fix: always capture default prefill CUDA graph (#33352)

This commit is contained in:
Mick
2026-08-08 19:24:49 +08:00
committed by GitHub
parent cf2d4fd679
commit db75dfe10f
3 changed files with 48 additions and 45 deletions
@@ -50,24 +50,6 @@ if TYPE_CHECKING:
logger = logging.getLogger(__name__)
# Measured immediately before prefill graph construction, after model weights,
# KV cache, and the eager runner's static buffers have been allocated. Below
# this budget, compiling/capturing a multi-bucket prefill graph is likely to
# OOM or make no forward progress for large models. Keep an explicitly chosen
# backend untouched: an operator may intentionally trade KV capacity for it.
_MIN_AUTO_PREFILL_CUDA_GRAPH_FREE_MEMORY_GB = 4.0
def should_skip_auto_prefill_cuda_graph_for_memory(
available_memory_gb: float,
cuda_graph_config_locked: set[tuple[str, str]],
) -> bool:
"""Return whether an auto-selected prefill graph lacks capture headroom."""
return (
(Phase.PREFILL, "backend") not in cuda_graph_config_locked
and available_memory_gb < _MIN_AUTO_PREFILL_CUDA_GRAPH_FREE_MEMORY_GB
)
class GraphCapture(msgspec.Struct, frozen=True, kw_only=True):
runner: Optional[BaseRunner]
@@ -396,20 +378,6 @@ def capture_prefill_graph(
tic = time.perf_counter()
before_mem = get_available_gpu_memory(model_runner.device, model_runner.gpu_id)
if should_skip_auto_prefill_cuda_graph_for_memory(
before_mem,
getattr(model_runner.server_args, "_cuda_graph_config_locked", set()),
):
logger.warning(
"Disabling auto-selected prefill CUDA graph: only %.2f GiB is free "
"after model/KV/eager-buffer allocation; at least %.2f GiB is "
"required for capture. Set an explicit prefill CUDA graph backend "
"to override this safety gate.",
before_mem,
_MIN_AUTO_PREFILL_CUDA_GRAPH_FREE_MEMORY_GB,
)
return result(eager_runner)
role = "draft" if model_runner.is_draft_worker else "target"
capture_name = f"{role} prefill"
logger.info(