Github user tedyu commented on the issue:
https://github.com/apache/flink/pull/5205
Looks like the PR needs to be rebased.
---
Github user tedyu commented on the issue:
https://github.com/apache/flink/pull/5205
+1 on @StephanEwen's suggestion.
---
Github user StephanEwen commented on the issue:
https://github.com/apache/flink/pull/5205
One think we may want to do is to change the return type of `pageSize` to
`long` (even if it can be at most MAX_INT). That should prevent most of these
bugs from the start.
---
Github user StephanEwen commented on the issue:
https://github.com/apache/flink/pull/5205
I think you can test the `MemoryManager` by simply setting pre-allocation
to false and calling the method to compute the memory size.
For the other two, not sure how to best do that...
Github user greghogan commented on the issue:
https://github.com/apache/flink/pull/5205
@StephanEwen of the five occurrences in `MemoryManager`,
`ChannelWriterOutputView`, and `InPlaceMutableHashTable` only one occurs in a
function that is called elsewhere. How would you write the
