On 04/23/2013 05:11 PM, Paolo Bonzini wrote:
Il 23/04/2013 22:15, Michael R. Hines ha scritto:
Also, don't be surprised if not everyone reviews the entire series;
sometimes reviewers like myself focus only on the portion of the series
that interacts with my current interests (I tend to review anything QMP,
because I want to make sure the design will be sane for libvirt
interaction, while overlooking things like migration internals because
they are black box ops to libvirt if the interface was sane).
Yes, Paolo has done a fantastic job of reviewing the internals.
But the internals have changed while I wasn't looking. :)
I'm very close to giving my Reviewed-by, I would just prefer to have
this last change reverted.
Paolo
The only new change was QEMUFileOps "qemu_get_max_size()" (patch #4).
During this patch series, I stopped using my 40G hardware, but when I
switched
back to my 40G hardware, I noticed that the max_size calculation was
artificially
limiting the max throughput achievable by RDMA.
Everything else is the same as we agreed in previous versions.
- Michael
