You are viewing a plain text version of this content. The canonical link for it is here.
Posted to reviews@spark.apache.org by jerryshao <gi...@git.apache.org> on 2017/08/30 13:39:14 UTC

[GitHub] spark pull request #19077: [SPARK-21860][core]Improve memory reuse for heap ...

Github user jerryshao commented on a diff in the pull request:

    https://github.com/apache/spark/pull/19077#discussion_r136069800
  
    --- Diff: common/unsafe/src/main/java/org/apache/spark/unsafe/memory/HeapMemoryAllocator.java ---
    @@ -47,23 +47,29 @@ private boolean shouldPool(long size) {
     
       @Override
       public MemoryBlock allocate(long size) throws OutOfMemoryError {
    -    if (shouldPool(size)) {
    +    int arraySize = (int)((size + 7) / 8);
    +    long alignedSize = arraySize * 8;
    +    if (shouldPool(alignedSize)) {
           synchronized (this) {
    -        final LinkedList<WeakReference<MemoryBlock>> pool = bufferPoolsBySize.get(size);
    +        final LinkedList<WeakReference<MemoryBlock>> pool = bufferPoolsBySize.get(alignedSize);
             if (pool != null) {
               while (!pool.isEmpty()) {
                 final WeakReference<MemoryBlock> blockReference = pool.pop();
                 final MemoryBlock memory = blockReference.get();
                 if (memory != null) {
    -              assert (memory.size() == size);
    +              assert ((int)((memory.size() + 7) / 8) == arraySize);
    +              memory.resetSize(size);
    --- End diff --
    
    Hmm, from my understanding the size of `MemoryBlock` is always the actual size, not the aligned size, so looks like we dont need to reset the size here.


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org