Fix OpenMP critical section contention in IndexBinaryHNSW search

[sharm235]

Summary:

Problem

We recently found 87% of CPU was being wasted on OpenMP lock contention in FlatHammingDis::~FlatHammingDis, not on useful computation.

The flame graph breakdown:

• 88.2% CPU in openmp_worker threads
• 87.2% in FlatHammingDis::~FlatHammingDis → __kmpc_critical_with_hint → __kmp_acquire_queuing_lock → __sched_yield (84% CPU spinning/yielding on lock)

Root Cause

The FlatHammingDis destructor used #pragma omp critical to accumulate a single size_t counter (hnsw_stats.ndis += ndis). Unnamed #pragma omp critical sections share a global serialization lock — when all OpenMP threads exit the #pragma omp parallel block in IndexBinaryHNSW::search() simultaneously, they ALL enter the destructor at the same time, serializing on that single lock.

With N threads, this means N sequential lock acquisitions where each thread spins/yields waiting for its turn. This is O(N) serialization at the end of every search call.

In IndexBinaryHNSWCagra::search() with base_level_only=true, the situation is even worse: FlatHammingDis is created and destroyed per query iteration inside #pragma omp parallel for, causing n × num_threads critical section entries.

Fix

Replace #pragma omp critical with #pragma omp atomic. Since hnsw_stats.ndis += ndis is a simple size_t addition, #pragma omp atomic compiles to a single hardware atomic instruction (lock xadd on x86-64) — orders of magnitude faster than a mutex-based critical section, with effectively zero contention.

For reference, the float HNSW path in IndexHNSW.cpp already uses the correct pattern: #pragma omp for reduction(+: n1, n2, ndis, nhops) with a single-threaded hnsw_stats.combine() call outside the parallel region.

Impact

• Eliminates ~87% CPU waste from lock contention in binary HNSW search
• Affects all users of IndexBinaryHNSW::search() and IndexBinaryHNSWCagra::search()
• No change to search results or statistics accuracy — #pragma omp atomic provides the same correctness guarantees as #pragma omp critical for a single += operation

Differential Revision: D95910991

[meta-codesync]

@sharm235 has exported this pull request. If you are a Meta employee, you can view the originating Diff in D95910991.

[meta-codesync]

This pull request has been merged in 5b83ec6.