DPDK patches and discussions
 help / color / mirror / Atom feed
From: Jerin Jacob <jerinjacobk@gmail.com>
To: "Kundapura, Ganapati" <ganapati.kundapura@intel.com>
Cc: "Jayatheerthan, Jay" <jay.jayatheerthan@intel.com>,
	dpdk-dev <dev@dpdk.org>
Subject: Re: [dpdk-dev] [PATCH] eventdev: configure the Rx event buffer size
Date: Mon, 19 Jul 2021 21:43:55 +0530	[thread overview]
Message-ID: <CALBAE1Nt1hX0aSsp4aJ=zy=auPRiOnZu8gb35-gL6J+hkqibOw@mail.gmail.com> (raw)
In-Reply-To: <CO1PR11MB488256E50332FAF0F5F3B71B87E19@CO1PR11MB4882.namprd11.prod.outlook.com>

On Mon, Jul 19, 2021 at 8:57 PM Kundapura, Ganapati
<ganapati.kundapura@intel.com> wrote:
>
> Hi Jerin,

HI Ganapati

>    Please find my response in lined.
>
> -----Original Message-----
> From: Jerin Jacob <jerinjacobk@gmail.com>
> Sent: 19 July 2021 12:14
> To: Kundapura, Ganapati <ganapati.kundapura@intel.com>
> Cc: Jayatheerthan, Jay <jay.jayatheerthan@intel.com>; dpdk-dev <dev@dpdk.org>
> Subject: Re: [dpdk-dev] [PATCH] eventdev: configure the Rx event buffer size
>
> On Fri, Jul 16, 2021 at 10:33 PM Ganapati Kundapura <ganapati.kundapura@intel.com> wrote:
> >
> > As of now Rx event buffer size is static and set to 128.
> >
> > This patch sets the Rx event buffer size to 192, configurable at
> > compile time and also errors out at run time if Rx event buffer size
> > is configured more than 16 bits.
> >
> > Signed-off-by: Ganapati Kundapura <ganapati.kundapura@intel.com>
> > ---
> >  config/rte_config.h                     |  1 +
> >  lib/eventdev/rte_event_eth_rx_adapter.c | 14 +++++++++++++-
> >  2 files changed, 14 insertions(+), 1 deletion(-)
> >
> > diff --git a/config/rte_config.h b/config/rte_config.h index
> > 590903c..3d938c8 100644
> > --- a/config/rte_config.h
> > +++ b/config/rte_config.h
> > @@ -77,6 +77,7 @@
> >  #define RTE_EVENT_ETH_INTR_RING_SIZE 1024  #define
> > RTE_EVENT_CRYPTO_ADAPTER_MAX_INSTANCE 32  #define
> > RTE_EVENT_ETH_TX_ADAPTER_MAX_INSTANCE 32
> > +#define RTE_EVENT_ETH_RX_ADAPTER_BUFFER_SIZE 128
>
> We are limiting any configuration to rte_config.h file.
> Could you make it dynamic with the default value and application can pass the value kind of scheme?
> [Ganapati]
> Making the Rx event buffer size dynamic seems to be a good idea but in case of rx adapter,
> either passing event buffer size to adapter create api requires api signature change which breaks ABI
> or by adding event buffer size in port_config parameter which comes from eventdev
> to adapter create function is not scalable as user can also call create_ext() with its own callback
> and parameter to callback is void * which is interpreted by user space callback function.
>
> I think one way to do the event buffer size dynamic is to add new api to set the event buffer size.
> If called, it'll set the event buffer size to the value passed otherwise rx adapter instance create api will do with
> default value.
>
> Let me know your opinion on this.

we can break the ABI in v21.11 so create API config structure can change.
Please send depreciation notice and submit the implementation for 21.11,

>

      reply	other threads:[~2021-07-19 16:14 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-07-16 17:02 Ganapati Kundapura
2021-07-19  6:43 ` Jerin Jacob
2021-07-19 15:26   ` Kundapura, Ganapati
2021-07-19 16:13     ` Jerin Jacob [this message]

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='CALBAE1Nt1hX0aSsp4aJ=zy=auPRiOnZu8gb35-gL6J+hkqibOw@mail.gmail.com' \
    --to=jerinjacobk@gmail.com \
    --cc=dev@dpdk.org \
    --cc=ganapati.kundapura@intel.com \
    --cc=jay.jayatheerthan@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).