Unverified Commit 9b8ebb27 authored by Lianmin Zheng's avatar Lianmin Zheng Committed by GitHub
Browse files

move more files under srt/utils (#11285)

parent 758b887a
...@@ -487,7 +487,7 @@ def make_layers( ...@@ -487,7 +487,7 @@ def make_layers(
# circula imports # circula imports
from sglang.srt.distributed import get_pp_indices from sglang.srt.distributed import get_pp_indices
from sglang.srt.layers.utils import PPMissingLayer from sglang.srt.layers.utils import PPMissingLayer
from sglang.srt.offloader import get_offloader from sglang.srt.utils.offloader import get_offloader
assert not pp_size or num_hidden_layers >= pp_size assert not pp_size or num_hidden_layers >= pp_size
start_layer, end_layer = ( start_layer, end_layer = (
......
...@@ -11,14 +11,14 @@ from sglang.srt.distributed.naive_distributed import ( ...@@ -11,14 +11,14 @@ from sglang.srt.distributed.naive_distributed import (
get_naive_distributed, get_naive_distributed,
set_naive_distributed, set_naive_distributed,
) )
from sglang.srt.host_shared_memory import ( from sglang.srt.layers.parameter import ModelWeightParameter
from sglang.srt.server_args import ServerArgs
from sglang.srt.utils import MultiprocessingSerializer, is_pin_memory_available
from sglang.srt.utils.host_shared_memory import (
HostSharedMemoryManager, HostSharedMemoryManager,
get_host_shared_memory_manager, get_host_shared_memory_manager,
set_host_shared_memory_manager, set_host_shared_memory_manager,
) )
from sglang.srt.layers.parameter import ModelWeightParameter
from sglang.srt.server_args import ServerArgs
from sglang.srt.utils import MultiprocessingSerializer, is_pin_memory_available
logger = logging.getLogger(__name__) logger = logging.getLogger(__name__)
......
...@@ -7,7 +7,6 @@ from pathlib import Path ...@@ -7,7 +7,6 @@ from pathlib import Path
import torch import torch
import triton import triton
from sglang.srt.bench_utils import bench_kineto
from sglang.srt.layers.quantization.fp8_kernel import ( from sglang.srt.layers.quantization.fp8_kernel import (
create_per_token_group_quant_fp8_output_scale, create_per_token_group_quant_fp8_output_scale,
) )
...@@ -16,6 +15,7 @@ from sglang.srt.layers.quantization.fp8_kernel import ( ...@@ -16,6 +15,7 @@ from sglang.srt.layers.quantization.fp8_kernel import (
) )
from sglang.srt.layers.quantization.fp8_kernel import sglang_per_token_group_quant_8bit from sglang.srt.layers.quantization.fp8_kernel import sglang_per_token_group_quant_8bit
from sglang.srt.utils import is_hip from sglang.srt.utils import is_hip
from sglang.srt.utils.bench_utils import bench_kineto
# CI environment detection # CI environment detection
IS_CI = ( IS_CI = (
......
...@@ -11,7 +11,7 @@ from sgl_kernel.testing.rotary_embedding import ( ...@@ -11,7 +11,7 @@ from sgl_kernel.testing.rotary_embedding import (
create_inputs, create_inputs,
) )
from sglang.srt.bench_utils import bench_kineto from sglang.srt.utils.bench_utils import bench_kineto
# CI environment detection # CI environment detection
IS_CI = ( IS_CI = (
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment