RFR: 8078193: BACKOUT: Rename and clean up the ParGCAllocBuffer class

Kim Barrett kim.barrett at oracle.com
Tue Apr 21 13:46:51 UTC 2015

On Apr 21, 2015, at 4:09 AM, Bengt Rutisson <bengt.rutisson at oracle.com> wrote:
> Doing hg backout does not apply cleanly for me.
> $ hg backout -r 8148
> reverting src/share/vm/gc_implementation/g1/g1Allocator.cpp
> reverting src/share/vm/gc_implementation/g1/g1Allocator.hpp
> reverting src/share/vm/gc_implementation/g1/g1CollectedHeap.hpp
> reverting src/share/vm/gc_implementation/parNew/parNewGeneration.cpp
> reverting src/share/vm/gc_implementation/parNew/parNewGeneration.hpp
> adding src/share/vm/gc_implementation/shared/parGCAllocBuffer.cpp
> adding src/share/vm/gc_implementation/shared/parGCAllocBuffer.hpp
> adding src/share/vm/gc_implementation/shared/parGCAllocBuffer.inline.hpp
> removing src/share/vm/gc_implementation/shared/plab.cpp
> removing src/share/vm/gc_implementation/shared/plab.hpp
> removing src/share/vm/gc_implementation/shared/plab.inline.hpp
> reverting src/share/vm/precompiled/precompiled.hpp
> reverting src/share/vm/runtime/globals.hpp
> merging src/share/vm/runtime/globals.hpp
> merging src/share/vm/runtime/globals.hpp failed!
> 142 files updated, 0 files merged, 1 files removed, 1 files unresolved
> use 'hg resolve' to retry unresolved file merges
> There is a conflict in globals.hpp. But the patch file provided in the webrev:
> http://cr.openjdk.java.net/~kbarrett/8078193/webrev.00/hotspot.changeset

Hm, I don’t get any such unresolved merge.  I do get a merge now, which I don’t remember
getting before.  That might be because there have been pushes since I cloned, or because
I’m mis-rememebering.

> applies cleanly for me and does the right thing to globals.hpp.
> So, I think the webrev looks fine. Reviewed.

Thanks for looking at this.

> Will you push the fix for "Rename and clean up the ParGCAllocBuffer class" with the correct bug number at the same time? Would be nice to minimize the window where the repo is in an intermediate state.

I’m not sure what you mean by pushing them “at the same time”.  My intent was to push this backout, and when it completed then setup and push the corrected fix, expecting to complete sometime this afternoon.  Are you suggesting something different?

More information about the hotspot-gc-dev mailing list