Skip to content

compiler: Schedule the readers of a waited-on Function after the wait #3037

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
mloubout wants to merge 3 commits into
base: main
Choose a base branch
from
Open

compiler: Schedule the readers of a waited-on Function after the wait #3037

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
bea92eb
compiler: Make every reader of a prefetched buffer wait for it
mloubout Sep 29, 2026
2d8bfe3
compiler: Keep a prefetch reader's scalar definitions in its wait
mloubout Sep 30, 2026
751250e
compiler: Schedule the readers of a waited-on Function after the wait
mloubout Sep 30, 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
7 changes: 7 additions & 0 deletions devito/passes/clusters/fusion.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -305,10 +305,17 @@ def _build_dag(self, cgroups, prefix):
# Track whether there is any fence between `cg0` and the current `cg1`.
fenced = cg0.scope.has_barrier

# Functions `cg0` waits on, e.g. a prefetched buffer: reading them
# is no data hazard, but no reader may be scheduled before the wait
waited = {s.target for s in flatten(cg0.syncs.values())

@FabioLuporini FabioLuporini Sep 30, 2026

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

so what's s.target, and why does it not appear among cg0's equations?

if isinstance(s, WaitLock)}

for n1, cg1 in enumerate(cgroups[n+1:], start=n+1):
fenced = fenced or cg1.scope.has_barrier

hazard = _fusion_hazards(cg0.scope, cg1.scope, prefix)
if waited & set(cg1.scope.reads):
dag.add_edge(cg0, cg1)
if not (hazard or fenced):
continue

Expand Down
Loading