Skip to content

Commit

Permalink
[Kernel] Update Cutlass int8 kernel configs for SM90 (vllm-project#5514)
Browse files Browse the repository at this point in the history 
Co-authored-by: Varun Sundar Rabindranath <varun@neuralmagic.com>
  • Loading branch information
@varun-sundar-rabindranath
varun-sundar-rabindranath and Varun Sundar Rabindranath authored Jun 20, 2024
1 parent 1b2eaac commit 111af1f
Showing 1 changed file with 143 additions and 22 deletions.
165 changes: 143 additions & 22 deletions csrc/quantization/cutlass_w8a8/scaled_mm_c3x.cu
View file
Original file line number Diff line number Diff line change
Expand Up @@ -234,38 +234,39 @@ void cutlass_gemm_caller(torch::Tensor& out, torch::Tensor const& a,
}

template <typename InType, typename OutType,
template <typename, typename, typename> typename Epilogue, int32_t M>
struct sm90_fp8_config {
template <typename, typename, typename> typename Epilogue>
struct sm90_fp8_config_default {
// M in (128, inf)
static_assert(std::is_same<InType, cutlass::float_e4m3_t>());
using KernelSchedule =
cutlass::gemm::KernelTmaWarpSpecializedPingpongFP8FastAccum;
using EpilogueSchedule = typename cutlass::epilogue::TmaWarpSpecialized;
using TileShape = Shape<_128, _128, _128>;
using ClusterShape = Shape<_2, _1, _1>;

using Cutlass3xGemm =
cutlass_3x_gemm<InType, OutType, Epilogue, TileShape, ClusterShape,
KernelSchedule, EpilogueSchedule>;
};

template <typename InType, typename OutType,
template <typename, typename, typename> typename Epilogue>
struct sm90_fp8_config<InType, OutType, Epilogue, 128> {
struct sm90_fp8_config_M128 {
// M in (64, 128]
static_assert(std::is_same<InType, cutlass::float_e4m3_t>());
using KernelSchedule =
cutlass::gemm::KernelTmaWarpSpecializedPingpongFP8FastAccum;
using EpilogueSchedule = typename cutlass::epilogue::TmaWarpSpecialized;
using TileShape = Shape<_64, _128, _128>;
using ClusterShape = Shape<_2, _1, _1>;

using Cutlass3xGemm =
cutlass_3x_gemm<InType, OutType, Epilogue, TileShape, ClusterShape,
KernelSchedule, EpilogueSchedule>;
};

template <typename InType, typename OutType,
template <typename, typename, typename> typename Epilogue>
struct sm90_fp8_config<InType, OutType, Epilogue, 64> {
struct sm90_fp8_config_M64 {
// M in [1, 64]
static_assert(std::is_same<InType, cutlass::float_e4m3_t>());
using KernelSchedule =
cutlass::gemm::KernelTmaWarpSpecializedPingpongFP8FastAccum;
Expand All @@ -278,6 +279,78 @@ struct sm90_fp8_config<InType, OutType, Epilogue, 64> {
KernelSchedule, EpilogueSchedule>;
};

template <typename InType, typename OutType,
template <typename, typename, typename> typename Epilogue>
struct sm90_int8_config_default {
// For M > 128 and any N
static_assert(std::is_same<InType, int8_t>());
using KernelSchedule =
typename cutlass::gemm::KernelTmaWarpSpecializedPingpong;
using EpilogueSchedule = typename cutlass::epilogue::TmaWarpSpecialized;
using TileShape = Shape<_128, _128, _128>;
using ClusterShape = Shape<_2, _1, _1>;
using Cutlass3xGemm =
cutlass_3x_gemm<InType, OutType, Epilogue, TileShape, ClusterShape,
KernelSchedule, EpilogueSchedule>;
};

template <typename InType, typename OutType,
template <typename, typename, typename> typename Epilogue>
struct sm90_int8_config_M128 {
// For M in (64, 128] and any N
static_assert(std::is_same<InType, int8_t>());
using KernelSchedule =
typename cutlass::gemm::KernelTmaWarpSpecializedPingpong;
using EpilogueSchedule = typename cutlass::epilogue::TmaWarpSpecialized;
using TileShape = Shape<_64, _128, _128>;
using ClusterShape = Shape<_2, _1, _1>;
using Cutlass3xGemm =
cutlass_3x_gemm<InType, OutType, Epilogue, TileShape, ClusterShape,
KernelSchedule, EpilogueSchedule>;
};

template <typename InType, typename OutType,
template <typename, typename, typename> typename Epilogue>
struct sm90_int8_config_M64 {
// For M in (32, 64] and any N
static_assert(std::is_same<InType, int8_t>());
using KernelSchedule = typename cutlass::gemm::KernelTmaWarpSpecialized;
using EpilogueSchedule = typename cutlass::epilogue::TmaWarpSpecialized;
using TileShape = Shape<_64, _64, _256>;
using ClusterShape = Shape<_1, _1, _1>;
using Cutlass3xGemm =
cutlass_3x_gemm<InType, OutType, Epilogue, TileShape, ClusterShape,
KernelSchedule, EpilogueSchedule>;
};

template <typename InType, typename OutType,
template <typename, typename, typename> typename Epilogue>
struct sm90_int8_config_M32_NBig {
// For M in [1, 32] and N >= 8192
static_assert(std::is_same<InType, int8_t>());
using KernelSchedule = typename cutlass::gemm::KernelTmaWarpSpecialized;
using EpilogueSchedule = typename cutlass::epilogue::TmaWarpSpecialized;
using TileShape = Shape<_64, _128, _256>;
using ClusterShape = Shape<_1, _4, _1>;
using Cutlass3xGemm =
cutlass_3x_gemm<InType, OutType, Epilogue, TileShape, ClusterShape,
KernelSchedule, EpilogueSchedule>;
};

template <typename InType, typename OutType,
template <typename, typename, typename> typename Epilogue>
struct sm90_int8_config_M32_NSmall {
// For M in [1, 32] and N < 8192
static_assert(std::is_same<InType, int8_t>());
using KernelSchedule = typename cutlass::gemm::KernelTmaWarpSpecialized;
using EpilogueSchedule = typename cutlass::epilogue::TmaWarpSpecialized;
using TileShape = Shape<_64, _64, _256>;
using ClusterShape = Shape<_1, _8, _1>;
using Cutlass3xGemm =
cutlass_3x_gemm<InType, OutType, Epilogue, TileShape, ClusterShape,
KernelSchedule, EpilogueSchedule>;
};

} // namespace

template <typename InType, typename OutType,
Expand All @@ -291,11 +364,12 @@ void cutlass_gemm_sm90_fp8_dispatch(torch::Tensor& out, torch::Tensor const& a,
TORCH_CHECK(b.dtype() == torch::kFloat8_e4m3fn);

using Cutlass3xGemmDefault =
typename sm90_fp8_config<InType, OutType, Epilogue, 0>::Cutlass3xGemm;
typename sm90_fp8_config_default<InType, OutType,
Epilogue>::Cutlass3xGemm;
using Cutlass3xGemmM64 =
typename sm90_fp8_config<InType, OutType, Epilogue, 64>::Cutlass3xGemm;
typename sm90_fp8_config_M64<InType, OutType, Epilogue>::Cutlass3xGemm;
using Cutlass3xGemmM128 =
typename sm90_fp8_config<InType, OutType, Epilogue, 128>::Cutlass3xGemm;
typename sm90_fp8_config_M128<InType, OutType, Epilogue>::Cutlass3xGemm;

uint32_t const m = a.size(0);
uint32_t const mp2 =
Expand All @@ -316,6 +390,61 @@ void cutlass_gemm_sm90_fp8_dispatch(torch::Tensor& out, torch::Tensor const& a,
}
}

template <typename InType, typename OutType,
template <typename, typename, typename> typename Epilogue,
typename... EpilogueArgs>
void cutlass_gemm_sm90_int8_dispatch(torch::Tensor& out, torch::Tensor const& a,
torch::Tensor const& b,
EpilogueArgs&&... args) {
static_assert(std::is_same<InType, int8_t>());
TORCH_CHECK(a.dtype() == torch::kInt8);
TORCH_CHECK(b.dtype() == torch::kInt8);

using Cutlass3xGemmDefault =
typename sm90_int8_config_default<InType, OutType,
Epilogue>::Cutlass3xGemm;
using Cutlass3xGemmM128 =
typename sm90_int8_config_M128<InType, OutType, Epilogue>::Cutlass3xGemm;
using Cutlass3xGemmM64 =
typename sm90_int8_config_M64<InType, OutType, Epilogue>::Cutlass3xGemm;
using Cutlass3xGemmM32NBig =
typename sm90_int8_config_M32_NBig<InType, OutType,
Epilogue>::Cutlass3xGemm;
using Cutlass3xGemmM32NSmall =
typename sm90_int8_config_M32_NSmall<InType, OutType,
Epilogue>::Cutlass3xGemm;

uint32_t const n = out.size(1);
bool const is_small_n = n < 8192;

uint32_t const m = a.size(0);
uint32_t const mp2 =
std::max(static_cast<uint32_t>(32), next_pow_2(m)); // next power of 2

if (mp2 <= 32) {
// m in [1, 32]
if (is_small_n) {
return cutlass_gemm_caller<Cutlass3xGemmM32NSmall>(
out, a, b, std::forward<EpilogueArgs>(args)...);
} else {
return cutlass_gemm_caller<Cutlass3xGemmM32NBig>(
out, a, b, std::forward<EpilogueArgs>(args)...);
}
} else if (mp2 <= 64) {
// m in (32, 64]
return cutlass_gemm_caller<Cutlass3xGemmM64>(
out, a, b, std::forward<EpilogueArgs>(args)...);
} else if (mp2 <= 128) {
// m in (64, 128]
return cutlass_gemm_caller<Cutlass3xGemmM128>(
out, a, b, std::forward<EpilogueArgs>(args)...);
} else {
// m in (128, inf)
return cutlass_gemm_caller<Cutlass3xGemmDefault>(
out, a, b, std::forward<EpilogueArgs>(args)...);
}
}

void cutlass_scaled_mm_sm90(torch::Tensor& out, torch::Tensor const& a,
torch::Tensor const& b,
torch::Tensor const& a_scales,
Expand All @@ -326,22 +455,14 @@ void cutlass_scaled_mm_sm90(torch::Tensor& out, torch::Tensor const& a,
if (a.dtype() == torch::kInt8) {
TORCH_CHECK(b.dtype() == torch::kInt8);

using TileShape = Shape<_128, _128, _128>;
using ClusterShape = Shape<_1, _2, _1>;
using KernelSchedule =
typename cutlass::gemm::KernelTmaWarpSpecializedPingpong;
using EpilogueSchedule = typename cutlass::epilogue::TmaWarpSpecialized;

if (out.dtype() == torch::kBFloat16) {
return cutlass_gemm_caller<cutlass_3x_gemm<
int8_t, cutlass::bfloat16_t, ScaledEpilogue, TileShape, ClusterShape,
KernelSchedule, EpilogueSchedule>>(out, a, b, a_scales, b_scales);
return cutlass_gemm_sm90_int8_dispatch<int8_t, cutlass::bfloat16_t,
ScaledEpilogue>(
out, a, b, a_scales, b_scales);
} else {
TORCH_CHECK(out.dtype() == torch::kFloat16);

return cutlass_gemm_caller<
cutlass_3x_gemm<int8_t, cutlass::half_t, ScaledEpilogue, TileShape,
ClusterShape, KernelSchedule, EpilogueSchedule>>(
return cutlass_gemm_sm90_int8_dispatch<int8_t, cutlass::half_t,
ScaledEpilogue>(
out, a, b, a_scales, b_scales);
}
} else {
Expand Down

0 comments on commit 111af1f

Please sign in to comment.