Skip to content

fix: [nvbugs/5289912][nvbugs/5232406] use thread pool for multi-thread weight loading in fused moe. #4699

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 3 commits into from
May 28, 2025
Merged

fix: [nvbugs/5289912][nvbugs/5232406] use thread pool for multi-thread weight loading in fused moe. #4699

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
55ab15b
fix: use thread pool for multi-thread weight loading in fused moe.
yuxianq May 27, 2025
f6e1854
unwaive tests.
yuxianq May 27, 2025
5b7f8cd
Fix rebase error.
yuxianq May 27, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
82 changes: 46 additions & 36 deletions tensorrt_llm/_torch/modules/fused_moe.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
import math
import os
import threading
from concurrent.futures import ThreadPoolExecutor
from enum import Enum
from typing import Dict, List, NamedTuple, Optional, Union

Expand Down Expand Up @@ -1278,45 +1278,55 @@ def load_expert_w2_weight(w2_weight,
# Even though CPython has global interpreter lock (GIL),
# it's still faster to load weights in parallel because it can utilize
# CPU memory bandwidth better.
threads = []
max_workers = min(
(self.expert_end - self.expert_start) * 2,
os.cpu_count() * 2,
16,
)

for expert_id in range(self.expert_start, self.expert_end):
expert_idx = expert_id - self.expert_start
with ThreadPoolExecutor(max_workers=max_workers) as executor:
futures = []

for expert_id in range(self.expert_start, self.expert_end):
expert_idx = expert_id - self.expert_start

if self.weight_loading_mode == MoEWeightLoadingMode.VANILLA:
w1_weight = weights[f"{expert_id}.w1.weight"]
w3_weight = weights[f"{expert_id}.w3.weight"]
w2_weight = weights[f"{expert_id}.w2.weight"]
elif self.weight_loading_mode == MoEWeightLoadingMode.FUSED_GATE_UP_PROJ:
w1_w3_weight = weights["gate_up_proj"][expert_id].transpose(
0, 1)
w1_weight, w3_weight = w1_w3_weight.chunk(2, dim=0)
w2_weight = weights["down_proj"][expert_id].transpose(
0, 1).contiguous()
else:
raise NotImplementedError(
f"Unknown weight loading mode in MoE: {self.weight_loading_mode}"
)

if self.weight_loading_mode == MoEWeightLoadingMode.VANILLA:
w1_weight = weights[f"{expert_id}.w1.weight"]
w3_weight = weights[f"{expert_id}.w3.weight"]
w2_weight = weights[f"{expert_id}.w2.weight"]
elif self.weight_loading_mode == MoEWeightLoadingMode.FUSED_GATE_UP_PROJ:
w1_w3_weight = weights["gate_up_proj"][expert_id].transpose(
0, 1)
w1_weight, w3_weight = w1_w3_weight.chunk(2, dim=0)
w2_weight = weights["down_proj"][expert_id].transpose(
0, 1).contiguous()
else:
raise NotImplementedError(
f"Unknown weight loading mode in MoE: {self.weight_loading_mode}"
is_trtllm_nvfp4 = self.is_trtllm(
) and self.quant_config.quant_mode.has_nvfp4()

future_w3_w1 = executor.submit(
load_expert_w3_w1_weight,
w1_weight,
w3_weight,
self.w3_w1_weight.data[expert_idx],
is_trtllm_nvfp4,
)
futures.append(future_w3_w1)

future_w2 = executor.submit(
load_expert_w2_weight,
w2_weight,
self.w2_weight.data[expert_idx],
is_trtllm_nvfp4,
)
futures.append(future_w2)

is_trtllm_nvfp4 = self.is_trtllm(
) and self.quant_config.quant_mode.has_nvfp4()

thread = threading.Thread(target=load_expert_w3_w1_weight,
args=(w1_weight, w3_weight,
self.w3_w1_weight.data[expert_idx],
is_trtllm_nvfp4))
thread.start()
threads.append(thread)

thread = threading.Thread(target=load_expert_w2_weight,
args=(w2_weight,
self.w2_weight.data[expert_idx],
is_trtllm_nvfp4))
thread.start()
threads.append(thread)

for thread in threads:
thread.join()
for future in futures:
future.result()

if self.quant_config and self.quant_config.quant_mode.has_any_quant(
exclude_kv_cache=True):
Expand Down
5 changes: 0 additions & 5 deletions tests/integration/test_lists/waives.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -484,11 +484,6 @@ test_e2e.py::test_ptp_quickstart_advanced_8gpus[Llama3.1-70B-BF16-llama-3.1-mode
test_e2e.py::test_ptp_quickstart_advanced_8gpus[Mixtral-8x7B-BF16-Mixtral-8x7B-v0.1] SKIP (https://nvbugs/5136994)
test_e2e.py::test_ptp_quickstart_advanced_8gpus[Llama3.1-70B-FP8-llama-3.1-model/Llama-3.1-70B-Instruct-FP8] SKIP (https://nvbugs/5289909)
test_e2e.py::test_ptp_scaffolding[DeepSeek-R1-Distill-Qwen-7B-DeepSeek-R1/DeepSeek-R1-Distill-Qwen-7B] SKIP (https://nvbugs/5289910)
disaggregated/test_disaggregated.py::test_disaggregated_deepseek_v3_lite_fp8_ucx[DeepSeek-V3-Lite-fp8] SKIP (https://nvbugs/5289912)
disaggregated/test_disaggregated.py::test_disaggregated_deepseek_v3_lite_fp8[DeepSeek-V3-Lite-fp8] SKIP (https://nvbugs/5232406)
disaggregated/test_disaggregated.py::test_disaggregated_deepseek_v3_lite_fp8_attention_dp[DeepSeek-V3-Lite-fp8] SKIP (https://nvbugs/5232406)
disaggregated/test_disaggregated.py::test_disaggregated_deepseek_v3_lite_fp8_attention_dp_one[DeepSeek-V3-Lite-fp8] SKIP (https://nvbugs/5232406)
disaggregated/test_disaggregated.py::test_disaggregated_deepseek_v3_lite_fp8_attention_dp_one_mtp[DeepSeek-V3-Lite-fp8] SKIP (https://nvbugs/5232406)
perf/test_perf.py::test_perf[mixtral_8x22b_v0.1-bench-float16-input_output_len:512,512-quant:fp8-tp:4] SKIP (https://nvbugspro.nvidia.com/bug/5274894)
perf/test_perf.py::test_perf[llama_v3.1_70b-bench-bfloat16-input_output_len:128,128-gpus:4] SKIP (https://nvbugspro.nvidia.com/bug/5274894)
perf/test_perf.py::test_perf[llama_v3.1_70b-bench-streaming-bfloat16-input_output_len:128,128-gpus:4] SKIP (https://nvbugspro.nvidia.com/bug/5274894)
Expand Down