fix: memory leak in delete

Signed-off-by: aaronzuo <anarionzuo@outlook.com>

Author: Anarion-zuo

.github/workflows/build.yml

@@ -47,6 +47,40 @@ jobs:
           fi
         shell: bash
 
+  leak_sanitizers:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v3
+      - uses: actions/setup-python@v4
+        with:
+          python-version: "3.11"
+
+      - name: Build and install (ASAN/LSAN)
+        env:
+          HNSWLIB_NO_NATIVE: "1"
+          CFLAGS: "-O1 -g -fno-omit-frame-pointer -fsanitize=address,leak"
+          CXXFLAGS: "-O1 -g -fno-omit-frame-pointer -fsanitize=address,leak"
+          LDFLAGS: "-fsanitize=address,leak"
+        run: |
+          python -m pip install -U pip setuptools wheel
+          python -m pip install -v --no-build-isolation .
+
+      - name: Python leak smoke test (ASAN/LSAN)
+        timeout-minutes: 15
+        env:
+          ASAN_OPTIONS: "detect_leaks=1:halt_on_error=1:alloc_dealloc_mismatch=1"
+        run: |
+          python -m unittest discover -v --start-directory tests/python --pattern "bindings_test_leaks.py"
+
+      - name: C++ leak smoke test (ASAN/LSAN)
+        timeout-minutes: 15
+        env:
+          ASAN_OPTIONS: "detect_leaks=1:halt_on_error=1:alloc_dealloc_mismatch=1"
+        run: |
+          cmake -S . -B build-asan -DHNSWLIB_EXAMPLES=ON -DCMAKE_BUILD_TYPE=RelWithDebInfo -DCMAKE_CXX_FLAGS="-O1 -g -fno-omit-frame-pointer -fsanitize=address,leak -std=c++11" -DCMAKE_EXE_LINKER_FLAGS="-fsanitize=address,leak"
+          cmake --build build-asan --target leak_smoke_test -j 2
+          ./build-asan/leak_smoke_test
+
       - name: Prepare test data
         run: |
           pip install numpy
@@ -77,4 +111,5 @@ jobs:
           ./test_updates update
           ./multivector_search_test
           ./epsilon_search_test
+          ./leak_smoke_test
         shell: bash

CMakeLists.txt

@@ -100,6 +100,9 @@ if(HNSWLIB_EXAMPLES)
     add_executable(multiThread_replace_test tests/cpp/multiThread_replace_test.cpp)
     target_link_libraries(multiThread_replace_test hnswlib)
 
+    add_executable(leak_smoke_test tests/cpp/leak_smoke_test.cpp)
+    target_link_libraries(leak_smoke_test hnswlib)
+
     add_executable(main tests/cpp/main.cpp tests/cpp/sift_1b.cpp)
     target_link_libraries(main hnswlib)
 endif()

python_bindings/bindings.cpp

@@ -392,19 +392,19 @@ class Index {
         }
 
         py::capsule free_when_done_l0(data_level0_npy, [](void* f) {
-            delete[] f;
+            free(f);
             });
         py::capsule free_when_done_lvl(element_levels_npy, [](void* f) {
-            delete[] f;
+            free(f);
             });
         py::capsule free_when_done_lb(label_lookup_key_npy, [](void* f) {
-            delete[] f;
+            free(f);
             });
         py::capsule free_when_done_id(label_lookup_val_npy, [](void* f) {
-            delete[] f;
+            free(f);
             });
         py::capsule free_when_done_ll(link_list_npy, [](void* f) {
-            delete[] f;
+            free(f);
             });
 
         /*  TODO: serialize state of random generators appr_alg->level_generator_ and appr_alg->update_probability_generator_  */
@@ -676,10 +676,10 @@ class Index {
             }
         }
         py::capsule free_when_done_l(data_numpy_l, [](void* f) {
-            delete[] f;
+            delete[] reinterpret_cast<hnswlib::labeltype*>(f);
             });
         py::capsule free_when_done_d(data_numpy_d, [](void* f) {
-            delete[] f;
+            delete[] reinterpret_cast<dist_t*>(f);
             });
 
         return py::make_tuple(
@@ -884,10 +884,10 @@ class BFIndex {
         }
 
         py::capsule free_when_done_l(data_numpy_l, [](void *f) {
-            delete[] f;
+            delete[] reinterpret_cast<hnswlib::labeltype*>(f);
         });
         py::capsule free_when_done_d(data_numpy_d, [](void *f) {
-            delete[] f;
+            delete[] reinterpret_cast<dist_t*>(f);
         });
 
 

tests/cpp/leak_smoke_test.cpp

@@ -0,0 +1,38 @@
+#include "assert.h"
+#include "../../hnswlib/hnswlib.h"
+#include <random>
+#include <vector>
+
+int main() {
+    int dim = 16;
+    int max_elements = 1000;
+    int M = 16;
+    int ef_construction = 200;
+
+    std::mt19937 rng(42);
+    std::uniform_real_distribution<float> distrib(0.0f, 1.0f);
+
+    std::vector<float> data(dim * max_elements);
+    for (size_t i = 0; i < data.size(); i++) {
+        data[i] = distrib(rng);
+    }
+
+    for (int iter = 0; iter < 5; iter++) {
+        hnswlib::L2Space space(dim);
+        hnswlib::HierarchicalNSW<float>* alg_hnsw = new hnswlib::HierarchicalNSW<float>(
+            &space, max_elements, M, ef_construction, 42 + iter);
+
+        for (int i = 0; i < max_elements; i++) {
+            alg_hnsw->addPoint(data.data() + (i * dim), i);
+        }
+
+        for (int i = 0; i < 50; i++) {
+            auto result = alg_hnsw->searchKnn(data.data() + (i * dim), 10);
+            assert(result.size() == 10);
+        }
+
+        delete alg_hnsw;
+    }
+
+    return 0;
+}

tests/python/bindings_test.py

@@ -44,6 +44,7 @@ def testRandomSelf(self):
         # Query the elements for themselves and measure recall:
         labels, distances = p.knn_query(data1, k=1)
         self.assertAlmostEqual(np.mean(labels.reshape(-1) == np.arange(len(data1))), 1.0, 3)
+        del labels, distances
 
         # Serializing and deleting the index:
         index_path = 'first_half.bin'
@@ -64,5 +65,6 @@ def testRandomSelf(self):
         labels, distances = p.knn_query(data, k=1)
 
         self.assertAlmostEqual(np.mean(labels.reshape(-1) == np.arange(len(data))), 1.0, 3)
+        del labels, distances
 
         os.remove(index_path)

tests/python/bindings_test_leaks.py

@@ -0,0 +1,32 @@
+import gc
+import pickle
+import unittest
+
+import numpy as np
+
+import hnswlib
+
+
+class LeakSmokeTestCase(unittest.TestCase):
+    def testLeakSmoke(self):
+        dim = 8
+        max_elements = 200
+
+        for _ in range(25):
+            data = np.float32(np.random.random((max_elements, dim)))
+
+            p = hnswlib.Index(space='l2', dim=dim)
+            p.init_index(max_elements=max_elements, ef_construction=100, M=16)
+            p.add_items(data)
+
+            labels, distances = p.knn_query(data[:25], k=5)
+            del labels, distances
+
+            payload = pickle.dumps(p)
+            del p
+            gc.collect()
+
+            p2 = pickle.loads(payload)
+            labels2, distances2 = p2.knn_query(data[:10], k=3)
+            del labels2, distances2, p2, payload, data
+            gc.collect()