ViperEkura
41dcf0feb9
- PagedAttentionParams uses flat KV pool + req_to_token + kv_indptr/qo_indptr instead of page_table - MMA split-KV decode and split-Q prefill kernels with indirect ragged-batch addressing - Prefill kernel accepts 4D mask (causal-aware); decode kernel supports 2D mask - CudaBackend is inference-only: kv_cache=None raises, no torch fallback - benchmark.py: required --ckpt, --backend/--compare options - Parallel build isolates build-temp/build-lib per subprocess - Standalone test covers decode/prefill with mask, 27 cases pass
46 lines
1.4 KiB
Plaintext
46 lines
1.4 KiB
Plaintext
#include "attn_dispatchers.cuh"
|
|
#include "attn_entry_utils.cuh"
|
|
|
|
torch::Tensor attn_paged_prefill(
|
|
torch::Tensor q,
|
|
torch::Tensor k_cache,
|
|
torch::Tensor v_cache,
|
|
torch::Tensor req_to_token,
|
|
torch::Tensor req_pool_indices,
|
|
torch::Tensor kv_indptr,
|
|
torch::Tensor qo_indptr,
|
|
c10::optional<torch::Tensor> mask,
|
|
int64_t max_q_len,
|
|
int64_t causal_offset,
|
|
double scale
|
|
) {
|
|
PagedAttentionParams<bf16> p;
|
|
attn_pack_paged_prefill_params(q, k_cache, v_cache,
|
|
req_to_token, req_pool_indices,
|
|
kv_indptr, qo_indptr, mask,
|
|
max_q_len, causal_offset, scale, p);
|
|
|
|
auto O = torch::empty({q.size(0), q.size(1), q.size(2)}, q.options());
|
|
p.o = (bf16*)O.data_ptr();
|
|
|
|
DISPATCH_HEAD_DIM(p.head_dim, dispatch_paged_prefill, p);
|
|
C10_CUDA_CHECK(cudaGetLastError());
|
|
return O;
|
|
}
|
|
|
|
PYBIND11_MODULE(TORCH_EXTENSION_NAME, m) {
|
|
m.def("attn_paged_prefill", &attn_paged_prefill,
|
|
py::arg("q"),
|
|
py::arg("k_cache"),
|
|
py::arg("v_cache"),
|
|
py::arg("req_to_token"),
|
|
py::arg("req_pool_indices"),
|
|
py::arg("kv_indptr"),
|
|
py::arg("qo_indptr"),
|
|
py::arg("mask") = py::none(),
|
|
py::arg("max_q_len"),
|
|
py::arg("causal_offset") = -1,
|
|
py::arg("scale") = 0.0,
|
|
"SGLang-style paged prefill: flat KV pool + ragged batch.");
|
|
}
|