Skip to content

Commit

Permalink
Replace c10::optional with std::optional
Browse files Browse the repository at this point in the history 
Signed-off-by: Lu Fang <[email protected]>
  • Loading branch information
@houseroad
houseroad committed Jan 4, 2025
1 parent e5d7ed0 commit 7ab7ce7
Show file tree
Hide file tree
Showing 24 changed files with 136 additions and 136 deletions.
4 changes: 2 additions & 2 deletions csrc/attention/paged_attention_v1.cu
View file
Original file line number Diff line number Diff line change
Expand Up @@ -53,7 +53,7 @@ void paged_attention_v1_launcher(
torch::Tensor& out, torch::Tensor& query, torch::Tensor& key_cache,
torch::Tensor& value_cache, int num_kv_heads, float scale,
torch::Tensor& block_tables, torch::Tensor& seq_lens, int max_seq_len,
const c10::optional<torch::Tensor>& alibi_slopes, float k_scale,
const std::optional<torch::Tensor>& alibi_slopes, float k_scale,
float v_scale, const int tp_rank, const int blocksparse_local_blocks,
const int blocksparse_vert_stride, const int blocksparse_block_size,
const int blocksparse_head_sliding_step) {
Expand Down Expand Up @@ -176,7 +176,7 @@ void paged_attention_v1(
torch::Tensor& block_tables, // [num_seqs, max_num_blocks_per_seq]
torch::Tensor& seq_lens, // [num_seqs]
int64_t block_size, int64_t max_seq_len,
const c10::optional<torch::Tensor>& alibi_slopes,
const std::optional<torch::Tensor>& alibi_slopes,
const std::string& kv_cache_dtype, double k_scale, double v_scale,
const int64_t tp_rank, const int64_t blocksparse_local_blocks,
const int64_t blocksparse_vert_stride, const int64_t blocksparse_block_size,
Expand Down
4 changes: 2 additions & 2 deletions csrc/attention/paged_attention_v2.cu
View file
Original file line number Diff line number Diff line change
Expand Up @@ -54,7 +54,7 @@ void paged_attention_v2_launcher(
torch::Tensor& tmp_out, torch::Tensor& query, torch::Tensor& key_cache,
torch::Tensor& value_cache, int num_kv_heads, float scale,
torch::Tensor& block_tables, torch::Tensor& seq_lens, int max_seq_len,
const c10::optional<torch::Tensor>& alibi_slopes, float k_scale,
const std::optional<torch::Tensor>& alibi_slopes, float k_scale,
float v_scale, const int tp_rank, const int blocksparse_local_blocks,
const int blocksparse_vert_stride, const int blocksparse_block_size,
const int blocksparse_head_sliding_step) {
Expand Down Expand Up @@ -187,7 +187,7 @@ void paged_attention_v2(
torch::Tensor& block_tables, // [num_seqs, max_num_blocks_per_seq]
torch::Tensor& seq_lens, // [num_seqs]
int64_t block_size, int64_t max_seq_len,
const c10::optional<torch::Tensor>& alibi_slopes,
const std::optional<torch::Tensor>& alibi_slopes,
const std::string& kv_cache_dtype, double k_scale, double v_scale,
const int64_t tp_rank, const int64_t blocksparse_local_blocks,
const int64_t blocksparse_vert_stride, const int64_t blocksparse_block_size,
Expand Down
8 changes: 4 additions & 4 deletions csrc/cpu/attention.cpp
View file
Original file line number Diff line number Diff line change
Expand Up @@ -386,7 +386,7 @@ void paged_attention_v1_impl_launcher(
torch::Tensor& out, torch::Tensor& query, torch::Tensor& key_cache,
torch::Tensor& value_cache, int num_kv_heads, float scale,
torch::Tensor& block_tables, torch::Tensor& seq_lens, int max_seq_len,
const c10::optional<torch::Tensor>& alibi_slopes) {
const std::optional<torch::Tensor>& alibi_slopes) {
int num_seqs = query.size(0);
int num_heads = query.size(1);
int head_size = query.size(2);
Expand Down Expand Up @@ -459,7 +459,7 @@ void paged_attention_v1(
torch::Tensor& out, torch::Tensor& query, torch::Tensor& key_cache,
torch::Tensor& value_cache, int64_t num_kv_heads, double scale,
torch::Tensor& block_tables, torch::Tensor& seq_lens, int64_t block_size,
int64_t max_seq_len, const c10::optional<torch::Tensor>& alibi_slopes,
int64_t max_seq_len, const std::optional<torch::Tensor>& alibi_slopes,
const std::string& kv_cache_dtype, double k_scale, double v_scale,
const int64_t tp_rank, const int64_t blocksparse_local_blocks,
const int64_t blocksparse_vert_stride, const int64_t blocksparse_block_size,
Expand Down Expand Up @@ -702,7 +702,7 @@ void paged_attention_v2_impl_launcher(
torch::Tensor& tmp_out, torch::Tensor& query, torch::Tensor& key_cache,
torch::Tensor& value_cache, int num_kv_heads, float scale,
torch::Tensor& block_tables, torch::Tensor& seq_lens, int block_size,
int max_seq_len, const c10::optional<torch::Tensor>& alibi_slopes) {
int max_seq_len, const std::optional<torch::Tensor>& alibi_slopes) {
int num_seqs = query.size(0);
int num_heads = query.size(1);
int head_size = query.size(2);
Expand Down Expand Up @@ -781,7 +781,7 @@ void paged_attention_v2(
torch::Tensor& tmp_out, torch::Tensor& query, torch::Tensor& key_cache,
torch::Tensor& value_cache, int64_t num_kv_heads, double scale,
torch::Tensor& block_tables, torch::Tensor& seq_lens, int64_t block_size,
int64_t max_seq_len, const c10::optional<torch::Tensor>& alibi_slopes,
int64_t max_seq_len, const std::optional<torch::Tensor>& alibi_slopes,
const std::string& kv_cache_dtype, double k_scale, double v_scale,
const int64_t tp_rank, const int64_t blocksparse_local_blocks,
const int64_t blocksparse_vert_stride, const int64_t blocksparse_block_size,
Expand Down
10 changes: 5 additions & 5 deletions csrc/cpu/quant.cpp
View file
Original file line number Diff line number Diff line change
Expand Up @@ -359,7 +359,7 @@ void int8_scaled_mm(torch::Tensor& c, // [M, OC], row-major
const torch::Tensor& b, // [IC, OC], column-major
const torch::Tensor& a_scales, // [1] or [M]
const torch::Tensor& b_scales, // [1] or [OC]
const c10::optional<torch::Tensor>& bias // [OC]
const std::optional<torch::Tensor>& bias // [OC]
) {
CPU_KERNEL_GUARD_IN(cutlass_scaled_mm)
// Checks for conformality
Expand Down Expand Up @@ -442,8 +442,8 @@ void int8_scaled_mm_azp(torch::Tensor& c, // [M, OC], row-major
const torch::Tensor& a_scales, // [1] or [M]
const torch::Tensor& b_scales, // [1] or [OC]
const torch::Tensor& azp_adj, // [OC]
const c10::optional<torch::Tensor>& azp, // [1] or [M]
const c10::optional<torch::Tensor>& bias // [OC]
const std::optional<torch::Tensor>& azp, // [1] or [M]
const std::optional<torch::Tensor>& bias // [OC]
) {
CPU_KERNEL_GUARD_IN(cutlass_scaled_mm_azp)
// Checks for conformality
Expand Down Expand Up @@ -561,7 +561,7 @@ void int8_scaled_mm_azp(torch::Tensor& c, // [M, OC], row-major
void static_scaled_int8_quant(torch::Tensor& out, // [..., hidden_size]
const torch::Tensor& input, // [..., hidden_size]
const torch::Tensor& scale,
c10::optional<torch::Tensor> const& azp) {
std::optional<torch::Tensor> const& azp) {
CPU_KERNEL_GUARD_IN(static_scaled_int8_quant)
TORCH_CHECK(input.is_contiguous());
TORCH_CHECK(out.is_contiguous());
Expand Down Expand Up @@ -590,7 +590,7 @@ void dynamic_scaled_int8_quant(
torch::Tensor& out, // [..., hidden_size]
const torch::Tensor& input, // [..., hidden_size]
torch::Tensor& scale, // [..., 1]
c10::optional<torch::Tensor> const& azp) {
std::optional<torch::Tensor> const& azp) {
CPU_KERNEL_GUARD_IN(dynamic_scaled_int8_quant)
TORCH_CHECK(input.is_contiguous());
TORCH_CHECK(out.is_contiguous());
Expand Down
6 changes: 3 additions & 3 deletions csrc/cpu/torch_bindings.cpp
View file
Original file line number Diff line number Diff line change
Expand Up @@ -9,14 +9,14 @@ std::string init_cpu_threads_env(const std::string& cpu_ids);
void int8_scaled_mm(torch::Tensor& c, const torch::Tensor& a,
const torch::Tensor& b, const torch::Tensor& a_scales,
const torch::Tensor& b_scales,
const c10::optional<torch::Tensor>& bias);
const std::optional<torch::Tensor>& bias);

void int8_scaled_mm_azp(torch::Tensor& c, const torch::Tensor& a,
const torch::Tensor& b, const torch::Tensor& a_scales,
const torch::Tensor& b_scales,
const torch::Tensor& azp_adj,
const c10::optional<torch::Tensor>& azp,
const c10::optional<torch::Tensor>& bias);
const std::optional<torch::Tensor>& azp,
const std::optional<torch::Tensor>& bias);

TORCH_LIBRARY_EXPAND(TORCH_EXTENSION_NAME, ops) {
// vLLM custom ops
Expand Down
6 changes: 3 additions & 3 deletions csrc/cutlass_extensions/epilogue/scaled_mm_epilogues_c2x.hpp
View file
Original file line number Diff line number Diff line change
Expand Up @@ -68,7 +68,7 @@ struct ScaledEpilogueBase {
// This overload handles the case where there might not be a tensor, in which
// case a nullptr is passed and a constant (0) is used.
template <typename Descriptor, typename T>
static auto args_from_tensor(c10::optional<torch::Tensor> const& tensor) {
static auto args_from_tensor(std::optional<torch::Tensor> const& tensor) {
static_assert(std::is_same_v<Descriptor, RowOrZeroLoad<T>>);
using Arguments = typename Descriptor::Arguments;
auto* data_ptr = tensor ? static_cast<T*>(tensor->data_ptr()) : nullptr;
Expand Down Expand Up @@ -223,7 +223,7 @@ struct ScaledEpilogueBiasAzp
static ArgumentType prepare_args(torch::Tensor const& a_scales,
torch::Tensor const& b_scales,
torch::Tensor const& azp_adj,
c10::optional<torch::Tensor> const& bias) {
std::optional<torch::Tensor> const& bias) {
auto a_args = SUPER::template args_from_tensor<ScaleA, float>(a_scales);
auto b_args = SUPER::template args_from_tensor<ScaleB, float>(b_scales);
auto bias_args = SUPER::template args_from_tensor<Bias, ElementD>(bias);
Expand Down Expand Up @@ -301,7 +301,7 @@ struct ScaledEpilogueBiasAzpToken
torch::Tensor const& b_scales,
torch::Tensor const& azp_adj,
torch::Tensor const& azp,
c10::optional<torch::Tensor> const& bias) {
std::optional<torch::Tensor> const& bias) {
auto a_args = SUPER::template args_from_tensor<ScaleA, float>(a_scales);
auto b_args = SUPER::template args_from_tensor<ScaleB, float>(b_scales);
auto bias_args = SUPER::template args_from_tensor<Bias, ElementD>(bias);
Expand Down
6 changes: 3 additions & 3 deletions csrc/cutlass_extensions/epilogue/scaled_mm_epilogues_c3x.hpp
View file
Original file line number Diff line number Diff line change
Expand Up @@ -67,7 +67,7 @@ struct ScaledEpilogueBase {
// This overload handles the case where there might not be a tensor, in which
// case a nullptr is passed and a constant (0) is used.
template <typename Descriptor, typename T>
static auto args_from_tensor(c10::optional<torch::Tensor> const& tensor) {
static auto args_from_tensor(std::optional<torch::Tensor> const& tensor) {
using Arguments = typename Descriptor::Arguments;
auto* data_ptr = tensor ? static_cast<T*>(tensor->data_ptr()) : nullptr;
static_assert(std::is_same_v<Descriptor, ColLoad<T, true>> ||
Expand Down Expand Up @@ -223,7 +223,7 @@ struct ScaledEpilogueBiasAzp
static ArgumentType prepare_args(torch::Tensor const& a_scales,
torch::Tensor const& b_scales,
torch::Tensor const& azp_adj,
c10::optional<torch::Tensor> const& bias) {
std::optional<torch::Tensor> const& bias) {
auto a_args = SUPER::template args_from_tensor<ScaleA, float>(a_scales);
auto b_args = SUPER::template args_from_tensor<ScaleB, float>(b_scales);
auto bias_args = SUPER::template args_from_tensor<Bias, ElementD>(bias);
Expand Down Expand Up @@ -299,7 +299,7 @@ struct ScaledEpilogueBiasAzpToken
torch::Tensor const& b_scales,
torch::Tensor const& azp_adj,
torch::Tensor const& azp,
c10::optional<torch::Tensor> const& bias) {
std::optional<torch::Tensor> const& bias) {
auto a_args = SUPER::template args_from_tensor<ScaleA, float>(a_scales);
auto b_args = SUPER::template args_from_tensor<ScaleB, float>(b_scales);
auto bias_args = SUPER::template args_from_tensor<Bias, ElementD>(bias);
Expand Down
2 changes: 1 addition & 1 deletion csrc/cutlass_extensions/torch_utils.hpp
View file
Original file line number Diff line number Diff line change
Expand Up @@ -97,7 +97,7 @@ static inline auto make_cute_layout(torch::Tensor const& tensor,

template <typename Stride>
static inline auto maybe_make_cute_layout(
c10::optional<torch::Tensor> const& tensor,
std::optional<torch::Tensor> const& tensor,
std::string_view name = "tensor") {
using Layout = decltype(make_cute_layout<Stride>(*tensor));

Expand Down
24 changes: 12 additions & 12 deletions csrc/mamba/causal_conv1d/causal_conv1d.cu
View file
Original file line number Diff line number Diff line change
Expand Up @@ -53,12 +53,12 @@ void set_conv_params_fwd(ConvParamsBase &params,
const at::Tensor x,
const at::Tensor weight,
const at::Tensor out,
const c10::optional<at::Tensor>& bias,
const std::optional<at::Tensor>& bias,
bool silu_activation,
int64_t pad_slot_id,
const c10::optional<at::Tensor>& query_start_loc = std::nullopt,
const c10::optional<at::Tensor>& cache_indices = std::nullopt,
const c10::optional<at::Tensor>& has_initial_state = std::nullopt) {
const std::optional<at::Tensor>& query_start_loc = std::nullopt,
const std::optional<at::Tensor>& cache_indices = std::nullopt,
const std::optional<at::Tensor>& has_initial_state = std::nullopt) {

// Reset the parameters
memset(&params, 0, sizeof(params));
Expand Down Expand Up @@ -93,11 +93,11 @@ void set_conv_params_fwd(ConvParamsBase &params,


void causal_conv1d_fwd(const at::Tensor &x, const at::Tensor &weight,
const c10::optional<at::Tensor> &bias_,
const c10::optional<at::Tensor> &conv_states,
const c10::optional<at::Tensor> &query_start_loc,
const c10::optional<at::Tensor> &cache_indices,
const c10::optional<at::Tensor> &has_initial_state,
const std::optional<at::Tensor> &bias_,
const std::optional<at::Tensor> &conv_states,
const std::optional<at::Tensor> &query_start_loc,
const std::optional<at::Tensor> &cache_indices,
const std::optional<at::Tensor> &has_initial_state,
bool silu_activation,
// used to identify padding entries if cache_indices provided
// in case of padding, the kernel will return early
Expand Down Expand Up @@ -194,10 +194,10 @@ void causal_conv1d_fwd(const at::Tensor &x, const at::Tensor &weight,
void causal_conv1d_update(const at::Tensor &x,
const at::Tensor &conv_state,
const at::Tensor &weight,
const c10::optional<at::Tensor> &bias_,
const std::optional<at::Tensor> &bias_,
bool silu_activation,
const c10::optional<at::Tensor> &cache_seqlens_,
const c10::optional<at::Tensor> &conv_state_indices_,
const std::optional<at::Tensor> &cache_seqlens_,
const std::optional<at::Tensor> &conv_state_indices_,
// used to identify padding entries if cache_indices provided
// in case of padding, the kernel will return early
int64_t pad_slot_id) {
Expand Down
22 changes: 11 additions & 11 deletions csrc/mamba/mamba_ssm/selective_scan_fwd.cu
View file
Original file line number Diff line number Diff line change
Expand Up @@ -402,14 +402,14 @@ void set_ssm_params_fwd(SSMParamsBase &params,
const torch::Tensor out,
const torch::Tensor z,
const torch::Tensor out_z,
const c10::optional<at::Tensor>& D,
const c10::optional<at::Tensor>& delta_bias,
const std::optional<at::Tensor>& D,
const std::optional<at::Tensor>& delta_bias,
const torch::Tensor ssm_states,
bool has_z,
bool delta_softplus,
const c10::optional<at::Tensor>& query_start_loc,
const c10::optional<at::Tensor>& cache_indices,
const c10::optional<at::Tensor>& has_initial_state,
const std::optional<at::Tensor>& query_start_loc,
const std::optional<at::Tensor>& cache_indices,
const std::optional<at::Tensor>& has_initial_state,
bool varlen,
int64_t pad_slot_id) {

Expand Down Expand Up @@ -504,13 +504,13 @@ void set_ssm_params_fwd(SSMParamsBase &params,

void selective_scan_fwd(const torch::Tensor &u, const torch::Tensor &delta,
const torch::Tensor &A, const torch::Tensor &B, const torch::Tensor &C,
const c10::optional<torch::Tensor> &D_,
const c10::optional<torch::Tensor> &z_,
const c10::optional<torch::Tensor> &delta_bias_,
const std::optional<torch::Tensor> &D_,
const std::optional<torch::Tensor> &z_,
const std::optional<torch::Tensor> &delta_bias_,
bool delta_softplus,
const c10::optional<torch::Tensor> &query_start_loc,
const c10::optional<torch::Tensor> &cache_indices,
const c10::optional<torch::Tensor> &has_initial_state,
const std::optional<torch::Tensor> &query_start_loc,
const std::optional<torch::Tensor> &cache_indices,
const std::optional<torch::Tensor> &has_initial_state,
const torch::Tensor &ssm_states,
// used to identify padding entries if cache_indices provided
// in case of padding, the kernel will return early
Expand Down
Loading

0 comments on commit 7ab7ce7

Please sign in to comment.