Skip to content

fix(worker): guard kernel_warmup against missing torchvision on OOT runtimes #386

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
cyber-pioneer merged 2 commits into from
Aug 17, 2026
Merged

fix(worker): guard kernel_warmup against missing torchvision on OOT runtimes #386

Changes from 1 commit
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
a48e02a
fix(worker): guard kernel_warmup against missing torchvision on OOT r…
tengqm Aug 17, 2026
78722f8
fix(worker): log detailed import error when skipping kernel_warmup
tengqm Aug 17, 2026
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
10 changes: 9 additions & 1 deletion vllm_fl/worker/worker.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -671,7 +671,15 @@ def compile_or_warm_up_model(self) -> CompilationTimes:
### NOTE(lms): can add gems kernel pretune here
# Warmup and tune the kernels used during model execution before
# cuda graph capture.
kernel_warmup(self)
try:
kernel_warmup(self)
except ImportError:
# vllm 0.24.0's kernel_warmup unconditionally imports
# minimax_m3_msa_warmup, whose chain reaches torchvision.
# torchvision is not installed on OOT runtimes (installing it
# would overwrite the vendor-matched torch matrix); the warmup
# is a no-op for any model other than MiniMaxM3, so skip it.
logger.warning("kernel_warmup skipped: torchvision unavailable")
Comment thread
tengqm marked this conversation as resolved.
Outdated

cuda_graph_memory_bytes = 0
if self.vllm_config.compilation_config.cudagraph_mode != CUDAGraphMode.NONE:
Expand Down
Loading