RFR (S) 8173361: various crashes in JvmtiExport::post_compiled_method_load

[Chris Plummer]

Hi Coleen,

Is it ok to end up missing some CompiledMethodLoad events? The spec says:

"Sent when a method is compiled and loaded into memory by the VM. If it 
is unloaded, the CompiledMethodUnload event is sent. If it is moved, the 
CompiledMethodUnload event is sent, followed by a new CompiledMethodLoad 
event. Note that a single method may have multiple compiled forms, and 
that this event will be sent for each form. "

So a method was still "compiled and loaded into memory", right? We just 
didn't get the event out before it was too late. Is the 
CompiledMethodUnload still sent?

thanks,

Chris

On 11/14/19 5:15 PM, coleen.phillimore at oracle.com wrote:
> Summary: Don't post information which uses metadata from unloaded 
> nmethods
>
> Tested tier1-3 and 100 times with test that failed (reproduced failure 
> without the fix).
>
> open webrev at http://cr.openjdk.java.net/~coleenp/2019/8173361.01/webrev
> bug link https://bugs.openjdk.java.net/browse/JDK-8173361
>
> Thanks,
> Coleen