Skip to content

[V1] Remove _get_cache_block_size #12214

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 1 commit into from
Jan 20, 2025
Merged

[V1] Remove _get_cache_block_size #12214

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
24 changes: 1 addition & 23 deletions vllm/v1/worker/gpu_worker.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -7,14 +7,13 @@
import torch.distributed

import vllm.envs as envs
from vllm.config import CacheConfig, ModelConfig, ParallelConfig, VllmConfig
from vllm.config import ParallelConfig, VllmConfig
from vllm.distributed import (ensure_model_parallel_initialized,
init_distributed_environment,
set_custom_all_reduce)
from vllm.logger import init_logger
from vllm.model_executor import set_random_seed
from vllm.platforms import current_platform
from vllm.utils import STR_DTYPE_TO_TORCH_DTYPE, LayerBlockType, get_dtype_size
from vllm.v1.core.scheduler import SchedulerOutput
from vllm.v1.kv_cache_interface import KVCacheConfig, KVCacheSpec
from vllm.v1.outputs import ModelRunnerOutput
Expand Down Expand Up @@ -231,24 +230,3 @@ def _check_if_gpu_supports_dtype(torch_dtype: torch.dtype):
f"of at least 8.0. Your {gpu_name} GPU {compute_str}. "
"You can use float16 instead by explicitly setting the"
"`dtype` flag in CLI, for example: --dtype=half.")


def _get_cache_block_size(
cache_config: CacheConfig,
model_config: ModelConfig,
parallel_config: ParallelConfig,
) -> int:
head_size = model_config.get_head_size()
num_heads = model_config.get_num_kv_heads(parallel_config)
num_attention_layers = model_config.get_num_layers_by_block_type(
parallel_config, LayerBlockType.attention)

key_cache_block = cache_config.block_size * num_heads * head_size
value_cache_block = key_cache_block
total = num_attention_layers * (key_cache_block + value_cache_block)
if cache_config.cache_dtype == "auto":
dtype = model_config.dtype
else:
dtype = STR_DTYPE_TO_TORCH_DTYPE[cache_config.cache_dtype]
dtype_size = get_dtype_size(dtype)
return dtype_size * total
Loading