Re: [Qemu-devel] [PULL v3 7/7] rdma: add documentation

[Eric Blake <eblake@redhat.com>]

[-- Attachment #1: Type: text/plain, Size: 1562 bytes --]

On 04/16/2013 10:20 PM, mrhines@linux.vnet.ibm.com wrote:

> +Experimental: Next, decide if you want dynamic page registration.
> +For example, if you have an 8GB RAM virtual machine, but only 1GB 
> +is in active use, then disabling this feature will cause all 8GB to 
> +be pinned and resident in memory. This feature mostly affects the
> +bulk-phase round of the migration and can be disabled for extremely
> +high-performance RDMA hardware using the following command:
> +
> +QEMU Monitor Command:
> +$ migrate_set_capability chunk_register_destination off # enabled by default

This needs updating to match the QMP capability name (or else the HMP
command needs some glue logic to map chunk_register_destination into the
QMP x-chunk-register-destination as a convenience to the user).

> +
> +TODO:
> +=====
> +1. Currently, cgroups swap limits and infinband memory pinning
> +   are not compatible with each other. We have no plans to unregister
> +   memory *during* the live migration as the benefit of doing so is not clear.
> +2. Use of the recent /proc/<pid>/pagemap would likely speed up
> +   the use of KSM and ballooning while using RDMA.
> +3. Also, some form of balloon-device usage tracking would also
> +   help alleviate some issues.

Maybe also mention that the protocol will be renamed from -incoming
x-rdma:... to -incoming rdma:... when it has been through more testing
to stabilize the protocol.

-- 
Eric Blake   eblake redhat com    +1-919-301-3266
Libvirt virtualization library http://libvirt.org


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 621 bytes --]