Github user JoshRosen commented on a diff in the pull request:
https://github.com/apache/spark/pull/9000#discussion_r41568609
--- Diff: core/src/main/scala/org/apache/spark/storage/MemoryStore.scala ---
@@ -489,17 +478,21 @@ private[spark] class MemoryStore(blockManager:
BlockManager, maxMemory: Long)
}
/**
- * Reserve additional memory for unrolling blocks used by this task.
- * Return whether the request is granted.
+ * Reserve memory for unrolling the given block for this task.
+ * @return whether the request is granted.
*/
- def reserveUnrollMemoryForThisTask(memory: Long): Boolean = {
+ def reserveUnrollMemoryForThisTask(
+ blockId: BlockId,
+ memory: Long,
+ droppedBlocks: mutable.Buffer[(BlockId, BlockStatus)]): Boolean = {
accountingLock.synchronized {
- val granted = freeMemory > currentUnrollMemory + memory
- if (granted) {
+ val acquired = memoryManager.acquireUnrollMemory(blockId, memory,
droppedBlocks)
+ val success = acquired == memory
+ if (success) {
val taskAttemptId = currentTaskAttemptId()
unrollMemoryMap(taskAttemptId) =
unrollMemoryMap.getOrElse(taskAttemptId, 0L) + memory
}
--- End diff --
AFAIK, `acquireUnrollMemory` can return less than the requested number of
bytes, so don't you have to free the acquired unroll memory in an `else` branch
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 [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
