Skip to content

Fix GCC-12.x prefetch-loop collapse in greedy_search neighbor prefetch #361

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
yuejiaointel wants to merge 6 commits into
base: main
Choose a base branch
from
Open

Fix GCC-12.x prefetch-loop collapse in greedy_search neighbor prefetch #361

Changes from all commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
067221b
Fix GCC-12.x prefetch-loop collapse in greedy_search neighbor prefetch
yuejiaointel Aug 7, 2026
3405406
Use portable #pragma GCC unroll instead of x86 inline asm
yuejiaointel Aug 7, 2026
47d552d
Revert to inline-asm (the version proven to work in the built module)
yuejiaointel Aug 7, 2026
621e902
Use portable std::atomic_signal_fence instead of inline asm
yuejiaointel Aug 7, 2026
288960b
Shorten prefetch fence comment
yuejiaointel Aug 7, 2026
a3f6f83
clang-format: order <atomic> include
yuejiaointel Aug 8, 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
4 changes: 4 additions & 0 deletions include/svs/lib/prefetch.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,6 +18,7 @@

#include "svs/lib/misc.h"

#include <atomic>
#include <cstdint>
#include <iterator>
#include <span>
Expand Down Expand Up @@ -56,8 +57,11 @@ template <typename T, size_t Extent> void prefetch_l0(std::span<T, Extent> span)
);
}

// The fence is a no-op at runtime; without it GCC 12 drops all but the first
// prefetch here, so neighbor vectors load cold from DRAM and search slows ~24%.
for (size_t i = 0; i < num_prefetches; ++i) {
prefetch_l0(base + CACHELINE_BYTES * i);
std::atomic_signal_fence(std::memory_order_seq_cst);
}
}

Expand Down
Loading