RFR (XS) 8213182: Minimal VM build failure after JDK-8212200 (assert when shared java.lang.Object is redefined by JVMTI agent)

David Holmes david.holmes at oracle.com
Wed Oct 31 10:33:38 UTC 2018


Looks good Aleksey!

Thanks,
David

On 31/10/2018 8:19 PM, Aleksey Shipilev wrote:
> Bug:
>    https://bugs.openjdk.java.net/browse/JDK-8213182
> 
> Minimal VM is built without JVMTI, and so lacks the definitions for new methods. These need to be
> stubbed out with the usual macros that handle non-JVMTI cases.
> 
> Fix:
> 
> # HG changeset patch
> # User shade
> # Date 1540980808 -3600
> #      Wed Oct 31 11:13:28 2018 +0100
> # Node ID 4b473b022238837a05637d3afebd38edaad330a4
> # Parent  9341b077bd5530734202134a90fb34d2dcc31ef2
> 8213182: Minimal VM build failure after JDK-8212200 (assert when shared java.lang.Object is
> redefined by JVMTI agent)
> Reviewed-by: XXX
> 
> diff -r 9341b077bd55 -r 4b473b022238 src/hotspot/share/prims/jvmtiExport.hpp
> --- a/src/hotspot/share/prims/jvmtiExport.hpp   Wed Oct 31 09:30:24 2018 +0100
> +++ b/src/hotspot/share/prims/jvmtiExport.hpp   Wed Oct 31 11:13:28 2018 +0100
> @@ -328,8 +328,8 @@
>       JVMTI_ONLY(return _should_post_class_file_load_hook);
>       NOT_JVMTI(return false;)
>     }
> -  static bool is_early_phase();
> -  static bool has_early_class_hook_env();
> +  static bool is_early_phase() NOT_JVMTI_RETURN_(false);
> +  static bool has_early_class_hook_env() NOT_JVMTI_RETURN_(false);
>     // Return true if the class was modified by the hook.
>     static bool post_class_file_load_hook(Symbol* h_name, Handle class_loader,
>                                           Handle h_protection_domain,
> 
> 
> Testing: x86_64 Minimal VM builds
> 
> Thanks,
> -Aleksey
> 


More information about the hotspot-runtime-dev mailing list