DPDK patches and discussions
 help / color / mirror / Atom feed
From: Jerin Jacob <jerinjacobk@gmail.com>
To: Ganapati Kundapura <ganapati.kundapura@intel.com>
Cc: jerinj@marvell.com, jay.jayatheerthan@intel.com,
	 s.v.naga.harish.k@intel.com, abhinandan.gujjar@intel.com,
	dev@dpdk.org
Subject: Re: [PATCH v1] eventdev/crypto: flush ops when circ buffer is full
Date: Wed, 2 Aug 2023 21:47:29 +0530	[thread overview]
Message-ID: <CALBAE1NVTZzm2V1NuSpBg8NaCF18g6dtavZF1ZpLwUyTVMZRQA@mail.gmail.com> (raw)
In-Reply-To: <20230801054457.1184208-1-ganapati.kundapura@intel.com>

On Tue, Aug 1, 2023 at 11:15 AM Ganapati Kundapura
<ganapati.kundapura@intel.com> wrote:
>
> crypto ops from the circ buffer are not getting flushed
> to crypto dev when crypto dev becomes busy and circ buffer
> gets full.
>
> This patch flushes ops from circ buffer when circ buffer is full
> instead of returning without flushing.


Since it is bug, Please add Fixes: tag , Also change description accordingly.

> Signed-off-by: Ganapati Kundapura <ganapati.kundapura@intel.com>
>
> diff --git a/lib/eventdev/rte_event_crypto_adapter.c b/lib/eventdev/rte_event_crypto_adapter.c
> index 52a28e5..1b435c9 100644
> --- a/lib/eventdev/rte_event_crypto_adapter.c
> +++ b/lib/eventdev/rte_event_crypto_adapter.c
> @@ -248,9 +248,18 @@ eca_circular_buffer_flush_to_cdev(struct crypto_ops_circular_buffer *bufp,
>                 n = *tailp - *headp;
>         else if (*tailp < *headp)
>                 n = bufp->size - *headp;
> -       else {
> -               *nb_ops_flushed = 0;
> -               return 0;  /* buffer empty */
> +       else { /* head == tail case */
> +               /* when head == tail,
> +                * circ buff is either full(tail pointer roll over) or empty
> +                */
> +               if (bufp->count != 0) {
> +                       /* circ buffer is full */
> +                       n = bufp->count;
> +               } else {
> +                       /* circ buffer is empty */
> +                       *nb_ops_flushed = 0;
> +                       return 0;  /* buffer empty */
> +               }
>         }
>
>         *nb_ops_flushed = rte_cryptodev_enqueue_burst(cdev_id, qp_id,
> --
> 2.6.4
>

  reply	other threads:[~2023-08-02 16:17 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-08-01  5:44 Ganapati Kundapura
2023-08-02 16:17 ` Jerin Jacob [this message]
2023-08-03  8:34   ` Kundapura, Ganapati
2023-08-03  8:32 ` [PATCH v2] eventdev/crypto: fix circular buffer full case Ganapati Kundapura
2023-08-08 12:59   ` Jerin Jacob

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=CALBAE1NVTZzm2V1NuSpBg8NaCF18g6dtavZF1ZpLwUyTVMZRQA@mail.gmail.com \
    --to=jerinjacobk@gmail.com \
    --cc=abhinandan.gujjar@intel.com \
    --cc=dev@dpdk.org \
    --cc=ganapati.kundapura@intel.com \
    --cc=jay.jayatheerthan@intel.com \
    --cc=jerinj@marvell.com \
    --cc=s.v.naga.harish.k@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).