Remove unimplemented InlineTree constructor definition from parse.hpp

Vladimir Kozlov vladimir.kozlov at oracle.com
Tue Mar 17 20:07:06 UTC 2020


Yes, we forgot to remove this lines when we removed implementation:

http://hg.openjdk.java.net/jdk9/jdk9/hotspot/rev/17ec2d5c43e8

Change looks good and it is trivial. I will sponsor it:

https://bugs.openjdk.java.net/browse/JDK-8241122

Thanks,
Vladimir K

On 3/17/20 12:43 PM, Charlie Gracie wrote:
> When looking through the code for InlineTree I noticed an unimplemented constructor definition. I spent a few mins looking for the
> implementation and it's uses so I figure it is best to get it removed so others do not spend time as well.
> 
> I tested compilation of release, fastdebug and slowdebug builds on my Mac and also verified the JVM still starts.
> 
> Here is the patch for tip:
> diff --git a/src/hotspot/share/opto/parse.hpp b/src/hotspot/share/opto/parse.hpp
> index 28e440f586..adbb5ba355 100644
> --- a/src/hotspot/share/opto/parse.hpp
> +++ b/src/hotspot/share/opto/parse.hpp
> @@ -107,9 +107,6 @@ public:
>     static InlineTree* build_inline_tree_root();
>     static InlineTree* find_subtree_from_root(InlineTree* root, JVMState* jvms, ciMethod* callee);
>   
> -  // For temporary (stack-allocated, stateless) ilts:
> -  InlineTree(Compile* c, ciMethod* callee_method, JVMState* caller_jvms, float site_invoke_ratio, int max_inline_level);
> -
>     // See if it is OK to inline.
>     // The receiver is the inline tree for the caller.
>     //
> 
> Thanks,
> Charlie
> 


More information about the hotspot-compiler-dev mailing list