Come to think of it, why not host the official mercurial SVN mirror repo on
Bitbucket for the same reason (lots of nice hg features on Bitbucket -- nicer
than hgweb/www.open-mpi.org's hosting).
There's no need to delete the original mercurial SVN mirror on www.open-mpi.org
(because I know of at
Chris --
I see that this was fixed on the trunk. Does it need to be CMR'ed to v1.7?
On Aug 15, 2012, at 11:56 PM, Open MPI wrote:
> #3161: openib udcm finalize causes open async thread to abort
> ---+---
> Reporter: cyeoh | Owner: ompi-rm1.7
>
Mike / Yevgeny --
Can you comment on what is going on here? It would be really good to
understand exactly what these 2 MLX4 parameters are (e.g., why you suggested
increasing one and not the other), and why there would be differences in
registering small numbers of large chunks of contiguous m
