DPDK patches and discussions
 help / color / mirror / Atom feed
From: "Hunt, David" <david.hunt@intel.com>
Cc: dev@dpdk.org, Marcin Hajkowski <marcinx.hajkowski@intel.com>
Subject: Re: [dpdk-dev] [PATCH v2 3/3] examples/power_guest: send request for specified core capabilities
Date: Mon, 30 Sep 2019 14:04:16 +0100	[thread overview]
Message-ID: <d938bf3b-627a-051f-0bc3-44c8659476e4@intel.com> (raw)
In-Reply-To: <20190930130043.2885-4-david.hunt@intel.com>


On 30/09/2019 14:00, David Hunt wrote:
> From: Marcin Hajkowski <marcinx.hajkowski@intel.com>
>
> Send request to power manager for core id provided
> by user to get related capabilities.
>
> Signed-off-by: Marcin Hajkowski <marcinx.hajkowski@intel.com>
> Tested-by: David Hunt <david.hunt@intel.com>
> Acked-by: Lee Daly <lee.daly@intel.com>
> ---
>   .../guest_cli/vm_power_cli_guest.c            | 119 +++++++++++++++++-
>   1 file changed, 117 insertions(+), 2 deletions(-)
>
--snip--

Some of the steps to test included blow:

In the Host, enable queries via the virtio-serial interface from the
guest to the host:

vmpower> set_query ubuntu3 enable

In an SST-BF enabled host, with 6 high priority cores per
socket, query the list of high prioroty cores with the
sst_bf.py python script
(https://github.com/intel/commspowermanagement):

root@silpixa00399952:~# sst_bf.py -l
1,6,7,8,9,16,21,26,27,28,29,30
0x7c2103c2

In the host, query the cores used by the VM:

vmpower> show_vm ubuntu3
...snip...
Virtual CPU(s): 8
   [0]: Physical CPU 28
   [1]: Physical CPU 29
   [2]: Physical CPU 30
   [3]: Physical CPU 31
   [4]: Physical CPU 32
   [5]: Physical CPU 33
   [6]: Physical CPU 34
   [7]: Physical CPU 35

So we now know that cores 28,29 and 30 should show up as high priority
in the VM. So, from the VM side, we use the new 'query_cpu_caps' command
to see the capabilities of each core:

vmpower(guest)> query_cpu_caps all
Capabilities of [0] vcore are: turbo possibility: 1, is priority core: 1.
Capabilities of [1] vcore are: turbo possibility: 1, is priority core: 1.
Capabilities of [2] vcore are: turbo possibility: 1, is priority core: 1.
Capabilities of [3] vcore are: turbo possibility: 1, is priority core: 0.
Capabilities of [4] vcore are: turbo possibility: 1, is priority core: 0.
Capabilities of [5] vcore are: turbo possibility: 1, is priority core: 0.
Capabilities of [6] vcore are: turbo possibility: 1, is priority core: 0.
Capabilities of [7] vcore are: turbo possibility: 1, is priority core: 0.

As expected, vcore 0,1,2 (pcore 28,29,30) are high priority cores.

Tested-by: David Hunt <david.hunt@intel.com>




  reply	other threads:[~2019-09-30 13:04 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-05-30 16:15 [dpdk-dev] [PATCH 0/3] Core capabilities query Hajkowski
2019-05-30 16:15 ` [dpdk-dev] [PATCH 1/3] power: add new packet type for capabilities Hajkowski
2019-09-30 10:55   ` Daly, Lee
2019-09-30 13:00   ` [dpdk-dev] [PATCH v2 0/3] Core capabilities query David Hunt
2019-09-30 13:00     ` [dpdk-dev] [PATCH v2 1/3] power: add new packet type for capabilities David Hunt
2019-09-30 13:00     ` [dpdk-dev] [PATCH v2 2/3] examples/power_manager: send cpu capabilities on vm request David Hunt
2019-09-30 13:00     ` [dpdk-dev] [PATCH v2 3/3] examples/power_guest: send request for specified core capabilities David Hunt
2019-09-30 13:04       ` Hunt, David [this message]
2019-10-27 20:16     ` [dpdk-dev] [PATCH v2 0/3] Core capabilities query Thomas Monjalon
2019-05-30 16:15 ` [dpdk-dev] [PATCH 2/3] examples/power_manager: send cpu capabilities on vm request Hajkowski
2019-09-30 10:56   ` Daly, Lee
2019-05-30 16:15 ` [dpdk-dev] [PATCH 3/3] examples/power_guest: send request for specified core capabilities Hajkowski
2019-09-30 10:54   ` Daly, Lee
2019-09-30 12:54   ` Daly, Lee
2019-07-04 19:58 ` [dpdk-dev] [PATCH 0/3] Core capabilities query Thomas Monjalon

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=d938bf3b-627a-051f-0bc3-44c8659476e4@intel.com \
    --to=david.hunt@intel.com \
    --cc=dev@dpdk.org \
    --cc=marcinx.hajkowski@intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).