RFR(s): 8218041: Assorted wrong/missing includes

[David Holmes]

Still good to me too. :)

David

On 30/01/2019 11:30 pm, Robbin Ehn wrote:
> Hi, here is v3.
> 
> http://cr.openjdk.java.net/~rehn/8218041/v03/inc/
> http://cr.openjdk.java.net/~rehn/8218041/v03/
> 
> Passes same compilations and t1.
> 
> /Robbin
> 
> On 2019-01-30 09:21, Robbin Ehn wrote:
>> Hi all, please review.
>>
>> Code:
>> http://cr.openjdk.java.net/~rehn/8218041/webrev/
>> Issue:
>> https://bugs.openjdk.java.net/browse/JDK-8218041
>>
>> After fixing these includes, there was a circular dependency via 
>> shenandoah
>> code. I moved try_cancel_gc to cpp where the only use was. So it never 
>> should
>> had been in the inline header in the first place.
>>
>> I listed why the include is needed below.
>>
>> Tier 1 and no pre-compiled.
>>
>> FYI: I was investigating why Handle::Handle(Thread*,oop) was not inlined.
>> gcc complains there being a local comdat symbol, forcing it to be 
>> inlined or
>> using clang there is no issue. So it looks like a gcc bug both in 7.3 
>> and 8.2.
>>
>> Thanks, Robbin
>>
>> src/hotspot/share/aot/aotLoader.cpp
>> runtime/os.inline.hpp      for os::dll_unload
>>
>> src/hotspot/share/c1/c1_Runtime1.cpp
>> runtime/handles.inline.hpp for Handle(Thread*, oop)
>>
>> src/hotspot/share/gc/z/zFuture.inline.hpp
>> runtime/interfaceSupport.inline.hpp not used.
>>
>> src/hotspot/share/prims/nativeLookup.cpp
>> runtime/os.inline.hpp      for os::dll_unload
>>
>> src/hotspot/share/runtime/handles.hpp
>> Forward declaration            Thread
>>
>> src/hotspot/share/runtime/handles.inline.hpp
>> runtime/thread.hpp       for Thread::current
>> oops/oop.inline.hpp        for oopDesc::is_a
>> oops/metadata.hpp          for is_valid
>>
>> src/hotspot/share/runtime/semaphore.inline.hpp
>> runtime/thread.hpp         for osthread
>>
>> src/hotspot/share/runtime/vframe.cpp
>> runtime/thread.inline.hpp  for JavaThread::class_to_be_initialized