mirror of
https://github.com/PaddlePaddle/FastDeploy.git
synced 2026-04-22 16:07:51 +08:00
yzwu
129 lines
5.1 KiB
Python
129 lines
5.1 KiB
Python
"""
|
|
# Copyright (c) 2025 PaddlePaddle Authors. All Rights Reserved.
|
|
#
|
|
# Licensed under the Apache License, Version 2.0 (the "License"
|
|
# you may not use this file except in compliance with the License.
|
|
# You may obtain a copy of the License at
|
|
#
|
|
# http://www.apache.org/licenses/LICENSE-2.0
|
|
#
|
|
# Unless required by applicable law or agreed to in writing, software
|
|
# distributed under the License is distributed on an "AS IS" BASIS,
|
|
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
# See the License for the specific language governing permissions and
|
|
# limitations under the License.
|
|
"""
|
|
|
|
from functools import partial
|
|
|
|
import paddle
|
|
from paddleformers.utils.log import logger
|
|
|
|
from fastdeploy import envs
|
|
from fastdeploy.config import FDConfig
|
|
from fastdeploy.model_executor.layers.attention import get_attention_backend
|
|
from fastdeploy.worker.gpu_model_runner import GPUModelRunner
|
|
|
|
|
|
def _patch_before_model_runner():
|
|
paddle.Tensor.pin_memory = paddle.Tensor.cpu
|
|
paddle.device.cuda.create_event = partial(paddle.device.custom_device.create_event, device_type="iluvatar_gpu")
|
|
|
|
def disable_record(self):
|
|
pass
|
|
|
|
paddle.device.custom_device.Event.record = disable_record
|
|
|
|
def disable_synchronize(self):
|
|
pass
|
|
|
|
paddle.device.custom_device.Event.synchronize = disable_synchronize
|
|
|
|
|
|
_patch_before_model_runner()
|
|
|
|
|
|
class IluvatarModelRunner(GPUModelRunner):
|
|
def __init__(
|
|
self,
|
|
fd_config: FDConfig,
|
|
device: str, # logic device
|
|
device_id: int, # physical device id
|
|
rank: int,
|
|
local_rank: int,
|
|
):
|
|
super(IluvatarModelRunner, self).__init__(
|
|
fd_config=fd_config, device=device, device_id=device_id, rank=rank, local_rank=local_rank
|
|
)
|
|
assert not self.speculative_decoding, "Iluvatar does not support speculative decoding"
|
|
assert self.guided_backend is None, "Iluvatar does not support guided decoding"
|
|
assert not self.cache_config.enable_prefix_caching, "Iluvatar does not support prefix caching"
|
|
self.mla_cache = envs.FD_ATTENTION_BACKEND == "MLA_ATTN"
|
|
assert not self.mla_cache, "Iluvatar does not support MLA"
|
|
self.dsa_cache = envs.FD_ATTENTION_BACKEND == "DSA_ATTN"
|
|
assert not self.dsa_cache, "Iluvatar does not support DSA_ATTN"
|
|
if self.enable_mm:
|
|
assert (
|
|
not self.cache_config.enable_chunked_prefill
|
|
), "Iluvatar does not support chunked prefill for VL model"
|
|
|
|
if self.model_config.model_type == "ernie4_5_moe_vl" and self.parallel_config.tensor_parallel_size > 1:
|
|
# ernie-vl does not support cuda graph for tp > 1
|
|
logger.warning("disable cudagraph since ernie-vl does not support cuda graph for tp > 1")
|
|
self.use_cudagraph = False
|
|
|
|
if hasattr(self.quant_config, "moe_quant_type") and self.quant_config.moe_quant_type == "wint4":
|
|
# Iluvatar does not support cuda graph for weight_only_int4 yet
|
|
logger.warning("disable cudagraph since iluvatar does not support cuda graph for weight_only_int4")
|
|
self.use_cudagraph = False
|
|
|
|
logger.info(f"self.use_cudagraph={self.use_cudagraph}")
|
|
# VL neox style = True
|
|
emb_shape = self.share_inputs["rope_emb"].shape
|
|
if emb_shape[-1] == self.model_config.head_dim // 2:
|
|
emb_shape[-1] = self.model_config.head_dim
|
|
self.share_inputs["rope_emb"] = paddle.full(
|
|
shape=emb_shape,
|
|
fill_value=0,
|
|
dtype="float32",
|
|
)
|
|
|
|
def _initialize_attn_backend(self) -> None:
|
|
"""
|
|
Initialize attention backends
|
|
"""
|
|
assert (
|
|
len(self.attn_backends) == 0
|
|
), f"attn_backends should be empty before initialization, got {len(self.attn_backends)} backends"
|
|
|
|
num_heads = self.model_config.num_attention_heads // self.parallel_config.tensor_parallel_size
|
|
self.model_config.kv_num_heads = max(
|
|
1,
|
|
int(self.model_config.num_key_value_heads) // self.parallel_config.tensor_parallel_size,
|
|
)
|
|
attn_cls = get_attention_backend()
|
|
attn_backend = attn_cls(
|
|
self.fd_config,
|
|
kv_num_heads=self.model_config.kv_num_heads,
|
|
num_heads=num_heads,
|
|
head_dim=self.model_config.head_dim,
|
|
)
|
|
self.attn_backends.append(attn_backend)
|
|
|
|
def initialize_kv_cache(self, profile: bool = False) -> None:
|
|
super(IluvatarModelRunner, self).initialize_kv_cache(profile)
|
|
paddle.device.empty_cache()
|
|
|
|
def initialize_forward_meta(self, is_dummy_or_profile_run=False):
|
|
super(IluvatarModelRunner, self).initialize_forward_meta(is_dummy_or_profile_run)
|
|
only_decode = self.forward_meta.attn_backend.prefill_len == 0
|
|
self.fd_config.model_config.moe_phase.phase = "decode" if only_decode else "prefill"
|
|
|
|
def clear_cache(self):
|
|
super(IluvatarModelRunner, self).clear_cache()
|
|
paddle.device.empty_cache()
|
|
|
|
def clear_parameters(self, pid):
|
|
super(IluvatarModelRunner, self).clear_parameters(pid)
|
|
paddle.device.empty_cache()
|