RFR: JDK-8319307: DCmds should not assert on truncation and should report truncation [v2]

[Thomas Stuefe]

On Tue, 28 Nov 2023 06:46:53 GMT, David Holmes <dholmes at openjdk.org> wrote:

>> src/hotspot/share/nmt/memReporter.cpp line 323:
>> 
>>> 321:   const MallocSite* malloc_site;
>>> 322:   int num_omitted = 0;
>>> 323:   while (!output()->was_truncated() && (malloc_site = malloc_itr.next()) != nullptr) {
>> 
>> I don't think clients should be aware of, or worrying about truncation, this way. It seems very intrusive.
>
> Again where do we inform the consumer of the stream that the content has been truncated?

Delayed, at the caller. NMT report is called via dcmd, and attach listener queries the truncation flag. Here, we only query the truncation flag to save unnecessary work.

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

PR Review Comment: https://git.openjdk.org/jdk/pull/16474#discussion_r1407318482