DPDK patches and discussions
 help / color / mirror / Atom feed
From: "Dewar, Alan" <ad759e@intl.att.com>
To: "'Dumitrescu, Cristian'" <cristian.dumitrescu@intel.com>,
	"'alangordondewar@gmail.com'" <alangordondewar@gmail.com>,
	"'Kantecki, Tomasz'" <tomasz.kantecki@intel.com>
Cc: "'dev@dpdk.org'" <dev@dpdk.org>, "'Alan Dewar'" <alan.dewar@att.com>
Subject: Re: [dpdk-dev] [PATCH v3] sched: make RED scaling configurable
Date: Tue, 26 Sep 2017 08:02:37 +0000	[thread overview]
Message-ID: <3F9268EEC0E43747A5FFFC6B48EF0321F90BE1@gbcdcmbx03.intl.att.com> (raw)
In-Reply-To: <3EB4FA525960D640B5BDFFD6A3D891267BAC0754@IRSMSX108.ger.corp.intel.com>

Hi Cristian,

No problem,  it doesn't sounds like it will be too difficult. 

Unfortunately I'm in the middle of another piece of work, so it will probably be a couple of weeks before I get around to it.

Regards
Alan

-----Original Message-----
From: Dumitrescu, Cristian [mailto:cristian.dumitrescu@intel.com] 
Sent: Monday, September 25, 2017 11:37 AM
To: alangordondewar@gmail.com; Kantecki, Tomasz <tomasz.kantecki@intel.com>
Cc: dev@dpdk.org; Alan Dewar <alan.dewar@att.com>
Subject: RE: [dpdk-dev] [PATCH v3] sched: make RED scaling configurable

<snip>...
 
> diff --git a/config/common_base b/config/common_base index 
> 5e97a08..2626557 100644
> --- a/config/common_base
> +++ b/config/common_base
> @@ -666,6 +666,13 @@ CONFIG_RTE_SCHED_COLLECT_STATS=n  
> CONFIG_RTE_SCHED_SUBPORT_TC_OV=n
>  CONFIG_RTE_SCHED_PORT_N_GRINDERS=8
>  CONFIG_RTE_SCHED_VECTOR=n
> +#
> +# RTE_RED_SCALING - number of fractional bits used for RED's moving
> average
> +# For every bit that RTE_RED_SCALING is reduced, the max-queue size 
> +can
> doubled
> +# RTE_RED_MAX_TH_MAX = (max-queue size - 1), max-queue size must be
> power of two
> +#
> +CONFIG_RTE_RED_SCALING=10
> +CONFIG_RTE_RED_MAX_TH_MAX=1023
> 

Alan, Tomasz,

Thank you for your work!

We generally want to avoid build-time configuration in favour of run-time configuration. Can you please rework this as run-time configuration?

I suggest we add a new API function to configure these parameters globally for all the RED objects (as opposed to per RED object), with the default values as the current values when this function is not called.

So NACK for now.

Regards,
Cristian

  reply	other threads:[~2017-09-26  8:03 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-09-01 13:41 [dpdk-dev] [PATCH] " alangordondewar
2017-09-11 15:51 ` Kantecki, Tomasz
2017-09-13 10:15 ` [dpdk-dev] [PATCH v2] " alangordondewar
2017-09-18 22:03   ` Kantecki, Tomasz
2017-09-20 13:12   ` [dpdk-dev] [PATCH v3] " alangordondewar
2017-09-25 10:36     ` Dumitrescu, Cristian
2017-09-26  8:02       ` Dewar, Alan [this message]
2017-10-03  9:21     ` [dpdk-dev] [PATCH v4] " alangordondewar
2017-10-03 17:15       ` Dumitrescu, Cristian
2018-01-02 16:21         ` Dumitrescu, Cristian
2018-01-02 16:43       ` Dumitrescu, Cristian
2018-01-03 14:29         ` Dewar, Alan
2018-01-03 16:20           ` Dumitrescu, Cristian
2018-01-04 13:34             ` Dewar, Alan
2018-01-04 18:25               ` Dumitrescu, Cristian

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=3F9268EEC0E43747A5FFFC6B48EF0321F90BE1@gbcdcmbx03.intl.att.com \
    --to=ad759e@intl.att.com \
    --cc=alan.dewar@att.com \
    --cc=alangordondewar@gmail.com \
    --cc=cristian.dumitrescu@intel.com \
    --cc=dev@dpdk.org \
    --cc=tomasz.kantecki@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).