Skip to content

fix(compiler): default cluster_dims=(1,1,1) for backends that omit it #1020

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

Open
tengqm wants to merge 1 commit into
base: main
Choose a base branch
from
Open

fix(compiler): default cluster_dims=(1,1,1) for backends that omit it #1020

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
6 changes: 5 additions & 1 deletion python/triton/compiler/compiler.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -434,7 +434,11 @@ def __init__(self, src, metadata_group, hash):
target = metadata['target']
metadata['target'] = GPUTarget(target['backend'], target['arch'], target['warp_size'])
# Restore tuple-typed metadata fields serialized as JSON arrays.
cluster_dims = metadata.get("cluster_dims")
# Backends that do not support cluster launches (e.g. hcu, mthreads)
# omit cluster_dims entirely; torch._inductor reads it unconditionally
# when the kernel carries metadata, so default it here rather than
# letting every such backend patch torch/inductor in its own adapter.
cluster_dims = metadata.setdefault("cluster_dims", (1, 1, 1))
if isinstance(cluster_dims, list):
metadata["cluster_dims"] = tuple(cluster_dims)
KernelMetadata = namedtuple('KernelMetadata', sorted(list(metadata.keys())))
Expand Down
Loading