[DSV4] Add BF16 and MXFP8 A2A support for flashinfer a2a one sided (#40960)
Signed-off-by: Yongye Zhu <zyy1102000@gmail.com> Signed-off-by: Zijing Liu <liuzijing2014@gmail.com> Co-authored-by: Zijing Liu <liuzijing2014@users.noreply.github.com> Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This commit is contained in:
co-authored by
Zijing Liu
Claude Opus 4.7
parent
526927be94
commit
b4806c8ee1
@@ -577,6 +577,8 @@ class FlashInferNVLinkOneSidedManager(All2AllManagerBase):
|
||||
top_k: int,
|
||||
num_experts: int,
|
||||
hidden_size: int,
|
||||
dispatch_dtype_bytes_per_elem: int = 0,
|
||||
dispatch_scale_bytes_per_token: int = 0,
|
||||
):
|
||||
"""Initialize the MoeAlltoAll workspace."""
|
||||
if self.initialized:
|
||||
@@ -607,9 +609,13 @@ class FlashInferNVLinkOneSidedManager(All2AllManagerBase):
|
||||
ep_config = MnnvlConfig(
|
||||
comm_backend=CustomCommunicator(self.cpu_group),
|
||||
)
|
||||
if dispatch_dtype_bytes_per_elem == 0:
|
||||
hidden_bytes = hidden_size // 2
|
||||
else:
|
||||
hidden_bytes = hidden_size * dispatch_dtype_bytes_per_elem
|
||||
total_dispatch_payload_size_per_token = (
|
||||
hidden_size // 2 # nvfp4 hidden states
|
||||
+ hidden_size // 16 # fp8 scaling factors
|
||||
hidden_bytes
|
||||
+ dispatch_scale_bytes_per_token
|
||||
+ top_k * 4 # int32 topks ids
|
||||
+ top_k * 4 # float32 topk weights
|
||||
)
|
||||
|
||||
Reference in New Issue
Block a user