harmony-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Pavel Rebriy" <pavel.reb...@gmail.com>
Subject Re: [drlvm][stability][winx64] finalization tests irregular failure
Date Fri, 30 Mar 2007 12:49:37 GMT
I attached patch which reverts revision 521690.
This patch repairs DRLVM, but could break gcv5.

On 29/03/07, Pavel Rebriy <pavel.rebriy@gmail.com> wrote:
>
> I've got revision 521612 of DRLVM and added all commits after it except
> 521690.
> Build works fine.
> I could check 521690 revision but it doesn't make sense because there is a
> lot of changes for WinX64 after it.
>
> On 29/03/07, Xiao-Feng Li <xiaofeng.li@gmail.com> wrote:
> >
> > Pavel, did you mean revision 521689 has no problem? Would you like to
> > check the revision of 521690? to use the code that doesn't include any
> > new checkins after 521690?
> >
> > Thanks,
> > xiaofeng
> >
> > On 3/29/07, Pavel Rebriy < pavel.rebriy@gmail.com> wrote:
> > > Xiao Feng,
> > >
> > > There is list of files in vm which influence by the patch in
> > HARMONY-3485:
> > >
> > > Index: vm/include/open/gc.h
> > > Index: vm/include/open/vm_gc.h
> > > Index: vm/include/open/hythread_ext.h
> > > Index: vm/thread/src/hythr.def
> > > Index: vm/thread/src/thread_native_condvar.c
> > > Index: vm/vmcore/include/finalizer_thread.h
> > > Index: vm/vmcore/include/ref_enqueue_thread.h
> > > Index: vm/vmcore/include/finalize.h
> > > Index: vm/vmcore/src/jni/jni.cpp
> > > Index: vm/vmcore/src/init/finalize.cpp
> > > Index: vm/vmcore/src/init/vm_init.cpp
> > > Index: vm/vmcore/src/init/finalizer_thread.cpp
> > > Index: vm/vmcore/src/init/ref_enqueue_thread.cpp
> > >
> > > If I role back the changes, tests pass.
> > >
> > > On 29/03/07, Xiao-Feng Li < xiaofeng.li@gmail.com> wrote:
> > > >
> > > > Pavel, the commit H3485 has nothing to do with the execution if GCv5
> > > > is not used. Please check the few lines of the commit in vmcore,
> > it's
> > > > only a declaration of an exported function, no real functionality is
> > > > introduced. I'd like to know how you came to your conclusion. For
> > > > example, have you tried with Harmony with revision before 521690?
> > > >
> > > > Thanks,
> > > > xiaofeng
> > > >
> > > > On 3/29/07, Pavel Rebriy <pavel.rebriy@gmail.com> wrote:
> > > > > Stability on WinX64 was drastically reduced after HARMONY-3485 was
> >
> > > > committed
> > > > > at revision 521690. I created a JIRA issue HARMONY-3523 [1].
> > > > >
> > > > > P.S. BTW why HARMONY-3485 is still open if the patch is already
> > > > committed?
> > > > >
> > > > > [1] https://issues.apache.org/jira/browse/HARMONY-3523
> > > > >
> > > > > --
> > > > > Best regards,
> > > > > Pavel Rebriy
> > > > >
> > > >
> > >
> > >
> > > --
> > > Best regards,
> > > Pavel Rebriy
> > >
> >
>
>
>
> --
> Best regards,
> Pavel Rebriy
>



-- 
Best regards,
Pavel Rebriy

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message