Skip to content

Small cleanup of lower_sync_info #2023

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
Oct 8, 2022
Merged

Small cleanup of lower_sync_info #2023

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
20 changes: 0 additions & 20 deletions torch/csrc/jit/codegen/cuda/lower_sync_information.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -410,26 +410,6 @@ void SyncMap::build(Fusion* fusion) {
}
}

// If same parallel type and mapped, no need for syncs unless
// producer is in smem, producer parallel type is a thread
// dimension, and consumer concretizes the dimension. This sync is
// due to the redundant predicate omission in lower thread
// predicate.
auto redundant_preds = GpuLower::current()
->threadPredMap()
.getPredicateInfo(producer)
.redundant_types;

if (p_id->isBroadcast() &&
GpuLower::current()->concretizedBroadcastDomains()->isConcretized(
p_id) &&
producer->getMemoryType() == MemoryType::Shared &&
redundant_preds.hasTID()) {
redundant_preds.clearAllBID();
raw_dims |= redundant_preds;
continue;
}

// When the producer axis is a broadcast, it is not really
// parallelized unless thread-predicated and concretized
if (isParallelTypeThread(producer_ptype) && p_id->isBroadcast() &&
Expand Down