netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: David Miller <davem@davemloft.net>
To: simon.guinot@sequanux.org
Cc: thomas.petazzoni@free-electrons.com, netdev@vger.kernel.org,
	musv@gmx.de, andreas.tobler@cloudguard.ch,
	gregory.clement@free-electrons.com,
	antoine.tenart@free-electrons.com, mw@semihalf.com,
	stable@vger.kernel.org
Subject: Re: [PATCH] net: mvneta: fix handling of the Tx descriptor counter
Date: Sat, 11 Nov 2017 18:45:04 +0900 (KST)	[thread overview]
Message-ID: <20171111.184504.2151388779461931723.davem@davemloft.net> (raw)
In-Reply-To: <20171108165835.13453-1-simon.guinot@sequanux.org>

From: Simon Guinot <simon.guinot@sequanux.org>
Date: Wed,  8 Nov 2017 17:58:35 +0100

> @@ -2413,8 +2416,7 @@ static int mvneta_tx(struct sk_buff *skb, struct net_device *dev)
>  		if (txq->count >= txq->tx_stop_threshold)
>  			netif_tx_stop_queue(nq);
>  
> -		if (!skb->xmit_more || netif_xmit_stopped(nq) ||
> -		    txq->pending + frags > MVNETA_TXQ_DEC_SENT_MASK)
> +		if (!skb->xmit_more || netif_xmit_stopped(nq))
>  			mvneta_txq_pend_desc_add(pp, txq, frags);
>  		else
>  			txq->pending += frags;

As David Laight said, you should not allow unlimited amounts of
->xmit_more frames to be queued without a TX doorbell update.

Therefore, please keep some kind of limit here otherwise latency
will spike in some circumstances.

  parent reply	other threads:[~2017-11-11  9:45 UTC|newest]

Thread overview: 22+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <20171018223425.42ce7a74@gmx.de>
2017-10-18 20:55 ` Problems with mvneta Thomas Petazzoni
2017-10-19 22:25   ` Sven Müller
2017-10-20  7:09     ` Thomas Petazzoni
2017-10-23  6:29       ` Andreas Tobler
2017-10-23  9:30         ` Sven Müller
2017-10-31 14:23           ` Sven Müller
2017-10-31 14:27             ` Thomas Petazzoni
2017-10-31 17:09               ` Simon Guinot
2017-10-31 20:23                 ` Thomas Petazzoni
2017-11-01  8:10                   ` Marcin Wojtas
2017-11-08 16:58                     ` [PATCH] net: mvneta: fix handling of the Tx descriptor counter Simon Guinot
2017-11-08 17:03                       ` David Laight
2017-11-08 17:17                       ` Simon Guinot
2017-11-09 19:19                         ` Andreas Tobler
2017-11-11  9:45                       ` David Miller [this message]
2017-11-13 14:51                         ` Simon Guinot
2017-11-13 14:54                           ` David Miller
2017-11-13 15:36                             ` Simon Guinot
2017-11-20 14:58                               ` David Laight
2017-11-13 15:27                       ` [PATCH v2] " Simon Guinot
2017-11-14 12:53                         ` David Miller
2017-11-10 10:22             ` [PATCH] " Sven Müller

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=20171111.184504.2151388779461931723.davem@davemloft.net \
    --to=davem@davemloft.net \
    --cc=andreas.tobler@cloudguard.ch \
    --cc=antoine.tenart@free-electrons.com \
    --cc=gregory.clement@free-electrons.com \
    --cc=musv@gmx.de \
    --cc=mw@semihalf.com \
    --cc=netdev@vger.kernel.org \
    --cc=simon.guinot@sequanux.org \
    --cc=stable@vger.kernel.org \
    --cc=thomas.petazzoni@free-electrons.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).