Delete PagedAttention (#47361)
Signed-off-by: mgoin <mgoin64@gmail.com>
This commit is contained in:
@@ -598,33 +598,6 @@ STABLE_TORCH_LIBRARY_FRAGMENT(_C, ops) {
|
||||
"Tensor? initial_state_idx,"
|
||||
"Tensor? cu_chunk_seqlen,"
|
||||
"Tensor? last_chunk_indices) -> ()");
|
||||
|
||||
// Attention ops
|
||||
// Compute the attention between an input query and the cached
|
||||
// keys/values using PagedAttention.
|
||||
ops.def(
|
||||
"paged_attention_v1("
|
||||
" Tensor! out, Tensor query, Tensor key_cache,"
|
||||
" Tensor value_cache, int num_kv_heads, float scale,"
|
||||
" Tensor block_tables, Tensor seq_lens, int block_size,"
|
||||
" int max_seq_len, Tensor? alibi_slopes,"
|
||||
" str kv_cache_dtype, Tensor k_scale, Tensor v_scale,"
|
||||
" int tp_rank, int blocksparse_local_blocks,"
|
||||
" int blocksparse_vert_stride, int blocksparse_block_size,"
|
||||
" int blocksparse_head_sliding_step) -> ()");
|
||||
|
||||
// PagedAttention V2.
|
||||
ops.def(
|
||||
"paged_attention_v2("
|
||||
" Tensor! out, Tensor! exp_sums, Tensor! max_logits,"
|
||||
" Tensor! tmp_out, Tensor query, Tensor key_cache,"
|
||||
" Tensor value_cache, int num_kv_heads, float scale,"
|
||||
" Tensor block_tables, Tensor seq_lens, int block_size,"
|
||||
" int max_seq_len, Tensor? alibi_slopes,"
|
||||
" str kv_cache_dtype, Tensor k_scale, Tensor v_scale,"
|
||||
" int tp_rank, int blocksparse_local_blocks,"
|
||||
" int blocksparse_vert_stride, int blocksparse_block_size,"
|
||||
" int blocksparse_head_sliding_step) -> ()");
|
||||
}
|
||||
|
||||
STABLE_TORCH_LIBRARY_IMPL(_C, CUDA, ops) {
|
||||
@@ -747,9 +720,6 @@ STABLE_TORCH_LIBRARY_IMPL(_C, CUDA, ops) {
|
||||
|
||||
// Mamba kernels
|
||||
ops.impl("selective_scan_fwd", TORCH_BOX(&selective_scan_fwd));
|
||||
|
||||
ops.impl("paged_attention_v1", TORCH_BOX(&paged_attention_v1));
|
||||
ops.impl("paged_attention_v2", TORCH_BOX(&paged_attention_v2));
|
||||
}
|
||||
|
||||
STABLE_TORCH_LIBRARY_IMPL(_C, CPU, ops) {
|
||||
|
||||
Reference in New Issue
Block a user