs_ms_ks: " << q_gs_ms_ks.mDesc << std::endl; std::cout << "q_gs_ms_ks: " << q_gs_ms_ks.mDesc << std::endl;
std::cout << "k_gs_ns_ks: " << k_gs_ns_ks.mDesc << std::endl; std::cout << "k_gs_ns_ks: " << k_gs_ns_ks.mDesc << std::endl;
...@@ -402,6 +419,8 @@ int run(int argc, char* argv[]) ...@@ -402,6 +419,8 @@ int run(int argc, char* argv[])
std::cout << "y_gs_ms_os: " << y_gs_ms_os.mDesc << std::endl; std::cout << "y_gs_ms_os: " << y_gs_ms_os.mDesc << std::endl;
std::cout << "lse_gs_ms_os: " << lse_gs_ms.mDesc << std::endl; std::cout << "lse_gs_ms_os: " << lse_gs_ms.mDesc << std::endl;
std::cout << "d_gs_ms_os: " << d_gs_ms.mDesc << std::endl; std::cout << "d_gs_ms_os: " << d_gs_ms.mDesc << std::endl;
std::cout << "kgrad_gs_ns_ks: " << kgrad_gs_ns_ks.mDesc << std::endl;
std::cout << "vgrad_gs_os_ns: " << vgrad_gs_os_ns.mDesc << std::endl;
z_gs_ms_ns.GenerateTensorValue(GeneratorTensor_1<InputDataType>{0}); z_gs_ms_ns.GenerateTensorValue(GeneratorTensor_1<InputDataType>{0});
switch(init_method) switch(init_method)
...@@ -478,10 +497,20 @@ int run(int argc, char* argv[]) ...@@ -478,10 +497,20 @@ int run(int argc, char* argv[])
q_gs_ms_ks.ForEach( q_gs_ms_ks.ForEach(
[&](auto& self, auto idx) { q_g_m_k(idx[0] * G1 + idx[1], idx[2], idx[3]) = self(idx); }); [&](auto& self, auto idx) { q_g_m_k(idx[0] * G1 + idx[1], idx[2], idx[3]) = self(idx); });
k_gs_ns_ks.ForEach( k_g_n_k.ForEach([&](auto& self, auto idx) {
[&](auto& self, auto idx) { k_g_n_k(idx[0] * G1 + idx[1], idx[2], idx[3]) = self(idx); }); const size_t& g0 = idx[0] / G1;
v_gs_os_ns.ForEach( const size_t& g1 = idx[0] % G1;
[&](auto& self, auto idx) { v_g_n_o(idx[0] * G1 + idx[1], idx[3], idx[2]) = self(idx); }); const size_t& g2 = g1 / (G1 / G2);
self(idx) = k_gs_ns_ks(g0, g2, idx[1], idx[2]);
});
v_g_n_o.ForEach([&](auto& self, auto idx) {
const size_t& g0 = idx[0] / G1;
const size_t& g1 = idx[0] % G1;
const size_t& g2 = g1 / (G1 / G2);
self(idx) = v_gs_os_ns(g0, g2, idx[2], idx[1]);
});
// qkv gradients have the same descriptor as with qkv // qkv gradients have the same descriptor as with qkv
DeviceMem q_device_buf(sizeof(InputDataType) * q_gs_ms_ks.mDesc.GetElementSpaceSize()); DeviceMem q_device_buf(sizeof(InputDataType) * q_gs_ms_ks.mDesc.GetElementSpaceSize());
...@@ -491,8 +520,8 @@ int run(int argc, char* argv[]) ...@@ -491,8 +520,8 @@ int run(int argc, char* argv[])
DeviceMem y_device_buf(sizeof(InputDataType) * y_gs_ms_os.mDesc.GetElementSpaceSize()); DeviceMem y_device_buf(sizeof(InputDataType) * y_gs_ms_os.mDesc.GetElementSpaceSize());
DeviceMem lse_device_buf(sizeof(LSEDataType) * lse_gs_ms.mDesc.GetElementSpaceSize()); DeviceMem lse_device_buf(sizeof(LSEDataType) * lse_gs_ms.mDesc.GetElementSpaceSize());
DeviceMem qgrad_device_buf(sizeof(OutputDataType) * q_gs_ms_ks.mDesc.GetElementSpaceSize()); DeviceMem qgrad_device_buf(sizeof(OutputDataType) * q_gs_ms_ks.mDesc.GetElementSpaceSize());
DeviceMem kgrad_device_buf(sizeof(OutputDataType) * k_gs_ns_ks.mDesc.GetElementSpaceSize()); DeviceMem kgrad_device_buf(sizeof(OutputDataType) * kgrad_gs_ns_ks.mDesc.GetElementSpaceSize());
DeviceMem vgrad_device_buf(sizeof(OutputDataType) * v_gs_os_ns.mDesc.GetElementSpaceSize()); DeviceMem vgrad_device_buf(sizeof(OutputDataType) * vgrad_gs_os_ns.mDesc.GetElementSpaceSize());
DeviceMem ygrad_device_buf(sizeof(InputDataType) * y_gs_ms_os.mDesc.GetElementSpaceSize()); DeviceMem ygrad_device_buf(sizeof(InputDataType) * y_gs_ms_os.mDesc.GetElementSpaceSize());
DeviceMem d_device_buf(sizeof(DDataType) * d_gs_ms.mDesc.GetElementSpaceSize()); DeviceMem d_device_buf(sizeof(DDataType) * d_gs_ms.mDesc.GetElementSpaceSize());
...@@ -533,6 +562,10 @@ int run(int argc, char* argv[]) ...@@ -533,6 +562,10 @@ int run(int argc, char* argv[])
y_gs_ms_os_lengths, y_gs_ms_os_lengths,
y_gs_ms_os_strides, y_gs_ms_os_strides,
lse_gs_ms_lengths, lse_gs_ms_lengths,
kgrad_gs_ns_ks_lengths,
kgrad_gs_ns_ks_strides,
vgrad_gs_os_ns_lengths,
vgrad_gs_os_ns_strides,
{}, // std::array<std::vector<ck::index_t>, 1>{acc0_biases_gs_ms_ns_lengths}, {}, // std::array<std::vector<ck::index_t>, 1>{acc0_biases_gs_ms_ns_lengths},
{}, // std::array<std::vector<ck::index_t>, 1>{acc0_biases_gs_ms_ns_strides}, {}, // std::array<std::vector<ck::index_t>, 1>{acc0_biases_gs_ms_ns_strides},
{}, // std::array<std::vector<ck::index_t>, 1>{acc1_biases_gs_ms_os_lengths}, {}, // std::array<std::vector<ck::index_t>, 1>{acc1_biases_gs_ms_os_lengths},
...@@ -581,6 +614,10 @@ int run(int argc, char* argv[]) ...@@ -581,6 +614,10 @@ int run(int argc, char* argv[])
y_gs_ms_os_lengths, y_gs_ms_os_lengths,
y_gs_ms_os_strides, y_gs_ms_os_strides,
lse_gs_ms_lengths, lse_gs_ms_lengths,
kgrad_gs_ns_ks_lengths,
kgrad_gs_ns_ks_strides,
vgrad_gs_os_ns_lengths,
vgrad_gs_os_ns_strides,
{}, // std::array<std::vector<ck::index_t>, 1>{acc0_biases_gs_ms_ns_lengths}, {}, // std::array<std::vector<ck::index_t>, 1>{acc0_biases_gs_ms_ns_lengths},
{}, // std::array<std::vector<ck::index_t>, 1>{acc0_biases_gs_ms_ns_strides}, {}, // std::array<std::vector<ck::index_t>, 1>{acc0_biases_gs_ms_ns_strides},
{}, // std::array<std::vector<ck::index_t>, 1>{acc1_biases_gs_ms_os_lengths}, {}, // std::array<std::vector<ck::index_t>, 1>{acc1_biases_gs_ms_os_lengths},
...@@ -763,12 +800,16 @@ int run(int argc, char* argv[]) ...@@ -763,12 +800,16 @@ int run(int argc, char* argv[])
#endif #endif
Tensor<OutputDataType> qgrad_gs_ms_ks_host_result(q_gs_ms_ks_lengths, q_gs_ms_ks_strides); Tensor<OutputDataType> qgrad_gs_ms_ks_host_result(q_gs_ms_ks_lengths, q_gs_ms_ks_strides);
Tensor<OutputDataType> kgrad_gs_ns_ks_host_result(k_gs_ns_ks_lengths, k_gs_ns_ks_strides); Tensor<OutputDataType> kgrad_gs_ns_ks_host_result(kgrad_gs_ns_ks_lengths,
Tensor<OutputDataType> vgrad_gs_os_ns_host_result(v_gs_os_ns_lengths, v_gs_os_ns_strides); kgrad_gs_ns_ks_strides);
Tensor<OutputDataType> vgrad_gs_os_ns_host_result(vgrad_gs_os_ns_lengths,
vgrad_gs_os_ns_strides);
Tensor<OutputDataType> qgrad_gs_ms_ks_device_result(q_gs_ms_ks_lengths, q_gs_ms_ks_strides); Tensor<OutputDataType> qgrad_gs_ms_ks_device_result(q_gs_ms_ks_lengths, q_gs_ms_ks_strides);
Tensor<OutputDataType> kgrad_gs_ns_ks_device_result(k_gs_ns_ks_lengths, k_gs_ns_ks_strides); Tensor<OutputDataType> kgrad_gs_ns_ks_device_result(kgrad_gs_ns_ks_lengths,
Tensor<OutputDataType> vgrad_gs_os_ns_device_result(v_gs_os_ns_lengths, v_gs_os_ns_strides); kgrad_gs_ns_ks_strides);
Tensor<OutputDataType> vgrad_gs_os_ns_device_result(vgrad_gs_os_ns_lengths,
vgrad_gs_os_ns_strides);
qgrad_device_buf.FromDevice(qgrad_gs_ms_ks_device_result.mData.data()); qgrad_device_buf.FromDevice(qgrad_gs_ms_ks_device_result.mData.data());
kgrad_device_buf.FromDevice(kgrad_gs_ns_ks_device_result.mData.data()); kgrad_device_buf.FromDevice(kgrad_gs_ns_ks_device_result.mData.data());
......
example/32_batched_gemm_scale_softmax_gemm/grouped_multihead_attention_backward_v3.cpp
View file @ 104aeabc
...@@ -24,7 +24,7 @@ Kernel outputs: ...@@ -24,7 +24,7 @@ Kernel outputs:
*/ */
#define USING_MASK 0 #define USING_MASK 0
#define DIM 32 // DIM should be a multiple of 8. #define DIM 128 // DIM should be a multiple of 8.
#include <iostream> #include <iostream>
#include <numeric> #include <numeric>
...@@ -273,6 +273,7 @@ int run(int argc, char* argv[]) ...@@ -273,6 +273,7 @@ int run(int argc, char* argv[])
// y_g0_m_g1_o = permute(y_g0_g1_m_o, [0, 2, 1, 3]) // y_g0_m_g1_o = permute(y_g0_g1_m_o, [0, 2, 1, 3])
float alpha = 1.f / std::sqrt(DIM); float alpha = 1.f / std::sqrt(DIM);
float p_drop = 0.0; float p_drop = 0.0;
int h_ratio = 1; // G1 / G2
bool input_permute = true; bool input_permute = true;
bool output_permute = true; bool output_permute = true;
...@@ -290,25 +291,26 @@ int run(int argc, char* argv[]) ...@@ -290,25 +291,26 @@ int run(int argc, char* argv[])
init_method = std::stoi(argv[2]); init_method = std::stoi(argv[2]);
time_kernel = std::stoi(argv[3]); time_kernel = std::stoi(argv[3]);
} }
else if(argc == 7) else if(argc == 8)
{ {
do_verification = std::stoi(argv[1]); do_verification = std::stoi(argv[1]);
init_method = std::stoi(argv[2]); init_method = std::stoi(argv[2]);
time_kernel = std::stoi(argv[3]); time_kernel = std::stoi(argv[3]);
p_drop = std::stof(argv[4]); p_drop = std::stof(argv[4]);
h_ratio = std::stof(argv[5]);
input_permute = std::stoi(argv[5]); input_permute = std::stoi(argv[6]);
output_permute = std::stoi(argv[6]); output_permute = std::stoi(argv[7]);
} }
else else
{ {
printf("arg1: verification (0=no, 1=yes)\n"); printf("arg1: verification (0=no, 1=yes)\n");
printf("arg2: initialization (0=no init, 1=integer value, 2=decimal value)\n"); printf("arg2: initialization (0=no init, 1=integer value, 2=decimal value)\n");
printf("arg3: time kernel (0=no, 1=yes)\n"); printf("arg3: time kernel (0=no, 1=yes)\n");
printf("arg4 to 11: M, N, K, O, G0, G1\n"); printf("arg4: p_drop\n");
printf("arg10: scale (alpha)\n"); printf("arg5: h_ratio\n");
printf("arg11 to 12: input / output permute\n"); printf("arg6 to 7: input / output permute\n");
exit(0); exit(0);
} }
...@@ -376,24 +378,25 @@ int run(int argc, char* argv[]) ...@@ -376,24 +378,25 @@ int run(int argc, char* argv[])
int K = DIM; int K = DIM;
int O = DIM; int O = DIM;
int G0 = rand() % 4 + 1; int G0 = rand() % 4 + 1;
int G1 = rand() % 4 + 1; int G2 = rand() % 4 + 1;
int G1 = G2 * h_ratio;
std::vector<ck::index_t> q_gs_ms_ks_lengths{G0, G1, M, K}; std::vector<ck::index_t> q_gs_ms_ks_lengths{G0, G1, M, K};
std::vector<ck::index_t> q_gs_ms_ks_strides = std::vector<ck::index_t> q_gs_ms_ks_strides =
input_permute input_permute
? std::vector<ck::index_t>{M * G1 * K, K, G1 * K, 1} // Q layout [G0, M, G1, K] ? std::vector<ck::index_t>{M * G1 * K, K, G1 * K, 1} // Q layout [G0, M, G1, K]
: std::vector<ck::index_t>{G1 * M * K, M * K, K, 1}; // Q layout [G0, G1, M, K] : std::vector<ck::index_t>{G1 * M * K, M * K, K, 1}; // Q layout [G0, G1, M, K]
std::vector<ck::index_t> k_gs_ns_ks_lengths{G0, G1, N, K}; std::vector<ck::index_t> k_gs_ns_ks_lengths{G0, G2, N, K};
std::vector<ck::index_t> k_gs_ns_ks_strides = std::vector<ck::index_t> k_gs_ns_ks_strides =
input_permute input_permute
? std::vector<ck::index_t>{N * G1 * K, K, G1 * K, 1} // K layout [G0, N, G1, K] ? std::vector<ck::index_t>{N * G2 * K, K, G2 * K, 1} // K layout [G0, N, G2, K]
: std::vector<ck::index_t>{G1 * N * K, N * K, K, 1}; // K layout [G0, G1, N, K] : std::vector<ck::index_t>{G2 * N * K, N * K, K, 1}; // K layout [G0, G2, N, K]
std::vector<ck::index_t> v_gs_os_ns_lengths{G0, G1, O, N}; std::vector<ck::index_t> v_gs_os_ns_lengths{G0, G2, O, N};
std::vector<ck::index_t> v_gs_os_ns_strides = std::vector<ck::index_t> v_gs_os_ns_strides =
input_permute input_permute
? std::vector<ck::index_t>{N * G1 * O, O, 1, G1 * O} // V layout [G0, N, G1, O] ? std::vector<ck::index_t>{N * G2 * O, O, 1, G2 * O} // V layout [G0, N, G2, O]
: std::vector<ck::index_t>{G1 * N * O, N * O, 1, O}; // V layout [G0, G1, N, O] : std::vector<ck::index_t>{G2 * N * O, N * O, 1, O}; // V layout [G0, G2, N, O]
std::vector<ck::index_t> y_gs_ms_os_lengths{G0, G1, M, O}; std::vector<ck::index_t> y_gs_ms_os_lengths{G0, G1, M, O};
std::vector<ck::index_t> y_gs_ms_os_strides = std::vector<ck::index_t> y_gs_ms_os_strides =
...@@ -406,6 +409,17 @@ int run(int argc, char* argv[]) ...@@ -406,6 +409,17 @@ int run(int argc, char* argv[])
input_permute input_permute
? std::vector<ck::index_t>{M * G1 * N, N, G1 * N, 1} // Z layout [G0, M, G1, N] ? std::vector<ck::index_t>{M * G1 * N, N, G1 * N, 1} // Z layout [G0, M, G1, N]
: std::vector<ck::index_t>{G1 * M * N, M * N, N, 1}; // Z layout [G0, G1, M, N] : std::vector<ck::index_t>{G1 * M * N, M * N, N, 1}; // Z layout [G0, G1, M, N]
std::vector<ck::index_t> kgrad_gs_ns_ks_lengths{G0, G1, N, K};
std::vector<ck::index_t> kgrad_gs_ns_ks_strides =
input_permute
? std::vector<ck::index_t>{N * G1 * K, K, G1 * K, 1} // KGrad layout [G0, N, G1, K]
: std::vector<ck::index_t>{G1 * N * K, N * K, K, 1}; // KGrad layout [G0, G1, N, K]
std::vector<ck::index_t> vgrad_gs_os_ns_lengths{G0, G1, O, N};
std::vector<ck::index_t> vgrad_gs_os_ns_strides =
input_permute
? std::vector<ck::index_t>{N * G1 * O, O, 1, G1 * O} // VGrad layout [G0, N, G1, O]
: std::vector<ck::index_t>{G1 * N * O, N * O, 1, O}; // VGrad layout [G0, G1, N, O]
// The softmax stat log-sum-exp (LSE) is used to speed up softmax calculation in backward // The softmax stat log-sum-exp (LSE) is used to speed up softmax calculation in backward
// pass Pi = exp(Si) / sum(exp(S0) + exp(S1) + ...) // pass Pi = exp(Si) / sum(exp(S0) + exp(S1) + ...)
// = exp(Si) / exp(log(sum(exp() + ...))) // = exp(Si) / exp(log(sum(exp() + ...)))
...@@ -427,6 +441,10 @@ int run(int argc, char* argv[]) ...@@ -427,6 +441,10 @@ int run(int argc, char* argv[])
y_gs_ms_os_strides, y_gs_ms_os_strides,
lse_gs_ms_lengths, lse_gs_ms_lengths,
lse_gs_ms_strides, lse_gs_ms_strides,
kgrad_gs_ns_ks_lengths,
kgrad_gs_ns_ks_strides,
vgrad_gs_os_ns_lengths,
vgrad_gs_os_ns_strides,
{}, // std::array<std::vector<ck::index_t>, 1>{acc0_biases_gs_ms_ns_lengths}, {}, // std::array<std::vector<ck::index_t>, 1>{acc0_biases_gs_ms_ns_lengths},
{}, // std::array<std::vector<ck::index_t>, 1>{acc0_biases_gs_ms_ns_strides}, {}, // std::array<std::vector<ck::index_t>, 1>{acc0_biases_gs_ms_ns_strides},
{}, // std::array<std::vector<ck::index_t>, 1>{acc1_biases_gs_ms_os_lengths}, {}, // std::array<std::vector<ck::index_t>, 1>{acc1_biases_gs_ms_os_lengths},
...@@ -451,6 +469,8 @@ int run(int argc, char* argv[]) ...@@ -451,6 +469,8 @@ int run(int argc, char* argv[])
Tensor<InputDataType> ygrad_gs_ms_os(y_gs_ms_os_lengths, y_gs_ms_os_strides); Tensor<InputDataType> ygrad_gs_ms_os(y_gs_ms_os_lengths, y_gs_ms_os_strides);
Tensor<LSEDataType> lse_gs_ms(lse_gs_ms_lengths, lse_gs_ms_strides); Tensor<LSEDataType> lse_gs_ms(lse_gs_ms_lengths, lse_gs_ms_strides);
Tensor<DDataType> d_gs_ms(lse_gs_ms_lengths, lse_gs_ms_strides); Tensor<DDataType> d_gs_ms(lse_gs_ms_lengths, lse_gs_ms_strides);
Tensor<OutputDataType> kgrad_gs_ns_ks(kgrad_gs_ns_ks_lengths, kgrad_gs_ns_ks_strides);
Tensor<OutputDataType> vgrad_gs_os_ns(vgrad_gs_os_ns_lengths, vgrad_gs_os_ns_strides);
if(i < 4) if(i < 4)
{ {
std::cout << "q_gs_ms_ks: " << q_gs_ms_ks.mDesc << std::endl; std::cout << "q_gs_ms_ks: " << q_gs_ms_ks.mDesc << std::endl;
...@@ -460,6 +480,8 @@ int run(int argc, char* argv[]) ...@@ -460,6 +480,8 @@ int run(int argc, char* argv[])
std::cout << "y_gs_ms_os: " << y_gs_ms_os.mDesc << std::endl; std::cout << "y_gs_ms_os: " << y_gs_ms_os.mDesc << std::endl;
std::cout << "lse_gs_ms_os: " << lse_gs_ms.mDesc << std::endl; std::cout << "lse_gs_ms_os: " << lse_gs_ms.mDesc << std::endl;
std::cout << "d_gs_ms_os: " << d_gs_ms.mDesc << std::endl; std::cout << "d_gs_ms_os: " << d_gs_ms.mDesc << std::endl;
std::cout << "kgrad_gs_ns_ks: " << kgrad_gs_ns_ks.mDesc << std::endl;
std::cout << "vgrad_gs_os_ns: " << vgrad_gs_os_ns.mDesc << std::endl;
} }
z_gs_ms_ns.GenerateTensorValue(GeneratorTensor_1<InputDataType>{0}); z_gs_ms_ns.GenerateTensorValue(GeneratorTensor_1<InputDataType>{0});
switch(init_method) switch(init_method)
...@@ -538,11 +560,19 @@ int run(int argc, char* argv[]) ...@@ -538,11 +560,19 @@ int run(int argc, char* argv[])
q_gs_ms_ks.ForEach([&](auto& self, auto idx) { q_gs_ms_ks.ForEach([&](auto& self, auto idx) {
q_g_m_k(idx[0] * G1 + idx[1], idx[2], idx[3]) = self(idx); q_g_m_k(idx[0] * G1 + idx[1], idx[2], idx[3]) = self(idx);
}); });
k_gs_ns_ks.ForEach([&](auto& self, auto idx) { k_g_n_k.ForEach([&](auto& self, auto idx) {
k_g_n_k(idx[0] * G1 + idx[1], idx[2], idx[3]) = self(idx); const size_t& g0 = idx[0] / G1;
const size_t& g1 = idx[0] % G1;
const size_t& g2 = g1 / h_ratio;
self(idx) = k_gs_ns_ks(g0, g2, idx[1], idx[2]);
}); });
v_gs_os_ns.ForEach([&](auto& self, auto idx) { v_g_n_o.ForEach([&](auto& self, auto idx) {
v_g_n_o(idx[0] * G1 + idx[1], idx[3], idx[2]) = self(idx); const size_t& g0 = idx[0] / G1;
const size_t& g1 = idx[0] % G1;
const size_t& g2 = g1 / h_ratio;
self(idx) = v_gs_os_ns(g0, g2, idx[2], idx[1]);
}); });
q_g_m_ks.push_back(q_g_m_k); q_g_m_ks.push_back(q_g_m_k);
...@@ -562,6 +592,8 @@ int run(int argc, char* argv[]) ...@@ -562,6 +592,8 @@ int run(int argc, char* argv[])
z_tensors.push_back(z_gs_ms_ns); z_tensors.push_back(z_gs_ms_ns);
lse_tensors.push_back(lse_gs_ms); lse_tensors.push_back(lse_gs_ms);
ygrad_tensors.push_back(ygrad_gs_ms_os); ygrad_tensors.push_back(ygrad_gs_ms_os);
kgrad_tensors.push_back(kgrad_gs_ns_ks);
vgrad_tensors.push_back(vgrad_gs_os_ns);
q_tensors_device.emplace_back( q_tensors_device.emplace_back(
std::make_unique<DeviceMem>(sizeof(InputDataType) * q_gs_ms_ks.GetElementSpaceSize())); std::make_unique<DeviceMem>(sizeof(InputDataType) * q_gs_ms_ks.GetElementSpaceSize()));
k_tensors_device.emplace_back( k_tensors_device.emplace_back(
...@@ -578,10 +610,10 @@ int run(int argc, char* argv[]) ...@@ -578,10 +610,10 @@ int run(int argc, char* argv[])
std::make_unique<DeviceMem>(sizeof(DDataType) * d_gs_ms.GetElementSpaceSize())); std::make_unique<DeviceMem>(sizeof(DDataType) * d_gs_ms.GetElementSpaceSize()));
qgrad_tensors_device.emplace_back( qgrad_tensors_device.emplace_back(
std::make_unique<DeviceMem>(sizeof(OutputDataType) * q_gs_ms_ks.GetElementSpaceSize())); std::make_unique<DeviceMem>(sizeof(OutputDataType) * q_gs_ms_ks.GetElementSpaceSize()));
kgrad_tensors_device.emplace_back( kgrad_tensors_device.emplace_back(std::make_unique<DeviceMem>(
std::make_unique<DeviceMem>(sizeof(OutputDataType) * k_gs_ns_ks.GetElementSpaceSize())); sizeof(OutputDataType) * kgrad_gs_ns_ks.GetElementSpaceSize()));
vgrad_tensors_device.emplace_back( vgrad_tensors_device.emplace_back(std::make_unique<DeviceMem>(
std::make_unique<DeviceMem>(sizeof(OutputDataType) * v_gs_os_ns.GetElementSpaceSize())); sizeof(OutputDataType) * vgrad_gs_os_ns.GetElementSpaceSize()));
ygrad_tensors_device.emplace_back( ygrad_tensors_device.emplace_back(
std::make_unique<DeviceMem>(sizeof(InputDataType) * y_gs_ms_os.GetElementSpaceSize())); std::make_unique<DeviceMem>(sizeof(InputDataType) * y_gs_ms_os.GetElementSpaceSize()));
q_tensors_device.back()->ToDevice(q_gs_ms_ks.data()); q_tensors_device.back()->ToDevice(q_gs_ms_ks.data());
...@@ -625,6 +657,7 @@ int run(int argc, char* argv[]) ...@@ -625,6 +657,7 @@ int run(int argc, char* argv[])
QKVElementOp{}, QKVElementOp{},
YElementOp{}, YElementOp{},
p_drop, p_drop,
h_ratio,
std::tuple<unsigned long long, unsigned long long>(seed, offset)); std::tuple<unsigned long long, unsigned long long>(seed, offset));
DeviceMem problem_desc_workspace(gemm.GetWorkSpaceSize(&argument)); DeviceMem problem_desc_workspace(gemm.GetWorkSpaceSize(&argument));
...@@ -674,6 +707,7 @@ int run(int argc, char* argv[]) ...@@ -674,6 +707,7 @@ int run(int argc, char* argv[])
QKVElementOp{}, QKVElementOp{},
YElementOp{}, YElementOp{},
p_drop, p_drop,
h_ratio,
std::tuple<unsigned long long, unsigned long long>(seed, offset)); std::tuple<unsigned long long, unsigned long long>(seed, offset));
DeviceMem problem_desc_workspace_verify(gemm.GetWorkSpaceSize(&argument)); DeviceMem problem_desc_workspace_verify(gemm.GetWorkSpaceSize(&argument));
gemm.SetWorkSpacePointer(&argument, problem_desc_workspace_verify.GetDeviceBuffer()); gemm.SetWorkSpacePointer(&argument, problem_desc_workspace_verify.GetDeviceBuffer());
...@@ -687,7 +721,7 @@ int run(int argc, char* argv[]) ...@@ -687,7 +721,7 @@ int run(int argc, char* argv[])
for(std::size_t i = 0; i < group_count; i++) for(std::size_t i = 0; i < group_count; i++)
{ {
int G1 = v_tensors[i].GetLengths()[1]; int G1 = q_tensors[i].GetLengths()[1];
// copy z matirx data form device // copy z matirx data form device
z_tensors_device[i]->FromDevice(z_tensors[i].mData.data()); z_tensors_device[i]->FromDevice(z_tensors[i].mData.data());
z_tensors[i].ForEach([&](auto& self, auto idx) { z_tensors[i].ForEach([&](auto& self, auto idx) {
...@@ -724,8 +758,8 @@ int run(int argc, char* argv[]) ...@@ -724,8 +758,8 @@ int run(int argc, char* argv[])
for(std::size_t i = 0; i < group_count; i++) for(std::size_t i = 0; i < group_count; i++)
{ {
int G0 = v_tensors[i].GetLengths()[0]; int G0 = q_tensors[i].GetLengths()[0];
int G1 = v_tensors[i].GetLengths()[1]; int G1 = q_tensors[i].GetLengths()[1];
int O = v_tensors[i].GetLengths()[2]; int O = v_tensors[i].GetLengths()[2];
int N = v_tensors[i].GetLengths()[3]; int N = v_tensors[i].GetLengths()[3];
int M = q_tensors[i].GetLengths()[2]; int M = q_tensors[i].GetLengths()[2];
...@@ -783,17 +817,17 @@ int run(int argc, char* argv[]) ...@@ -783,17 +817,17 @@ int run(int argc, char* argv[])
Tensor<OutputDataType> qgrad_gs_ms_ks_host_result(q_tensors[i].GetLengths(), Tensor<OutputDataType> qgrad_gs_ms_ks_host_result(q_tensors[i].GetLengths(),
q_tensors[i].GetStrides()); q_tensors[i].GetStrides());
Tensor<OutputDataType> kgrad_gs_ns_ks_host_result(k_tensors[i].GetLengths(), Tensor<OutputDataType> kgrad_gs_ns_ks_host_result(kgrad_tensors[i].GetLengths(),
k_tensors[i].GetStrides()); kgrad_tensors[i].GetStrides());
Tensor<OutputDataType> vgrad_gs_os_ns_host_result(v_tensors[i].GetLengths(), Tensor<OutputDataType> vgrad_gs_os_ns_host_result(vgrad_tensors[i].GetLengths(),
v_tensors[i].GetStrides()); vgrad_tensors[i].GetStrides());
Tensor<OutputDataType> qgrad_gs_ms_ks_device_result(q_tensors[i].GetLengths(), Tensor<OutputDataType> qgrad_gs_ms_ks_device_result(q_tensors[i].GetLengths(),
q_tensors[i].GetStrides()); q_tensors[i].GetStrides());
Tensor<OutputDataType> kgrad_gs_ns_ks_device_result(k_tensors[i].GetLengths(), Tensor<OutputDataType> kgrad_gs_ns_ks_device_result(kgrad_tensors[i].GetLengths(),
k_tensors[i].GetStrides()); kgrad_tensors[i].GetStrides());
Tensor<OutputDataType> vgrad_gs_os_ns_device_result(v_tensors[i].GetLengths(), Tensor<OutputDataType> vgrad_gs_os_ns_device_result(vgrad_tensors[i].GetLengths(),
v_tensors[i].GetStrides()); vgrad_tensors[i].GetStrides());
qgrad_tensors_device[i]->FromDevice(qgrad_gs_ms_ks_device_result.data()); qgrad_tensors_device[i]->FromDevice(qgrad_gs_ms_ks_device_result.data());
kgrad_tensors_device[i]->FromDevice(kgrad_gs_ns_ks_device_result.data()); kgrad_tensors_device[i]->FromDevice(kgrad_gs_ns_ks_device_result.data());
......
include/ck/tensor_operation/gpu/device/impl/device_batched_mha_bwd_xdl_cshuffle_qloop_light_v2.hpp
View file @ 104aeabc
...@@ -132,14 +132,17 @@ __global__ void ...@@ -132,14 +132,17 @@ __global__ void
const CElementwiseOperation c_element_op, const CElementwiseOperation c_element_op,
const AGridDesc_AK0_M_AK1 a_grid_desc_ak0_m_ak1, const AGridDesc_AK0_M_AK1 a_grid_desc_ak0_m_ak1,
const BGridDesc_BK0_N_BK1 b_grid_desc_bk0_n_bk1, const BGridDesc_BK0_N_BK1 b_grid_desc_bk0_n_bk1,
const BGridDesc_BK0_N_BK1 bgrad_grid_desc_bk0_n_bk1,
const D0GridDescriptor_M0_N0_M1_M2_N1_M3 d0_grid_desc_m0_n0_m1_m2_n1_m3, const D0GridDescriptor_M0_N0_M1_M2_N1_M3 d0_grid_desc_m0_n0_m1_m2_n1_m3,
const ZGridDescriptor_M0_N0_M1_N1_M2_N2_M3_M4_M5_N3 const ZGridDescriptor_M0_N0_M1_N1_M2_N2_M3_M4_M5_N3
c_grid_desc_m0_n0_m1_n1_m2_n2_m3_m4_m5_n3, c_grid_desc_m0_n0_m1_n1_m2_n2_m3_m4_m5_n3,
const B1GridDesc_BK0_N_BK1 b1_grid_desc_bk0_n_bk1, const B1GridDesc_BK0_N_BK1 b1_grid_desc_bk0_n_bk1,
const B1GridDesc_BK0_N_BK1 b1grad_grid_desc_bk0_n_bk1,
const LSEGridDescriptor_M lse_grid_desc_m, const LSEGridDescriptor_M lse_grid_desc_m,
const YGradGridDesc_M0_O_M1 ygrad_grid_desc_m0_o_m1, const YGradGridDesc_M0_O_M1 ygrad_grid_desc_m0_o_m1,
const Block2CTileMap block_2_ctile_map, const Block2CTileMap block_2_ctile_map,
const index_t batch_count, const index_t batch_count,
const index_t h_ratio,
const index_t nblock, const index_t nblock,
const ComputeBasePtrOfStridedBatch compute_base_ptr_of_batch, const ComputeBasePtrOfStridedBatch compute_base_ptr_of_batch,
const C0MatrixMask c0_matrix_mask, const C0MatrixMask c0_matrix_mask,
...@@ -155,21 +158,26 @@ __global__ void ...@@ -155,21 +158,26 @@ __global__ void
const index_t num_blocks_per_batch = const index_t num_blocks_per_batch =
__builtin_amdgcn_readfirstlane(get_grid_size() / batch_count); __builtin_amdgcn_readfirstlane(get_grid_size() / batch_count);
const index_t g_idx = __builtin_amdgcn_readfirstlane(get_block_1d_id() / num_blocks_per_batch); const index_t g_idx = __builtin_amdgcn_readfirstlane(get_block_1d_id() / num_blocks_per_batch);
const index_t gkv_idx = __builtin_amdgcn_readfirstlane(g_idx / h_ratio);
// NOTE: assumes QKVY has the same layout as dQ/dK/dV/dY therefore being able to reuse batch // NOTE: assumes QKVY has the same layout as dQ/dK/dV/dY therefore being able to reuse batch
// offsets // offsets
const long_index_t a_batch_offset = __builtin_amdgcn_readfirstlane( const long_index_t a_batch_offset = __builtin_amdgcn_readfirstlane(
static_cast<long_index_t>(compute_base_ptr_of_batch.GetABasePtr(g_idx))); static_cast<long_index_t>(compute_base_ptr_of_batch.GetABasePtr(g_idx)));
const long_index_t b_batch_offset = __builtin_amdgcn_readfirstlane( const long_index_t b_batch_offset = __builtin_amdgcn_readfirstlane(
static_cast<long_index_t>(compute_base_ptr_of_batch.GetBBasePtr(g_idx))); static_cast<long_index_t>(compute_base_ptr_of_batch.GetBBasePtr(gkv_idx)));
const long_index_t z_batch_offset = __builtin_amdgcn_readfirstlane( const long_index_t z_batch_offset = __builtin_amdgcn_readfirstlane(
static_cast<long_index_t>(compute_base_ptr_of_batch.GetZBasePtr(g_idx))); static_cast<long_index_t>(compute_base_ptr_of_batch.GetZBasePtr(g_idx)));
const long_index_t b1_batch_offset = __builtin_amdgcn_readfirstlane( const long_index_t b1_batch_offset = __builtin_amdgcn_readfirstlane(
static_cast<long_index_t>(compute_base_ptr_of_batch.GetB1BasePtr(g_idx))); static_cast<long_index_t>(compute_base_ptr_of_batch.GetB1BasePtr(gkv_idx)));
const long_index_t c_batch_offset = __builtin_amdgcn_readfirstlane( const long_index_t c_batch_offset = __builtin_amdgcn_readfirstlane(
static_cast<long_index_t>(compute_base_ptr_of_batch.GetCBasePtr(g_idx))); static_cast<long_index_t>(compute_base_ptr_of_batch.GetCBasePtr(g_idx)));
const long_index_t lse_batch_offset = __builtin_amdgcn_readfirstlane( const long_index_t lse_batch_offset = __builtin_amdgcn_readfirstlane(
static_cast<long_index_t>(compute_base_ptr_of_batch.GetLSEBasePtr(g_idx))); static_cast<long_index_t>(compute_base_ptr_of_batch.GetLSEBasePtr(g_idx)));
const long_index_t bgrad_batch_offset = __builtin_amdgcn_readfirstlane(
static_cast<long_index_t>(compute_base_ptr_of_batch.GetBGradBasePtr(g_idx)));
const long_index_t b1grad_batch_offset = __builtin_amdgcn_readfirstlane(
static_cast<long_index_t>(compute_base_ptr_of_batch.GetB1GradBasePtr(g_idx)));
ck::philox ph(seed, 0, offset); ck::philox ph(seed, 0, offset);
ZDataType* z_matrix_ptr = (p_z_grid == nullptr ? nullptr : p_z_grid + z_batch_offset); ZDataType* z_matrix_ptr = (p_z_grid == nullptr ? nullptr : p_z_grid + z_batch_offset);
...@@ -206,9 +214,9 @@ __global__ void ...@@ -206,9 +214,9 @@ __global__ void
p_d_grid + lse_batch_offset, p_d_grid + lse_batch_offset,
p_ygrad_grid + c_batch_offset, p_ygrad_grid + c_batch_offset,
p_qgrad_grid + a_batch_offset, p_qgrad_grid + a_batch_offset,
p_kgrad_grid + b_batch_offset, p_kgrad_grid + bgrad_batch_offset,
tmp_p_d0grad_grid, tmp_p_d0grad_grid,
p_vgrad_grid + b1_batch_offset, p_vgrad_grid + b1grad_batch_offset,
p_shared, p_shared,
a_element_op, a_element_op,
b_element_op, b_element_op,
...@@ -217,9 +225,11 @@ __global__ void ...@@ -217,9 +225,11 @@ __global__ void
c_element_op, c_element_op,
a_grid_desc_ak0_m_ak1, a_grid_desc_ak0_m_ak1,
b_grid_desc_bk0_n_bk1, b_grid_desc_bk0_n_bk1,
bgrad_grid_desc_bk0_n_bk1,
d0_grid_desc_m0_n0_m1_m2_n1_m3, d0_grid_desc_m0_n0_m1_m2_n1_m3,
c_grid_desc_m0_n0_m1_n1_m2_n2_m3_m4_m5_n3, c_grid_desc_m0_n0_m1_n1_m2_n2_m3_m4_m5_n3,
b1_grid_desc_bk0_n_bk1, b1_grid_desc_bk0_n_bk1,
b1grad_grid_desc_bk0_n_bk1,
lse_grid_desc_m, lse_grid_desc_m,
ygrad_grid_desc_m0_o_m1, ygrad_grid_desc_m0_o_m1,
block_2_ctile_map, block_2_ctile_map,
...@@ -243,9 +253,9 @@ __global__ void ...@@ -243,9 +253,9 @@ __global__ void
p_d_grid + lse_batch_offset, p_d_grid + lse_batch_offset,
p_ygrad_grid + c_batch_offset, p_ygrad_grid + c_batch_offset,
p_qgrad_grid + a_batch_offset, p_qgrad_grid + a_batch_offset,
p_kgrad_grid + b_batch_offset, p_kgrad_grid + bgrad_batch_offset,
tmp_p_d0grad_grid, tmp_p_d0grad_grid,
p_vgrad_grid + b1_batch_offset, p_vgrad_grid + b1grad_batch_offset,
p_shared, p_shared,
a_element_op, a_element_op,
b_element_op, b_element_op,
...@@ -254,9 +264,11 @@ __global__ void ...@@ -254,9 +264,11 @@ __global__ void
c_element_op, c_element_op,
a_grid_desc_ak0_m_ak1, a_grid_desc_ak0_m_ak1,
b_grid_desc_bk0_n_bk1, b_grid_desc_bk0_n_bk1,
bgrad_grid_desc_bk0_n_bk1,
d0_grid_desc_m0_n0_m1_m2_n1_m3, d0_grid_desc_m0_n0_m1_m2_n1_m3,
c_grid_desc_m0_n0_m1_n1_m2_n2_m3_m4_m5_n3, c_grid_desc_m0_n0_m1_n1_m2_n2_m3_m4_m5_n3,
b1_grid_desc_bk0_n_bk1, b1_grid_desc_bk0_n_bk1,
b1grad_grid_desc_bk0_n_bk1,
lse_grid_desc_m, lse_grid_desc_m,
ygrad_grid_desc_m0_o_m1, ygrad_grid_desc_m0_o_m1,
block_2_ctile_map, block_2_ctile_map,
...@@ -287,13 +299,16 @@ __global__ void ...@@ -287,13 +299,16 @@ __global__ void
ignore = c_element_op; ignore = c_element_op;
ignore = a_grid_desc_ak0_m_ak1; ignore = a_grid_desc_ak0_m_ak1;
ignore = b_grid_desc_bk0_n_bk1; ignore = b_grid_desc_bk0_n_bk1;
ignore = bgrad_grid_desc_bk0_n_bk1;
ignore = d0_grid_desc_m0_n0_m1_m2_n1_m3; ignore = d0_grid_desc_m0_n0_m1_m2_n1_m3;
ignore = c_grid_desc_m0_n0_m1_n1_m2_n2_m3_m4_m5_n3; ignore = c_grid_desc_m0_n0_m1_n1_m2_n2_m3_m4_m5_n3;
ignore = b1_grid_desc_bk0_n_bk1; ignore = b1_grid_desc_bk0_n_bk1;
ignore = b1grad_grid_desc_bk0_n_bk1;
ignore = lse_grid_desc_m; ignore = lse_grid_desc_m;
ignore = ygrad_grid_desc_m0_o_m1; ignore = ygrad_grid_desc_m0_o_m1;
ignore = block_2_ctile_map; ignore = block_2_ctile_map;
ignore = batch_count; ignore = batch_count;
ignore = h_ratio;
ignore = nblock; ignore = nblock;
ignore = compute_base_ptr_of_batch; ignore = compute_base_ptr_of_batch;
ignore = c0_matrix_mask; ignore = c0_matrix_mask;
...@@ -704,6 +719,8 @@ struct DeviceBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2 ...@@ -704,6 +719,8 @@ struct DeviceBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2
const ZGridDesc_G_M_N& z_grid_desc_g_m_n, const ZGridDesc_G_M_N& z_grid_desc_g_m_n,
const B1GridDesc_G_N_K& b1_grid_desc_g_n_k, const B1GridDesc_G_N_K& b1_grid_desc_g_n_k,
const CGridDesc_G_M_N& c_grid_desc_g_m_n, const CGridDesc_G_M_N& c_grid_desc_g_m_n,
const BGridDesc_G_N_K& bgrad_grid_desc_g_n_k,
const B1GridDesc_G_N_K& b1grad_grid_desc_g_n_k,
index_t BatchStrideLSE) index_t BatchStrideLSE)
: a_grid_desc_g_m_k_(a_grid_desc_g_m_k), : a_grid_desc_g_m_k_(a_grid_desc_g_m_k),
b_grid_desc_g_n_k_(b_grid_desc_g_n_k), b_grid_desc_g_n_k_(b_grid_desc_g_n_k),
...@@ -711,6 +728,8 @@ struct DeviceBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2 ...@@ -711,6 +728,8 @@ struct DeviceBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2
z_grid_desc_g_m_n_(z_grid_desc_g_m_n), z_grid_desc_g_m_n_(z_grid_desc_g_m_n),
b1_grid_desc_g_n_k_(b1_grid_desc_g_n_k), b1_grid_desc_g_n_k_(b1_grid_desc_g_n_k),
c_grid_desc_g_m_n_(c_grid_desc_g_m_n), c_grid_desc_g_m_n_(c_grid_desc_g_m_n),
bgrad_grid_desc_g_n_k_(bgrad_grid_desc_g_n_k),
b1grad_grid_desc_g_n_k_(b1grad_grid_desc_g_n_k),
BatchStrideLSE_(BatchStrideLSE) BatchStrideLSE_(BatchStrideLSE)
{ {
} }
...@@ -729,6 +748,7 @@ struct DeviceBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2 ...@@ -729,6 +748,7 @@ struct DeviceBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2
{ {
return d0_grid_desc_g_m_n_.CalculateOffset(make_multi_index(g_idx, 0, 0)); return d0_grid_desc_g_m_n_.CalculateOffset(make_multi_index(g_idx, 0, 0));
} }
__host__ __device__ constexpr long_index_t GetZBasePtr(index_t g_idx) const __host__ __device__ constexpr long_index_t GetZBasePtr(index_t g_idx) const
{ {
return z_grid_desc_g_m_n_.CalculateOffset(make_multi_index(g_idx, 0, 0)); return z_grid_desc_g_m_n_.CalculateOffset(make_multi_index(g_idx, 0, 0));
...@@ -749,6 +769,16 @@ struct DeviceBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2 ...@@ -749,6 +769,16 @@ struct DeviceBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2
return g_idx * static_cast<long_index_t>(BatchStrideLSE_); return g_idx * static_cast<long_index_t>(BatchStrideLSE_);
} }
__host__ __device__ constexpr long_index_t GetBGradBasePtr(index_t g_idx) const
{
return bgrad_grid_desc_g_n_k_.CalculateOffset(make_multi_index(g_idx, 0, 0));
}
__host__ __device__ constexpr long_index_t GetB1GradBasePtr(index_t g_idx) const
{
return b1grad_grid_desc_g_n_k_.CalculateOffset(make_multi_index(g_idx, 0, 0));
}
private: private:
AGridDesc_G_M_K a_grid_desc_g_m_k_; AGridDesc_G_M_K a_grid_desc_g_m_k_;
BGridDesc_G_N_K b_grid_desc_g_n_k_; BGridDesc_G_N_K b_grid_desc_g_n_k_;
...@@ -756,6 +786,8 @@ struct DeviceBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2 ...@@ -756,6 +786,8 @@ struct DeviceBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2
ZGridDesc_G_M_N z_grid_desc_g_m_n_; ZGridDesc_G_M_N z_grid_desc_g_m_n_;
B1GridDesc_G_N_K b1_grid_desc_g_n_k_; B1GridDesc_G_N_K b1_grid_desc_g_n_k_;
CGridDesc_G_M_N c_grid_desc_g_m_n_; CGridDesc_G_M_N c_grid_desc_g_m_n_;
BGridDesc_G_N_K bgrad_grid_desc_g_n_k_;
B1GridDesc_G_N_K b1grad_grid_desc_g_n_k_;
index_t BatchStrideLSE_; index_t BatchStrideLSE_;
}; };
...@@ -874,6 +906,10 @@ struct DeviceBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2 ...@@ -874,6 +906,10 @@ struct DeviceBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2
const std::vector<index_t>& c_gs_ms_gemm1ns_lengths, // c_gs_ms_os_lengths const std::vector<index_t>& c_gs_ms_gemm1ns_lengths, // c_gs_ms_os_lengths
const std::vector<index_t>& c_gs_ms_gemm1ns_strides, // c_gs_ms_os_strides const std::vector<index_t>& c_gs_ms_gemm1ns_strides, // c_gs_ms_os_strides
const std::vector<index_t>& lse_gs_ms_lengths, const std::vector<index_t>& lse_gs_ms_lengths,
const std::vector<index_t>& bgrad_gs_ns_ks_lengths,
const std::vector<index_t>& bgrad_gs_ns_ks_strides,
const std::vector<index_t>& b1grad_gs_gemm1ns_gemm1ks_lengths,
const std::vector<index_t>& b1grad_gs_gemm1ns_gemm1ks_strides,
const std::vector<ck::index_t>& acc0_bias_gs_ms_ns_lengths, const std::vector<ck::index_t>& acc0_bias_gs_ms_ns_lengths,
const std::vector<ck::index_t>& acc0_bias_gs_ms_ns_strides, const std::vector<ck::index_t>& acc0_bias_gs_ms_ns_strides,
const std::vector<ck::index_t>& const std::vector<ck::index_t>&
...@@ -904,9 +940,13 @@ struct DeviceBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2 ...@@ -904,9 +940,13 @@ struct DeviceBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2
DeviceOp::MakeAGridDescriptor_AK0_M_AK1(a_gs_ms_ks_lengths, a_gs_ms_ks_strides)}, DeviceOp::MakeAGridDescriptor_AK0_M_AK1(a_gs_ms_ks_lengths, a_gs_ms_ks_strides)},
b_grid_desc_bk0_n_bk1_{ b_grid_desc_bk0_n_bk1_{
DeviceOp::MakeBGridDescriptor_BK0_N_BK1(b_gs_ns_ks_lengths, b_gs_ns_ks_strides)}, DeviceOp::MakeBGridDescriptor_BK0_N_BK1(b_gs_ns_ks_lengths, b_gs_ns_ks_strides)},
bgrad_grid_desc_bk0_n_bk1_{DeviceOp::MakeBGridDescriptor_BK0_N_BK1(
bgrad_gs_ns_ks_lengths, bgrad_gs_ns_ks_strides)},
z_grid_desc_m_n_{MakeZGridDescriptor_M_N(z_gs_ms_ns_lengths, z_gs_ms_ns_strides)}, z_grid_desc_m_n_{MakeZGridDescriptor_M_N(z_gs_ms_ns_lengths, z_gs_ms_ns_strides)},
b1_grid_desc_bk0_n_bk1_{DeviceOp::MakeVGridDescriptor_O0_N_O1( b1_grid_desc_bk0_n_bk1_{DeviceOp::MakeVGridDescriptor_O0_N_O1(
b1_gs_gemm1ns_gemm1ks_lengths, b1_gs_gemm1ns_gemm1ks_strides)}, b1_gs_gemm1ns_gemm1ks_lengths, b1_gs_gemm1ns_gemm1ks_strides)},
b1grad_grid_desc_bk0_n_bk1_{DeviceOp::MakeVGridDescriptor_O0_N_O1(
b1grad_gs_gemm1ns_gemm1ks_lengths, b1grad_gs_gemm1ns_gemm1ks_strides)},
y_grid_desc_m_o_{Transform::MakeCGridDescriptor_M_N(c_gs_ms_gemm1ns_lengths, y_grid_desc_m_o_{Transform::MakeCGridDescriptor_M_N(c_gs_ms_gemm1ns_lengths,
c_gs_ms_gemm1ns_strides)}, c_gs_ms_gemm1ns_strides)},
d_y_grid_desc_m_o_{DTransform::MakeCGridDescriptor_M_N(c_gs_ms_gemm1ns_lengths, d_y_grid_desc_m_o_{DTransform::MakeCGridDescriptor_M_N(c_gs_ms_gemm1ns_lengths,
...@@ -927,6 +967,10 @@ struct DeviceBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2 ...@@ -927,6 +967,10 @@ struct DeviceBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2
c_gs_ms_gemm1ns_strides)}, c_gs_ms_gemm1ns_strides)},
z_grid_desc_g_m_n_{ z_grid_desc_g_m_n_{
Transform::MakeC0GridDescriptor_G_M_N(z_gs_ms_ns_lengths, z_gs_ms_ns_strides)}, Transform::MakeC0GridDescriptor_G_M_N(z_gs_ms_ns_lengths, z_gs_ms_ns_strides)},
bgrad_grid_desc_g_n_k_{Transform::MakeB0GridDescriptor_G_N_K(bgrad_gs_ns_ks_lengths,
bgrad_gs_ns_ks_strides)},
b1grad_grid_desc_g_n_k_{Transform::MakeB1GridDescriptor_G_N_K(
b1grad_gs_gemm1ns_gemm1ks_lengths, b1grad_gs_gemm1ns_gemm1ks_strides)},
block_2_ctile_map_{GridwiseGemm::MakeDefaultBlock2CTileMap(k_grid_desc_n_k_)}, block_2_ctile_map_{GridwiseGemm::MakeDefaultBlock2CTileMap(k_grid_desc_n_k_)},
d_block_2_ctile_map_{ d_block_2_ctile_map_{
GridwiseYDotYGrad::MakeDefaultBlock2CTileMap(d_y_grid_desc_m_o_)}, GridwiseYDotYGrad::MakeDefaultBlock2CTileMap(d_y_grid_desc_m_o_)},
...@@ -950,6 +994,7 @@ struct DeviceBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2 ...@@ -950,6 +994,7 @@ struct DeviceBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2
c_mz_gemm1nz_strides_{c_gs_ms_gemm1ns_strides[NumDimG + NumDimM - 1], c_mz_gemm1nz_strides_{c_gs_ms_gemm1ns_strides[NumDimG + NumDimM - 1],
c_gs_ms_gemm1ns_strides[NumDimG + NumDimM + NumDimO - 1]}, c_gs_ms_gemm1ns_strides[NumDimG + NumDimM + NumDimO - 1]},
batch_count_{c_grid_desc_g_m_n_.GetLength(I0)}, batch_count_{c_grid_desc_g_m_n_.GetLength(I0)},
h_ratio_{c_grid_desc_g_m_n_.GetLength(I0) / b_grid_desc_g_n_k_.GetLength(I0)},
p_drop_{p_drop} p_drop_{p_drop}
{ {
// TODO: implement bias addition // TODO: implement bias addition
...@@ -979,6 +1024,8 @@ struct DeviceBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2 ...@@ -979,6 +1024,8 @@ struct DeviceBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2
z_grid_desc_g_m_n_, z_grid_desc_g_m_n_,
b1_grid_desc_g_n_k_, b1_grid_desc_g_n_k_,
c_grid_desc_g_m_n_, c_grid_desc_g_m_n_,
bgrad_grid_desc_g_n_k_,
b1grad_grid_desc_g_n_k_,
type_convert<index_t>(lse_grid_desc_m_.GetElementSpaceSize())); type_convert<index_t>(lse_grid_desc_m_.GetElementSpaceSize()));
seed_ = std::get<0>(seeds); seed_ = std::get<0>(seeds);
...@@ -1005,7 +1052,7 @@ struct DeviceBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2 ...@@ -1005,7 +1052,7 @@ struct DeviceBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2
<< b_grid_desc_g_n_k_.GetLength(I1) << ", " << b_grid_desc_g_n_k_.GetLength(I1) << ", "
<< b_grid_desc_g_n_k_.GetLength(I2) << '\n'; << b_grid_desc_g_n_k_.GetLength(I2) << '\n';
// b_grid_desc_g_n_k_.Print(); // b_grid_desc_g_n_k_.Print();
std::cout << "b1_grid_desc_g_o_n_: " << b1_grid_desc_g_n_k_.GetLength(I0) << ", " std::cout << "b1_grid_desc_g_n_k_: " << b1_grid_desc_g_n_k_.GetLength(I0) << ", "
<< b1_grid_desc_g_n_k_.GetLength(I1) << ", " << b1_grid_desc_g_n_k_.GetLength(I1) << ", "
<< b1_grid_desc_g_n_k_.GetLength(I2) << '\n'; << b1_grid_desc_g_n_k_.GetLength(I2) << '\n';
// b1_grid_desc_g_n_k_.Print(); // b1_grid_desc_g_n_k_.Print();
...@@ -1018,6 +1065,17 @@ struct DeviceBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2 ...@@ -1018,6 +1065,17 @@ struct DeviceBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2
std::cout << "ygrad_grid_desc_m0_o_m1_: " << ygrad_grid_desc_m0_o_m1_.GetLength(I0) std::cout << "ygrad_grid_desc_m0_o_m1_: " << ygrad_grid_desc_m0_o_m1_.GetLength(I0)
<< ", " << ygrad_grid_desc_m0_o_m1_.GetLength(I1) << ", " << ", " << ygrad_grid_desc_m0_o_m1_.GetLength(I1) << ", "
<< ygrad_grid_desc_m0_o_m1_.GetLength(I2) << '\n'; << ygrad_grid_desc_m0_o_m1_.GetLength(I2) << '\n';
std::cout << "d0_grid_desc_g_m_n_: " << d0_grid_desc_g_m_n_.GetLength(I0) << ", "
<< d0_grid_desc_g_m_n_.GetLength(I1) << ", "
<< d0_grid_desc_g_m_n_.GetLength(I2) << '\n';
std::cout << "bgrad_grid_desc_g_n_k_: " << bgrad_grid_desc_g_n_k_.GetLength(I0) << ", "
<< bgrad_grid_desc_g_n_k_.GetLength(I1) << ", "
<< bgrad_grid_desc_g_n_k_.GetLength(I2) << '\n';
// bgrad_grid_desc_g_n_k_.Print();
std::cout << "b1grad_grid_desc_g_n_k_: " << b1grad_grid_desc_g_n_k_.GetLength(I0)
<< ", " << b1grad_grid_desc_g_n_k_.GetLength(I1) << ", "
<< b1grad_grid_desc_g_n_k_.GetLength(I2) << '\n';
// b1grad_grid_desc_g_n_k_.Print();
} }
// pointers // pointers
...@@ -1038,9 +1096,11 @@ struct DeviceBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2 ...@@ -1038,9 +1096,11 @@ struct DeviceBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2
// tensor descriptor // tensor descriptor
AGridDesc_AK0_M_AK1 a_grid_desc_ak0_m_ak1_; AGridDesc_AK0_M_AK1 a_grid_desc_ak0_m_ak1_;
BGridDesc_BK0_N_BK1 b_grid_desc_bk0_n_bk1_; BGridDesc_BK0_N_BK1 b_grid_desc_bk0_n_bk1_;
BGridDesc_BK0_N_BK1 bgrad_grid_desc_bk0_n_bk1_;
typename GridwiseGemm::D0GridDescriptor_M0_N0_M1_M2_N1_M3 d0_grid_desc_m0_n0_m1_m2_n1_m3_; typename GridwiseGemm::D0GridDescriptor_M0_N0_M1_M2_N1_M3 d0_grid_desc_m0_n0_m1_m2_n1_m3_;
ZGridDesc_M_N z_grid_desc_m_n_; ZGridDesc_M_N z_grid_desc_m_n_;
B1GridDesc_BK0_N_BK1 b1_grid_desc_bk0_n_bk1_; B1GridDesc_BK0_N_BK1 b1_grid_desc_bk0_n_bk1_;
B1GridDesc_BK0_N_BK1 b1grad_grid_desc_bk0_n_bk1_;
YGridDesc_M_O y_grid_desc_m_o_; YGridDesc_M_O y_grid_desc_m_o_;
DYGridDesc_M_O d_y_grid_desc_m_o_; DYGridDesc_M_O d_y_grid_desc_m_o_;
LSEGridDesc_M lse_grid_desc_m_; LSEGridDesc_M lse_grid_desc_m_;
...@@ -1055,6 +1115,8 @@ struct DeviceBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2 ...@@ -1055,6 +1115,8 @@ struct DeviceBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2
B1GridDesc_G_N_K b1_grid_desc_g_n_k_; B1GridDesc_G_N_K b1_grid_desc_g_n_k_;
CGridDesc_G_M_N c_grid_desc_g_m_n_; CGridDesc_G_M_N c_grid_desc_g_m_n_;
ZGridDesc_G_M_N z_grid_desc_g_m_n_; ZGridDesc_G_M_N z_grid_desc_g_m_n_;
BGridDesc_G_N_K bgrad_grid_desc_g_n_k_;
B1GridDesc_G_N_K b1grad_grid_desc_g_n_k_;
typename GridwiseGemm::ZGridDescriptor_M0_N0_M1_N1_M2_N2_M3_M4_M5_N3 typename GridwiseGemm::ZGridDescriptor_M0_N0_M1_N1_M2_N2_M3_M4_M5_N3
c_grid_desc_m0_n0_m1_n1_m2_n2_m3_m4_m5_n3_; c_grid_desc_m0_n0_m1_n1_m2_n2_m3_m4_m5_n3_;
...@@ -1083,6 +1145,7 @@ struct DeviceBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2 ...@@ -1083,6 +1145,7 @@ struct DeviceBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2
std::vector<index_t> c_mz_gemm1nz_strides_; std::vector<index_t> c_mz_gemm1nz_strides_;
index_t batch_count_; index_t batch_count_;
index_t h_ratio_;
ComputeBasePtrOfStridedBatch compute_base_ptr_of_batch_; ComputeBasePtrOfStridedBatch compute_base_ptr_of_batch_;
float p_drop_; float p_drop_;
...@@ -1208,13 +1271,16 @@ struct DeviceBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2 ...@@ -1208,13 +1271,16 @@ struct DeviceBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2
arg.c_element_op_, arg.c_element_op_,
arg.a_grid_desc_ak0_m_ak1_, arg.a_grid_desc_ak0_m_ak1_,
arg.b_grid_desc_bk0_n_bk1_, arg.b_grid_desc_bk0_n_bk1_,
arg.bgrad_grid_desc_bk0_n_bk1_,
arg.d0_grid_desc_m0_n0_m1_m2_n1_m3_, arg.d0_grid_desc_m0_n0_m1_m2_n1_m3_,
arg.c_grid_desc_m0_n0_m1_n1_m2_n2_m3_m4_m5_n3_, arg.c_grid_desc_m0_n0_m1_n1_m2_n2_m3_m4_m5_n3_,
arg.b1_grid_desc_bk0_n_bk1_, arg.b1_grid_desc_bk0_n_bk1_,
arg.b1grad_grid_desc_bk0_n_bk1_,
arg.lse_grid_desc_m_, arg.lse_grid_desc_m_,
arg.ygrad_grid_desc_m0_o_m1_, arg.ygrad_grid_desc_m0_o_m1_,
arg.block_2_ctile_map_, arg.block_2_ctile_map_,
arg.batch_count_, arg.batch_count_,
arg.h_ratio_,
arg.block_2_ctile_map_.CalculateGridSize(arg.k_grid_desc_n_k_), arg.block_2_ctile_map_.CalculateGridSize(arg.k_grid_desc_n_k_),
arg.compute_base_ptr_of_batch_, arg.compute_base_ptr_of_batch_,
arg.c0_matrix_mask_, arg.c0_matrix_mask_,
...@@ -1280,13 +1346,15 @@ struct DeviceBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2 ...@@ -1280,13 +1346,15 @@ struct DeviceBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2
// Check if C permute dimension matches GEMM + GEMM shape // Check if C permute dimension matches GEMM + GEMM shape
const index_t c_g = arg.c_grid_desc_g_m_n_.GetLength(I0); // unpadded const index_t c_g = arg.c_grid_desc_g_m_n_.GetLength(I0); // unpadded
const index_t b_g = arg.b_grid_desc_g_n_k_.GetLength(I0);
const index_t c_m = arg.y_grid_desc_m_o_.GetLength(I0); const index_t c_m = arg.y_grid_desc_m_o_.GetLength(I0);
const index_t c_gemm1n = arg.y_grid_desc_m_o_.GetLength(I1); const index_t c_gemm1n = arg.y_grid_desc_m_o_.GetLength(I1);
const index_t a_m = arg.a_grid_desc_ak0_m_ak1_.GetLength(I1); const index_t a_m = arg.a_grid_desc_ak0_m_ak1_.GetLength(I1);
const index_t b1_gemm1n = const index_t b1_gemm1n =
arg.b1_grid_desc_bk0_n_bk1_.GetLength(I0) * arg.b1_grid_desc_bk0_n_bk1_.GetLength(I2); arg.b1_grid_desc_bk0_n_bk1_.GetLength(I0) * arg.b1_grid_desc_bk0_n_bk1_.GetLength(I2);
if(!(c_g == arg.batch_count_ && c_m == a_m && c_gemm1n == b1_gemm1n)) if(!(c_g == arg.batch_count_ && c_m == a_m && c_gemm1n == b1_gemm1n && c_g % b_g == 0 &&
b_g <= c_g))
{ {
return false; return false;
} }
...@@ -1380,6 +1448,10 @@ struct DeviceBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2 ...@@ -1380,6 +1448,10 @@ struct DeviceBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2
const std::vector<index_t>& c_gs_ms_gemm1ns_lengths, // c_gs_ms_os_lengths const std::vector<index_t>& c_gs_ms_gemm1ns_lengths, // c_gs_ms_os_lengths
const std::vector<index_t>& c_gs_ms_gemm1ns_strides, // c_gs_ms_os_strides const std::vector<index_t>& c_gs_ms_gemm1ns_strides, // c_gs_ms_os_strides
const std::vector<index_t>& lse_gs_ms_lengths, const std::vector<index_t>& lse_gs_ms_lengths,
const std::vector<index_t>& bgrad_gs_ns_ks_lengths,
const std::vector<index_t>& bgrad_gs_ns_ks_strides,
const std::vector<index_t>& b1grad_gs_gemm1ns_gemm1ks_lengths,
const std::vector<index_t>& b1grad_gs_gemm1ns_gemm1ks_strides,
const std::vector<ck::index_t>& acc0_bias_gs_ms_ns_lengths, const std::vector<ck::index_t>& acc0_bias_gs_ms_ns_lengths,
const std::vector<ck::index_t>& acc0_bias_gs_ms_ns_strides, const std::vector<ck::index_t>& acc0_bias_gs_ms_ns_strides,
const std::vector<ck::index_t>& const std::vector<ck::index_t>&
...@@ -1420,6 +1492,10 @@ struct DeviceBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2 ...@@ -1420,6 +1492,10 @@ struct DeviceBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2
c_gs_ms_gemm1ns_lengths, // c_gs_ms_os_lengths c_gs_ms_gemm1ns_lengths, // c_gs_ms_os_lengths
c_gs_ms_gemm1ns_strides, // c_gs_ms_os_strides c_gs_ms_gemm1ns_strides, // c_gs_ms_os_strides
lse_gs_ms_lengths, lse_gs_ms_lengths,
bgrad_gs_ns_ks_lengths,
bgrad_gs_ns_ks_strides,
b1grad_gs_gemm1ns_gemm1ks_lengths,
b1grad_gs_gemm1ns_gemm1ks_strides,
acc0_bias_gs_ms_ns_lengths, acc0_bias_gs_ms_ns_lengths,
acc0_bias_gs_ms_ns_strides, acc0_bias_gs_ms_ns_strides,
acc1_bias_gs_ms_gemm1ns_lengths, // acc1_bias_gs_ms_os_lengths acc1_bias_gs_ms_gemm1ns_lengths, // acc1_bias_gs_ms_os_lengths
...@@ -1464,6 +1540,10 @@ struct DeviceBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2 ...@@ -1464,6 +1540,10 @@ struct DeviceBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2
const std::vector<index_t>& c_gs_ms_gemm1ns_lengths, // c_gs_ms_os_lengths const std::vector<index_t>& c_gs_ms_gemm1ns_lengths, // c_gs_ms_os_lengths
const std::vector<index_t>& c_gs_ms_gemm1ns_strides, // c_gs_ms_os_strides const std::vector<index_t>& c_gs_ms_gemm1ns_strides, // c_gs_ms_os_strides
const std::vector<index_t>& lse_gs_ms_lengths, const std::vector<index_t>& lse_gs_ms_lengths,
const std::vector<index_t>& bgrad_gs_ns_ks_lengths,
const std::vector<index_t>& bgrad_gs_ns_ks_strides,
const std::vector<index_t>& b1grad_gs_gemm1ns_gemm1ks_lengths,
const std::vector<index_t>& b1grad_gs_gemm1ns_gemm1ks_strides,
const std::vector<ck::index_t>& acc0_bias_gs_ms_ns_lengths, const std::vector<ck::index_t>& acc0_bias_gs_ms_ns_lengths,
const std::vector<ck::index_t>& acc0_bias_gs_ms_ns_strides, const std::vector<ck::index_t>& acc0_bias_gs_ms_ns_strides,
const std::vector<ck::index_t>& const std::vector<ck::index_t>&
...@@ -1505,6 +1585,10 @@ struct DeviceBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2 ...@@ -1505,6 +1585,10 @@ struct DeviceBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2
c_gs_ms_gemm1ns_lengths, // c_gs_ms_os_lengths c_gs_ms_gemm1ns_lengths, // c_gs_ms_os_lengths
c_gs_ms_gemm1ns_strides, // c_gs_ms_os_strides c_gs_ms_gemm1ns_strides, // c_gs_ms_os_strides
lse_gs_ms_lengths, lse_gs_ms_lengths,
bgrad_gs_ns_ks_lengths,
bgrad_gs_ns_ks_strides,
b1grad_gs_gemm1ns_gemm1ks_lengths,
b1grad_gs_gemm1ns_gemm1ks_strides,
acc0_bias_gs_ms_ns_lengths, acc0_bias_gs_ms_ns_lengths,
acc0_bias_gs_ms_ns_strides, acc0_bias_gs_ms_ns_strides,
acc1_bias_gs_ms_gemm1ns_lengths, acc1_bias_gs_ms_gemm1ns_lengths,
......
include/ck/tensor_operation/gpu/device/impl/device_grouped_mha_bwd_xdl_cshuffle_qloop_light_v1.hpp
View file @ 104aeabc
...@@ -103,6 +103,7 @@ __global__ void ...@@ -103,6 +103,7 @@ __global__ void
kernel_grouped_multihead_attention_backward_qloop_xdl_cshuffle_light_v1( kernel_grouped_multihead_attention_backward_qloop_xdl_cshuffle_light_v1(
const void CK_CONSTANT_ADDRESS_SPACE* group_kernel_args, const void CK_CONSTANT_ADDRESS_SPACE* group_kernel_args,
const index_t group_count, const index_t group_count,
const index_t h_ratio,
const AElementwiseOperation a_element_op, const AElementwiseOperation a_element_op,
const BElementwiseOperation b_element_op, const BElementwiseOperation b_element_op,
const AccElementwiseOperation acc_element_op, const AccElementwiseOperation acc_element_op,
...@@ -141,19 +142,26 @@ __global__ void ...@@ -141,19 +142,26 @@ __global__ void
const index_t num_blocks_per_batch = arg_ptr[group_id].num_blocks_per_batch_; const index_t num_blocks_per_batch = arg_ptr[group_id].num_blocks_per_batch_;
const index_t g_idx = __builtin_amdgcn_readfirstlane( const index_t g_idx = __builtin_amdgcn_readfirstlane(
(block_id - arg_ptr[group_id].block_start_) / (Deterministic ? 1 : num_blocks_per_batch)); (block_id - arg_ptr[group_id].block_start_) / (Deterministic ? 1 : num_blocks_per_batch));
const index_t gkv_idx = __builtin_amdgcn_readfirstlane(g_idx / h_ratio);
const long_index_t a_batch_offset = __builtin_amdgcn_readfirstlane( const long_index_t a_batch_offset = __builtin_amdgcn_readfirstlane(
static_cast<long_index_t>(arg_ptr[group_id].compute_base_ptr_of_batch_.GetABasePtr(g_idx))); static_cast<long_index_t>(arg_ptr[group_id].compute_base_ptr_of_batch_.GetABasePtr(g_idx)));
const long_index_t b_batch_offset = __builtin_amdgcn_readfirstlane( const long_index_t b_batch_offset = __builtin_amdgcn_readfirstlane(static_cast<long_index_t>(
static_cast<long_index_t>(arg_ptr[group_id].compute_base_ptr_of_batch_.GetBBasePtr(g_idx))); arg_ptr[group_id].compute_base_ptr_of_batch_.GetBBasePtr(gkv_idx)));
const long_index_t z_batch_offset = __builtin_amdgcn_readfirstlane( const long_index_t z_batch_offset = __builtin_amdgcn_readfirstlane(
static_cast<long_index_t>(arg_ptr[group_id].compute_base_ptr_of_batch_.GetZBasePtr(g_idx))); static_cast<long_index_t>(arg_ptr[group_id].compute_base_ptr_of_batch_.GetZBasePtr(g_idx)));
const long_index_t b1_batch_offset = __builtin_amdgcn_readfirstlane(static_cast<long_index_t>( const long_index_t b1_batch_offset = __builtin_amdgcn_readfirstlane(static_cast<long_index_t>(
arg_ptr[group_id].compute_base_ptr_of_batch_.GetB1BasePtr(g_idx))); arg_ptr[group_id].compute_base_ptr_of_batch_.GetB1BasePtr(gkv_idx)));
const long_index_t c_batch_offset = __builtin_amdgcn_readfirstlane( const long_index_t c_batch_offset = __builtin_amdgcn_readfirstlane(
static_cast<long_index_t>(arg_ptr[group_id].compute_base_ptr_of_batch_.GetCBasePtr(g_idx))); static_cast<long_index_t>(arg_ptr[group_id].compute_base_ptr_of_batch_.GetCBasePtr(g_idx)));
const long_index_t lse_batch_offset = __builtin_amdgcn_readfirstlane(static_cast<long_index_t>( const long_index_t lse_batch_offset = __builtin_amdgcn_readfirstlane(static_cast<long_index_t>(
arg_ptr[group_id].compute_base_ptr_of_batch_.GetLSEBasePtr(g_idx))); arg_ptr[group_id].compute_base_ptr_of_batch_.GetLSEBasePtr(g_idx)));
const long_index_t bgrad_batch_offset =
__builtin_amdgcn_readfirstlane(static_cast<long_index_t>(
arg_ptr[group_id].compute_base_ptr_of_batch_.GetBGradBasePtr(g_idx)));
const long_index_t b1grad_batch_offset =
__builtin_amdgcn_readfirstlane(static_cast<long_index_t>(
arg_ptr[group_id].compute_base_ptr_of_batch_.GetB1GradBasePtr(g_idx)));
const index_t global_thread_id = get_thread_global_1d_id(); const index_t global_thread_id = get_thread_global_1d_id();
ck::philox ph(seed, global_thread_id, offset); ck::philox ph(seed, global_thread_id, offset);
...@@ -168,6 +176,7 @@ __global__ void ...@@ -168,6 +176,7 @@ __global__ void
const long_index_t d0_batch_offset = const long_index_t d0_batch_offset =
__builtin_amdgcn_readfirstlane(static_cast<long_index_t>( __builtin_amdgcn_readfirstlane(static_cast<long_index_t>(
arg_ptr[group_id].compute_base_ptr_of_batch_.GetD0BasePtr(g_idx))); arg_ptr[group_id].compute_base_ptr_of_batch_.GetD0BasePtr(g_idx)));
if(arg_ptr[group_id].p_d0_grid_ != nullptr) if(arg_ptr[group_id].p_d0_grid_ != nullptr)
tmp_p_d0_grid = arg_ptr[group_id].p_d0_grid_ + d0_batch_offset; tmp_p_d0_grid = arg_ptr[group_id].p_d0_grid_ + d0_batch_offset;
if(arg_ptr[group_id].p_d0grad_grid_) if(arg_ptr[group_id].p_d0grad_grid_)
...@@ -187,9 +196,9 @@ __global__ void ...@@ -187,9 +196,9 @@ __global__ void
arg_ptr[group_id].p_d_grid_ + lse_batch_offset, arg_ptr[group_id].p_d_grid_ + lse_batch_offset,
arg_ptr[group_id].p_ygrad_grid_ + c_batch_offset, arg_ptr[group_id].p_ygrad_grid_ + c_batch_offset,
arg_ptr[group_id].p_qgrad_grid_ + a_batch_offset, arg_ptr[group_id].p_qgrad_grid_ + a_batch_offset,
arg_ptr[group_id].p_kgrad_grid_ + b_batch_offset, arg_ptr[group_id].p_kgrad_grid_ + bgrad_batch_offset,
tmp_p_d0grad_grid, tmp_p_d0grad_grid,
arg_ptr[group_id].p_vgrad_grid_ + b1_batch_offset, arg_ptr[group_id].p_vgrad_grid_ + b1grad_batch_offset,
p_shared, p_shared,
a_element_op, a_element_op,
b_element_op, b_element_op,
...@@ -198,9 +207,11 @@ __global__ void ...@@ -198,9 +207,11 @@ __global__ void
c_element_op, c_element_op,
arg_ptr[group_id].a_grid_desc_ak0_m_ak1_, arg_ptr[group_id].a_grid_desc_ak0_m_ak1_,
arg_ptr[group_id].b_grid_desc_bk0_n_bk1_, arg_ptr[group_id].b_grid_desc_bk0_n_bk1_,
arg_ptr[group_id].bgrad_grid_desc_bk0_n_bk1_,
arg_ptr[group_id].d0_grid_desc_m0_n0_m1_m2_n1_m3_, arg_ptr[group_id].d0_grid_desc_m0_n0_m1_m2_n1_m3_,
arg_ptr[group_id].c_grid_desc_m0_n0_m1_n1_m2_n2_m3_m4_m5_n3_, arg_ptr[group_id].c_grid_desc_m0_n0_m1_n1_m2_n2_m3_m4_m5_n3_,
arg_ptr[group_id].b1_grid_desc_bk0_n_bk1_, arg_ptr[group_id].b1_grid_desc_bk0_n_bk1_,
arg_ptr[group_id].b1grad_grid_desc_bk0_n_bk1_,
arg_ptr[group_id].lse_grid_desc_m_, arg_ptr[group_id].lse_grid_desc_m_,
arg_ptr[group_id].ygrad_grid_desc_o0_m_o1_, arg_ptr[group_id].ygrad_grid_desc_o0_m_o1_,
arg_ptr[group_id].block_2_ctile_map_, arg_ptr[group_id].block_2_ctile_map_,
...@@ -225,9 +236,9 @@ __global__ void ...@@ -225,9 +236,9 @@ __global__ void
arg_ptr[group_id].p_d_grid_ + lse_batch_offset, arg_ptr[group_id].p_d_grid_ + lse_batch_offset,
arg_ptr[group_id].p_ygrad_grid_ + c_batch_offset, arg_ptr[group_id].p_ygrad_grid_ + c_batch_offset,
arg_ptr[group_id].p_qgrad_grid_ + a_batch_offset, arg_ptr[group_id].p_qgrad_grid_ + a_batch_offset,
arg_ptr[group_id].p_kgrad_grid_ + b_batch_offset, arg_ptr[group_id].p_kgrad_grid_ + bgrad_batch_offset,
tmp_p_d0grad_grid, tmp_p_d0grad_grid,
arg_ptr[group_id].p_vgrad_grid_ + b1_batch_offset, arg_ptr[group_id].p_vgrad_grid_ + b1grad_batch_offset,
p_shared, p_shared,
a_element_op, a_element_op,
b_element_op, b_element_op,
...@@ -236,9 +247,11 @@ __global__ void ...@@ -236,9 +247,11 @@ __global__ void
c_element_op, c_element_op,
arg_ptr[group_id].a_grid_desc_ak0_m_ak1_, arg_ptr[group_id].a_grid_desc_ak0_m_ak1_,
arg_ptr[group_id].b_grid_desc_bk0_n_bk1_, arg_ptr[group_id].b_grid_desc_bk0_n_bk1_,
arg_ptr[group_id].bgrad_grid_desc_bk0_n_bk1_,
arg_ptr[group_id].d0_grid_desc_m0_n0_m1_m2_n1_m3_, arg_ptr[group_id].d0_grid_desc_m0_n0_m1_m2_n1_m3_,
arg_ptr[group_id].c_grid_desc_m0_n0_m1_n1_m2_n2_m3_m4_m5_n3_, arg_ptr[group_id].c_grid_desc_m0_n0_m1_n1_m2_n2_m3_m4_m5_n3_,
arg_ptr[group_id].b1_grid_desc_bk0_n_bk1_, arg_ptr[group_id].b1_grid_desc_bk0_n_bk1_,
arg_ptr[group_id].b1grad_grid_desc_bk0_n_bk1_,
arg_ptr[group_id].lse_grid_desc_m_, arg_ptr[group_id].lse_grid_desc_m_,
arg_ptr[group_id].ygrad_grid_desc_o0_m_o1_, arg_ptr[group_id].ygrad_grid_desc_o0_m_o1_,
arg_ptr[group_id].block_2_ctile_map_, arg_ptr[group_id].block_2_ctile_map_,
...@@ -253,6 +266,7 @@ __global__ void ...@@ -253,6 +266,7 @@ __global__ void
#else #else
ignore = group_kernel_args; ignore = group_kernel_args;
ignore = group_count; ignore = group_count;
ignore = h_ratio;
ignore = a_element_op; ignore = a_element_op;
ignore = b_element_op; ignore = b_element_op;
ignore = acc_element_op; ignore = acc_element_op;
...@@ -366,6 +380,12 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1 ...@@ -366,6 +380,12 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1
std::vector<index_t> lse_gs_ms_lengths; std::vector<index_t> lse_gs_ms_lengths;
std::vector<index_t> lse_gs_ms_strides; std::vector<index_t> lse_gs_ms_strides;
std::vector<index_t> bgrad_gs_ns_ks_lengths;
std::vector<index_t> bgrad_gs_ns_ks_strides;
std::vector<index_t> b1grad_gs_gemm1ns_gemm1ks_lengths;
std::vector<index_t> b1grad_gs_gemm1ns_gemm1ks_strides;
std::vector<index_t> acc0_bias_gs_ms_ns_lengths; std::vector<index_t> acc0_bias_gs_ms_ns_lengths;
std::vector<index_t> acc0_bias_gs_ms_ns_strides; std::vector<index_t> acc0_bias_gs_ms_ns_strides;
...@@ -576,7 +596,6 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1 ...@@ -576,7 +596,6 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1
static auto MakeD0GridDescriptor_M_N(const std::vector<ck::index_t>& acc0_bias_gs_ms_ns_lengths, static auto MakeD0GridDescriptor_M_N(const std::vector<ck::index_t>& acc0_bias_gs_ms_ns_lengths,
const std::vector<ck::index_t>& acc0_bias_gs_ms_ns_strides) const std::vector<ck::index_t>& acc0_bias_gs_ms_ns_strides)
{ {
return Transform::MakeC0GridDescriptor_M_N(acc0_bias_gs_ms_ns_lengths, return Transform::MakeC0GridDescriptor_M_N(acc0_bias_gs_ms_ns_lengths,
acc0_bias_gs_ms_ns_strides); acc0_bias_gs_ms_ns_strides);
} }
...@@ -585,7 +604,6 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1 ...@@ -585,7 +604,6 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1
MakeD0GridDescriptor_G_M_N(const std::vector<ck::index_t>& acc0_bias_gs_ms_ns_lengths, MakeD0GridDescriptor_G_M_N(const std::vector<ck::index_t>& acc0_bias_gs_ms_ns_lengths,
const std::vector<ck::index_t>& acc0_bias_gs_ms_ns_strides) const std::vector<ck::index_t>& acc0_bias_gs_ms_ns_strides)
{ {
return Transform::MakeC0GridDescriptor_G_M_N(acc0_bias_gs_ms_ns_lengths, return Transform::MakeC0GridDescriptor_G_M_N(acc0_bias_gs_ms_ns_lengths,
acc0_bias_gs_ms_ns_strides); acc0_bias_gs_ms_ns_strides);
} }
...@@ -625,7 +643,7 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1 ...@@ -625,7 +643,7 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1
using D0GridDesc_G_M_N = decltype(MakeD0GridDescriptor_G_M_N({}, {})); using D0GridDesc_G_M_N = decltype(MakeD0GridDescriptor_G_M_N({}, {}));
using B1GridDesc_G_N_K = decltype(Transform::MakeB1GridDescriptor_G_N_K({}, {})); using B1GridDesc_G_N_K = decltype(Transform::MakeB1GridDescriptor_G_N_K({}, {}));
using CGridDesc_G_M_N = decltype(Transform::MakeCGridDescriptor_G_M_N({}, {})); using CGridDesc_G_M_N = decltype(Transform::MakeCGridDescriptor_G_M_N({}, {}));
using ZGridDesc_G_M_N = decltype(Transform::MakeCGridDescriptor_G_M_N({}, {})); using ZGridDesc_G_M_N = decltype(Transform::MakeC0GridDescriptor_G_M_N({}, {}));
using KGridDesc_N_K = decltype(Transform::MakeB0GridDescriptor_N_K({}, {})); using KGridDesc_N_K = decltype(Transform::MakeB0GridDescriptor_N_K({}, {}));
using D0GridDesc_M_N = decltype(MakeD0GridDescriptor_M_N({}, {})); using D0GridDesc_M_N = decltype(MakeD0GridDescriptor_M_N({}, {}));
...@@ -660,6 +678,8 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1 ...@@ -660,6 +678,8 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1
const ZGridDesc_G_M_N& z_grid_desc_g_m_n, const ZGridDesc_G_M_N& z_grid_desc_g_m_n,
const B1GridDesc_G_N_K& b1_grid_desc_g_n_k, const B1GridDesc_G_N_K& b1_grid_desc_g_n_k,
const CGridDesc_G_M_N& c_grid_desc_g_m_n, const CGridDesc_G_M_N& c_grid_desc_g_m_n,
const BGridDesc_G_N_K& bgrad_grid_desc_g_n_k,
const B1GridDesc_G_N_K& b1grad_grid_desc_g_n_k,
index_t batch_stride_lse) index_t batch_stride_lse)
: a_grid_desc_g_m_k_(a_grid_desc_g_m_k), : a_grid_desc_g_m_k_(a_grid_desc_g_m_k),
b_grid_desc_g_n_k_(b_grid_desc_g_n_k), b_grid_desc_g_n_k_(b_grid_desc_g_n_k),
...@@ -667,6 +687,8 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1 ...@@ -667,6 +687,8 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1
z_grid_desc_g_m_n_(z_grid_desc_g_m_n), z_grid_desc_g_m_n_(z_grid_desc_g_m_n),
b1_grid_desc_g_n_k_(b1_grid_desc_g_n_k), b1_grid_desc_g_n_k_(b1_grid_desc_g_n_k),
c_grid_desc_g_m_n_(c_grid_desc_g_m_n), c_grid_desc_g_m_n_(c_grid_desc_g_m_n),
bgrad_grid_desc_g_n_k_(bgrad_grid_desc_g_n_k),
b1grad_grid_desc_g_n_k_(b1grad_grid_desc_g_n_k),
batch_stride_lse_(batch_stride_lse) batch_stride_lse_(batch_stride_lse)
{ {
} }
...@@ -706,6 +728,16 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1 ...@@ -706,6 +728,16 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1
return g_idx * static_cast<long_index_t>(batch_stride_lse_); return g_idx * static_cast<long_index_t>(batch_stride_lse_);
} }
__host__ __device__ constexpr long_index_t GetBGradBasePtr(index_t g_idx) const
{
return bgrad_grid_desc_g_n_k_.CalculateOffset(make_multi_index(g_idx, 0, 0));
}
__host__ __device__ constexpr long_index_t GetB1GradBasePtr(index_t g_idx) const
{
return b1grad_grid_desc_g_n_k_.CalculateOffset(make_multi_index(g_idx, 0, 0));
}
private: private:
AGridDesc_G_M_K a_grid_desc_g_m_k_; AGridDesc_G_M_K a_grid_desc_g_m_k_;
BGridDesc_G_N_K b_grid_desc_g_n_k_; BGridDesc_G_N_K b_grid_desc_g_n_k_;
...@@ -713,6 +745,8 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1 ...@@ -713,6 +745,8 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1
ZGridDesc_G_M_N z_grid_desc_g_m_n_; ZGridDesc_G_M_N z_grid_desc_g_m_n_;
B1GridDesc_G_N_K b1_grid_desc_g_n_k_; B1GridDesc_G_N_K b1_grid_desc_g_n_k_;
CGridDesc_G_M_N c_grid_desc_g_m_n_; CGridDesc_G_M_N c_grid_desc_g_m_n_;
BGridDesc_G_N_K bgrad_grid_desc_g_n_k_;
B1GridDesc_G_N_K b1grad_grid_desc_g_n_k_;
index_t batch_stride_lse_; index_t batch_stride_lse_;
}; };
...@@ -817,9 +851,11 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1 ...@@ -817,9 +851,11 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1
// tensor descriptors for block/thread-wise copy // tensor descriptors for block/thread-wise copy
AGridDesc_AK0_M_AK1 a_grid_desc_ak0_m_ak1_; AGridDesc_AK0_M_AK1 a_grid_desc_ak0_m_ak1_;
BGridDesc_BK0_N_BK1 b_grid_desc_bk0_n_bk1_; BGridDesc_BK0_N_BK1 b_grid_desc_bk0_n_bk1_;
BGridDesc_BK0_N_BK1 bgrad_grid_desc_bk0_n_bk1_;
typename GridwiseGemm::D0GridDescriptor_M0_N0_M1_M2_N1_M3 d0_grid_desc_m0_n0_m1_m2_n1_m3_; typename GridwiseGemm::D0GridDescriptor_M0_N0_M1_M2_N1_M3 d0_grid_desc_m0_n0_m1_m2_n1_m3_;
ZGridDesc_M_N z_grid_desc_m_n_; ZGridDesc_M_N z_grid_desc_m_n_;
B1GridDesc_BK0_N_BK1 b1_grid_desc_bk0_n_bk1_; B1GridDesc_BK0_N_BK1 b1_grid_desc_bk0_n_bk1_;
B1GridDesc_BK0_N_BK1 b1grad_grid_desc_bk0_n_bk1_;
YGridDesc_M_O y_grid_desc_m_o_; YGridDesc_M_O y_grid_desc_m_o_;
typename GridwiseGemm::ZGridDescriptor_M0_N0_M1_N1_M2_N2_M3_M4_M5_N3 typename GridwiseGemm::ZGridDescriptor_M0_N0_M1_N1_M2_N2_M3_M4_M5_N3
...@@ -861,6 +897,7 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1 ...@@ -861,6 +897,7 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1
std::vector<index_t> c_mz_gemm1nz_strides_; std::vector<index_t> c_mz_gemm1nz_strides_;
// for gridwise gemm check // for gridwise gemm check
BGridDesc_G_N_K b_grid_desc_g_n_k_;
CGridDesc_G_M_N c_grid_desc_g_m_n_; CGridDesc_G_M_N c_grid_desc_g_m_n_;
index_t batch_count_; index_t batch_count_;
...@@ -893,13 +930,15 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1 ...@@ -893,13 +930,15 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1
B1ElementwiseOperation b1_element_op, B1ElementwiseOperation b1_element_op,
CElementwiseOperation c_element_op, CElementwiseOperation c_element_op,
float p_drop, float p_drop,
index_t h_ratio,
std::tuple<unsigned long long, unsigned long long> seeds) std::tuple<unsigned long long, unsigned long long> seeds)
: a_element_op_{a_element_op}, : a_element_op_{a_element_op},
b_element_op_{b_element_op}, b_element_op_{b_element_op},
acc_element_op_{acc_element_op}, acc_element_op_{acc_element_op},
b1_element_op_{b1_element_op}, b1_element_op_{b1_element_op},
c_element_op_{c_element_op}, c_element_op_{c_element_op},
p_dropout_{p_drop} p_dropout_{p_drop},
h_ratio_{h_ratio}
{ {
seed_ = std::get<0>(seeds); seed_ = std::get<0>(seeds);
offset_ = std::get<1>(seeds); offset_ = std::get<1>(seeds);
...@@ -960,6 +999,8 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1 ...@@ -960,6 +999,8 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1
problem_desc.a_gs_ms_ks_lengths, problem_desc.a_gs_ms_ks_strides); problem_desc.a_gs_ms_ks_lengths, problem_desc.a_gs_ms_ks_strides);
const auto b_grid_desc_bk0_n_bk1 = DeviceOp::MakeBGridDescriptor_BK0_N_BK1( const auto b_grid_desc_bk0_n_bk1 = DeviceOp::MakeBGridDescriptor_BK0_N_BK1(
problem_desc.b_gs_ns_ks_lengths, problem_desc.b_gs_ns_ks_strides); problem_desc.b_gs_ns_ks_lengths, problem_desc.b_gs_ns_ks_strides);
const auto bgrad_grid_desc_bk0_n_bk1 = DeviceOp::MakeBGridDescriptor_BK0_N_BK1(
problem_desc.bgrad_gs_ns_ks_lengths, problem_desc.bgrad_gs_ns_ks_strides);
std::vector<index_t> tmp_d0_gs_ms_ns_lengths; std::vector<index_t> tmp_d0_gs_ms_ns_lengths;
std::vector<index_t> tmp_d0_gs_ms_ns_strides; std::vector<index_t> tmp_d0_gs_ms_ns_strides;
...@@ -982,6 +1023,9 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1 ...@@ -982,6 +1023,9 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1
const auto b1_grid_desc_bk0_n_bk1 = DeviceOp::MakeVGridDescriptor_O0_N_O1( const auto b1_grid_desc_bk0_n_bk1 = DeviceOp::MakeVGridDescriptor_O0_N_O1(
problem_desc.b1_gs_gemm1ns_gemm1ks_lengths, problem_desc.b1_gs_gemm1ns_gemm1ks_lengths,
problem_desc.b1_gs_gemm1ns_gemm1ks_strides); problem_desc.b1_gs_gemm1ns_gemm1ks_strides);
const auto b1grad_grid_desc_bk0_n_bk1 = DeviceOp::MakeVGridDescriptor_O0_N_O1(
problem_desc.b1grad_gs_gemm1ns_gemm1ks_lengths,
problem_desc.b1grad_gs_gemm1ns_gemm1ks_strides);
const auto y_grid_desc_m_o = Transform::MakeCGridDescriptor_M_N( const auto y_grid_desc_m_o = Transform::MakeCGridDescriptor_M_N(
problem_desc.c_gs_ms_gemm1ns_lengths, problem_desc.c_gs_ms_gemm1ns_strides); problem_desc.c_gs_ms_gemm1ns_lengths, problem_desc.c_gs_ms_gemm1ns_strides);
...@@ -1005,6 +1049,11 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1 ...@@ -1005,6 +1049,11 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1
problem_desc.b1_gs_gemm1ns_gemm1ks_strides); problem_desc.b1_gs_gemm1ns_gemm1ks_strides);
const auto c_grid_desc_g_m_n = Transform::MakeCGridDescriptor_G_M_N( const auto c_grid_desc_g_m_n = Transform::MakeCGridDescriptor_G_M_N(
problem_desc.c_gs_ms_gemm1ns_lengths, problem_desc.c_gs_ms_gemm1ns_strides); problem_desc.c_gs_ms_gemm1ns_lengths, problem_desc.c_gs_ms_gemm1ns_strides);
const auto bgrad_grid_desc_g_n_k = Transform::MakeB0GridDescriptor_G_N_K(
problem_desc.bgrad_gs_ns_ks_lengths, problem_desc.bgrad_gs_ns_ks_strides);
const auto b1grad_grid_desc_g_n_k = Transform::MakeB1GridDescriptor_G_N_K(
problem_desc.b1grad_gs_gemm1ns_gemm1ks_lengths,
problem_desc.b1grad_gs_gemm1ns_gemm1ks_strides);
typename GridwiseGemm::ZGridDescriptor_M0_N0_M1_N1_M2_N2_M3_M4_M5_N3 typename GridwiseGemm::ZGridDescriptor_M0_N0_M1_N1_M2_N2_M3_M4_M5_N3
c_grid_desc_m0_n0_m1_n1_m2_n2_m3_m4_m5_n3; c_grid_desc_m0_n0_m1_n1_m2_n2_m3_m4_m5_n3;
const index_t BlockStart = grid_size_; const index_t BlockStart = grid_size_;
...@@ -1027,6 +1076,8 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1 ...@@ -1027,6 +1076,8 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1
z_grid_desc_g_m_n, z_grid_desc_g_m_n,
b1_grid_desc_g_n_k, b1_grid_desc_g_n_k,
c_grid_desc_g_m_n, c_grid_desc_g_m_n,
bgrad_grid_desc_g_n_k,
b1grad_grid_desc_g_n_k,
type_convert<index_t>(lse_grid_desc_m.GetElementSpaceSize())); type_convert<index_t>(lse_grid_desc_m.GetElementSpaceSize()));
// C0 mask // C0 mask
...@@ -1073,9 +1124,11 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1 ...@@ -1073,9 +1124,11 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1
p_vgrad_grid, p_vgrad_grid,
a_grid_desc_ak0_m_ak1, a_grid_desc_ak0_m_ak1,
b_grid_desc_bk0_n_bk1, b_grid_desc_bk0_n_bk1,
bgrad_grid_desc_bk0_n_bk1,
d0_grid_desc_m0_n0_m1_m2_n1_m3, d0_grid_desc_m0_n0_m1_m2_n1_m3,
z_grid_desc_m_n, z_grid_desc_m_n,
b1_grid_desc_bk0_n_bk1, b1_grid_desc_bk0_n_bk1,
b1grad_grid_desc_bk0_n_bk1,
y_grid_desc_m_o, y_grid_desc_m_o,
c_grid_desc_m0_n0_m1_n1_m2_n2_m3_m4_m5_n3, c_grid_desc_m0_n0_m1_n1_m2_n2_m3_m4_m5_n3,
lse_grid_desc_m, lse_grid_desc_m,
...@@ -1119,6 +1172,7 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1 ...@@ -1119,6 +1172,7 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1
problem_desc.b1_gs_gemm1ns_gemm1ks_strides[NumDimG + NumDimO + NumDimN - 1]}, problem_desc.b1_gs_gemm1ns_gemm1ks_strides[NumDimG + NumDimO + NumDimN - 1]},
{problem_desc.c_gs_ms_gemm1ns_strides[NumDimG + NumDimM - 1], {problem_desc.c_gs_ms_gemm1ns_strides[NumDimG + NumDimM - 1],
problem_desc.c_gs_ms_gemm1ns_strides[NumDimG + NumDimM + NumDimO - 1]}, problem_desc.c_gs_ms_gemm1ns_strides[NumDimG + NumDimM + NumDimO - 1]},
b_grid_desc_g_n_k,
c_grid_desc_g_m_n, c_grid_desc_g_m_n,
batch_count, batch_count,
d0_n_length_stride}); d0_n_length_stride});
...@@ -1145,6 +1199,7 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1 ...@@ -1145,6 +1199,7 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1
index_t grid_size_; index_t grid_size_;
index_t group_count_; index_t group_count_;
index_t h_ratio_;
std::vector<GroupKernelArg> group_kernel_args_; std::vector<GroupKernelArg> group_kernel_args_;
std::vector<GroupDeviceArg> group_device_args_; std::vector<GroupDeviceArg> group_device_args_;
...@@ -1224,6 +1279,7 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1 ...@@ -1224,6 +1279,7 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1
0, 0,
cast_pointer_to_constant_address_space(arg.p_workspace_), cast_pointer_to_constant_address_space(arg.p_workspace_),
arg.group_count_, arg.group_count_,
arg.h_ratio_,
arg.a_element_op_, arg.a_element_op_,
arg.b_element_op_, arg.b_element_op_,
arg.acc_element_op_, arg.acc_element_op_,
...@@ -1292,13 +1348,15 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1 ...@@ -1292,13 +1348,15 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1
const auto& device_arg = arg.group_device_args_[i]; const auto& device_arg = arg.group_device_args_[i];
// Check if C permute dimension matches GEMM + GEMM shape // Check if C permute dimension matches GEMM + GEMM shape
const index_t c_g = device_arg.c_grid_desc_g_m_n_.GetLength(I0); // unpadded const index_t c_g = device_arg.c_grid_desc_g_m_n_.GetLength(I0); // unpadded
const index_t b_g = device_arg.b_grid_desc_g_n_k_.GetLength(I0);
const index_t c_m = kernel_arg.y_grid_desc_m_o_.GetLength(I0); const index_t c_m = kernel_arg.y_grid_desc_m_o_.GetLength(I0);
const index_t c_gemm1n = kernel_arg.y_grid_desc_m_o_.GetLength(I1); const index_t c_gemm1n = kernel_arg.y_grid_desc_m_o_.GetLength(I1);
const index_t a_m = kernel_arg.a_grid_desc_ak0_m_ak1_.GetLength(I1); const index_t a_m = kernel_arg.a_grid_desc_ak0_m_ak1_.GetLength(I1);
const index_t b1_gemm1n = kernel_arg.b1_grid_desc_bk0_n_bk1_.GetLength(I0) * const index_t b1_gemm1n = kernel_arg.b1_grid_desc_bk0_n_bk1_.GetLength(I0) *
kernel_arg.b1_grid_desc_bk0_n_bk1_.GetLength(I2); kernel_arg.b1_grid_desc_bk0_n_bk1_.GetLength(I2);
if(!(c_g == device_arg.batch_count_ && c_m == a_m && c_gemm1n == b1_gemm1n)) if(!(c_g == device_arg.batch_count_ && c_m == a_m && c_gemm1n == b1_gemm1n &&
c_g % b_g == 0 && c_g / b_g == arg.h_ratio_))
{ {
return false; return false;
} }
...@@ -1395,6 +1453,7 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1 ...@@ -1395,6 +1453,7 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1
B1ElementwiseOperation b1_element_op, B1ElementwiseOperation b1_element_op,
CElementwiseOperation c_element_op, CElementwiseOperation c_element_op,
float p_drop, float p_drop,
index_t h_ratio,
std::tuple<unsigned long long, unsigned long long> seeds) std::tuple<unsigned long long, unsigned long long> seeds)
{ {
return Argument{p_As, return Argument{p_As,
...@@ -1419,6 +1478,7 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1 ...@@ -1419,6 +1478,7 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1
b1_element_op, b1_element_op,
c_element_op, c_element_op,
p_drop, p_drop,
h_ratio,
seeds}; seeds};
} }
...@@ -1449,6 +1509,7 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1 ...@@ -1449,6 +1509,7 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1
B1ElementwiseOperation b1_element_op, B1ElementwiseOperation b1_element_op,
CElementwiseOperation c_element_op, CElementwiseOperation c_element_op,
float p_drop, float p_drop,
index_t h_ratio,
std::tuple<unsigned long long, unsigned long long> seeds) // override std::tuple<unsigned long long, unsigned long long> seeds) // override
{ {
return std::make_unique<Argument>(p_As, return std::make_unique<Argument>(p_As,
...@@ -1473,6 +1534,7 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1 ...@@ -1473,6 +1534,7 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1
b1_element_op, b1_element_op,
c_element_op, c_element_op,
p_drop, p_drop,
h_ratio,
seeds); seeds);
} }
......
include/ck/tensor_operation/gpu/device/impl/device_grouped_mha_bwd_xdl_cshuffle_qloop_light_v2.hpp
View file @ 104aeabc
...@@ -102,6 +102,7 @@ __global__ void ...@@ -102,6 +102,7 @@ __global__ void
kernel_grouped_multihead_attention_backward_qloop_xdl_cshuffle_light_v2( kernel_grouped_multihead_attention_backward_qloop_xdl_cshuffle_light_v2(
const void CK_CONSTANT_ADDRESS_SPACE* group_kernel_args, const void CK_CONSTANT_ADDRESS_SPACE* group_kernel_args,
const index_t group_count, const index_t group_count,
const index_t h_ratio,
const AElementwiseOperation a_element_op, const AElementwiseOperation a_element_op,
const BElementwiseOperation b_element_op, const BElementwiseOperation b_element_op,
const AccElementwiseOperation acc_element_op, const AccElementwiseOperation acc_element_op,
...@@ -140,19 +141,26 @@ __global__ void ...@@ -140,19 +141,26 @@ __global__ void
const index_t num_blocks_per_batch = arg_ptr[group_id].num_blocks_per_batch_; const index_t num_blocks_per_batch = arg_ptr[group_id].num_blocks_per_batch_;
const index_t g_idx = __builtin_amdgcn_readfirstlane( const index_t g_idx = __builtin_amdgcn_readfirstlane(
(block_id - arg_ptr[group_id].block_start_) / (Deterministic ? 1 : num_blocks_per_batch)); (block_id - arg_ptr[group_id].block_start_) / (Deterministic ? 1 : num_blocks_per_batch));
const index_t gkv_idx = __builtin_amdgcn_readfirstlane(g_idx / h_ratio);
const long_index_t a_batch_offset = __builtin_amdgcn_readfirstlane( const long_index_t a_batch_offset = __builtin_amdgcn_readfirstlane(
static_cast<long_index_t>(arg_ptr[group_id].compute_base_ptr_of_batch_.GetABasePtr(g_idx))); static_cast<long_index_t>(arg_ptr[group_id].compute_base_ptr_of_batch_.GetABasePtr(g_idx)));
const long_index_t b_batch_offset = __builtin_amdgcn_readfirstlane( const long_index_t b_batch_offset = __builtin_amdgcn_readfirstlane(static_cast<long_index_t>(
static_cast<long_index_t>(arg_ptr[group_id].compute_base_ptr_of_batch_.GetBBasePtr(g_idx))); arg_ptr[group_id].compute_base_ptr_of_batch_.GetBBasePtr(gkv_idx)));
const long_index_t z_batch_offset = __builtin_amdgcn_readfirstlane( const long_index_t z_batch_offset = __builtin_amdgcn_readfirstlane(
static_cast<long_index_t>(arg_ptr[group_id].compute_base_ptr_of_batch_.GetZBasePtr(g_idx))); static_cast<long_index_t>(arg_ptr[group_id].compute_base_ptr_of_batch_.GetZBasePtr(g_idx)));
const long_index_t b1_batch_offset = __builtin_amdgcn_readfirstlane(static_cast<long_index_t>( const long_index_t b1_batch_offset = __builtin_amdgcn_readfirstlane(static_cast<long_index_t>(
arg_ptr[group_id].compute_base_ptr_of_batch_.GetB1BasePtr(g_idx))); arg_ptr[group_id].compute_base_ptr_of_batch_.GetB1BasePtr(gkv_idx)));
const long_index_t c_batch_offset = __builtin_amdgcn_readfirstlane( const long_index_t c_batch_offset = __builtin_amdgcn_readfirstlane(
static_cast<long_index_t>(arg_ptr[group_id].compute_base_ptr_of_batch_.GetCBasePtr(g_idx))); static_cast<long_index_t>(arg_ptr[group_id].compute_base_ptr_of_batch_.GetCBasePtr(g_idx)));
const long_index_t lse_batch_offset = __builtin_amdgcn_readfirstlane(static_cast<long_index_t>( const long_index_t lse_batch_offset = __builtin_amdgcn_readfirstlane(static_cast<long_index_t>(
arg_ptr[group_id].compute_base_ptr_of_batch_.GetLSEBasePtr(g_idx))); arg_ptr[group_id].compute_base_ptr_of_batch_.GetLSEBasePtr(g_idx)));
const long_index_t bgrad_batch_offset =
__builtin_amdgcn_readfirstlane(static_cast<long_index_t>(
arg_ptr[group_id].compute_base_ptr_of_batch_.GetBGradBasePtr(g_idx)));
const long_index_t b1grad_batch_offset =
__builtin_amdgcn_readfirstlane(static_cast<long_index_t>(
arg_ptr[group_id].compute_base_ptr_of_batch_.GetB1GradBasePtr(g_idx)));
const index_t global_thread_id = get_thread_global_1d_id(); const index_t global_thread_id = get_thread_global_1d_id();
ck::philox ph(seed, global_thread_id, offset); ck::philox ph(seed, global_thread_id, offset);
...@@ -166,7 +174,6 @@ __global__ void ...@@ -166,7 +174,6 @@ __global__ void
const long_index_t d0_batch_offset = const long_index_t d0_batch_offset =
__builtin_amdgcn_readfirstlane(static_cast<long_index_t>( __builtin_amdgcn_readfirstlane(static_cast<long_index_t>(
arg_ptr[group_id].compute_base_ptr_of_batch_.GetD0BasePtr(g_idx))); arg_ptr[group_id].compute_base_ptr_of_batch_.GetD0BasePtr(g_idx)));
if(arg_ptr[group_id].p_d0_grid_ != nullptr) if(arg_ptr[group_id].p_d0_grid_ != nullptr)
tmp_p_d0_grid = arg_ptr[group_id].p_d0_grid_ + d0_batch_offset; tmp_p_d0_grid = arg_ptr[group_id].p_d0_grid_ + d0_batch_offset;
if(arg_ptr[group_id].p_d0grad_grid_) if(arg_ptr[group_id].p_d0grad_grid_)
...@@ -187,9 +194,9 @@ __global__ void ...@@ -187,9 +194,9 @@ __global__ void
arg_ptr[group_id].p_d_grid_ + lse_batch_offset, arg_ptr[group_id].p_d_grid_ + lse_batch_offset,
arg_ptr[group_id].p_ygrad_grid_ + c_batch_offset, arg_ptr[group_id].p_ygrad_grid_ + c_batch_offset,
arg_ptr[group_id].p_qgrad_grid_ + a_batch_offset, arg_ptr[group_id].p_qgrad_grid_ + a_batch_offset,
arg_ptr[group_id].p_kgrad_grid_ + b_batch_offset, arg_ptr[group_id].p_kgrad_grid_ + bgrad_batch_offset,
tmp_p_d0grad_grid, tmp_p_d0grad_grid,
arg_ptr[group_id].p_vgrad_grid_ + b1_batch_offset, arg_ptr[group_id].p_vgrad_grid_ + b1grad_batch_offset,
p_shared, p_shared,
a_element_op, a_element_op,
b_element_op, b_element_op,
...@@ -198,9 +205,11 @@ __global__ void ...@@ -198,9 +205,11 @@ __global__ void
c_element_op, c_element_op,
arg_ptr[group_id].a_grid_desc_ak0_m_ak1_, arg_ptr[group_id].a_grid_desc_ak0_m_ak1_,
arg_ptr[group_id].b_grid_desc_bk0_n_bk1_, arg_ptr[group_id].b_grid_desc_bk0_n_bk1_,
arg_ptr[group_id].bgrad_grid_desc_bk0_n_bk1_,
arg_ptr[group_id].d0_grid_desc_m0_n0_m1_m2_n1_m3_, arg_ptr[group_id].d0_grid_desc_m0_n0_m1_m2_n1_m3_,
arg_ptr[group_id].c_grid_desc_m0_n0_m1_n1_m2_n2_m3_m4_m5_n3_, arg_ptr[group_id].c_grid_desc_m0_n0_m1_n1_m2_n2_m3_m4_m5_n3_,
arg_ptr[group_id].b1_grid_desc_bk0_n_bk1_, arg_ptr[group_id].b1_grid_desc_bk0_n_bk1_,
arg_ptr[group_id].b1grad_grid_desc_bk0_n_bk1_,
arg_ptr[group_id].lse_grid_desc_m_, arg_ptr[group_id].lse_grid_desc_m_,
arg_ptr[group_id].ygrad_grid_desc_m0_o_m1_, arg_ptr[group_id].ygrad_grid_desc_m0_o_m1_,
arg_ptr[group_id].block_2_ctile_map_, arg_ptr[group_id].block_2_ctile_map_,
...@@ -225,9 +234,9 @@ __global__ void ...@@ -225,9 +234,9 @@ __global__ void
arg_ptr[group_id].p_d_grid_ + lse_batch_offset, arg_ptr[group_id].p_d_grid_ + lse_batch_offset,
arg_ptr[group_id].p_ygrad_grid_ + c_batch_offset, arg_ptr[group_id].p_ygrad_grid_ + c_batch_offset,
arg_ptr[group_id].p_qgrad_grid_ + a_batch_offset, arg_ptr[group_id].p_qgrad_grid_ + a_batch_offset,
arg_ptr[group_id].p_kgrad_grid_ + b_batch_offset, arg_ptr[group_id].p_kgrad_grid_ + bgrad_batch_offset,
tmp_p_d0grad_grid, tmp_p_d0grad_grid,
arg_ptr[group_id].p_vgrad_grid_ + b1_batch_offset, arg_ptr[group_id].p_vgrad_grid_ + b1grad_batch_offset,
p_shared, p_shared,
a_element_op, a_element_op,
b_element_op, b_element_op,
...@@ -236,9 +245,11 @@ __global__ void ...@@ -236,9 +245,11 @@ __global__ void
c_element_op, c_element_op,
arg_ptr[group_id].a_grid_desc_ak0_m_ak1_, arg_ptr[group_id].a_grid_desc_ak0_m_ak1_,
arg_ptr[group_id].b_grid_desc_bk0_n_bk1_, arg_ptr[group_id].b_grid_desc_bk0_n_bk1_,
arg_ptr[group_id].bgrad_grid_desc_bk0_n_bk1_,
arg_ptr[group_id].d0_grid_desc_m0_n0_m1_m2_n1_m3_, arg_ptr[group_id].d0_grid_desc_m0_n0_m1_m2_n1_m3_,
arg_ptr[group_id].c_grid_desc_m0_n0_m1_n1_m2_n2_m3_m4_m5_n3_, arg_ptr[group_id].c_grid_desc_m0_n0_m1_n1_m2_n2_m3_m4_m5_n3_,
arg_ptr[group_id].b1_grid_desc_bk0_n_bk1_, arg_ptr[group_id].b1_grid_desc_bk0_n_bk1_,
arg_ptr[group_id].b1grad_grid_desc_bk0_n_bk1_,
arg_ptr[group_id].lse_grid_desc_m_, arg_ptr[group_id].lse_grid_desc_m_,
arg_ptr[group_id].ygrad_grid_desc_m0_o_m1_, arg_ptr[group_id].ygrad_grid_desc_m0_o_m1_,
arg_ptr[group_id].block_2_ctile_map_, arg_ptr[group_id].block_2_ctile_map_,
...@@ -253,6 +264,7 @@ __global__ void ...@@ -253,6 +264,7 @@ __global__ void
#else #else
ignore = group_kernel_args; ignore = group_kernel_args;
ignore = group_count; ignore = group_count;
ignore = h_ratio;
ignore = a_element_op; ignore = a_element_op;
ignore = b_element_op; ignore = b_element_op;
ignore = acc_element_op; ignore = acc_element_op;
...@@ -373,6 +385,12 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2 ...@@ -373,6 +385,12 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2
std::vector<index_t> lse_gs_ms_lengths; std::vector<index_t> lse_gs_ms_lengths;
std::vector<index_t> lse_gs_ms_strides; std::vector<index_t> lse_gs_ms_strides;
std::vector<index_t> bgrad_gs_ns_ks_lengths;
std::vector<index_t> bgrad_gs_ns_ks_strides;
std::vector<index_t> b1grad_gs_gemm1ns_gemm1ks_lengths;
std::vector<index_t> b1grad_gs_gemm1ns_gemm1ks_strides;
std::vector<index_t> acc0_bias_gs_ms_ns_lengths; std::vector<index_t> acc0_bias_gs_ms_ns_lengths;
std::vector<index_t> acc0_bias_gs_ms_ns_strides; std::vector<index_t> acc0_bias_gs_ms_ns_strides;
...@@ -639,7 +657,6 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2 ...@@ -639,7 +657,6 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2
static auto MakeD0GridDescriptor_M_N(const std::vector<ck::index_t>& acc0_bias_gs_ms_ns_lengths, static auto MakeD0GridDescriptor_M_N(const std::vector<ck::index_t>& acc0_bias_gs_ms_ns_lengths,
const std::vector<ck::index_t>& acc0_bias_gs_ms_ns_strides) const std::vector<ck::index_t>& acc0_bias_gs_ms_ns_strides)
{ {
return Transform::MakeC0GridDescriptor_M_N(acc0_bias_gs_ms_ns_lengths, return Transform::MakeC0GridDescriptor_M_N(acc0_bias_gs_ms_ns_lengths,
acc0_bias_gs_ms_ns_strides); acc0_bias_gs_ms_ns_strides);
} }
...@@ -648,7 +665,6 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2 ...@@ -648,7 +665,6 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2
MakeD0GridDescriptor_G_M_N(const std::vector<ck::index_t>& acc0_bias_gs_ms_ns_lengths, MakeD0GridDescriptor_G_M_N(const std::vector<ck::index_t>& acc0_bias_gs_ms_ns_lengths,
const std::vector<ck::index_t>& acc0_bias_gs_ms_ns_strides) const std::vector<ck::index_t>& acc0_bias_gs_ms_ns_strides)
{ {
return Transform::MakeC0GridDescriptor_G_M_N(acc0_bias_gs_ms_ns_lengths, return Transform::MakeC0GridDescriptor_G_M_N(acc0_bias_gs_ms_ns_lengths,
acc0_bias_gs_ms_ns_strides); acc0_bias_gs_ms_ns_strides);
} }
...@@ -723,6 +739,8 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2 ...@@ -723,6 +739,8 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2
const ZGridDesc_G_M_N& z_grid_desc_g_m_n, const ZGridDesc_G_M_N& z_grid_desc_g_m_n,
const B1GridDesc_G_N_K& b1_grid_desc_g_n_k, const B1GridDesc_G_N_K& b1_grid_desc_g_n_k,
const CGridDesc_G_M_N& c_grid_desc_g_m_n, const CGridDesc_G_M_N& c_grid_desc_g_m_n,
const BGridDesc_G_N_K& bgrad_grid_desc_g_n_k,
const B1GridDesc_G_N_K& b1grad_grid_desc_g_n_k,
index_t BatchStrideLSE) index_t BatchStrideLSE)
: a_grid_desc_g_m_k_(a_grid_desc_g_m_k), : a_grid_desc_g_m_k_(a_grid_desc_g_m_k),
b_grid_desc_g_n_k_(b_grid_desc_g_n_k), b_grid_desc_g_n_k_(b_grid_desc_g_n_k),
...@@ -730,6 +748,8 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2 ...@@ -730,6 +748,8 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2
z_grid_desc_g_m_n_(z_grid_desc_g_m_n), z_grid_desc_g_m_n_(z_grid_desc_g_m_n),
b1_grid_desc_g_n_k_(b1_grid_desc_g_n_k), b1_grid_desc_g_n_k_(b1_grid_desc_g_n_k),
c_grid_desc_g_m_n_(c_grid_desc_g_m_n), c_grid_desc_g_m_n_(c_grid_desc_g_m_n),
bgrad_grid_desc_g_n_k_(bgrad_grid_desc_g_n_k),
b1grad_grid_desc_g_n_k_(b1grad_grid_desc_g_n_k),
BatchStrideLSE_(BatchStrideLSE) BatchStrideLSE_(BatchStrideLSE)
{ {
} }
...@@ -769,6 +789,16 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2 ...@@ -769,6 +789,16 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2
return g_idx * static_cast<long_index_t>(BatchStrideLSE_); return g_idx * static_cast<long_index_t>(BatchStrideLSE_);
} }
__host__ __device__ constexpr long_index_t GetBGradBasePtr(index_t g_idx) const
{
return bgrad_grid_desc_g_n_k_.CalculateOffset(make_multi_index(g_idx, 0, 0));
}
__host__ __device__ constexpr long_index_t GetB1GradBasePtr(index_t g_idx) const
{
return b1grad_grid_desc_g_n_k_.CalculateOffset(make_multi_index(g_idx, 0, 0));
}
private: private:
AGridDesc_G_M_K a_grid_desc_g_m_k_; AGridDesc_G_M_K a_grid_desc_g_m_k_;
BGridDesc_G_N_K b_grid_desc_g_n_k_; BGridDesc_G_N_K b_grid_desc_g_n_k_;
...@@ -776,6 +806,8 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2 ...@@ -776,6 +806,8 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2
ZGridDesc_G_M_N z_grid_desc_g_m_n_; ZGridDesc_G_M_N z_grid_desc_g_m_n_;
B1GridDesc_G_N_K b1_grid_desc_g_n_k_; B1GridDesc_G_N_K b1_grid_desc_g_n_k_;
CGridDesc_G_M_N c_grid_desc_g_m_n_; CGridDesc_G_M_N c_grid_desc_g_m_n_;
BGridDesc_G_N_K bgrad_grid_desc_g_n_k_;
B1GridDesc_G_N_K b1grad_grid_desc_g_n_k_;
index_t BatchStrideLSE_; index_t BatchStrideLSE_;
}; };
...@@ -888,9 +920,11 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2 ...@@ -888,9 +920,11 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2
// tensor descriptors for block/thread-wise copy // tensor descriptors for block/thread-wise copy
AGridDesc_AK0_M_AK1 a_grid_desc_ak0_m_ak1_; AGridDesc_AK0_M_AK1 a_grid_desc_ak0_m_ak1_;
BGridDesc_BK0_N_BK1 b_grid_desc_bk0_n_bk1_; BGridDesc_BK0_N_BK1 b_grid_desc_bk0_n_bk1_;
BGridDesc_BK0_N_BK1 bgrad_grid_desc_bk0_n_bk1_;
typename GridwiseGemm::D0GridDescriptor_M0_N0_M1_M2_N1_M3 d0_grid_desc_m0_n0_m1_m2_n1_m3_; typename GridwiseGemm::D0GridDescriptor_M0_N0_M1_M2_N1_M3 d0_grid_desc_m0_n0_m1_m2_n1_m3_;
ZGridDesc_M_N z_grid_desc_m_n_; ZGridDesc_M_N z_grid_desc_m_n_;
B1GridDesc_BK0_N_BK1 b1_grid_desc_bk0_n_bk1_; B1GridDesc_BK0_N_BK1 b1_grid_desc_bk0_n_bk1_;
B1GridDesc_BK0_N_BK1 b1grad_grid_desc_bk0_n_bk1_;
YGridDesc_M_O y_grid_desc_m_o_; YGridDesc_M_O y_grid_desc_m_o_;
typename GridwiseGemm::ZGridDescriptor_M0_N0_M1_N1_M2_N2_M3_M4_M5_N3 typename GridwiseGemm::ZGridDescriptor_M0_N0_M1_N1_M2_N2_M3_M4_M5_N3
...@@ -932,6 +966,7 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2 ...@@ -932,6 +966,7 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2
std::vector<index_t> c_mz_gemm1nz_strides_; std::vector<index_t> c_mz_gemm1nz_strides_;
// for gridwise gemm check // for gridwise gemm check
BGridDesc_G_N_K b_grid_desc_g_n_k_;
CGridDesc_G_M_N c_grid_desc_g_m_n_; CGridDesc_G_M_N c_grid_desc_g_m_n_;
index_t batch_count_; index_t batch_count_;
...@@ -964,13 +999,15 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2 ...@@ -964,13 +999,15 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2
B1ElementwiseOperation b1_element_op, B1ElementwiseOperation b1_element_op,
CElementwiseOperation c_element_op, CElementwiseOperation c_element_op,
float p_drop, float p_drop,
index_t h_ratio,
std::tuple<unsigned long long, unsigned long long> seeds) std::tuple<unsigned long long, unsigned long long> seeds)
: a_element_op_{a_element_op}, : a_element_op_{a_element_op},
b_element_op_{b_element_op}, b_element_op_{b_element_op},
acc_element_op_{acc_element_op}, acc_element_op_{acc_element_op},
b1_element_op_{b1_element_op}, b1_element_op_{b1_element_op},
c_element_op_{c_element_op}, c_element_op_{c_element_op},
p_dropout_{p_drop} p_dropout_{p_drop},
h_ratio_{h_ratio}
{ {
seed_ = std::get<0>(seeds); seed_ = std::get<0>(seeds);
offset_ = std::get<1>(seeds); offset_ = std::get<1>(seeds);
...@@ -1031,6 +1068,8 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2 ...@@ -1031,6 +1068,8 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2
problem_desc.a_gs_ms_ks_lengths, problem_desc.a_gs_ms_ks_strides); problem_desc.a_gs_ms_ks_lengths, problem_desc.a_gs_ms_ks_strides);
const auto b_grid_desc_bk0_n_bk1 = DeviceOp::MakeBGridDescriptor_BK0_N_BK1( const auto b_grid_desc_bk0_n_bk1 = DeviceOp::MakeBGridDescriptor_BK0_N_BK1(
problem_desc.b_gs_ns_ks_lengths, problem_desc.b_gs_ns_ks_strides); problem_desc.b_gs_ns_ks_lengths, problem_desc.b_gs_ns_ks_strides);
const auto bgrad_grid_desc_bk0_n_bk1 = DeviceOp::MakeBGridDescriptor_BK0_N_BK1(
problem_desc.bgrad_gs_ns_ks_lengths, problem_desc.bgrad_gs_ns_ks_strides);
std::vector<index_t> tmp_d0_gs_ms_ns_lengths; std::vector<index_t> tmp_d0_gs_ms_ns_lengths;
std::vector<index_t> tmp_d0_gs_ms_ns_strides; std::vector<index_t> tmp_d0_gs_ms_ns_strides;
...@@ -1053,6 +1092,9 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2 ...@@ -1053,6 +1092,9 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2
const auto b1_grid_desc_bk0_n_bk1 = DeviceOp::MakeVGridDescriptor_O0_N_O1( const auto b1_grid_desc_bk0_n_bk1 = DeviceOp::MakeVGridDescriptor_O0_N_O1(
problem_desc.b1_gs_gemm1ns_gemm1ks_lengths, problem_desc.b1_gs_gemm1ns_gemm1ks_lengths,
problem_desc.b1_gs_gemm1ns_gemm1ks_strides); problem_desc.b1_gs_gemm1ns_gemm1ks_strides);
const auto b1grad_grid_desc_bk0_n_bk1 = DeviceOp::MakeVGridDescriptor_O0_N_O1(
problem_desc.b1grad_gs_gemm1ns_gemm1ks_lengths,
problem_desc.b1grad_gs_gemm1ns_gemm1ks_strides);
const auto y_grid_desc_m_o = Transform::MakeCGridDescriptor_M_N( const auto y_grid_desc_m_o = Transform::MakeCGridDescriptor_M_N(
problem_desc.c_gs_ms_gemm1ns_lengths, problem_desc.c_gs_ms_gemm1ns_strides); problem_desc.c_gs_ms_gemm1ns_lengths, problem_desc.c_gs_ms_gemm1ns_strides);
...@@ -1076,6 +1118,11 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2 ...@@ -1076,6 +1118,11 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2
problem_desc.b1_gs_gemm1ns_gemm1ks_strides); problem_desc.b1_gs_gemm1ns_gemm1ks_strides);
const auto c_grid_desc_g_m_n = Transform::MakeCGridDescriptor_G_M_N( const auto c_grid_desc_g_m_n = Transform::MakeCGridDescriptor_G_M_N(
problem_desc.c_gs_ms_gemm1ns_lengths, problem_desc.c_gs_ms_gemm1ns_strides); problem_desc.c_gs_ms_gemm1ns_lengths, problem_desc.c_gs_ms_gemm1ns_strides);
const auto bgrad_grid_desc_g_n_k = Transform::MakeB0GridDescriptor_G_N_K(
problem_desc.bgrad_gs_ns_ks_lengths, problem_desc.bgrad_gs_ns_ks_strides);
const auto b1grad_grid_desc_g_n_k = Transform::MakeB1GridDescriptor_G_N_K(
problem_desc.b1grad_gs_gemm1ns_gemm1ks_lengths,
problem_desc.b1grad_gs_gemm1ns_gemm1ks_strides);
typename GridwiseGemm::ZGridDescriptor_M0_N0_M1_N1_M2_N2_M3_M4_M5_N3 typename GridwiseGemm::ZGridDescriptor_M0_N0_M1_N1_M2_N2_M3_M4_M5_N3
c_grid_desc_m0_n0_m1_n1_m2_n2_m3_m4_m5_n3; c_grid_desc_m0_n0_m1_n1_m2_n2_m3_m4_m5_n3;
const index_t BlockStart = grid_size_; const index_t BlockStart = grid_size_;
...@@ -1098,6 +1145,8 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2 ...@@ -1098,6 +1145,8 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2
z_grid_desc_g_m_n, z_grid_desc_g_m_n,
b1_grid_desc_g_n_k, b1_grid_desc_g_n_k,
c_grid_desc_g_m_n, c_grid_desc_g_m_n,
bgrad_grid_desc_g_n_k,
b1grad_grid_desc_g_n_k,
type_convert<index_t>(lse_grid_desc_m.GetElementSpaceSize())); type_convert<index_t>(lse_grid_desc_m.GetElementSpaceSize()));
// C0 mask // C0 mask
...@@ -1144,9 +1193,11 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2 ...@@ -1144,9 +1193,11 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2
p_vgrad_grid, p_vgrad_grid,
a_grid_desc_ak0_m_ak1, a_grid_desc_ak0_m_ak1,
b_grid_desc_bk0_n_bk1, b_grid_desc_bk0_n_bk1,
bgrad_grid_desc_bk0_n_bk1,
d0_grid_desc_m0_n0_m1_m2_n1_m3, d0_grid_desc_m0_n0_m1_m2_n1_m3,
z_grid_desc_m_n, z_grid_desc_m_n,
b1_grid_desc_bk0_n_bk1, b1_grid_desc_bk0_n_bk1,
b1grad_grid_desc_bk0_n_bk1,
y_grid_desc_m_o, y_grid_desc_m_o,
c_grid_desc_m0_n0_m1_n1_m2_n2_m3_m4_m5_n3, c_grid_desc_m0_n0_m1_n1_m2_n2_m3_m4_m5_n3,
lse_grid_desc_m, lse_grid_desc_m,
...@@ -1190,6 +1241,7 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2 ...@@ -1190,6 +1241,7 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2
problem_desc.b1_gs_gemm1ns_gemm1ks_strides[NumDimG + NumDimO + NumDimN - 1]}, problem_desc.b1_gs_gemm1ns_gemm1ks_strides[NumDimG + NumDimO + NumDimN - 1]},
{problem_desc.c_gs_ms_gemm1ns_strides[NumDimG + NumDimM - 1], {problem_desc.c_gs_ms_gemm1ns_strides[NumDimG + NumDimM - 1],
problem_desc.c_gs_ms_gemm1ns_strides[NumDimG + NumDimM + NumDimO - 1]}, problem_desc.c_gs_ms_gemm1ns_strides[NumDimG + NumDimM + NumDimO - 1]},
b_grid_desc_g_n_k,
c_grid_desc_g_m_n, c_grid_desc_g_m_n,
batch_count, batch_count,
d0_n_length_stride}); d0_n_length_stride});
...@@ -1216,6 +1268,7 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2 ...@@ -1216,6 +1268,7 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2
index_t grid_size_; index_t grid_size_;
index_t group_count_; index_t group_count_;
index_t h_ratio_;
std::vector<GroupKernelArg> group_kernel_args_; std::vector<GroupKernelArg> group_kernel_args_;
std::vector<GroupDeviceArg> group_device_args_; std::vector<GroupDeviceArg> group_device_args_;
...@@ -1294,6 +1347,7 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2 ...@@ -1294,6 +1347,7 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2
0, 0,
cast_pointer_to_constant_address_space(arg.p_workspace_), cast_pointer_to_constant_address_space(arg.p_workspace_),
arg.group_count_, arg.group_count_,
arg.h_ratio_,
arg.a_element_op_, arg.a_element_op_,
arg.b_element_op_, arg.b_element_op_,
arg.acc_element_op_, arg.acc_element_op_,
...@@ -1362,13 +1416,15 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2 ...@@ -1362,13 +1416,15 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2
const auto& device_arg = arg.group_device_args_[i]; const auto& device_arg = arg.group_device_args_[i];
// Check if C permute dimension matches GEMM + GEMM shape // Check if C permute dimension matches GEMM + GEMM shape
const index_t c_g = device_arg.c_grid_desc_g_m_n_.GetLength(I0); // unpadded const index_t c_g = device_arg.c_grid_desc_g_m_n_.GetLength(I0); // unpadded
const index_t b_g = device_arg.b_grid_desc_g_n_k_.GetLength(I0);
const index_t c_m = kernel_arg.y_grid_desc_m_o_.GetLength(I0); const index_t c_m = kernel_arg.y_grid_desc_m_o_.GetLength(I0);
const index_t c_gemm1n = kernel_arg.y_grid_desc_m_o_.GetLength(I1); const index_t c_gemm1n = kernel_arg.y_grid_desc_m_o_.GetLength(I1);
const index_t a_m = kernel_arg.a_grid_desc_ak0_m_ak1_.GetLength(I1); const index_t a_m = kernel_arg.a_grid_desc_ak0_m_ak1_.GetLength(I1);
const index_t b1_gemm1n = kernel_arg.b1_grid_desc_bk0_n_bk1_.GetLength(I0) * const index_t b1_gemm1n = kernel_arg.b1_grid_desc_bk0_n_bk1_.GetLength(I0) *
kernel_arg.b1_grid_desc_bk0_n_bk1_.GetLength(I2); kernel_arg.b1_grid_desc_bk0_n_bk1_.GetLength(I2);
if(!(c_g == device_arg.batch_count_ && c_m == a_m && c_gemm1n == b1_gemm1n)) if(!(c_g == device_arg.batch_count_ && c_m == a_m && c_gemm1n == b1_gemm1n &&
c_g % b_g == 0 && c_g / b_g == arg.h_ratio_))
{ {
return false; return false;
} }
...@@ -1471,6 +1527,7 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2 ...@@ -1471,6 +1527,7 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2
B1ElementwiseOperation b1_element_op, B1ElementwiseOperation b1_element_op,
CElementwiseOperation c_element_op, CElementwiseOperation c_element_op,
float p_drop, float p_drop,
index_t h_ratio,
std::tuple<unsigned long long, unsigned long long> seeds) std::tuple<unsigned long long, unsigned long long> seeds)
{ {
return Argument{p_As, return Argument{p_As,
...@@ -1495,6 +1552,7 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2 ...@@ -1495,6 +1552,7 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2
b1_element_op, b1_element_op,
c_element_op, c_element_op,
p_drop, p_drop,
h_ratio,
seeds}; seeds};
} }
...@@ -1525,6 +1583,7 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2 ...@@ -1525,6 +1583,7 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2
B1ElementwiseOperation b1_element_op, B1ElementwiseOperation b1_element_op,
CElementwiseOperation c_element_op, CElementwiseOperation c_element_op,
float p_drop, float p_drop,
index_t h_ratio,
std::tuple<unsigned long long, unsigned long long> seeds) // override std::tuple<unsigned long long, unsigned long long> seeds) // override
{ {
return std::make_unique<Argument>(p_As, return std::make_unique<Argument>(p_As,
...@@ -1549,6 +1608,7 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2 ...@@ -1549,6 +1608,7 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2
b1_element_op, b1_element_op,
c_element_op, c_element_op,
p_drop, p_drop,
h_ratio,
seeds); seeds);
} }
......
include/ck/tensor_operation/gpu/grid/gridwise_batched_mha_bwd_xdl_cshuffle_qloop_b2t_light_v1.hpp
View file @ 104aeabc
...@@ -1440,10 +1440,12 @@ struct GridwiseBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1 ...@@ -1440,10 +1440,12 @@ struct GridwiseBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1
const CElementwiseOperation& c_element_op, const CElementwiseOperation& c_element_op,
const QGridDesc_K0_M_K1& q_grid_desc_k0_m_k1, const QGridDesc_K0_M_K1& q_grid_desc_k0_m_k1,
const KGridDesc_K0_N_K1& k_grid_desc_k0_n_k1, const KGridDesc_K0_N_K1& k_grid_desc_k0_n_k1,
const KGridDesc_K0_N_K1& kgrad_grid_desc_k0_n_k1,
const D0GridDescriptor_M0_N0_M1_M2_N1_M3& d0_grid_desc_m0_n0_m1_m2_n1_m3, const D0GridDescriptor_M0_N0_M1_M2_N1_M3& d0_grid_desc_m0_n0_m1_m2_n1_m3,
const ZGridDescriptor_M0_N0_M1_N1_M2_N2_M3_M4_M5_N3& const ZGridDescriptor_M0_N0_M1_N1_M2_N2_M3_M4_M5_N3&
z_grid_desc_m0_n0_m1_n1_m2_n2_m3_m4_m5_n3, z_grid_desc_m0_n0_m1_n1_m2_n2_m3_m4_m5_n3,
const VGridDesc_O0_N_O1& v_grid_desc_o0_n_o1, const VGridDesc_O0_N_O1& v_grid_desc_o0_n_o1,
const VGridDesc_O0_N_O1& vgrad_grid_desc_o0_n_o1,
const LSEGridDesc_M& lse_grid_desc_m, const LSEGridDesc_M& lse_grid_desc_m,
const YGradGridDesc_O0_M_O1& ygrad_grid_desc_o0_m_o1, const YGradGridDesc_O0_M_O1& ygrad_grid_desc_o0_m_o1,
const Block2CTileMap& block_2_ctile_map, const Block2CTileMap& block_2_ctile_map,
...@@ -1474,11 +1476,11 @@ struct GridwiseBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1 ...@@ -1474,11 +1476,11 @@ struct GridwiseBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1
const auto ygrad_grid_buf = make_dynamic_buffer<AddressSpaceEnum::Global>( const auto ygrad_grid_buf = make_dynamic_buffer<AddressSpaceEnum::Global>(
p_ygrad_grid, ygrad_grid_desc_o0_m_o1.GetElementSpaceSize()); p_ygrad_grid, ygrad_grid_desc_o0_m_o1.GetElementSpaceSize());
auto vgrad_grid_buf = make_dynamic_buffer<AddressSpaceEnum::Global>( auto vgrad_grid_buf = make_dynamic_buffer<AddressSpaceEnum::Global>(
p_vgrad_grid, v_grid_desc_o0_n_o1.GetElementSpaceSize()); p_vgrad_grid, vgrad_grid_desc_o0_n_o1.GetElementSpaceSize());
auto qgrad_grid_buf = make_dynamic_buffer<AddressSpaceEnum::Global>( auto qgrad_grid_buf = make_dynamic_buffer<AddressSpaceEnum::Global>(
p_qgrad_grid, q_grid_desc_k0_m_k1.GetElementSpaceSize()); p_qgrad_grid, q_grid_desc_k0_m_k1.GetElementSpaceSize());
auto kgrad_grid_buf = make_dynamic_buffer<AddressSpaceEnum::Global>( auto kgrad_grid_buf = make_dynamic_buffer<AddressSpaceEnum::Global>(
p_kgrad_grid, k_grid_desc_k0_n_k1.GetElementSpaceSize()); p_kgrad_grid, kgrad_grid_desc_k0_n_k1.GetElementSpaceSize());
// divide block work by [N, K] // divide block work by [N, K]
const auto block_work_idx = const auto block_work_idx =
...@@ -1628,7 +1630,7 @@ struct GridwiseBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1 ...@@ -1628,7 +1630,7 @@ struct GridwiseBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1
// dV: transform input and output tensor descriptors // dV: transform input and output tensor descriptors
auto vgrad_grid_desc_nblock_nperblock_oblock_operblock = auto vgrad_grid_desc_nblock_nperblock_oblock_operblock =
MakeVGradGridDesc_NBlock_NPerBlock_OBlock_OPerBlock(v_grid_desc_o0_n_o1); MakeVGradGridDesc_NBlock_NPerBlock_OBlock_OPerBlock(vgrad_grid_desc_o0_n_o1);
// dK: A matrix blockwise copy // dK: A matrix blockwise copy
auto kgrad_gemm_tile_sgrad_blockwise_copy = auto kgrad_gemm_tile_sgrad_blockwise_copy =
...@@ -1657,7 +1659,7 @@ struct GridwiseBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1 ...@@ -1657,7 +1659,7 @@ struct GridwiseBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V1
// dK: transform input and output tensor descriptors // dK: transform input and output tensor descriptors
auto kgrad_grid_desc_nblock_nperblock_oblock_operblock = auto kgrad_grid_desc_nblock_nperblock_oblock_operblock =
MakeKGradGridDesc_NBlock_NPerBlock_OBlock_OPerBlock(k_grid_desc_k0_n_k1); MakeKGradGridDesc_NBlock_NPerBlock_OBlock_OPerBlock(kgrad_grid_desc_k0_n_k1);
// //
// set up dQ Gemm (type 3 crr) // set up dQ Gemm (type 3 crr)
......
include/ck/tensor_operation/gpu/grid/gridwise_batched_mha_bwd_xdl_cshuffle_qloop_b2t_light_v2.hpp
View file @ 104aeabc
...@@ -1531,10 +1531,12 @@ struct GridwiseBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2 ...@@ -1531,10 +1531,12 @@ struct GridwiseBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2
const CElementwiseOperation& c_element_op, const CElementwiseOperation& c_element_op,
const QGridDesc_K0_M_K1& q_grid_desc_k0_m_k1, const QGridDesc_K0_M_K1& q_grid_desc_k0_m_k1,
const KGridDesc_K0_N_K1& k_grid_desc_k0_n_k1, const KGridDesc_K0_N_K1& k_grid_desc_k0_n_k1,
const KGridDesc_K0_N_K1& kgrad_grid_desc_k0_n_k1,
const D0GridDescriptor_M0_N0_M1_M2_N1_M3& d0_grid_desc_m0_n0_m1_m2_n1_m3, const D0GridDescriptor_M0_N0_M1_M2_N1_M3& d0_grid_desc_m0_n0_m1_m2_n1_m3,
const ZGridDescriptor_M0_N0_M1_N1_M2_N2_M3_M4_M5_N3& const ZGridDescriptor_M0_N0_M1_N1_M2_N2_M3_M4_M5_N3&
z_grid_desc_m0_n0_m1_n1_m2_n2_m3_m4_m5_n3, z_grid_desc_m0_n0_m1_n1_m2_n2_m3_m4_m5_n3,
const VGridDesc_O0_N_O1& v_grid_desc_o0_n_o1, const VGridDesc_O0_N_O1& v_grid_desc_o0_n_o1,
const VGridDesc_O0_N_O1& vgrad_grid_desc_o0_n_o1,
const LSEGridDesc_M& lse_grid_desc_m, const LSEGridDesc_M& lse_grid_desc_m,
const YGradGridDesc_M0_O_M1& ygrad_grid_desc_m0_o_m1, const YGradGridDesc_M0_O_M1& ygrad_grid_desc_m0_o_m1,
const Block2CTileMap& block_2_ctile_map, const Block2CTileMap& block_2_ctile_map,
...@@ -1565,11 +1567,11 @@ struct GridwiseBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2 ...@@ -1565,11 +1567,11 @@ struct GridwiseBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2
const auto ygrad_grid_buf = make_dynamic_buffer<AddressSpaceEnum::Global>( const auto ygrad_grid_buf = make_dynamic_buffer<AddressSpaceEnum::Global>(
p_ygrad_grid, ygrad_grid_desc_m0_o_m1.GetElementSpaceSize()); p_ygrad_grid, ygrad_grid_desc_m0_o_m1.GetElementSpaceSize());
auto vgrad_grid_buf = make_dynamic_buffer<AddressSpaceEnum::Global>( auto vgrad_grid_buf = make_dynamic_buffer<AddressSpaceEnum::Global>(
p_vgrad_grid, v_grid_desc_o0_n_o1.GetElementSpaceSize()); p_vgrad_grid, vgrad_grid_desc_o0_n_o1.GetElementSpaceSize());
auto qgrad_grid_buf = make_dynamic_buffer<AddressSpaceEnum::Global>( auto qgrad_grid_buf = make_dynamic_buffer<AddressSpaceEnum::Global>(
p_qgrad_grid, q_grid_desc_k0_m_k1.GetElementSpaceSize()); p_qgrad_grid, q_grid_desc_k0_m_k1.GetElementSpaceSize());
auto kgrad_grid_buf = make_dynamic_buffer<AddressSpaceEnum::Global>( auto kgrad_grid_buf = make_dynamic_buffer<AddressSpaceEnum::Global>(
p_kgrad_grid, k_grid_desc_k0_n_k1.GetElementSpaceSize()); p_kgrad_grid, kgrad_grid_desc_k0_n_k1.GetElementSpaceSize());
// divide block work by [N, K] // divide block work by [N, K]
const auto block_work_idx = const auto block_work_idx =
...@@ -1742,7 +1744,7 @@ struct GridwiseBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2 ...@@ -1742,7 +1744,7 @@ struct GridwiseBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2
// dV: transform input and output tensor descriptors // dV: transform input and output tensor descriptors
auto vgrad_grid_desc_nblock_nperblock_oblock_operblock = auto vgrad_grid_desc_nblock_nperblock_oblock_operblock =
MakeVGradGridDesc_NBlock_NPerBlock_OBlock_OPerBlock(v_grid_desc_o0_n_o1); MakeVGradGridDesc_NBlock_NPerBlock_OBlock_OPerBlock(vgrad_grid_desc_o0_n_o1);
// dK: transform input and output tensor descriptors // dK: transform input and output tensor descriptors
const auto q_grid_desc_m0_k_m1 = const auto q_grid_desc_m0_k_m1 =
...@@ -1775,7 +1777,7 @@ struct GridwiseBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2 ...@@ -1775,7 +1777,7 @@ struct GridwiseBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_Light_V2
// dK: transform input and output tensor descriptors // dK: transform input and output tensor descriptors
auto kgrad_grid_desc_nblock_nperblock_oblock_operblock = auto kgrad_grid_desc_nblock_nperblock_oblock_operblock =
MakeKGradGridDesc_NBlock_NPerBlock_OBlock_OPerBlock(k_grid_desc_k0_n_k1); MakeKGradGridDesc_NBlock_NPerBlock_OBlock_OPerBlock(kgrad_grid_desc_k0_n_k1);
// //
// set up dQ Gemm (type 3 crr) // set up dQ Gemm (type 3 crr)
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment