allocate the correct amount of chunks for caching (missing was the run count) and add them to the queue for each run

qdp_project/src/Benchmark.cpp

@@ -163,10 +163,9 @@ int main() {
         CACHE_.SetFlags(cache_flags);
         CACHE_.Init(CachePlacementPolicy, CopyMethodPolicy, MemoryAllocator_Allocate, MemoryAllocator_Free);
 
-        for (uint32_t i = 0; i < GROUP_COUNT * TC_AGGRJ; i++) {
+        for (uint32_t i = 0; i < GROUP_COUNT * TC_AGGRJ * RUN_COUNT; i++) {
             void* ptr = numa_alloc_onnode(AGGRJ_CHUNK_SIZE_B, MEM_NODE_HBM);
             cache_memory_locations.emplace_back(ptr);
-            CACHE_LOCATIONS_.push(reinterpret_cast<uint8_t*>(ptr));
             memset(ptr, 0xAB, AGGRJ_CHUNK_SIZE_B);
         }
     }
@@ -178,6 +177,10 @@ int main() {
         std::promise<void> launch_promise;
         LAUNCH_ = launch_promise.get_future();
 
+	for (void* ptr : cache_memory_locations) {
+            CACHE_LOCATIONS_.push(reinterpret_cast<uint8_t*>(ptr));
+	}
+
         std::vector<std::thread> filter_pool;
         std::vector<std::thread> copy_pool;
         std::vector<std::thread>  agg_pool;