RFR: 8308144: Uncontrolled memory consumption in SSLFlowDelegate.Reader [v2]

Daniel Fuchs dfuchs at openjdk.org
Mon Nov 20 10:53:33 UTC 2023


On Mon, 20 Nov 2023 05:14:59 GMT, Jaikiran Pai <jpai at openjdk.org> wrote:

>> src/java.net.http/share/classes/jdk/internal/net/http/common/SSLFlowDelegate.java line 99:
>> 
>>> 97:  *              |   downWriter                          |   downReader
>>> 98:  *              | supplied to constructor               |  supplied to constructor
>>> 99:  *              v                                       v
>> 
>> Can we keep the original diagram too? The new diagram is very nice but gives the impression that encrypt/decrypt are completely independent threads of execution and they're not - since consuming some data may probe the engine to request sending of some data.
>
> Hello Daniel, I've updated the PR to slightly improve the current proposed form to address the part that the encrypt/decrypt are independent. To me, the old diagram wasn't clear enough on what represented upstream/downstream (given that the directions were horizontal), plus what kind of data the upstream/downstream received.
> 
> Do you think in the updated version of this PR, the diagram addresses this review comment? If not, I'll update the PR further to include the original diagram too.

I think you will need to invert the right hand side of the new diagram: for reading, upstream is below, while for writing, upstream is above.

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

PR Review Comment: https://git.openjdk.org/jdk/pull/16704#discussion_r1399015201


More information about the net-dev mailing list