Implement AVX2 SIMD optimization #10
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
- Add l2_simd with AVX2 intrinsics - Enable -march=native for vectorization
5000user5000
commented
Nov 9, 2025
Owner
Author
5000user5000
left a comment
5000user5000
left a comment
There was a problem hiding this comment.
實作 AVX SIMD 加速
- 增加 -match=native 啟用 CPU 指令集
- 原有 L2 計算新增 AVX SIMD 計算
- 將舊的 L2 函式名替換成新的
| @@ -1,5 +1,5 @@ | |||
| CXX := g++ | |||
| CXXFLAGS := -std=c++17 -O3 -fPIC -fopenmp | |||
| CXXFLAGS := -std=c++17 -O3 -fPIC -march=native -fopenmp | |||
Owner
Author
There was a problem hiding this comment.
增加 -march=native , 啟用 CPU 支援指令集
Comment on lines
+6
to
+36
| inline float l2_simd(const float* __restrict a, | ||
| const float* __restrict b, | ||
| size_t dim) { | ||
| #if defined(__AVX2__) | ||
| const size_t step = 8; // 8 × 32-bit floats | ||
| __m256 acc = _mm256_setzero_ps(); | ||
| size_t i = 0; | ||
| for (; i + step - 1 < dim; i += step) { | ||
| __m256 va = _mm256_loadu_ps(a + i); | ||
| __m256 vb = _mm256_loadu_ps(b + i); | ||
| __m256 diff = _mm256_sub_ps(va, vb); | ||
| acc = _mm256_fmadd_ps(diff, diff, acc); // acc += diff² | ||
| } | ||
| float buf[step]; | ||
| _mm256_storeu_ps(buf, acc); | ||
| float d = 0.f; | ||
| for (int j = 0; j < step; ++j) d += buf[j]; | ||
|
|
||
| for (; i < dim; ++i) { | ||
| float diff = a[i] - b[i]; | ||
| d += diff * diff; | ||
| } | ||
| return d; | ||
| #else | ||
| float d = 0.f; | ||
| for (size_t i = 0; i < dim; ++i) { | ||
| float diff = a[i] - b[i]; | ||
| d += diff * diff; | ||
| } | ||
| return d; | ||
| #endif |
Owner
Author
There was a problem hiding this comment.
L2 計算新增 AVX SIMD 版本,如果不支援 AVX2,則會退回原版
| #pragma omp parallel for schedule(static) | ||
| for (size_t c = 0; c < nlist_; ++c) { | ||
| float d = l2_naive(query.data(), centroids_[c].data(), dimension_); | ||
| float d = l2_simd(query.data(), centroids_[c].data(), dimension_); |
Owner
Author
There was a problem hiding this comment.
將既有 l2_naive 替換成新的 l2_simd
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
實作 AVX2 SIMD 加速