netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Stephen Hemminger <shemminger@osdl.org>
To: David Miller <davem@davemloft.net>
Cc: netdev@vger.kernel.org
Subject: [PATCH 5/8] netpoll deferred transmit path
Date: Thu, 26 Oct 2006 15:46:53 -0700	[thread overview]
Message-ID: <20061026225646.186971212@osdl.org> (raw)
In-Reply-To: 20061026225535.443288276@osdl.org

[-- Attachment #1: netpoll-defer-tx.patch --]
[-- Type: text/plain, Size: 1618 bytes --]

When the netpoll beast got busy, he tended to babble.
Instead of talking out of his large mouth as normal,
he tended to try to snort out other orifices. This lead
to words (skbs) ending up in odd places (like NIT) that
he did not intend.

The normal way of talking wouldn't work, but he could
at least change to using the same tone all the time.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
---
 net/core/netpoll.c |   21 +++++++++++++++++++--
 1 file changed, 19 insertions(+), 2 deletions(-)

--- linux-2.6.orig/net/core/netpoll.c
+++ linux-2.6/net/core/netpoll.c
@@ -55,9 +55,25 @@ static void queue_process(void *p)
 	struct netpoll_info *npinfo = p;
 	struct sk_buff *skb;
 
-	while ((skb = skb_dequeue(&npinfo->txq)))
-		dev_queue_xmit(skb);
+	while ((skb = skb_dequeue(&npinfo->txq))) {
+		struct net_device *dev = skb->dev;
 
+		if (!netif_device_present(dev) || !netif_running(dev)) {
+			__kfree_skb(skb);
+			continue;
+		}
+
+		netif_tx_lock_bh(dev);
+		if (netif_queue_stopped(dev) ||
+		    dev->hard_start_xmit(skb, dev) != NETDEV_TX_OK) {
+			skb_queue_head(&npinfo->txq, skb);
+			netif_tx_unlock_bh(dev);
+
+			schedule_delayed_work(&npinfo->tx_work, HZ/10);
+			return;
+		}
+		netif_tx_unlock_bh(dev);
+	}
 }
 
 void netpoll_queue(struct sk_buff *skb)
@@ -756,6 +772,7 @@ void netpoll_cleanup(struct netpoll *np)
 			if (atomic_dec_and_test(&npinfo->refcnt)) {
 				skb_queue_purge(&npinfo->arp_tx);
  				skb_queue_purge(&npinfo->txq);
+				cancel_rearming_delayed_work(&npinfo->tx_work);
  				flush_scheduled_work();
 
 				kfree(npinfo);

--
Stephen Hemminger <shemminger@osdl.org>


  parent reply	other threads:[~2006-10-26 23:00 UTC|newest]

Thread overview: 16+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-10-26 22:46 [PATCH 0/8] netpoll: A Halloween horror mystery Stephen Hemminger
2006-10-26 22:46 ` [PATCH 1/8] netpoll: skb private pool management Stephen Hemminger
2006-10-27  0:12   ` David Miller
2006-10-27  1:04     ` Stephen Hemminger
2006-10-27  1:08       ` David Miller
2006-10-27  2:29         ` Stephen Hemminger
2006-11-14  1:00           ` David Miller
2006-11-14 21:55             ` netpoll patches Stephen Hemminger
2006-11-15  4:43               ` David Miller
2006-10-26 22:46 ` [PATCH 2/8] netpoll info leak Stephen Hemminger
2006-10-26 22:46 ` [PATCH 3/8] netpoll per device txq Stephen Hemminger
2006-10-26 22:46 ` [PATCH 4/8] netpoll setup error handling Stephen Hemminger
2006-10-26 22:46 ` Stephen Hemminger [this message]
2006-10-26 22:46 ` [PATCH 6/8] netpoll retry cleanup Stephen Hemminger
2006-10-26 22:46 ` [PATCH 7/8] netpoll queue cleanup Stephen Hemminger
2006-10-26 22:46 ` [PATCH 8/8] netpoll header cleanup Stephen Hemminger

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=20061026225646.186971212@osdl.org \
    --to=shemminger@osdl.org \
    --cc=davem@davemloft.net \
    --cc=netdev@vger.kernel.org \
    /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).