EngineX/xc-llm-ascend
3
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

[main] [refactor] refactor fused_moe.py to enable token_dispatchers (#2570)

Browse Source 
### What this PR does / why we need it?
Enable token_dispatcher to replace fused_experts_with_xxx in eager mode
### Does this PR introduce _any_ user-facing change?
no
### How was this patch tested?
e2e & ut


- vLLM version: v0.10.1.1
- vLLM main:
704432af3c

Signed-off-by: Pr0Wh1teGivee <calvin_zhu0210@outlook.com>
Co-authored-by: sherie <963372609@qq.com>
Co-authored-by: weijinqian0 <12153182+weijinqian0@users.noreply.github.com>
Co-authored-by: shiyuan680 <72335504+shiyuan680@users.noreply.github.com>
This commit is contained in:
weichen
2025-08-28 10:13:35 +08:00
committed by GitHub
parent 936c102105
commit 320edde2df
10 changed files with 1066 additions and 1639 deletions
Download Patch File Download Diff File Expand all files Collapse all files

840
vllm_ascend/ops/fused_moe.py
View File

@@ -16,14 +16,14 @@
# Adapted from vllm/tests/kernels/test_moe.py
import os
from typing import Any, Callable, Optional, Tuple, Union
from typing import Any, Callable, Optional
import torch
import torch.distributed as dist
import torch_npu
from torch import nn
from vllm.config import get_current_vllm_config
from vllm.distributed import (GroupCoordinator, get_tensor_model_parallel_rank,
from vllm.distributed import (get_tensor_model_parallel_rank,
get_tensor_model_parallel_world_size,
tensor_model_parallel_all_reduce)
from vllm.distributed.parallel_state import (get_dp_group, get_ep_group,
@@ -49,9 +49,8 @@ from vllm_ascend.ops.layers.experts_selector import select_experts
from vllm_ascend.ops.moe_dispatcher.token_dispatcher import (
MoEAlltoAllSeqOverLapDispatcher, MoEDispatcherConfig)
from vllm_ascend.ops.sequence_parallel import MetadataForPadding
from vllm_ascend.utils import (ACL_FORMAT_FRACTAL_NZ, AscendSocVersion,
dispose_tensor, get_all_reduce_merge_state,
get_ascend_soc_version,
from vllm_ascend.utils import (ACL_FORMAT_FRACTAL_NZ, dispose_tensor,
get_all_reduce_merge_state,
get_rm_router_logits_state, is_310p)
MOE_ALL2ALL_BUFFER: bool = envs_ascend.MOE_ALL2ALL_BUFFER
@@ -122,149 +121,6 @@ def process_topk_ids(topk_ids: torch.Tensor, expert_num: int, ep_size: int,
return topk_ids_pad, unpad_indices
def fused_experts_with_mc2(
hidden_states: torch.Tensor,
w1: torch.Tensor,
w2: torch.Tensor,
topk_weights: torch.Tensor,
topk_ids: torch.Tensor,
top_k: int,
moe_parallel_config: FusedMoEParallelConfig,
expert_map: torch.Tensor = None,
moe_all_to_all_group_name: Optional[str] = None,
shared_experts: Optional[Any] = None,
mc2_mask: Optional[torch.Tensor] = None,
) -> Union[torch.Tensor, Tuple[torch.Tensor, torch.Tensor]]:
quant_mode = 0
ep_rank_id = moe_parallel_config.ep_rank
ep_world_size = moe_parallel_config.ep_size
# NOTE: Currently, when in A3 or in torchair graph, we need to pass in some extra param into dispatch & combine
need_extra_args = (get_ascend_soc_version() == AscendSocVersion.A3)
# NOTE: Currently, when in A3, we need to pass in some extra param into dispatch & combine
a3_need_extra_args = get_ascend_soc_version() == AscendSocVersion.A3
enable_dispatch_v2 = hasattr(torch_npu, "npu_moe_distribute_dispatch_v2")
moe_expert_num = len(expert_map)
kwargs_mc2 = {
"x": hidden_states,
"expert_ids": topk_ids,
"expert_shard_type": 0,
"shared_expert_rank_num": 0,
"moe_expert_num": moe_expert_num,
"global_bs": 0,
}
stage1_kwargs = {
"scales": None,
"quant_mode": quant_mode,
"group_ep": moe_all_to_all_group_name,
"ep_world_size": ep_world_size,
"ep_rank_id": ep_rank_id,
}
if need_extra_args:
stage1_kwargs.update({
"group_tp": moe_all_to_all_group_name,
"tp_world_size": 1,
"tp_rank_id": 0,
})
if a3_need_extra_args and enable_dispatch_v2:
stage1_kwargs.update({
"x_active_mask": mc2_mask,
})
kwargs_mc2.update(stage1_kwargs)
output = torch_npu.npu_moe_distribute_dispatch_v2(
**kwargs_mc2
) if enable_dispatch_v2 else torch_npu.npu_moe_distribute_dispatch(
**kwargs_mc2)
# comm_stream.wait_stream(torch.npu.current_stream())
expand_x, dynamic_scale, assist_info_for_combine, expert_token_nums, ep_recv_counts = output[
0:5]
if shared_experts is not None:
shared_gate_up, _ = shared_experts.gate_up_proj(hidden_states)
shared_act = shared_experts.act_fn(shared_gate_up)
w1 = w1.transpose(1, 2)
group_list = expert_token_nums.to(torch.int64)
gate_up_out_list = torch_npu.npu_grouped_matmul(
x=[expand_x],
weight=[w1],
split_item=2,
# 1 means count mode, to avoid cumulative operation of the group list
group_list_type=1,
group_type=0,
group_list=group_list,
)[0]
gate_up_out = torch_npu.npu_swiglu(gate_up_out_list)
w2 = w2.transpose(1, 2)
down_out_list = torch_npu.npu_grouped_matmul(
x=[gate_up_out],
weight=[w2],
split_item=2,
group_list_type=1,
group_type=0,
group_list=group_list,
)[0]
# moeCombine
kwargs_mc2 = {
"expand_x": down_out_list,
"expert_ids": topk_ids,
"expert_scales": topk_weights.to(torch.float32),
"expert_shard_type": 0,
"shared_expert_rank_num": 0,
"moe_expert_num": moe_expert_num,
"global_bs": 0,
}
tp_recv_counts = output[5]
stage3_kwargs = {
"ep_send_counts": ep_recv_counts,
"group_ep": moe_all_to_all_group_name,
"ep_world_size": ep_world_size,
"ep_rank_id": ep_rank_id,
}
if enable_dispatch_v2:
stage3_kwargs.update({
"assist_info_for_combine":
assist_info_for_combine,
})
else:
stage3_kwargs.update({
"expand_idx": assist_info_for_combine,
})
if need_extra_args:
stage3_kwargs.update({
"tp_send_counts": tp_recv_counts,
"group_tp": moe_all_to_all_group_name,
"tp_world_size": 1,
"tp_rank_id": 0,
})
if a3_need_extra_args and enable_dispatch_v2:
stage3_kwargs.update({
"x_active_mask": mc2_mask,
})
kwargs_mc2.update(stage3_kwargs)
hidden_states = torch_npu.npu_moe_distribute_combine_v2(
**kwargs_mc2
) if enable_dispatch_v2 else torch_npu.npu_moe_distribute_combine(
**kwargs_mc2)
if shared_experts is None:
return hidden_states
else:
shared_hidden_states, _ = shared_experts.down_proj(shared_act)
return hidden_states, shared_hidden_states
def apply_mlp(
hidden_states: torch.Tensor,
w1: torch.Tensor,
@@ -318,248 +174,6 @@ def apply_mlp(
return hidden_states
# currently expert parallelism implemented with all2all
# is under-optimized.
def fused_experts_with_all2all(
hidden_states: torch.Tensor,
w1: torch.Tensor,
w2: torch.Tensor,
topk_weights: torch.Tensor,
topk_ids: torch.Tensor,
row_idx: torch.Tensor,
top_k: int,
expert_map: torch.Tensor = None,
ep_group: GroupCoordinator = None,
):
original_shape = hidden_states.shape
if len(original_shape) == 3:
hidden_states = hidden_states.view(-1, hidden_states.shape[-1])
num_tokens, _ = hidden_states.shape
num_experts = w1.shape[0]
if expert_map is not None:
global_num_experts = len(expert_map)
local_num_experts = global_num_experts // ep_group.world_size
hidden_states, expanded_row_idx, expanded_expert_idx = torch_npu.npu_moe_init_routing(
hidden_states,
row_idx=row_idx,
expert_idx=topk_ids,
active_num=num_tokens)
global_expert_tokens = torch.bincount(expanded_expert_idx,
minlength=global_num_experts)
scatter_sizes = global_expert_tokens.view(ep_group.world_size,
-1).sum(-1)
gather_sizes = torch.empty_like(scatter_sizes)
dist.all_to_all_single(gather_sizes,
scatter_sizes,
group=ep_group.device_group)
scatter_size_list = scatter_sizes.cpu().tolist()
gather_size_list = gather_sizes.cpu().tolist()
expanded_expert_idx = expanded_expert_idx % local_num_experts
hidden_states = ep_group.all_to_all(hidden_states, 0, 0,
scatter_size_list,
gather_size_list)
local_expert_idx = ep_group.all_to_all(expanded_expert_idx, 0, 0,
scatter_size_list,
gather_size_list)
sorted_local_expert_idx, sorted_idx = torch.sort(local_expert_idx)
expert_tokens = torch_npu.npu_moe_compute_expert_tokens(
sorted_local_expert_idx, local_num_experts).to(torch.int64)
hidden_states = hidden_states[sorted_idx]
else:
hidden_states, expanded_row_idx, expanded_expert_idx = torch_npu.npu_moe_init_routing(
hidden_states,
row_idx=row_idx,
expert_idx=topk_ids,
active_num=num_tokens)
expert_tokens = torch_npu.npu_moe_compute_expert_tokens(
expanded_expert_idx, num_experts)
expert_tokens = expert_tokens.to(torch.int64)
w1 = w1.transpose(1, 2)
gate_up_out_list = torch_npu.npu_grouped_matmul(
x=[hidden_states],
weight=[w1],
split_item=2,
group_list_type=0,
group_type=0,
group_list=expert_tokens,
)[0]
hidden_states = torch_npu.npu_swiglu(gate_up_out_list)
w2 = w2.transpose(1, 2)
hidden_states = torch_npu.npu_grouped_matmul(
x=[hidden_states],
weight=[w2],
split_item=2,
group_list_type=0,
group_type=0,
group_list=expert_tokens,
)[0]
if expert_map is not None:
resorted_idx = torch.argsort(sorted_idx)
hidden_states = hidden_states[resorted_idx]
hidden_states = ep_group.all_to_all(hidden_states, 0, 0,
gather_size_list,
scatter_size_list)
final_hidden_states = torch_npu.npu_moe_finalize_routing(
hidden_states,
skip1=None,
skip2=None,
bias=None,
scales=topk_weights,
expanded_src_to_dst_row=expanded_row_idx,
export_for_source_row=topk_ids,
)
else:
# TODO: Reorder device memory 2 times here, replace the current
# implementation here when suitable operators become available.
final_hidden_states = torch_npu.npu_moe_finalize_routing(
hidden_states,
skip1=None,
skip2=None,
bias=None,
scales=topk_weights,
expanded_src_to_dst_row=expanded_row_idx,
export_for_source_row=topk_ids,
)
if len(original_shape) == 3:
final_hidden_states = final_hidden_states.view(original_shape)
return final_hidden_states
# currently expert parallelism implemented with all2all
# is under-optimized.
def fused_experts_with_all2all_buffer(
hidden_states: torch.Tensor,
w1: torch.Tensor,
w2: torch.Tensor,
topk_weights: torch.Tensor,
topk_ids: torch.Tensor,
row_idx: torch.Tensor,
top_k: int,
max_model_len: int,
global_batch_size: int,
expert_map: torch.Tensor = None,
ep_group: GroupCoordinator = None,
):
original_shape = hidden_states.shape
if len(original_shape) == 3:
hidden_states = hidden_states.view(-1, hidden_states.shape[-1])
num_tokens, _ = hidden_states.shape
global_num_experts = len(expert_map)
local_num_experts = global_num_experts // ep_group.world_size
row_idx_len = num_tokens * top_k
hidden_states, expanded_row_idx, expanded_expert_idx = torch_npu.npu_moe_init_routing(
hidden_states,
row_idx=row_idx,
expert_idx=topk_ids,
active_num=num_tokens)
max_row_per_ep_rank = (-(-global_batch_size // ep_group.world_size) *
max_model_len // ep_group.world_size +
1) * top_k * 2
expert_idx_buffer_scatter, unpad_indices = process_topk_ids(
expanded_expert_idx, global_num_experts, ep_group.world_size,
max_row_per_ep_rank, num_tokens, top_k)
hidden_states_pad_idx = torch.zeros(
expert_idx_buffer_scatter.shape,
dtype=expert_idx_buffer_scatter.dtype,
device=expert_idx_buffer_scatter.device)
non_pad_len = torch.sum((expert_idx_buffer_scatter
!= global_num_experts).to(torch.int32))
hidden_states_pad_idx[expert_idx_buffer_scatter !=
global_num_experts] = torch.arange(
non_pad_len,
dtype=expert_idx_buffer_scatter.dtype,
device=hidden_states.device)
hidden_states_buffer_scatter = hidden_states[hidden_states_pad_idx]
expert_idx_buffer_gather = torch.empty_like(
expert_idx_buffer_scatter,
dtype=expert_idx_buffer_scatter.dtype,
device=expert_idx_buffer_scatter.device)
hidden_states_buffer_gather = torch.empty_like(
hidden_states_buffer_scatter,
dtype=hidden_states_buffer_scatter.dtype,
device=hidden_states_buffer_scatter.device)
dist.all_to_all_single(expert_idx_buffer_gather,
expert_idx_buffer_scatter,
group=ep_group.device_group)
dist.all_to_all_single(hidden_states_buffer_gather,
hidden_states_buffer_scatter,
group=ep_group.device_group)
mask = expert_idx_buffer_gather != global_num_experts
local_expert_idx = expert_idx_buffer_gather[mask] - ep_group.rank * (
global_num_experts // ep_group.world_size)
hidden_states = hidden_states_buffer_gather[mask]
idx_type = local_expert_idx.dtype
sorted_local_expert_idx, sorted_idx = torch.sort(local_expert_idx.float())
sorted_local_expert_idx = sorted_local_expert_idx.to(idx_type)
expert_tokens = torch_npu.npu_moe_compute_expert_tokens(
sorted_local_expert_idx, local_num_experts).to(torch.int64)
hidden_states = hidden_states[sorted_idx]
group_list_type = 0
hidden_states = apply_mlp(hidden_states,
w1,
w2,
expert_tokens,
group_list_type=group_list_type)
resorted_idx = torch.argsort(sorted_idx.float()).to(sorted_idx.dtype)
hidden_states = hidden_states[resorted_idx]
hidden_states_scatter = torch.zeros(
(mask.shape[0], hidden_states.shape[1]),
dtype=hidden_states.dtype,
device=hidden_states.device)
hidden_states_scatter[mask] = hidden_states
hidden_states_gatter = torch.empty_like(
hidden_states_scatter,
dtype=hidden_states_scatter.dtype,
device=hidden_states_scatter.device)
dist.all_to_all_single(hidden_states_gatter,
hidden_states_scatter,
group=ep_group.device_group)
hidden_states_gatter = hidden_states_gatter[expert_idx_buffer_scatter !=
global_num_experts]
if hidden_states_gatter.shape[0] != row_idx_len:
hidden_states = torch.zeros((row_idx_len, hidden_states.shape[1]),
dtype=hidden_states.dtype,
device=hidden_states.device)
hidden_states[unpad_indices != -1] = hidden_states_gatter
else:
# TODO: Reorder device memory 2 times here, replace the current
hidden_states = hidden_states_gatter
final_hidden_states = torch_npu.npu_moe_finalize_routing(
hidden_states,
skip1=None,
skip2=None,
bias=None,
scales=topk_weights,
expanded_src_to_dst_row=expanded_row_idx,
export_for_source_row=topk_ids,
)
if len(original_shape) == 3:
final_hidden_states = final_hidden_states.view(original_shape)
return final_hidden_states
def fused_experts_moge(
hidden_states: torch.Tensor,
w1: torch.Tensor,
@@ -651,188 +265,228 @@ def fused_experts_moge(
return final_hidden_states
def fused_experts_with_all2allv(
token_dispatcher,
probs,
routing_map,
hidden_states: torch.Tensor,
w1: torch.Tensor,
w2: torch.Tensor,
):
# Enable moe alltoallv, it's a balanced policy for precision and efficiency.
(share_experts_output, dispatched_input,
tokens_per_expert) = (token_dispatcher.token_permutation(
hidden_states, probs, routing_map))
expert_output = apply_mlp(dispatched_input, w1, w2, tokens_per_expert)
output, mlp_bias = token_dispatcher.token_unpermutation(expert_output)
return output
def fused_experts(
hidden_states: torch.Tensor,
w1: torch.Tensor,
w2: torch.Tensor,
topk_weights: torch.Tensor,
topk_ids: torch.Tensor,
row_idx: torch.Tensor,
top_k: int,
expert_map: torch.Tensor = None,
apply_router_weight_on_input: bool = False,
max_num_tokens: Optional[int] = None,
) -> torch.Tensor:
"""
Fused experts with top-k routing.
Args:
hidden_states: Hidden states of shape (num_tokens, hidden_size).
w1: Expert weights1 of shape (num_experts, intermediate_size * 2, hidden_size).
w2: Expert weights2 of shape (num_experts, hidden_size, intermediate_size).
topk_weights: Routing weights of shape (num_tokens, top_k).
topk_ids: Selected expert IDs of shape (num_tokens, top_k).
top_k: Number of experts to select.
expert_map: Expert mapping of shape (num_experts,).
Returns:
hidden_states: Hidden states after routing.
"""
"""
# Check constraints.
assert hidden_states.shape[1] == w1.shape[2], "Hidden size mismatch"
assert topk_weights.shape == topk_ids.shape, "topk shape mismatch"
assert hidden_states.is_contiguous(), "Hidden_states must be contiguous"
assert w1.is_contiguous(), "Expert weights1 must be contiguous"
assert w2.is_contiguous(), "Expert weights2 must be contiguous"
"""
# if torch.distributed.get_rank() == 0:
# print(w1.shape)
# print(hidden_states.shape)
original_shape = hidden_states.shape
# assert len(original_shape) == 2
num_tokens = hidden_states.shape[:-1].numel()
num_experts = w1.shape[0]
dtype = hidden_states.dtype
device = hidden_states.device
# assert dtype in [torch.float32, torch.float16, torch.bfloat16
# ], "Only float32, float16, and bfloat16 are supported"
if apply_router_weight_on_input:
assert (topk_weights.dim() == 2
), "`topk_weights` should be in shape (num_tokens, topk)"
_, topk = topk_weights.shape
assert (
topk == 1
), "Only support topk=1 when `apply_router_weight_on_input` is True"
hidden_states = hidden_states * topk_weights.to(hidden_states.dtype)
if expert_map is not None:
# Generate token indices and flatten
token_indices = (torch.arange(num_tokens,
device=device,
dtype=torch.int64).unsqueeze(1).expand(
-1, top_k).reshape(-1))
# Flatten token-to-expert mappings and map to local experts
weights_flat = topk_weights.view(-1)
experts_flat = topk_ids.view(-1)
local_experts_flat = expert_map[experts_flat]
# Filter valid token-expert pairs
mask = local_experts_flat != -1
filtered_weights = torch.where(
mask, weights_flat, torch.zeros_like(weights_flat)).to(dtype)
filtered_experts = torch.where(
mask, local_experts_flat,
torch.full_like(local_experts_flat,
num_experts)).to(topk_ids.dtype)
# Sort by local expert IDs
sort_indices = torch.argsort(filtered_experts.view(torch.float32))
sorted_token_indices = token_indices[sort_indices]
sorted_weights = filtered_weights[sort_indices]
# Compute token counts with minlength of num_experts
# This is equivalent to but faster than:
# >>> token_counts = torch.bincount(filtered_experts, minlength=num_experts)[:-1]
token_counts = torch.zeros(num_experts + 1,
device=device,
dtype=torch.int64)
ones = torch.ones_like(filtered_experts, dtype=torch.int64)
token_counts.scatter_add_(0, filtered_experts.to(torch.int64), ones)
token_counts = token_counts[:num_experts]
expert_tokens = torch.cumsum(token_counts, dim=0, dtype=torch.int64)
# Rearrange hidden_states
sorted_hidden_states = hidden_states[sorted_token_indices]
def quant_apply_mlp(hidden_states: torch.Tensor,
w1: torch.Tensor,
w1_scale: torch.Tensor,
w2: torch.Tensor,
w2_scale: torch.Tensor,
group_list: torch.Tensor,
dynamic_scale: torch.Tensor = None,
group_list_type: int = 1,
w1_scale_bias: torch.Tensor = None,
w2_scale_bias: torch.Tensor = None) -> torch.Tensor:
if dynamic_scale is None:
unquantized_hidden_states = hidden_states
hidden_states, pertoken_scale = torch_npu.npu_dynamic_quant(
hidden_states)
# Dispose the original unquantized hidden states
# to save npu memory because they're no longer used.
dispose_tensor(unquantized_hidden_states)
else:
active_num = max_num_tokens if max_num_tokens is not None else num_tokens
sorted_hidden_states, expanded_row_idx, expanded_expert_idx = torch_npu.npu_moe_init_routing(
hidden_states,
row_idx=row_idx,
expert_idx=topk_ids,
active_num=active_num)
pertoken_scale = dynamic_scale
expert_tokens = torch_npu.npu_moe_compute_expert_tokens(
expanded_expert_idx, num_experts)
expert_tokens = expert_tokens.to(torch.int64)
bias1, bias2 = None, None
_output_dtype = w2_scale.dtype
is_mc2 = get_forward_context().fused_moe_state == FusedMoEState.MC2
if w1_scale_bias is None and is_mc2:
w1_scale = w1_scale.to(torch.float32)
# gmm1: gate_up_proj
hidden_states = torch_npu.npu_grouped_matmul(
x=[hidden_states],
weight=[w1],
split_item=3,
group_list_type=group_list_type,
group_type=0,
group_list=group_list,
output_dtype=torch.int32)[0]
# act_fn: swiglu
hidden_states, swiglu_out_scale = torch_npu.npu_dequant_swiglu_quant(
x=hidden_states,
weight_scale=w1_scale,
activation_scale=pertoken_scale,
bias=None,
quant_scale=None,
quant_offset=None,
group_index=group_list,
activate_left=True,
quant_mode=1,
)
# gmm2: down_proj
hidden_states = torch_npu.npu_grouped_matmul(
x=[hidden_states],
weight=[w2],
scale=[w2_scale],
per_token_scale=[swiglu_out_scale],
split_item=2,
group_list_type=group_list_type,
group_type=0,
group_list=group_list,
output_dtype=w2_scale.dtype)[0]
else:
if w1_scale_bias is not None:
if group_list_type == 0:
group_list = torch.cat(
[group_list[:1],
torch.diff(group_list, dim=0)])
group_list_type = 1
bias1 = [w1_scale_bias]
bias2 = [w2_scale_bias]
# TODO w4a8 scene: dynamic acquisition of dtype in the future
_output_dtype = torch.bfloat16
# gmm1: gate_up_proj
hidden_states = torch_npu.npu_grouped_matmul(
x=[hidden_states],
weight=[w1],
scale=[w1_scale],
bias=bias1,
per_token_scale=[pertoken_scale],
split_item=2,
group_list_type=group_list_type,
group_type=0,
group_list=group_list,
output_dtype=_output_dtype)[0]
# act_fn: swiglu
hidden_states = torch_npu.npu_swiglu(hidden_states)
hidden_states, swiglu_out_scale = torch_npu.npu_dynamic_quant(
hidden_states)
# gmm2: down_proj
hidden_states = torch_npu.npu_grouped_matmul(
x=[hidden_states],
weight=[w2],
scale=[w2_scale],
bias=bias2,
per_token_scale=[swiglu_out_scale],
split_item=2,
group_list_type=group_list_type,
group_type=0,
group_list=group_list,
output_dtype=_output_dtype)[0]
return hidden_states
def unquant_apply_mlp(
hidden_states: torch.Tensor,
w1: torch.Tensor,
w2: torch.Tensor,
group_list: torch.Tensor,
group_list_type: int = 1,
topk_scales: Optional[torch.Tensor] = None) -> torch.Tensor:
w1 = w1.transpose(1, 2)
gate_up_out_list = torch_npu.npu_grouped_matmul(
x=[sorted_hidden_states],
gate_up_out = torch_npu.npu_grouped_matmul(
x=[hidden_states],
weight=[w1],
split_item=2,
group_list_type=0,
group_list_type=group_list_type,
group_type=0,
group_list=expert_tokens,
group_list=group_list,
)[0]
if is_310p():
gate_up_out = torch_npu.npu_swiglu(gate_up_out.to(torch.float32)).to(
torch.float16)
else:
gate_up_out = torch_npu.npu_swiglu(gate_up_out)
gate_up_out = torch_npu.npu_swiglu(gate_up_out_list)
if topk_scales is not None:
gate_up_out *= topk_scales
w2 = w2.transpose(1, 2)
down_out_list = torch_npu.npu_grouped_matmul(
hidden_states = torch_npu.npu_grouped_matmul(
x=[gate_up_out],
weight=[w2],
split_item=2,
group_list_type=0,
group_list_type=group_list_type,
group_type=0,
group_list=expert_tokens,
group_list=group_list,
)[0]
return hidden_states
if expert_map is not None:
weighted_down_out = down_out_list * sorted_weights.unsqueeze(1)
final_hidden_states = torch.zeros(*original_shape,
device=hidden_states.device,
dtype=dtype)
# TODO: npu_grouped_matmul output random values at [num_valid_tokens:, ...]
# This created multiple NaN and index_add_ will mix them up which harms accuracy
# remove this mask and filter after it being fixed
num_valid_tokens = mask.sum()
valid_token_mask = torch.arange(
0, sorted_token_indices.shape[0],
device=device).unsqueeze(1) < num_valid_tokens
valid_output = torch.where(
valid_token_mask, weighted_down_out,
torch.zeros_like(weighted_down_out)).to(dtype)
final_hidden_states.index_add_(0, sorted_token_indices, valid_output)
def unified_apply_mlp(
hidden_states: torch.Tensor,
w1: torch.Tensor,
w1_scale: torch.Tensor,
w2: torch.Tensor,
w2_scale: torch.Tensor,
group_list: torch.Tensor,
dynamic_scale: torch.Tensor = None,
group_list_type: int = 1,
w1_scale_bias: torch.Tensor = None,
w2_scale_bias: torch.Tensor = None,
topk_scales: Optional[torch.Tensor] = None) -> torch.Tensor:
if get_forward_context().with_quant:
return quant_apply_mlp(hidden_states=hidden_states,
w1=w1,
w1_scale=w1_scale,
w2=w2,
w2_scale=w2_scale,
group_list=group_list,
dynamic_scale=dynamic_scale,
group_list_type=group_list_type,
w1_scale_bias=w1_scale_bias,
w2_scale_bias=w2_scale_bias)
else:
scales = torch.ones_like(
topk_weights) if apply_router_weight_on_input else topk_weights
# TODO: Reorder device memory 2 times here, replace the current
# implementation here when suitable operators become available.
final_hidden_states = torch_npu.npu_moe_finalize_routing(
down_out_list,
skip1=None,
skip2=None,
bias=None,
scales=scales,
expanded_src_to_dst_row=expanded_row_idx,
export_for_source_row=topk_ids,
)
return unquant_apply_mlp(hidden_states=hidden_states,
w1=w1,
w2=w2,
group_list=group_list,
group_list_type=group_list_type,
topk_scales=topk_scales)
def unified_fused_experts_eager(hidden_states: torch.Tensor,
w1: torch.Tensor,
w2: torch.Tensor,
topk_weights: torch.Tensor,
topk_ids: torch.Tensor,
row_idx: torch.Tensor,
expert_map: Optional[torch.Tensor] = None,
log2phy: Optional[torch.Tensor] = None,
global_redundant_expert_num: int = 0,
w1_scale: Optional[torch.Tensor] = None,
w1_scale_bias: Optional[torch.Tensor] = None,
w2_scale: Optional[torch.Tensor] = None,
w2_scale_bias: Optional[torch.Tensor] = None,
shared_experts: Optional[torch.Tensor] = None,
shared_gate_up: Optional[Any] = None,
shared_dequant_scale: Optional[Any] = None,
mc2_mask: Optional[torch.Tensor] = None,
apply_router_weight_on_input: bool = False):
token_dispatcher = get_forward_context().token_dispatcher
results = token_dispatcher.token_dispatch(
hidden_states=hidden_states,
topk_weights=topk_weights,
topk_ids=topk_ids,
row_idx=row_idx,
expert_map=expert_map,
log2phy=log2phy,
global_redundant_expert_num=global_redundant_expert_num,
shared_experts=shared_experts,
shared_gate_up=shared_gate_up,
shared_dequant_scale=shared_dequant_scale,
mc2_mask=mc2_mask,
apply_router_weight_on_input=apply_router_weight_on_input)
expert_output = unified_apply_mlp(
hidden_states=results["hidden_states"],
w1=w1,
w1_scale=w1_scale,
w2=w2,
w2_scale=w2_scale,
group_list=results["group_list"],
dynamic_scale=results.get("dynamic_scale"),
group_list_type=results.get("group_list_type"),
w1_scale_bias=w1_scale_bias,
w2_scale_bias=w2_scale_bias,
topk_scales=results.get("topk_scales"))
final_hidden_states = token_dispatcher.token_combine(expert_output)
return final_hidden_states
@@ -914,65 +568,16 @@ class AscendUnquantizedFusedMoEMethod(UnquantizedFusedMoEMethod):
if enable_force_load_balance and not self.use_aclgraph:
topk_ids = torch.randint_like(topk_ids, 0, global_num_experts)
fused_moe_state = get_forward_context().fused_moe_state
if fused_moe_state == FusedMoEState.MC2:
return fused_experts_with_mc2(
hidden_states=x,
w1=layer.w13_weight,
w2=layer.w2_weight,
moe_parallel_config=self.moe.moe_parallel_config,
topk_weights=topk_weights,
topk_ids=topk_ids,
top_k=top_k,
expert_map=expert_map,
moe_all_to_all_group_name=self.moe_all_to_all_group_name,
shared_experts=shared_experts,
mc2_mask=kwargs.get("mc2_mask", None))
elif fused_moe_state in [
FusedMoEState.AllGather, FusedMoEState.NaiveMulticast
]:
return fused_experts(hidden_states=x,
w1=layer.w13_weight,
w2=layer.w2_weight,
topk_weights=topk_weights,
topk_ids=topk_ids,
row_idx=row_idx,
top_k=top_k,
expert_map=expert_map)
elif MOE_ALL2ALL_BUFFER:
return fused_experts_with_all2all_buffer(
hidden_states=x,
w1=layer.w13_weight,
w2=layer.w2_weight,
topk_weights=topk_weights,
topk_ids=topk_ids,
row_idx=row_idx,
top_k=top_k,
max_model_len=self.max_model_len,
global_batch_size=self.global_batch_size,
expert_map=expert_map,
ep_group=get_ep_group())
elif fused_moe_state == FusedMoEState.All2AllSeq:
token_dispatcher = kwargs.get("token_dispatcher")
return fused_experts_with_all2allv(
token_dispatcher=token_dispatcher,
probs=topk_weights,
routing_map=topk_ids,
hidden_states=x,
w1=layer.w13_weight,
w2=layer.w2_weight,
)
else:
return fused_experts_with_all2all(hidden_states=x,
w1=layer.w13_weight,
w2=layer.w2_weight,
topk_weights=topk_weights,
topk_ids=topk_ids,
row_idx=row_idx,
top_k=top_k,
expert_map=expert_map,
ep_group=get_ep_group())
return unified_fused_experts_eager(hidden_states=x,
w1=layer.w13_weight,
w2=layer.w2_weight,
topk_weights=topk_weights,
topk_ids=topk_ids,
row_idx=row_idx,
expert_map=expert_map,
shared_experts=shared_experts,
mc2_mask=kwargs.get(
"mc2_mask", None))
class AscendFusedMoE(FusedMoE):
@@ -1154,6 +759,19 @@ class AscendFusedMoE(FusedMoE):
self.token_dispatcher, token_dispatcher1
]
ep_size = (get_ep_group().world_size if
vllm_config.parallel_config.enable_expert_parallel else 1)
with_quant = quant_config is not None
from vllm_ascend.ops.moe_dispatcher.token_dispatcher import \
setup_token_dispatchers
setup_token_dispatchers(
ep_size,
top_k=self.top_k,
num_experts=self.global_num_experts,
num_global_redundant_experts=self.global_redundant_expert_num,
num_local_experts=self.local_num_experts,
with_quant=with_quant)
def naive_multicast(self, x: torch.Tensor,
cu_tokens_across_dp_cpu: torch.Tensor):
assert (len(x.shape) == 2)

288
vllm_ascend/ops/moe_dispatcher/token_dispatcher.py
View File

@@ -22,21 +22,18 @@
# limitations under the License.
from abc import ABC, abstractmethod
from typing import Optional
from typing import Any, Dict, Optional
import torch
import torch_npu
from vllm.distributed.parallel_state import get_ep_group
from vllm.forward_context import get_forward_context
from vllm_ascend.ascend_config import get_ascend_config
from vllm_ascend.distributed.parallel_state import get_mc2_group
from vllm_ascend.distributed.tensor_parallel import (
all_gather_last_dim_from_tensor_parallel_region, all_to_all_hp2sp,
all_to_all_sp2hp, gather_from_sequence_parallel_region,
reduce_scatter_last_dim_to_tensor_parallel_region)
from vllm_ascend.ops.comm_utils import async_all_to_all
from vllm_ascend.torchair.utils import npu_stream_switch, npu_wait_tensor
from vllm_ascend.utils import AscendSocVersion, get_ascend_soc_version
@@ -460,6 +457,31 @@ class MoEAlltoAllSeqOverLapDispatcher(MoEDispatcher):
return output, None
_Dispatchers: Dict[str, Any] = {}
def _register_token_dispatcher(dispatcher: Any):
_Dispatchers[dispatcher.__class__.__name__] = dispatcher
def get_token_dispatcher(name: str):
return _Dispatchers.get(name)
def setup_token_dispatchers(ep_size: int, **kwargs):
existing_dispatchers = set(_Dispatchers.keys())
if ep_size == 1 and "TokenDispatcherWithAllGather" not in existing_dispatchers:
_register_token_dispatcher(TokenDispatcherWithAllGather(**kwargs))
elif ep_size < 16 and "TokenDispatcherWithAll2AllV" not in existing_dispatchers:
_register_token_dispatcher(TokenDispatcherWithAll2AllV(**kwargs))
elif ep_size >= 16:
if "TokenDispatcherWithAll2AllV" not in existing_dispatchers:
_register_token_dispatcher(TokenDispatcherWithAll2AllV(**kwargs))
if "TokenDispatcherWithMC2" not in existing_dispatchers:
_register_token_dispatcher(TokenDispatcherWithMC2(**kwargs))
class MoETokenDispatcher(ABC):
def __init__(self, **kwargs) -> None:
@@ -484,18 +506,19 @@ class MoETokenDispatcher(ABC):
return get_ep_group().world_size
@abstractmethod
def token_dispatch(
self,
hidden_states: torch.Tensor,
topk_weights: torch.Tensor,
topk_ids: torch.Tensor,
expert_map: torch.Tensor,
log2phy: Optional[torch.Tensor] = None,
global_redundant_expert_num: int = 0,
shared_gate_up: Optional[torch.Tensor] = None,
shared_dequant_scale: Optional[torch.Tensor] = None,
shared_experts: Optional[torch.Tensor] = None,
):
def token_dispatch(self,
hidden_states: torch.Tensor,
topk_weights: torch.Tensor,
topk_ids: torch.Tensor,
row_idx: torch.Tensor,
expert_map: Optional[torch.Tensor] = None,
log2phy: Optional[torch.Tensor] = None,
global_redundant_expert_num: int = 0,
shared_experts: Optional[torch.Tensor] = None,
shared_gate_up: Optional[torch.Tensor] = None,
shared_dequant_scale: Optional[torch.Tensor] = None,
mc2_mask: Optional[torch.Tensor] = None,
apply_router_weight_on_input: bool = False):
raise NotImplementedError("Dispatch function not implemented.")
@abstractmethod
@@ -516,40 +539,39 @@ class TokenDispatcherWithMC2(MoETokenDispatcher):
self.moe_all_to_all_group_name = backend.get_hccl_comm_name(local_rank)
self.ep_rank_id = get_mc2_group().rank_in_group
self.ep_world_size = get_mc2_group().world_size
ascend_config = get_ascend_config()
self.torchair_graph_enabled = ascend_config.torchair_graph_config.enabled
self.enable_dispatch_v2 = hasattr(torch_npu,
"npu_moe_distribute_dispatch_v2")
self.need_extra_args = (get_ascend_soc_version() == AscendSocVersion.A3
or self.torchair_graph_enabled)
self.need_extra_args = (
get_ascend_soc_version() == AscendSocVersion.A3)
# NOTE: Currently, when in A3, we need to pass in some extra param into dispatch & combine
self.a3_need_extra_args = \
get_ascend_soc_version() == AscendSocVersion.A3
self.output = None
self.dynamic_scale = None
self.assist_info_for_combine = None
self.ep_recv_counts = None
self.shared_act = None
self.topk_ids = None
self.topk_weights = None
self.shared_experts = None
self.mc2_mask = None
def get_dispatch_mc2_kwargs(self,
hidden_states: torch.Tensor,
topk_weights: torch.Tensor,
topk_ids: torch.Tensor,
expert_map: torch.Tensor,
global_redundant_expert_num: int = 0):
quant_mode = 0
forward_context = get_forward_context()
mc2_mask = forward_context.mc2_mask
def get_dispatch_mc2_kwargs(
self,
hidden_states: torch.Tensor,
topk_weights: torch.Tensor,
topk_ids: torch.Tensor,
expert_map: torch.Tensor,
global_redundant_expert_num: int = 0,
):
if self.with_quant:
quant_mode = 2
if (expert_map is not None):
moe_expert_num = len(expert_map) + global_redundant_expert_num
else:
moe_expert_num = global_redundant_expert_num
else:
quant_mode = 0
moe_expert_num = len(expert_map)
kwargs_mc2 = {
"x": hidden_states,
@@ -575,28 +597,30 @@ class TokenDispatcherWithMC2(MoETokenDispatcher):
})
if self.a3_need_extra_args and self.enable_dispatch_v2:
stage1_kwargs.update({
"x_active_mask": mc2_mask,
"x_active_mask": self.mc2_mask,
})
kwargs_mc2.update(stage1_kwargs)
return kwargs_mc2
def token_dispatch(
self,
hidden_states: torch.Tensor,
topk_weights: torch.Tensor,
topk_ids: torch.Tensor,
expert_map: torch.Tensor,
log2phy: Optional[torch.Tensor] = None,
global_redundant_expert_num: int = 0,
shared_gate_up: Optional[torch.Tensor] = None,
shared_dequant_scale: Optional[torch.Tensor] = None,
shared_experts: Optional[torch.Tensor] = None,
):
def token_dispatch(self,
hidden_states: torch.Tensor,
topk_weights: torch.Tensor,
topk_ids: torch.Tensor,
row_idx: torch.Tensor,
expert_map: Optional[torch.Tensor] = None,
log2phy: Optional[torch.Tensor] = None,
global_redundant_expert_num: int = 0,
shared_experts: Optional[torch.Tensor] = None,
shared_gate_up: Optional[torch.Tensor] = None,
shared_dequant_scale: Optional[torch.Tensor] = None,
mc2_mask: Optional[torch.Tensor] = None,
apply_router_weight_on_input: bool = False):
self.expert_map = expert_map
self.topk_ids = topk_ids
self.topk_weights = topk_weights
self.shared_experts = shared_experts
self.mc2_mask = mc2_mask
kwargs_mc2 = self.get_dispatch_mc2_kwargs(hidden_states, topk_weights,
topk_ids, expert_map,
@@ -606,28 +630,27 @@ class TokenDispatcherWithMC2(MoETokenDispatcher):
) if self.enable_dispatch_v2 else torch_npu.npu_moe_distribute_dispatch(
**kwargs_mc2)
# comm_stream.wait_stream(torch.npu.current_stream())
expand_x, self.dynamic_scale, self.assist_info_for_combine, \
expand_x, dynamic_scale, self.assist_info_for_combine, \
expert_token_nums, self.ep_recv_counts = self.output[0:5]
if self.with_quant:
if shared_experts is not None:
with npu_stream_switch("moe_secondary", 0):
npu_wait_tensor(shared_gate_up, expand_x)
shared_act_out = shared_experts.act_fn(
(shared_gate_up, shared_dequant_scale))
self.shared_act, self.swiglu_out_scale = \
shared_act_out[0], shared_act_out[1]
shared_act_out = shared_experts.act_fn(
(shared_gate_up, shared_dequant_scale))
self.shared_act, self.swiglu_out_scale = \
shared_act_out[0], shared_act_out[1]
else:
if shared_experts is not None:
with npu_stream_switch("moe_secondary", 0):
npu_wait_tensor(hidden_states, topk_weights)
shared_gate_up, _ = shared_experts.gate_up_proj(
hidden_states)
npu_wait_tensor(shared_gate_up, expand_x)
self.shared_act = shared_experts.act_fn(shared_gate_up)
shared_gate_up, _ = shared_experts.gate_up_proj(hidden_states)
self.shared_act = shared_experts.act_fn(shared_gate_up)
group_list_type = 1
return group_list_type, expand_x, expert_token_nums
return {
"group_list_type": group_list_type,
"hidden_states": expand_x,
"group_list": expert_token_nums,
"dynamic_scale": dynamic_scale,
}
def get_combine_mc_kwargs(self, hidden_states: torch.Tensor):
assert self.expert_map is not None
@@ -635,8 +658,6 @@ class TokenDispatcherWithMC2(MoETokenDispatcher):
assert self.topk_ids is not None
assert self.output is not None
moe_expert_num = len(self.expert_map)
forward_context = get_forward_context()
mc2_mask = forward_context.mc2_mask
# moeCombine
kwargs_mc2 = {
"expand_x": hidden_states,
@@ -677,7 +698,7 @@ class TokenDispatcherWithMC2(MoETokenDispatcher):
})
if self.a3_need_extra_args and self.enable_dispatch_v2:
stage3_kwargs.update({
"x_active_mask": mc2_mask,
"x_active_mask": self.mc2_mask,
})
kwargs_mc2.update(stage3_kwargs)
return kwargs_mc2
@@ -685,7 +706,6 @@ class TokenDispatcherWithMC2(MoETokenDispatcher):
def token_combine(self,
hidden_states: torch.Tensor,
bias: torch.Tensor = None):
kwargs_mc2 = self.get_combine_mc_kwargs(hidden_states)
hidden_states = torch_npu.npu_moe_distribute_combine_v2(
**kwargs_mc2
@@ -695,15 +715,11 @@ class TokenDispatcherWithMC2(MoETokenDispatcher):
return hidden_states
else:
if self.with_quant:
with npu_stream_switch("moe_secondary", 0):
npu_wait_tensor(self.shared_act, hidden_states)
shared_hidden_states, _ = self.shared_experts.down_proj(
(self.shared_act, self.swiglu_out_scale))
shared_hidden_states, _ = self.shared_experts.down_proj(
(self.shared_act, self.swiglu_out_scale))
else:
with npu_stream_switch("moe_secondary", 0):
npu_wait_tensor(self.shared_act, hidden_states)
shared_hidden_states, _ = self.shared_experts.down_proj(
self.shared_act)
shared_hidden_states, _ = self.shared_experts.down_proj(
self.shared_act)
return hidden_states, shared_hidden_states
@@ -711,13 +727,9 @@ class TokenDispatcherWithAllGather(MoETokenDispatcher):
def __init__(self, **kwargs):
super().__init__(**kwargs)
self.apply_router_weight_on_input = kwargs.get(
"apply_router_weight_on_input")
self.top_k = kwargs.get("top_k")
self.apply_router_weight_on_input = False
self.max_num_tokens = kwargs.get("max_num_tokens")
ep_size = kwargs.get("ep_size")
if ep_size is not None:
self.num_experts_local = self.num_experts // ep_size
self.num_experts_local = kwargs.get("num_local_experts", 0)
self.sorted_weights = None
self.expanded_row_idx = None
self.sorted_token_indices = None
@@ -727,20 +739,20 @@ class TokenDispatcherWithAllGather(MoETokenDispatcher):
self.topk_weights = None
self.topk_ids = None
def token_dispatch(
self,
hidden_states: torch.Tensor,
topk_weights: torch.Tensor,
topk_ids: torch.Tensor,
expert_map: torch.Tensor,
log2phy: Optional[torch.Tensor] = None,
global_redundant_expert_num: int = 0,
shared_gate_up: Optional[torch.Tensor] = None,
shared_dequant_scale: Optional[torch.Tensor] = None,
shared_experts: Optional[torch.Tensor] = None,
):
def token_dispatch(self,
hidden_states: torch.Tensor,
topk_weights: torch.Tensor,
topk_ids: torch.Tensor,
row_idx: torch.Tensor,
expert_map: Optional[torch.Tensor] = None,
log2phy: Optional[torch.Tensor] = None,
global_redundant_expert_num: int = 0,
shared_experts: Optional[torch.Tensor] = None,
shared_gate_up: Optional[torch.Tensor] = None,
shared_dequant_scale: Optional[torch.Tensor] = None,
mc2_mask: Optional[torch.Tensor] = None,
apply_router_weight_on_input: bool = False):
self.original_shape = hidden_states.shape
# assert len(original_shape) == 2
num_tokens = hidden_states.shape[:-1].numel()
dtype = hidden_states.dtype
@@ -748,9 +760,7 @@ class TokenDispatcherWithAllGather(MoETokenDispatcher):
self.expert_map = expert_map
self.topk_weights = topk_weights
self.topk_ids = topk_ids
# assert dtype in [torch.float32, torch.float16, torch.bfloat16
# ], "Only float32, float16, and bfsloat16 are supported"
self.apply_router_weight_on_input = apply_router_weight_on_input
if self.apply_router_weight_on_input:
assert (topk_weights.dim() == 2
), "`topk_weights` should be in shape (num_tokens, topk)"
@@ -803,19 +813,13 @@ class TokenDispatcherWithAllGather(MoETokenDispatcher):
sorted_hidden_states = hidden_states[self.sorted_token_indices]
if self.with_quant:
group_list_type = 1
expert_tokens = token_counts
else:
expert_tokens = torch.cumsum(token_counts,
dim=0,
dtype=torch.int64)
group_list_type = 0
else:
row_idx_len = num_tokens * self.top_k
row_idx = (torch.arange(0,
row_idx_len,
dtype=torch.int32,
device=device).view(self.top_k,
-1).permute(
1, 0).contiguous())
active_num = self.max_num_tokens if self.max_num_tokens is not None else num_tokens
sorted_hidden_states, self.expanded_row_idx, expanded_expert_idx = torch_npu.npu_moe_init_routing(
hidden_states,
@@ -827,18 +831,23 @@ class TokenDispatcherWithAllGather(MoETokenDispatcher):
expanded_expert_idx, self.num_experts_local)
expert_tokens = expert_tokens.to(torch.int64)
group_list_type = 0
return group_list_type, sorted_hidden_states, expert_tokens
return {
"group_list_type": group_list_type,
"hidden_states": sorted_hidden_states,
"group_list": expert_tokens,
}
def token_combine(self,
hidden_states: torch.Tensor,
bias: torch.Tensor = None):
assert self.mask is not None
assert self.sorted_token_indices is not None
assert self.sorted_weights is not None
assert self.original_shape is not None
dtype = hidden_states.dtype
device = hidden_states.device
if self.expert_map is not None:
assert self.mask is not None
assert self.sorted_token_indices is not None
assert self.sorted_weights is not None
weighted_down_out = hidden_states * \
self.sorted_weights.unsqueeze(1)
@@ -887,7 +896,6 @@ class TokenDispatcherWithAllGather(MoETokenDispatcher):
expanded_src_to_dst_row=self.expanded_row_idx,
export_for_source_row=self.topk_ids,
)
return final_hidden_states
@@ -895,29 +903,27 @@ class TokenDispatcherWithAllGather(MoETokenDispatcher):
class UnquantizedTokenDispatcherWithFusedExpertsMoge(MoETokenDispatcher):
def __init__(self, **kwargs):
super(MoETokenDispatcher, self).__init__(**kwargs)
self.apply_router_weight_on_input = kwargs.get(
"apply_router_weight_on_input")
ep_size = kwargs.get("ep_size")
self.local_ep = ep_size
assert self.local_ep is not None
super().__init__(**kwargs)
self.apply_router_weight_on_input = False
self.local_ep = 1
self.local_num_experts = self.num_experts // self.local_ep
self.local_num_group = self.top_k // self.local_ep
self.bsz = None
def token_dispatch(
self,
hidden_states: torch.Tensor,
topk_weights: torch.Tensor,
topk_ids: torch.Tensor,
expert_map: torch.Tensor,
log2phy: Optional[torch.Tensor] = None,
global_redundant_expert_num: int = 0,
shared_gate_up: Optional[torch.Tensor] = None,
shared_dequant_scale: Optional[torch.Tensor] = None,
shared_experts: Optional[torch.Tensor] = None,
):
def token_dispatch(self,
hidden_states: torch.Tensor,
topk_weights: torch.Tensor,
topk_ids: torch.Tensor,
row_idx: torch.Tensor,
expert_map: Optional[torch.Tensor] = None,
log2phy: Optional[torch.Tensor] = None,
global_redundant_expert_num: int = 0,
shared_experts: Optional[torch.Tensor] = None,
shared_gate_up: Optional[torch.Tensor] = None,
shared_dequant_scale: Optional[torch.Tensor] = None,
mc2_mask: Optional[torch.Tensor] = None,
apply_router_weight_on_input: bool = False):
self.apply_router_weight_on_input = apply_router_weight_on_input
if self.apply_router_weight_on_input:
assert (topk_weights.dim() == 2
), "`topk_weights` should be in shape (num_tokens, topk)"
@@ -932,7 +938,7 @@ class UnquantizedTokenDispatcherWithFusedExpertsMoge(MoETokenDispatcher):
flatten_topk_ids = topk_ids.view(-1)
self.sorted_topk_ids = torch.argsort(flatten_topk_ids.float())
self.sorted_topk_ids = self.sorted_topk_ids.to(torch.int32)
self.sorted_hidden_states = hidden_states.index_select(
sorted_hidden_states = hidden_states.index_select(
0, self.sorted_topk_ids // self.local_num_group)
experts_id = torch.arange(0,
@@ -942,15 +948,20 @@ class UnquantizedTokenDispatcherWithFusedExpertsMoge(MoETokenDispatcher):
num_tokens_per_expert = (
flatten_topk_ids.unsqueeze(-1) == experts_id).to(
torch.float32).sum(0)
self.topk_scales = topk_weights.view(-1).index_select(
topk_scales = topk_weights.view(-1).index_select(
0, self.sorted_topk_ids).unsqueeze(-1)
group_list = num_tokens_per_expert.cumsum(dim=0).to(torch.int64)
return hidden_states, group_list
group_list_type = 0
return {
"group_list_type": group_list_type,
"hidden_states": sorted_hidden_states,
"group_list": group_list,
"topk_scales": topk_scales,
}
def token_combine(self,
hidden_states: torch.Tensor,
bias: torch.Tensor = None):
assert self.local_ep is not None
unsorted_topk_ids = torch.argsort(self.sorted_topk_ids.float()).to(
torch.int32)
unsorted_hidden_states = hidden_states.index_select(
@@ -1009,18 +1020,19 @@ class TokenDispatcherWithAll2AllV(MoETokenDispatcher):
self.local_expert_indices[i + 1] -
1), "local_expert_indices must be continuous"
def token_dispatch(
self,
hidden_states: torch.Tensor,
topk_weights: torch.Tensor,
topk_ids: torch.Tensor,
expert_map: torch.Tensor,
log2phy: Optional[torch.Tensor] = None,
global_redundant_expert_num: int = 0,
shared_gate_up: Optional[torch.Tensor] = None,
shared_dequant_scale: Optional[torch.Tensor] = None,
shared_experts: Optional[torch.Tensor] = None,
):
def token_dispatch(self,
hidden_states: torch.Tensor,
topk_weights: torch.Tensor,
topk_ids: torch.Tensor,
row_idx: torch.Tensor,
expert_map: Optional[torch.Tensor] = None,
log2phy: Optional[torch.Tensor] = None,
global_redundant_expert_num: int = 0,
shared_experts: Optional[torch.Tensor] = None,
shared_gate_up: Optional[torch.Tensor] = None,
shared_dequant_scale: Optional[torch.Tensor] = None,
mc2_mask: Optional[torch.Tensor] = None,
apply_router_weight_on_input: bool = False):
self.hidden_shape = hidden_states.shape
self.topk_weights = topk_weights
assert topk_weights.dim() == 2, "Expected 2D tensor for topk_weights"