RFR: 8332124: Jcmd processing should accept the "help" sub option as command argument [v5]

Kevin Walls kevinw at openjdk.org
Wed Jul 3 12:08:21 UTC 2024


On Tue, 2 Jul 2024 15:21:49 GMT, Sonia Zaldana Calles <szaldana at openjdk.org> wrote:

>> Hi all, 
>> 
>> This PR addresses [8332124](https://bugs.openjdk.org/browse/JDK-8332124) enabling jcmd to accept "help" as an argument to subcommands. 
>> 
>> Testing: 
>> - [x] Verified running `jcmd 4711 VM.metaspace help` works along with other subcommands. 
>> - [x] Added test case passes. 
>> 
>> Thanks, 
>> Sonia
>
> Sonia Zaldana Calles has updated the pull request incrementally with one additional commit since the last revision:
> 
>   Making enabling help more restrictive. Will not accept -help

Hi,

1.
The JBS issue and PR title need an update. Maybe something like:

  Jcmd should recognise options that look like requests for help


2.
I would encourage recognising "-help" in this change.  This is contrary to Thomas' suggestion, and I see the clash with posix style --options, but we already have many many single-minus-prefixed options.  This change as I see it is an undocumented attempt to help lost users.  If we start a migration to posix-style, much effort will be needed so this clash, which could be fixed then, will seem insignificant.


3.
I want to mention the alternative to this change: print the command help whenever showing an error.  The downside would be that it can be quite long, and maybe you don't want to see several screens of help when you mistype one option to jcmd JFR.start...
I think we should complete what we have here already, but we should note that there are alternatives.


4.
sun/tools/jcmd/TestJcmdSubcommandHelp.java
The summary at line 28 needs updating, whether you add "-help" or not.

Would be great to test "jcmd VM.command -hello" and check it does not show help.  I added one and tried it out, will paste it in here somehow.


Thanks!

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

PR Comment: https://git.openjdk.org/jdk/pull/19776#issuecomment-2205920976


More information about the hotspot-runtime-dev mailing list