[Kernel][Attention] Separate Attention.kv_scale into k_scale and v_scale (#6081)

This commit is contained in:
Michael Goin
2024-07-16 15:31:32 -07:00
committed by GitHub
parent 160e1d8c99
commit 978aed5300
33 changed files with 317 additions and 185 deletions
+6 -6
View File
@@ -423,11 +423,11 @@ void paged_attention_v1(
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,
const std::string& kv_cache_dtype, double kv_scale, const int64_t tp_rank,
const int64_t blocksparse_local_blocks,
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,
const int64_t blocksparse_head_sliding_step) {
TORCH_CHECK(kv_scale == 1.0f);
TORCH_CHECK(k_scale == 1.0f && v_scale == 1.0f);
TORCH_CHECK(blocksparse_vert_stride <= 1,
"CPU backend does not support blocksparse attention yet.");
VLLM_DISPATCH_FLOATING_TYPES(query.scalar_type(), "paged_attention_v1_impl",
@@ -742,11 +742,11 @@ void paged_attention_v2(
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,
const std::string& kv_cache_dtype, double kv_scale, const int64_t tp_rank,
const int64_t blocksparse_local_blocks,
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,
const int64_t blocksparse_head_sliding_step) {
TORCH_CHECK(kv_scale == 1.0f);
TORCH_CHECK(k_scale == 1.0f && v_scale == 1.0f);
TORCH_CHECK(blocksparse_vert_stride <= 1,
"CPU backend does not support blocksparse attention yet.");
VLLM_DISPATCH_FLOATING_TYPES(query.scalar_type(), "paged_attention_v2_impl",
+3 -2
View File
@@ -107,8 +107,9 @@ void copy_blocks(std::vector<torch::Tensor> const& key_caches,
void reshape_and_cache(torch::Tensor& key, torch::Tensor& value,
torch::Tensor& key_cache, torch::Tensor& value_cache,
torch::Tensor& slot_mapping,
const std::string& kv_cache_dtype, double kv_scale) {
TORCH_CHECK(kv_scale == 1.0f);
const std::string& kv_cache_dtype, double k_scale,
double v_scale) {
TORCH_CHECK(k_scale == 1.0f && v_scale == 1.0f);
int num_tokens = key.size(0);
int num_heads = key.size(1);
+5 -5
View File
@@ -16,8 +16,8 @@ TORCH_LIBRARY_EXPAND(TORCH_EXTENSION_NAME, ops) {
" 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, float kv_scale, int tp_rank,"
" int blocksparse_local_blocks,"
" str kv_cache_dtype, float k_scale, float v_scale,"
" int tp_rank, int blocksparse_local_blocks,"
" int blocksparse_vert_stride, int blocksparse_block_size,"
" int blocksparse_head_sliding_step) -> ()");
ops.impl("paged_attention_v1", torch::kCPU, &paged_attention_v1);
@@ -30,8 +30,8 @@ TORCH_LIBRARY_EXPAND(TORCH_EXTENSION_NAME, ops) {
" 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, float kv_scale, int tp_rank,"
" int blocksparse_local_blocks,"
" str kv_cache_dtype, float k_scale, float v_scale,"
" int tp_rank, int blocksparse_local_blocks,"
" int blocksparse_vert_stride, int blocksparse_block_size,"
" int blocksparse_head_sliding_step) -> ()");
ops.impl("paged_attention_v2", torch::kCPU, &paged_attention_v2);
@@ -103,7 +103,7 @@ TORCH_LIBRARY_EXPAND(CONCAT(TORCH_EXTENSION_NAME, _cache_ops), cache_ops) {
" Tensor! key_cache, Tensor! value_cache,"
" Tensor slot_mapping,"
" str kv_cache_dtype,"
" float kv_scale) -> ()");
" float k_scale, float v_scale) -> ()");
cache_ops.impl("reshape_and_cache", torch::kCPU, &reshape_and_cache);
}