GitHub user zhijiangW opened a pull request:
https://github.com/apache/flink/pull/6417
[FLINK-9913][runtime] Improve output serialization only once in RecordWriter
## What is the purpose of the change
*This pull request improves the output serialization only once for multi
Github user zhijiangW commented on the issue:
https://github.com/apache/flink/pull/5381
Yes, I think it can make better use of floating buffers.
As last confirmation with Piotr, this PR will not be merged into
release-1.5, because there were still some issues not confirmed
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/6254#discussion_r201293862
--- Diff:
flink-runtime/src/main/java/org/apache/flink/runtime/io/network/partition/consumer/RemoteInputChannel.java
---
@@ -594,22 +626,22 @@ public
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/6254#discussion_r201293052
--- Diff:
flink-runtime/src/main/java/org/apache/flink/runtime/io/network/partition/consumer/RemoteInputChannel.java
---
@@ -479,6 +508,9 @@ void
Github user zhijiangW commented on the issue:
https://github.com/apache/flink/pull/6272
LGTM!
---
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/6272#discussion_r201292263
--- Diff:
flink-runtime/src/main/java/org/apache/flink/runtime/io/network/partition/consumer/RemoteInputChannel.java
---
@@ -360,32 +360,45 @@ public
Github user zhijiangW commented on the issue:
https://github.com/apache/flink/pull/6272
Thanks for fixing this potential bug.
It makes sense to handle exception during `notifyBufferAvailable` on
listener side. Just some thoughts on my side above. :)
---
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/6272#discussion_r201247182
--- Diff:
flink-runtime/src/test/java/org/apache/flink/runtime/io/network/partition/consumer/RemoteInputChannelTest.java
---
@@ -687,7 +690,65 @@ public
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/6272#discussion_r201242662
--- Diff:
flink-runtime/src/main/java/org/apache/flink/runtime/io/network/partition/consumer/RemoteInputChannel.java
---
@@ -360,32 +360,45 @@ public
Github user zhijiangW commented on the issue:
https://github.com/apache/flink/pull/6271
LGTM!
---
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/6257#discussion_r200287796
--- Diff:
flink-runtime/src/main/java/org/apache/flink/runtime/io/network/buffer/LocalBufferPool.java
---
@@ -251,27 +257,56 @@ private MemorySegment
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/6257#discussion_r200287730
--- Diff:
flink-runtime/src/main/java/org/apache/flink/runtime/io/network/buffer/LocalBufferPool.java
---
@@ -251,27 +257,56 @@ private MemorySegment
Github user zhijiangW commented on the issue:
https://github.com/apache/flink/pull/6257
Thanks for fixing this bug, and we also solve this problem in this way.
This solution seems more lightweight than the way in
[6254](https://github.com/apache/flink/pull/6254), and I also
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/6257#discussion_r200252253
--- Diff:
flink-runtime/src/main/java/org/apache/flink/runtime/io/network/buffer/LocalBufferPool.java
---
@@ -251,27 +257,56 @@ private MemorySegment
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/6257#discussion_r200251500
--- Diff:
flink-runtime/src/main/java/org/apache/flink/runtime/io/network/buffer/LocalBufferPool.java
---
@@ -251,27 +257,56 @@ private MemorySegment
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/6254#discussion_r200245464
--- Diff:
flink-runtime/src/main/java/org/apache/flink/runtime/io/network/partition/consumer/RemoteInputChannel.java
---
@@ -594,22 +626,22 @@ public
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/6254#discussion_r200247127
--- Diff:
flink-runtime/src/main/java/org/apache/flink/runtime/io/network/partition/consumer/RemoteInputChannel.java
---
@@ -479,6 +508,9 @@ void
Github user zhijiangW commented on the issue:
https://github.com/apache/flink/pull/6238
ð
---
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/6238#discussion_r199676237
--- Diff:
flink-runtime/src/main/java/org/apache/flink/runtime/io/network/buffer/NetworkBufferPool.java
---
@@ -147,7 +151,12 @@ public void recycle
Github user zhijiangW commented on the issue:
https://github.com/apache/flink/pull/6238
Thanks for fixing this bug. :)
I think there is another potential bug in polling segments from queue
during task canceling process, and I pointed out it in the above codes.
---
Github user zhijiangW commented on the issue:
https://github.com/apache/flink/pull/5923
Thanks for improving it, looks good from my point. :)
---
Github user zhijiangW commented on the issue:
https://github.com/apache/flink/pull/5916
LGTM :)
---
Github user zhijiangW commented on the issue:
https://github.com/apache/flink/pull/5915
The points is already `rebalance()` in line 131, so it is no need to
`rebalance()` again in line 137?
---
Github user zhijiangW commented on the issue:
https://github.com/apache/flink/pull/5747
Thanks for fixing this problem. The `notifyReaderCreated` should be called
after both views are created correctly, otherwise it will cause inconsistent.
LGTM ð
---
Github user zhijiangW commented on the issue:
https://github.com/apache/flink/pull/5708
Thanks piotr, I agree with it.
---
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/5317#discussion_r171575083
--- Diff:
flink-core/src/main/java/org/apache/flink/configuration/TaskManagerOptions.java
---
@@ -269,15 +269,21 @@
public static final
Github user zhijiangW commented on the issue:
https://github.com/apache/flink/pull/4529
Thanks for telling me the plan of it. I will do that if necessary. :)
---
Github user zhijiangW commented on the issue:
https://github.com/apache/flink/pull/4529
Hey @NicoK , would this PR be covered in FLINK-1.5? We experienced the
netty direct memory out of memory sometimes in production cased by
`extractFrame`, so we expect this improvement. :)
---
Github user zhijiangW commented on the issue:
https://github.com/apache/flink/pull/5317
@pnowojski , thanks for your messages and I already updated the commit as
you suggested.
---
Github user zhijiangW commented on the issue:
https://github.com/apache/flink/pull/5558
Thanks for your reviews! I already addressed above comments in a separate
commit.
Regarding the tests I also think the same with you. :)
---
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/5558#discussion_r170148337
--- Diff:
flink-runtime/src/test/java/org/apache/flink/runtime/io/network/partition/consumer/RemoteInputChannelTest.java
---
@@ -396,32 +399,49 @@ public
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/5558#discussion_r170148132
--- Diff:
flink-runtime/src/main/java/org/apache/flink/runtime/io/network/partition/consumer/RemoteInputChannel.java
---
@@ -337,6 +337,11 @@ public int
Github user zhijiangW commented on the issue:
https://github.com/apache/flink/pull/5558
@NicoK , as I described in the purpose of the change, there are two
scenarios for this issue. And I only modified the previous
`RemoteInputChannelTest#testAvailableBuffersLessThanRequiredBuffers
GitHub user zhijiangW opened a pull request:
https://github.com/apache/flink/pull/5558
[FLINK-8747][bugfix] The tag of waiting for floating buffers in
RemoteInputChannel should be updated properly
## What is the purpose of the change
*In credit-based flow control mode
Github user zhijiangW commented on the issue:
https://github.com/apache/flink/pull/5317
@NicoK , I found most of the codes in this PR are already merged into
master by commit `0093bcbe771f296baf3857ef15fe7ec9b22bbc34` in your
`FLINK-8425`. Maybe I only need to add the `config.md` in
Github user zhijiangW commented on the issue:
https://github.com/apache/flink/pull/5317
Yes, I also found that and was rebasing on the latest codes.
Because I am on the Spring Festival vacation, it can be ready later today
or on Thursday.
---
Github user zhijiangW commented on the issue:
https://github.com/apache/flink/pull/5400
@pnowojski , I have changed the `EXACTLY_ONCE_BLOCKING_DATA_ENABLED` as
true and squashed the commits.
---
Github user zhijiangW commented on the issue:
https://github.com/apache/flink/pull/5400
Thanks for rebasing the conflicts.
Yes, the default value can be changed to true after the credit-based is
totally merged. If need any changes on my side after all, pls let me know. :)
---
Github user zhijiangW commented on the issue:
https://github.com/apache/flink/pull/5400
@pnowojski , I have submitted the updates for above comments.
---
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/5400#discussion_r167582971
--- Diff:
flink-streaming-java/src/main/java/org/apache/flink/streaming/runtime/io/BufferBlocker.java
---
@@ -0,0 +1,61 @@
+/*
+ * Licensed to
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/5400#discussion_r167582882
--- Diff:
flink-streaming-java/src/test/java/org/apache/flink/streaming/runtime/io/BarrierBufferTest.java
---
@@ -18,1426 +18,40 @@
package
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/5400#discussion_r167582763
--- Diff:
flink-streaming-java/src/main/java/org/apache/flink/streaming/runtime/io/BufferBlocker.java
---
@@ -0,0 +1,61 @@
+/*
+ * Licensed to
Github user zhijiangW commented on the issue:
https://github.com/apache/flink/pull/5400
@pnowojski , thanks for suggestions and I totally agree with that.
That abstraction indeed makes the code simple. I will update the codes
ASAP.
---
Github user zhijiangW commented on the issue:
https://github.com/apache/flink/pull/5400
@pnowojski , I have submitted a separate commit to address above comments.
---
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/5400#discussion_r166175837
--- Diff:
flink-streaming-java/src/main/java/org/apache/flink/streaming/runtime/io/StreamInputProcessor.java
---
@@ -131,10 +131,14 @@ public
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/5400#discussion_r166174743
--- Diff:
flink-streaming-java/src/main/java/org/apache/flink/streaming/runtime/io/CreditBasedBarrierBuffer.java
---
@@ -0,0 +1,529
Github user zhijiangW commented on the issue:
https://github.com/apache/flink/pull/5400
@pnowojski , thanks for reviews!
I understand your concerns and I should deduplicate some common utils in
these tests. I will do that tomorrow together with other comments!
---
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/5400#discussion_r165998584
--- Diff:
flink-streaming-java/src/main/java/org/apache/flink/streaming/runtime/io/StreamInputProcessor.java
---
@@ -131,10 +131,14 @@ public
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/5400#discussion_r165997853
--- Diff:
flink-streaming-java/src/main/java/org/apache/flink/streaming/runtime/io/CreditBasedBarrierBuffer.java
---
@@ -0,0 +1,529
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/5400#discussion_r165983607
--- Diff:
flink-streaming-java/src/main/java/org/apache/flink/streaming/runtime/io/CreditBasedBarrierBuffer.java
---
@@ -0,0 +1,529
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/5400#discussion_r165983714
--- Diff:
flink-streaming-java/src/main/java/org/apache/flink/streaming/runtime/io/CreditBasedBarrierBuffer.java
---
@@ -0,0 +1,529
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/5400#discussion_r165983496
--- Diff:
flink-core/src/main/java/org/apache/flink/configuration/TaskManagerOptions.java
---
@@ -184,6 +184,18 @@
key
GitHub user zhijiangW opened a pull request:
https://github.com/apache/flink/pull/5400
[FLINK-8547][network] Implement CheckpointBarrierHandler not to spill data
for exactly-once
## What is the purpose of the change
*Currently in exactly-once mode, the BarrierBuffer would
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/5381#discussion_r164993308
--- Diff:
flink-runtime/src/main/java/org/apache/flink/runtime/io/network/partition/consumer/SingleInputGate.java
---
@@ -485,6 +494,23 @@ public void
Github user zhijiangW commented on the issue:
https://github.com/apache/flink/pull/5381
Thanks for reviews and suggestions! :)
I will add some unit tests first to verify the related logics. For itcase,
I will consider the necessary and feasibility.
---
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/5381#discussion_r164990652
--- Diff:
flink-runtime/src/main/java/org/apache/flink/runtime/io/network/partition/consumer/UnionInputGate.java
---
@@ -138,6 +148,24 @@ public void
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/5381#discussion_r164988630
--- Diff:
flink-runtime/src/main/java/org/apache/flink/runtime/io/network/partition/consumer/SingleInputGate.java
---
@@ -315,6 +322,7 @@ public void
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/5381#discussion_r164988413
--- Diff:
flink-runtime/src/main/java/org/apache/flink/runtime/io/network/partition/consumer/SingleInputGate.java
---
@@ -133,8 +134,13
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/5381#discussion_r164988217
--- Diff:
flink-runtime/src/main/java/org/apache/flink/runtime/io/network/partition/consumer/RemoteInputChannel.java
---
@@ -360,8 +360,9 @@ public
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/5381#discussion_r164987333
--- Diff:
flink-runtime/src/main/java/org/apache/flink/runtime/io/network/partition/consumer/InputChannel.java
---
@@ -165,6 +168,14 @@ protected void
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/5381#discussion_r164986882
--- Diff:
flink-runtime/src/main/java/org/apache/flink/runtime/io/network/partition/consumer/InputChannel.java
---
@@ -66,6 +66,9 @@
/** The
Github user zhijiangW commented on the issue:
https://github.com/apache/flink/pull/5381
There is one issue I am thinking.
After the input channel is unblocked, we can check whether this input
channel needs request floating buffer from pool (if `availableNum <
numRequiredBuff
Github user zhijiangW commented on the issue:
https://github.com/apache/flink/pull/5381
@pnowojski I have submitted the whole process in one commit.
After you verify the implementation is feasible, I will submit a separate
commit for adding unit tests based on this process.
---
GitHub user zhijiangW opened a pull request:
https://github.com/apache/flink/pull/5381
[FLINK-8523][network] Stop assigning floating buffers for blocked input
channels in exactly-once mode
## What is the purpose of the change
In exactly-once mode, the input channel is set
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/5317#discussion_r163155607
--- Diff: docs/ops/config.md ---
@@ -290,6 +290,12 @@ The following parameters configure Flink's JobManager
and TaskMan
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/5317#discussion_r163155437
--- Diff: docs/ops/config.md ---
@@ -290,6 +290,12 @@ The following parameters configure Flink's JobManager
and TaskMan
GitHub user zhijiangW opened a pull request:
https://github.com/apache/flink/pull/5317
[FLINK-8458] Add the switch for keeping both the old mode and the new
credit-based mode
## What is the purpose of the change
*After the whole feature of credit-based flow control is done
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/4552#discussion_r162266243
--- Diff:
flink-runtime/src/main/java/org/apache/flink/runtime/io/network/partition/SpilledSubpartitionView.java
---
@@ -31,7 +31,6 @@
import
Github user zhijiangW commented on the issue:
https://github.com/apache/flink/pull/4552
1. Thanks for you FLINK08425.
2. I would have thought the tests for
`ResultSubpartition#nextBufferIsEvent` which have already been covered before.
The test for `BufferAndBacklog
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/4552#discussion_r162259781
--- Diff:
flink-runtime/src/main/java/org/apache/flink/runtime/io/network/netty/PartitionRequestClientFactory.java
---
@@ -164,11 +165,13 @@ private
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/4552#discussion_r162259728
--- Diff:
flink-runtime/src/main/java/org/apache/flink/runtime/io/network/netty/PartitionRequestServerHandler.java
---
@@ -82,10 +83,17 @@ protected void
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/4552#discussion_r162259766
--- Diff:
flink-runtime/src/main/java/org/apache/flink/runtime/io/network/netty/PartitionRequestQueue.java
---
@@ -125,11 +126,11 @@ private void
Github user zhijiangW commented on the issue:
https://github.com/apache/flink/pull/4552
@NicoK , I have submitted the switch for keeping the old mode and the new
credit-based mode.
---
Github user zhijiangW commented on the issue:
https://github.com/apache/flink/pull/4552
@NicoK , thanks for your reviews!
I have submitted all the patches you provided offline to address above
issues.
1. Remove `FLINK-8425` from this PR.
2. Do you think I should add
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/4552#discussion_r161667346
--- Diff:
flink-tests/src/test/java/org/apache/flink/test/misc/SuccessAfterNetworkBuffersFailureITCase.java
---
@@ -59,7 +59,7 @@ public void
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/4552#discussion_r161667234
--- Diff:
flink-runtime/src/test/java/org/apache/flink/runtime/taskmanager/TaskCancelAsyncProducerConsumerITCase.java
---
@@ -84,7 +84,7 @@ public void
Github user zhijiangW commented on the issue:
https://github.com/apache/flink/pull/4552
@NicoK , I have submitted all the modifications based on the patches you
provided.
The tests for `nextBufferIsEvent` will be added in a new commit tomorrow.
---
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/4552#discussion_r161450876
--- Diff:
flink-runtime/src/main/java/org/apache/flink/runtime/io/network/partition/SpillableSubpartitionView.java
---
@@ -199,6 +199,19 @@ public
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/4552#discussion_r160849525
--- Diff:
flink-runtime/src/main/java/org/apache/flink/runtime/io/network/netty/PartitionRequestQueue.java
---
@@ -250,10 +304,12 @@ private void
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/4552#discussion_r160849478
--- Diff:
flink-runtime/src/main/java/org/apache/flink/runtime/io/network/netty/PartitionRequestQueue.java
---
@@ -88,6 +94,37 @@ public void run
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/4552#discussion_r160371761
--- Diff:
flink-runtime/src/main/java/org/apache/flink/runtime/io/network/netty/PartitionRequestQueue.java
---
@@ -88,6 +93,35 @@ public void run
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/4552#discussion_r160333139
--- Diff:
flink-runtime/src/main/java/org/apache/flink/runtime/io/network/netty/PartitionRequestQueue.java
---
@@ -88,6 +93,35 @@ public void run
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/4552#discussion_r160333068
--- Diff:
flink-runtime/src/main/java/org/apache/flink/runtime/io/network/netty/PartitionRequestQueue.java
---
@@ -88,6 +93,35 @@ public void run
Github user zhijiangW commented on the issue:
https://github.com/apache/flink/pull/4509
Already close it
---
Github user zhijiangW closed the pull request at:
https://github.com/apache/flink/pull/4509
---
Github user zhijiangW closed the pull request at:
https://github.com/apache/flink/pull/4533
---
Github user zhijiangW commented on the issue:
https://github.com/apache/flink/pull/4533
Sure, I close it.
---
Github user zhijiangW commented on the issue:
https://github.com/apache/flink/pull/4559
@NicoK , I have submitted the `hotfix` commit to address above comments.
---
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/4559#discussion_r157709904
--- Diff:
flink-runtime/src/main/java/org/apache/flink/runtime/io/network/partition/PipelinedSubpartition.java
---
@@ -52,6 +54,10 @@
/** Flag
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/4559#discussion_r157706951
--- Diff:
flink-runtime/src/main/java/org/apache/flink/runtime/io/network/partition/ResultSubpartition.java
---
@@ -99,6 +82,23 @@ protected Throwable
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/4559#discussion_r157694294
--- Diff:
flink-runtime/src/test/java/org/apache/flink/runtime/io/network/partition/SubpartitionTestBase.java
---
@@ -47,7 +48,14 @@ public void
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/4559#discussion_r157693477
--- Diff:
flink-runtime/src/test/java/org/apache/flink/runtime/io/network/partition/SpillableSubpartitionTest.java
---
@@ -181,10 +182,27 @@ public void
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/4559#discussion_r157691096
--- Diff:
flink-runtime/src/main/java/org/apache/flink/runtime/io/network/partition/PipelinedSubpartition.java
---
@@ -52,6 +54,10 @@
/** Flag
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/4559#discussion_r157686388
--- Diff:
flink-runtime/src/main/java/org/apache/flink/runtime/io/network/partition/ResultSubpartition.java
---
@@ -99,6 +82,23 @@ protected Throwable
Github user zhijiangW commented on the issue:
https://github.com/apache/flink/pull/4559
@NicoK, I have submitted two `[hotfix]` commits for the above issues.
One is for `Nullable` annotation and tests of backlog statistics. And the
other is for updating backlog in thread-safe
Github user zhijiangW commented on the issue:
https://github.com/apache/flink/pull/4559
@NicoK , thanks for suggestions.
I understand your point of wrapping the buffer and backlog together in a
new structure returned by `getNextBuffer()` and it really makes sense for
Github user zhijiangW commented on the issue:
https://github.com/apache/flink/pull/4559
Or we make the backlog as `AtomicInteger` to keep the current process,
otherwise we may need to call `decreaseStatistics` in different parts in the
region of `synchronized(buffers)` . What do you
Github user zhijiangW commented on the issue:
https://github.com/apache/flink/pull/4559
@NicoK @pnowojski , for backlog thread-safe issue, the current
implementation is not thread-safe and we should restore my previous
implementation that calls `decreaseStatistics` inside the method
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/4559#discussion_r155458048
--- Diff:
flink-runtime/src/main/java/org/apache/flink/runtime/io/network/partition/SpillableSubpartitionView.java
---
@@ -145,6 +145,10 @@ public Buffer
Github user zhijiangW commented on a diff in the pull request:
https://github.com/apache/flink/pull/4559#discussion_r155454935
--- Diff:
flink-runtime/src/main/java/org/apache/flink/runtime/io/network/partition/ResultSubpartitionView.java
---
@@ -22,32 +22,57 @@
import
1 - 100 of 274 matches
Mail list logo
