RFR: 8319873: Add windows implementation for jcmd System.map and System.dump_map [v10]

Simon Tooke stooke at openjdk.org
Wed Sep 11 17:40:11 UTC 2024


On Wed, 11 Sep 2024 17:26:15 GMT, Thomas Stuefe <stuefe at openjdk.org> wrote:

>> I was thinking about this - my original intent was to be fatal only on a debug build, not release.  I have added a message in the output, and changed these to warnings (for both release and debug)
>
> I would prefer an assert here. It will crash in debug builds, which is fine, since if it crashes it is either a bug or an hitherto unknown state. In any case something that should be fixed.

I believe we can all agree that fatal() was too harsh.
Is there a macro that is the equivalent of "assert(false, ....)"?  
my current unchecked in code has 'warning()' which will log the problem.  I am happy to replace this with assert().
@tstuefe How do you feel about a message in the output even in release builds?

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

PR Review Comment: https://git.openjdk.org/jdk/pull/20597#discussion_r1755223901


More information about the serviceability-dev mailing list