Skip to content

perf(placement): prefetch the table the scoring loop probes #117

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鈥檒l occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
amkram merged 1 commit into from
Jul 13, 2026
Merged

perf(placement): prefetch the table the scoring loop probes #117

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
8 changes: 4 additions & 4 deletions src/placement.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -268,8 +268,9 @@ void placement::NodeMetrics::computeChildMetrics(placement::NodeMetrics& childMe
const size_t initialPrefetchCount = std::min(numChanges, PREFETCH_DISTANCE);
for (size_t i = 0; i < initialPrefetchCount; ++i) {
__builtin_prefetch(&H[i], 0, 0);
state.seedFreqInReads.prefetch(H[i]);
state.seedInverseGenomeCounts.prefetch(H[i]);
// Prefetch the table this loop actually probes (logReadCounts), not the
// read-processing tables that are never read here.
state.logReadCounts.prefetch(H[i]);
}

// Batch for instruction pipelining and cache utilization
Expand All @@ -281,8 +282,7 @@ void placement::NodeMetrics::computeChildMetrics(placement::NodeMetrics& childMe
if (i + PREFETCH_DISTANCE < numChanges) [[likely]] {
const uint64_t nextHash = H[i + PREFETCH_DISTANCE];
__builtin_prefetch(&H[i + PREFETCH_DISTANCE], 0, 0);
state.seedFreqInReads.prefetch(nextHash);
state.seedInverseGenomeCounts.prefetch(nextHash);
state.logReadCounts.prefetch(nextHash);
}

const uint64_t seedHash = H[i];
Expand Down
Loading