MDEV-40827: prefetch MHNSW neighbours during search

[chanztuying]

Summary

Prefetch unseen MHNSW neighbour allocations before their distance evaluations.

Correctness

• Clean debug and release builds; the non-GNU #else path also compiles.
• --do-test=vector and main.mysqld--help pass in both.
• Exact brute-force recall is identical with and without the patch in every cell below.

Performance

Counterbalanced AB/BA, warm cache, both binaries querying the same persisted graph (200k x 1024, cosine, M=6); 200 queries, 30 paired observations per cell.
AMD EPYC 7713, RelWithDebInfo, 8-CPU cgroup quota on a shared host.

ef_search	upstream	prefetch	paired median spe
40	1364.2	1518.2	+10.1% (+8.0%, +14.4%)
160	795.2	913.8	+13.8% (+11.8%, +20.0%)

The gain grows with ef_search: larger ef expands more nodes, so more neighbour groups are scanned and more memory latency is available to hide.

[CLAassistant]

All committers have signed the CLA.

[gkodinov]

Thank you for your contribution! This is a preliminary review.

Please consider increasing the compiler coverage as suggested below. And stand by for the final review.

[gkodinov]

please enable this for clang and msvc. This is what grok says should be done:

#if defined(__GNUC__) || defined(__clang__)
    __builtin_prefetch(link, 0, 3);   // rw: 0=read, 1=write; locality: 0–3
#elif defined(_MSC_VER) && (defined(_M_IX86) || defined(_M_X64))
    #include <xmmintrin.h>   // or <intrin.h>
    _mm_prefetch((const char*)link, _MM_HINT_T0);   // _MM_HINT_T0, _MM_HINT_T1, _MM_HINT_T2, _MM_HINT_NTA
#else
    // no-op or do nothing
    (void)addr;
#endif

[vaintroub]

There is also PreFetchCacheLine , which can be used just with _WIN32 , i.e does not depend on architecture

[chanztuying]

There is also PreFetchCacheLine , which can be used just with _WIN32 , i.e does not depend on architecture

Good idea, added to make prefetch portable!

[chanztuying]

Addressed in the latest commit. Now supports GCC/Clang, MSVC x86/x64, and MSVC ARM64.

[Copilot]

Pull request overview

Adds portable CPU cache prefetching for unseen MHNSW neighbors during vector search.

Changes:

• Prefetch neighbor allocations before distance evaluation.
• Add cross-platform my_prefetch() support.

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated no comments.

File	Description
sql/vector_mhnsw.cc	Prefetches unseen MHNSW neighbors during search.
include/my_cpu.h	Provides portable prefetch helper implementations.

Suppressed comments (1)

sql/vector_mhnsw.cc:1360

• This assumes that the vector begins in the cache line immediately after the node's line, but the allocation is FVectorNode + gref + tref + vector, and the supported reference length can be much larger than one cache line. For such indexes this prefetch only warms the node/reference area; distance_to() still starts on a cold vector, so the optimization misses its target. Prefetch the actual vector address (using link->vec when loaded, otherwise derive it from tref() + tref_len() with FVector::align_ptr()).

          my_prefetch(link);
          my_prefetch(reinterpret_cast<const char*>(link)
                      + CPU_LEVEL1_DCACHE_LINESIZE);

---

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.