I'm going to persevere with this a bit more.. but first I need to better understand "safe publication"
Thanks Andrew Andrew Leonard Java Runtimes Development IBM Hursley IBM United Kingdom Ltd internet email: andrew_m_leon...@uk.ibm.com From: Alan Bateman <alan.bate...@oracle.com> To: Andrew Leonard <andrew_m_leon...@uk.ibm.com> Cc: serviceability-dev@openjdk.java.net Date: 14/06/2019 17:06 Subject: Re: RFR JDK-8225474: JDI connector accept fails "Address already in use" with concurrent listeners On 14/06/2019 15:52, Andrew Leonard wrote: In doing the recent changes I applied knowledge of how the ConnectorImpl and its defaultArguments are used to decide on what is necessary from a threading perspective. I also am only considering the "listening" concurrency issue, at this stage, and was not considering making all the jdi classes thread-safe. Sure but default arguments exposed in the ListeningConnector API are mutable. It may be that debuggers and other tools aren't changing them but we can't say that the GenericListeningConnector and its implementations are safe for by use by concurrent threads without studying this further. Also the comments about changing fields to final are not optimizations, that are suggestions to allow GenericListeningConnector objects be safely published. -Alan Unless stated otherwise above: IBM United Kingdom Limited - Registered in England and Wales with number 741598. Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU
