RFR: 8276779: (ch) InputStream returned by Channels.newInputStream should have fast path for SelectableChannels [v14]

[Markus KARG]

On Sun, 21 Nov 2021 08:32:49 GMT, Alan Bateman <alanb at openjdk.org> wrote:

>> Markus KARG has refreshed the contents of this pull request, and previous commits have been removed. The incremental views will show differences compared to the previous content of the PR.
>
> test/jdk/java/nio/channels/Channels/TransferTo.java line 198:
> 
>> 196: 
>> 197:         // IllegalBlockingMode must be thrown when trying to perform a transfer
>> 198:         assertThrows(IllegalBlockingModeException.class, () -> is2.transferTo(os2));
> 
> Thanks for adding this test. It looks like it leaks 5 file descriptors (the FileChannel, and the source/sink of two pipes). Can you update these to close the channels? The test current runs in othervm mode but if it is changed back to agentvm mode then leaving channels open could impact tests that are run later in the same VM.

I sketched a possibly solution in https://github.com/openjdk/jdk/pull/5179/commits/26ed670be318669edf1ef9a7101c1136273f2b82, but looking at the PipeImpl internals I doubt that *all* file descriptors actually get closed. To me it feels like this being actually a design flaw of PipeImpl: If one side of a pipe was never actually claimed then it should not keep a descriptor open. So IMHO the final solution would be to file a PR for PipeImpl which either defers opening of the file descriptors until they are claimed (which I doubt would be easily possible), or which cleans up both sides if as soon as one side is closed and the other side never got claimed (which is my personal preference).

-------------

PR: https://git.openjdk.java.net/jdk/pull/5179