DPDK patches and discussions
 help / color / mirror / Atom feed
From: David Marchand <david.marchand@redhat.com>
To: "Hunt, David" <david.hunt@intel.com>,
	Feifei Wang <feifei.wang2@arm.com>,
	 Ruifeng Wang <ruifeng.wang@arm.com>
Cc: dev@dpdk.org, nd@arm.com, Thomas Monjalon <thomas@monjalon.net>
Subject: Re: [PATCH v1 3/3] examples/l3fwd-power: enable PMD power mgmt on Arm
Date: Mon, 3 Oct 2022 09:12:13 +0200	[thread overview]
Message-ID: <CAJFAV8yDparS_jXRBdx0mxhCWNZi-hWC=WmY5sE+GLs5SJH6wg@mail.gmail.com> (raw)
In-Reply-To: <9039d076-aad2-d741-5724-a163085b470d@intel.com>

On Mon, Aug 29, 2022 at 2:48 PM Hunt, David <david.hunt@intel.com> wrote:
> On 25/08/2022 07:42, Feifei Wang wrote:
> > For Arm aarch, power monitor uses WFE instruction to enable, which can
> > not exit automatically within the time limit. This means
> > 'rte_power_monitor_wakeup' API needs to be called to wake up sleep cores
> > if there is no store operation to monitored address.
> >
> > Furthermore, we disable power monitor feature on the main core so that it
> > can be used to wake up other sleeping cores when it receives SIGINT
> > siginal.
> >
> > Signed-off-by: Feifei Wang <feifei.wang2@arm.com>
> > Reviewed-by: Ruifeng Wang <ruifeng.wang@arm.com>
> > ---
> >   examples/l3fwd-power/main.c | 30 +++++++++++++++++++++++++++++-
> >   1 file changed, 29 insertions(+), 1 deletion(-)
> >
> > diff --git a/examples/l3fwd-power/main.c b/examples/l3fwd-power/main.c
> > index 887c6eae3f..2bd0d700f0 100644
> > --- a/examples/l3fwd-power/main.c
> > +++ b/examples/l3fwd-power/main.c
> > @@ -432,8 +432,16 @@ static void
> >   signal_exit_now(int sigtype)
> >   {
> >
> > -     if (sigtype == SIGINT)
> > +     if (sigtype == SIGINT) {
> > +#if defined(RTE_ARCH_ARM64)

Having a arch specific behavior in the application shows that there is
something wrong either in the API, or in the Arm implementation of the
API.
I don't think this is a good solution.

Can't we find a better alternative? By changing the API probably?


> > +     /**
> > +      * wake_up api does not need input parameter on Arm,
> > +      * so 0 is meaningless here.
> > +      */
> > +             rte_power_monitor_wakeup(0);
> > +#endif
> >               quit_signal = true;
> > +     }
> >
> >   }
> >
> > @@ -2885,6 +2893,25 @@ main(int argc, char **argv)
> >                                               "Error setting scaling freq max: err=%d, lcore %d\n",
> >                                                       ret, lcore_id);
> >
> > +#if defined(RTE_ARCH_ARM64)
> > +                             /* Ensure the main lcore does not enter the power-monitor state,
> > +                              * so that it can be used to wake up other lcores on ARM.
> > +                              * This is due to WFE instruction has no timeout wake-up mechanism,
> > +                              * and if users want to exit actively, the main lcore is needed
> > +                              * to send SEV instruction to wake up other lcores.
> > +                              */
> > +                             unsigned int main_lcore = rte_get_main_lcore();
> > +                             if (lcore_id != main_lcore ||
> > +                                             pmgmt_type != RTE_POWER_MGMT_TYPE_MONITOR) {
> > +                                     ret = rte_power_ethdev_pmgmt_queue_enable(
> > +                                                     lcore_id, portid, queueid,
> > +                                                     pmgmt_type);
> > +                                     if (ret < 0)
> > +                                             rte_exit(EXIT_FAILURE,
> > +                                                     "rte_power_ethdev_pmgmt_queue_enable: err=%d, port=%d\n",
> > +                                                     ret, portid);
> > +                             }
> > +#else
> >                               ret = rte_power_ethdev_pmgmt_queue_enable(
> >                                               lcore_id, portid, queueid,
> >                                               pmgmt_type);
> > @@ -2892,6 +2919,7 @@ main(int argc, char **argv)
> >                                       rte_exit(EXIT_FAILURE,
> >                                               "rte_power_ethdev_pmgmt_queue_enable: err=%d, port=%d\n",
> >                                                       ret, portid);
> > +#endif
> >                       }
> >               }
> >       }
>
>
> Hi Feifei,
>
> Acked-by: David Hunt <david.hunt@intel.com>


-- 
David Marchand


  reply	other threads:[~2022-10-03  7:12 UTC|newest]

Thread overview: 39+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-08-25  6:42 [PATCH v1 0/3] Enable PMD power management " Feifei Wang
2022-08-25  6:42 ` [PATCH v1 1/3] eal: add 8 bits case for wait scheme Feifei Wang
2022-08-25  6:42 ` [PATCH v1 2/3] eal: add power mgmt support on Arm Feifei Wang
2022-08-25  6:42 ` [PATCH v1 3/3] examples/l3fwd-power: enable PMD power mgmt " Feifei Wang
2022-08-29 12:48   ` Hunt, David
2022-10-03  7:12     ` David Marchand [this message]
2022-10-11  7:56       ` 回复: " Feifei Wang
2022-10-20 20:41         ` Thomas Monjalon
2022-10-27  9:38           ` 回复: " Feifei Wang
2022-10-20 22:09   ` Stephen Hemminger
2022-10-27  9:43     ` 回复: " Feifei Wang
2022-11-07  7:04 ` [PATCH v2 0/3] Enable PMD power management " Feifei Wang
2022-11-07  7:04   ` [PATCH v2 1/3] eal: add 8 bits case for wait scheme Feifei Wang
2022-11-07  7:04   ` [PATCH v2 2/3] eal: add power mgmt support on Arm Feifei Wang
2022-11-07  7:04   ` [PATCH v2 3/3] examples/l3fwd-power: enable PMD power monitor " Feifei Wang
2022-11-07 16:01     ` Stephen Hemminger
2022-11-08  3:25       ` 回复: " Feifei Wang
2022-11-11  7:26 ` [PATCH v3 0/3] Enable PMD power management " Feifei Wang
2022-11-11  7:26   ` [PATCH v3 1/3] eal: add 8 bits case for wait scheme Feifei Wang
2022-11-11  7:26   ` [PATCH v3 2/3] eal: add power mgmt support on Arm Feifei Wang
2022-11-11  7:26   ` [PATCH v3 3/3] examples/l3fwd-power: enable PMD power monitor " Feifei Wang
2022-11-11  8:22     ` Thomas Monjalon
2022-11-11 10:21       ` 回复: " Feifei Wang
2022-11-11 10:20 ` [PATCH v4 0/4] Enable PMD power management " Feifei Wang
2022-11-11 10:20   ` [PATCH v4 1/4] eal: add 8 bits case for wait scheme Feifei Wang
2022-11-11 10:20   ` [PATCH v4 2/4] eal: add power mgmt support on Arm Feifei Wang
2022-11-11 10:20   ` [PATCH v4 3/4] power: add power monitor support check Feifei Wang
2022-11-11 10:20   ` [PATCH v4 4/4] examples/l3fwd-power: add power monitor wake up API Feifei Wang
2022-12-14  8:14 ` [PATCH v5 0/2] Enable PMD power management on Arm Feifei Wang
2022-12-14  8:14   ` [PATCH v5 1/2] eal: add 8 bits case for wait scheme Feifei Wang
2022-12-14  8:14   ` [PATCH v5 2/2] eal: add power mgmt support on Arm Feifei Wang
2023-02-17 16:23     ` Stephen Hemminger
2023-02-20  1:56       ` 回复: " Feifei Wang
2023-02-17  8:26   ` 回复: [PATCH v5 0/2] Enable PMD power management " Feifei Wang
2023-02-20  8:51 ` [PATCH v6 " Feifei Wang
2023-02-20  8:51   ` [PATCH v6 1/2] eal: add 8 bits case for wait scheme Feifei Wang
2023-02-20  8:51   ` [PATCH v6 2/2] eal: add power mgmt support on Arm Feifei Wang
2023-02-20 12:07   ` [PATCH v6 0/2] Enable PMD power management " David Marchand
2023-02-21  2:37     ` 回复: " Feifei Wang

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='CAJFAV8yDparS_jXRBdx0mxhCWNZi-hWC=WmY5sE+GLs5SJH6wg@mail.gmail.com' \
    --to=david.marchand@redhat.com \
    --cc=david.hunt@intel.com \
    --cc=dev@dpdk.org \
    --cc=feifei.wang2@arm.com \
    --cc=nd@arm.com \
    --cc=ruifeng.wang@arm.com \
    --cc=thomas@monjalon.net \
    /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).