[Feature] Support Decode Context Parallel (DCP) for MLA (#23734)

Signed-off-by: hongchao <hongchao@msh.team>
Signed-off-by: youkaichao <youkaichao@gmail.com>
Co-authored-by: hongchao <hongchao@msh.team>
Co-authored-by: youkaichao <youkaichao@gmail.com>
This commit is contained in:
yzds
2025-09-06 13:24:05 +08:00
committed by GitHub
co-authored by hongchao youkaichao
parent 3c529fc994
commit ac201a0eaf
27 changed files with 999 additions and 230 deletions
-14
View File
@@ -1625,20 +1625,6 @@ def concat_and_cache_mla(
scale)
def cp_fused_concat_and_cache_mla(
kv_c: torch.Tensor,
k_pe: torch.Tensor,
cp_local_token_select_indices: torch.Tensor,
kv_cache: torch.Tensor,
slot_mapping: torch.Tensor,
kv_cache_dtype: str,
scale: torch.Tensor,
) -> None:
torch.ops._C_cache_ops.cp_fused_concat_and_cache_mla(
kv_c, k_pe, cp_local_token_select_indices, kv_cache, slot_mapping,
kv_cache_dtype, scale)
def copy_blocks(key_caches: list[torch.Tensor],
value_caches: list[torch.Tensor],
block_mapping: torch.Tensor) -> None: