[dpdk-dev] [PATCH v4] Policy Based Power Control for Guest
david.hunt at intel.com
Wed Oct 4 11:15:06 CEST 2017
This patchset adds the facility for a guest VM to send a policy down to the
host that will allow the host to scale up/down cpu frequencies
depending on the policy criteria independently of the DPDK app running in
the guest. This differs from the previous vm_power implementation where
individual scale up/down requests were send from the guest to the host via
V4 patchset changes:
* None, replying into correct email thread. V3 was a reply to the
turbo patch set, should have been inband policy power patchset.
V3 patchset changes:
* Changed to using is_same_ether_addr() instead of looping through
the mac address bytes to compare them.
* Tweaked some comments and working in the i40e patch after review.
* Added a patch to the set to add new i40e function to map file, so
as to allow shared library builds. The power library API needs a cleanup
in next release, so will add API/ABI warning for this cleanup in a
V2 patchset changes:
* Removed API's in ethdev layer.
* Now just a single new API in the i40e driver for mapping VF MAC to
* Moved new function from rte_rxtx.c to rte_pmd_i40e.c
* Removed function for reading i40e register, moved to using the
standard stats API.
* Renamed i40e function to rte_pmd_i40e_query_vfid_by_mac
* Cleaned up policy generation code.
It's a modification of the vm_power_manager app that runs in the host, and
the guest_vm_power_app example app that runs in the guest. This allows the
guest to send down a policy to the host via virtio-serial, which then allows
the host to scale up/down based on the criteria in the policy, resulting in
quicker scale up/down than individual requests coming from the guest.
It also means that the DPDK application running in the guest does not need
to be modified in any way, it is unaware that it's cores are being scaled
up/down, reducing the effort in implementing a power-aware infrastructure.
The usage model is as follows:
1. Set up the VF's and assign to the guest in the usual way.
2. run vm_power_manager on the host, creating a channel to the guest.
3. Start the guest_vm_power_mgr app on the guest, which establishes
a virtio-serial channel to the host.
4. Send down the profile for the guest using the "send_profile now" command.
There is an example profile hard-coded into guest_vm_power_mgr.
5. Stop the guest_vm_power_mgr and run your normal power-unaware application.
6. Send traffic into the VFs at varying traffic rates.
Observe the frequency change on the host (turbostat -i 1)
The sequence of code changes are as follows:
A new function has been aded to the i40e driver to allow mapping of
a VF MAC to VF index.
Next we make an addition to librte_power that adds an extra command to allow
the passing of a policy structure from the guest to the host. This struct
contains information like busy/quiet hour, packet throughput thresholds, etc.
The next addition adds functionality to convert the virtual CPU (vcpU0 IDs to
physical CPU (pcpu) IDs so that the host can scale up/down the cores used
in the guest.
The remaining patches are functionality to process the policy, and take action
when the relevant trigger occurs to cause a frequency change.
[1/9] net/i40e: add API to convert VF MAC to VF id
[2/9] lib/librte_power: add extra msg type for policies
[3/9] examples/vm_power_mgr: add vcpu to pcpu mapping
[4/9] examples/vm_power_mgr: add scale to medium freq fn
[5/9] examples/vm_power_mgr: add policy to channels
[6/9] examples/vm_power_mgr: add port initialisation
[7/9] power: add send channel msg function to map file
[8/9] examples/guest_cli: add send policy to host
[9/9] examples/vm_power_mgr: set MAC address of VF
More information about the dev