RFR: 8263464: NMT: assert in gtest os.release_multi_mappings_vm [v7]

[Johan Sjölen]

On Tue, 23 May 2023 09:50:35 GMT, Johan Sjölen <jsjolen at openjdk.org> wrote:

>> Gerard Ziemski has updated the pull request incrementally with one additional commit since the last revision:
>> 
>>   Johan's feedback: add comments, handle pesky NULL
>
> src/hotspot/share/services/virtualMemoryTracker.cpp line 536:
> 
>> 534:       reserved_rgn->exclude_region(addr, size);
>> 535:       return true;
>> 536:     } else {
> 
> OK! We split the region into two, that looks good. What happens to the committed regions that are part of both the removed reserved region and one of the surviving regions (or why not both?).
> 
> 
>           *   &
>           |   |
> [---------------]
>        [----]
> ~>
> [-----]     [---]
> 
> 
> What if we have a committed region starting by the `*` and ending at the `&`, what happens to it?

Easier example:


- = reserved
C = committed
D = to be deleted
[---------------]
  [CCCCCCCCC]
     [DDD]
Are the results:
[----]   [-------]
  [CC]   [CC]
Where the left C is part of the left - and the right C is part of the right -
Because it looks like the results might be:
[----]   [-------]
  [CCCCCCCCC]
where the left - owns the committed region

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

PR Review Comment: https://git.openjdk.org/jdk/pull/13813#discussion_r1201949862