netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
To: Xin Long <lucien.xin@gmail.com>
Cc: network dev <netdev@vger.kernel.org>,
	linux-sctp@vger.kernel.org, davem@davemloft.net,
	Neil Horman <nhorman@tuxdriver.com>
Subject: Re: [PATCH net 2/3] sctp: abandon the whole msg if one part of a fragmented message is abandoned
Date: Mon, 27 Nov 2017 10:26:38 -0200	[thread overview]
Message-ID: <20171127122638.GB3473@localhost.localdomain> (raw)
In-Reply-To: <527417b933f70998528cbbfcdf4f6a8405fb0292.1511615658.git.lucien.xin@gmail.com>

On Sat, Nov 25, 2017 at 09:18:35PM +0800, Xin Long wrote:
> As rfc3758#section-3.1 demands:
> 
>    A3) When a TSN is "abandoned", if it is part of a fragmented message,
>        all other TSN's within that fragmented message MUST be abandoned
>        at the same time.
> 
> Besides, if it couldn't handle this, the rest frags would never get
> assembled in peer side.
> 
> This patch supports it by adding abandoned flag in sctp_datamsg, when
> one chunk is being abandoned, set chunk->msg->abandoned as well. Next
> time when checking for abandoned, go checking chunk->msg->abandoned
> first.
> 
> Signed-off-by: Xin Long <lucien.xin@gmail.com>

Acked-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>

> ---
>  include/net/sctp/structs.h |  3 ++-
>  net/sctp/chunk.c           |  7 +++++++
>  net/sctp/outqueue.c        | 12 ++++++++----
>  3 files changed, 17 insertions(+), 5 deletions(-)
> 
> diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h
> index 16f949e..2f8f93d 100644
> --- a/include/net/sctp/structs.h
> +++ b/include/net/sctp/structs.h
> @@ -503,7 +503,8 @@ struct sctp_datamsg {
>  	/* Did the messenge fail to send? */
>  	int send_error;
>  	u8 send_failed:1,
> -	   can_delay;	    /* should this message be Nagle delayed */
> +	   can_delay:1,	/* should this message be Nagle delayed */
> +	   abandoned:1;	/* should this message be abandoned */
>  };
>  
>  struct sctp_datamsg *sctp_datamsg_from_user(struct sctp_association *,
> diff --git a/net/sctp/chunk.c b/net/sctp/chunk.c
> index 7b261af..9213805 100644
> --- a/net/sctp/chunk.c
> +++ b/net/sctp/chunk.c
> @@ -53,6 +53,7 @@ static void sctp_datamsg_init(struct sctp_datamsg *msg)
>  	msg->send_failed = 0;
>  	msg->send_error = 0;
>  	msg->can_delay = 1;
> +	msg->abandoned = 0;
>  	msg->expires_at = 0;
>  	INIT_LIST_HEAD(&msg->chunks);
>  }
> @@ -304,6 +305,9 @@ int sctp_chunk_abandoned(struct sctp_chunk *chunk)
>  	if (!chunk->asoc->peer.prsctp_capable)
>  		return 0;
>  
> +	if (chunk->msg->abandoned)
> +		return 1;
> +
>  	if (SCTP_PR_TTL_ENABLED(chunk->sinfo.sinfo_flags) &&
>  	    time_after(jiffies, chunk->msg->expires_at)) {
>  		struct sctp_stream_out *streamout =
> @@ -316,6 +320,7 @@ int sctp_chunk_abandoned(struct sctp_chunk *chunk)
>  			chunk->asoc->abandoned_unsent[SCTP_PR_INDEX(TTL)]++;
>  			streamout->ext->abandoned_unsent[SCTP_PR_INDEX(TTL)]++;
>  		}
> +		chunk->msg->abandoned = 1;
>  		return 1;
>  	} else if (SCTP_PR_RTX_ENABLED(chunk->sinfo.sinfo_flags) &&
>  		   chunk->sent_count > chunk->sinfo.sinfo_timetolive) {
> @@ -324,10 +329,12 @@ int sctp_chunk_abandoned(struct sctp_chunk *chunk)
>  
>  		chunk->asoc->abandoned_sent[SCTP_PR_INDEX(RTX)]++;
>  		streamout->ext->abandoned_sent[SCTP_PR_INDEX(RTX)]++;
> +		chunk->msg->abandoned = 1;
>  		return 1;
>  	} else if (!SCTP_PR_POLICY(chunk->sinfo.sinfo_flags) &&
>  		   chunk->msg->expires_at &&
>  		   time_after(jiffies, chunk->msg->expires_at)) {
> +		chunk->msg->abandoned = 1;
>  		return 1;
>  	}
>  	/* PRIO policy is processed by sendmsg, not here */
> diff --git a/net/sctp/outqueue.c b/net/sctp/outqueue.c
> index 7029f8b..4ab164b 100644
> --- a/net/sctp/outqueue.c
> +++ b/net/sctp/outqueue.c
> @@ -364,10 +364,12 @@ static int sctp_prsctp_prune_sent(struct sctp_association *asoc,
>  	list_for_each_entry_safe(chk, temp, queue, transmitted_list) {
>  		struct sctp_stream_out *streamout;
>  
> -		if (!SCTP_PR_PRIO_ENABLED(chk->sinfo.sinfo_flags) ||
> -		    chk->sinfo.sinfo_timetolive <= sinfo->sinfo_timetolive)
> +		if (!chk->msg->abandoned &&
> +		    (!SCTP_PR_PRIO_ENABLED(chk->sinfo.sinfo_flags) ||
> +		     chk->sinfo.sinfo_timetolive <= sinfo->sinfo_timetolive))
>  			continue;
>  
> +		chk->msg->abandoned = 1;
>  		list_del_init(&chk->transmitted_list);
>  		sctp_insert_list(&asoc->outqueue.abandoned,
>  				 &chk->transmitted_list);
> @@ -404,10 +406,12 @@ static int sctp_prsctp_prune_unsent(struct sctp_association *asoc,
>  	q->sched->unsched_all(&asoc->stream);
>  
>  	list_for_each_entry_safe(chk, temp, &q->out_chunk_list, list) {
> -		if (!SCTP_PR_PRIO_ENABLED(chk->sinfo.sinfo_flags) ||
> -		    chk->sinfo.sinfo_timetolive <= sinfo->sinfo_timetolive)
> +		if (!chk->msg->abandoned &&
> +		    (!SCTP_PR_PRIO_ENABLED(chk->sinfo.sinfo_flags) ||
> +		     chk->sinfo.sinfo_timetolive <= sinfo->sinfo_timetolive))
>  			continue;
>  
> +		chk->msg->abandoned = 1;
>  		sctp_sched_dequeue_common(q, chk);
>  		asoc->sent_cnt_removable--;
>  		asoc->abandoned_unsent[SCTP_PR_INDEX(PRIO)]++;
> -- 
> 2.1.0
> 
> --
> To unsubscribe from this list: send the line "unsubscribe linux-sctp" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> 

  parent reply	other threads:[~2017-11-27 12:26 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-11-25 13:18 [PATCH net 0/3] a couple of fixes for chunks abandoned in prsctp Xin Long
2017-11-25 13:18 ` [PATCH net 1/3] sctp: only update outstanding_bytes for transmitted queue when doing prsctp_prune Xin Long
2017-11-25 13:18   ` [PATCH net 2/3] sctp: abandon the whole msg if one part of a fragmented message is abandoned Xin Long
2017-11-25 13:18     ` [PATCH net 3/3] sctp: do not abandon the other frags in unsent outq if one msg has outstanding frags Xin Long
2017-11-27 12:26       ` Marcelo Ricardo Leitner
2017-11-27 12:26     ` Marcelo Ricardo Leitner [this message]
2017-11-27 12:26   ` [PATCH net 1/3] sctp: only update outstanding_bytes for transmitted queue when doing prsctp_prune Marcelo Ricardo Leitner
2017-11-27 13:58 ` [PATCH net 0/3] a couple of fixes for chunks abandoned in prsctp Neil Horman
2017-11-27 19:52 ` David Miller

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=20171127122638.GB3473@localhost.localdomain \
    --to=marcelo.leitner@gmail.com \
    --cc=davem@davemloft.net \
    --cc=linux-sctp@vger.kernel.org \
    --cc=lucien.xin@gmail.com \
    --cc=netdev@vger.kernel.org \
    --cc=nhorman@tuxdriver.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).