RFR: 8319873: Add windows implementation for jcmd System.map and System.dump_map [v8]

[Simon Tooke]

On Tue, 10 Sep 2024 15:09:03 GMT, Sonia Zaldana Calles <szaldana at openjdk.org> wrote:

>> Simon Tooke has updated the pull request incrementally with one additional commit since the last revision:
>> 
>>   fix overly restrictive test
>
> I am not a Reviewer but this looks mostly good to me.

@SoniaZaldana thank you for looking at my PR.  I've fixed the issues you've found.  

As this "reopens" the PR, I also took the opportunity to fix an issue when no filename is given to the '-F' parameter in System.dump_map (i.e. when "jcmd 999 System.dump_map -F" is issued, for example).  Previously, the Windows implementation would terminate the VM (an error in fileStream::fileStream(const char*) presumably) and crash jcmd with a java stack trace.  (this should be addressed in a separate PR).

Now an error message is sent to the jcmd user: 
`C:\work>.\jdk\build\fastdebug\jdk\bin\jcmd TT System.dump_map -F=`
`13008:`
`filename is empty or not specified.  No file written`

I'm sorry to ask, @tstuefe , but this impacts the Linux implementation too, so you might want to look at this new behaviour in SystemDumpMapDCmd::execute() (diagnosticComand.cpp:1196)

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

PR Comment: https://git.openjdk.org/jdk/pull/20597#issuecomment-2343726145