You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@flink.apache.org by NicoK <gi...@git.apache.org> on 2017/08/25 12:53:15 UTC

[GitHub] flink pull request #4593: [FLINK-7516][memory] do not allow copies into a re...

GitHub user NicoK opened a pull request:

    https://github.com/apache/flink/pull/4593

    [FLINK-7516][memory] do not allow copies into a read-only ByteBuffer

    ## What is the purpose of the change
    
    `HybridMemorySegment#get(int, ByteBuffer, int)` allows writing into a read-only `ByteBuffer` but this operation should be forbidden.
    
    ## Brief change log
    
    - throw a `ReadOnlyBufferException` if trying to write into a read-only `ByteBuffer`
    
    ## Verifying this change
    
    This change added additional test methods to `MemorySegmentTestBase`.
    
    ## Does this pull request potentially affect one of the following parts:
    
      - Dependencies (does it add or upgrade a dependency): (no)
      - The public API, i.e., is any changed class annotated with `@Public(Evolving)`: (no)
      - The serializers: (no)
      - The runtime per-record code paths (performance sensitive): (yes)
      - Anything that affects deployment or recovery: JobManager (and its components), Checkpointing, Yarn/Mesos, ZooKeeper: (no)
    
    ## Documentation
    
      - Does this pull request introduce a new feature? (no)
      - If yes, how is the feature documented? (JavaDocs)
    


You can merge this pull request into a Git repository by running:

    $ git pull https://github.com/NicoK/flink flink-7516

Alternatively you can review and apply these changes as the patch at:

    https://github.com/apache/flink/pull/4593.patch

To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:

    This closes #4593
    
----
commit f2eb28d3cdbbe7ef4d2e66da263e47336b6f105d
Author: Nico Kruber <ni...@data-artisans.com>
Date:   2017-08-22T16:33:55Z

    [FLINK-7516][memory] do not allow copies into a read-only ByteBuffer

----


---
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.
---

[GitHub] flink pull request #4593: [FLINK-7516][memory] do not allow copies into a re...

Posted by NicoK <gi...@git.apache.org>.
Github user NicoK commented on a diff in the pull request:

    https://github.com/apache/flink/pull/4593#discussion_r148209947
  
    --- Diff: flink-core/src/main/java/org/apache/flink/core/memory/HybridMemorySegment.java ---
    @@ -306,6 +307,9 @@ public final void get(int offset, ByteBuffer target, int numBytes) {
     		if ((offset | numBytes | (offset + numBytes)) < 0) {
     			throw new IndexOutOfBoundsException();
     		}
    +		if (target.isReadOnly()) {
    --- End diff --
    
    you are right - the non-direct buffers path is based on `ByteBuffer#array()` which will throw a `ReadOnlyBufferException` for read-only buffers, so it really is enough in the direct buffers code path where the `UNSAFE.copyMemory` is not checking the source pointer (how should it?!)


---

[GitHub] flink pull request #4593: [FLINK-7516][memory] do not allow copies into a re...

Posted by asfgit <gi...@git.apache.org>.
Github user asfgit closed the pull request at:

    https://github.com/apache/flink/pull/4593


---

[GitHub] flink pull request #4593: [FLINK-7516][memory] do not allow copies into a re...

Posted by pnowojski <gi...@git.apache.org>.
Github user pnowojski commented on a diff in the pull request:

    https://github.com/apache/flink/pull/4593#discussion_r148204666
  
    --- Diff: flink-core/src/main/java/org/apache/flink/core/memory/HybridMemorySegment.java ---
    @@ -306,6 +307,9 @@ public final void get(int offset, ByteBuffer target, int numBytes) {
     		if ((offset | numBytes | (offset + numBytes)) < 0) {
     			throw new IndexOutOfBoundsException();
     		}
    +		if (target.isReadOnly()) {
    --- End diff --
    
    Isn't this check redundant? Shouldn't the `ByteBuffer`s validate it on their own like `DirectByteBufferR` do? Putting this check here, it complicates the code and we have to pay for it on each call, even on happy path (cost should be very tiny). 
    
    Maybe it should be put only in `if (target.isDirect())` branch to check `read-only` only before unsafe copy?


---