* Alex Snast <asn...@gmail.com> wrote:
> What's the benefit of having that diverge check script as on every commit
> you'll
> either add the new stuff to tools/include/linux/rbtree.h or add an exception
> to
> that script as in rb_link_node_rcu case.
The benefit is that things do not diverge - diff or md5sum is fast enough.
I don't think exceptions should be added, we should find ways to make the files
100% identical.
> And are you suggesting to check for divergence for everything under tools/
> include/linux recursively?
No, on a case by case basis, for bigger 'facility files' we copied from the
kernel
source.
rbtree.h and rbtree.c certainly applies: for example we've got interval rbtree
additions in -tip that would be nice to merge upstream:
tools/kvm/include/kvm/rbtree-interval.h
tools/kvm/util/rbtree-interval.c
... but which have bitrotten meanwhile, making the copy out of sync and harder
to
merge. Making sure the files are properly upstreamed during build time makes
sure
things don't get out of sync.
Thanks,
Ingo
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
