RFR: 8193513: add support for printing a stack trace on class loading [v7]

[Doug Simon]

On Wed, 5 Jul 2023 06:48:25 GMT, David Holmes <dholmes at openjdk.org> wrote:

>> src/hotspot/share/oops/instanceKlass.cpp line 3867:
>> 
>>> 3865:         // We have printed the native stack in platform-specific code,
>>> 3866:         // so nothing else to do in this case.
>>> 3867:       } else {
>> 
>> Why did you do this? Did you observe problems on Windows? This is too low-level for this code. If we need to do this then it needs to be pushed down to a lower-level and encapsulated more cleanly. We do not do this in `JavaThread::print_jni_stack()`.
>> 
>> I need to revoke my approval while this is resolved.
>
> I'm also investigating the details of the stack printing code - use of `os::platform_print_native_stack` is very inconsistent.

I was under the impression that the only way to get native stack printing on windows is with `os::platform_print_native_stack`. I'm trying to set up a Windows machine to test this assumption now.

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

PR Review Comment: https://git.openjdk.org/jdk/pull/14553#discussion_r1252866376